INTERNATIONAL POLICY ANALYSIS The Corridor Model – Relaunched Short Version KLAUS BUSCH July 2011 „ „ The basic idea underlying the corridor model, developed in the 1990s, is the maintenance of a close connection between levels of economic and welfare state development in the EU member states. During periods of crisis in which drastic cuts are made in social security systems European regulation of this kind is crucial. This concept can also be used to prevent social dumping between member states and to facilitate welfare state catch-up processes on the part of less developed member states. „ „ The corridor model was originally developed with reference to the»social expenditure ratio« indicator. It can be presented more easily on the basis of the»social protection expenditure per capita in purchasing power standards(PPS)« indicator, however, which is related extremely closely to»per capita income in PPS«. Instead of several corridors for different country income groups only one corridor is needed for all states. „ „ There is a close connection between the quantity of financial expenditure and the quality of provision with regard to the various functions of the welfare state. This can also be demonstrated empirically for the health care sector. As a result, the quantitative approach of the corridor model does not conflict with qualitative claims on the welfare state. „ „ The criticism that the corridor model cannot be reconciled with a progressive welfare state philosophy is a misunderstanding. The model neither prescribes nor sanctions upper limits for the welfare state.

Klaus Busch| The Corridor Model – Relaunched Content Introduction ............................................................ 3 1. The Corridor Model: 20 Years Ago and Today .............................. 3 2. Criticisms of the Corridor Model ......................................... 6 2.1 The Connection between Quantity and Quality in the Welfare State and its Significance for the Corridor Model.................................... 6 2.2 Corridor instead of Minimum Standards?................................ 7 3. Outlook ............................................................ 8 Bibliography ............................................................ 9 1

Klaus Busch| The Corridor Model – Relaunched Introduction In the wake of the global economic crisis the welfare state in many European countries has found itself in difficulties unprecedented since the end of the Second World War. In particular in heavily indebted countries the financial markets are enforcing austerity policies bound up with wage cuts, higher unemployment – especially youth unemployment – and drastic cuts in social security, primarily pensions(Heise/Lierse 2011). In the course of their efforts to combat the crisis in the Euro Area(Eurozone) the EU member states agreed a Euro-Plus Pact that for the first time looks at the development of national competitiveness and national wage, pension and health care policies at the European level. This does not involve convergence»while maintaining the improvements made« but a European agreement on the dynamics of the cuts to be made. Germany’s hegemony means that the rest of Europe has to dance to its tune. In this situation it is all the more incumbent on Europe’s left-wing parties and the European trade unions to present clear programmatic alternatives for European economic and financial policy and European wage and social policy. A fourfold paradigm change – involving a radical reorientation of European growth strategy, the EU’s fiscal policy architecture, European coordination of wage, social and tax policy and European regulations on the financing of public debt – is urgently required(Busch/ Hirschel 2011). The corridor model developed at the end of the 1990s is to be regarded as an important building block in this context of alternative policy conceptions(Busch 1998). Its aim is to ensure that a country’s level of economic output and level of welfare state development are linked in the EU, as well as to prevent social dumping and avoid the overburdening of less developed states. The aim of this article is once more to subject the corridor model to a critical examination, to submit new findings, to present a new indicator for the model and to reply to critical objections. 1. The Corridor Model: 20 Years Ago and Today The basic idea underlying the concept developed in the 1990s is to assign member states with different per capita incomes different social expenditure ratios in accordance with their level of economic development, thereby ensuring a close relationship between economy and welfare state(Busch 1998: 178). The aims of an agreement on specific ranges or corridors for the social expenditure ratios of EU member states falling into different per capita income groups are as follows: „ „ Clamping down on social dumping policies. Individual countries would not be able to gain a competitive advantage by maintaining a social expenditure ratio which is below average given their level of per capita income. „ „ This form of social policy regulation would make it possible to avoid the less developed economies becoming economically overburdened. They would have to provide only a level of social expenditure in keeping with their income level. „ „ In the course of the less developed countries’ economic catch-up process social expenditure ratios in the EU would converge; spending on old age, illness, incapacity to work and unemployment would be brought into line not only relatively but also absolutely. The corridors of lower and middle income groups would therefore move upwards on the regression line. „ „ Quantitative regulation of social policy at the EU level would initially be confined to a minimum, with no redistributive elements. Since only aggregated magnitudes (social expenditure ratios) would be regulated in this way member state autonomy in the allocation of social spending to various areas(pensions, health care, unemployment, family support) would be unaffected. In 1993, three income groups were clearly distinguished in the EU11(excluding Luxembourg) and assigned to corresponding corridors with regard to social expenditure ratios. As a consequence of EU enlargement, today(data from 2007) nine groups would be needed, and the delimitation of individual groups would be much more difficult than in the 1990s. Furthermore, the ratio between 3

