Executive summary  1

1.       Scene setter: economic situation and outlook  3

2.       Imbalances, risks and adjustment 10

2.1.   Productivity and the allocation of resources 11

2.2.   External sustainability and competitiveness 25

2.3.   General government debt, fiscal framework and public sector governance  37

2.4.   Corporate and household debt and financial sector stability  51

3.       Other structural issues 63

3.1.   Social policy, pensions and active labour market policies 65

3.2.   Healthcare system   73

3.3.   Network industries and environment 77

3.4.   Education and innovation  81

3.5.   European Structural and Investment Funds  85

AA.   Overview Table  87

AB.   Standard Tables 93

LIST OF Tables

1.1.     Key economic, financial and social indicators 8

1.2.     MIP scoreboard   9

AB.1.  Macroeconomic indicators 93

AB.2.  Financial market indicators 94

AB.3.  Taxation indicators 95

AB.4.  Labour market and social indicators 96

AB.5.  Expenditure on social protection benefits (% of GDP) 97

AB.6.  Product market performance and policy indicators 98

AB.7.  Green growth  99

LIST OF Graphs

1.1.     Croatia: External and domestic demand, contribution to growth  3

1.2.     Real imports and real exports in Croatia and EU-10  4

1.3.     Registered employed population and unemployment rate  4

1.4.     Wage and price growth  5

1.5.     Real estate price  6

2.1.1.  Growth accounting, 2002-2008 average (%) 11

2.1.2.  Growth accounting, 2008-2013 average (%) 11

2.1.3.  Productivity and employment dynamics (2008-14) 12

2.1.4.  Distribution of firms according to productivity and employment dynamics 12

2.1.5.  Migration flows and total population  14

2.1.6.  Productivity vis-à-vis EU-28 and average earnings in Croatia, EU and other OECD countries (2014) 16

2.1.7.  Share of temporary and recently employed (less than three months) workers 17

2.1.8.  Average monthly earnings by educational attainment (2010) 18

2.1.9.  Unemployment rate by education level (2013) 18

2.2.1.  Net lending position – transaction side  25

2.2.2.  Net lending position - financing side  25

2.2.3.  Real Effective Exchange Rates, ULC- and export prices-deflated   26

2.2.4.  Competitiveness and export performance (2008-2014) 26

2.2.5.  Elasticity of exports to income, price and distance  27

2.2.6.  Decomposition of changes in export market shares 27

2.2.7.  SITC structure of goods exports in Croatia and peer economies (percentage of total exports, 2013) 28

2.2.8.  BEC structure of exports of goods in Croatia and peer economies (percentage of total exports, 2013) 29

2.2.9.  Import content of exports (2005) 29

2.2.10.Internationalisation relevant indicators – distance from EU average (std. dev.) 30

2.2.11.Export of services 31

2.2.12.Revenue from tourism and number of nights in tourism establishments 31

2.2.13.BEC Structure of imports, 2013  32

2.2.14.Import elasticities with respect to final demand components 34

2.2.15.Balance of primary and secondary incomes 35

2.2.16.Net international investment position (NIIP) 35

2.2.17.CA surplus necessary to stabilise NIIP at -60% of GDP by 2024 - sensitivity analysis 36

2.3.1.  General government debt 37

2.3.2.  Development of general government debt and deficit and Maastricht criteria   37

2.3.3.  Contributions to change in general government debt-to-GDP ratio  38

2.3.4.  Structure of expenditure by COFOG function (2012) 41

2.3.5.  Outstanding liabilities of hospitals 41

2.3.6.  Structure of expenditure by ESA category (2013) 42

2.3.7.  Total factor productivity by sector and ownership (average for 2006-13) 49

2.4.1.  Money (M1) and domestic credit growth  51

2.4.2.  Consolidated debt-to-GDP ratio, Non-Financial Corporations 52

2.4.3.  Corporate foreign debt 53

2.4.4.  Debt-to-GDP ratio - Households sector 57

2.4.5.  Domestic credit institutions’ claims on households by type and currency  58

2.4.6.  Financial sector stability and profitability  60

3.1.1.  Labour market and social indicators 65

3.1.2.  Population structure  68

3.1.3.  Many Member States achieve higher pensions with lower expenditure  69

3.1.4.  New pensioners (excluding special pensions) 69

3.1.5.  Pensioners (Q3‑2014) 70

3.4.1.  Strategies judged very important by innovative and non-innovative enterprises (2012) 82

3.4.2.  R&D intensity by sector (2013) 83

LIST OF Boxes

1.1.     Economic surveillance process 7

2.1.1.  Doing business in Croatia   20

2.3.1.  Long-term projections of general government debt 39

2.4.1.  Corporate debt in private and publicly owned companies 54

LIST OF Maps

No table of contents entries found.

In 2014, Croatia’s economy contracted for its sixth year in a row and although the recession is expected to come to an end in 2015, the economic outlook remains bleak. The pace of the contraction abated over the course of 2014, bringing the overall fall in GDP to -0.5 %. Growth is set to be just above zero in 2015 and to pickup timidly to 1% in 2016. Against this background, the unemployment rate is not expected to decline significantly from the current 17%. Internal demand should progressively start contributing positively to growth on the back of investments spurred by EU funds, while the export performance should remain strong as the recovery progresses in the EU. Significant fiscal consolidation and deleveraging needs nevertheless weigh on the growth perspectives.

In March 2014, the Commission concluded that Croatia was experiencing excessive macroeconomic imbalances. More specifically the risks stemming from high external liabilities, declining export performance, highly leveraged firms and fast increasing general government debt, all in a context of low growth and poor adjustment capacity, required specific monitoring and strong policy action. The identified imbalances strongly informed the country-specific recommendations issued to Croatia by the Council in June 2014. This Country Report assesses Croatia’s economy against the background of the Commission’s Annual Growth Survey which recommends three main pillars for the EU’s economic and social policy in 2015: investment, structural reforms, and fiscal responsibility. In line with the Investment Plan for Europe, it also explores ways to maximise the impact of public resources and unlock private investment. Finally, it assesses Croatia in light of the findings of the 2015 Alert Mechanism Report, in which the Commission found it useful to further examine the persistence of imbalances or their unwinding. The main findings of the In-Depth Review contained in this Country Report are:

· Subdued growth, delayed restructuring of firms and the dismal performance of employment are rooted in inefficiencies in the allocation of resources. Low employment is partly related to labour market institutions and policy settings. The unfavourable business environment is a major drag on the adjustment capacity of the economy.

· Weak external competiveness and the large negative international investment position threaten external sustainability. Croatia’s export of goods remains far below its potential. Cost and non-cost competiveness factors partly contribute to the weak export base. The large stock of liabilities fuels the outflow of investment income and is a source of vulnerability.

· Rising general government debt is a source of concern, which is aggravated by weaknesses in public sector governance. The prolonged recession coupled with significant assumption of liabilities of public enterprises have resulted in a sharp increase in public debt. The high and rising interest expenditure may signal the kicking-in of a snow-ball effect. Administrative fragmentation and weak public sector governance leads to low efficiency of spending and loose management of public finances.

· Household and corporate debt holds back consumption and investment. Both sectors are deleveraging at a moderate pace, but corporate debt remains highly concentrated in weakly profitable corporations and the relatively high rates of non-performing loans – especially in the corporate-sector need monitoring.

The Country Report also analyses other macroeconomic and structural issues:

· Current policies insufficiently address the needs of vulnerable workers and discourage participation of elderly workers.

· The healthcare system is a source of fiscal risks.

· Network industries are a bottleneck to the economy’s growth potential.

· Weaknesses in the education system together with scarce innovation capacity may undermine long-term productivity growth.

· Limited administrative capacity threatens the full absorption of EU funds.

There has been some progress in addressing the eight MIP relevant Country Specific Recommendations. The second phase of the labour market reform was fully implemented and the additional supervisory diagnostic exercises on banks have been satisfactorily completed, confirming the resilience of the Croatian banking sector. In the majority of areas covered by the recommendations some or limited progress was registered. Moreover, the MIP relevant country-specific recommendation to reduce access to early retirement has remained unaddressed.

The Country Report discusses the policy challenges stemming from the analysis of macro-economic imbalances:

· Administrative fragmentation and loose public finances management, including in the health sector, contribute to significant inefficiencies in the public sector. A complex functional distribution of governance structures, including a suboptimal articulation of the responsibilities of local tiers of government result in weak control of public finances, low efficiency of public spending and weak corruption prevention mechanisms. 

· The rigid business environment is a bottleneck to growth. Strict regulation and high administrative and para-fiscal burdens drive up the cost of entrepreneurial activities. Weaknesses in the application of the regulatory impact assessment and legislative planning contribute to regulatory instability. Competition on the internal market is limited by the near monopoly regime in which some companies operate. The large share of public enterprises in the economy creates an uneven level playing field for private businesses. Inefficiencies in the justice system drive up the cost of litigation.

· Institutional and policy shortcomings affect the performance of the labour market. The severance payment regime hinders labour mobility and discourages the use of open-ended contracts. The multi-layered social benefits system and generous early retirement options create disincentives to work. The wage setting regime is not conducive to aligning wage dynamics to macroeconomic conditions. 

· As regard corporate debt, the challenge is to ensure quick access to restructuring and to reconcile the protection of creditors and the efficiency of the deleveraging process, without undermining financial stability.  

· The financial sector faces challenges from the degree of government intervention in the economy. Recent policies, including the freeze of the kuna/Swiss franc exchange rate for Swiss franc indexed loans, highlight the need for a more predicable legal system, a proportionate and equitable intervention of the state in the financial sector and the implementation of a permanent scheme for dealing with personal insolvency.

· Further investments in network infrastructure could enable Croatia to leverage its geo-strategic position.

· Mismatches between labour demand and supply hold back employment. Innovation capacity is negatively affected by a low level of private and public investment in R&D, but also by weaknesses in the governance of public research and in the design of support instruments. 

· Inadequate administrative capacities hinder the full absorption of the EU funds. In particular the capacities of public administration are weak and the transparency and efficiency of public procurement at both central and local levels are low.  

Several new Member States in central and eastern Europe (the EU-10) experienced a rapid building up of imbalances in the years that preceded the crisis ([1]). EU accession signalled prospects for faster income convergence and ushered in large-scale bank-intermediated foreign capital flows which fuelled domestic demand booms. Dynamic internal demand spilled over into imports and consequently widened current account deficits. Large increases in wages and prices eroded the cost competitiveness and compressed the tradable sector.  The result was an excessive build-up of debt, mostly owed to foreigners, without a corresponding debt-servicing capacity, deterioration of competitiveness, and inadequate policy space.

Graph 1.1:     Croatia: External and domestic demand, contribution to growth

Source: European Commission

The unwinding of macroeconomic imbalances was relatively swift, and most EU-10 Member States resumed growth in 2010. The external adjustment was facilitated by sudden capital outflows, sharp wage contraction, bold fiscal consolidation policies and in some cases currency depreciation. Relatively flexible labour markets allowed for wage adjustment and sector reallocations, and — following the great trade collapse of 2008 — exports accelerated swiftly again, countering the shrinking internal demand. Tightening credit conditions in the private sector forced a sharp adjustment in the indebtedness of non-financial corporations and households, while pressures to restore market confidence encouraged bolt fiscal consolidation measures. By 2010, however, internal demand had finished contracting, heralding the start of a new, more balanced, expansionary phase.

Despite not joining the EU until 2013, Croatia experienced similar adverse macroeconomic trends in the pre-crisis years. Strong capital inflows — partly channelled through Croatia’s largely foreign-owned banking sector — underpinned the robust growth up to the 2008 global financial crisis. Foreign direct investment (FDI), including cross-border intercompany lending from parent holdings, was also sizable. To a greater extent than peer economies in the region, the sizeable FDI inflows largely bypassed the tradable sector. The investment-led internal demand contributed to rapid import penetration. Though price and wages dynamics were contained in relative terms, subdued productivity dynamics resulted in increasing unit labour costs (ULC). In a context of a tightly managed euro exchange rate, cost competitiveness losses accumulated, and the already poor export performance deteriorated further. As a result, by 2008 Croatia registered an overall negative net international investment position (NIIP) of over 75% of GDP and a record current account deficit of 8.9% of GDP. At 110.8% and 36% of GDP respectively, corporate and household debt and general government debt were not particularly high when the crisis flared up, though gross external debt was already in excess of 100 % of GDP.

In contrast to its regional peers, six years after the bust of the financial bubble, Croatia is still mired in recession and struggling with the rebalancing of its economy. The prolonged recession has reduced real GDP by about 12.5%, while the unemployment rate has almost doubled from 8.9% in 2008 to 17.0% in 2014. Investment activity was hit first and hardest: from a peak of 28% of GDP in 2008, investments plummeted to 19 % in 2014, a real decline entailing a particularly steep drop in construction activity. However, the deterioration of the labour market and falling disposable income, combined with worsening consumer sentiment and tighter bank lending conditions rapidly affected household consumption which contracted over the same period by almost 13 percentage points (pps.). As a result of falling internal demand, imports contracted sharply, but since only limited investment had gone into productive domestic assets and economic transformation, exports failed to rebound. Loose fiscal policies partly cushioned the impact of the crisis but general government debt, spiralling from 36 % of GDP in 2008 to 81.4 % in 2014, rapidly emerged as a new concern. Leverage ratios in the private sector also continued to deteriorate, up to 2010. Growing public and private indebtedness resulted in a worsening of the NIIP and a further increase in gross external debt.

Graph 1.2:     Real imports and real exports in Croatia and EU-10

Source: European Commission

The depth and duration of the crisis revealed structural weaknesses that had accumulated during the previous expansionary phase. Between 2002 and 2008, growth was mainly driven by labour input, while labour productivity growth was subdued in comparison with the catching-up EU-10 economies. Given the positive contribution of capital deepening, low productivity dynamics are entirely attributable to the weak growth of total factor productivity (TFP). Between 2002 and 2008, TFP growth averaged only 0.7% per year – as opposed to an average of 3.0% in EU-10. As in most Member States, the contribution of TFP turned negative with the crisis. Whereas five years after the crisis TFP growth recovered in EU-10, in the case of Croatia the average growth for 2008-2013 was still negative. Low productivity growth is typically associated with different dimensions of allocative inefficiency. The inefficient allocation of resources is also likely to have contributed to the build-up of macroeconomic imbalances and is slowing down the adjustment process.

In the absence of wide-ranging micro-structural reforms, the rebalancing of the economy is set to be slow and will continue to drag on economic growth. The economy is currently set to have further contracted by 0.5 in 2014 and is expected to stagnate at around 0.2% in 2015. In 2016, growth should be in firmer positive territory, but, at just 1%, it will still be relatively feeble. Also, these forecasts present risks to the downside as they do not factor in the fiscal consolidation measures needed to bring the spiralling public debt dynamic under control.

Graph 1.3:     Registered employed population and unemployment rate

Source: Central Bureau of Statistics

The negative NIIP was only marginally affected by the turnaround in the current account and gross external debt remains a factor of vulnerability. Following a sharp import contraction, the current-account balance turned to a small surplus in 2012. On the back of recent improvement in export performance, the current-account balance surplus is set to reach a surplus of 0.9% of GDP in 2014. As a consequence, the negative NIIP broadly stabilised in Q3-2014 at an estimated ‑88.2% of GDP, considerably below the peak values registered at the beginning of 2011. The reduction was partly supported by positive valuation effects. With the kuna relatively stable against the euro, these effects probably reflect a deterioration of confidence in the Croatian economy and the good performance of most EU stock markets in the first half of 2014, but also the depreciation of the kuna against the dollar in the third quarter. The stabilisation of the NIIP is accompanied by a broad stabilisation of the net external debt, as the accumulation of reserves by the central bank offset the government’s increasing external indebtedness. 

The fall in the employment rate might be close to bottoming out, but future employment growth is likely to be slowed down by the sluggish recovery. Labour force survey (LFS) data for Q3-2014 signalled an increase in the activity rate (67.6%) and employment rate (56.9%): both registered an increase with respect to the previous quarter (65.6% and 55.6% respectively) and the same quarter of previous year (64.7% and 53.7% respectively). Unemployment also decreased in the third quarter, reaching 15.8%, down from 16.7% in the previous quarter and 17% in the same quarter of the previous year. Administrative data likewise show a decrease in the unemployment. Despite an increase in the monthly unemployment rate from 18.7% to 19.2% in December 2014, the figure was well below the unemployment rate of 21% registered in the same month of the previous year. These figures are nevertheless likely to be affected by the shortening of the time window available for registering as unemployed. In contrast to the LFS figures, registered employment decreased mildly throughout 2014, which is probably more consistent with the contracting economy. Despite the contrasting indicators, it is likely that employment destruction is coming to an end and unemployment will see a moderate reduction in 2015 and 2016

Real wage adjustment is hindered by the subdued price dynamics. Nominal wages in 2014 have been more responsive to the deterioration of labour-market conditions and substantially stagnated in nominal terms. Between January and October 2014, the average monthly gross earnings per person in paid employment amounted to HRK 7 934 (approximately EUR 1 044), which represented a nominal increase of 0.1% and a real increase of 0.3% as compared with the same period in 2013. With a growth rate of 0.2% year-on-year, price dynamics continued to be extremely subdued in 2014 according to the Harmonised Index of Consumer Prices (HICP), whereas the national consumer price index (CPI) currently used by Croatian authorities is already in negative territory. HICP inflation is expected to be -0.3% in 2015 on the back of falling energy and commodity prices, whereas a slight up-take is expected with the gradual recovery of oil prices and with economic growth on firmer positive ground in 2016.

Graph 1.4:     Wage and price growth

Source: Central Bureau of Statistics

While household debt broadly stabilised in the third quarter of 2014, deleveraging in non-financial corporations seems to have come to a halt. Total household debt broadly stabilised at around 40.4% of GDP. The total amount of outstanding loans to households further decreased in the second half of 2014, despite historically low interest rates, reflecting negative expectations on the labour market as well as tight credit conditions from the banks. The debt-to-GDP ratio (non-consolidated) of non-financial corporations, on the other hand, increased marginally from 88.2% at the end of 2013 to 89.2% at the end of the third quarter of 2014.

Public debt is set to continue on its steep upward trend. The general government deficit for 2014 should reach 5.0% of GDP in 2014 (compared with 5.2% of GDP in 2013). The slippage from the EDP target is attributable to lower revenue collection, partly explained by the weak tax-base dynamics, while expenditure appears to have evolved mostly in line with the authorities’ plans, notwithstanding some additional outlays from flood-related reconstruction costs that materialised in the second half of 2014. In the absence of further measures, the budget deficit is set to deteriorate to 5.5% of GDP in 2015 and remain around the same level in 2016, in spite of the moderate pick-up in growth. The debt profile was strongly affected by the inclusion (under the new accounting rules) of two major SOEs, which in the period 2008-2013 increased the debt-to-GDP ratio by an average 7.8 pps. The general government debt is expected to have increased from 75.7% of GDP in 2013 to 81.4% of GDP in 2014. In 2015 and 2016, the debt-to-GDP ratio is expected to increase to 84.9% and 88.7% of GDP as a result of the underlying deficit trends.

Graph 1.5:     Real estate price

Source: Croatian National Bank

The financial sector has withstood the recession, though the accumulation of impaired assets could constrain the provision of credit to the private sector in the future. By the end of September 2014, the non-performing loans (NPL) ratio reached 17%. For the corporate sector, which has been particularly affected by the crisis, the NPL ratio recently reached 30%, but NPL from the household sector have also started to increase. Although high capital levels and the recent review of the banking sector portfolio have provided reassurance concerning the stability of the financial sector, administrative measures governing the setting of interest rates and the freezing of the exchange rate of the kuna against the Swiss franc in loan contracts in response to the sudden appreciation of the franc may also impact the banks’ ability to lend to some categories of borrowers and reduce their profitability.

The correction in the housing market has been sharp, but recent indicators suggest that the real-estate prices are stabilising. The real-estate price index is now more or less back to the value at the end of 2004, thus fully offsetting the sharp increase registered before the crisis. Following a 0.6% increase in Q2-2014, the real estate price index grew further by 4.7% in the third quarter, on the back of a strong rebound of the real-estate prices along the Adriatic coast (12.9% quarter-on-quarter), whereas the Zagreb index showed a more moderate progression (1.1%). In year‑on‑year terms, the index registered its first increase since the offset of the crisis (2.5%).

Overall, recent outlook indicators present a mixed picture. The process of internal devaluation is bearing some fruits, specifically in terms of external rebalancing. The latter, however, relied more on a compression of imports than an expansion of exports. Moreover, strong headwinds stemming from the deleveraging needs in parts of the private sector and in the public sector (including publicly owned enterprises) and tight credit conditions still hamper the recovery of the internal demand. Subdued price dynamics hinder wage adjustment and increase the burden of real debt. More fundamentally, the Croatian economy does not seem to have engaged in the significant process of capital and labour reallocation required to unwind internal and external imbalances and return to growth.

 Box 1.1: Economic surveillance process The Commission’s Annual Growth Survey, adopted in November 2014, started the 2015 European Semester, proposing that the EU pursue an integrated approach to economic policy built around three main pillars: boosting investment, accelerating structural reforms and pursuing responsible growth-friendly fiscal consolidation. The Annual Growth Survey also presented the process of streamlining the European Semester to increase the effectiveness of economic policy coordination at the EU level through greater accountability and by encouraging greater ownership by all actors. In line with streamlining efforts this Country Report includes an In-Depth Review — as per Article 5 of Regulation no. 1176/2011 — to determine whether macroeconomic imbalances still exist, as announced in the Commission’s Alert Mechanism Report published on November 2014. Based on the 2014 IDR for Croatia published in March 2014, the Commission concluded that Croatia was experiencing macroeconomic imbalances monitoring and policy action, in particular, developments in the areas of household debt, linked to the high levels of mortgage debt and structural characteristics of the housing market, as well as unfavourable developments in export market shares. This Country Report includes an assessment of progress towards the implementation of the 2014 Country-Specific Recommendations adopted by the Council in July 2014. The Country-Specific Recommendations for Croatia concerned public finances, retirement and disability, labour market and education, the tax and benefits system, business environment and public administration, state owned enterprises, corporate pre-bankruptcy procedures and financial sector stability.

Table 1.1:       Key economic, financial and social indicators

(1) Domestic banking groups and stand-alone banks. (2) Domestic banking groups and stand-alone banks, foreign-controlled (EU and non-EU) subsidiaries and branches. (3) Real effective exchange rate (*) Indicates BPM5 and/or ESA95 Source: ECB and European Commission Source:

Table 1.2:       MIP scoreboard

(1) Figures highlighted are those falling outside the threshold established in the European Commission's Alert Mechanism Report. For REER and ULC, the first threshold applies to euro area Member States. (2) Figures in italics are calculated according to the old standards (ESA95/BPM5). (3) Export market share data: total world exports are based on the fifth edition of the Balance of Payments Manual (BPM5). Source: European Commission

Croatia’s long-term output growth is relatively modest when compared with the growth rates of more successful transition economies. By 2013, Croatia had erased around a quarter of the GDP per capita gap to the weighted EU-15 average, broadly keeping pace for instance with the Czech Republic and Hungary. However, its convergence performance has been overshadowed by the more successful peer transition economies such as Slovakia, Poland and the Baltic States.

Graph 2.1.1:   Growth accounting, 2002-2008 average (%)

Source: European Commission

Unlike most new Member States, Croatia’s GDP growth was driven essentially by employment growth, with limited productivity gains. Real GDP growth can be broken down into productivity growth and employment growth — the latter being the outcome of demographic dynamics and labour market performance. While a bigger employed population can sustain growth in the short-to-medium term, productivity is the only source of growth in the long run. Against this background, it is worrying to note that Croatia’s productivity growth had already been relatively weak before the crisis (see Figure 2.1.1). It indicates that the common factors that shaped economic transition in the new Member States have been put in Croatia to less productive use than in some other central and eastern European economies.

The weak growth fundamentals surfaced fully during the protracted recession. While the low productivity growth was masked and partly offset inter alia by sizeable capital inflows and a buoyant external environment in the pre-crisis period, the weak foundations of growth manifested themselves fully once these positive demand factors receded. In view of the persistently weak productivity growth, the exceptionally long recession ([2]) is re-interpreted as the product of deep-rooted inefficiencies that depress potential growth and delay recovery in demand, rather than being an exclusively demand-driven phenomenon. The goal of this section is to understand the nature of these inefficiencies, with a particular emphasis on those created by the labour market institutions and the business environment.

Graph 2.1.2:   Growth accounting, 2008-2013 average (%)

Source: European Commission

Adjustment dynamics

The long recession has not delivered the requisite restructuring of the economy. Graph 2.1.3 tracks developments in labour productivity and employment in the post-crisis period in selected EU-10 countries. It shows that most have registered an increase in labour productivity along with a decrease in employment. This appears not to be the case for Croatia. Despite the depth of the recession and a loss of more than 200 000 jobs (13 % of overall employment), Croatia has not yet profited from the opportunity to restructure for higher productivity. Sluggish productivity growth is evident also in firm-level data which, moreover, confirm that most economically significant activities, including manufacturing, trade and construction, posted low total factor productivity (TFP) growth already in pre-crisis years.

Graph 2.1.3:   Productivity and employment dynamics (2008-14)

Source: European Commission

The poor adjustment dynamics are a result of economy-wide weaknesses rather than sector-specific shocks. Despite the massive layoffs, restructuring has yet to result in value-added and productivity gains in most sectors. The only sectors where value added in 2013 exceeded the 2008 level were public administration (by 3.5 %), real-estate activities (by 2.3 %) and financial services (by 1.5 %). In other sectors, aggregate value-added growth was persistently negative or too weak to reach the 2008 level. A breakdown of value-added growth into labour productivity growth and employment growth reveals that, of the market-oriented sectors, only the ICT sector achieved a higher level of productivity than in 2008, while others were still at (manufacturing, agriculture) or far below (construction, business services) the pre-crisis level.

Graph 2.1.4:   Distribution of firms according to productivity and employment dynamics

Source: European Commission, ORBIS

At micro-level, the low dynamics in firms can be traced back to delayed restructuring. An analysis based on the ORBIS database of firm-level data shows that in 2007, the year before the onset of the crisis, over half (54 %) of all companies were in ‘productivity enhancing’ mode, expanding productivity and employment. Around 20 % were ‘successfully restructuring’, i.e. cutting jobs with a positive impact on productivity, while only 8 % were ‘restructuring’, i.e. cutting jobs without succeeding in increasing productivity. Finally, around 18 % of firms were undermining productivity, i.e. expanding employment despite decreasing productivity. In 2009, the proportion of ‘productivity enhancing’ companies collapsed and it has continued to fall since, which illustrates the depth of the recession. The majority of companies turned to down-sizing, yet it is apparent that most of them took a wait-and-see approach rather than starting to reshape themselves for higher productivity. The proportion of ‘successfully restructuring’ businesses started to consolidate only in 2013. Sustaining this trend is a necessary precondition for erasing the accumulated efficiency losses and achieving higher aggregate growth and convergence going forward.

At the extensive margin, the contribution to growth from the company entries and exits is surprisingly negative. A study on the relationship between firm dynamism (net entry) and productivity in Croatia finds that business churn was relatively low in Croatia in the 2008-12 as compared with its regional peers. In particular, on average only 5.5 % of companies were new to the market every year, against 9-18 % newcomers in the EU-10 ([3]). Also, the exit dynamics were found to be muted (6.5 % of firms exited every year, as compared with 7-26 % in the EU-10). While typically the entry of new firms and the exit of old ones enhances productivity, the market selection process in Croatia appears inefficient, as the net productivity dynamics contribution of net entry was found to be negative. While, as expected, new entrants helped to enhance productivity, firm exits actually had the opposite effect. This is a sign of inefficiency in the market selection process, which is in line with the evidence of inefficient pre-insolvency and insolvency frameworks presented in section 2.4.

The adjustment was particularly slow in small- and medium-sized companies. Only limited evidence can be gathered as regards the performance of companies of different sizes over time. Yet, data for the manufacturing and transport sectors show that in 2008-11 the productivity growth of the smallest companies took the longest-lasting hit, while larger companies found it easier to initiate the necessary restructuring. Access to credit is likely to be a factor in productivity patterns differing over time. However, on the policy front, this evidence also underlies the importance of facilitating the doing of business across the board so that small and medium-sized companies can benefit, as opposed to taking selective measures targeting large-scale companies and projects.

All in all, low productivity growth appears to be a structural feature of the Croatian economy. The reallocation of resources towards their most productive use seems to be hampered both within firms (as evidenced by delays in restructuring), and between viable and unviable companies as suggested by the apparent inefficiencies in the market selection process.

Labour market

Labour force in Croatia is negatively affected by the rapidly ageing population coupled with low activity rates of older persons. The natural population change was negative between 2000 and 2013, with an average decline of 10 000 citizens a year. According to the European Commission 2015 Ageing Report, demographic projections involve a further drop of almost 7 % of the Croatian population by 2045 (from 4.2 million in 2014 to 4.0 million) and almost 14 % by 2060 (down to 3.7 million). The negative effects of demographic changes on the labour force are likely to emerge already in the medium term – with a decrease of 50 000 by 2020. At present, activity rates are particularly low for workers above 50 (52 % in Croatia, 64 % in the EU-28 in 2013), which is partly a result of the presence of numerous early retirement options. While disability and special pension schemes (mainly for war veterans) were used as the main avenue to early retirement in the previous decade, the generous standard early retirement scheme has been undermining labour supply in recent years. Current projections incorporating inter alia changes in the pension system forecast an increase in the participation rate of older workers from currently 43.2 % to 44.5 % in 2020, progressively reaching 50.9 % by 2060. See section 3.1 for a detailed analysis of the pension system.

