State Aid in the New EU Member States

In the early phase of transition, which began in the 1990s, Central and Eastern European countries (CEECs) pursued economic restructuring that involved massive injections of state support. With reference to the history of state aid in centrally planned economies, we investigate state aid practices of CEECs since attaining full EU membership. We analyse whether their state aid policies during and after transition challenged European state aid legislation, and whether these fit into the EU strategy of &#8216;less but better targeted aid&#8217;. The data&#8208;based analysis is complemented with some indicative insights from state aid in the steel industry as well as the financial service sector to suggest that there is today no significant difference in state aid law application between East and West any more &#8211; the new EU members have further caught up by better aligning to the objectives of the State Aid Action Plan.


Introduction
Initially, non-compliance with European state aid rules was commonplace in many Member State governments during the 1960s and 1970s. It took the Commission many years to establish and enforce even the simple rule of notification for state aid schemes. For many years, the Member States were reluctant to honour the Commission's decisions or European Court of Justice (ECJ) rulings in such cases. This changed only gradually with the introduction of the common market from the 1980s onwards. In 2004 and 2007, ten new Member States from Central and Eastern Europe Countries (CEECs) joined the EU. These countries all had to transform from centrally planned economieswhere state aid is an essential elementto market economies. With the accession to the European Union, the acceding countries transferred EU law into national law and changed their political practices regarding levels of state aid and the objectives for which they are used. Almost simultaneously with the Eastern enlargement, the Commission announced its plans for a comprehensive reform of the EU state aid rules and procedures (the State Aid Action Plan, SAAP), with a view on a redirection of state aid towards horizontal objectives.
The objective of the article is to review the trends in state aid in new the EU Member States. The analysis shows that the new EU Member States have gradually improved compatibility with European state aid objectives. Using selected case studies in the steel sector, the article establishes that increasing compliance is a result of a redirection of state aid to growth-enhancing instruments by national state aid policy in new Member States *We wish to thank the two anonymous referees and the Editor, Michelle Cini, for their helpful comments and suggestions. We wish to acknowledge support by the Economic and Social Research Council project 'Access to Finance for SMEs' under Grant number ES/M002462/1. The copyright line for this article was changed on February 15, 2017 after original online publication.
which was accompanied at the same time by investigations and sanctions by DG Competition. The financial sector, even though apparently moving away from the otherwise applicable European objectives of state aid is well in line with policy in other European states. In an earlier study, Hölscher and Stephan (2009) revealed that the effectiveness of implementation of competition law and policy is at lower levels in CEECs compared with the 'old' EU-15 Member States. With regard to state aid, CEECs have a strong tradition of government intervention resulting from the planned economy system. With the start of transition, state aid continued to be an important element in national policies. State aid then smoothed the process of (re-)gaining competitiveness and therefore mitigated negative social externalities. Although the need for state aid during that time was recognized by the Commission (Sharp, 2003), the accession treaties required adjustment of national policies according to European state aid rules in the acquis communautaire.
Even at the dawn of EU accession, CEECs' levels of public support to certain industries and individual firms were significantly higher than in the EU-15 Member States. This raised the expectation of political discussions at a European level (see Röller and Hirschhausen, 1996, bearing in mind the experience from East Germany's attempts to revive its industries by massive state intervention and expenditure).
Contrary to these expectations, the compliance literature in other policy fields suggests that there is no significant lag in application and institutional enforcement of EU law among the CEECs either prior to or after accession. Falkner and Treib (2008) found a good pattern of transposition of EU law into national law and Sedelmeier (2008) found that CEECs even outperform EU-15 Member States regarding the number of infringements of EU law. With respect to state aid rules, Blauberger (2009) found that compliance of the new Member States does not happen immediately. It is, rather, the result of the enforcement powers of the Commission setting in with accession to the EU. Blauberger draws the conclusion that the 'carrot of conditionality has been replaced by the more forceful stick of regular Commission control' (Blauberger, 2009, p. 1,043).