Klaus Busch| The Corridor Model – Relaunched per capita income and social expenditure ratio has also fallen significantly during the period in question(based on the coefficient of determination), from 81 per cent to 56 per cent. Based on new data from Eurostat on per capita social protection expenditure in the 2000s(PPS), the corridor model can now be presented using a new indicator. Based on the new data the regression»social protection expenditure per capita in PPS« in relation to»per capita income in PPS« can now be calculated(see Figure 1). For the years 2001 and 2007 we get coefficients of determination of an astonishing 87 per cent(2001) and 86 per cent(2007), respectively. 1 That means that the connection between level of economic development and level of the welfare state remains extremely close and much stronger than the data calculated on the basis of the social expenditure ratio would suggest. The difference between the two calculations can be explained as follows: the coefficient of determination is lower in the calculations with the indicator»social expenditure ratio« because here social protection expenditure is related first of all to income in order to be related to income once again in the regression. The variable»income« loses explanatory power as a result of this double»reflection«. Given this very close connection between per capita social protection expenditure and per capita income the corridor model is readily applicable to the indicator»social protection expenditure per capita in PPS«. For each value of per capita income in PPS a corridor could be established with regard to per capita social protection expenditure in PPS which individual states would have to keep to. Instead of a plethora of corridors, which were necessary when using the indicator»social expenditure ratio«, a single corridor is sufficient based on the indicator»social protection expenditure per capita in PPS«. The new approach can easily be explained by means of the regression in Figure 2. In contrast to Figure 1, the outlying country Ireland is here left out of the calculation. In this way, the regression line is turned a little anticlockwise, to some extent functioning as a bisector. The correlation coefficient is now 0.96, thereby showing an extremely strong relationship between the two variables 1. The regression calculations end in 2007 because the data from 2008 onwards are severely impaired by the collapse in economic output in many EU member states(per capita GDP). Cyclically adjusted calculations are therefore needed, but to date none have been made available. (the coefficient of determination is 92 per cent). A point on the y-axis can now be assigned to every point on the x-axis using the formula given above the regression lines: y= 0.3672x – 2.9719. The regression line is the locus of the intersection of the y-target values for social protection expenditure per capita and the x-actual values for GDP per capita for the various countries. The corridor model could now be deployed by extending a line both above and below the regression lines, constructed as follows: for each y-target value for individual states two points are calculated which are five per cent(7.5 per cent, 10 per cent …) higher or lower than the y-target value(the particular percentages would be decided on politically). In this way a range of variation would be laid down for each member state for the values of social protection expenditure per capita whose mid-point is represented by the value on the regression lines. The distance of the two lines of deviation from the regression lines would increase as per capita income grows because the absolute amounts of deviation increase with higher incomes at the same percentage of deviation. The corridor would no longer be laid down for individual income groups; instead, there would be a single corridor around the regression line and no groups of states would be needed, either. Given the ongoing changes in per capita income and the large number of states whose incomes are spread out along the whole length of the regression line this is in any case the more elegant solution. In Figure 2, a corridor of plus/minus 10 per cent is shown by way of illustration. Based on this range, the following member states would exceed the corridor: Sweden (marginally), France, Portugal, Hungary, Poland, Bulgaria and Romania. Deviating below it would be the UK, Spain, Cyprus, Malta, the Czech Republic, Estonia, Lithuania (marginally), Latvia and, self-evidently, Ireland, for which based on per capita income a target value for social protection expenditure of plus/minus 10 per cent can be calculated. Within the framework of the current euro-crisis the situation in the UK, Ireland and Spain is particularly precarious given that their social expenditure veered downwards as early as 2007. Given the tough austerity measures demanded by the international financial markets and the EU, welfare state spending has come under further pressure. The three Baltic states are in a similar position: they have been hard hit by the global economic crisis and their 4