Substantial migration flows in the pre-crisis years broadly offset the decline in the overall population. In the first decade of the 2000s sustained growth in low-skilled labour-intensive sectors ushered in sizeable immigration flows, mainly from Bosnia-Herzegovina and other countries in the region, which broadly compensated for the shrinkage in the natural population.

Graph 2.1.5:   Migration flows and total population

Source: European Commission

Worsening labour market conditions after 2008 triggered a reversal of migration flows and resulted in a substantial decline in labour market participation, particularly by young people, women, the long-term unemployed and older workers. In 2009, net migration flows turned negative, contributing to an acceleration of the demographic trend. Also, the low labour market participation of young people contributed strongly to the decline in activity rates. Low labour demand combined with a discouragement effect influenced young workers either to withdraw or not to enter the labour market. Some prolonged their stay in the education system, whereas others contributed to the growing number of not in education, employment or training (NEET rate)  (up to 22.3 % in 2013, EU average 15.9 %). Similarly, the inability to find employment led older workers either to become inactive or to take early retirement. Deterioration of the overall labour market situation, accentuated the employment gender gap and long-term unemployment, particularly for the most vulnerable groups on the labour market.

An inefficiently targeted and multi-layered social benefits system creates disincentives to work and contributes to the high inactivity rate. There are currently over 80 different social benefits and programmes in Croatia, partly administered by eight ministries and partly by county and municipal authorities and provided through a complex network of more than 500 offices operating under different jurisdictions. The lack of a clear overview and adequate targeting of social protection benefits, coupled with their ineffective and untransparent allocation, often results in inactivity traps. These stem from discrepancies in the provision of benefits at the local level with those at the central level. The poor design of the existing benefit scheme also results in inadequate income support. This is reflected in the comparably high proportion (37 %) of the unemployed and poor not covered by social safety nets. A more detailed discussion of social policy follows in section 3.1.

Consolidation of social benefits is progressing, aiming at better targeting, rationalisation and effectiveness. The process is taking place at two levels: consolidation of the guaranteed minimum benefit (GMB) into a guaranteed minimum standard (GMS) and the establishing of a ‘one‑stop shop’ (OSS) — a single payment centre using single national database systems. The GMB was introduced in 2013, and merged four existing benefits with stricter means-testing, introduced mandatory activation measures offering incentives to work, and a social benefits ceiling based on minimum gross salary per household. The impact of the introduction of the GMB on activity rates is as yet unclear, but it could be a positive step towards reducing inactivity traps and intensifying work incentives, while achieving greater efficiency in social spending.

Further efforts to reform the benefit system and reduce inactivity traps are expected in 2015. The mapping of social protection benefits to identify which could be merged under the GMS was completed by the end of 2014, while work on the ‘categorisation’ of benefits should be completed by spring 2015. A consolidation action plan for the overhaul of the benefits system, due for adoption by early summer 2015, will include proposed alternative models, information about political governance, and a legislative framework up to 2020, including plans for decentralisation. The planned consolidation of social protection benefits is ambitious and goes in the right direction. Once the OSS and its detailed tracking and monitoring system are in place, this should limit the accumulation of benefits and ensure that they are better targeted. Following the compilation of the review and assessment of reform options it is paramount that authorities move to the implementation phase.

The situation on the labour market remains challenging on the back of the enduring recession and the limited adjustment capacity of the economy. The sharp fall in internal and external demand goes a long way to explaining the significant surge in unemployment. Fiscal consolidation needs in the general government sector – which includes several public companies – also play a significant role, given the high proportion of workers employed by the broad public sector. Nevertheless, the unemployment rate had also been consistently high in the pre-crisis period, signalling structural weaknesses in the functioning of the labour market. Despite recent timid signs of improvement, these are set to weigh on future employment growth.

In recent years, wages have seen only moderate growth in Croatia, but, gross earnings remain relatively high, considering productivity. Gross real wages (deflated by GDP deflator) increased by less than 8 % between 2002 and 2014, as compared with almost 30 % in the EU-10. Despite slow productivity growth, this resulted in a moderate decrease of 5 % in real unit labour costs (ULCs). The adjustment in real ULCs was sharper in the EU-10, however, where it averaged 10 % over the same period. Considering current productivity levels, gross earnings in Croatia are still relatively high (especially since aggregate productivity is likely to be overestimated given the public sector’s higher wages and large share in the economy).

The public sector and public enterprises pay higher wages than the private sector and offer a range of benefits that distort the reservation wage and encourage early exits from the labour force. Analysis based on Labour Force Survey data shows that the average net hourly wage in 2012 was 25 % and 18 % higher in the public sector and in public enterprises, respectively, than in the private sector. Controlling for characteristics of employees and jobs, the wage premium was about 5 % in the public sector and 7 % in publicly owned enterprises ([4]). In addition, according to information provided by the authorities, some SOEs offer excessively generous retirement bonuses (the equivalent of a year’s salary) and other bonuses. Such bonuses, if not well designed, may not only negatively affect retirement decisions but also drive up the reservation wage in the entire economy.

Limited wage adjustment in the post-crisis period exacerbated the negative impacts on employment. Real wages adjusted only slowly to the changing macroeconomic conditions. As unemployment started to increase in 2009, real wages were still rising and (albeit moderately) the trend has continued since. The relatively high degree of unionisation in the labour force, combined with an uncoordinated and decentralised wage bargaining system, contributes to the limited responsiveness of wages to changes in the macroeconomic environment. An analysis of wage dynamics in Croatia in the period 2000-13 concludes that wage setting institutions are prone to generate spillover effects from the tradable to the non-tradable sector. In particular, weak competition in the internal market generates wage pressures which spill over to companies subject to competition from foreign firms, thus eroding their cost competitiveness. The authors suggest that authorities could consider further decentralisation of wage bargaining accompanied by stronger enforcement mechanisms, or a step towards a more centralised system characterised by enhanced coordination ([5]). A third possibility – among the available models used in European Member States – would be to combine those two alternatives so as to provide for enhanced coordination with enough room for actual decentralisation: the former to deal better with economy-wide shocks, the latter to allow room for adjustment to more local or firm-specific shocks. Either model, if successfully implemented, could allow achieving a better alignment of wages and productivity developments in all the sectors of the economy.

Graph 2.1.6:   Productivity vis-à-vis EU-28 and average earnings in Croatia, EU and other OECD countries (2014)

Source: European Commission, based on 2015 winter forecast

The review of the wage-setting system resulted from the need to align wage dynamics with productivity developments and macroeconomic conditions. In cooperation with external experts, the authorities have completed a comprehensive analysis of wage determination and wage-setting practices in both the private and the public sectors and in SOEs, with the intention to inform a tripartite discussion with social partners. The review suggests that there is scope to foster wage dynamics better aligned with productivity developments and more attuned to macroeconomic and social conditions, including the way wage-setting dampens or cushion changes in economic activity. In this context, the authors of the review call for more coordination in the wage-bargaining system. In early 2015 the authorities have initiated a debate on possible measures of either legislative or administrative nature. An inter-ministerial consultation on establishing centralised coordination of collective bargaining in the public sector and SOEs is under way. The reform of the current institutional setup – especially with regards to the existing collective agreements within SOEs – may prove challenging. Nevertheless policy action is warranted, following the wage review.

To address institutional deficiencies of the labour market, Croatia implemented two labour market reforms in 2013 and 2014. Before these reforms, the Croatian labour market was characterised by relatively strict employment protection legislation. While it is still too early to assess the impact of the legislative changes, initial signs indicate that they go in the right direction. The labour law reform also resulted in Croatia’s employment protection legislation (EPL) index (as per the OECD methodology) falling from 2.6 to 2.3 after the first phase and to 2.2 after the second, which compares well with the situation in most other OECD countries.

The 2013 reform tackled the issue of flexibility in the use of temporary contracts. The 2013 amendments to the Labour Act allowed for the use of fixed-term contracts concluded for the first time for a period of over three years. This resulted in increased flexibility as compared with the previous regulation. In order to protect workers’ rights, however, the amendments limit the duration of any successive fixed-term contract to three years. Furthermore, the new legislation lifts restrictions on reasons for concluding the first fixed-term contract, and restrictions now apply only for the termination of successive fixed-term contracts.

The second phase of the labour law reform addressed low contractual flexibility and rigid employment protection legislation. The new Labour Act, adopted in August 2014, increases flexibility in the organisation of working time, facilitates the use of flexible types of work (distance work, part-time work, seasonal work and agency work), and simplifies in various ways the procedures for terminating employment contracts. The reform also took a first step in pointing to the need to facilitate the renegotiation of outdated collective agreements. A thorough re-examination of the rules on re-negotiation, administrative extensions of sectoral agreements to non-signatory parties, enforcement-related issues (especially in the private sector) and overlapping agreements are all areas in which there is scope for further policy action. A working group for monitoring the implementation and impacts of the labour market reform has been established with the social partners, with the first regular report envisaged for January 2015.

Graph 2.1.7:   Share of temporary and recently employed (less than three months) workers

Source: European Commission

Increases in temporary employment signal enhanced flexibility on the labour market but could bear a risk of growing segmentation. Croatia had been performing close to the EU average in temporary employment (14.5 % of total employment in 2013), but recent trends show rapid increases in the use of temporary contracts (in 2013 and 2014, 93 % of all newly employed persons were employed on a temporary basis). Temporary employment is highest for young people and has been steadily increasing to 45.1 % in Q3-2014. According to a survey of employers, 85.1 % of all new employment in 2014 was planned to be temporary, half of it (49.9 %) seasonal. In this context, the sensible increase in mobility registered following the reform (measured as the share of employees having started a job in the last three months) is most likely reflecting a higher turnover of temporary workers. One of the challenges ahead is to maintain relatively high levels of labour mobility while preventing labour market segmentation, which is detrimental to productivity growth.

Despite the recent reforms, current labour market regulations still place a number of barriers to hiring and job-to-job mobility. The labour market is still characterised by hiring constraints, including a temporary ban on hiring for firms that have recently dismissed workers. This protection measure could constitute a bottleneck to labour mobility, considering the on-going pressures for economic restructuring (see previous section). Also, the severance payment arrangements, characterised by non-portable entitlements, have been criticised for hindering labour mobility. Various reform proposals have recently been submitted to the government, including a switch towards the ʻAustrian model’ ([6]). Lastly, the current level of minimum wage might create barriers to employment of low skilled and younger workers. In 2014, the minimum wage – set by the government at HRK 3 017.61 – is estimated to have attained 38 % of average wage, and a further increase to HRK 3 029.55 was implemented by decree on 1 January 2015. ([7]) It should be noted that in PPS terms the Croatian minimum wage is higher than in the EU-10, with the exception of Poland and Slovenia. Moreover, based on the 2010 Structure of Earnings Survey, at 9.2 % Croatia is one of the countries with the highest share of employees earning less than 105 percent of the monthly minimum wage. A different modulation of the minimum wage for vulnerable categories of workers and the removal of nominal rigidities in its adjustment could reconcile the needs of workers protection with enhanced opportunities for all workers and more flexibility in wage dynamics.

Graph 2.1.8:   Average monthly earnings by educational attainment (2010)

Source: European Commission, Structure of earnings survey, 2010

Relatively high returns to education are coupled with high unemployment rate of the tertiary educated. Gross wage premia associated with tertiary education are especially high, with the employees in question earning on average 50 % more than workers with secondary education.([8]) The weak absorption capacity of the domestic production system is witnessed by the relatively high unemployment rate among graduates of tertiary education. Issues related to the quality and relevance of education and training are addressed in section 3.4.

Graph 2.1.9:   Unemployment rate by education level (2013)

Source: European Commission

The size of the informal economy remains a persisting challenge in Croatia and results in a high incidence of undeclared work and ʻenvelope wages’. Steps have been taken to combat undeclared work in Croatia. In January 2014, the single tax, surtax and contributions form (JOPPD) was introduced, which incorporates all data on income from employment, income tax, surtax and contributions for compulsory insurance, data on other income (contracts, royalties, income in kind) and corresponding taxes and contributions, as well as income from property rights, capital or insurance. In the framework of labour law reform, it is now possible to work part-time with a second employer alongside a full-time job and employees must now be registered before they start work. In August 2014, a ʻcommission to combat undeclared work’ was established as an expert advisory body in charge of assessing measures taken in the previous two years to combat undeclared work, evaluating their implementation, monitoring future measures and  (if necessary) proposing amendments to existing measures. In September 2014, the commission issued a consolidated report on measures taken to combat undeclared work, which provides an overview of action taken by the relevant bodies. An analysis of the underlying causes of undeclared work was completed by the end of 2014. It covered all measures that have been implemented and identified where there is room for improvement.

Labour market institutions have been broadly reformed over the past two years, but challenges remain. The two phases of labour-market reform aimed at increasing labour market flexibility and as soon as positive growth returns, visible improvements in the performance of the labour market should materialise. Nevertheless, the impact of the recent changes are yet to work through. Moreover, remaining shortcomings continue to slow down employment creation, prevent labour mobility, fuel labour market segmentation and enable for wage dynamics which are responding insufficiently to existing macroeconomic and social conditions.

The business environment

Croatia’s business environment suffers from major institutional shortcomings. The seriousness of the challenges investors and entrepreneurs are facing is borne out by a number of international reviews. The World Bank’s 2015 Doing Business report ranks Croatia 65th out of 189 countries globally and 27th in the EU-28. Compared with 2014, Croatia improved its global ranking by two places but its position in the EU-28 context stayed the same. The 2014-15 Global Competitiveness Report of the World Economic Forum ranks Croatia 87th as regards institutional quality overall and 141st (out of 144) in terms of the burden of government regulation. The value of the institutional quality index remained unchanged as compared with the 2013-14 report but the rank improved by six places. Croatia is the worst performing EU Member State in the OECD product market regulation Index, which is largely due to a high degree of public ownership and government involvement in network sectors and in general business operations.

A number of specific issues give rise to this assessment, including: regulatory instability and a lack of strong ex ante legislative assessment; high compliance costs, discriminatory practices and rigid regulation in services; high administrative burdens; the plethora of parafiscal charges, low transparency and predictability in the working of administrative bodies (in particular at local level) and unevenly developed electronic communication channels; and long judicial proceedings. These issues are discussed below in detail. Related issues pertaining to the functioning of public administration, effectiveness of public spending, corruption and the governance of public enterprises are discussed in section 2.3. The functioning of the pre-insolvency and insolvency frameworks (relevant for the resolution of the debt overhang) is analysed in section 2.4.

Weaknesses in legislative planning and in the application of impact assessments weigh on the quality of new regulations. Half of the laws adopted in 2014 were not anticipated in the legislative plan. The extensive unplanned legislative activity squeezes the scope for ex ante regulatory impact assessment (RIA), with negative impact on the quality of the assessment. Earmarking of acts for RIA appears to reflect their socio-economic significance. In practice, however, the RIA is frequently circumvented by resorting to fast‑track legislative procedure, under which it is not mandatory. RIA was actually carried on only nine out of 45 legislative acts earmarked for it in 2014. Moreover, when used, RIA appears to be rather perfunctory due to weak quality control, inadequate capacities and a lack of high-level commitment to impact assessment. Inadequate assessment of legislative acts weighs on the quality of new regulations and leads to frequent amendments to address unintended (but predictable) consequences of adopted legislation. Regulatory instability, in turn, complicates business decisions, in particular for SMEs and young companies, which lack the resources to deal with it. The authorities are planning to introduce the ‘two out, one in’ check for new regulations but no specific measures are planned to improve the quality of impact assessment overall.

 Box 2.1.1: Doing business in Croatia The table below presents the development over time of the most relevant indicators covered by the World Bank’s Doing Business Report. Apart from reporting on the individual numerical scores (see the report for interpretation of the values), it shows Croatia’s ranking vis-à-vis the other Member States. Dark red indicates under-performance, yellow average and green better-than-average performance. The 2015 Doing Business Report substantially altered the assessment of Croatia following revision of data and refinement of some of the underlying indicators. The series was updated retrospectively. In the revised assessment, Croatia’s 2014 rank is 67th, as opposed to 89th originally. Two most important changes concerned: (a) protecting minority investors: the 2014 ranking improved from 156th to 62nd as a result of the indicator being extended to include factors favourable to Croatia (the role of shareholders in corporate decision–making, safeguards against undue board control, and corporate transparency); (b) resolving insolvency: the 2014 ranking improved from 98th to 56th, again due to a number of new sub-indicators such as the involvement of creditors in reorganisation proceedings and the protection for dissenting creditors. Even with the improvement on the basis of the revised data, Croatia’s rank in terms of ‘ease of doing business’ is still the second lowest in the EU-28 (after Malta).

Sector-specific regulations pose excessive burdens on entrepreneurs in services. For instance, laws applicable to specific services require prior authorisation from a competent authority in Croatia, including entry in a register or registration with a professional body or association. Some laws, for instance in the tourism sector and construction sectors, also set rules for minimum number of employees or impose specific legal form requirements (this applies particularly to educational institutions and legal professions). Moreover, regulated professions such as the legal profession, accounting, architecture and engineering are strictly regulated by the legislator and by professional bodies. Both quantitative and qualitative restrictions on the number of entrants into the profession (e.g. educational requirements, quotas) apply. Conduct regulations further determine tariffs and the organisational structure of businesses providing professional services, while restricting advertising and cooperation between professionals. The OECD’s product market regulation index for professional services ranks Croatia relatively unfavourably as regards entry restrictions in all regulated professions it covers and among the three worst Member States in terms of conduct restrictions in all professions but accounting services.

Another problem is that discriminatory practices against foreigners limit competition on the domestic market. Practices that have been reported include additional taxes on non-residents or a direct requirement of residence in Croatia in order to benefit from specific services, e.g. in the maritime field.

The ability of the Competition Authority to foster more intense competition in the Croatian business environment is hampered by a shortage of staff and other resources. The Authority is small compared with its counterparts in other Member States of similar size, as is its budget (approximately EUR 1.7 million in 2015 for a staff of 47, of which 31 work in the Competition Division). These factors significantly restrict its capacity for more comprehensive enforcement and thus the effectiveness of competition policy.

Administrative burdens on companies are substantial. According to preliminary values of the OECD’s 2013 product market review indicators, Croatia imposes the highest start-up administrative costs on corporations as well as on limited liability companies. However, companies with no more than three members can be established through a streamlined procedure as a ‘simple limited liability’ company. The World Bank’s 2015 Doing Business report notes that while the one-stop-shop ‘HITRO’ allows for company incorporation, it is not as common as registration through a notary (which takes at least 10-15 days), who inter alia certifies the non-existence of debt claims. The ‘silence is consent’ rule applies only to some licences required to open up a business and the issuance/acceptance of relevant notifications and licenses is not performed via a common access point. In addition to the start-up costs, the Doing Business report highlights severe bottlenecks in trading across borders (despite an electronic customs system) and securing construction permits (see box 2.1.1). Other pockets of heavy administrative burden complicate the process of registering property changes (length of procedure) and enforcing contracts (procedural complexity).

The authorities have embarked on a process of auditing administrative burdens with a view to alleviating them. In November 2014, the government decided to pilot measurement of administrative burdens using the standard cost model. The project concerns administrative requirements under the Trade Act and the Mediation in Immovable Property Transactions Act. It is carried out by the Agency for Investment and Competitiveness, which reports to the Ministry of Economy, in close co-operation with experts from the Netherlands. The pilot phase is expected to be completed by the end of February 2015, following which an action plan for the simplification of administrative burdens will be presented for the areas in question.

Progress on removing or decreasing parafiscal charges is considerably slower than expected. At the beginning of 2014, the register of parafiscal charges contained 244 items, generating revenue of HRK 6.4 billion (1.9 % of GDP). The 2014 Implementation Plan signalled the removal or reduction of 58 charges in 2014, with an expected decrease in revenues by 0.16 % of GDP. However, this happened in only 15 cases, decreasing the parafiscal burden by 0.03 % of GDP. In total, 63 levies were written off the registry as it turned out that they were not ‘parafiscal’ but other payments, such as concession fees, and additional 12 minor levies were added to the register. The fact that the authorities lack a solid overview of charges imposed on businesses undermines the legitimacy of the payments and makes it difficult to assess the ambition of the authorities’ efforts. Big fluctuations in the registry of charges are due to the fact that the coordinating entity (the Ministry of Finance) checks the classification only after a payment is entered into the registry. It is important to note that the removal of charges is not centrally coordinated, which means that there is insufficient guarantee that the charges being removed are those that are most problematic for businesses.

The burden parafiscal charges impose on businesses is not only financial. Administering frequent, although relatively small, payments to numerous entities creates an additional administrative burden for businesses. The authorities plan to reduce the administrative burden associated with paying the charges by reducing the number of individual payments, but they have yet to present a specific plan. Joint accounts shared by several institutions are being used for some sectoral charges. In general, though, the fragmented organisation of collection, verification of compliance and enforcement by the recipients (public administration, SOEs and agencies) take up significant resources in return for limited revenue.

The public administration in Croatia is more costly than that in many EU Member States, but does not perform well in terms of effectiveness. Despite a public wage bill close to 12 % of GDP, the EU Public Administration Scoreboard (Member States’ Competitiveness Report) shows that Croatia scores low on the EU Public Administration Scoreboard (part of the Member States’ Competitiveness Report) particularly on the indicators of effective implementation ([9]), cost and time to export, and irregular payments and bribes. Inconsistencies in the decision‑making of regional and local units across the territory, and the lack of a clear timeframe for issuing binding opinions in tax issues are a source of concern for businesses ([10]). High turnover rates, limited training opportunities, the lack of transparency in staff recruitment and the absence of evidence-based approaches to reform, remain weak points in the public administration. Proposed amendments to the Public Servants Act (from mid‑2014) were aimed at simplifying recruitment and dismissal procedures, and redefining decision‑making in the civil service with a view to improving efficiency. However, in the light of observations from a number of institutions in public consultation, the legislative procedure has been put on hold and a new proposal is under preparation.

Modern communication channels with the public administration are unevenly developed. To tackle the very low online interaction between the public administration and citizens (ranked 25th in the EU-28), the e-citizens web portal was launched in June 2014. The system comprises a central state web portal, a personalised electronic mailbox and a national identification and authentication system. The portal is developed and expanded on an ongoing basis, with the introduction of 12 more services by the end of 2014 and further upgrades planned for 2015. The Ministry of Interior will be coming on board, so the e‑citizens portal will be used to issue a wider range of e‑certificates and for paying of administrative charges. As regards businesses, limited online services, in particular for users from other Member States, are provided by the Croatian Chamber of Commerce, the ‘one‑stop shop’ HITRO and the Financial Agency (FINA). These services are administered by different institutions, which creates challenges both for the authorities (ensuring compatibility) and for users (having to familiarise themselves with different digital interfaces). The authorities plan to develop a new platform, www.poduzetnik.hr, which would integrate all existing services. In the area of physical planning, an advanced information system (ISPU) is in place. This includes a new e‑construction permit application, which allows permits to be issued within 30 days.

Improving the efficiency and quality of the justice system remains a challenge, despite some positive developments. In 2014, the disposition time (a calculated measure of the length of trials) remained comparatively long at first instance both in litigious civil and commercial cases (378 days) and in administrative cases (427 days). At first‑instance civil and commercial courts in litigious cases, the situation has improved slightly as compared with 2013 in terms of both backlog (a reduction by 6 %) and disposition time. These improvements can, however, be partly attributed to a big drop in the number of incoming cases (by 11 %) rather than a higher number of resolved cases. At first‑instance administrative courts, backlog increased by 17 %, despite more cases being resolved. At the High Commercial Court, the situation worsened seriously both in terms of backlog (an increase by 16 %) and disposition time, which reached 1 243 days in 2014, up from 1 142 days in the previous year. A significant drop in new cases (by 30 %) at the High Administrative Court partly helped maintain a positive trend, with disposition time falling markedly from 224 days in 2013 to 68 in 2014. In enforcement cases, disposition time increased slightly.

Implementing the reform of the judicial map could improve the quality and efficiency of specific courts. The reform, affecting municipal, misdemeanour and county courts was adopted in October 2014 and is expected to be implemented from April 2015. Primarily, it aims to increase specialisation and balance out the uneven workload of judges by merging courts. The reform involves a change in the allocation of cases before second‑instance civil courts: cases on appeal will be allocated to all second‑instance courts, with some courts specialising in particular types of case. At first‑instance administrative courts, where lack of staff remains an issue, there are plans to provide judges from other areas with incentives to become administrative judges. At first‑ and second‑instance commercial courts, there is a lack of streamlined business processes (including case management) in the handling of urgent and complex cases, and this has a negative effect on the length of trials. Evaluating the effects of past reforms, particularly the amendments to the Civil Procedure Act, is important in order to correctly identify challenges in the civil justice system.

Alternative dispute resolution (ADR) methods are underused. According to the 2014 ‘Flash Eurobarometer’ survey on consumer attitudes towards cross-border trade and consumer protection, Croatia had the fifth lowest percentage of consumers in the EU who found it easy to resolve disputes with businesses through courts and out-of-court bodies. Efforts have been made to promote ADR methods, but their impact is difficult to predict.

Courts are implementing information and communication technology (ICT). In 2014, the upgraded ‘ICMS’ system was introduced in all first‑instance courts except administrative courts, which use a separate system, and the High Commercial Court and Supreme Court, which are due to be included in 2015. However, the use of ICT systems in communication between courts and parties is underdeveloped, as claims and undisputed debt recovery can still not be processed electronically.

Overall, the business environment stands out as a major reason for the low adjustment capacity of the Croatian economy and a bottleneck for investment and growth going forward. Specific issues contributing to the high cost of doing business include strict regulation in the product and service markets, and high administrative, regulatory and parafiscal burdens. Regulatory instability is aggravated by weaknesses in regulatory impact assessment and legislative planning. Long judicial proceedings erode the efficiency of the legal framework. Despite several concrete measures, only limited progress has been made on improving the quality of the business environment overall.

The competitiveness losses accumulated in the years before the global financial crisis resulted in the build-up of large external imbalances. In the pre-crisis expansionary cycle, the Croatian economy benefited from large-scale partly bank-intermediated foreign capital inflows, though the latter were essentially directed towards the non-tradable sectors, fuelling an unsustainably high level of internal demand. Expectations of rapid income convergence led to increases in wages and prices, which eroded Croatia’s tradable sector and its competitive position. The dynamics leading to the build-up of external imbalances and their ongoing unwinding can be analysed by focussing on real transactions (trade and transfers) or looking at the financing side. In the following sections, the former approach is followed and the determinants of the trade balance and the other balances that produce the net lending position are discussed. Specifically, the analysis will cover the performance of exports of goods and services. The following subsection analyses the drivers of the high import dependence. Finally the analysis turns to the balance of primary and secondary incomes and capital transactions, and vulnerabilities stemming from the high NIIP and external debt.

Graph 2.2.1:   Net lending position – transaction side

* 2014 is an estimate based on quarterly data. Source: European Commission

Exports of goods

The weak performance of goods exports is a long-lasting weakness of the Croatian economy. After a short-lived expansionary phase at the beginning of the last decade, export market shares started contracting from 2004 and losses have intensified since 2008. The contraction in exports in recent years was to some extent a consequence of sector-specific shocks (mainly the restructuring of the important ship-building sector) and temporary trade flow diversions linked to giving up preferential CEFTA trade agreements on accession to the EU (preferences were restored the following year through association agreements). The acceleration of exports in 2014 and expected positive development in 2015 and 2016 signal that Croatia may start to benefit from accession to the EU market. Yet Croatia’s export base is small. Export of goods corresponds to just 21.6% of GDP, as opposed to more than 50% in the export-oriented economies of the EU-10, but also well below the EU (above 30%). Several factors contribute to the low share of exports, including the relatively high wages, substantial non-price competiveness weaknesses and rigid business environment. The above factors – together with delayed EU accession – have held back FDI in the tradable sector (especially in capital intensive sectors) while terms of trade are penalised by currency inflows from the tourism and remittances.

Graph 2.2.2:   Net lending position - financing side

* 2014 is an estimate based on quarterly data. Source: European Commission

Several indicators suggest that Croatia suffered from an erosion of competitiveness in the years before the crisis. Standard ex ante competitiveness indicators, such as the real effective exchange rate (REER) highlight an erosion of competitiveness until 2009, irrespective of the deflator used. From 2010, however, cost competitiveness vis-à-vis trading partners improved visibly, though export price-deflated REER stagnated, suggesting a difficulty in transposing lower unit labour costs (ULCs) into lower export prices and/or higher volumes. Export market shares shrank significantly over the whole period.