This article contributes to the literature on European integration and compliance in the field of state aid law. Following a review of post-accession compliance of new Member States, differences regarding the discretion of state aid policy in East and West are analyzed. First, we present the state aid practices of CEECs with reference to the history of state aid in centrally planned economies; second, we analyze whether CEEC policies challenged European state aid legislation and the objectives of the SAAP. Furthermore, infringements of state aid rules and the orientation of national state aid objectives are explored. The analysis is essentially based on critically reviewing the policy debate around the reforms of European state aid policy and its enforcement. This is complemented by data analysis 1 of state aid in East and West. This study is essentially of an empirical nature, but attempts to embed descriptive statistical analysis into explanatory frameworks and concepts. 1 We based our analysis in general on data provided by the European Commission, published in the annual State Aid Scoreboard. There are other sources providing figures on state aids such as the System of National Accounts (SNA), OECD and WTO data sources. However, in order to provide comparable data over time and countries, we based our analysis on the EU data, as only EU Member States are the subject of this study. The SNA does not distinguish between different sectors; neither are types of measures separated. EU data afford dependable comparison.

State Aid in Centrally-planned Economies
The countries in Eastern Europe have a strong tradition of government intervention resulting from the planned economy system. State aid to industry has been a common support for state-owned enterprises. These countries were used to compensating firms for the losses incurred by producing under a distorted price system and to ensure full employment of the labour force as a political objective. On top of this, private households received indirect support, stemming from social and equity considerations, leading to heavily subsidized consumer goods, living space and energy supply (OECD, 1992). Thus, the scale of total subsidies was immense as both the price subsidies and the subsidies granted to state-owned enterprises constituted a large share of public expenditure (Schaffer, 1995).
Even though some CEECs had already commenced reforms and cut their state aid in the 1980s, the most significant changes began with the transition in the early 1990s. Although there was a dramatic drop in the level of subsidies (to 3-5 per cent of GDP, Schaffer, 1995), they still remained well above the Western European average (Schwartz and Clements, 1999). Besides the general cut in state aid, in the early 1990s CEECs focused their remaining public support on new activities which had not existed in the Communist system. In order to develop a market economy with a competitive and dynamic firm structure, state aid was redirected at SME programmes (tax incentives), attracting foreign direct investment and export promotion (OECD, 1998). Thus, an initial change towards more horizontal objectives could be observed. Nonetheless, the difficulties during the transition phase led to further state aid programmes directed at regional development and, especially, at rescue and restructuring measures. Such ad hoc aid measures to enterprises had been crucial in the transition of CEECs, as state aid mitigated negative social externalities (Nitsche and Heidhues, 2006). However, restructuring measures to enterprises were not connected to any long-term development policy but to rescuing enterprises in difficulty 'in order to speed up the adjustment process and save as many ... enterprises as possible' (Hashi, 2004, p. 10). At this stage, tax arrears became a workable instrument of state aid policy. Schaffer (1995) found that mainly state-owned companies with low profitability received such indirect tax subsidies, which enabled them to continue to operate.

State Aid and the Accession of CEECs to the EU
While the transition from planned to market economies initiated a first change in CEECs state aid practices, the signing of the Europe Agreements led to a second wave of major changes. The Europe Agreements, which were signed between 1991-96, required CEECs to align their legislation and institutions with the large and complex body of EU state aid legislation in order to prepare for accession. National industrial policies in general had to be redirected and aligned with EU policy objectives (see Botta and Schwellnus, 2015 for a qualitative comparative analysis of EU conditionality on the development of state aid in candidate countries). Furthermore, state aid expenditures had to be made transparent (Atanasiu, 2001). Between 1997 and 2001, 2 CEECs adopted national state aid laws and started to build up their own state aid monitoring authorities, either within national competition authorities or within the Ministries of Finance. 3 Although their mandate of state aid control was limited in time, because the authority would be transferred to the European Commission upon accession, monitoring by national institutions and the prospect of EU membership were a necessary complement of effective state aid control at the national level (Botta, 2013). This process allowed CEECs to learn to play the (state aid control) game (Cremona, 2003, p. 287).
In practice, these state aid monitoring authorities faced the problem that they were not endowed with a proper legal status and thus had no power to enforce the state aid rules: the Hungarian State Aid Monitoring Office, for example, lacked the power to stop state aid practices by the government, to enforce negative decisions or to recover unlawful aid (Hargita and Filep, 2004).
