![Sample our Politics & International Relations journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5947/TOC-Subject-Sample-2021-Politics-International-Relations.jpg"})
Abstract
This article challenges the conventional wisdom of weak market discipline in Economic and Monetary Union (EMU). In so doing, we empirically analyse the dynamics of market discipline for all 27 EU member states between 1992 and 2007. The existing literature tends to assert that markets discipline governments, without measuring whether the interest punishment markets impose actually has the purported effect on government policy. To better grasp the dynamics of market discipline it is essential to consider both sides. Market discipline is thus understood as a two-sided phenomenon. On the one hand, financial investors react to policy developments. On the other hand, policy-makers react to market signals. We find strong evidence that although the impact of fiscal policy developments on market punishment slightly decreases with monetary integration, government responsiveness to market punishment increases. This runs counter to the conventional narrative of policy-makers banking on bailout from fellow EMU members.
Acknowledgements
Special thanks are due to Iain Hardie. I wish to thank the German National Merit Foundation (Studienstiftung des deutschen Volkes) for generous financial support. The research leading to this article has received funding from the European Research Council under the European Union’s Seventh Framework Programme (FP7/2007-2013)/ERC grant agreement no. 291733. Earlier versions of this article were presented at the workshop ‘The European Sovereign Debt Crisis’ at the University of Luxembourg and at the International Political Economy Research Group at the University of Edinburgh. Comments and helpful suggestions by participants are gratefully acknowledged.
Notes on contributor
Charlotte Rommerskirchen is Lecturer in International Political Economy at the University of Edinburgh. Her research interests lie in the politics of financial and economic crisis, monetary integration and financial globalisation. She has recently published in West European Politics, Contemporary European Politics and Party Politics.
Notes
1. Hardie (Citation201Citation1: 143) defines financialisation as the increased ability to trade risk. The broader trend of financialisation can be attributed to the financialisation of the financial actors in the market as well as the financialisation of the structure of the government bond market itself (Hardie Citation2011).
2. Market punishment is hardly the result of a normative or pedagogical agenda of market participants, but instead, first and foremost, the result of any portfolio model with standard preferences for risk and return. Put crudely, market discipline is therefore primarily concerned with supply and demand, not crime and punishment.
3. Moral hazard refers to a situation where the provision of insurance, by diminishing the incentives to prevent a particular outcome, may make this outcome more, not less, likely.
4. The SGP sets budgetary rules which apply to all EU Member States. Specifically, it stipulates a deficit to GDP threshold of 3 per cent and a debt level threshold of 60 per cent of GDP. The challenges of fiscal restraint are not E(M)U specific. On the international level, for instance various studies have highlighted the weakness of International Monetary Fund conditionality, as a form of political discipline to bring about ‘sound' fiscal policies, with ‘the clear majority of Fund programmes [ … ] uncompleted' (Bird Citation2002: 847).
5. For a distinction between market discipline and capital discipline, see Milios and Sotiropoulos (Citation2010).
6. Robert Mundell (Citation1961, see also Snaith Citation2013) argues that three factors determine whether or not two (or more) countries should share the same currency; the degree of economic integration, the degree of asymmetry between countries and the existence of correction mechanisms to correct divergences.
7. A ‘credit default swap spread’ for a particular bond issuer refers to the quoted market price to enter into a credit default swap written on that issuer's bonds. A credit default swap is a financial instrument negotiated between a buyer and a seller in which the seller agrees to pay to the buyer the recovery value of the issuer's bonds, contingent upon its default. It thus functions as a form of insurance against the risk of default.
8. Debtor moral hazard refers to a situation where debtor countries have weak incentives to implement politically costly reforms or pursue ‘conservative’ economic policies expecting to be bailed out in the event of crisis and thus raising the probability of such crisis occurring.
9. Investor moral hazard refers to excessive risk taking which is linked to the perceived insurance of future bail-outs.
10. Corsetti and Dedola (Citation2012) argue that the moment investors anticipate inflationary financing, interest rates would rise, reducing the gains from debt monetisation. Instead, the main constraint of monetary union stems from the national central bank's inability to swap debt for interest-bearing reserves, which according to their estimates does not necessarily lead to a burst of inflation. The core of the ‘original sin’ argument of monetary union however still holds.
11. Gelpern (Citation2012) uses the term ‘quasi-sovereign debt' to describe sub-state borrowing in the USA. She explicitly relates one of the main arguments, namely that the absence of monetary policy results in ‘a hard budget constraint, as in a country that can only borrow in a foreign currency' (Gelpern Citation2012: 917), to EMU.