Graph 2.2.3:   Real Effective Exchange Rates, ULC- and export prices-deflated

Source: European Commission

Weak export-market performance cannot be unequivocally traced back to worsening competitiveness indicators. This was probably due to the progressive improvement in the quality of products. Croatia was also able to increase its export market shares – though to a much more limited extent. The weaker performance is likely to be related not as much to the dynamics over the period, but rather by the relatively high starting level of wages. Conversely, the post-2008 reduction in REERs in Croatia was associated with a strong deterioration in market shares, whereas developments in regional peer economies were in line with standard theory. This is slightly puzzling, as it suggests that the gradual erosion of competitiveness and its subsequent recovery have limited implications for the dynamics of export‑market shares. In its 2012 country report, the IMF analyses ULCs and export performance at disaggregated level and highlights how (unlike that of its peer economies) Croatia’s export performance was disappointing despite relatively subdued ULC increases — particularly in exporting industries ([11]). In the light of the above, is it doubtful that wage compression per se would be sufficient to deliver significant improvements in export performance.

Graph 2.2.4:   Competitiveness and export performance (2008-2014)

(1) 2014 data based on forecast Source: European Commission

Poor export performance is more likely linked to structural features of the Croatian economy and loss of non-cost competitiveness. Beyond relatively high labour costs (in absolute terms), the concentration of Croatian exports in non-growing product and geographical markets, the limited degree of integration in global value-chains and inefficiencies in product and factor markets hinder the penetration of Croatian firms in foreign markets. These weaknesses translate into a lower demand for Croatian goods. While almost all central and eastern European Member States have an export income elasticity of about 1, the value for Croatia is around 0.6 (see Graph 2.2.5). Differences in price elasticity and in elasticity to trade‑partner proximity are less relevant in statistical terms and are a smaller factor in Croatia’s export gap. Lower income elasticity signifies that Croatia is less able than its competitors to benefit from the proximity of rich markets ([12]).

Graph 2.2.5:   Elasticity of exports to income, price and distance

(1)Histograms represent the absolute value of the elasticities. Elasticities with respect to distance and relative prices are negative. Source: European Commission

The unfavourable initial product specialisation and destination of Croatian exports does not explain the recent weak export performance. In its 2014 in‑depth review, the Commission highlighted the role of disadvantageous specialisation in less dynamic regions and industries. Croatia was penalised by reliance on the depressed Italian, Bosnian and Slovenian markets, limited presence in fast‑growing market segments and a heavy reliance on depressed sectors, including machinery, metals and vessels. However, updated analysis reveals that, if anything, the initial structure contributed positively to export shares in 2012-13, so that the contraction of world export‑market shares in that period was entirely attributable to weak competiveness (see graph 2.2.6).

Graph 2.2.6:   Decomposition of changes in export market shares

Source: European Commission

Apart from a relatively high share of high-value-added goods, the export structure appears biased towards labour- and resource-intensive sectors. The share of high value added exports is slightly higher than in the EU-10. Croatia however features a significantly higher share of low-value-added exports in labour-intensive or raw-material intensive sectors, such as wood and cork, construction materials, leather products and footwear and animal products. These sectors are likely to have been more penalised by the relatively moderate increases in labour costs and increasing trade from developing countries ([13]). As shown in graph 2.2.7, Croatia falls significantly behind its peer economies in trade in capital‑intensive and easily imitable R&D‑based goods ([14]).

In the higher value-added range, Croatian firms also appear to be struggling to scale up the quality of their products. Despite higher labour costs in medium-to-low- and high‑technology-intensive industries, Slovenian firms outperform their Croatian counterparts in terms of product quality. Croatian firms appear to be struggling with quality enhancement and rely excessively on the cost dimension to improve their competitive edge ([15]). This shows that cost moderation can help short- and medium‑term performance but not remedy more lasting weaknesses.

Graph 2.2.7:   SITC structure of goods exports in Croatia and peer economies (percentage of total exports, 2013)

Source: European Commission

The limited integration of Croatian firms in global value‑chains hinders the improvement of non-cost competitiveness and is likely to increase sensitivity to price increases. As a consequence of the fragmentation of previously integrated production process, trade in intermediate goods — i.e. input that has itself been produced and is used up in production – has significantly increased over the past years. This trade is particularly relevant for catching up economies, since the shift from price- towards quality-driven competitiveness is challenging for firms and industries in transition economies due to a lack of relevant skills and knowledge, and obstacles in accessing technology. In most of the countries in the region, imports of intermediate goods and machinery drove the changes in export structure. Local enterprises acquired foreign inputs and know-how in order to improve production quality and expand exports to the EU ([16]). Outsourcing and offshoring is likely to have been used to a more limited extent in the case of Croatia. Consequently the volume of trade in intermediate and capital goods (which includes parts and accessories of equipment) is significantly less developed than in peer economies (see Graph 2.2.8). The limited internationalisation of Croatian firms is also evident by the low import content of exports. As shown in Graph 2.2.9 this was only 22.5% compared with an average of 37.4% for the central and eastern European new Member States, or about half the share of Slovakia – the regional peer with the highest import content of exports) ([17]).

Graph 2.2.8:   BEC structure of exports of goods in Croatia and peer economies (percentage of total exports, 2013)

Source: European Commission

The weak internationalisation of Croatian firms is also explained by FDI, which, although sizeable, bypassed export-oriented sectors. FDI has contributed significantly to export restructuring in central and eastern Europe, but the impact has not been even. The EU-10 succeeded in increasing exports predominantly in higher-end technology industries, while Croatia and other non-EU counties specialised in exporting the product of lower-end technology industries. The available evidence suggests that early EU accession played a significant role in shaping the scale and the nature of the FDI ([18]). As already documented in the 2014 in-depth review, FDI in Croatia largely bypassed the tradable sector.

Graph 2.2.9:   Import content of exports (2005)

Source: European Commission

In the absence of a clear strategy, FDI was not systematically directed towards sectors with high growth potential. It was essentially linked to the privatisation process, with limited greenfield investments ([19]). An FDI strategy to accompany the ambitious industrial strategy currently being elaborated by the government and further privatisation plans could ensure that FDI increasingly targeted exporting industries. Greenfield investments should be prioritised however given the limited number of SOEs operating in the tradable sector. Measures tackling specific bottlenecks in product and factor market will be key to making Croatia more attractive for FDI. Equally important is the development of a communication strategy on its comparative advantages in this respect, due to its strategic location potentially linking central and eastern European countries to the unstable, though potentially high-growth economies in Europe’s southern and eastern neighbourhood.

Inefficiencies in the national innovation system undermine firms’ innovation capacity and efforts to enhance the quality of domestic production. As will be discussed in section 3.4, Croatia is currently stuck in a low‑innovation growth model. The national innovation system is characterised by its closed nature, sub-scale investment, excessive fragmentation and badly defined policies and supporting instruments. Whereas endogenous innovation capacity appears at the current juncture to be the binding constraint for Croatia, exporting industries would benefit from a more supportive innovation system that would enhance technology absorption capacity and non-R&D-based innovation. 

Structural bottlenecks in product and factor markets affect firms’ capacity to penetrate foreign markets. Recent literature has highlighted that aggregate export performance can be better explained by firms’ microeconomic behaviour, rather than by aggregate indexes of competitiveness. Croatian exporting firms are on average more productive, have higher sales and have more capital-intensive processes. These positive characteristics improve further following a breakthrough on export markets ([20]). Ultimately, a significant bottleneck to exports from Croatia is likely to be represented by the small number of firms with productivity levels above the export threshold. The very same factors that are limiting allocative efficiency – including the restrictive business environment – therefore also contribute to the weak performance of exports. A particular role in this context is played by regulations hindering the internationalisation of business. A recent European Commission review has highlighted how, in comparative perspective, Croatia scores particularly badly on business indicators of the cost, time and administrative requirements for import and export (see Graph 2.2.10).

Graph 2.2.10: Internationalisation relevant indicators – distance from EU average (std. dev.)

Source: European Commission (2013): ’2014 SBA Fact Sheet – Croatia

Exports of services

Croatia outperforms most of its regional peers in the export of services, thanks mainly to its all-important tourism sector. Croatia has been able to preserve its environment - 47% of its land and 39% of its maritime territory is designated as specially protected areas and areas of conservation and it hosts almost 20 national parks. Its natural beauty draws in millions of tourists each year, with tourism revenues representing around 15% of the country’s GDP. Only the two Mediterranean island-states of Malta and Cyprus have comparable tourism sectors in terms of contribution to GDP. Analysis of revealed comparative advantages shows an improvement in the performance of the sector, though the two high-potential sectors of transportations and other services (which mostly include services to enterprises) have deteriorated in recent years. Revenues from tourism have performed well in the past decade and, despite weaker performance in previous years, 2014 was another excellent year.

Graph 2.2.11: Export of services

Source: European Commission

The Croatian tourism sector is struggling to evolve from a model of mass tourism to a model of tourism based on higher quality. Despite a significant increase in the number of nights spent in tourist accommodations, the increase in tourism revenues has been relatively modest in recent years, suggesting that increases in volumes are being achieved by lowering the average price of the services ([21]).The structural weaknesses of the tourism industry include outdated products, poor traffic infrastructure, old−fashioned forms of accommodation and overall tourist infrastructure, low‑quality accommodation and other services, high seasonality and the absence of branding. Scaling-up the quality of services will involve significant structural changes, but also a change in the target group, moving away from large group-based, seasonal, price-conscious, passive tourists to value-conscious smaller groups and families seeking higher quality.

Graph 2.2.12: Revenue from tourism and number of nights in tourism establishments

(1) The size of the circle represents the share of the tourism sector as a percentage of GDP Source: European Commission

A large tourism sector is likely to generate negative spillover effects on other industries, and risks crowding-out the tradable sector. Broz and Dubravčić suggest that Croatia might suffer from industrial decline due to its oversized tourism sector ([22]).

The Croatian authorities show awareness of the need to scale up the quality of tourist services, but have so far showed limited active involvement. Because of the specificity of the tourist sector, public authorities play a crucial role that ranges from promoting and preserving the attractiveness of the country and/or specific regions, to developing infrastructure and structuring the various private-sector service providers. Cooperation and partnership between many organisations and interest groups (public, private and non-profit) is the basis of destination management. Environmental sustainability has been central to Croatian strategic planning on tourism for over two decades, but little has been done to move away from a low value‑added equilibrium. There are few examples of local government participating with other sector stakeholders in the development of the industry Going forward the key challenge is to provide the sector with market (domestic competition, internationalisation) and non-market (regulation, public procurement) stimuli and framework conditions to facilitate diversification and avoid a lock-in into low-value-added and adverse terms of trade over the long term. The Travel and Tourism Competitiveness report also notes that in order to improve the sector’s competitiveness, there is room to upgrade its ground railroad transport infrastructure, its ports, and its air transport infrastructure. More efforts should also be devoted to ensuring that policy rules and regulations are support the development of the sector with improvements in areas such as FDI, property rights protection and red tape. A high-value-added tourism sector could to increase the multiplier effect – so that the impact of tourism spreads to the whole economy – thus overcoming the negative externalities referred to above.

Graph 2.2.13: BEC Structure of imports, 2013

Source: European Commission

Imports of goods and services

The comparatively low import penetration rate signals that Croatia is not leveraging industry‑specific comparative advantages typical of competitive small and open economies. At about 44 %, Croatia’s import penetration rate is substantially smaller than the corresponding average figure in regional peer economies (62.5 %) and far below the 78 % average of the small and open economies of the Baltic States ([23]). However, this rate essentially reflects the relatively closed nature of the Croatian economy and the fact that Croatian firms produce mainly for the domestic market, rather than specialising in export-oriented sectors by leveraging country-specific comparative advantages.

The structure of imports reveals weaknesses in the competitiveness of domestic production. Croatia’s imports are characterised by a relatively high proportion of import of consumer goods. Inter-industry trade is far below that of its regional peers, as witnessed by the low proportion of trade in intermediate and capital goods (especially in the latter category which include equipment and parts and accessories thereof). Another prominent feature is the high proportion of fuels-related imports. This is partly explained by the high energy dependence, but also by intra-industry trade, since Croatia also features a relatively high share of exports of fuels (see Graph 2.2.8). The high rate of penetration of finished goods signals gaps in the productive structure and/or domestic firms’ inability to compete with international firms. Analysis shows that the import content of final demand components (i.e. households and government consumption and investment) is higher than the corresponding values for Hungary, Poland, the Czech Republic and Slovakia, while (see above) the export content is significantly lower. ([24])

Evidence of structural weaknesses and gaps in the domestic production system are highlighted by the limited capacity of domestic producers to meet increases in demand and/or provide cheaper substitutes. Several studies have highlighted relatively high import elasticity with respect to income and relatively low price elasticity. In particular, high income elasticity (i.e. higher than one) signals domestic producers’ inability to accommodate increases in domestic demand, as imports increase proportionally more than the increase in aggregate demand. Low price elasticity (i.e. lower than one), on the other hand, signals that domestic producers do not avail themselves fully of their cost advantage, since reductions in domestic aggregate prices lead to a less-than-proportional import substitution effects. Taken together, these parameters indicate gaps in domestic production (i.e. incapacity to address all segments of domestic demand) and possibly weaker non-cost competitiveness ([25]). Reininger (2007) compares import elasticity for most central and eastern European countries, but disaggregates elasticities by demand component ([26]). The study finds that Croatia stands out in two respects: very high elasticity of imports with respect to consumption and low elasticity with respect to exports (see Graph 2.2.14). This result is consistent with the low import content of exports discussed above.

Graph 2.2.14: Import elasticities with respect to final demand components

(1) Shaded histograms represent non-significant coefficients Source: Reininger T., Factors Driving Import Demand in Central and Eastern European EU Member States, Oesterreichische Nationalbank Workshops, No. 14, 2007, pp. 163-183.

Balance of primary and secondary incomes and capital transfers

The accumulation of external liabilities generates a conspicuous outflow of investment income. Privatisations and follow-up investments by parent companies ushered in flows of FDI flows averaging 6% of GDP in the years before the crisis. Long-term loans to corporates and foreign banks’ deposits were broadly equal. As a consequence of these inflows and valuation effects driven by growing domestic asset prices, the value in the economy of the stock of foreign-held liabilities grew more than five-fold between 2000 and mid-2007, when the NIIP reached almost -95% of GDP. Consequently, an increasing amount of national income was flowing out of the Croatian economy to remunerate foreign owned assets. The contraction of the economy has nevertheless taken a toll on the profits of domestic and foreign-owned companies, bringing about a stabilisation of the investment income outflows at less than 2% of GDP.

Remittances and compensation for employees have broadly offset the negative balance of investment income. Typically a relatively minor component in the current-account composition of advanced economies, remittances from old and new emigrants and compensation of seasonal and frontier workers play a significant role in the case of Croatia, which (given its large diaspora and recent migration outflows) receives the highest net remittances per person in the EU (almost EUR 500 per person residing in Croatia in 2013).

In the long-term a relatively high share of remittances can have both positive and negative effects. Positive microeconomic effects of high remittances can be offset by negative macroeconomic long-term impacts, though most empirical evidence is based on emerging economies. On the positive side, remittances improve the welfare of individual households lifting families out of poverty. They are often used for consumption smoothening, thus insuring households against income shocks. On the negative side, however, remittance flows, especially if relatively permanent, may lead to the currency appreciation and distort capital and labour allocation. By increasing consumption, decreasing labour supply and slowing down capital accumulation, remittances are associated with worsening deficits in the trade balance. Evidence for Croatia suggests that remittances have played a significant impact in reducing inequality and the severity (if not the extent) of poverty. However, there is also evidence that a significant proportion of remittances were channelled to the construction sector and real-estate investments, thus contributing to the asset bubble ([27]). Moreover, abundant remittances contributed, together with tourism and FDIs, to the relative appreciation of the kuna and the real effective exchange rate, negatively affecting industry and exports.

The absorption of EU funds will have a tangible positive effect on the capital and income transfer balance, likely to be partly offset by higher imports. Government transfers contributed negatively to the balance of primary and secondary incomes in 2013 and is forecasted to have continued to do so in 2014, due to a lag between contributions to and from the EU budget. With the take-up of EU funds potentially totalling up to 3 pps. of GDP per year at cruising speed, the balance of government transfers, primary incomes and capital transfers are all set to improve significantly. If anything, the above discussion on FDI, the tourism sector and remittances should give rise to caution as to the potentially distortionary effects that such inflows can have on a relatively small open economy and tight policy control to ensure that funds boost productive investment.

Graph 2.2.15: Balance of primary and secondary incomes

Source: Croatian National Bank

Net investment position and external debt

The slow-down of capital inflows and sizeable valuation effects have stabilised the negative net investment position. The burst of the financial bubble transformed the external financing of the economy. Between 2008 and 2009, direct equity investment dropped by almost 65%, while (with a year lag) direct loans plummeted by more than 90%. In the meantime, as NFCs and especially SOEs shifted financing from foreign to domestic sources, the general government increased its external financing. However, it was the sharp negative valuation effects of assets held by foreigners and, to a minor extent, the still high nominal growth that drove the initial contraction in the NIIP. The contraction was particularly sharp in 2008, when the value of liabilities shrunk by more than 20% of GDP. After a short-lived rebound, valuation effects have continued to reduce the value of the stock of outstanding direct liabilities in the past two years. As a result, the negative NIIP has been trending around 90% of GDP, despite current account deficits continuing until 2013.

Graph 2.2.16: Net international investment position (NIIP)

Source: Croatian National Bank

The changing pattern of external financing and the strong valuation effects have partly altered the structure of the NIIP, while gross external debt is still increasing. Increased foreign borrowing by the sovereign and, to a lesser extent, cross-border, inter-company lending have put gross external debt on a rapidly increasing trend (almost 108% of GDP in Q3-2014). Consequently, whereas by 2007 net external debt was less than half of the total stock of net external liabilities, it is now trending at about 2/3. Note that this change in composition has taken place despite significant debt-to-equity swaps. Between 2007 and the end of 2014 the cumulative amount of debt to equity swaps totalled almost 5% of GDP. The accumulation of official reserves by the Croatian National Bank contributes to the stabilisation of net external debt.

The external vulnerability of the economy is increasing and Croatia will need to generate substantial surpluses to reduce its negative NIIP. According to simulations by Commission services, in order to stabilise the NIIP at -60% of GDP by 2024, the average trade-account surplus should average 3.2% of GDP. This assumption is based on average real growth and inflation of 1.5%, average domestic yield of 3% and average foreign yield of 2%. The high and rising level of gross external debt requires close monitoring, however. The simulations show that even a small (1 pp) increase in the yield on foreign‑owned domestic assets requires an increase of 1.2 pps. of GDP in the trade‑account surplus if the NIIP is to stabilise at -60% of GDP.

Graph 2.2.17: CA surplus necessary to stabilise NIIP at -60% of GDP by 2024 - sensitivity analysis

Source: European Commission

In conclusion, Croatia still faces significant challenges to achieve external balance. In particular macroeconomic risks stem from the high stock of external liabilities and specifically the high share of debt instruments. The turnaround in current-account will progressively contribute to reduce the negative NIIP, but until recently the adjustment has relied excessively on import compression. Despite a positive performance in 2014 exports are still far below potential, owing to weak cost and especially non-cost competiveness. Expanding the diversified, but narrow, export base requires addressing bottlenecks in the product and factor markets, and enhances integration in global value-chain, including through a renewed FDI strategy specifically targeting the tradable sector and greenfield investments. The tourist sector and the remittances counter the negative trade-balance, however, if not well managed, conspicuous currency inflows can also have negative spill over effects and contribute to the erosion of competitiveness.

The level and the dynamics of the Croatian general government debt, aggravated by weaknesses in public sector governance, is increasingly becoming a source of concern. The protracted recession coupled with sizable transfers of liabilities of state-owned enterprises have resulted in a sharp increase of public debt-to-GDP ratio. In addition, the rising share of interest expenditure, despite the relatively low interest rate, signals the kicking-in of a snow-ball effect. All this is aggravated by relatively weak public sector governance, which leads to low efficiency of spending (including in health and pensions) and loose management of public finances.

General government debt

General government debt was not substantially higher than regional peers' at the beginning of the last decade. By 2002, general government debt as a proportion of GDP stood at 35.1% — broadly in line with the average of peer economies. Fiscal policy maintained an expansionary stance throughout the favourable business cycle and by 2008 the debt-to-GDP ratio stood at 36% of GDP, above that of regional peers who had generally taken advantage of the pre-crisis robust growth to reduce general government debt.

Graph 2.3.1:   General government debt

Source: European Commission

The enduring recession took a toll on public finances. Croatian general government debt more than doubled between 2008 and 2013. At the end of Q3‑2014, general government debt is estimated to have been at 78% of GDP, well above regional peers and EU averages. Only Hungary and Slovenia had debt levels close to 80% of GDP, however, in Hungary it is on a declining path and in Slovenia it has been strongly affected by recent bank recapitalisations. The strong increase in the general government debt between 2008 and 2013 in Croatia was a combination of the accumulation of budget deficits, some stock-flow adjustments and subdued nominal growth over the same period. Budget deficits in that period reflect a relatively rapid adjustment of revenues to the declining level of economic activity, and authorities’ reluctance and/or inability to enact more decisive expenditure cuts. The debt level was also affected by the debt assumptions of some SOE’s such as the (now‑privatised) shipyards. A significant adjustment also followed the switchover to ESA2010 rules and the inclusion in the general government sector of large public motorway companies. For details, see Eurostat's note "Revisions to government deficit and debt of EU Member States for 2010-2013". The doubling of the general government debt in a short time was a blow to Croatia’s chances of meeting the Maastricht criteria for euro adoption.

Graph 2.3.2:   Development of general government debt and deficit and Maastricht criteria

Source: European Commission, 2015 Winter forecast

The debt snowball effect is undermining Croatia’s efforts to consolidate public finances. After a record high of 4.8% of GDP in 2011, the primary deficit has narrowed in recent years, but its persistence suggests a structural gap revenue and expenditure levels. The proportion of general government non-interest expenditure is higher in Croatia (43.6 % in 2013) than in most of its peer economies, except neighbouring Slovenia and Hungary ([28]). The revenue-to-GDP ratio has been trending at about 2.5 pps. lower than the ratio of non-interest expenditure to GDP. Although the primary deficit decreased in recent years, interest expenditure has risen sharply. The enduring recession, which affected nominal GDP growth, and the rising stock debt are generating a sizeable snowball effect, even in a climate of relatively low interest rates. The rapid expansion of public debt in recent years means that it is becoming a major burden for Croatia, negatively affecting overall growth prospects and the competitiveness of the entire economy.

The rapidly increasing stock of internal and external general government debt is a source of vulnerability. The risk premium of government debt in a small open economy dependent on foreign financing is affecting private‑sector borrowing costs, reducing investment and creating a drag on the economic recovery. A possible worsening of global market liquidity or the return of risk-averse sentiment could severely weigh on the sustainability of public debt. An additional risk stems from a relatively high proportion of public debt issued or denominated in foreign currency, which exposes the sovereign to exchange rate risk. The significant proportion of debt held by domestic institutional investors (close to two thirds of public debt stock) is a risk-mitigating factor, but also raises the potential issue of the crowding-out of credit to the private sector. In addition, the fact that a third of the debt stock is held by the banks makes the banking sector also vulnerable to any adverse public finance developments. Compared with the EU‑10 countries, Croatia’s short‑term debt is at the higher end and the average maturity of the debt is at the lower end of the distribution. However, compared with the emerging markets that attract similar investors to Croatia’s, the former is relatively low and the latter is high (HNB, Financial Stability Report, 2014)

Graph 2.3.3:   Contributions to change in general government debt-to-GDP ratio

Source: European Commission

Decisive fiscal consolidation is necessary to put the public debt on a declining path. Under a no‑policy‑change assumption, public debt would keep increasing throughout the forecast period. Unless a medium-term consolidation strategy is adhered to, the debt stock will continue to rise due to high budget deficits relative to growth and inflation developments. In addition, different sensitivity tests show that adverse economic events (such as interest rate increases or negative inflation shock) would have a significant negative impact on debt dynamics (see Box 2.3.1). Importantly, concerns from various stakeholders, such as ratings agencies, investors and international organisations, as to Croatia’s capacity to deliver on budget plans and carry out structural reforms might lead to a change in market sentiment. The steep rise of the debt ratio in the previous period could even provoke an overreaction by the markets.

Box 2.3.1: Long-term projections of general government debt The public debt trajectory has been simulated under alternative scenarios. The baseline scenario is based on the Commission winter 2015 forecast, where short- and long‑term interest rates are set in line with the forecast implicit interest rate, the proportions of short- and long-term public debt, and data on maturing debt. This scenario incorporates a number of technical assumptions, as follows: - the structural primary balance stays constant beyond the forecast horizon at the 2016 value. The cyclical component of the primary balance is calculated using the (country‑specific) budget balance sensitivities to the cycle until assumed output gap closure (2019); - the long-term interest rate on new and rolled-over debt is assumed to converge to 3 % in real terms by the end of the projection horizon (2025), while the short-term rate converges to an end-of-projection value consistent with the 3 % long-term interest rate and the value of the euro‑area yield curve; - the inflation rate of the GDP deflator is assumed to converge in a linear fashion to 2 % in 2019 and remain constant thereafter; - a short-term temporary feedback effect on GDP growth is introduced in the consolidation scenario (a 1 pp. of GDP consolidation effort has a negative impact on baseline GDP growth of 0.5 pp. in the same year); - the stock-flow adjustment is set to zero after 2016. Medium-term growth projections are based on the T+10 methodology agreed with the EPC and assumed to average 0.4 % between 2013 and 2020 and to increase to 0.7 % on average between 2020 and 2025. Source: European Commission

(Continued on the next page)

Box (continued)

The following sensitivity tests are conducted around the baseline: i)       a standardised (permanent) negative shock (-1 pp.) to the short- and long-term interest rates on newly issued and rolled‑over debt; ii)      a standardised (permanent) positive shock (+1 pp.) to the short- and long-term interest rates on newly issued and rolled‑over debt; iii)     a standardised (permanent) negative shock (-0.5 pp.) on GDP growth; iv)     a standardised (permanent) positive shock (+0.5 pp.) on GDP growth; v)      a standardised (permanent) negative shock (-0.5 pp.) on inflation; vi)     a standardised (permanent) positive shock (+0.5 pp.) on inflation; vii)    a standardised (permanent) negative shock on the PB equal to 50 % of the forecast cumulative change over the two years in question; and viii)   an enhanced (permanent) positive shock (+2 pps./+1 pp.) to the short- and long-term interest rates on newly issued and rolled‑over debt. The simulations show that general government debt is likely to keep increasing in the medium term in the absence of consolidation measures. Favourable assumptions on interest rates and growth would bring the debt ratio down somewhat, to just over 105 % of GDP by the end of the projection horizon. By contrast, under unfavourable scenarios on interest rates or inflation, debt would reach around 118 % of GDP. The ‘Stability and Growth Pact (SGP) institutional scenario’ in the graph assumes that the structural primary balance is adjusted in line with the fiscal efforts recommended by the Council (with improvements in the structural primary balance by 0.9 % and 0.7 % of GDP in 2015 and 2016 respectively). Under this scenario, the debt ratio would start to fall slowly after 2016, to 80 % of GDP in 2025.

Expenditure

Apart from its higher level, expenditure in Croatia differs markedly in structure from that of regional peers due to the higher proportion of non-productive components. Compared with peer economies, public expenditure by function in Croatia in 2012 (the only year for which data is available) was higher than EU‑10 average in several categories, such as general public services, defence, public order and safety and economic affairs, while less was spent in the areas of environment, community amenities and recreation. Health expenditure in Croatia is also higher than in the EU‑10, although the reported figure for 2012 contrasts sharply with more recent estimates ([29]). Comparison of expenditure by ESA categories to the EU‑10 countries reveals that Croatia is spending relatively more on compensation of employees, intermediate consumption and subsidies. On the other hand, the proportion of investment expenditure is markedly lower, which is worrying and gives rise to concern as to how it might affect the country’s capacity to match the necessary EU co-financing. In order to fully benefit from the funds available, additional funds will be needed for the co-financing of projects. This would necessitate redirection of spending from other categories, due to the limited fiscal space on the revenue side and already high budget deficits.

Graph 2.3.4:   Structure of expenditure by COFOG function (2012)

Source: European Commission

An additional source of fiscal risk are the recurrent financial injections to the healthcare sector. The sector continuously generates arrears, which necessitate ad hoc payments from the state budget. According to the authorities, such injections were made in 17 out of the past 20 years. Most of the arrears are generated in hospitals in relation to payments for drugs and material supplies, and are a result of the reimbursement scheme, inefficiencies and the hospital’s rigid cost-structure. Despite a one-off payment of HRK 3.2 billion (1 % of GDP) from the state budget in 2014 to cover the outstanding obligations of both hospitals and the Health Insurance Fund, the total level of arrears reached HRK 3.1billion (0.9% of GDP) in November 2014.

Graph 2.3.5:   Outstanding liabilities of hospitals

Note: The graph excludes arrears generated in other segments of the healthcare system. Source: Croatian authorities

Empirical analysis suggests that the binding constraint may not relate to the high level of public expenditure, but rather its inefficiency, which weighs on growth potential. A recent analysis of optimal government size concludes that Croatia could cut government expenditure in order to achieve higher growth rates. (Bađun, Pribičević, Deskar-Škrbić, 2014) However, in their view, the binding constraint is not so much the overall size of the government, but rather the low comparative efficiency of public spending in most of the areas that affect economic growth, such as education, healthcare, public administration and public investment. This inefficiency is the result of the poor quality of public governance and strategic planning, corruption, poorly defined property rights and a lack of transparency in government policy‑making. Substantial inefficiencies are found also in the pension system (see section 4.1), where they result in a low average level of pensions.