Klaus Busch| The Corridor Model – Relaunched Figure 1: Regression of social protection expenditure per capita in 1000 PPS in relation to GDP per capita in 1000 PPS for the EU26(excluding Luxembourg), 2007 Social protection expenditure per capita in 1000 PPS 12 y= 0.3276x – 2.2141 R ² = 0.85759 10 NL SE FR 8 DE DK AT BE FI IT UK IE 6 EL ES 4 2 BG RO PT SI CY CZ HU MT PL SK LT EE LV 0 8 13 18 23 28 33 38 43 GDP per capita in 1000 PPS Source: Author’s calculations based on Eurostat data. Figure 2: Example of a corridor(plus/minus 10 per cent) around the regression line social protection expenditure per capita in 1000 PPS/GDP per capita in 1000 PPS, EU25, 2007 12 Health care expenditure per capita in 1000 PPS y= 0.3672x – 2.9719 10 R²= 0.9238 8 6 4 2 BG RO PT HU CZ MT PL SK LT EE LV SE NL DK AT FR DE BE FI IT UK EL ES SI CY 0 8 13 18 23 28 33 38 GDP per capita in 1000 PPS Source: Author’s calculation of the regression and the corridor based on Eurostat data. 5

Klaus Busch| The Corridor Model – Relaunched public consolidation programmes have targeted their social security systems. 2. Criticisms of the Corridor Model The corridor model has gained some recognition in the political sphere over the past 10 years. The SPD has taken up the concept in its Hamburg programme; the trade unions ver.di and IG Metall advocate it in their European policy documents; and the Party of European Socialists (Hacker/Maass 2011) and UNI Europe, the European trade union federation of the private service sector, consider the idea eminently worthy of discussion. However, the model has encountered a number of criticisms, the two most important being as follows: on the one hand, the model has been criticised as exclusively quantitative, neglecting qualitative dimensions; on the other hand, there is resistance to the notion of a corridor because it is regarded as a limitation on the progressive development of the welfare state. In the next two sections I shall discuss these two principal objections to the model. 2.1 The Connection between Quantity and Quality in the Welfare State and its Significancefor the Corridor Model If, when examining the issue of quality, we restrict our focus to the two main functions of the welfare state, namely pensions(old-age pensions, survivors’ pensions) and health care spending(illness, invalidity) – which in 2007 made up almost 85 per cent of social protection expenditure in the EU15 and in the EU27 – the answer with regard to pension expenditure is easy to find. The pensionable age, reductions for early retirement and the income replacement rate in retirement – in other words, purely quantitative factors – determine the quality of this function of the welfare state. With regard to the health sector this simple relationship between quantity and quality is somewhat dubious. It could be that countries with above-average health care spending per capita waste money due to inefficiencies in the system and thus the quality of their provision is no better than that of other countries which commit less financial resources. Conversely, countries whose commitment of financial resources is below average, in comparison to their level of economic development, could provide a quality of provision in line with the average provision in economically comparable countries. Turning to this question we first encounter the fact that the connection between a country’s level of economic development and the level of its health care spending is very close. Figure 3 compares government health care spending and the per capita income of EU states for 2007. The coefficient of determination is extremely high, at 93 per cent, with a correlation coefficient of 96 per cent. With such a close connection between economic and social levels the question about the relationship between quantity and quality of provision answers itself. As prosperity grows states would not increase their health care spending in both absolute and relative terms unless that meant a qualitative improvement in medical provision. If higher quality were obtainable at a lower price at least some of the richer EU states would take this route. Even Ireland, which is far below the EU average with regard to social protection expenditure in general, is close to the regression lines with regard to health care spending per capita. This impression, which in any case follows from the overwhelmingly close connection between health care spending and level of economic development, is borne out by an examination of qualitative indicators related to the health care systems of EU states. Although the debate on qualitative indicators in health care is still in its infancy internationally, and a multitude of methodological problems pertain here, nevertheless, it is also here that we encounter the first evidence underlining the close relationship between quantity and quality(WHO 2000; BASYS 2007; OECD 2007; Greß/Wasem 2009). Reviewing a whole mass of qualitative indicators in the EU and the OECD, including, besides life expectancy, infant mortality, perinatal mortality, 2 vaccination rates for children and older people, survival rates for different kinds of cancer, chances of surviving heart attacks and strokes and the condition of the teeth of 12 year-old children, the following trends can be discerned: the states with the highest health care spending per capita exhibit a significantly better level of health among the population than states 2. Perinatal mortality means the total number of stillborn babies and early neonatal mortality(within the first seven days of life). 6

