Towards a Sustainable Growth Model for Europe: Institutional Framework * SEBASTIAN DULLIEN The Problem: Dangerous Debt Trends The current financial and economic crisis has highlighted once again that a sustainable growth path cannot be achieved with permanent growth of indebtedness in single sectors or countries. A debt-driven expansion, reflected in ever-rising debt levels, has to come to an end at some point and usually harbors the danger of a debt crisis. Such crises are usually associated with major disruptions of normal business activities; sharp cutbacks in expenditure in individual countries – necessary to reverse unsustainable debt trends – risk triggering a recession in their trading partners as well. As the imf confirms in its Autumn 2010 World Economic Outlook, a financial crisis also often leads to permanent losses in output, thus making it even more undesirable than normal recessions. Based on the experience of the past 24 months, within the European context, three levels of debt at the national level seem in need of control: net government debt, gross debt in the financial sector, and net liabilities of a country against the rest of the world. Besides the argument that excessive debt levels of single countries or sectors will at some time stifle growth and thus are not desirable, there is an additional rationale for controlling the government debt of emu partner countries, namely direct external costs to the rest of emu . The danger of a default by the government of one emu member state must be considered an issue of common concern in the monetary union. In fact, this argument has become even stronger after the crisis: as the events of the crisis have shown, the financial institutions of the Eurozone are very much interlinked. Hence, problems in one country’s banking sector will immediately have dire consequences for the rest of the monetary union. * The paper is based on an impulse statement at the international seminar»Cohesive Growth in Europe after the Crisis«, organized by the Friedrich-Ebert-Stiftung, Berlin, October 23, 2009. 36 Dullien, A Growth Model for Europe ipg 1/2010

At the same time, all of the emu countries(except, perhaps, Luxembourg, Malta, Cyprus, and Slovenia) are big enough for a default on their government debt to have the potential to cause major problems in the banking system. Thus, out of economic self-interest, other emu countries can be expected to bail out any government which gets into trouble. This point was made by then – German finance minister Peer Steinbrück in February 2009 when he told the media that smaller emu countries would be bailed out during the current crisis. The formal nobail-out clause in Art. 103 of the eu treaty has thus morphed into an informal bail-out guarantee. Thus, it is now in the direct interest of all emu countries to prevent a default by any of the other member states. The necessity of limiting the gross debt in a country’s financial sector, as well as the net debt of emu countries can also be easily deduced from this conclusion: no emu or eu member state can afford the bankruptcy of its banking system, which is indeed the backbone of the economy – the negative impacts on growth, employment, and future tax revenues would simply be too severe. It is for this reason that, in a serious financial crisis, governments are very likely to assume the liabilities of the national financial sector – as recently took place in Great Britain and Ireland, as well as in past financial crises in Latin America and Asia. The same is probably true when significant parts of the business sector are threatened with bankruptcy. In case of a crisis, the government would probably assume the obligations rather than risk the collapse of large parts of the private sector. In times of high unemployment and especially in the run-up to elections, governments find themselves under intense pressure to take action. In the extreme case, the sum of private debt could comprise many times the sum of previous government debt – and, as can be seen in Ireland, it is quite possible for a country with sound public finances to become a basket case practically overnight. When, at the same time, the logic of economic linkages within emu makes it likely that other countries will have to take over the liabilities of another country in a dire situation, the rest of emu might be forced, de facto, to pick up indirectly the liabilities of the private sector of any single emu country. ipg 1/2010 Dullien, A Growth Model for Europe 37

The First Pillar of Debt Control: EU-level Financial Market Regulation Before trying to find an institutional solution to the problem of unsustainable debt trends in emu (or the eu ), one must distinguish between different types of dangerous debt accumulation. First, there is the danger that the financial sector in one country, expanding excessively into international business, will accumulate large gross debts in the process. This debt at first does not appear to be very burdensome as the financial institutions also accumulate assets of similar value and the net debt level therefore remains low. However, this large gross debt is dangerous, as it is a sign of increasing leverage and therefore of the increasing loss potential of the financial sector in question. In times of crisis, this loss potential can easily turn into outright losses which might then, in a second step, burden the government budget, as has been the case in Ireland. This specific danger(of, as some might say, a country turning into a hedge fund) is usually not strongly related to the macroeconomic development of the country in question, but is rather based on its regulatory approach. For example, the large increase in the Irish financial sector’s balance sheet over the past few years was probably due, to a large extent, to the lax regulation of financial institutions based in Ireland. Consequently, the right approach here would be to tighten financial regulation. As any unsustainable debt trend in a national financial system might result in costs to the partner countries in the eu and emu , the only logical solution would be to establish both the regulation of financial markets and financial institutions and the enforcement of these rules at the European level. The current proposals of the eu Commission for a new pan-European supervisory framework are a step in the right direction but even more centralization is probably needed. Completely centralized financial market regulation and financial market supervision(perhaps with national branches, as in the System of European Central Banks) would thus be the first pillar of effective debt control. 38 Dullien, A Growth Model for Europe ipg 1/2010