Graph 2.3.6:   Structure of expenditure by ESA category (2013)

Source: European Commission

Croatia is currently reviewing its expenditure. In 2014 Croatia was recommended to carry out a thorough expenditure review with a view to achieving efficiencies (in particular in wage, social security and subsidy outlays) and providing sufficient fiscal space for prioritising growth‑enhancing expenditure and investment. Preparation of the expenditure review has advanced. A government decision on it has been endorsed. Five sub-committees have been established, each responsible for designing options for potential savings in a particular policy area. The authorities plan to review spending on state‑budget wages, healthcare, public subsidies, agencies’ operations and tax expenditure. The sub-committees are expected to submit their reports, containing a package of measures which should lead to a 10 % reduction in expenditure, to the central committee by 1 February 2015. By 1 March, the government is due to decide which of the proposed measures will be implemented. However, it appears that a clear commitment to implement fully the findings of the review is still lacking.

Taxation

Measures have been taken recently to lower taxes on labour, but not always with a clear focus. Croatia first lowered social security contributions for employers in 2012 — the rate for health insurance contribution for SSC employers was decreased from 15% to 13%, but this was reversed in April 2014. As from 1 January 2015, the threshold for the highest personal income tax rate of 40% was raised to HRK 13 200 (from HRK 8 800), while the basic personal allowance rose to HRK 2 600 (from HRK 2 200 per month). Due to the design of the measure, the cuts are likely to benefit mostly middle and high income earners, thereby limiting their intended effect of boosting personal consumption. On the other hand, the direct budgetary impact is a reduction in general government revenues by around 0.6 % of GDP. In addition, from 1 January 2015, Croatia has introduced a new tax incentive to hire young workers. Employers offering a permanent contract to staff under 30 will be exempted from paying social security contributions for up to five years. This reduces the labour tax burden of employing young workers, one of the groups most affected by unemployment. The budgetary impact is expected to be limited, as most of the newly recruited staff would otherwise have been unemployed and not pay social security contributions.

Labour taxation in Croatia seems comparatively low, but labour supply might be affected by the interplay of the tax and benefit system and labour taxation. Theoretically, further lowering the tax burden on labour could improve cost competitiveness, create and preserve new jobs, encourage job-taking and open up new investment opportunities. However, it is debatable to what extent labour taxation, taken as the effective tax rate on labour earnings, explains the poor performance of the Croatian labour market. Estimates from 2013 point to a tax wedge of about 34.9% for a single worker earning 67% of the average salary, lower than the value in most of peer countries at the time. As regards the implicit tax rate on labour, Croatia scored relatively low against the EU Member States in 2012 (22nd of 28) and the rate has decreased over time (EC, Taxation Trends in the European Union, 2014). Overall, however, the interplay between labour taxation and the benefit systems and what that implies for effective marginal tax rates may be affecting labour supply. To shed light on this important question, Croatia was recommended to review its tax and benefits systems by the end of 2014. A dedicated working group to investigate the matter was set up but its work is still ongoing.

Statutory and effective corporate income taxation rates are comparatively low; several tax breaks and incentives apply. Croatia has a state-wide corporate income tax rate of 20%, with no local or regional surcharges. Looking at the base, Croatia offers several tax breaks and incentives: for R&D, for interest on debt (up to a certain limit) and, tax exemptions for business activity in areas of the country particularly affected by war damage, such as the city of Vukovar and less developed municipalities. Also, in 2014 Croatia adjusted a tax break for reinvested profits, which has been in effect since January 2013, by introducing two new criteria: employers are not allowed to reduce their workforce and must increase their capital stock. Also, there is a plan to introduce a new tax relief for investment in innovative start-up companies and innovation activities, employing researchers, and technology transfers. The debt bias in Croatia does not appear particularly high according to data from a ZEW study on effective corporate tax rates in the EU. (EC, Tax Reforms Report, 2014)

It is important that tax systems are supportive of productive investments, especially when it comes to innovation and technology. To stimulate research and innovation, Croatia has put in place R&D tax incentives and these have been evaluated positively.(CPB, EC, A Study on R&D Tax Incentives, 2014) However, it could be useful to review the access to these incentives especially for young innovative companies. The administration of the incentive could be made more business‑friendly, e.g. Croatia could provide e‑services for taxpayers wishing to make use of it.

Croatia plans to reform property taxation in 2016. Progress has been made in removing technical obstacles to meeting the timetable, but the key issue remains the lack of clear political commitment. Recurrent taxes on real estate are among the least detrimental to economic growth, mainly due to the immobile tax base. Croatia’s revenues from such taxes as a proportion of GDP are second lowest in the EU. (EC, Taxation Trends in the European Union, 2014) To contribute to growth-friendly fiscal consolidation, Croatia was recommended in 2014 to present a concrete strategy to reform recurrent property taxation. The World Bank estimates potential additional revenues from a modern property tax of about 1.5 % of GDP, based on assumptions of an updated cadastre and transparent valuations (World Bank, Croatia Public Finance Review, 2014). In the 2014 Convergence Programme of Croatia, the government provides a more conservative estimate of 0.4 % of GDP. The final outcome will depend on the actual design of the tax; exemptions are currently planned inter alia for government properties and families on low incomes. Croatia has drafted a plan, defined the roles and responsibilities of the various actors involved and drawn up a communication strategy to accompany the introduction of the new tax. A significant amount of (mostly preparatory and technical) work has been done and some good practices taken over from other Member States. However, it appears that the focus has so far been on planning rather than implementation. The authorities have decided to use the joint information system, comprising cadastre and land register data, as a basis for assigning tax liability. This will be complemented and corrected by information from local government databases on communal fees and the national database of electricity consumers. As a next step, the plan is to phase in the property tax through a pilot project in two municipalities (local units). Uncertainty remains as to whether the new property tax will be in effect as of January 2016. While logistical and technological issues still pose a risk to the timely introduction of the tax, the key issue for fully delivering on this reform is sufficient political endorsement.

A review of tax expenditure is currently under way with a view of broadening the tax base. While consumption taxation is already relatively broad‑based ([30]), direct taxes are (according to a public finance review by the World Bank) eroded by large tax exemptions for both households and businesses. Broadening tax bases could result in the additional revenues that are so necessary to help with fiscal consolidation. As an example (according to the same World Bank document) rationalising tax expenditure by streamlining child allowances could increase revenues by +1 % of GDP in the personal income tax (PIT) system alone. On the other hand, the effect of any base-broadening measure on redistribution and welfare must be carefully taken into account, especially as nearly one third of the population is at risk of poverty and social exclusion. The authorities have completed a ‘catalogue’ of tax expenditure and have defined the obligation of a regular reporting of tax expenditure.

Croatia is taking steps to modernise and improve the efficiency of its tax administration. Issues calling for attention include a lack of end‑to‑end and uniform compliance risk management across the whole country, the limited use of ICT to support core tax administration processes (in particular, tax audit and investigation, risk analysis, tax arrears management) and the fact that the extent of tax evasion is not measured. Compliance costs based on time to comply are above the EU average (World Bank, Doing Business 2015, 2014World Bank, 2014b). More efficient tax administration would also provide better support to tax reforms and help those who want to invest and do business in the country.

Steps have been taken to improve tax compliance. In particular, Croatia stepped up efforts to recover overdue tax debts. The focus was put on the enforced recovery of debts, with special attention to debtors who have delayed repayments and/or are paying in instalments. Following a preliminary analysis of VAT compliance risks in September, the country prepared an action plan to tackle VAT fraud. According to the Croatian Tax Administration, implementation of the action plan is already producing positive results. The average collection rate of declared VAT in 2014 was close to 94 % of assessed liability, as compared with 91 % in 2013. The amount of VAT collected in 2014 rose by HRK 0.6 bn (0.2 % of GDP). However, the authorities were unable to isolate the effect of the action plan from other contributing factors, such as the increase of the intermediate rate from 10% to 13% in January 2014 or the impact of underlying changes in economic activity. Croatia continued to reform the tax administration by reducing strongly the number of regional offices and expand the use of ICT. Other control measures include broader information obligations as regards the acquisition of real estate (IBFD, Croatia Real estate transfer tax law amendments proposed, 2014). A key result has been the establishment of a new VAT anti-fraud unit. Although this is welcome, it is still too early to evaluate the impact of this and of the reorganisation. 

Fiscal framework

The fiscal framework has been bolstered by recent reforms, but significant challenges remain. Croatia’s fiscal framework has been reformed in recent years. The Budget Act was amended in 2009 and the Fiscal Responsibility Act adopted in 2011, partly to meet pre-accession requirements. The latter was further amended in January 2014 in order to transpose Council Directive 2011/85/EU on requirements for budgetary frameworks of the Member States. Following the wide-ranging CSR on Croatia’s fiscal framework in 2014 both laws are undergoing further amendment. Specifically the CSR called for:

 (1)          reinforcement of the budgetary planning process (in particular by improving the accuracy of macroeconomic and budgetary forecasts and strengthening the binding nature of the annual and medium‑term expenditure ceilings);

 (2)          improvement in the design of fiscal rules; and

 (3)          grounding in law the Fiscal Policy Commission, strengthening its independence and broadening its mandate.

The status of the national fiscal watchdog is improving; however the independence of its institutional set-up is still not warranted. The Fiscal Policy Commission is fiscal monitoring body. It was established in 2011, reformed at the end of 2013 via a parliament decision and may be grounded in law in the envisaged amendments to the Fiscal Responsibility Act. The Commission is now hosted within the Parliament, which appoints Fiscal Policy Commission board members and is expected to fund it on its budget. The Commission is chaired by the head of the Finance and State Budget Committee and the other six members, all of whom have a five-year mandate and should not be politically affiliated, are experts from the Croatian National Bank, the State Audit Office, research institutes and academia. The fact that the Fiscal Policy Commission is not established in law at this stage, that it is chaired by the head of the Finance and State Budget Committee and the absence of dedicated non-board staff put a serious question mark over its autonomy within the host institution.

Furthermore, numerical fiscal rules subject to legislative amendments in January 2014 could be further improved. As currently defined, the medium-term objective is not sufficiently operational and does not ensure full compliance with the Stability and Growth Pact (a structural balance that ensures that the general budget deficit is not higher than 3% of GDP and that the debt-to-GDP ratio does not exceed 60%). Also, escape clauses are to be triggered where the annual growth rate of real GDP is ‘significantly below’ potential GDP growth, but the definition of ‘significantly below’ is insufficiently precise.. The risk of excessive use of the clauses could be mitigated to some extent by the requirement that recourse to them be assessed by the Fiscal Policy Commission, but greater precision is still desirable. In its November 2014 position paper on the draft budget for 2015, the Fiscal Policy Commission sees a risk of non-compliance with this fiscal rule in 2015. The Government did not react specifically to the points raised.

Weaknesses in relation to effective control over expenditure and consistent application of budgetary constraints negatively impact fiscal policy‑making. Proposed changes to the legislation in this area, such as not allowing the introduction of new budgetary programmes until the financing of existing programmes is fully secured, are only minor improvements significant further progress in needed.

Reforms, budgetary planning and consistent application of fiscal rules continue to be challenging. Although the 2015 annual budget law was endorsed on time, there was, once again, serious deviation from the legislative timeline for budgetary planning. The Guidelines for Economic and Fiscal Policy, a three-year document that serves as the basis for annual budgetary planning and draft budgets was submitted almost six months after the legal deadline. Work is underway to produce ESA aligned budgetary projections by the time the 2015 Convergence Programme is submitted, but significant discrepancies persist between budgetary targets and actual outcomes. Moreover, the authorities seem incapable of establishing an accurate real-time estimate of the budgetary situation according to ESA standards; this significantly complicates budgetary planning. In 2014, there were two revisions of the central government budget: the first, in the first quarter of 2014, after a consolidation package had been put together in response to the 2014 EDP recommendation; the second, rectifying revision in the fourth quarter of the year. Specifically, the second revision increased the planned central government deficit by 0.7 pp. of GDP, only for it to be cut by 0.9 pp. of GDP just two months later, in the light of the preliminary central budget outturn. Little explanation was published to detail the drivers for such revisions. In its November 2014 position paper on the draft budget for 2015, the Fiscal Policy Commission considers the macroeconomic forecasts for 2015 to be optimistic. A thorough independent evaluation of macroeconomic and budgetary forecasts leading to tangible improvements, enhanced transparency on the nature of revisions and more safeguards against overspending would benefit budgetary planning greatly by strengthening its quality and binding nature. The existing debt rule under the 2009 Budget Act is undermined by the absence of preventive mechanisms and a track record of non‑compliance. Although the rule stipulates that the central government debt-to-GDP ratio should increase only if it did not exceed 60 % of GDP the previous year, the debt level keeps rising and was close to 80 % at the end of 2014. The credibility of the debt rule could be improved by extending its coverage to the entire general government, through independent monitoring, a preventive mechanism where there is a risk of non-compliance and a correction mechanism in actual cases of non‑compliance.

Fiscal frameworks at local level also need to be strengthened. The General Budget Act lays down procedures, a calendar and reporting obligations of local government units, central government and other budgetary users. The Local Government Units Financing Act identifies the sources of revenue at local government level. The current arrangements could be improved so as to ensure better coordination between budgetary timelines and reporting obligations at local and central level, and a more systematic approach to local governments’ remits.

Public sector governance

The distribution of competencies between administrative levels of national and local government is complex and characterised by a high degree of fragmentation. Formerly a relatively centralised state, Croatia reviewed its local governance structure in 1992 and again in 2000. Nevertheless the process was characterised by neglect for economic and fiscal considerations and to some extent a transposition of existing models of local governance without adequate contextualisation ([31]). This resulted in blurring of the roles, responsibilities and lines of accountability of local and decentralised administrative bodies. A very complex network has been created, of 1 279 state administrative departments at local level, working to 21 state administration offices at county level and responsible for the direct application of state laws (exercising administrative supervision, resolving administrative cases, managing various registers and issuing various certificates). Amendments to the State Administration System Act were drafted in September 2014 to address these inefficiencies, but their adoption by parliament has been delayed. The amendments involved merging 20 regional administration management structures into five.

The suboptimal decentralisation policy undermines management of public finances and efficient public expenditure. The complex and non-transparent attribution of policy functions across different levels of government contributes to excess spending since local governments do not bear the full costs of their decisions ([32]). Administrative weaknesses in the local authorities directly affect a number of areas, such as tax collection, the provision of social benefits, the management of European Structural and Investment Funds (ESIF), public procurement and the provision of public services. The fragmentation of sub-national governance units, with 428 municipalities, 127 cities, 20 counties and the City of Zagreb (special status), impairs their administrative capacity and thus weighs on their economic efficiency. Reallocation and coordination of regional and local competencies that could lead to better efficiency of fiscal policies remains a priority.

The gap between political and fiscal decentralisation contributes to expenditure pressures at local level ([33]). On aggregate, 60 % of local government revenue is from taxes, 25 % is other revenue (mostly administrative fees and property income) and up to 15 % takes the form of grants, which are mostly transfers from central government. The allocation of revenue to local government units is based on revenue‑sharing arrangements under the Act on the Financing of Local Units. Two main taxes shared between local and central government are personal income tax and real estate‑transfer tax. Of local government tax revenue 90 % is from personal income tax and surtax on personal income tax. The remaining 10% is split broadly equally between real‑estate transfer taxes and the special tax on goods and services. The distribution of personal income tax revenue is very complex: it depends on the number of decentralised functions that local government has decided to take over, but also on the level of development of the local unit. In general, cities and municipalities receive 60 % and counties receive 16.5 % of the personal income tax collected in their jurisdiction. They can receive up to 6 % in addition, depending on the extent of any additional decentralised functions they assume, such as primary or secondary schooling, social security, health and fire protection services. Most of the remaining funds go to the joint equalisation fund used to provide grants to the local units, with only a small part being used to provide grants for local EU‑financed projects. Of the real estate transfer tax revenue, 80 % goes to the cities or municipalities and 20 % to central government. Overall, however, due to the overall structure of taxation in Croatia, 90 % of general government revenue goes to central government. This relatively centralised revenue allocation, coupled with the complex division of expenditure responsibilities across different tiers of governance, gives rise to vertical and horizontal fiscal imbalances offset by central government transfers. The magnitude of funds provided to the local governments from the central budget due to their insufficient fiscal capacity can be estimated on the basis of the level of ‘current grants’ in the central budget ([34]). Almost exclusively these comprise transfers from the central budget to lower government levels. In 2004-13, following changes to the legislative framework for local government, they averaged around 0.8 %. of GDP. Besides suggesting insufficient fiscal capacity in some local units, the existence of such fixed transfers directly weakens the incentives for local authorities to reduce spending. Recent research finds that only three counties and the City of Zagreb have a positive net fiscal position. A breakdown of general government revenues suggests that Croatia’s fragmented regional organisation is inefficient and probably un-sustainable in the long term ([35]).

The 2015-20 Strategy for the Development of Public Administration is to be adopted by June 2015. The Strategy defines reform objectives as regards the allocation of responsibilities of government units at different levels, the provision of public services and the management of human resources in the public administration. The measures proposed in the Strategy remain very broad. Following a public consultation, a final proposal is to be presented by the end of March 2015 and discussed by parliament in June. The authorities are currently gathering data and analysing the distribution of functions, costs and administrative capacities of local units. The results should inform the development of structural reform options and an action and implementation plan, to be prepared by all relevant ministries under the supervision of a central coordination unit. Progress on reform in this area has been slow in recent years.

Corruption continues to present a pressing challenge. Corruption reduces expenditure efficiency, impedes productive public investment and undermines business confidence. A recent evaluation report, focusing on corruption prevention in respect of members of parliament, judges and prosecutors, singled out a number of specific shortcomings, including the absence of a code of conduct for Members of Parliament, inadequate technical and personnel resources of the Commission for the Prevention of Conflicts of Interest and a lack of a proactive approach in its preventive role ([36]).The protection of whistle-blowers, i.e. persons who report corruption, also appears weak.

The new anti-corruption strategy, awaiting adoption by the parliament, is broad in scope but lacks focus and a sufficient level of detail. The Croatian government recently published a draft anti-corruption strategy for 2015-20, public consultation was concluded on 30 November 2014 and parliamentary adoption is expected in the course of 2015. The strategy outlines, at a very general level, 31 horizontal and 33 sector-specific measures. While the ambition and the coverage of the strategy are broadly commensurate with the perceived significance of identified risks, there is little clarity on the main priorities and how the main risks would be specifically addressed in the forthcoming action plans.

Tools to detect irregularities in public procurement, including at local level, appear weak. While the legislative framework for public procurement is broadly adequate, the authorities appear to lack a comprehensive overview and robust tools to detect irregularities in public procurement. For instance, information on tenders below the national thresholds is not sufficiently detailed to detect the splitting of contracts into smaller parts. Information is also lacking as regards the proportion of contracts with only one bidder and the benchmarking of prices tendered by different units or over time. Detection of prior agreements between tender participants is rather weak. Average bidder participation is around 2.5 bids per tender, below the EU average, and this has an impact on public procurement outcomes ([37]). Local knowledge and capacity are under‑developed. The authorities envisage placing more staff into relevant institutions in 2015 to reinforce inter alia control over public procurement in EU-funded projects (see section 3.5).

Measures are being introduced at central level. The authorities are successfully expanding the scope of central procurement to a wider range of central-government institutions, which contributes to increasing efficiency of public procurement and to building a knowledge base within the State Office for Centralised Procurement. Apart from measures that will be specified in the above-mentioned action plan implementing the anti-corruption strategy, amendments to the Public Procurement Act are being prepared (by the end of 2015) to transpose provisions of the relevant EU Directives. Criteria for using the ‘most economically advantageous’ tender specifications are planned to be refined.

Transparency in public procurement is being improved through initiatives led by the non‑profit sector. In March 2013, a local non‑profit organisation launched a web portal on public procurement, co-financed from EU funds and accessible free of charge, which consolidates information on the implementation of public procurement procedures and the companies involved. It also contains information on the assets and interests of public officials, in line with asset disclosure rules. Such aggregated data allow the carrying‑out of useful cross-checks ([38]).

Public enterprises

Public enterprises continue to pose fiscal risks. Notwithstanding recent progress, such as the privatisation of the loss-making shipyards, Croatia retains a significant degree of state involvement in the economy. The number of companies with partial or full state ownership has further increased as a result of the debt-to-equity swaps agreed under the pre-bankruptcy settlement procedure. Moreover, box 2.4.1 shows that there are highly concentrated pockets of indebtedness among public enterprises. Given the track-record in assuming public enterprises’ debts and arrears, such situation poses fiscal risks. In addition, according to the EDP notification, most companies that are classified in the general government sector are running deficits; the aggregate deficit reached 0.6 % of GDP on average in 2010-13.

Local authorities contribute to the accumulation of debt in public companies. Recent research by the Croatian Institute of Public Finance ([39]) suggests that fiscal risks to the government budget accrue from ‘communal enterprises’ owned by the cities and municipalities. Due to the relatively strict debt restrictions imposed on the local units, some have leveraged such enterprises as vehicles for loans to fund infrastructure investment. The indirect indebtedness of the enterprises is nevertheless limited, with the exception of the City of Zagreb’s.

Public enterprises weigh on aggregate productivity. According to ORBIS database data, public enterprises employ up to one worker in eight and therefore play a significant role in aggregate productivity ([40]). It is important to note that these companies are in industries where private, domestic or foreign companies are also operating, and may negatively affect competition and input costs. Public ownership is particularly common in the electricity and gas sector, water industry and transportation, but also in construction and real-estate activities. Controlling for company size and sector of activity, public enterprises’ productivity was on average 29-33 % lower in 2006-13 ([41]). To put it another way, of two companies of equal size operating in the same specific sector and using the same measured quantity of labour and capital, the one with a public ownership share is likely to generate, on average, only about two thirds of value‑added achieved by the other, fully private company. The economic performance of private companies is superior to that of public ones in all sectors of activity, including as regards financial performance indicators, such as return on capital employed. The biggest differences are found in the utility sectors and construction. The measured performance gap of public enterprises is problematic in view of their large share in the economy.

Graph 2.3.7:   Total factor productivity by sector and ownership (average for 2006-13)

Source: European Commission, ORBIS

Both the fiscal risks and the productivity gap are exacerbated by a number of weaknesses in the governance framework. A sound governance framework is critical to ensuring that public enterprises contribute positively to a country’s overall economic efficiency and competitiveness. Several elements of the Croatian governance framework fall short of ‘best practice’, whereby the state should act as an informed and active owner, establish a clear and consistent ownership policy and ensure that governance is exercised in a transparent and accountable manner, with the necessary degree of professionalism and effectiveness (OECD, Guidelines on Corporate Governance of State-Owned Enterprises, 2005). Weaknesses identified include: the absence of monitoring and supervision over the entire public corporate sector; obstacles in listing minority packages of company shares; the absence of medium-term benchmarks against which the administrators of the companies (in particular those with a strong commercial objective) could be held accountable; and low transparency in managerial appointments. Improving the performance of public enterprises will be key to managing the fiscal risks as analysed above.

Despite its importance in the economy as a whole, supervision of public enterprises is fragmented and incomplete. The institutional framework for public enterprises is defined in the 2013 Act on the Management and Disposal of State Assets. According to the Act, the ownership function for 59 companies or agencies of ‘strategic’ or ‘special’ interest for Croatia is shared between competent ministries and the State Office for State Property Management (DUUDI), with a somewhat stronger role for competent ministries in ‘strategic’ companies ([42]). The ownership function for the other, 500+ companies owned by central government, most with a minority public share, is performed by the Centre for Restructuring and Sale (CERP). In 2013, it was decided that these companies would be sold or liquidated in two to three years, a target which is turning out impossible to meet. Subsidiaries owned by the above-mentioned SOEs and companies established by local entities and their subsidiaries are not considered state property and are not directly and systematically monitored by any public authority, which is a serious loophole in the Act. While differentiation in the exercise of ownership rights according to the level of ownership is appropriate, it makes the monitoring framework too narrow in view of the central government’s responsibility to ensure the proper functioning of the local labour and product markets and the fact that it ultimately bears the fiscal risks.

The authorities plan to reduce the number of ‘strategic’ companies, which is a positive step. One of the differences between ‘strategic’ and ‘special’ public companies is that the ‘strategic’ ones are not envisaged to be listed on a regulated financial market. International evidence suggests that placing even a modest part of company shares on the stock market can significantly improve the quality and nature of governance since it fosters compliance with disclosure requirements, securities’ regulation and governance codes ([43]). In this respect, maintaining the distinction between the two company types may become a bottleneck to reaching higher efficiency. Shares of 13 companies are already listed on a stock market.

The governance framework suffers from the absence of a medium-term anchor. While, in principle, the companies of ‘special’ interest are expected to maximize profits, the benchmark against which this goal is assessed has not been made sufficiently operational. Annual asset management plans report only short-term projections of each company’s profit. These plans are assessed both within the year and ex post, which is considered good practice (the authorities plan to strengthen reporting standards, which could contribute to improving the still weak implementation record). However, medium-term objectives are not defined, which decreases the transparency of target‑setting and makes assessment vulnerable to short-term fluctuations. Their absence also weighs on the reliability of medium-term projections of the general government. Benchmarking against the performance of similar companies in the private sector is not performed.

The nomination and appointment of directors and supervisory board members lack transparency. These positions are not publicly advertised, which restricts the pool of candidates. Nominations are made by competent ministries with DUUDI’s prior approval. Candidates are then assessed by committees set up by competent ministries and DUUDI, with the government formally empowered to make the appointments. Competency requirements of nominees are relatively weak at present. The combination of low transparency in hiring and weak competency requirements is not conducive to the sound management of public property.

Partial improvements are being prepared. Following the release of a register of SOE board members and management in September 2014, the authorities announced an overhaul of managerial nomination, to be adopted in the first half of 2015, and a revision of remuneration rules by the end of 2015. The proposed nomination procedure for CEOs aims at allowing the (optional) use of professional staffing agencies to select or recommend managers of SOEs. The proposal will require further work to ensure that top candidates are recruited in a transparent manner, while their performance is assessed against clear objectives.

Privatisation proceeded slowly in 2014, but several projects were initiated at the end of the year. In 2014, the government sold a minority package (39%) in the insurance company Croatia Osiguranje to a private investor, reducing the public share in the company to 28%. Attempts to sell other companies failed in 2013 or 2014 (e.g. Croatia Airlines, HPB, strategic partnership for Petrokemija). Several smaller SOEs were put on sale at the end of the year: two hotel operators, Jadroplov (a shipping firm), Meiso (a footwear producer), Nacionalna Veletrznica (a consultancy) and 3. Maj Motori i dizalice (engine and deck crane manufacturer).

Private sector debt increased rapidly in the years before the crisis, albeit from a relatively low base. In the previous expansionary phase, the underlying weaknesses in the economy were partially blurred by the investment-led boom. Significant capital inflows, partly driven by high global liquidity, over-optimistic expectations and strong risk appetite for emerging markets, were partially channelled through Croatia’s largely foreign owned banking sector. Other liabilities were accumulated directly by the non‑financial corporations (NFCs) through cross‑border loans from parent banks and inter‑company lending.

Graph 2.4.1:   Money (M1) and domestic credit growth

Source: Central National Bank, European Commission

The ongoing adjustment process is proceeding relatively slowly, on the back of loose fiscal policy and efforts to stimulate bank lending. The slow-down of foreign capital inflows initiated a process of deleveraging, which was partially offset by high liquidity provision to the banking sector and a relaxing of the conservative macro-prudential policies that had characterised the previous expansionary cycle. The accumulation of debt in the private sector therefore continued in earnest in 2010 and 2011. Eventually, however, credit started to contract on the back of both demand and supply factors. Deleveraging nevertheless progressed relatively slowly, in comparative perspective.

The worsening macroeconomic environment took a toll on banks’ profitability, but the financial sector is withstanding the prolonged recession. Before the outbreak of the global financial crisis, the levels of private-sector debt in several central and eastern European countries were not considered excessively high. The sustained credit dynamic was mainly explained by the process of financial deepening and convergence towards EU standards. Despite this prevailing lenient assessment, monetary authorities in several countries, including Croatia, activated macro-prudential policies aimed at limiting credit growth. This conservative stance prevented the accumulation of excessive risk exposures in the domestic banking sector. The build-up of the central bank’s official reserves and capital and foreign currency liquidity buffers in the banking sector underpinned both currency stability and the resilience of the financial system throughout the recession. Although the recent review of major banks’ balance sheets confirmed the resilience of the Croatian financial sector, the high level of non-performing loans (NPL) remains a source of concern.

Recent indicators point to a slowdown in private-sector deleveraging, but the financial sector remains resilient. This section builds on previous year’s in-depth Review and looks at deleveraging in the non-financial corporations and in the household sector, focussing in particular on recent dynamics. It also assesses financial-sector stability, with a focus on recent reviews of banking sector resilience.