In order to give candidate countries enough time to align their state aid practices with the acquis communautaire, two concepts were installed: existing aid mechanisms and transitional arrangements. The existing aid mechanism-concept had already been used in 1995, when Austria, Finland and Sweden joined the EU and rendered legal all aid that was granted by acceding countries before accession (Regulation 659/1999). Such 'existing aid' was protected from recovery and to qualify, it had to fulfil one of the three requirements 4 : (a) the measure had to be put into effect before 10 December 1994; (b) the measure had to be listed in the Annex of the Accession Treaty as lex specialis; and (c) the measure was approved by the State Aid Monitoring Authority prior to the Accession; measures approved by the Commission under the 'interim-procedure' are also considered as existing aid.
The first category, sometimes referred to as 'grandfather measures' (Roebling, 2003), are today negligible in their scope, after all, state aid measures rarely remain unchanged over a period of ten years or longer (Geradin, 2004). The second type of existing aid (special arrangements) comprised 222 measures regarding aid in the Czech Republic, Hungary, Slovenia, Cyprus and Poland. The interim mechanism under the third category took account of the structural adjustments during transition in the CEECs.
Next to the existing aid concept, a number of additional transitional arrangements were negotiated with each candidate country. The transitional framework included: temporary rules on incompatible fiscal aid for Poland, Hungary and Slovakia as well as sector specific exemptions regarding the restructuring of the steel industry in Poland and the Czech Republic (for an overview, see Birnstiel and Heinrich, 2011). Poland was granted a number of additional exemptions regarding state aid for environmental protection. Only the Baltic States and Slovenia did not conclude transitional agreements.
Transitional arrangements for the steel industry allowed state aid in the steel sector within the first five years after the entry into force of the Europe Agreements, the so-called 'grace period', which was extended until the end of 2003 (Hancher et al., 2012). They were agreed upon despite the rather strong opinion of the Commission against state aid in the steel sector and its strict enforcement in the West (Steel Aid Code, 5 even reinforced in 2002 with the Rescue and Restructuring guidelines for the steel sector 6 ). Yet, the steel industry in candidate countries was in dire need of significant restructuring to become internationally competitive, and transitional arrangements were designed to assist here, but some of the old Member States perceived it as a competitive disadvantage that CEECs were granted less strict rules (Trappmann, 2013, p. 70). In order to reduce the impact of the unlevel playing field, several conditions were attached to those transitional arrangements, in particular with a view to re-establishing commercial viability and a commitment to reduce capacity.

The Modernization of European State Aid Rules
Almost simultaneously with the accession of the CEECs, the State Aid Action Plan with its objective of 'less and better targeted state aid' (Wishlade, 2006, p. 233) gave rise to a third change in CEECs state aid-policy in 2005. The SAAP was aimed at encouraging Member States to reduce overall state aid levels whilst redirecting state aid to objectives more clearly targeting community interests. Yet, this was not a pure coincidence, as the inclusion of the new Member States necessitated a significant overhaul of EU institutions to prepare for enlargement, to change regulations to account for the increased diversity brought in by new members, to 'ensure an effective control in the enlarged Union' (European Commission, 2005, p. 5). To improve the effectiveness of enforcement of state aid regulations not least in the new Member States, the reforms aimed at more transparency, intensified co-operation between the Commission and the national competition agencies, and a stricter focus on cases with the biggest impact on competition. A further challenge for SAAP was the need for a renewed impetus to the so-called Lisbon Strategy and the assumed role of a reformed state aid policy for this (redeployment of aid in favour of research and innovation, human capital, and regional aid measures to reduce disparities, or in general: overcoming market failures to untap Europe's potentials). Of the many measures discussed in this consultation document, SAAP suggests the de minimis rule, 7 the General Block Exemption Regulation (GBER) 8 and refocused regional aid policy. Figure 1 shows that the amount of aid as a share of GDP in the candidate countries had been reduced to some 1.1 per cent of GDP in the year 2000. In comparison, the EU-15 had already reached a level of some 0.4 per cent per GDP at the end of the 1990s and have maintained this level ever since. The convergence in aid intensities until 2000 between East and West experienced a temporary setback in the years just before accession. This contradicts the assumption of better compliance in the East based on the incentive of accession (Schimmelfennig and Sedelmeier, 2004). It is, however, explained by intensifying the restructuring activities of large state-owned enterprises (SOEs) which had, so far, not attracted investors in privatization programmes and now had to be reorganized. This is 7 In 2006, the Commission published the Regulation on small amounts of aid exempting aid of a very low amount from the notification obligation. 8 In 2008, the Commission adopted its GBER specifying the rules for cases with limited effect on trade (R&D, environment, training, etc.). In May 2014, the Commission revised the GBER and extended the scope of exemptions from prior notification of state aid granted to companies. The Commission estimates that the GBER covers approximately 60 per cent of all aid measures and slightly more than 30 per cent of the total aid granted each year in the EU (European Commission, 2014). mainly driven by Poland, where over 7.6 billion Euros were handed out in 2003 (up from 1 billion the year before and 2.8 billion in 2005), and to some extent in Romania with 2.3 billion Euros in 2003 and nearly 2.9 billion in 2004 (yet here, levels have dropped sharply thereafter). Institutionally, this surge in the run up to full membership was, in fact, mostly consistent with EU law, as the above-mentioned transitional arrangements allowed several CEECs to phase out state aid with the aim of restructuring industries such as steel and shipbuilding. Still, it remained highly controversial and in the end also may have had a role in triggering the SAAP.