Klaus Busch| The Corridor Model – Relaunched Figure 3: Regression of public health care expenditure per capita in 1000 PPS in relation to per capita income in 1000 PPS for the EU 26(excluding Luxembourg), 2007 3.5 y= 0.1023x – 0.8028 3 R²= 0.9259 NL Social protection expenditure per capita in 1000 PPS 2.5 FR DE UK SE BE AT 2 DK IT FI SI ES 1.5 EL PT CZ 1 HU MT CY SK LT EE 0.5 RO PL LV BG 0 5 10 15 20 25 30 35 GDP per capita in 1000 PPS Source: Author’s calculations based on Eurostat data. IE 40 with lower spending. Virtually all indicators show this. The internal differentiation among the countries of Central and Eastern Europe(CEE) illustrates this particularly well. The CEE countries in this group with above average health care spending per capita – that is, Slovenia, the Czech Republic and Hungary – score significantly better in almost every instance on all the abovementioned indicators than the three Baltic states and Poland, whose spending is much lower(no details are available as yet for Romania and Bulgaria). 2.2 Corridor instead of Minimum Standards? The second objection to the corridor model holds that taking into account lower and upper limits on social spending is inconsistent with a progressive philosophy. This argument(Erdmenger/Gran/Kowalsky/Polzer 2009: 10) is based on a fundamental misunderstanding. First of all, it is contradictory to object against the corridor model that standards graded in accordance with economic output can be laid down perfectly well in the minimum standard model. The concept of differentiated standards is the very essence of the corridor model, and is not to be found in general minimum standards(Göbel 2002), so-called»social snakes«(Dispersyn/Van der Vorst 2002) nor in the open method of coordination(De la Porte/Pochet 2003; De la Porte/Nanz 2004; Busch/Hacker 2009). With the exception of the corridor model no approach has yet been put forward for the EU which lays down differentiated standards in accordance with level of economic development. To that extent, the objectors are already borrowing from the corridor model without realising it. To take upper limits to be a brake on progress is fundamentally to misunderstand the model. Needless to say, countries can deviate upwards as much as they wish, in accordance with a particular government’s own assessment. However, states which, for example, suffer from adverse demographic conditions cannot be obligated to put up with such competitive disadvantages. States disadvantaged or benefited in such ways should either receive compensation from the EU budget system or have to shoulder a bigger burden. To that extent, upper limits are not a brake on progress but a construct for avoiding excessive hardship for individual countries. In any case, these objections are at odds with empirical re7

Klaus Busch| The Corridor Model – Relaunched ality in Europe. The connection between social protection expenditure overall and level of economic development is – as we have repeatedly shown – so close that the criticism presented here has no empirical foundation. If, by contrast, one bases one’s objections on the high social expenditure ratios in the Scandinavian countries it has to be borne in mind that the latter’s lead levels off based on a net calculation. The Scandinavian countries fund social spending with contributions and taxation to an aboveaverage extent and based on a net calculation in fact fall below France and Germany(Adema 2001; Kemmerling 2001; OECD 2009). 3. Outlook The euro-crisis has by no means been surmounted. More and more countries are in need of rescue. After support loans were granted to Greece and Ireland, now Portugal has applied for EU assistance. It is to be expected that Spain will soon follow in their footsteps. Since this lending is coupled with tough austerity conditions, leading to drastic cuts in social security provisions, the future of the European Social Model is in doubt. The European Left urgently requires a consistent and comprehensive alternative approach with regard to growth policy, the architecture of the EU’s economic regime, European coordination of wage policy, the financing of public debt and the coordination of welfare states in the EU. This would enable it to oppose the newly invigorated neoliberal economic philosophy in the political arena. The model proposed here is a contribution to this debate which is so important for the future of Europe. 8