Corporate sector debt

Robust investment rates in the pre-crisis years resulted in a rapid accumulation of corporate debt. As a result of expansionary credit dynamics, the aggregate NFC indebtedness – measured as the consolidated debt-to-GDP ratio– increased by more than 45 pps. between 2001 and 2010 when it reached a peak of 83.8% of GDP. This figure, which has been recently updated to comply with new ESA2010 regulations, points at a much lower debt-to-GDP ratio, than the ESA1995 figure. The adjustment reflects the reclassification of two highly indebted SOEs in the general government sector. Faced with the constraints imposed by the Croatian National Bank’s (HNB) conservative macro‑prudential policy, non-financial corporations partially bypassed the domestic banking sector, resorting to cross-border loans by foreign parent banks and loans from parent companies. By 2010, therefore, almost 60% of total NFC debt was foreign debt. Domestic loans were to a great extent denominated in or indexed to foreign currencies – predominantly the euro.

Graph 2.4.2:   Consolidated debt-to-GDP ratio, Non-Financial Corporations

Source: European Commission

Deleveraging in non-financial corporations progressed at a moderate pace. With the burst of the financial bubble, capital flows contracted sharply and CNB relaxed macro-prudential policy.  Credit growth to NFCs therefore remained positive until the first half of 2012. Thereafter, demand and supply factors contributed to a reduction in domestic credit. The contracting economy significantly reduced investment opportunities while on the back of rapidly increasing NPLs, the banking sector adopted a more conservative approach to credit supply. According to the EIB’s CESEE Bank Lending Survey (first half of 2014), Croatian banks reported that declining demand and decreasing-to-neutral supply were behind the negative credit growth experienced between October 2013 and March 2014. According to the Croatian National Bank quarterly Bank lending survey, however, in the third quarter of 2014, for the first time since the introduction of the survey, the banks eased their credit standards as applied to the approval of most groups of loans to enterprises and households. The recent fall in credit activities of banks was heavily influenced by weakened demand, particularly in the segment of loans to small and medium-sized enterprises and loans for house purchase.

Lending patterns differed markedly between public and private sector corporations. As highlighted in the box 2.4.1, a significant share of corporate debt owned by public companies is concentrated in few unprofitable sectors (such as construction and utilities). Indebtedness in the private sector is less concentrated, both in terms of sector and number of enterprises. Leverage ratios in public and private owned enterprise result from different trends in borrowing. Negative credit flows put private corporations on a deleveraging path in the second half of 2012. On the other hand, as credit risk of other NFC increased, SOEs might have appeared to banks as a safer and cheaper investment. Credit institutions’ placement in this sector continued to grow until recently. In 2014, however, the trend changed: domestic credit institutions’ placements with SOEs went down by more than 10% on year-on-year – whereas the contraction in placements was less than 1% for private NFCs. Domestic debt write-off have so far played a limited role. A record high of about HRK 2 bn (around EUR 250 m) was registered in 2012, which decreased by about 40% in 2013 and by another 50% in 2014. The significant number of procedures launched under the new pre-bankruptcy legislation resulted in a lagged rise in bankruptcies – which may drive up write-offs in the near future.

External financing has been an important source of financing for domestic companies during the crisis. Although at a more moderate pace than in the past, Croatian private corporations continued to increase their foreign debt exposure. In the third quarter 2014, this was almost EUR 11.2 bn as compared with EUR 10.5 bn at the end of 2011. The increase in total foreign debt was partially off-set by the conversion of cross border loans from mother companies into equity. In the meantime, however, SOEs strongly reduced their foreign indebtedness (almost EUR 1 bn over the same period).

Graph 2.4.3:   Corporate foreign debt

Source: Central National Bank, European Commission

Overall aggregate deleveraging pressures in the corporate sector are likely to be contained, but a broader capital re-allocation process has still to take place. As a result of the above described trends, by the third quarter of 2014 the non-consolidated debt-to-GDP ratio stabilised at about 89%. Net of this effect, the reduction in the aggregate debt levels was less intense than in other peer economies, but it took place in the context of falling economic activity. Active deleveraging through negative credit flows was partly offset by the shrinking economy and the subdued price dynamics. More recently changes in aggregate corporate debt were accompanied by noticeable composition effects – especially between private and public enterprises. This process is nevertheless still in its infancy and, beyond a strong deleveraging in the construction sector, there is limited evidence of a broader cross-sector re-allocation of capital.

A smoother process of capital reallocation would be supported by more efficient restructuring of businesses undergoing difficulties. In legal terms, Croatia’s pre-insolvency framework is among the least efficient in the EU. Putting in place an efficient and transparent early-rescue and insolvency framework is a key prerequisite for alleviating the considerable deleveraging pressure faced by Croatian businesses and fostering a culture of early restructuring and ’second chance’. The current pre-insolvency framework, established in October 2012, has introduced a semi-formal restructuring procedure so as to allow debtors to address their financial difficulties with a low degree of judicial intervention, coupled with an over-reliance on intervention by administrative bodies. An assessment of the pre-insolvency legislation in EU Member States against international best practices shows that the ex ante efficiency of the current legislative framework is among the lowest in the EU ([44]).

While the legislative framework puts relatively few constraints on the operations of companies undergoing restructuring, access to effective restructuring is very limited. The framework performs well, for instance, as regards providing debtors relatively long moratoria against creditors’ action, which helps achieve balance between the parties’ interests. Also, the administration of companies remains in the hands of the entrepreneur rather than an appointed commissioner. The framework falls short of best practice in most other dimensions, however. Most importantly, restructuring possibilities are rather difficult to access. The conditions for initiating the pre-bankruptcy procedure are excessively strict (it starts when the debtor is practically already insolvent) and the legislative framework does not offer alternative pre-insolvency procedures to cater for different needs and situations. The framework also lacks early-warning tools for SMEs that would help them identify financial distress before actually becoming insolvent.

 Box 2.4.1: Corporate debt in private and publicly owned companies This box discusses the indebtedness in private and publicly owned non-financial corporations. The results on the distribution of corporate debt are based on a firm-level dataset from Bureau Van Dijk’s Orbis database. The data refer to the 2013 fiscal year, which on the date of the download (December 2014) were available in Orbis for a large majority of firms, but not for all. Subsidiaries of resident companies with consolidated financials were excluded to avoid double-counting. Firms operating in finance and insurance, public administration, health and social services, and education, were excluded. Debt is defined as the sum of loans and non-current liabilities. Capital employed is the sum of debt and equity. Earnings before interest, taxes, depreciation and amortisation (EBITDA) are directly taken from the database. The thresholds for debt/capital employed (70 % and 90 %) and debt/EBITDA (6x and 12x) are approximately equal to the 75th and 90th percentile across the pooled sample of firms from 15 EU countries (which include vulnerable and core countries). Reported figures represent the share of debt held by firms in a given solvency bucket, as a percentage of the total amount of debt. The definition of publicly owned enterprises is based on the analysis already performed in section 3.1 with respect to productivity. Publicly owned enterprises are defined as enterprises where the share of public ownership is 25 % or more. Note that the comparison with national accounts is not straightforward due to statistical coverage (micro enterprises are not included in the firm-level data), different definition of variables and ultimately the fact that some SOEs are partly classified in the general government sector and partly under non-financial corporations. However, according to 2013 data, corporate liabilities in national accounts totalled some EUR 38 bn as opposed to EUR 34 bn held by firms in the sample – which suggests a broadly comparable order of magnitude. The analysis of micro data allows us to further qualify the analysis of the current section which is essentially based on aggregated data. Firstly, the disaggregated analysis confirms that debt is highly concentrated in the public owned enterprises. Almost 30% of total debt in the sample - i.e. around EUR 10 bn – is held by publicly owned enterprises. At micro level, leverage ratios can be assessed against two benchmarks: the stock of capital employed and the flow of earnings before tax. The lower the capital employed and the lower the flow of earnings, the higher the debt exposure of a firm, due to limited capacity to sell capital or rely on earnings to face its financial obligations. In the following tables, the total stock of debt is broken down by its risk profile and by ownership. The tables report percentages of the total debt. For example, first table on the left shows that more than a third of total corporate debt can be considered as having a high risk profile since companies holding the debt have either high debt to capital ratios or high debt to earnings ratios, or both. Specifically, 26.6% of the debt is held by companies with both high capital and earnings to debt ratios, 6.4% of the debt is held by companies with a high earnings to debt ratio, but medium capital to debt ratio and 2.7% of debt is held by companies with high debt to capital ratios but medium earnings to debt ratios. Source: European Commission, ORBIS The breakdown by ownership shows that private corporations are more polarised: there is a higher proportion of debt held by weakly leveraged companies (28%), but also a higher proportion of debt held by highly indebted companies (about 45%). The situation in public corporations is mixed. While only about 13% of  debt can be classified as being

(Continued on the next page)

Box (continued)

held by highly leveraged companies, only slightly more than 10% of debt is held by financially sound corporations – i.e. corporations with a low debt-to-earnings and debt-to-capital ratios. What is striking, in the case of public owned enterprises, is the concentration of debt (46.1%) in corporations having relatively low debt-to-capital ratio, but a high debt-to-earnings ratio. The reason for the high capital-to-debt ratio of public companies becomes clear from the graphs below: medium leveraged public companies holding the overwhelming majority of total public company debt operate in two sectors: construction and energy and utilities. Private corporate debt, on the other hand, is more evenly spread across sectors. Distribution of debt – private corporations || Distribution of debt – publicly owned corporations

||

Note: Debt is coloured in red, yellow and green according to the debt to capital and debt to earnings ratios defined in the tables above.

Source: European Commission, ORBIS

The concentration of debt in a few sectors is a source of financial and ultimately fiscal risks, though the latter are mitigated by the quasi-monopoly enjoyed by utilities and public construction companies operate. An additional source of risk stems from the fact that in both the utility and the construction sector, debt is held by only a handful of companies: i.e. just over 40 utility companies, and about 20 construction companies. Within both groups, moreover, an even smaller group of firms hold the overwhelming majority of debt.

This high concentration of debt makes public companies 'too big to fail'. Vulnerabilities are therefore likely to be transferred to consumers through higher fees (especially in utilities) or to all the taxpayers.  By contrast, debt in the private sector is distributed among a larger number of companies – so insolvency risks are somewhat less concentrated. An assessment of concentration of this risky private debt between main creditors would be useful, but is beyond the scope of this box.

The above analysis highlights another dimension in which public enterprises may act as a drag on the whole of the economy. This underlines the need to address their governance framework, as discussed in the previous section. 

An ex post assessment of concluded pre-insolvency cases confirms a number of weaknesses in the practical application of the framework. According to a recent analysis, the total value of pre-bankruptcy settlements concluded by March 2014 reached nearly 8% of GDP ([45]). The analysis points out that the framework does not facilitate sustainable deleveraging, as firms undergoing restructuring under the pre-bankruptcy framework typically failed to achieve the planned decrease in the debt-to-asset ratio. Overall, restructuring plans proved to be unrealistic. For instance, 60% of debtors failed (by a wide margin) to meet the expected revenue targets. The authors found that the claims of public institutions and SOEs were written off to the greatest extent, which (together with the inherent conflict of interest due to the Ministry of Finance being the largest creditor and simultaneously the administrator of the procedure) weighs on the transparency of the process. Court data for 2014, however, suggest some improvement in implementing the law as compared with 2013. While the number of new pre-insolvency cases remained broadly stable, the number of resolved cases increased considerably; this resulted in a slight reduction in the backlog ([46]).

The draft Insolvency Act is a step in the right direction. The draft Insolvency Act, which incorporates the pre-insolvency and insolvency procedures in a common framework, was adopted by the government in December 2014 and is due to be adopted by parliament in the first quarter of 2015. Its objective is to encourage viable businesses to restructure at an early stage so as to prevent insolvency and thus to contribute to improving Croatia’s business environment and stimulate the investment necessary for growth to resume. Specifically, the reform aims to facilitate earlier access to the procedure when there is a risk of insolvency by making the insolvency/illiquidity test somewhat less strict than in the current law. The reform also gives a stronger role to commercial courts as regards verifying the reported claims and validating restructuring plans that appear to undermine the efficiency and credibility of the current arrangements. It is, however, uncertain whether this change will result in tangible improvements in view of the limited resources (and hence efficiency) of commercial. The reform is likely to provide for stricter deadlines and therefore lead to a reduction in the length of proceedings. The procedure also limits to 10 % the size of the ʻhaircut’ that dissenting creditors (not accepting the settlement) may be forced to accept.

However, a number of concerns remain. First, it remains to be seen whether the 10 % limit strikes the right balance between the protection of creditors and the efficiency of the deleveraging process. Secondly, the stay on individual enforcement actions has no bearing on the secured creditor, who will be able to enforce their security during the procedure. Thirdly, the classification of creditors does not seem to be appropriate for the purposes of voting on a restructuring plan. Last but not least, the effectiveness of the procedure in practice will depend on pro-active involvement by the commercial courts, which pre-supposes that they are suitably equipped, in particular as regards training for judges and court staff on the new provisions in the Insolvency Act, and the business processes in courts (including case management) in the light of the very tight deadlines for several procedural steps.

The draft Insolvency Act also aims to streamline the insolvency/liquidation process. The World Bank 2015 Doing Business report points to considerable weaknesses in the insolvency framework such as a very low recovery rate, and high cost and length of the procedure. The main new elements in the Insolvency Act are limiting the effects of parallel cases on the insolvency proceeding and streamlining the process of asset liquidation, which could help to shorten the lengthy procedure. The inclusion of a provision requiring courts to appoint insolvency practitioners randomly from a list is still under discussion with local stakeholders ([47]).

Household sector debt

In the years of the construction boom, real estate prices generated expectations of future capital gains and fuelled a sharp increase in households’ debt. The housing and construction bubble fed a sustained demand for mortgage loans. Rapidly increasing demand was met by relatively loose lending policies on the part of banks. The growing indebtedness also led to increasing foreign exchange risk exposure, due to the widespread practice of banks to extend loans indexed to euros and, to a lesser extent, to Swiss Francs, to unhedged households (essentially in view of matching the currency structure of domestic deposits also denominated in foreign currencies).

Households deleveraging started somewhat earlier than in the corporate sector, but progressed slowly on the back of high loan maturity and an increasing debt repayment burden. Households started to adjust sooner (in 2009), probably on the back of falling property prices, when credit growth rates adjusted sharply. Deleveraging was however partially offset by adverse foreign exchange dynamics, and the debt-to- GDP ratio continued to rise until 2010, when it reached 41.8% of GDP. As about almost half of the loans were contracted for real-estate purchases, the average duration of the stock of outstanding household debt is still high. The sharp deterioration of the labour market and falling disposable income have hindered a swift re-payment of household debt and increased the burden of debt repayment. Yet active deleveraging (i.e. though negative credit growth) has intensified since the second half of 2012.

In the absence of personal debt foreclosure legislation, write off in the household sector has been so far limited. Household debt write off reached a peak of HRK 525 m in 2013 (about EUR 70 m) and was down again at about HRK 290 m (i.e. about EUR 38 m) in 2014. In the absence of a mechanism for dealing with personal insolvency, the government is increasingly recurring to ad-hoc and temporary measures. A new measure currently discussed by the government is likely to increase somewhat the writing-off of non‑mortgage loans by poor households. The measure is likely to have a broader impact on NFCs (mainly utilities) than on the banking sector ([48]). The recent decision of Swiss authorities to let the Swiss Franc appreciate against the Euro was set to further increase the debt burden of households who had contracted Swiss franc indexed loans. Before the recent appreciation, Swiss franc indexed loans stood at about 16.2% of total outstanding loans to households – i.e. about HRK 20.5 bn or EUR 2.7 bn. The government intervention to freeze the exchange rate on loans to pre-appreciation levels has shifted the burden on to the banking sector.

Graph 2.4.4:   Debt-to-GDP ratio - Households sector

Source: European Commission

The recovering labour and housing market are set to ease the household sector’s residual deleveraging pressures. In a recent debt sustainability analysis, the CNB has concluded that although Croatian households face residual deleveraging pressures, further debt reductions should be relatively contained. The study looks in particular at the dynamics of debt vis-à-vis underlying macroeconomic variables such as the unemployment rate, GDP growth, interest rates and real-estate prices. Going forward, improving employment and stabilising real estate prices are set to ease residual deleveraging needs in the household sector.

The introduction of a personal insolvency procedure would allow a more orderly write-off of debt by insolvent households, limiting the reliance on one-off and temporary measures. There is currently no bankruptcy procedure available for consumers in Croatia. Against the backdrop of the Swiss franc crisis and other one-off measures currently planned by the authorities, however, establishing such a procedure is the best way of addressing consumers’ inability to service their debts without jeopardising the stability of the financial system or consumer confidence in the financial market. The authorities plan to enhance the early rescue/insolvency framework further by introducing personal insolvency to allow natural persons and micro-enterprises to discharge their debts.

Graph 2.4.5:   Domestic credit institutions’ claims on households by type and currency

Source: CNB, European Commission

Financial sector stability

Conservative macro-prudential policies have bolstered the stability of the Croatian financial system. The structure of private sector debt in Croatia, with a high proportion of domestic bank lending directed to the household sector and a large proportion of corporate lending in the form of cross-border loans by foreign parent banks, can be explained by the conservative macro-prudential supervision by CNB. Against the background of a high share of assets and liabilities denominated or indexed to the euro (i.e. the high ʻeuro-isation’ of the economy), and the constraints this implies for the CNB’s role as a lender of last resort, the authorities adopted macro-prudential measures in the previous ʻgood times’ to reduce the risks associated with volatile capital flows ([49]). Some of the macro-prudential measures have since been relaxed in the course of the crisis with a view to attracting capital inflows and stimulate bank lending, but together with high capital requirements, they have helped to strengthen the resilience of the banking sector’s resilience.

Given the high share of euro-denominated domestic and foreign lending, a broadly stable exchange rate of the kuna against the euro is important from a financial stability perspective. While direct balance-sheet risks to the banking sector are limited as most of the banks’ liabilities are also denominated in euro, lenders are still subject to exchange-rate-induced indirect credit risk. A depreciation of the kuna against the euro would increase the real debt burden of households and corporates whose income is denominated in kuna, which would adversely impact on banks’ asset quality. In the past few years, the kuna has remained broadly stable against the euro, fluctuating between HRK 7.4 and 7.7/EUR. By contrast, the kuna has depreciated by about 15% against the Swiss franc during the second half of January 2015, following the decision of the Swiss National Bank to discontinue the exchange rate ceiling against the euro. International reserves (12.7 bn in December 2014, covering around 8 months of imports and roughly 100% of short-term debt) appear adequate, taking into account the high share of external debt held by the non-financial corporate sector and banks’ own foreign currency reserves ([50]).

The recent reviews of banking sector portfolios have confirmed the picture of a stable financial sector, albeit one facing challenges from a weak economic environment. The asset quality review (AQR) under the single supervisory mechanism (SSM) has led to a relatively small revision of the NPL, coverage and capital‑adequacy ratios of the four Croatian credit institutions included in the exercise ([51]). As a result of the SSM AQR’s additional impairments, their December 2013 aggregate consolidated capital-adequacy ratio decreases from 20.8% to 20.2%, which is still significantly above the legally prescribed minimum and regulatory requirements of the CNB ([52]). In line with the country-specific recommendations for Croatia under the European Semester, the CNB has also undertaken a portfolio screening exercise (PSE) as a complementary activity to the SSM AQR. The PSE looks at: (i) important portfolios in the largest banks that were excluded from the AQR because they were not considered to be important from a banking-group perspective; and (ii) extends the exercise to smaller and mid-size banks as well ([53]).  The results appear to be rather comforting for the large banks that are part of international groups, but point to some weaknesses in risk management and governance for a small number of mid-size and smaller banks.

Despite the relative strength of most domestic credit institutions, the operating environment remains challenging, notably due to increased government interference in the sector, such as the recent exceptional measures taken by the government to address the appreciation of the Swiss franc. The Swiss national bank’s decision on 15 January 2015 to scrap the floor on the CHF/EUR exchange rate resulted in an immediate appreciation of the franc against the euro and the kuna (from HRK 6.39 on 15 January 2015 to HRK 7.52 on 28 January 2015), directly impacting the instalments due on the HRK 22bn CHF-indexed assets (of which approximately 80% are housing loans) held by Croatian banks. On 23 January 2015 the Croatian Parliament adopted a law to freeze the exchange rate at HRK 6.39/CHF for one year in loan contracts, with the costs (around HRK 400mn end of January) to be borne by the banks. This law goes significantly beyond what had been proposed by the banking association, i.e. a voluntary three-month freeze for clients experiencing difficulties and a commitment to restructure mortgages for clients facing the most severe payment difficulties as a result of the franc appreciation. Should the government decide to extend the freeze until the maturity of bank’s total franc portfolio, banks’ losses could reach HRK 3.5bn according to CNB’s current estimate and current FX market conditions. The risks associated with unanticipated exchange rate movements for borrowers, especially households, with large unhedged positions in a foreign currency have been increasingly stressed by European institutions in recent years. However, the Croatian authorities did not reduce the FX risk faced by households before the risk actually materialised. Instead, they decided to assign the whole loss to the banks ex post. While it is legitimate to find solutions for borrowers who are in payment difficulties, any remedy should be proportional, equitable and have a solid legal basis with a focus on helping generally distressed borrowers. This principle should also be borne in mind when designing the permanent solution that will replace this temporary exchange-rate freeze.. Regarding the legal basis, it should be noted that state intervention in existing loan contracts can seriously undermine legal certainty and send a negative signal to investors. Regarding proportionality and equity, targeted relief towards more vulnerable borrowers and those whose payment capacities have clearly deteriorated would be the most appropriate way forward. This can avoid the pitfalls of blanket, one-size-fits-all approaches, which may disproportionately benefit higher-income households and those with properties purchased for investment purposes.

The various measures governing the setting of interest rates, and aimed at protecting households and small stakeholders, might impact banks’ profitability, even if some measures may not be binding in the current environment of low demand for credit. These existing measures on bank interest rates might also introduce market rigidities and reduce competition The government has been considering introducing a new administrative measure for consumer protection limiting further the setting of interest rates by lowering penalty interest rates on overdue loans. There is however a risk that any such measure would make it unattractive for banks to lend to some customers, including potentially innovative SMEs that can make a valuable contribution to Croatia’s recovery. The government’s intention to protect borrowers from unfair practices by banks could be achieved by fully implementing existing EU legislation in the field of consumer protection (e.g. by increasing products’ transparency and consumers’ information) instead of regulating interest rates, while ensuring that households and other weak stakeholders are adequately protected. It is also necessary to assess, in this context, the degree to which this new law could overlap with existing regulations (Consumer Credit Act and Financial Operation and Pre-Bankruptcy Settlement Act). A significant overlap could be confusing for economic agents, and could undermine legal certainty, and, therefore, discourage investment.

Croatia’s largest banks report strong capital ratios compared to banks in neighbouring countries, but the sector remains vulnerable, due to the country’s fragile macroeconomic outlook. In September 2014 the NPL ratio had increased to 17.2%, from 15.3% a year earlier. The highly-indebted corporate sector (the corporate NPL ratio stands at 30.6%) represents a major source of risk for the financial system, despite the significant level of external debt and debt covered by state contingent liabilities. The introduction of the pre-bankruptcy settlement at the end of 2012, revealed hidden illiquid and insolvent non-financial corporations in 2013 and 2014. The household sector, in volume terms the largest beneficiary of Croatian banks loans (representing 44% of total loans granted domestically), has also reached a relatively high NPL ratio of 12%. The coverage of NPLs with provisions, which used to be below the levels of peer countries, has increased rapidly over the last two years, thanks to new measures introduced by the CNB, in addition to international financial reporting standards (IFRS) provisioning rules. In June 2014 the provision coverage ratio stood at 50.3%, as compared with 44.2% a year before, and is now above the EU average (46%). High NPL ratios are weighing on banks’ profitability may hinder needed corporate restructuring in the Croatian economy. Furthermore, it cannot be excluded that the quality of banks’ assets will continue to deteriorate in 2015, even with a return to growth. Despite these concerns, Croatia’s banking sector has been resilient up to now, and overall the sector remains well capitalized, as confirmed by the results of the comprehensive assessment under the SSM and CNB’s additional PSE.

Graph 2.4.6:   Financial sector stability and profitability

Source: European Commission

There is still uncertainty surrounding the Hvratska Postanska Banka, the publicly owned Croatian postal savings bank. Hvratska Postanska Banka (HPB) is the seventh largest bank in Croatia and one of two publicly owned banks, with approximately 5% of banking assets, and the government plans to privatise it (a previous privatisation attempt failed in 2013, as the government did not find the sole bid to be acceptable). The CNB recently ordered HPB to increase provisions on legacy bad loans made under a previous management team. As a result of this increase in provisions, HPB made a loss for the first nine months of 2014 and fell below minimum regulatory capital levels. To begin to remedy this situation, the new management had planned to issue a EUR 60 m subordinated bond in December 2014, but these plans had to be abandoned due to lack of investor interest.

There is scope for improving the supervision of the Croatian Bank for Reconstruction and Development and enhancing the accountability of its management in order to strengthen its capacity to support Croatia’s recovery. The mission of the Croatian Bank for Reconstruction and Development (HBOR) is to support exports, finance infrastructure and provide credit to SMEs. It is 100% state-owned and benefits from the government’s unconditional, explicit and irrevocable guarantee, which allows it to raise funds on favourable terms. Credit facilities for SMEs are provided either through loans via commercial banks or, increasingly, through direct lending. While, in principle, HBOR does not assume any direct credit risk when on-lending via commercial banks ([54]), it has to manage borrowers’ credit risk fully when lending directly. HBOR’s direct lending has been increasing recently (up from 24% of total loans in 2011 to 44% in 2013 based on HBOR data), due to commercial banks unwillingness to increase their exposure to riskier borrowers as a result of the crisis. Despite engaging in significant direct lending activity, HBOR is not subject to any kind of prudential supervision by the CNB. This deficiency means that Croatia does not apply best practice for national promotional banks in the EU. In Germany, for example, KfW is subject to prudential supervision by the German supervisory authorities (BaFin and the Bundesbank) for both its direct lending and its on-lending activities. The draft HBOR Act currently before the Croatian Parliament would be the ideal opportunity to remedy this. Furthermore, in order to be able to fulfil its mandate over the medium term, and also given the non-negligible contingent liability it represents to public finances, it is important that HBOR operations are based on strong corporate governance, operational independence and a high degree of transparency. More structured and transparent accountability arrangements are needed, therefore, e.g. with respect to the fitness and probity of HBOR management board members. Again, the draft HBOR Act would be the ideal opportunity to strengthen HBOR’s governance.

The domestic banking system probably has limited capacity to absorb additional public debt. Croatian banks’ lending to the general government has been on a slightly downward trend since the second quarter of 2013, with a temporary drop in September 2014. The total exposure of banks to the sovereign is 24% of GDP or 18% of total bank assets (from 17% one year earlier). Such a level seems to be close to the internal exposure limit set by the largest banks (which are also the most exposed). Thus, the banking sector may not have strong capacity to absorb much more Croatian government debt. Additionally, there is also the loan exposure to SOEs, which accounts for 22% of domestic corporate lending. Most of SOE loans enjoy a state guarantee (72%, based on CNB data from end 2013) which adds to the broad government exposure as a contingent liability. The national financial system is also exposed to the central government through the insurance companies and the pension funds, which typically invest most of their funds in government securities. In Croatia, they jointly hold in their investment portfolios around HRK 55.9 bn in sovereign bonds, corresponding to 17% of GDP.

The future capital treatment of EUR-denominated Croatian sovereign bonds will have an impact on the life insurance sector. A specific feature of Croatian life insurers is their high exposure to EUR-denominated Croatian sovereign bonds (HRK 9.7 bn, corresponding to 34% of total investments), because a large part of their liabilities is denominated in EUR. Unlike sovereign bonds in domestic currency, sovereign bonds in foreign currency are not considered as risk-free by the delegated acts of Solvency II (entry into force: 1 January 2016), which is consistent with their treatment in the banking Capital Requirements Regulation (CRR). The impact of these changes needs to be monitored.

In conclusion, deleveraging in previous sector has progressed, though at a slow pace. The reclassification of two large SOEs into the general government sector with the switchover to ESA 2010 already brought about a significant correction in corporate leverage ratios. The micro data analysis shows that a large stock of debt is still held by a handful of weakly profitable publicly owned enterprises concentrated in the utility sector and in the construction sector. Nevertheless leverage ratios in private corporations remain also substantial, though less concentrated. Moreover a pick-up of economic activity and turnaround in profitability is set to lead to ease deleverage pressure. Household deleveraging also progressed despite the adverse macroeconomic conditions and the situation on the labour market.

Despite being dominated by large, foreign-owned and well capitalised banks, Croatia’s financial sector faces a number of challenges that still need to be addressed in order for the financial sector to play its full role in supporting the recovery. The operating environment is marked by an increased amount of state intervention, with, most recently, the recent freezing of the exchange rate of the kuna against the Swiss franc in loan contracts in response to the sudden appreciation of the latter, and the various administrative measures governing the setting of interest rates. Supervisory diagnostic exercises seem to confirm the high level of capital of the largest banks, but have revealed some weaknesses in governance and risk management practices in a small number of mid-size and smaller institutions. The high level of non-performing loans, especially for corporate borrowers, also remains a major concern, as does the banking sector’s high overall exposure to the state (both government debt and loans to state-owned enterprises). HBOR can play a potentially useful role in the current environment by ensuring that creditworthy small and medium-sized enterprises have adequate access to credit, but its governance and supervisory arrangements are not yet in line with European best practice.