'Less Aid'
By the time of accession to the EU in 2004, the aid level per GDP among the CEECs had dropped sharply and had almost reached Western European levels, suggesting that CEECs by then had further aligned the intensities of state aid policies to EU standards. Yet, the aim of the SAAP to further reduce the level of aid in all Member States has not yet been reached. In 2013, levels of aid in relation to GDP reached 0.38 per cent in the EU-15 countries and 0.66 per cent in the CEECs. Since 2005, when the SAAP had been published, the intensity of aid has increased slightly in CEECs and remained stagnant in the EU-15. Important to note is that this interpretation of GDP-related development of state aid over time masks differences in GDP growth during this period of time between East and West: Poland, in parallel to its high GDP growth did little to reduce its state aid expenditures in monetary terms, likewise the Czech Republic and Hungary retained levels of non-crisis aid for industry and services (total non-crisis aid less agriculture, fisheries and transport) at levels around 1 billion Euros. Other countries, like Lithuania and Slovenia increased their state aid expenditure year-by-year. And yet, the objective of a level playing field has to be tested against aid intensities (as aid expenditures per GDP) and turns out to be rather heterogeneous between CEECs ( Figure 2). Country-variations exist between Hungary, Slovenia, Poland and the Czech Republic at the higher end of the spectrum, and Estonia and Bulgaria at the lower end.
Whereas some of the CEECs show a declining trend in state aid intensities, others (Slovenia, the Czech Republic and Latvia) show an upward trend in public expenditure per GDP. Whilst Hungary used to show the highest intensities of aid among all CEECs until 2010, the country has been surpassed in 2012 and 2013 by Latvia and Slovenia, as aid intensities in these two countries increased from 2011 to 2013 (Figure 3).  The increasing aid intensities raise the question of whether this development is triggered by the fact that the disciplining effect of the prospect of conditional EU membership has vanished following accession (Schimmelfennig and Sedelmeier, 2004). The economic effects of membership may also have forced the new members to increase their aid in order to cope with the more competitive situation. In Slovenia, Lithuania and the Czech Republic, the increase can be traced back to an increase in aid devoted to regional development. In Latvia, the increase in 2012 and 2013 is caused by aid for cultural activities. In Hungary, the amount of aid intensities decreased in 2011 and 2012, as aid for sectoral development had been cut from €900 million in 2010 to €34 million in 2013, whilst aid for R&D increased from €30 million to €300 million. However, state aid in Hungary is still twice that of the EU-15 average at around 1.0 per cent of GDP.
To sum up, the centrally planned economies of Eastern Europe pursued enormous amounts of aid prior to their transition in the 1990s. Two major factors have determined the development of their state aid policies: transition to market economies and accession to the EU. This has led to a framework of formal institutions which did not, at least initially, correspond to informal norms and values of their socialist legacy: still, their aid intensities decreased to 0.7 per cent of GDP (compared with 0.4 per cent in the old Member States). However, the EU-15 Member States did not further reduce their state aid following the announcement of the SAAP in 2005. In 2013, all EU Member States (EU-27) still spent €54 billion (without crisis measures) to support their national industries. More insights on the objectives of the various measures will be given in the next section.