The Second Pillar of Debt Control: Monitoring Current Account Balances 1 The second pressing debt issue in Europe, that of unsustainable debt accumulation by individual countries, is much harder to tackle. While the neoclassical textbook model of current account determination usually argues that current account deficits or surpluses stem from the individual inter-temporal consumption and investment choices of households and firms and should thus not be seen as a problem, recent experience casts doubt on this view. According to the textbook account, capital will flow from developed, capital-abundant economies to less developed, labor-abundant economies. This helps the less developed countries to build capital stock and expand their production frontier. Part of the additional production can then later be used for servicing the debt incurred in the process. There are two problems with this approach: first, as recent research in behavioral finance shows, most individuals do not have a good grasp of the long-run consequences of their saving decisions. These decisions are thus very susceptible to the influence of psychological factors rather than microeconomic utility maximization. 2 As a consequence, it is very possible that individuals will exhibit saving and borrowing trends which will lead to an unsustainable debt trend for the macroeconomy. The problem is that, from the headline current account figure alone, one cannot distinguish whether the capital imports are being used for additional investment in a country, which helps to expand productive capacity and hence the ability to service the incurred external debt, or whether they are just being used(directly or indirectly in the form of investment in housing) for domestic consumption. In the context of the European Monetary Union, one must also take into account the possibility that, due to a loss of competitiveness, a country loses the ability to generate sufficient income to maintain its current level of consumption. If individuals do not understand this problem, however, and mistake the permanent loss of income for a temporary one and continue to borrow in order to keep up their individual level of consumption, this will lead to a large current account deficit without an increasing capital stock to service this debt at 1. This section builds on S. Dullien/ D. Schwarzer(2009): The Euro Zone Needs an External Stability Pact, swp Comments 2009/C 09(July 2009). 2. For an accessible overview, see Chapter 10 of G. A. Akerlof and R. J. Shiller(2009): Animal Spirits. Princeton: Princeton University Press. ipg 1/2010 Dullien, A Growth Model for Europe 39

a later date. Macroeconomically, we may witness a phenomenon which is well known from the microeconomic level: an economy then basically engages in distress borrowing, borrowing merely to prevent a necessary, but painful, adjustment(or, on a microeconomic level, becoming insolvent) without the prospect of repaying the debt in the future. Second, the simple neoclassical model of international capital flows does not distinguish between investment in tradables and non-tradables. Empirically, a large part of capital inflows in countries with large current account deficits ends up in investment in the non-tradable sector or in construction investment for the housing sector. While these investments might increase the capital stock and hence, in principle, shift the consumption possibilities frontier outward, they do not contribute at all to the individual country’s ability to service its foreign debt later on. After all, the foreign investors in, for example, Latvia expected to be repaid in euros, not in apartments in Riga or mobile phone calls in Latvia. Closer examination of national and external account data of recent years reveals that most of the net borrowing of emu countries with large current account deficits is actually due to different trends in consumption growth and housing investment, as well as diverging levels of competitiveness, rather than just investors from more developed emu countries chasing higher returns in less capital-abundant countries. For example, it is sensible to argue that the construction boom in Spain was fueled by the favorable tax treatment of mortgages in Spain, while in Germany low wage increases and the government’s decision to subsidize retirement savings has contributed to stagnating consumption. Thus, in order to tackle the unsustainable debt trends, some kind of policy measure seems to be desirable. A first best solution for controlling excessive debt trends would thus be to control the underlying reasons for them. This would mean tackling the divergence of competitiveness and coordinating policies that influence the national savings and investment rates. With regard to competitiveness, this would imply a coordination of wage bargaining in Europe, so that on trend, wages in each country increase with the rate of productivity growth plus the target rate of inflation. For fiscal and economic policies, it would mean close cooperation between countries, which will probably be workable only if one centralizes large parts of economic and fiscal policy-making in Brussels. However, such centralization seems politically unrealistic at the moment. Another option would be to force coordination by introducing an»External Economic Stability Pact.« It 40 Dullien, A Growth Model for Europe ipg 1/2010