Weak labour market performance continues to be a challenge. As discussed in section 2.1 a number of structural weaknesses and the negative macroeconomic environment weigh on employment. Croatia is facing one of the lowest labour market participation rates in the EU, particularly affecting young people, the low-skilled, older workers and women. Low activity rates amongst older workers reflect relatively generous exit paths from the labour market. Low female participation is affected also by limited access to care structures. This is against a backdrop of high unemployment, including high and rising long-term unemployment. The unemployment rate is also higher for the population groups showing weaker activity rates, which suggests that inactivity partly reflects discouragement effects and insufficient investments in active labour market policies.

Graph 3.1.1:   Labour market and social indicators

Source: European Commission

Social protection

Besides high unemployment and low labour market participation, the inadequacy of social benefits and services contribute to high levels of poverty and social exclusion. While in-work poverty in Croatia is low and the minimum wage is above the poverty threshold, older persons, the unemployed, households with dependent children and households with low work intensity as well as disabled persons remains particularly vulnerable.

Specifically, at 29.9 % in 2013 the proportion of the population at risk of poverty and social exclusion (AROPE) remained well above the EU average. The rate decreased somewhat from 32.3 % in 2012. However, the decrease in the share of people at risk of poverty is partly explained by a decrease in median incomes and hence a lower poverty threshold. The proportion of the population at risk of poverty (AROP) has been trending at around 20%.

The enduring crisis is putting pressure on social expenditure. The already low total social protection expenditure in Croatia has further decreased, while the social benefits reform aiming at improving effectiveness and targeting of social assistance is progressing only slowly. In particular, the total social protection expenditure in Croatia continued to fall from 21.2 % of GDP in 2012 to 18.2 % of GDP in 2013 which is well below the EU average (29 % in 2012). Non-means-tested social protection benefits (cash and non-cash) accounted for 93.1 % of all social protection benefits, which may mean that there is scope for better targeting spending to those most in need.

The overall labour market and social situation is aggravated by significant regional disparities. These result in lower income and education levels, poorer housing conditions and lower living standards in disadvantaged and remote areas. Social protection benefits provided by local communities vary significantly according to the resources available. The introduction of the guaranteed minimum standard (GMS) is aimed at tackling the differences in standards of living based on regional variation. However, the consolidation of social benefits is planned for 2017 only and local authorities will continue to play an important role in providing benefits not included in the unified GMS. Also, substantial local, regional and categorical benefits are not covered by the established ‘one-stop shop’. January 2015 saw the entry into force of the Regional Development Act, which established a regional development index to serve as the basis for classifying regional and local governments and preparing the regional development strategy. No timescale has been given for this, however.

The Croatian authorities are taking measures to improve the targeting of social welfare programmes. To improve targeting, Croatia will implement a project, ‘Strengthening institutional capacity in the social welfare system to improve social welfare targeting and reduce poverty’, aimed at strengthening the administrative capacities of social services to increase social inclusion and tackle poverty, particularly for families and children. As of 2014/15, one year of mandatory pre-schooling was introduced, but early childhood care has not been tackled. With the introduction of the Guaranteed Minimum Benefit the benefits for households with three or more children decreased. However, early childhood services remain underdeveloped.

Improving employment service capacities and active labour market policies

Despite real increases in the scope of, and expenditure on, active labour market policies (ALMPs), they remain insufficiently targeted, in particular as regards outreach to older persons and the low‑skilled. Participation in ALMP measures has increased in recent years, but 2014 saw a considerable fall in the number of new participants as compared with 2013. The level of ALMP expenditure increased, but remains quite low and represented only 0.24 % of GDP by October 2014. National ALMP expenditure for 2015 stands at 0.3 % of GDP (including training allowances), or 0.5 % of GDP when the ESF/Youth Employment Initiative pipeline is taken into account. The Croatian Employment Service (CES) plans to commission an external evaluation of ALMP measures in 2010-13, with a view to improving the effectiveness and quality of services and the measures themselves. It is expected that the outcomes of the evaluation will be fed into preparations for the new employment support programmes. 

Guidelines have been prepared for the development and implementation of ALMPs in 2015-17. An interdepartmental working group has set implementation, monitoring and reporting standards, and a methodology for the development of annual plans for all institutions involved in providing ALMPs. CES’ weak administrative capacities and business processes hinder its capacity to improve the effectiveness of ALMP measures and enhance outreach to the groups most vulnerable on the labour market. The restructuring and reorganisation of CES is under way, the aim being to strengthen administrative capacities, increase efficiency and offer new services. New career guidance centres (CISOK) are being established to provide tailor-made services, counselling and support in life‑long career guidance, focusing particularly on young people, including NEETs. The ‘forum for lifelong learning and career guidance’ was established in September 2014 as a national coordination body to promote, coordinate and improve lifelong professional guidance and to prepare a strategy for lifelong learning and career guidance, expected in October 2015. In addition, an advanced labour market information system (ALMIS) is being put in place, to serve a wide range of stakeholders (CISOK centres, institutes, schools and universities) as a vocational guidance and career development tool (using register data) and provide indicators on occupations and the labour‑market situation and trends. To promote participation in lifelong learning, a system for recognising and validating non-formal and informal learning will be developed in early 2015. In addition, training programmes for special counsellors will be carried out and a pilot CES portal for career development has been established to encompass employment and education attainment indicators. A new categorisation of clients will be established, along with a new breakdown of services provided (to facilitate the provision of tailor-made services). Also, a new pilot project tracking the employment of graduates is geared to the networking of data on graduates with the CES and the Croatian Institute for Pension Insurance.

Young people are particularly vulnerable on the labour market. Youth unemployment has been increasing since 2008 and reached 50 % in 2013, half of whom were long-term unemployed. However, youth employment increased to 21.3 %, resulting in youth unemployment falling substantially, to 41.4 % in Q3-2014. Despite wide quarterly fluctuations, this could be partially attributed to the positive effects of stronger ALMP measures targeting young people in recent years (in 2014, young people represented 67 % of all new ALMP participants). The high and increasing number of NEETs remains a challenge in Croatia (22.3 % in 2013 as compared with an EU average of 15.9 %) and a tracking or outreach mechanism has yet to be put in place. Overall, Croatia took steps to address the situation of young people on the labour market. However, additional measures are necessary, particularly as regards matching skills and education to labour market needs, promoting apprenticeships, strengthening CES capacities and better outreach to non-registered NEETs, in line with the objectives of the Youth Guarantee.

Croatia has been taking a range of measures to address youth unemployment, some of which are part of the Youth Guarantee Implementation Plan. The YGIP Council was established in September 2014 to implement and monitor the YGIP; a first progress report is expected in January 2015. In total, 11 new ALMP measures were set up in 2014 and grouped under the ʻYoung and Creative’ package, which now consists of 23 measures aimed at facilitating youth employment, including employment and self‑employment subsidies, training and specialisation subsidies, traineeships for work without employment, community service with an emphasis on added value, and job preservation.

ALMP measures providing young people with work experience have been at the forefront. The main measure, ʻoccupational training without commencing employment’, funds work experience and is aimed at improving school‑to‑work transitions. In 2014, 47.5 % of all new ALMP participants were covered by this measure. In September 2014, the CES lowered the eligibility criteria for participants and their financial allowance was raised from HRK 1 600 (EUR 208) to HRK 2 400 (EUR 312) as of January 2015. The long duration of the measure and the recently removed requirement for private‑sector employers to offer employment to the participants draws attention to potential misuse. Amendments to the Act on Contributions that entered into force on 28 November 2014 were aimed inter alia at preventing labour‑market segmentation of young people. Employers employing a young person under a permanent contract were exempted from paying contributions (17.2 % of the wage) for five years (see also section 2.3 on taxation). Overall, partly in view of the length of this time period, the efficiency of this measure draws upon the need for its close monitoring. The Workplace Training initiative to be rolled out as of March 2015 aims at placing 500 young unemployed individuals in the working environment for a six-month period. Both programme participants and employers will be entitled to a financial subvention during the training period, and the young people will remain on the unemployment register.

There are plans to develop a NEETs tracking system in 2015 to address increases in the numbers of NEETs. This will be part of a comprehensive human resources register, which will provide information on individual education and career development. Once a database of matching sources and data has been established, the NEETs group will be analysed and the results used to inform and shape outreach activities. First results are not expected until March 2015, but this seems a step in the right direction.

Pensions and older workers’ participation

Croatia is facing considerable demographic change, but pension expenditure is projected to decrease in the long run. By 2035, the labour force aged 15-65 is projected to decrease by 7 %, while the number of people over 65 is expected to rise by 29 %. Consequently, the ratio of the population over 65 to the working-age population (the old-age dependency ratio) is expected to rise from 27 % today to 42 %, in line with developments in the EU-28 as a whole. These trends are set to continue thereafter, although less dramatically. Despite the demographic pressure, pension expenditure was projected to decrease over the long term in national pre-accession projections. This is likely to be confirmed in forthcoming pension projections for the 2015 Ageing Report.

Graph 3.1.2:   Population structure

(1) Old-age dependency ratio is the share of population aged 65 and over in the population aged 15-64. Source: The 2015 Ageing Report, Underlying Assumptions and Projection Methodologies, European Economy 8, 2014

The projected decrease in public expenditure is largely a result of the low valorisation of pension rights and the anticipated decrease in the level of new public pensions. A number of factors help explain the projected decrease in public expenditure. First, under the current valorisation rule, the value of accumulated pension rights is eroded over time (relative to wages) because the valorisation rate is lower than the projected rate of wage growth. Secondly, the gradual maturing of the compulsory second pillar will partially alleviate the pressure on public finances, as more people will receive part of their pensions from elsewhere. Importantly, as second-pillar pensioners are, under current legislation, not entitled to receive the 27 % pension supplement to their first-pillar pension, the level of individual pensions paid out from the first pillar is set to fall significantly ([55]). Other factors contributing to lower public spending include a smaller proportion of new disability and war-veteran pensions and a gradual rise in the retirement age.

In the context of a rapidly ageing population, the planned decrease in pension levels gives rise to strong adequacy concerns and makes the pension system vulnerable to policy reversals. The benefit ratio, i.e. the average pension benefit divided by an economy-wide average wage, currently stands at 31 %, which is one of the lowest in the EU. This is also reflected in a high proportion of people aged 65+ living in poverty or social exclusion (31.9 % in 2013, as compared with 18.3 % in the EU-28 as a whole). The low valorisation of past earnings entering the pension formula, the planned phasing-out of the pension supplement and the expected changes in the composition of the pensioned population are set to reduce the benefit ratio to 27 % in 2030 (including the second-pillar benefit), and further to 22 % in 2060. Future pension benefits would thus be on average nearly 30 % lower than today, which, considering their already low current level and the widespread old-age poverty, represents a considerable challenge. It also increases the likelihood of policy reversals, which expands the margins of uncertainty around the current estimates of underlying fiscal pressure.

Some expenditure-increasing policy changes are already envisaged. The pension supplement is due to be extended to second-pillar pensioners, but only for qualifying periods prior to the introduction of the second pillar in 2002. These changes are not factored in in the latest projections, which again increases the margin of uncertainty around them. Given the introduction of the second pillar, it is unlikely that reversing the expected decrease in the benefit ratio would expose Croatia to high sustainability risks. It could, however, substantially reduce or fully absorb the currently expected decrease in pension spending. (This underlines the usefulness of projections based on alternative policy scenarios).

Graph 3.1.3:   Many Member States achieve higher pensions with lower expenditure

(1) Red markers depict countries with lower public pension expenditure but higher pensions (DE, DK, EE, ES, LT, LU, NL, RO, SE, SK), blue markers the remaining ones (AT, BE, FI, FR, IE, LV, PL, PT, UK), for which preliminary estimates indicate that data remained broadly stable in 2013. HR data are estimates for 2013. Source: 2012 Ageing Report, Ministry of Labour and Pensions.

The pension system suffers from a number of inefficiencies. These include a large number of early retirees, an excessively generous system of early pensions for arduous and hazardous professions and numerous special pension schemes. In addition, it is grappling with the consequences of the past explosion in disability retirement. These factors, which are described in greater depth below, together account for the fact that the benefit ratio provided by the Croatian pension system is lower than in most other Member States with comparable or lower pension spending (see Graph 3.1.3).

Addressing early retirement remains a challenge. Graph 3.1.4 shows the composition of the new pensioner population over time and points to a steady increase in the proportion of persons retiring before statutory retirement age, largely compensated by a decrease in the number of new disability pensioners as a result of stricter medical disability assessment. In the first nine months of 2014, over 35 % of all retirements under the general pension system involved early retirement, with ‘standard’ early pensions accounting for around three quarters of these and retirements from arduous or hazardous professions for the remainder.

Graph 3.1.4:   New pensioners (excluding special pensions)

(1)2014 is a projection based on Q1 to Q3-2014. Source: Croatian Institute for Pension Insurance

The high share of early retirement is related to the weak incentives to work until the statutory retirement age. Considering the extent of exemptions, existing penalties for early retirement and the late retirement bonus offer little incentive to work until (let alone beyond) the statutory retirement age. The average penalty for a year of early retirement in Member States that apply such penalties is above 5 %. In Croatia, the penalty is only 1.2-4.1 %, depending on the length of the contributory period. The length of the contributory period is typically not considered a relevant factor in determining the level of the penalty in other Member States. Workers with more than 41 contributing years and workers with 2 and more years of unemployment (and eligible for early retirement) are not penalised at all. Moreover, the five years gap between early and statutory retirement age remains wide compared with the EU average, decreasing labour supply and the sustainability of the pension system ([56]). These factors contribute to the very short duration of working lives (31 years as against 35 years in the EU-28 as a whole in 2013). There have been no measures taken to make early retirement less attractive.

A reform of arduous professions has been delayed and a tightening of standard early retirement is not envisaged. The list of arduous or hazardous professions currently includes more than 100 professions and, excluding employees of defence, interior and justice ministries, accounts for 2.7 % of people insured in the pension system, half of whom are heavy vehicle drivers, ships’ crew members or bus drivers. A reform of these pension entitlements (including a transfer of their pension assets to the first pillar), announced initially for December 2014, has been postponed to the third quarter of 2015.

Men’s and women’s retirement ages will not be harmonised until 2030. The current statutory retirement age is 65 for men and 61 year and three months for women. The latter is set to rise to 65 by 2030 and, following the reform that entered in force in January 2014, the retirement age for both genders will rise further to 67 by 2038. By 2020, retirement ages for men and women will be equalised in all but six Member States. Several Member States with a higher statutory retirement age for women than Croatia’s (e.g. IT, EL and UK) have progressed faster on this front. One key to reducing the pension gender gap (on average, men’s pensions are 24 % bigger than women’s) will therefore be for women to retire later and work longer. For men, the statutory retirement age will remain constant until 2030, which is not in line with the projected increase, by more than two years, in life expectancy at 65.

The number of new disability pensions has fallen recently, following the stricter checking of medical assessments. New disability pensions in 2014 accounted for around 7 % of all pensions granted under the Pension Insurance Act, down from nearly 25 % in 2008. The decrease follows the tightening of checks on medical assessments in recent years. As of January 2015, the first-instance assessment of disability claims was moved from the Croatian Pension Insurance Institute to the new Single Expert Evaluation Body, which centralises expert evaluations in several areas previously scattered among various government departments (pensions, social security, war veterans, education, and health care). The Ministry of Labour and the Pension System continues to audit every disability pension granted by Single Expert Evaluation Body.

Graph 3.1.5:   Pensioners (Q3‑2014)

Homeland war veteran pensions include pensions acquired in the general scheme, calculated according to homeland war veterans legislation Source: Croatian Institute for Pension Insurance

The pension system is fragmented into a high number of special pension schemes, which gives rise to unequal treatment and further contributes to early exits. Pensions under special schemes were paid to 14 % of all pension beneficiaries in the first nine months of 2014 (see Graph 3.1.5). The largest groups of special pensioners are war veterans and their family members. However, special pension schemes exist also for military and police personnel, authorised officials working in the judiciary, and members of the government, parliament and the Croatian Academy of Arts and Sciences. Most of these beneficiaries enjoy an above-average salary (thus also a pension), so the rationale for providing them with even higher pension levels under special rules is unclear. The system of special pensions is a source of inequality among pensioners and repeated ad hoc amendments to the pension system, most recently illustrated by the transfer to the state budget of the pension assets of active military personnel, police officers and authorised officials. A transitional cut and a freeze on the indexation of the highest privileged pensions were put in place in 2013 ([57]), but the planned harmonisation of special pension rights of military and police officers, announced for December 2014, has been postponed. Future steps to harmonise special and standard pensions will need to reflect the August 2014 ruling of the Constitutional Court, which declared unconstitutional the removal of special pension rights for MPs and high-level state officials.

The above-mentioned weaknesses erode labour market participation of older workers. The employment rate among older workers reached 37.8 % in 2013, much below the EU-28 average of 50.2 %. While the unemployment rate (9.9 %) was closer to the EU average, the long-term unemployment rate peaked at 83.5 % in 2013 (EU average 60.6 %). The slight increase in the employment rate of the population aged 55–64 observed in 2014 is largely due to the gradual increase in the statutory retirement age. Average years of service have not changed significantly since 2012, however; the figure was 30 years and 11 months in mid-2014, with an average retirement age of 64for men and 61 for women (excluding disability and special pension schemes). According to the 2014 National Reform Programme, only 12.8 % of retirees have a full career of 40 contributory years behind them.

There is no comprehensive active ageing strategy or measure for prolonging working lives, which would encourage and enable workers to stay longer in the labour market. Longer working lives are crucial to improving the adequacy of pensions and reducing the future risk of poverty in old age. The participation of the older population in lifelong learning is low (0.3 % as compared with an EU average of 5.7 % in 2013) and their participation in ALMP measures is the lowest of all age groups and has fallen in recent years. In 2014, the ALMP package for older persons (‘Experience matters’) included 12 ALMP measures, but none exclusively targeted this group. According to the draft 2014-17 guidelines for ALMP measures, one measure will exclusively target persons over 50: subsidies for the employment of older workers. Two measures have been implemented that aim to support part-time employment for pensioners.

All in all, the pension system, while financially sustainable, poses considerable social challenges and hinders efforts to extend working lives. Short working lives (made possible by generous early retirement rules) and unfavourable indexation erode pension levels and consequently heighten the risk of old-age poverty. The system provides privileged treatment to a number of specific professions, which gives rise to unequal treatment and is a source of destabilising interventions.

Care facilities

The high incidence of care in the family in Croatia may have particularly negative impacts on women’s labour-market participation. Overall, early childhood services remain inadequate and further cooperation is needed with local stakeholders to improve access. The coverage of formal childcare services for children aged between three and compulsory school age was much lower than the EU average. Low provision of childcare services may hinder female labour market participation or increase early retirement, and thus heighten the risk of poverty and social exclusion. A key challenge is the unequal access to services, with substantial regional differences in access to pre-school education. The poorest counties tend to have the lowest coverage ([58]).

There is no integrated approach to long-term care in Croatia. The fact that such services are spread between healthcare and social welfare systems results in inefficiencies and reduces their accessibility and adequacy. The social welfare legislation provides for supplements for assistance and care for elderly and frail persons, at home, in day-care and in institutional accommodation. Two non-institutional forms of care are provided to the elderly: day-care services and in-home assistance. As long-term care is split between the healthcare and social welfare systems, it is impossible to estimate the level of funds provided exactly. According to the ‘unmet needs’ indicator, Croatia faces no important access challenges; waiting lists for nursing homes remain long, however, while the services of private providers are not accessible to the majority of population. There is a considerable gap between the number of dependent people (around 300 000) and those who have actually received some kind of care (estimated at 50 000). The scale of family care for the elderly is above the EU average: Around 17 % of persons aged 35–49 reported having to care for elderly relatives at least several times a week. This is likely to negatively impact the labour-market participation of older women in particular.

There have been no significant recent changes in the area of long-term care. Care of the elderly and other persons in need of assistance for daily living is fragmented and perceived either as healthcare or as a social welfare problem. There is a scope for increasing the efficiency and adequacy of long-term care by better integrating various services and benefits, shifting from institutionalised to community-based care or improving care coordination. There are no uniform procedures for needs assessment and the focus on prevention, rehabilitation and independent living remains weak.

The health status of the Croatian population is close to the EU average and access to healthcare is improving. Life expectancy at birth is lower than in most Member States and mortality associated with cardiovascular diseases and cancer is relatively high. However, lifestyle factors (such as smoking or drinking) may have a more negative impact in Croatia than elsewhere ([59]). On the other hand, healthy life expectancy is above-average at birth and average at the age of 50. The overall accessibility of health services is reasonably good and has shown consistent improvement. Nevertheless, patients persistently report challenges with travelling distance, hinting at suboptimal care provision patterns at regional level. Access to outpatient medical goods is comparable with the rest of the EU. As regards hospital care, the total number of curative care beds is broadly average but Croatia stands out in terms of average length of stay and there are also large regional differences in bed occupancy rates. The number of general practitioners and nurses are below EU average. These indicators point to potential efficiencies that could be achieved through finding an optimal mix between inpatient care to prevention and outpatient care. An additional issue is informal payment for obtaining privileged access ([60]).

The mismatch between healthcare revenues and expenditure puts considerable pressure on the public budget, which is likely to worsen in the medium run. As argued in section 2.3, public expenditure on health is higher than in most central and eastern European Member States and the system continuously generates arrears, which require sizeable ad hoc payments from the state budget. According to the preliminary estimates for the 2015 Ageing Report, public expenditure on health is projected to increase significantly already in the medium run due to the mounting pressure of population ageing.

To prevent a further build-up of arrears, the authorities are increasing the budget allocation for hospitals. Specifically, the healthcare budget has been increased by 10%, most of which will be attributed to hospitals. The increase partially institutionalises past ad hoc injections, which should help to stabilise the system financially, but does not cover them fully. Savings on the part of hospitals will therefore be needed to avoid a further accumulation and gradual repayment of arrears. Furthermore, health contributions have been removed from the Treasury as of January 2015 and channelled to the Health Insurance Fund (HZZO). The authorities argue that this was a prerequisite to the reforms described below. It is yet to be established whether greater autonomy of the fund will help ensure financial discipline.

A number of measures are being implemented to strengthen the cost-effectiveness of the healthcare sector, including hospitals. The main measures include the introduction of a new reimbursement scheme for hospitals, the 2014-16 National Hospital Development Plan and the joint procurement project. These measures are assessed in greater detail below. In addition, the authorities are enhancing quality assurance and prevention programmes and expect to achieve savings by further strengthening referrals, sick-leave applications, e-health and primary care financing ([61]). The National Plan to Develop Human Resources in Health Care, due by May 2015, will assess the overall situation and the need for human resources development.

The authorities are putting in place new contracting arrangements for hospital care. Under the current hospital financing system, hospitals have an incentive to overspend to ensure that they meet the ex-ante spending limits. A transition to a partial retrospective payment system based on diagnosis-related groups (DRG) is scheduled for 2015, with the ex-ante component (independent of actual level of care provided) decreasing to 45 % by November 2015. The new system will include a performance-based top-up of up to 5 %.

The hospital financing reform is a step in the right direction, but its implementation may be challenging. The new system aims to channel funds to where care is provided and thus provide an incentive to rationalise existing, but underused, capacities. The authorities expect that, where the new payment mechanism (and ongoing monitoring) reveals unsustainable financial positions, the hospital management will opt for a functional integration of the units concerned with another hospital in the region or achieve savings in another way. The introduction of more competitive and performance-based contracts for hospital managers to support these changes is under discussion. However, the enforceability of changes involving reallocation of staff, adjustments in operational processes or physical capacities in the given timeframe (a year) may prove difficult, which constitutes a residual risk for the financial stability of the system.

International experience shows that introducing a DRG-based financing system and adapting it to the local specificities is a long-term process. Financing based on the DRG system in the Croatian hospitals does not fully reflect the structure of actual expenses and services, which points to a scope for improvement in the accounting and administration capacities. The incentive structure under a retrospective payment system also needs to be carefully designed so as to prevent early discharges and readmissions. A recent legislation change widens the responsibilities of the Agency for Quality Control, which is managing the currently voluntary system of quality assurance and accreditation.

The 2014-16 National Hospital Development Plan, adopted by the government in November 2014, identifies and initiates much-needed changes to the healthcare system, but may be difficult to implement fully. The Plan, of which parliamentary adoption is pending, aims to reduce the acute hospital treatment rate and increase the number of patients treated in outpatient and day care setting. As a result, it is expected to support the harmonisation and rationalisation of the bed occupancy rate, reduce the length of hospital treatment and improve access to hospital care ([62]). The implementation of the Plan is not intended to result in sizeable aggregate savings. In order to make the Plan operational, the authorities have established a reporting system (the National Registry) that keeps track of hospital capacities and will be used as a basis for contracting by the HZZO so as to meet the Plan’s objectives. Under the Plan, some units will need to become functionally integrated at regional level to meet their targets. Since hospital managers will ultimately be designing the appropriate adjustment in individual hospitals, close monitoring of the Plan’s implementation and strict enforcement will be needed ([63]). Furthermore, switching to outpatient and day care setting requires training and infrastructural adjustments in the hospitals as well as strengthened outpatient services, which could become a bottleneck for reaping full benefits from the Plan. The plan is partially supported by the ERDF and the ESF.

Joint public procurement for hospitals is already delivering savings. Central procurement was launched for 15 groups of goods and services in October 2012. Public hospitals were directed to form joint purchasing bodies for items such as medicines, medical devices and energy. A decentralised approach was adopted, whereby a number of hospitals were assigned to procure categories of goods for all participating hospitals. Hospitals that had previously achieved best value for money for certain procurement categories were selected to be the central purchasers. The reform is proving to be successful in reducing prices, achieving savings and standardising the quality of procured goods. According to the authorities, estimated savings achieved so far amount to HRK 485 million (0.15 % of GDP). However, there is scope to extend the system to other items and hospitals. The authorities expect the proportion of public spending on medical consumables, drugs, and devices for hospitals made through centralised procurement to reach 60 % in 2015, up from 30 % in 2013.

Co-payments for healthcare goods and services are in place; however is unclear whether they play any demand-management role in the current system. About 60 % of the population is insured against co-payments, mainly by the HZZO. For a third of the insured the insurance fee is covered by the state. The monthly cost of the insurance is around EUR 10. The authorities admit that the insurance system is vulnerable, as policy-holders include a large proportion of elderly persons (with the highest healthcare costs). Despite this unfavourable structure, the system is profitable. Given also the high number of exempt conditions, co-payments are unlikely to result in high unexpected out–of-pocket payments for patients.

The dominant role of the HZZO in the complementary insurance market is problematic. The HZZO is a public institution, so the financial risk inherent in the complementary insurance product (although it is currently profitable) is effectively passed on to the state. There are plans to separate the complementary scheme, but not to change the coverage. The effects of the co-payments and the complementary insurance on demand for health services and service use patterns are unclear.

Energy

Croatia’s economy is more energy and carbon‑intensive than the EU average, while import dependence is in line with the rest of the EU. Dependence on energy imports has nevertheless been increasing in recent years — especially for gas — pointing to the importance of further diversifying import sources. Transport is the sector that consumes the most energy (34 % of gross consumption). In order to reduce transport energy consumption as of 2015, the authorities have put in place a national programme that involves new financial incentives to procure electric and hybrid vehicles. The residential sector accounts for 30 % of total energy consumption. Croatia is mobilising public and private investment for the renovation of the national building stock. However, improving energy efficiency in buildings remains a challenge requiring significant support for such investment.

The development of the gas sector has seen new players gradually entering the market. However, price regulation hinders the sector’s development. In 2003, there was only one company bringing gas into the country, while in 2013 there were five. Gas retailers increased from three to 42 over the same period, so that the market share of the incumbent is now about 40 % below the EU average of over 50 %. In spite of this, end users’ prices have increased steeply since 2008, particularly because of a doubling of VAT and a sharp rise in gas supply costs, which have gone up by 160 % for industry and by 100 % for households. In Purchasing power standards terms, Croatia’s gas prices are the third highest in the EU for industry, while for households they are just above the EU average. The current Gas Market Act is not conducive to competition and investment in the gas markets, and could undermine the basis for Croatia’s further integration in the internal market. The Act allows for retail price regulation, which hinders competition at household level by discouraging consumers from switching to alternative gas providers and ensuring a 100 % market share for the incumbent retail suppliers. Additional regulation of gas prices was introduced by government decisions in February 2014. The gas producer INA has been placed under an obligation to offer its domestic production at a regulated price primarily to the incumbent state-owned wholesaler Hrvatska Elektroprivreda, which in turn re-sold it to retail distributors at a regulated price. Furthermore, the restriction on exports of domestically produced gas, which is not compatible with internal market rules, was retained in the Gas Market Act.