'Better Aid'
Besides the aim of less aid, the Commission further formulated in its SAAP the aim of 'better targeted aid' (Wishlade, 2006, p. 233). According to this, Member States are required to reduce aid awarded to particular sectors and firms, as these types of aid distort competition the most, leave unviable firms in the market and hinder the efficient allocation of resources. Instead, governments are required to redirect state aid towards horizontal measuresin particular, those which contribute to the aims of the Lisbon Agenda (and the successive Europe 2020 Strategy). 'Aid shall support flagship initiatives of the Europe 2020 strategy for smart, sustainable and inclusive growth in a pro-competitive way, maintaining a level playing field in the internal market' (European Commission, 2012a). The focus of horizontal measures is mainly on R&D, environmental protection, training and employment measuresthat is, more medium-to long-term measures. As of July 2013, the definition had been enlarged to include new forms of state aid, such as, for example, the conservation of cultural heritage, repair of damages caused by natural disasters. Regional aid had received renewed attention in SAAP, and in fact, for long had been deemed as being largely non-distorting, due to its focus on the less developed regions in Europe (below 75 per cent of the EU average) and because a large share was spent under the general block exemption regulation.
Prior to their accession to the EU, most CEECs pursued high levels of sectoral aid, as these countries first had to overcome the problems resulting from transition. Thus, disparities between East and West before accession can be explained in part by the restructuring of industries in order to reach commercial viability and to complete the process of privatization.
As Table 1 shows, the share of sectoral aid in the CEECs increased prior to accession, whereas the share of horizontal aid decreased. This can be partly explained by transitional rules which focused on aid to specific sectors. Yet, following accession the contribution of aid awarded to sectoral objectives had already halved between 2004 and 2005, and decreased further in the subsequent years. Since 2003, the year showing the highest level, the share of sectoral aid to industry decreased from 82 per cent to 7 per cent in 2013. Accordingly, the share of horizontal aid increased from 11 per cent in 2003 to 58 per cent in 2013. Due to the reform of the definition of horizontal aid, the shares for 2013 may exaggerate the positive trend. Other changes in the methodology of the Scoreboard data make it difficult to analyze trends over time (and differences between countries), as, for example, from 2012, some of the aid for services was categorized as 'Services of General Economic Interest' (hospitals, healthcare, childcare, social housing, airports and ports, etc.) and hence not deemed as aid, producing a statistical downward trend effect.
From an individual country perspective, almost all Member States have reduced sectoral aid to a minimum in 2013, with the exception of Spain (35 per cent) and Ireland (31 per cent). The most pronounced differences in this respect are not observed between CEECs and the EU-15 Member States but, rather, between individual member countries ( Figure 4).
Such a disparity between countries has been also observed in other publications. Börzel (2002) argues that Southern Member States lag behind in promoting European market liberalization and, instead, pursue strong national industrial policies leading to higher levels of state aid. Molina and Rhodes (2007) found that Southern countries stand between liberalized market economies and co-ordinated systems of economic governance. Yet, Zahariadis (2010), comparing state aid in Greece with the other Western EU members, argues that Greece (as, in fact, other countries showing a 'Southern Problem', Pridham and Cini, 1994) cannot be compared with most Northern EU countries due to the divergent political, economic and institutional developments between Europe's North and South. Figure 5 shows that the priorities in the type of horizontal aid vary significantly between CEECs and the EU-15. Yet, even though the share of horizontal aid doubled among CEECs from 2000 to 2013, employment aid remained an important focussupport for SMEs fell 9 at the expense of environmental measures. Among the EU-15 Member States, a large share of horizontal aid is devoted to R&D projects and environmental issues, their shares increased between 2000 and 2013, whereas SME support programmes  also decreased in the West. These trends regarding the distribution of horizontal aid measures and their consistency with the Lisbon Agenda and the Europe 2020 Strategy are the subject of the next section.