Klaus Busch| The Corridor Model – Relaunched Bibliography Adema, Willem (2001): Net Social Expenditure, 2nd edition, OECD Labour Market and Social Policy. Occasional Papers No. 52, Paris. BASYS (2007): Indikatoren der OMK im Gesundheitswesen und der Langzeitpflege. Augsburg. Busch, Klaus (1998): Das Korridormodell – ein Konzept zur Weiterentwicklung der EU-Sozialpolitik. In: Schmid, J./ Niketta, R.(eds), Wohlfahrtsstaat. Krise und Reform im Vergleich, 273–298. Marburg: Metropolis-Verlag. Busch, Klaus (2001): Economic Integration and Welfare State. The Corridor Model as a Strategy for an European Social Policy. In: Haller, M.(ed.), The Making of the European Union, 25–42. Berlin/Heidelberg/New York: Springer. Busch, Klaus/Hacker, Björn (2009): Die Methode der offenen Koordinierung: Entwicklungen, Perspektiven und Grenzen. In: Devetzi, Stamatia/ Platzer, Hans-Wolfgang(eds), Offene Methode der Koordinierung und das Europäische Sozialmodell, 27–49. Hannover: Ibidem Verlag. Busch, Klaus/Hirschel, Dierk (2011): Europa am Scheideweg. Wege aus der Krise. Internationale Politikanalyse, Friedrich-Ebert-Stiftung, March. De la Porte, Caroline/ Pochet, Philippe (2003): Building Social Europe through the Open Method of Coordination, Brussels: Peter Lang. De la Porte, Caroline/Nanz, Patrizia (2004): The OMC – A Deliberative-Democratic Mode of Governance? The Cases of Employment and Pensions. In: Journal of European Public Policy 11, 2: 267–288. Dispersyn, Michel/Van der Vorst, Pierre et al. (1990): La construction d’un serpent social européen. In: Revue Belge de Sécurité Sociale 12: 889–980. Dispersyn, Michel/Van der Vorst, Pierre et al. (1992): La construction d’un serpent social européen – Étude de �fa��is�a��bilité. In: Revue Belge de Sécurité Sociale 36: 315–656. Döring Diether/Dudenhöfer, Bettina/Herdt, Jürgen (2005): Europäische Gesundheitssysteme unter Globalisierungsdruck. Vergleichende Betrachtung der Finanzierungsstrukturen und Reformoptionen in den EU 15-Staaten und der Schweiz. Wiesbaden. Erdmenger, K./Gran, S./Kowalsky, W. /Polzer, U. (2009): Die soziale Dimension der EU. Internationale Politikanalyse, Friedrich Ebert Stiftung, July. Eurostat (2011): http://epp.eurostat.ec.europa.eu/portal/page/portal/social_protection/introduction Göbel, Marcus (2002): Von der Konvergenzstrategie zur Methode der offenen Koordinierung. Baden-Baden: Nomos. Greß, Stefan/Wasem, Jürgen (2009): Indikatoren zur Leistungsfähigkeit von Gesundheitssystemen und die Methode der offenen Koordinierung. Unpublished manuscript. Kassel. Hacker, Björn/Maaß, Gero (2011): Ein Grundsatzprogramm für die SPE: Baustellen, Gemeinsamkeiten und Eckpunkte aus deutscher Perspektive. Internationale Politikanalyse, Friedrich-Ebert-Stiftung, December. Heise, Arne/Lierse, Hanna (2011): Haushaltskonsolidierung und das Europäische Sozialmodell. Auswirkungen der europäischen Sparprogramme auf die Sozialsysteme. Internationale Politikanalyse, Friedrich-Ebert-Stiftung, March 2011. Kemmerling, Achim (2001): Die Messung sozialstaatlicher Leistungen. Zu den Folgen der Nettosozialleistungsquote Willem Ademas für die Diskussion um staatliche Sozialpolitik. Statistisches Bundesamt, Wiesbaden. OECD (2007): Health at a Glance 2007. Paris. OECD (2009): Society at a Glance 2009. Paris. Tiemann, Susanne (2006): Gesundheitssysteme in Europa – Experimentierfeld zwischen Staat und Markt. Köln. Wendt, Claus (2003): Krankenversicherung oder Gesundheitsversorgung? Gesundheitssysteme im Vergleich. Wiesbaden: VS Verlag für Sozialwissenschaften. WHO (2000): World Health Report 2000 – Health Systems: Improving Performance. Geneva. 9

About the author Dr Klaus Busch is professor(emeritus) of European Studies at the University of Osnabrück and adviser on European policy to the trade union ver.di. Imprint Friedrich-Ebert-Stiftung International Policy Analysis Hiroshimastraße 28| 10785 Berlin| Germany Responsible: Dr. Gero Maaß, Head, International Policy Analysis Tel.:++49-30-269-35-7745| Fax:++49-30-269-35-9248 www.fes.de/ipa To order publications: info.ipa@fes.de The views expressed in this publication are not necessarily those of the Friedrich-Ebert-Stiftung or of the organization for which the author works. This publication is printed on paper from sustainable forestry. ISBN 978-3-86872-795-1