would limit current account imbalances in emu outright by stipulating that no Eurozone country may have a current account imbalance of more than three percent of gdp , either as a deficit or as a surplus. This three percent could be adjusted to take into account inflows or outflows of foreign direct investment in new factories(greenfield investments), which would allow countries to import more during an investment boom, aimed at improving the ability to produce tradable goods. The figure of three percent is chosen not merely in analogy of the Stability Pact, but is derived from the mathematics of debt calculation: if a country shows an annual balance-of-payments deficit of three percent of gdp , then, at an expected average nominal growth rate of five percent a year, the level of foreign debt will stabilize at below 60 percent of gdp . This figure should be seen as acceptable based on experiences with previous financial crises, in which higher levels of debt have frequently resulted in balance-of-payments crises. The adjustment to include foreign direct investment is permissible because it does not affect a country’s debt levels, but can even help to improve the balance-of-payments position if investments have been made in the production of tradable goods. As with the existing Stability Pact, exceeding the threshold for balance-of-payments imbalances would initially result in warnings to the member state, followed by appropriate measures to bring the imbalance down. If this adjustment does not occur, sanctions could be imposed – given a correspondingly adapted legal framework – for example, halting financial transfers from the eu budget, stopping the provision of interestfree loans, or even imposing fines on the member states. The External Stability Pact should oblige every country to orient its own national economic policies(both fiscal and more regulatory) towards achieving an external economic balance. In comparison to the existing form of economic policy coordination, or to the centralized coordination of European fiscal and economic policy at the emu level, the External Stability Pact would offer three advantages. First, it would lead to a broader coordination of economic policy than is currently the case. This would have an impact especially on the wage-setting systems of individual countries – governments would be compelled to use national legislation and wage settlements in the public sector to influence wage policy in such a way that imbalances among the Eurozone countries can be avoided. This would give countries like Spain – which still partly index wage increases to the inflation rate, thereby exacerbating boom and bust cycles – another argument to use in the domestic policy ipg 1/2010 Dullien, A Growth Model for Europe 41

debate with a view to prohibiting such contract clauses. This would not be a direct infringement on the negotiating autonomy of the bargaining parties, but would force governments to create an emu -compatible framework. Second, the Pact would be an instrument that would oblige individual eu countries to take the consequences for other member states into account when designing their own national economic reforms. Each country would have to ensure that any effects on neighbors(for instance, in the form of excessively high surpluses on their own side) would not be too severe. If a surplus country wanted to lower non-wage labor costs and increase value-added tax in order to boost domestic economic competitiveness, under the new rules it would simultaneously have to compensate for the negative effects on its partners’ foreign trade through an expansive fiscal policy. Third, within the framework of the new regulations, individual countries would retain the authority to design their own economic and fiscal policies. Each one would have the freedom to decide what means they might want to use to avoid massive foreign trade imbalances. The Spanish government, for example, could have met the building boom and the foreign trade deficit with tax increases – or by urging the social partners to exercise wage restraint. As an alternative, it could have intervened by instituting planning regulations or a legal lending limit for mortgage loans. The Third Pillar of Debt Control: A Revised Fiscal Stability Pact Finally, there needs to be a mechanism to control the level of government debt in emu . There has been considerable debate on this issue in recent years, so this section will be confined to some basic observations. The Stability and Growth Pact( sgp ) in its current form still seems to have a number of problems, both in its basic construction and in its enforcement. The focus of the Pact seems to be too much on the numerical value of three percent and less on the development of debt over the medium term. In addition, the definitions of possible exceptions to this rule in case of an economic downturn still seem to be too strict. While the sgp is not being applied in the current crisis, since in most countries the recession is deep enough to trigger the automatic suspension of the deficit 42 Dullien, A Growth Model for Europe ipg 1/2010

rules, one must keep in mind that this crisis is probably a once-in-a-century event and that less severe recessions might also call for fiscal actions which might conflict with the sgp . Thus, one should think along the lines proposed in earlier reform debates of actually replacing the threepercent threshold with a medium-term target for the debt-togdp ratio, possibly with an independent board of experts at the emu level which would decide on exceptions in cases of severe recessions. No Early EMU Membership In the context of the current crisis, it has sometimes been argued that countries such as Estonia or Latvia should be allowed to enter emu early and without fulfilling the convergence criteria. By doing so, debts and financial assets in national currencies would be transformed into euros, while the banks would be given access to the ecb ’s lender-of-last-resort facilities. This would end speculation of a devaluation in these countries, thus taking some pressure off them. However, this argument does not hold up to closer scrutiny of possible side effects. The large current account deficits of the countries in question(in some cases more than 20 percent of gdp ), as well as strong real appreciation in recent years, hint that these economies have more problems than merely a lack of credibility of their monetary and exchange rate regime. The danger is that they would continue to have a very hard time adjusting, even if they were allowed to enter emu at once at the current exchange rate. The current overvaluation would require a severe domestic deflation if they joined emu . This would entail nominal wage cuts which, in turn, might cause defaults by mortgage owners, as well as insolvencies in the nontradable sector. The sad case of Portugal(where the convergence process of per-capita income was reversed in 2002) is a reminder of the dangers of being in emu at an overvalued exchange rate. While the country continues to run a current account deficit of about 10 percent of gdp , it has not been able to correct its real overvaluation, even though wage growth has slowed considerably. The convergence process has thus stalled at less than 60 percent of the average emu per-capita-income. For countries outside emu , a strong nominal depreciation seems to be a better path to adjustment than entering emu and hoping for an adjustment by domestic deflation. One could even argue that the old convergence criteria completely neglect the dangers of unsustainable debt ipg 1/2010 Dullien, A Growth Model for Europe 43

trends. One should thus add a new criterion which only allows countries to enter if their current account does not show an imbalance of more than three percent of gdp . 44 Dullien, A Growth Model for Europe ipg 1/2010