State Aid and the Lisbon Strategy
Since the early 2000s, the attitudes towards state aid have increasingly been influenced by the aim of making the EU 'the most competitive and dynamic knowledge-based economy in the world capable of sustainable economic growth with more and better jobs and greater social cohesion' (see Lisbon Agenda and the successive Europe 2020 Strategy, European Commission, 2000). The intellectual basis of this is to be found in the new growth theory (Romer, 1986(Romer, , 1990 which implies that long-term positive effects are possible if government interventions promote R&D and human, social and public capital. Investments in R&D contribute to competitiveness through product and process innovations and may induce spillovers between enterprises which, in turn, are essential for long-term economic growth (Grossman and Helpman, 1992). Market failures lead to underinvestments in R&D, because private companies remain unable to achieve the full internalization of the spillovers generated by their knowledge-activities. Public funds to support R&D are one strategy to induce a higher, economically efficient level of R&D by remunerating the generation of externalities. In its SAAP, the Commission identified the following policy areas which are essential for growth and motivated by market failures: research, development and innovation, environmental renewal, lifelong training, SMEs and social equality. Whilst levels of state aid had, in fact, been reduced and sectoral aid had gradually been replaced by horizontal aid, was horizontal aid also redirected towards measures that support the aims of the Lisbon Strategy or the Europe 2020 Strategy? Figure 6 shows that EU-15 members did, indeed, increase the amount of aid directed towards such objectives after 2000. However, this trend stopped in 2005, when the SAAP was published. In the CEECs, the amount of 'better aid' as a percentage of GDP did not increase much, whilst there is a long-term trend in the proportion of 'better aid' in total aid. In addition, the CEEC's level of 'better aid' is still far below those of the EU-15 Member States.
Within those country groups, there is no clear East-West divide: Latvia, Slovenia and Slovakia are among the 10 Member States spending more than 50 per cent of state aid towards Lisbon objectives, whereas Greece and Portugal are among those countries spending less than 10 per cent towards 'better aims'. Whilst Greece supports regional schemes, the majority of Portugal's aid went into the financial sector (these figures still do not include the crisis measures since 2008).
The reorientation of industrial policy is necessarily a time-consuming process. Public funds to support R&D were almost non-existent among EU Member States up until the 1970s. The launch of the first R&D programmes was, to a large extent, a reaction to the economic crisis of the 1970s. However, this has led to a minor increase in measures directed at R&D activities (European Commission, 1989, point 48). A more sizeable trend towards public support of R&D activities can only be observed from the year 2005 onwards. The same protracted development holds for public support to SMEs, accounting for more than 90 per cent of all enterprises in the EU and employing more than 50 per cent of employees (European Commission, 2008). Yet, for a long time, state aid measures were not targeted at the specific needs of SMEs. The 1970s especially were characterized as the era of national champions, predominantly in the steel, shipbuilding and car manufacturing industries (Maincent and Navarro, 2006;Sharp, 2003).
As the examples of Latvia, Slovakia and Slovenia show, an initial reorientation of CEEC industrial policies towards objectives supporting the Lisbon Strategy can be observed. This reorientation was however challenged from 2008 onwards when the economic and financial crisis induced many Member States to redirect state budgets to support their financial sectors. The following section will assess whether there were any noticeable differences between CEECs and old Member States.

Effects of the Financial Crisis on the Target to Spend 'Less and Better Aid'
Prior to the 2008 economic and financial crisis, the banking sector was only twice subject to state aid investigations by the European Commission. In 1997, the European Commission found in the Credit Lyonnais case (France, with costs of up to 2.5 per cent of GDP) that the aid measures were compatible with the common market. The second case concerned the German WestLB and required the recovery of up to 3 billion Euro.
In contrast to these two reference cases, the financial crisis affected a large number of banks and questioned the state aid principles for banks in force so far. In general, state aid to banks can generate positive externalities, because banks are interdependent, in some cases possibly relevant for the whole financial system and possibly beyond (systemic importance). State aid to ailing banks can provide the necessary liquidity to avoid contagion (see Vives, 2011) and re-establish trust in the whole financial system. However, this policy still distorts competition and might, in the worst of cases, lead to good banks being outcompeted by bad banks which receive support (Vives, 2011). The issue of state aid in the financial sector during a financial crisis allows for two opposing views (see Beck et al., 2010). One view prioritizes financial stability and fights contagion. The other view focuses on the risk that a generous state aid policy leads to massive distortions of competition. By offering a temporary framework for crisis measures, DG Competition found a balance by allowing more state aid, yet only temporarily to account for the crisis and the aid was conditional on adherence to three main principles: viability, burden sharing and competition (Kok, 2015). Some observers argue that the Commission's approach was rather lenient, it allowed exceptional rescue measures (Hasan and Marinc, 2013). Besides, the phasing out of the temporary measures was twice postponed, based on the perception that the financial crisis was not over.