Gas companies’ level investment efforts have dropped off sharply since 2008. This followed a drop in gas consumption of 6 % overall and as much as 30 % in the industrial sector between 2008 and 2012. The poor investment climate negatively affects the development of projects of common interest , such as the LNG terminal on the island of Krk, which could play an important role in the diversification of supply in the region. The EU‑wide list of projects of common interests includes 13 projects in Croatia; these are at various stages of development, from feasibility level to building permits level. The Croatian gas transmission network is connected with Hungary and Slovenia and there are plans to extend it to Bosnia‑Herzegovina by 2018 (in the south, via the trans-Adriatic pipeline) and Serbia by 2023. Croatia is one of the Member States that supports the Ionian Adriatic Pipeline, to be commissioned in 2018, which should bring Caspian Sea gas to Europe via the trans-Adriatic pipeline).

The process of unbundling of the state-owned electricity utility Hrvatska Elektroprivreda is under way. Between 2008 and 2013 electricity prices for households increased by about 8%, mainly because of an increase in the energy supply component while prices for industry remained unchanged. In terms of Purchasing power standards, Croatia’s prices for both industry and households in 2013 were just above the EU average. The state-owned transmission system operator Hrvatska Elektroprivreda is responsible for generation, transmission and distribution systems in electricity. Since the market was opened up in July 2013, several providers, both domestic and international, have entered the market previously exclusively supplied by Hrvatska Elektroprivreda, which remains by far the dominant electricity supplier with a share of over 85 % of the national market. The unbundling obligations contained in the 2013 Electricity Market Act introduced changes in the capital and structure of the vertically integrated utility Hrvatska Elektroprivreda (now named HOPS). However, the process of unbundling is still ongoing.

Croatia is on track to achieving its national renewable energy target of 20 % by 2020. Under the EU 2020 strategy, Croatia committed itself to a greenhouse gas target in 2020 of 11 % of emissions as compared with 2005 levels. The percentage of renewable energy was 16.8 % in 2012, up from 12.6 % in 2005, which is above the indicative trajectory in the Renewable Energy Directive. According to national projections, Croatia will reduce its non-ETS emissions by 2020 by 6 % as compared with 2005, i.e. in line with its national target. In November 2014, the government adopted a plan for the use of funds from the auctioning of emission allowances in 2014‑16; this estimates that HRK 688 million will be available. However, the transport sector (which should contribute 10 % to the overall target) is lagging behind, with only 0.23 % renewables in 2011 and 0.40 % in 2012.

Croatia is likely to achieve its energy efficiency target, which was however set at a rather unambitious level.  Taking into account the latest national projections and existing measures, the target of -6 % by 2020 (as compared with 2005 levels) is likely to be achieved. Progress has been made in implementing specific energy efficiency measures in the public sector at local and regional levels, although full transposition of the Energy Efficiency Directive has not yet been achieved. In recent years, Croatia has used national and EU funding to support refurbishment programmes for public buildings, energy audits, energy efficiency improvement investments in industry and information campaigns. Nevertheless, further energy‑saving opportunities remain untapped across all sectors. The transport sector’s share in GHG emissions amounts to around 25 %, of which over 70 % is generated by road transport. The Ministry of Environment and Nature Protection launched a package of measures aimed at raising awareness of the need to keep the air clean. Lower excise duties have been introduced for environment‑friendly vehicles; the procurement of such vehicles by citizens and companies is co-financed to the tune of HRK 15.5 million; ‘green public transport’ is subsidised. The Green Line project supplies eco vehicles and vessels in national parks and nature parks. A fairer model for collecting eco tax will enter into force next year: the amount will be determined on the basis of ecological criteria (average carbon dioxide emissions and type of fuel), rather than just the value of the vehicle.

Transport

The quality of transport infrastructure in Croatia varies heavily from mode to mode. Road infrastructure is well developed. Since 2000, more than 4 % of GDP has been spent annually to develop the motorway infrastructure. Motorways have been constructed mainly by state-owned enterprises, while private companies carry out maintenance under concession contracts. There are plans for a public-private partnership aimed at relieving the public debt of the liabilities incurred by the public motorway companies. The first round of bidding for the concession of 1 250 km of motorways has ended and talks with the bidders are ongoing.

The railway infrastructure is in dire need of modernisation. A period of sustained investment prior to the crisis has been followed by a sharp fall in investment rates both for network expansion and maintenance. While electrified rail density is similar to the EU average, several important routes remain non-electrified and single-tracked, and feature steep single-tracked and over-steep gradients, which leads to lower average speeds. The railway system faces challenges both operationally and financially. There is a significant maintenance backlog in rail substructure and superstructure and the network needs modernisation to improve operational capacity. Rail density and electrification have been increasing slowly, even though traffic intensity has contracted sharply since 2008 (by 30 % for freight and 38 % for passenger services). Rail freight and passenger traffic are below the EU average: 545 tonnes per capita and 258 passenger km per capita respectively in 2012, against EU levels of 805 tonnes and 827 km. Due to the decreasing traffic and the high labour costs, in the light of productivity, Croatian Railways’ financial situation remains problematic. Companies operating in the railway sector have faced increasing losses since 2008 – negative returns on capital employed are particularly significant in freight and passenger services (less so in infrastructure management). However, negative profitability in railways has been a common phenomenon in most Member States since the beginning of the crisis. In 2012, the state-owned Croatian Railways was broken down into three independent companies, but greater liberalisation in the market is still necessary to increase competition and attract investment. 

Croatia’s maritime and river ports would need qualitative and technological modernisation of basic infrastructure to satisfy existing and expected transport demand. Croatian ports are not connected with the main road and rail corridors which undermines the preconditions for the development of intermodal transport and limits the integration of the maritime regions with the economic hinterland. As far as inland shipping is concerned, the current situation is characterised by unbalanced market demand for cargo transport between the Sava and Danube rivers. Generally, this is a consequence of unreliable navigation conditions on the Sava, where important inland ports are located. The plans are underway to upgrade of the Sava to inland waterway class IV to ensure smoother and safer navigation. But the project has been delayed due to the 2014 floods and financing issues in neighbouring countries.

Broadband and the digital economy

Croatia faces a number of challenges in its progress towards a digital economy. Although by the end of 2013, 97 % of homes in Croatia had fixed broadband, penetration across the country is unequal. Požeško-Slavonska county has the lowest fixed broadband penetration rate (14.4 %) and the City of Zagreb and Zagrebačka county have the highest. In rural areas in general, only 77 % of homes had fixed broadband (as compared with 90 % in the EU as a whole) ([64]). Possible reasons for the low take-up (subscriptions) are the lack of investment in fixed broadband infrastructure, insufficient e-skills among citizens and the very high cost of broadband subscriptions in relation to gross income. In terms of affordability, Croatia ranks last out of the 28 Member States. This could be related to the market being dominated by the incumbent operator, Hrvatski Telekom, especially in fixed and broadband (69 % market share at the end of 2013) and a consequent lack of competition.

The limited take-up of digital technology holds back the development of e-commerce. Fewer than 3 in 10 consumers (28% compared to the EU average of 50%) have bought goods or services online in 2014, with little progress (2 percentage points) since 2013. Consumers’ confidence to buy online domestically or cross border is by far the lowest in the EU.  It can also be noted that in terms of Use of Internet Croatians are very reluctant to make transactions involving the use of payment devices. This might be linked to a general mistrust of the security of online payment systems. This is all the more disturbing given that a quarter of Croatian SMEs sell online.

Policy measures have improved the regulatory framework, but a low level of competition holds back further development of the digital economy. In 2011, the government adopted a 2012-15 Strategy for Broadband Development and the 2012-13 programme for implementing the Strategy. In 2014-20, Croatia intends to use EUR 308 million from the ESIFs. Croatia has thus made some progress with respect to broadband as far as administrative and financial decisions are concerned, but challenges remain as regards availability of fast broadband infrastructures and making internet access more affordable. Greater competition in the market could also reduce the costs of the internet for citizens and encourage greater take-up.

Environment

Waste and water management remain a challenge in Croatia. Croatia is performing particularly poorly against recycling targets, with only 15 % of municipal waste being recycled. The efficiency of public water supply is low with high losses from the system. Croatia makes limited use of various economic instruments that help recover costs and promote more efficient use of resources.

Education

The skill-composition of the Croatian workforce drags on aggregate productivity. The share of the population with tertiary education is one of the lowest in the EU in 2013. Low educational attainment also results in lower overall employment rates. The weak absorption capacity of the domestic production system, the skill mismatch phenomenon and the weaknesses of the education and training system, all act as deterrents for higher education attainment, which results in the loss of labour market potential. 

Skill mismatch contributes to gaps between labour demand and supply. The mismatches mainly concern workers with a low level and a high level of education ([65])., whereas those with a medium education level are found to be better matched. The so called ʻsubstitution assumption’ – which implies that low skilled workers are replaced by those with an immediately higher education level – does not apply for the case of Croatia. The unemployment rate for the low-skilled does not appear extremely high in comparative perspective. Finally, Croatia is among the countries with relatively high horizontal mismatches i.e. mismatches that relate not as much to the level of education attained, but rather to the field of study relative to the labour market needs ([66]).

Increasing the relevance and quality of education remains priority to address labour market needs. Structural weaknesses in the design of vocational education and training and adult education systems make the supply of non-compulsory education highly unresponsive to the needs of the labour market. The transition from school to labour market is hampered by outdated vocational education and training curricula and limited opportunities for work-based learning, lack of career guidance and low engagement by employers. Higher education insufficiently reflects labour market needs with an overproduction of diplomas in humanities. Low relevance of education compounds with quality issues. There is no centralised quality assurance mechanism in tertiary education for quality of teaching or learning outcomes of students. Croatia has one of the lowest tertiary attainment rates in the EU (25.6% and 36.9% respectively for population 30-34 in 2013) and high dropout rates result from lack of necessary entry competences, limited academic and career counselling, and lack of financial means. In addition, high access barriers exist for students from low socio-economic backgrounds as well as older students and part-time students. The system of adult education in Croatia is weakly governed and participation in life-long learning is extremely low. The incentives for employers in the form of tax deductions for adult education and to offer training costs are insufficiently used, due to low awareness and complexity of administrative procedures.

Croatia is currently tackling deep rooted weaknesses in its vocational education system and further reforms are in the making in tertiary education. The reform of the Croatian Qualifications Framework and the modernisation of vocational education and training curricula are progressing, aimed at modernising all study programmes leading to qualifications based on sector skills analysis. Updated curricula have been introduced in some vocational schools, but the impact is not visible yet. A vocational education and training development plan, to be adopted by the end of 2015, will be the first step in initiating a systematic vocational education and training reform, in line with the Croatian Qualifications Framework methodology, which is designed to close the skills gap between education and the labour market. To promote participation in lifelong learning, a system for recognition and validation of non-formal and informal learning is planned to be developed in early 2015 as part of the work on the qualifications framework.

The current weaknesses in the education and training system – including in primary and secondary education – could have adverse effects in the long-term perspective. The levels of basic and transversal skills (i.e. skills that prepare individuals for varied and unpredictable career paths and are usually taught across all subjects) are insufficient, reflected by low educational achievements of 15 year olds (measured by PISA), most strongly in mathematics. In addition, lack of adequate competencies is also prevalent in the use of ICT and digital technology, particularly among students in primary and vocational education. Underlying problems of learning outcomes also relate to teacher effectiveness, quality assurance and the curricula. Croatia’s public expenditure on education and training in 2012 was 5% of GDP (EU average 5.3%) and planned education budget decreased by almost 2% between 2013 and 2014. Failing to secure adequate funding to support the modernisation of school curricula could result in disruptive impacts of basic skills shortages on secondary school graduates’ prospects for employment or continuing education.

The Strategy for Education, Science and Technology aims to reform education and training system, however timely implementation is key. The Strategy adopted in October 2014 envisages a wide-reaching reform of the education system including an enhanced system of quality assurance, development of national competence standards for teachers and introducing a new national framework curriculum and gradually introducing work-based learning into all vocational education and training programmes. In higher education, the Strategy announces a revision of study programmes, differentiation of learning outcomes between professionally-oriented and academic university courses, harmonisation of study programmes in line with labour market needs, integration of transversal competences in study programmes, improvement of the social dimension of studying and stronger internationalisation of higher education. The implementation action plan has not yet been adopted and is expected in the beginning of 2015. The Strategy has the potential to significantly improve the quality of learning outcomes, however, no clear budget line or funding has been guaranteed for its implementation, which raises concerns about the commitment to its effective implementation.

Graph 3.4.1:   Strategies judged very important by innovative and non-innovative enterprises (2012)

Source: European Commission

Research and Innovation

Research and innovation in Croatian enterprises is hindered by scarce capacities, unfavourable incentives structures and limited internationalisation. Despite low aggregate business expenditure in R&D, available data suggest that faced with relatively high labour costs, Croatian enterprises face ([67]) lack of information on technology along with lack of qualified personnel which are the most relevant predictors of abandonments and delay innovation. An additional constraint to Croatia innovative capacity is likely to result from the weak internationalisation of companies pursuing innovation investments.

Graph 3.4.2:   R&D intensity by sector (2013)

Source: European Commission

Insufficient public R&D investments contribute to the current ʻlow-level equilibrium’ innovation system. In a low-level equilibrium, business sector’s weak innovation capabilities and low investment in R&D&I leads to little effective demand for and supply of innovation-related services and research ([68]). In such ʻlocked-in’ state, public policies can play a significant role in boosting the economy’s innovative capacity by fostering the accumulation of innovation capabilities. Yet Croatia stands out for limited public support for research. In this regard, a worrying development is the sharp decrease in public R&D intensity from 0.50 % of GDP in 2008 to 0.41 % in 2013, even with a shrinking GDP.

Public policies in support of innovation are characterised by weak commitment, slow reform pace and inefficient governance structures. The incapacity to mobilise public resources for R&D witnesses an underestimation of the competitive pressures stemming from more cost-competitive peer economies. In its review of Croatia’s innovation policy, the OECD highlights how policy measures, including funding decisions, with their strong focus on high-tech start-ups and technology transfer through instruments such as technology parks, incubators and public-private partnerships, might have been excessively influenced by the dominant paradigm in more advanced economies without an adequate contextualisation of the domestic constraints. According to the OECD report, available evidence overwhelmingly supports the view that the binding constraint that Croatia’s is facing is not at the interface but at the core of public-sector and business-sector, which still suffer from a sub-critical accumulation of innovation capabilities. These shortcomings compound with weaknesses in the governance of the research policy, public research institutions and higher education institutions. Croatia has made an attempt to fund research and higher education institutions according to performance-based criteria but the percentage public funding allocated on this basis is very low. Sub-critical scale, excessive fragmentation, relative closure and a persistent mismatch between academic curricula and labour market needs dominate the public research landscape.

Building a performing national innovation system and benefiting from EU funds will require an overhaul of the governance of public and publicly funded research. Given that Croatia is expected to receive significant funding for R&D, the 2014-2020 ESIF programming period could be used to stimulate Croatia’s transition to a knowledge-intensive economy through targeted capacity building and an integration of existing areas of scientific excellence and industry clusters. Such targeted capacity-building is of key importance to ensure a high leverage effect of the public funding on business investment. However, the national Smart Specialisation Strategy has not been adopted yet, and the thematic areas identified within the preliminary competitiveness analysis underlying the Strategy remain very broad. Moreover, the governance of public research and higher education remains a challenge to ensure higher efficiency and effectiveness of investments. In addition, focus on capacity building through quality investment also requires policies targeting non-R&D innovation and technology adoption.

Given the currently limited fiscal space and a highly leveraged corporate sector (see section 2.3), the ESIFs are a crucial financial and investment resource for Croatia in its efforts to achieve its EU 2020 targets, implement structural reforms, strengthen innovation and competitiveness, and ensure full alignment with the EU acquis. By the end of 2014, the Croatian Government and the Commission had prepared and adopted a Partnership Agreement and two operational programmes establishing the ESIF investment framework for 2014-20. The new allocation is EUR 10.68 bn, i.e. over 3 % of GDP on annual basis or about ten times as much as that for 2007-13. This represents a huge opportunity, provided that the authorities can prioritise investments with high economic (and social) return (see also section 2.2 on the macroeconomic risks of strong currency inflows). If the funding is to be managed effectively and absorbed in the coming years, Croatia will need to make major efforts to ensure adequate capacities for strategic programming, procurement, implementation, fund management and control, monitoring and evaluation, and mobilise stakeholders to prepare and implement results-oriented and quality projects.

The main short-term challenges for Croatia stem from the need to absorb the 2007-13 allocation by the end of 2016 at the same time as launching the new 2014-20 programmes. Poor public governance leads to ineffective management of EU funds. Limited administrative and technical capacities in management bodies, at local and beneficiary level, combined with difficulties linked to the switchover from the IPA to the ESIF, have contributed to delays in the absorption of the 2007-13 allocation. Also, contracting and implementation of the 2007-13 programme overlapped with intensive work to prepare the 2014-20 Partnership Agreement. The legal framework for implementation of the 2014‑20 programmes is now in place. However, the new programmes include a number of new sectors (ICT, energy, climate change, health and social inclusion, education) which will require specific technical capacities in the management bodies and targeted support for project beneficiaries.

Strong inter-ministerial and strategic coordination has been key to work on a long‑term strategy for investing the ESIFs. Throughout 2014, ministries worked together intensively at political and operational levels to prepare the Partnership Agreement and the operational programmes. Going forward, stepping up such efforts would ensure that the ESIF funding is allocated on the basis of a clear long-term and productive investment plan. Timely implementation of the action plans would mean that the necessary regulatory, strategic and capacity pre-conditions are in place for the funding to be invested and managed effectively. Implementation will include adopting a number of sectoral investment strategies (e.g. a smart specialisation strategy, an integrated transport strategy, a digital strategy and a strategy for public administration). The success of the strategies will depend on how effective they are in establishing cross-sectoral links, identifying clear investment priorities and intended results, and using strong monitoring tools to assess capacity gaps. Although the 2014-20 programmes emphasise results‑oriented programming and implementation, experience has shown that the analytical and monitoring capacities of the Croatian ESIF coordinating bodies and managing authorities are limited, and need to be bolstered, particularly as regards the new sectors to be covered.

Good range of high-quality, mature projects will be needed for the new programming period. The low absorption of the 2007-13 ESIFs is partly due to the restricted supply of ready-to-implement eligible projects. The amount of time taken to prepare projects and go through the tendering procedures delayed implementation. Also, beneficiaries did not always have access to the targeted support they needed to be able to prepare and implement projects in line with EU rules. The inclusion of new sectors in 2014-20 and the eligibility of new beneficiary profiles (in particular in the private sector) call for specific measures at the level of the coordinating and managing authorities and at project level. Initiatives in the past two years include establishing specific units within management bodies to support, and tailored on-demand training and coaching for, project beneficiaries. The further development of these initiatives would have to be in line with the increased ESIF allocation.

Further steps have been taken to improve administrative capacity and strategic planning as regards management of the ESIF funding. On the basis of a workload analysis by the management bodies, the authorities have decided to recruit 349 additional staff across various ministries and agencies in the course of 2015 to deal with the ESIF. Depending on individual needs, this will include technical experts. Throughout the year, the new staff will be given targeted training which will cover the effective and regular application of public procurement rules, and the technical skills needed in the new funding sectors. Other targeted measures will include: i) developing a common methodology to ensure that funds are managed consistently and efficiently across ESIF-funded programmes; and setting up an IT platform to address the fragmentation of the system and to improve coordination between stakeholders; ii) internal reorganisation of the Ministry of Regional Development and EU Funds and the other management authorities; and iii) a regulation on eligible expenditure for the 2014-20 programming period. An action plan for strengthening capacities and public procurement procedures in the management and implementation of ESIF funding was adopted in December 2014. Due to be implemented from November 2015, it should ensure the effective and transparent application of the public procurement rules, and improve monitoring and the detection of irregularities. In 2014, all budgetary users included in their 2015-17 financial and strategic plans the projects in their area of competence that are eligible for ESIF financing. A detailed analysis was carried out of planned public investment under the 2015-17 state budget. Work is ongoing to identify priority projects for contracting and implementation in 2015 and 2016. The first ‘programme complements’ (setting out timelines and the content of calls for proposals for the 2014‑16 strategic projects) are due to be published in May 2015.

Commitments || Summary assessment ([69])

2014 country‑specific recommendations (CSRs)

CSR1: Fully implement the budgetary measures adopted for 2014. Reinforce the budgetary strategy, further specifying announced measures for 2015 and 2016, and considering additional permanent, growth-friendly measures in order to ensure a sustainable correction of the excessive deficit by 2016. At the same time, ensure that the structural adjustment effort as specified in the Council recommendation under the Excessive Deficit Procedure is delivered. Align programme projections with ESA standards and Stability and Growth Pact requirements. Take measures to reinforce control over expenditure. By March 2015, carry out a thorough expenditure review. Reinforce the budgetary planning process, in particular by improving the accuracy of macroeconomic and budgetary forecasts and strengthening the binding nature of the annual and medium-term expenditure ceilings and improve the design of fiscal rules. By October 2014, ground in law the newly established Fiscal Policy Commission, strengthen its independence from all budgetary authorities, broaden its mandate, in particular with respect to the monitoring of all fiscal rules and the ex ante and ex post assessment of forecasts, and ensure adequate resourcing. Building on plans outlined in the National Reform Programme, present a concrete strategy to reform recurrent property taxation. Initiate a process of reporting and reviewing of tax expenditures. Improve tax compliance, in particular by further enhancing the efficiency of the tax administration; present an action plan to this end by the end of 2014. || Croatia has made limited progress in addressing CSR 1 of the Council recommendation (this overall assessment of CSR 1 excludes an assessment of compliance with the Stability and Growth Pact): Limited progress in aligning programme projections with ESA standards and Stability and Growth Pact requirements. Limited progress on measures to improve control over expenditure. Some progress in carrying out a thorough expenditure review. Some progress in improving the budgetary planning process. Limited progress in grounding in law the newly established Fiscal Policy Commission, strengthening its independence from all budgetary authorities and broadening its mandate. Limited progress in building on plans outlined in the National Reform Programme and presenting a concrete strategy to reform recurrent property taxation. Some progress in initiating a process of reporting and reviewing tax expenditures. Some progress in improving tax compliance and presenting an action plan on this by the end of 2014.

CSR2: Reduce access to early retirement. Adopt legislation by March 2015 to accelerate the planned harmonisation of statutory retirement ages of women and men and to advance the planned increase of the statutory retirement age to 67 years. Ensure enforcement of tighter disability pensions assessments and controls and accelerate the integration of pensions under special schemes into the general pension system. Strengthen the cost-effectiveness of the healthcare sector, including hospitals. || Croatia has made limited progress in addressing CSR 2 of the Council recommendation: No progress in reducing access to early retirement, accelerating the planned harmonisation of statutory retirement ages, bringing forward the planned rise in the statutory retirement age to 67 years or integrating special scheme pensions more quickly into the general pension system. The number of disability pensions granted in 2014 is expected to have been substantially higher than in 2013 but lower than in previous years. Some progress in making the healthcare sector more cost-effective.

CSR3: Implement the second phase of the labour law reform, following consultation with the social partners, in particular as regards conditions for dismissals and working time, and with a view to preventing further labour market segmentation including for young people, by March 2015. Review the wage-setting system with a view to better aligning productivity developments and wage conditions. Present the conclusions of this review by the end of 2014. Strengthen the effectiveness and reach of active labour market policies by reinforcing the administrative capacities of the public employment services, including at regional level, and by increasing the coverage of the young, long-term unemployed and older workers. Prioritise outreach to non-registered youth and mobilise the private sector to offer more apprenticeships, in line with the objectives of a youth guarantee. Outline plans, by the end of 2014, to address undeclared work. Implement measures to improve the labour market relevance and quality of education outcomes by modernising the qualification systems, by putting in place quality assurance mechanisms and by improving school-to-work transitions, in particular through strengthening vocational education and work-based learning. || Croatia has made some progress in addressing CSR 3 of the Council recommendation: Fully addressed implementation of the second phase of the labour law reform. Some progress in reviewing the wage‐setting system. The conclusions lack concrete policy proposals, however. Some progress in strengthening the effectiveness and reach of active labour market policies. Substantial progress in prioritising outreach to non-registered youth and mobilising the private sector to offer more apprenticeships. Some progress in outlining plans to address undeclared work. Some progress in implementing measures to improve the labour market relevance and quality of education outcomes.

CSR4: Review tax and benefits systems by the end of 2014, and present an action plan to improve the reactivation of inactive and unemployed persons. Strengthen the effectiveness and transparency of the social protection system by further consolidating benefits, unifying eligibility criteria and linking data from all relevant levels and government entities in the ‘one-stop shop’. Improve the effectiveness and adequacy of social assistance benefits through their better targeting. || Croatia has made some progress in addressing CSR 4 of the Council recommendation: Some progress in reviewing tax and benefits systems and presenting an action plan to improve the reactivation of inactive and unemployed persons. Limited progress in strengthening the effectiveness and transparency of the social protection system and improving the effectiveness and adequacy of social assistance benefits through better targeting.

CSR5: Take further measures to improve the business environment. In particular, by March 2015 set a target for considerably lowering administrative requirements, including para-fiscal charges. Address the high level of fragmentation and overlapping responsibilities by streamlining administrative processes and by clarifying the decision-making and accountability framework across various levels of government and at central government level between ministries and agencies. Improve administrative capacity and strategic planning of units entrusted with the management of European Structural and Investment Funds and provide them with adequate and stable staffing levels. || Croatia has made limited progress in addressing CSR 5 of the Council recommendation: Limited progress in improving the business environment overall. The authorities initiated measurement of administrative burdens. Progress on reducing para-fiscal charges has been considerably slower than expected. Limited progress in addressing the high level of fragmentation and overlapping responsibilities across various levels of government. Some progress in improving the administrative capacity and strategic planning of units managing European Structural and Investment Funds and providing them with adequate and stable staffing levels.

CSR6: Present, by October 2014, a detailed plan for public property management for 2015. Ensure that companies under state control are governed in a transparent and accountable manner, in particular, strengthen the competency requirements for members of management and supervisory boards nominated by the State and introduce a public register for appointments. Reinforce prevention of corruption in public administration and state-owned and state-controlled enterprises, including by increasing the verification powers of the Conflict of Interest Commission. Strengthen transparency and efficiency of public procurement at both central and local levels, and the capacity to monitor implementation and to detect irregularities. || Croatia has made limited progress in addressing CSR 6 of the Council recommendation: Limited progress on improving the governance of companies under state control overall. The 2015 State Asset Management Plan was adopted in November 2014.  The public appointments register has been made public. Limited progress in reinforcing prevention of corruption. The new anti-corruption strategy has been adopted but it lacks a sufficient level of detail as to measures that will be implemented.

CSR7: By the end of 2014, reinforce the role of commercial courts in the monitoring of transparency and legality in the application of the corporate pre-bankruptcy procedure. Review the compulsory test of insolvency/illiquidity to access pre-bankruptcy settlement proceedings and streamline the insolvency/liquidation process to reduce its length. Improve the quality and efficiency of the judicial system, in particular by providing incentives to resolve proceedings in litigious civil and commercial cases and in administrative cases in a timely manner and to resort to out-of-court settlement especially for smaller claims. || Croatia has made limited progress in addressing CSR 7 of the Council recommendation: Some progress in improving the pre‑insolvency and insolvency framework for corporate entities. The new Insolvency Law is expected to be adopted by parliament in the first quarter of 2015. The reform reinforces the role of commercial courts, facilitates access to the procedure and streamlines the insolvency/liquidation process. Limited progress in improving the quality and efficiency of the judicial system overall. Implementing the reform of the judicial map could bring progress in improving the quality and efficiency of municipal, misdemeanour and county courts. Information and communication technology systems are being implemented in courts. No sufficient measures have been adopted to improve efficiency in litigious commercial cases. Some measures are planned to address efficiency in (first‑instance) administrative cases.

CSR8: Complement the 2014 European Central Bank’s asset quality reviews and stress test exercises, undertake a comprehensive portfolio screening exercise designed specifically for the Croatian financial sector, with a focus on important portfolios that are not covered by the European Central Bank exercise and including key mid-size and smaller banks. || Croatia has made substantial progress in addressing CSR 8 of the Council recommendation.

Europe 2020 (national targets and progress)

Employment rate target: 62.9 %, || The employment rate in Croatia is one of the lowest in the EU, but since 2012 there have been some moderate positive trends. It increased from 55.4 % in 2012 to 57.2 % in 2013 and the trend continued in 2014, with 61.6 % in Q3‑2014. Croatia may achieve its national target of 62.9 %, but this is far below the level in other EU countries.

Europe2020 R&D target 1.4 % of GDP || The level of investment in research and development in Croatia increased from 0.75 % of GDP in 2012 to 0.81 % in 2013 (partly due to contraction in GDP), but it is still significantly below the national 1.40 % target for 2020. Public R&D intensity decreased sharply from 0.5 % in 2009 to 0.41 % in 2013. Business R&D intensity remained stable at 0.34 % between 2009 and 2012 and rose to 0.41 % in 2013. Limited progress has been achieved in relation to the Europe 2020 target, but Croatia is not on track to reach it.

