ABSTRACT
The articles in this special issue identify the most important institutional shortcomings and design flaws of EMU and provide plenty of ideas on how they could be amended. Giavazzi and Wyplosz, for instance, set off with a systematic and honest assessment of whether a similar analysis in 1998 stood the test of time; De Grauwe and Ji and Begg et al. discuss more
recent lessons from the crisis. While there is no complete agreement, nobody proposes that EMU governance should leave more room for market discipline. How times change. Fiscal surveillance in the euro area was premised on the idea that if it
would signal the right information to market observers, their reactions would support the building of a hard currency union. Already the Maastricht convergence criteria invited market forces to sit on the judging panel to decide who was ‘fit’ for EMU: Convergence of interest and inflation rates could not be simply manipulated by authorities. Yet, far from disciplining budgetary policies, capital flows financed deficits at historically low interest rates. It is safe to assume that the EU’s fiscal rules could have been more easily enforced if financial market actors had supported the Commission’s fiscal surveillance and, for instance, credit rating agencies (CRAs) had downgraded the bonds of offending member states reliably and investors imposed rising risk premia accordingly. Some economists warned of the lack of market discipline (Faini 2006;
Jones). They could point to the convergence of government bond spreads of member states despite different or deteriorating fiscal positions. Yet, those who believed in market discipline could point out that this might be due to diminishing exchange rate risks, leaving only liquidity and default risks to price (Manganelli and Wolswijk 2007: 10-11, 24-25). The empirical evidence before the 2007-2008 crisis suggested that bond spreads reflected different ratings but the methodology used could obviously not quantify whether the pricing was correct. Moreover, expansionary fiscal policy by a member state seemed to spillover into the average level of interest rates, suggesting that fiscal rules were needed additionally to prevent this externality (Faini 2006: 464, 469). The regulation of credit ratings was briefly discussed in Europe after the Asian crisis and the Enron scandal but was resisted by international business and banks which depended for their investments and also their own ratings on a sovereign benchmark (Abdelal 2007: 175). The lack of support for EU surveillance from CRAs should come as a
surprise even to those who do not believe that efficiency is an inherent feature of financial markets. After all, sovereign credit rating is a multi-billion dollar business, dominated by three agencies (Fitch, Moody’s, and Standard & Poor’s) that constantly provide information to markets about the credit risk of government bond issues. By mid-2010, Standard & Poor’s (S&P) rated 125 governments, the other two over a 100 each (IMF 2010: 87). This makes reputation for accurate ratings extremely valuable. Moreover, there should also have been incentives for the monitoring of these monitors. Credit ratings were built into the regulation of financial institutions, in that the risk weights of the assets could be based on such credit ratings and thus determine the capital requirements under Basel II (IMF 2010: 91-92). Since highly rated bonds of OECD countries carried a zero risk weight and were therefore very attractive to meet capital requirements, it should at the same time have given independent financial regulators considerable incentive to make sure that the ratings of government bonds signaled problems of fiscal sustainability accurately. Last but not least, the
834 Zsofia Barta and Waltraud Schelkle
European Central Bank (ECB), like other central banks, used these credit ratings in refinancing operations: In return for central bank credit, banks had to put up collateral that met a certain credit rating standard in order to be acceptable to the ECB. In sum, both financial and institutional incentives for reputation of commercial and technocratic monitors of public finances should have been aligned. These considerations can be supported by at least two conceptualizations
of the European integration project. There is, first, the interpretation of the EU as a regulatory polity that suggests that it constitutes a ‘fourth branch of government’ for supranational economic regulation (Majone 1993, 1996). This theory seeks to explain why sovereign governments delegated considerable regulatory powers to independent bodies, such as the Commission or the ECB. In this view, delegation of policy-making powers guards against well-known failures of democracy to either represent the majority or to consider the legitimate interests of outsiders of national democratic processes, be it foreigners or future generations. In the mediumto-long run, these failures impoverish countries and the interdependence of economies makes national democratic failure a problem for other countries. Hence, governments agree on supranational institutions as restraint on destabilizing macroeconomic policies. The analogy of the regulatory polity and the institutionalization of CRAs as regulators is striking: Governments, not only in Europe but internationally, gave self-regulatory authority to financial markets they had liberalized and the markets ceased some of their power to the agencies (Abdelal 2007: 165). Far from privatizing regulatory authority, CRAs seemed to fit into an innovative approach of non-politicized regulation of global markets in which sovereigns could not claim privileges. At the opposite end of this functionalist view of EU policy-making is a
constructivist interpretation that sees European economic integration as putting into practice neo-or ordoliberal ideas. This followed the perceived German success story at a time when Keynesian interventions seemed to fail (McNamara 1998). A singularly independent central bank and fiscal rules were introduced to tie governments’ hands. The recent failure of financial markets on a colossal scale did not end the neoliberal reign, on the contrary, an ‘austerity delusion’ has taken hold of EU policy-makers that is manifestly counterproductive for economic recovery (Blyth 2013: ch.3). In this view, the fiscal surveillance process can be seen as directly feeding financial investors with data, inviting them to sanction government behavior that would prioritize domestic redistribution and stimulus over the repayment of bonds. The role of credit ratings in the Basel II framework of financial regulation is further evidence for a close alignment of business interests and neoliberal ideas of government. CRAs invented ‘a common language of credit risk’ and governments proceeded to make ‘the bond market’s private authority public’ (Abdelal 2007: 174). The empirical corollary of these theories is that the international-
commercial assessments of sovereign creditworthiness and the supranational-technocratic surveillance of budgets should be fairly aligned. Prospective bond buyers had in the Commission a potential ally and the
At Cross-purposes 835
regulatory/neoliberal polity in fiscal policy should have directly fed into the assessment of sovereign creditworthiness by rating agencies. The commercial agencies’ interest in the probability of sovereign default can be seen as a specification of the EU’s interest in externalities of public debt accumulation, such as higher interest rates and inflationary pressures. For our study of rating agency behavior, we concentrate on Standard
and Poor’s, which is the most active and influential agency in sovereign credit rating (De Haan and Amtenbrink 2011: 5, 9). We rely both on publicly available data by analyzing the content of Rating Methodologies and individual rating reports for individual countries issued by S&P and on interview evidence.1 For the study of technocratic fiscal surveillance, we analyze the ways in which Eurostat’s statistical accounting translates data on pension savings and banking rescues into information about fiscal sustainability. So we turn the theoretical puzzle into the empirically researchable question: Why do market actors and a supranational economic bureaucracy have different assessments of these government interventions? We have chosen pension reforms and bank rescue programs because they
have strong fiscal consequences in contrast to, for instance, cuts in unemployment benefits.2 They also have sufficiently rich and complex budgetary effects so that their assessment is a non-trivial task. Their long-term effects can be different from the short-term. Pensions have been a construction site of reforms over several decades now, while systemic bank rescues in developed countries took government watchers by surprise. This allows us to see how commercial and technocratic monitoring of sovereign debt interpret complex data and construct information about fiscal sustainability or sovereign creditworthiness, respectively. The next three sections analyze the rating and statistical accounting of
the two policy measures as well as recent changes to the approach of CRAs and to fiscal surveillance in the wake of the economic and financial crisis. In the section before the conclusions, we discuss our findings and we try to answer the overall question of why EU fiscal governance by regulation only is so unsatisfactory, both from the point of view of those who are concerned about high debt and those who are concerned about constraints on fiscal stimulus. Our conclusions argue that markets and distributional conflicts can overstretch governments’ capacity to control their budgets.