With the start of the economic and financial crisis, the European Commission has taken 22 decisions in 2008 and 81 decisions in 2009most of them approved without objections. In the period between 2008 and 2014, the Commission approved aid to the financial sector with an overall amount of 4.9 trillion Euro (5 per cent of EU GDP), of which around 2 trillion were actually used (Figure 7).
Subsidies to the financial sector were less sizeable in the new Member States as compared to old Member States of the EU. Most financial institutions in CEECs are in fact subsidiaries of western banks, and their loan portfolio initially was of a lesser top-risk character and hence did not look as fragile as in most western banks. With recessions (or in the case of Poland exchange rate devaluations) setting in, those banks likewise had to file for support.

Sectoral Aid in the Steel Industry
The steel industry in CEECs was subject to several investigations by the European Commission after accession. The second largest Polish steel producer 'Huta Czestochowa' (HC) was subject to one of the first investigations. HC was unable to benefit from the transitional rules negotiated by Poland for its steel industry, because the company was not listed in Protocol No. 8 of the Accession Treaty, which contained all companies active in the Polish steel industry. Had the company been on the list, it would have been entitled to receive restructuring aid (in fact, the company was taken off the list of beneficiaries in the last minute due to its financial difficulties). Initially, it was the aim of the Polish government to liquidate the company (Saryusz-Wolska, 2010). However, the government decided in 2003 to restructure the company with a view to a subsequent privatisation.
In 2004, the Commission opened a formal investigation concerning state aid to HC. The final decision of 5 July 2005 found that certain measures in the period prior to the restructuring (1997)(1998)(1999)(2000)(2001)(2002) were incompatible with the internal market and the Commission ordered recovery of 4 million Euro.
The case has been novel as it concerned a pre-accession aid measure and was debatable due to the fact that an exhaustive list of other companies in the Polish steel sector were allowed to receive state aid by being listed in the relevant protocol. HC argued before the General Court that the Commission did not have the power to investigate, as the state aid was granted between 1997 and 2002. The Commission on the other side argued that Protocol No. 8 is a lex specialis and extends monitoring to any aid granted for the restructuring of the Polish steel industry between 1997 and 2006. The Court agreed with the Commission and thus Protocol No. 8 builds the legal basis for the Commission to order recovery.
In 2006, the Polish steel industry was subject to another investigation. Polish steel producer 'Huta Warszawa' (HW) received about 50 million Euro of state aid mainly in the form of a state guarantee for a loan to pursue its investment plan. However, HW had used around 30 million Euro of the loan in 2004 to pay off old debts. The Commission found that this measure was neither indicated in the restructuring plan nor necessary for the restructuring, but would rather endanger the company's return to viability. The Commission therefore concluded that part of the aid had been misused and was incompatible with the Single Market. In 2005, Arcelor had taken over HW and repaid the guaranteed loan and updated the business plan of the company in order to ensure long-term viability. The Commission however held that the advantage stemming from the guarantee which they claimed constituted an interest subsidy amounting to around €2 million. In order to settle the case, the company (AHW) agreed to repay also this amount.