Non-ETS emission reduction target: +11 % compared to 2005 emissions. || According to preliminary estimates, emissions decreased by 9 % between 2005 and 2013. According to the latest national projections and taking into account existing measures, it is expected that the target will be achieved: -6 % in 2020 as compared with 2005 (a 17 pps margin).

2020 Renewable energy target: 20 % Share of renewable energy in all modes of transport: 10 % || With an RES share of 16.8 % in 2012, Croatia is on track to achieve its 20 % target in 2020. There is no specific 2030 RES target.

Energy Efficiency target: 9.2 Mtoe expressed in primary energy consumption (7.8 Mtoe expressed in final energy consumption) || Although Croatia’s current primary energy consumption (7.6 Mtoe in 2012) is below its 2020 target, additional efforts on EE are needed to keep primary energy consumption at this level or to minimise its increase if GDP increases again over the next five years.

Early school leaving target: 4 % || The proportion of early school leavers in Croatia has been increasing since 2008 and reached 4.2 % in 2012 and 4.5 % in 2013, thus moving away from the national target of 4 %. Future developments will need to be closely monitored. However, the rate remains one of the lowest in the EU and is well below the EU target of 10 %.

Tertiary education target: 35 % || Tertiary attainment of 30-34 year olds has been rising continuously and reached 23.7 % in 2012 and 25.6 % in 2013. However, Croatia still performed well below the EU average of 36.9 % in 2013 and, in view of the growth rates, may not reach its 35 % national target by 2020.

Target on the reduction of population at risk of poverty or social exclusion: by 150.000. || The number of people at risk of poverty fell from 1 384 000 in 2012 to 1 271 000 in 2013. Croatia is likely to meet its national target of reducing the number of people at risk of poverty or social exclusion by 150 000 to 1 220 000.

Table AB.1:     Macroeconomic indicators

(1) The output gap constitutes the gap between the actual and potential gross domestic product at 2010  market prices. (2)  The indicator of domestic demand includes stocks. (3)  Unemployed persons are all those who were not employed, had actively sought work and were ready to begin working immediately or within two weeks. The labour force is the total number of people employed and unemployed. The unemployment rate covers the age group 15-74. Source:  European Commission 2015 winter forecast; Commission calculations

Table AB.2:     Financial market indicators

(1) Latest data November 2014. (2) Latest data Q3 2014 (3) Latest data September 2014. (4) Latest data June 2014.  Monetary authorities, monetary and financial institutions are not included. (*)  Measured in basis points. Source: IMF (financial soundness indicators); European Commission (long-term interest rates); World Bank (gross external debt); ECB (all other indicators).

Table AB.3:     Taxation indicators

1) Tax revenues are broken down by economic function, i.e. according to whether taxes are raised on consumption, labour or capital. See European Commission (2014), Taxation trends in the European Union, for a more detailed explanation. (2) This category comprises taxes on energy, transport and pollution and resources included in taxes on consumption and capital. (3)VAT efficiency is measured via the VAT revenue ratio. It is defined as the ratio between the actual VAT revenue collected and the revenue that would be raised if VAT was applied at the standard rate to all final (domestic) consumption expenditures, which is an imperfect measure of the theoretical pure VAT base. A low ratio can indicate a reduction of the tax base due to large exemptions or the application of reduced rates to a wide range of goods and services (‘policy gap’) or a failure to collect all tax due to e.g. fraud (‘collection gap’). It should be noted that the relative scale of cross-border shopping (including trade in financial services) compared to domestic consumption also influences the value of the ratio, notably for smaller economies. For a more detailed discussion, see European Commission (2012), Tax Reforms in EU Member States, and OECD (2014), Consumption tax trends. Source:  European Commission

Table AB.4:                Labour market and social indicators

(1) Unemployed persons are all those who were not employed, but had actively sought work and were ready to begin working immediately or within two weeks. The labour force is the total number of people employed and unemployed. Data on the unemployment rate of 2014 includes the last release by Eurostat in early February 2015. (2) Long-term unemployed are persons who have been unemployed for at least 12 months. Source:  European Commission (EU Labour Force Survey and European National Accounts)

Table AB.5:     Expenditure on social protection benefits (% of GDP)

(1) People at risk of poverty or social exclusion (AROPE): individuals who are at risk of poverty (AROP) and/or suffering from severe material deprivation (SMD) and/or living in households with zero or very low work intensity (LWI). (2) At-risk-of-poverty rate (AROP): proportion of people with an equivalised disposable income below 60 % of the national equivalised median income. (3) Proportion of people who experience at least four of the following forms of deprivation: not being able to afford to i) pay their rent or utility bills, ii) keep their home adequately warm, iii) face unexpected expenses, iv) eat meat, fish or a protein equivalent every second day, v) enjoy a week of holiday away from home once a year, vi) have a car, vii) have a washing machine, viii) have a colour TV, or ix) have a telephone. (4) People living in households with very low work intensity: proportion of people aged 0-59 living in households where the adults (excluding dependent children) worked less than 20 % of their total work-time potential in the previous 12 months. (5) For EE, CY, MT, SI and SK, thresholds in nominal values in euros; harmonised index of consumer prices (HICP) = 100 in 2006 (2007 survey refers to 2006 incomes) (6) 2014 data refer to the average of the first three quarters. Source:  For expenditure for social protection benefits ESSPROS; for social inclusion EU-SILC.

Table AB.6:     Product market performance and policy indicators

(1) Labour productivity is defined as gross value added (in constant prices) divided by the number of persons employed. (2) Patent data refer to applications to the European Patent Office (EPO). They are counted according to the year in which they were filed at the EPO. They are broken down according to the inventor’s place of residence, using fractional counting if multiple inventors or IPC classes are provided to avoid double counting. (3) The methodologies, including the assumptions, for this indicator are presented in detail here:  HYPERLINK "http://www.doingbusiness.org/methodology" . (4) Index: 0 = not regulated; 6 = most regulated. The methodologies of the OECD product market regulation indicators are presented in detail here:  HYPERLINK "http://www.oecd.org/competition/reform/indicatorsofproductmarketregulationhomepage.htm" (5)  Aggregate OECD indicators of regulation in energy, transport and communications (ETCR). Source:  European Commission; World Bank — Doing Business (for enforcing contracts and time to start a business); OECD (for the product market regulation indicators)

Table AB.7:     Green growth

(1) 2013 is not included in the table due to lack of data All macro intensity indicators are expressed as a ratio of a physical quantity to GDP (in 2000 prices) Energy intensity: gross inland energy consumption (in kgoe) divided by GDP (in EUR) Carbon intensity: Greenhouse gas emissions (in kg CO2 equivalents) divided by GDP (in EUR) Resource intensity: Domestic material consumption (in kg) divided by GDP (in EUR) Waste intensity: waste (in kg) divided by GDP (in EUR) Energy balance of trade: the balance of energy exports and imports, expressed as % of GDP Energy weight in HICP: the proportion of "energy" items in the consumption basket used for the construction of the HICP Difference between energy price change and inflation: energy component of HICP, and total HICP inflation (annual % change). Environmental taxes over labour or total taxes: from DG TAXUD’s database ‘Taxation trends in the European Union’ Industry energy intensity: final energy consumption of industry (in kgoe) divided by gross value added of industry (in 2005 EUR) Share of energy-intensive industries in the economy: share of gross value added of the energy-intensive industries in GDP Electricity and gas prices for medium-sized industrial users: consumption band 500–2000MWh and 10000–100000 GJ; figures excl. VAT. Recycling rate of municipal waste: ratio of recycled municipal waste to total municipal waste Public R&D for energy or for the environment: government spending on R&D (GBAORD) for these categories as % of GDP Proportion of GHG emissions covered by ETS: based on greenhouse gas emissions (excl. LULUCF) as reported by Member States to the European Environment Agency Transport energy intensity: final energy consumption of transport activity (kgoe) divided by transport industry gross value added (in 2005 EUR) Transport carbon intensity: greenhouse gas emissions in transport activity divided by gross value added of the transport sector Energy import dependency: net energy imports divided by gross inland energy consumption incl. consumption of international bunker fuels Diversification of oil import sources: Herfindahl index (HHI), calculated as the sum of the squared market shares of countries of origin Diversification of the energy mix: Herfindahl index over natural gas, total petrol products, nuclear heat, renewable energies and solid fuels Renewable energy share of energy mix: %-share of gross inland energy consumption, expressed in tonne oil equivalents (*) European Commission and European Environment Agency (**) For 2007 average of S1 & S2 for DE, HR, LU, NL, FI, SE & UK. Other countries only have S2. (***) For 2007 average of S1 & S2 for HR, IT, NL, FI, SE & UK. Other countries only have S2. Source: European Commission unless indicated otherwise; European Commission Calculations

([1])  In this and the following sections, several performance indicators will be compared with those of the ʻnew’ Member States from central and eastern Europe. The group of peer economies in the region (Bulgaria, the Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Slovakia, Slovenia and Romania) is referred to as the EU‑10.

([2])  While the total accumulated income loss (12.5 % between 2008 and 2013) does not stand out in international terms, long-run data on annual GDP growth (WB World Development Indicators, 1960-2013) suggest that in 96.5 % of all cases negative annual growth comes to an end after five consecutive years.

([3])  Iooty, Correa, Radas and Skrinjaric, Stylised Facts on Productivity Growth, Policy Research Working Paper, No 6990, World Bank, 2014.

([4])  Nestic, D,. Rubil, I., and Tomic, I., An Analysis of Wage Level and Wage Structure in Croatia, report prepared at the request of the Ministry of Labour and Pension System, 2014. See also See Nikolic J., Rubil I., Tomić I., Changes in Public and Private Sector Wage Structures in Two Emerging Market Economies during the Crisis, EIZ-WP-1403, p. 1-32, 2014.

([5])  Orsini, K. and Ostojić. V., Wage dynamics in Croatia, leaders and followers, European Commission, Directorate General for Economic and Financial Affairs, Country Focus, forthcoming.

([6])  Gotovac V., Drezgić S., Schuh U., Analysis of the severance pay scheme in the Republic of Croatia: current arrangements and changes to be considered, World Bank report No 81626, 2013, p. 1-69. In the meantime, the authorities have repealed the ‘Austrian model’ option because of a lack of evidence of the new scheme having a positive impact on labour mobility.

([7])  In the period 1999-2008 minimum wage has been kept relatively stable at the level of 33 percent of the average wage. In July 2008, minimum wage level increased in line with the 2008 Minimum Wage Act to reach 36.4 % of the average wage. Following the 2013 reform, the minimum wage is no longer indexed to GDP and inflation, but it is set by the Government decree following consultations with social partners and by taking into account “increase in the share of the minimum to the average wage”, whereas it cannot be set at lower level than it was in the previous year. This feature is not desirable considering the need of wages to flexibly adjust to prevailing macroeconomic conditions, especially in a deflationary environment. For more details see Nestic, D,. Rubil, I., and Tomic, I., An Analysis of Wage Level and Wage Structure in Croatia, Report prepared at the request of the Ministry of Labour and Pension System, 2014.

([8])  Bečić estimates an earnings function with 2005 and 2010 data from the Labour Force Survey to research the effect of education on wages in the Croatian labour market. The analysis suggests that the net wage premium (i.e. when controlling for other individual characteristics) from one additional year of schooling results into an 8.5 % increase in the average wage. The results also show that females' wages are on average 16 % lower than males’, all other things being equal. Bečić, M. The Effects of Education on Wages in the Croatian Labour Market, Department of Economics and Business Economics, University of Dubrovnik.

([9])  This includes government efficiency, ministerial compliance, monitoring of line ministries, monitoring of agencies and administrations, task funding, constitutional discretion and the existence of public service standards.

([10]) See, White Book 2014, Foreign Investors Council in Croatia (FICC), 2014.

([11]) IMF, Republic of Croatia 2012 Selected Issues, IMF Country Report No 12/303, 2012.

([12]) These estimates are derived from regression analysis of the exports of Croatia and peer economies on the income of import countries, the price ratio and the geographical distance between the capital of the exporting country and its trade partners. This model – typically referred as export gravitation model – allows also performing a simple counterfactual analysis: if the Croatian economy were able to achieve the price and income elasticity of neighbouring Slovenia, exports of goods would increase by around 80%.

([13]) Buturac, G. and Gržnić, J., The Competitiveness of Croatian Export to EU Markets, Zagreb International Review of Economics & Business, Vol. 12, No 1, 2009, p. 39-51.

([14]) The goods categorisation is based on Bahri Y., Turkey’s competitiveness in the European Union: A comparison with five candidate countries — Bulgaria, the Czech Republic, Hungary, Poland, Romania — and the EU-15, Ezoneplus Working Paper No 12, 2003.

([15]) Stojčić N., Bečić M. and Vojinić P., The Competitiveness of Exports from Manufacturing Industries in Croatia and Slovenia to the EU-15 Market: A Dynamic Panel Analysis, Croatian Economic Survey, Vol. 14, No 1, 2012, p. 69-105.

([16]) Hoekman, B. and Djankov, S., Determinants of the Export Structure of Countries in Central and Eastern Europe, The World Bank Economic Review, Vol. 11, No 3, 1997, pp. 471-487.

([17]) The foreign value added content of exports refers to the value of intermediate inputs of foreign origin which are, both directly and indirectly, embedded in the goods and services exported by a country. The indicator shows how much of foreign value added a country’s exports embodies.

([18]) Incidentally, this dichotomous export restructuring has also played a crucial role in determining long-run productivity growth. Countries attracting FDI to industries of higher-end technology intensity have consequently succeeded in substantially higher productivity growth. Jože, D., Kostevc, Č. & Rojec, M., Global Supply Chains at Work in Central and Eastern European Countries: Impact of FDI on export restructuring and productivity growth, Vives discussion paper series No 37, 2013, Katholieke Universiteit Leuven, Faculteit Economie en Bedrijfswetenschappen, Vives.

([19]) Estrin S., Uvalic M., Foreign Direct Investment into transition economies: Are the Balkans different?, London School of Economics, LSE ’Europe in question’ - Discussion Paper Series, Paper No. 64, 2013, p. 1-44. The bias of FDI towards the non-tradable sector does not only represent a missed opportunity. The distortion of relative prices in favour of real estate and construction has stifling effects on export performance, as the expansion of a non-tradable sector in a country with limited supply of production factors can have a detrimental effect on the ability of the tradable sector to increase its output and compete in international markets. See Tkalec, M. and Vizek, M., Real Estate Boom and Export Performance Bust in Croatia, Proceedings of Rijeka Faculty of Economics, Journal of Economics and Business, Vol. 32, No. 1, 2014, pp. 11-34  

([20]) Valdec M., Zrnc J., The Direction of Causality between Exports and Firm Performance; microeconomic evidence from Croatia using the matching approach, The Ninth Young Economists’ Seminar, Croatian National Bank, 2013.

([21]) The negative relationship between volumes and prices is also found in a recent panel-based estimate of hotel services supply. The study confirms a relatively high income elasticity – which is particularly positive in light of the high growth potential of most of the land-locked neighbouring regions. However the study also highlights the relatively high negative price elasticity (about -0.7) implying a difficulty in increasing the number of tourists without lowering the price the services. See Škuflić L., Štoković I., Demand Function for Croatian Tourist Product: A Panel Data Approach, Scientific Research, Modern Economy, 2011, 2, pp.49-53

([22]) Broz T., Dubravčić D., The Dutch Disease in Unwonted Places–Why has Croatia been infected while Slovenia Remains in Good Health?, South-Eastern Europe Journal of Economics 1, 2011, pp. 47-66.

([23]) Import penetration ratios are the ratio between the value of imports as a percentage of total domestic demand, where domestic demand is GDP minus exports plus imports [D = GDP-X+M)]. A low import penetration rate may reflect either import barriers or a good matching of output to domestic demand by competitive domestic firms capable of confronting foreign competition. Conversely, a high import penetration rate may signal the incapacity of domestic producers to compete with international competitors. A high penetration rate can also be associated with positive performance, especially for strongly export-oriented small economies that are leveraging industry-specific comparative advantages.

([24]) See Bussière, M., Callegari, G., Ghironi,F.,  Sestieri, G. Yamano, N., Estimating Trade Elasticities: Demand Composition and the Trade Collapse of 2008-09, NBER Working Paper No. 17712, 2011.

([25]) For the period 2000-2007, income elasticity and price elasticity of imports of goods are estimated at 2.0 and -0.9 respectively. These values confirm the results obtained in earlier estimates for Croatia and evidence for other catching-up economies. The high income elasticity and lower than unity price elasticity – especially in the considered timeframe – are indeed a signal of high import dependence and weak adjustment capacity. See Bobić  V., Income and Price Elasticities of Croatian Trade – A Panel Data Approach, Croatian National Bank, Working Papers W – 25, 2010, pp.1-26.

([26]) Reininger T., Factors Driving Import Demand in Central and Eastern European EU Member States, Oesterreichische Nationalbank Workshops, No. 14, 2007, pp. 163-183.

([27]) Poprzenovic A., Remittances and Income Inequality in Croatia, School of Economics and Management, Lund University, Department of Economics, Master Thesis, 2007.

([28]) In a recent analysis of government size and efficiency in the EU Member States (and Iceland and Norway), the authors find that Hungary, Slovenia and Croatia are the ʻnew’ Member States with the most oversized governments. Source: Bađun, Pribičević, Deskar-Škrbić (2014), Government size and efficiency as constraints to economic growth: comparing Croatia with other European countries, Post-Communist Economies, 26:3, pp. 297-323.

([29]) The fact that health account data are available only from 2011 and that the sickness and maternity benefits are part of the National Health Insurance Fund’s (HZZO) budget make a precise assessment difficult. According to the information provided by the authorities, the expenditure of the HZZO in 2014 amounted to 7% of GDP (including the financial injection from the state budget). Excluding payments for sickness and maternity benefits and benefits for was-veterans, public spending of the HZZO amounted to 6.6% of GDP. According to the System of Health Accounts (Eurostat), which provides data on health expenditure for 2011 (21 Member States), general government expenditure on health care was 5.5% of GDP, above the (simple) average of 9 other EU11 countries (5.0%) but below the average of 21 EU28 countries (6.3%).

([30]) According to an analysis on the efficiency of VAT, 2012 VAT expenditure was less than 4 %, of GDP, one of the lowest proportions in the EU, especially if compared with the situation of EU-15 countries. However, the situation has arguably improved since then, partly as a result of EU accession (importantly, Croatia repealed its zero VAT rate in 2013, replacing it with a 5 % rate). Source: Sopek, P. (2012). Tax Expenditures and the Efficiency of the Croatian Value Added Tax. Financial Theory and Practice, 36(3).

([31]) UNDP report prepared by the Andrew Young School of Policy Studies at Georgia State University under contract for the UNDP Bratislava Regional Centre. http://www.cea.org.mk/documents/studii/FiscalDecentralization2005%20UNDP.pdf

([32]) Jafarov and Gunnarson report, for example, how hospital staff are hired at central government level while hospitals are managed by local governments, which thus have limited incentives to rationalise over-capacity. Similarly, decisions to establish schools are made by cities and municipalities, while teachers are hired and financed by central government. Source: IMF, Government Spending on Health Care and Education in Croatia: Efficiency and Reform Options, 2008.

([33]) Significant changes to the legal framework for the financing of local government units were made in 2001 and 2003, paving the way for a more proactive role for local government in providing services to citizens. Some education, healthcare and social care activities were devolved to local level. However, since this is at the discretion of the local unit, the division of responsibilities is unequal and inconsistent. Also, some local units in ’special state‑supported areas’ receive additional central government support.

([34]) A quantitative assessment of the fiscal situation at local level is complicated by the absence of data. This is partly a reflection of a weak statistical base: until 2013, the budgets of only the 53 biggest local units were consolidated into the ’general government’ budget, as defined in the national methodology, which muddied the budgetary situation at local level. As a result of applying ESA methodology, consolidated general government figures now include all local units. However, the numbers are available only as aggregates and there is still no publicly available central database of the budgets of all local units.

([35]) Bajo, Primorac, Sopek, Vuco (2014) Neto fiskalni položaj županija od 2011 do 2013, IJF.

([36]) See GRECO, Fourth Evaluation Round – Corruption prevention in respect of members of parliament, judges and prosecutors – Evaluation Report – Croatia, 2014.

([37]) Evaluation Report: Impact and Effectiveness of EU Public Procurement Legislation               http://ec.europa.eu/internal_market/publicprocurement/docs/modernising_rules/executive-summary_en.pdf.

([38]) DG HOME, EU Anti-Corruption Report (annex on Croatia), 2014.

([39]) Bajo and Primorac (2014) Dug i fiskalni rizici jedinica lokalne i područne (regionalne) samouprave, IJF.

([40]) The ILO Laborsta database puts employment in public enterprises at 12.5 % of total employment in 2008 and results obtained from ORBIS firm-level data come close to this estimate. An ad hoc non-exhaustive survey carried out by the authorities in the context of implementing the anti-corruption strategy pointed to a lower proportion (7.2 %).

([41]) The figures refer to alternative definitions of public ownership (>25 %, >50 % and >90 %) and sectoral classifications (19 and 60 NACE activities). The estimates are based on a regression of the average (log) level of TFP, in constant prices, on a series of dummies capturing the sector of activity, average number of employees and public ownership. An alternative study, using different data and methods, estimates that in 2012 labour productivity in public enterprises was 40 % lower than in private enterprises. See Iooty, Correa, Radas and Skrinjaric, ’Stylised Facts on Productivity Growth’, Policy Research Working Paper 6990, World Bank, 2014.

([42]) ʻStrategic’ companies of those with revenue-generating powers in which the state participates in determining the prices of products or services. They include companies charged with road or waterway maintenance, railway maintenance and passenger transport, electricity generation, oil and gas transport and storage, forestry, the national lottery and military trade. Companies of ’special’ interest are other large companies such as airports, ports, Croatia Airlines, hotel resorts, Croatian Post, Railways Cargo, banks, etc.

([43]) Frederick, W. R., Enhancing the Role of the State Boards of Directors of SOEs, OECD Corporate Governance Working Papers, No 2, Paris, 2011.

([44]) Carpus Carcea, M., Ciriaci, D., Cuerpo, C., Lorenzani, D., and Pontuch, P., Economic impact of rescue and recovery frameworks, note for LIME working group, 2013.

([45]) Koscak, S., and Besevic-Vajo, M., Predstečajne nagodbe: Svi planovi podbacili;             http://www.banka.hr/komentari-i-analize/predstecajne-nagodbe-svi-planovi-podbacili, downloaded 7.10.2014.

([46]) In 2014, there were 1 246 pre-insolvency cases, as compared with 1 266 in 2013. Resolved pre-insolvency cases increased from 700 in 2013 to 1 285 in 2014. In the same period, pending pre-insolvency cases decreased from 566 to 527 (data from the Ministry of Justice). In 2014, the average length of proceedings in pre-insolvency cases was 130 days.

([47]) The number of incoming insolvency proceedings decreased significantly from 6 219 in 2013 to 2 641 in 2014 and, while the number of resolved cases decreased slightly (from 3 900 to 3 239), the backlog was reduced by 12 %. In 2014, the average length of insolvency proceedings was 421 days. Source: Ministry of Justice.

([48]) On January 15 the government approved a scheme to write off the debts of the poorest citizens. Under the programme, debts worth no more than HRK 35 000 (EUR 4 550) will be scrapped for citizens who have monthly incomes of less than HRK1 250.

([49]) The measures included ceilings to lending growth, reserve requirements on banks’ foreign liabilities and a liquidity reserve on banks’ foreign currency-denominated liabilities. Against this background, the total lending of the Croatian banking sector is below the EU average, at around 70% of GDP, part of the risk related to the high leverage of the corporate sector remains with foreign parent banks, and domestic banks are equipped with a pool of foreign currency liquidity, which can be seen as complementary to the CNB’s official reserves.

([50]) An improvement in macroeconomic fundamentals, notably export competitiveness and a broadening of the export base, is essential for underpinning exchange rate, and thus financial, stability in the medium- to long term.

([51]) Zagrebačka banka d.d., Privredna banka Zagreb d.d., Erste&Steiermärkische Bank d.d. and Raiffeisenbank Austria d.d. are subsidiaries of Italian and Austrian groups and therefore belong to euro-area banking groups. They account for two-thirds of the total assets of the banking system in Croatia.

([52]) As a result of the reclassification of loans and the review of collateral values under the AQR, the NPL ratio has increased from 11.3% to 12.7%, while the coverage ratio has fallen from 57.1% to 54.6%.

([53]) The methodology is similar to that adopted under the SSM AQR, but differs from it in a number of respects, e.g. approach to sampling. The final report was communicated to the Commission at the end of January.

([54]) In practice, there is some degree of credit risk associated with HBOR’s on-lending activities. For example, in its Credit Opinion of 29 July 2014, Moody’s notes "that high levels of customer delinquencies in the system have prompted HBOR to agree to restructure some loans it had granted to banks in order to assist these institutions in restructuring end-borrower debts".

([55]) The supplement was introduced in 2007 to reduce differences in pension levels for pensioners who retired before and after the 1999 reform. The supplement is 4 % for pensions acquired in 1999 and 27 % for those acquired in 2010 and after. It does not accrue to pensioners who are insured under the second pillar or belong to special categories.

([56]) Only 8 other Member States have the early retirement gap equal or larger than 5 years. The average gap is below three years. See DG ECFIN, Identifying fiscal sustainability challenges in the areas of pension, health care and long-term care policies, European Economy, October 2014..

([57]) Pensions from special schemes above HRK 5 000 were cut by 10 % as of 2014 (resulting in savings of 0.1 % of GDP), albeit on a temporary basis, while their indexation has been tied to a GDP trigger. The reduction will be automatically revoked once real GDP growth exceeds 2 % over three consecutive quarters compared with the same quarter in the previous year, and the budget deficit is below 3 %. The cut does not apply to war veterans with 100 % disability or to surviving children of war veterans, but the indexation freeze does.

([58]) Stubbs, P. and S. Zrinščak, Investing in Children: Breaking the cycle of disadvantage, country report, Croatia, EU Network of Independent Experts on Social Inclusion, 2014.

([59]) Croatia has one of the highest shares of adults reporting to smoke daily (OECD, ‘Health at a Glance: Europe 2014’). For the importance of life-style factors, see Erasmus University, Erasmus Medical Center and the Dutch Institute for Public Health and the Environment (RIVM), Comparative efficiency of health systems, corrected for selected lifestyle factors, 2015.

([60]) According to Transparency International’s Global Corruption Barometer, 2013, 61% of respondents in Croatia felt that medical and health services were corrupt or extremely corrupt. The issue receives attention in the new anti-corruption strategy (measure No 1 in chapter 2.2.5 "Healthcare" concerns informal payments).

([61]) The World Bank and Croatia have initiated a ‘Programme for Results’ project for improving the quality and efficiency of healthcare. Action aimed at reducing hospital referrals and increasing the accessibility of health services are also supported by EU funds.

([62]) Specifically, by 2016 acute hospital care should decrease by 10 %, day cases and outpatient care should increase by 10 %, the average length of hospital stays should be reduced by 10-40 % in individual hospitals and the average bed occupancy rate should rise to 85 %.

([63]) From February 2015, hospital managers will be expected to submit detailed implementation plans to the Ministry of Health, which will evaluate them and propose changes as needed.

([64]) Next‑generation access capable of providing high‑speed internet (at least 30 Mbps download) was available to only 33 % of homes (EU: 62 %). The price for fixed broadband is 37 % higher than the EU average and subscriptions are 21 % lower. Competition between fixed internet platforms is 17 % lower than the EU average. 68 % of households had a broadband subscription in 2014 (EU average: 78 %).

([65]) Arandarenko M. and Bartlett W., Labour Market and Skills in the Western Balkans, Foundation for the Advancement of Economics, LSEE, London, 2012, p. 1-224.

([66]) See Arpaia, A, Kiss, A. and Turrini A., Is unemployment structural or cyclical? Main features of job matching in the EU after the crisis. European Economy, Economic Papers 527, 2014; and Bejaković, P. and Mrnjavac, Ž. Skill Mismatches and Anticipation of the Future Labour Market Need: Case of Croatia, Zagreb International Review of Economics & Business, Vol. 17, No. 1, pp. 47-68, 2014. A horizontal mismatch refers to a mismatch between the field of study and the job, whereas a vertical mismatch refers to a mismatch between the level of education and the job. 

([67]) Božić L. J., Constraints to innovation activities in Croatian Enterprises, The Institute of Economics, Zagreb, No 62 (3-4), 2011, p. 177-189.

([68]) OECD, OECD Reviews of Innovation Policy: Croatia 2013, 2014.

([69]) The following categories are used to assess progress in implementing the 2014 CSRs of the Council Recommendation: No progress: The Member State (MS) has neither announced nor adopted any measures to address the CSR. This category also applies if the MS has commissioned a study group to evaluate possible measures.            Limited progress: The MS has announced some measures to address the CSR, but these appear insufficient and/or their adoption/implementation is at risk.           Some progress: The MS has announced or adopted measures to address the CSR. These are promising, but not all of them have been implemented yet and it is not certain that all will be.              Substantial progress: The MS has adopted measures, most of which have been implemented. They go a long way towards addressing the CSR.             Fully addressed: The MS has adopted and implemented measures that address the CSR appropriately.