Another case worth mentioning is state aid to the Bulgarian steel producer Kremikovtzi. Comparable transitional rules allowing for restructuring aid in the steel industry were also negotiated by Bulgaria. Accordingly, the largest Bulgarian steel producer was expected to undergo a comprehensive technical modernization in order to regain viability until the end of 2006 and therefore obtained restructuring aid totalling 222 million Euro. The initial restructuring period was extended by two years until 2008. However, Kremikovtzi was declared insolvent in August 2008 as the company was no longer able to service its debts. The Commission came to the conclusion that the business plan to restore long-term viability was never implemented properly. Only 43 per cent of the investment plan was realized. The Commission also argued that the fact that the company was bankrupt should be evidence enough that the restoration of long-term viability could not be achieved (European Commission, 2009). Besides Poland and Bulgaria, steel producers in the Czech Republic were also subject to state aid investigations. Yet, in contrast to the other cases above, state aid measures regarding steel companies Frydek Mistek and Trinecke Zelezarny were approved by the Commission. The first company requested a six-month postponement of the required capacity reductions and the Commission had no objections, because this postponement remained within the restructuring period until the end of 2006. In addition, an improved economic situation led to temporary profitability of the steel mill which helped it to improve long-term viability. The second Czech case concerns aid measures to Trinecke Zelezarny (TZ), the only producer of rails in the Czech Republic. TZ was privatized in the mid-1990s and had been fully restructured without state support. In 2004, the Czech government planned several aid measures to TZ, namely aid measures for training activities (1.5 million Euro), for a closure project (0.14 million Euro) and aid for the purchase of shares held by TZ. In fact, the training aid was approved by the Commission, and the closure aid was never paid out.
The cases show that the candidate countries indeed made use of their transitional rules negotiated in the Europe Agreements. However, the cases also show that CEECs were not treated any more leniently than their Western counterparts beyond those transitional arrangements.
Even though the European Steel Aid code prohibits aid to the steel sector in general, the database of the European Commission shows that Western European steel producers also regularly receive aid (Figure 8). Aid measures which are allowed comprise measures related to R&D activities or environmental protection. For example, of Source: European Commission (2016), State Aid Cases Database, own research the seven cases regarding Austrian steel producers, only one case concerned investment aid (and was recovered by the Austrian government). The other cases were in line with EU rules, in as much as they defined the aid to be for environmental protection. Figure 8 compares aid measures to the steel industry between CEECs and EU-15 Member States and shows that EU-15 members largely grant aid in order to support environmental or R&D measures while a large proportion of aid cases among CEECs are subsumed as regional aid.
However, there were also aid recipients among Western European members which received rescue or restructuring aid. Out of the 17 cases in the period between 2000 and 2014, the Commission found that aid measures in eight cases were incompatible with the common market, whereof Huta Czestochowa was the only case among CEECs, the other cases were from East and West Germany, Spain and Belgium. The 'Eisenguss Torgelow GmbH' (EGT) in Germany was one of the companies where aid measures regarding restructuring were found to be unlawful. EGT received aid measures between 1998 and 2000 that were not prior notified to the Commission as was required. After privatization in 1993, the company filed for bankruptcy in 1997. The company was sold to a group of investors but restructuring failed and the company filed for bankruptcy again in 2001. On 1 May 2001, EGT was finally declared bankrupt. During the restructuring periods, EGT received state aid measures totalling 9 million Euro, whereof 1.5 million are under scrutiny. The Commission decided that the initial restructuring plan would never have been able to restore commercial viability. This argument is supported by the second insolvency in 2001. Thus, aid to the amount of 1.5 million Euro had to be recovered.

Conclusion
Despite the difficulties in comparing state aid across time and between different countries, certain trends can be identified. State aid intensities declined substantially in Western Europe with the introduction of the Single Market in the mid-1980s. In Eastern Europe, the trend towards lower intensities of public support started after the transformation from the formerly planned economies towards market economies and was further pursued with the accession to the European Union.
In addition, industrial subsidies are increasingly being directed towards horizontal measures, such as R&D and SMEs, rather than to specific sectors. The analysis has further shown that the intended target of the Commission's SAAP of less and better aid could only partially be attained. Member States have, indeed, reduced their overall intensities of state aid, albeit no redirection in the EU-15 could be observed after 2005 when the SAAP was published. In CEECs, there has been an increase towards horizontal objectives in line with the Lisbon Strategy, even if its share in total aid is still far below the standard in the EU-15 Member States and leaves room for improvement. The example of the steel industry has also shown, that sectoral aid is still being used in both the East and West, for example, where it supports R&D measures or serves environmental protection.
As shown, most Member States have progressed in aligning their state aid policies closer to the aims of the Lisbon Strategy. In fact, no worries about the Eastern European new members are justified regarding state aid policy today. It is, rather, the policies of individual and single member countries that warrant concern about a level playing field. Yet, the current financial problems increasingly force governments to redirect and focus their public spending aimed at raising their growth potential.