12. History obviously offers ample evidence of sovereign default where governments were in control of the printing press (see Tomz Citation2007).
13. Real long-term bond yields from the IMF Data Mapper. Correlation results for 10-year government bond yields point to a similar relationship ((r(139) = 0.39, p < .01)). DataStream 10-year government bond yields for Eurozone countries: Austria and Belgium (2003–2013); France, Finland, Germany, Greece, Ireland, Italy, the Netherlands, Portugal and Spain (2002–2013); Malta (2012 and 2013) and Slovenia (2007–2013).
14. The residual maturity is the time from the reference date until the contractual redemption date of an instrument.
15. Across the Organisation for Economic Co-operation and Development (OECD), governments have followed a more market-oriented approach to debt management. This led governments to privatise the administration of sovereign debt and to use new financial instruments. For an account of the institutional transformation of the German public debt agency, see Trampusch (Citation2015).
16. An additional option would be to take the log of Interest to account for non-linear effects. This is not possible for the variable Budget, as it takes negative values. Given our joint estimation of both effects, we opted for the identical variable transformation and thus included the squared term. Testing the misspecification of the functional form, regression equation specification tests reject the addition of the cubic term for both dependent variables. Using the log form transformation of Interest does lead to similar results which can be obtained upon request.
17. Luxembourg does not publish its short-term interest rates. For Luxembourg the Belgian short-term rates are therefore taken, which is an obvious choice given the long-standing monetary policy cooperation between both countries. To test whether this approach biases results, we excluded the variable Short; results are similar and can be obtained upon request.
18. Bulgaria's debt servicing costs stood at above 21 per cent of GDP in 1996 (see Dobrinsky Citation2000 on the Bulgarian ‘transition crisis’ of the 1990s). Note that this outlier figure, both within the sample and for the country (interest expenditure fell to 7.29 per cent in the following year), is not included in any model due to the lack of available data for several of the independent variables.
19. 3SLS has the advantage of modelling the contemporaneous feedback effects between budget outcomes and market punishment. 3SLS is based on Two-Stage Least-Squares where an instrumental variable technique for the endogenous variable provides consistent estimators. The instrumental variable technique regresses all exogenous variables on the endogenous variable. It creates a ‘proxy variable’ that resembles the endogenous variable but is not correlated with the disturbance term, therefore producing unbiased and consistent coefficients.
20. Eichengreen and Mody (Citation2000) distinguish between ‘push’ (external) and ‘pull’ (internal) factors and argue that the former are as important, if not more, for developing countries' financing conditions as the latter.
21. There is evidence that the responsiveness of both market participants and policy-makers may be delayed (Balassone et al. Citation2004, Ardagna Citation2009). We therefore check the sensitivity of our estimations by taking the lag of the two dependent variables as well as the macro-economic controls. The results are similar to the findings presented in and can be obtained upon request
22. Results for the EU27 (1999–2007) and the Non-Eurozone EU (1992–2007) are very similar and can be obtained upon request.
23. The IMF's (Citation2009: Table 8.) Fiscal Monitor lists the total financing needs of selected advanced economies. Japan stands out with a 59.4 per cent of GDP financing needs, followed by Italy with 30.1 per cent Germany and Sweden (8.5 per cent and 4.7 per cent) conversely have a comparatively low financing need.
24. Correlation results suggest as much. For the 1992–2008 period, the correlation between 10-year government bond yields (IMF) and government debt (AMECO) is strong and statistically significant only for non-Eurozone Member States ((r(185) = 0.43, p < .01.) vs. (r(98) = 0.11, p < .26)).
25. Bail-out expectations exist also within the EU27. Indeed three of the eight EU Member States who received financial assistance since 2008 are not Eurozone members. See Gray (Citation2013: Chapter 4) for empirical evidence of financial markets’ bail-out expectations for the EU27. A broader perspective on investor moral hazard beyond the EU can be found in inter alia, Vaubel (Citation1983) and Breen (Citation2013).
26. The squared term of Punishment is neither individually nor jointly with the linear term statistically significant. Overall the squared terms add to the fit of the model and their inclusion is theoretically justified. Even where statistically not significant the squared form is kept, which furthermore makes the models easier to compare. The exclusion of non-linear terms does not change the results qualitatively.
27. Currency composition is a crucial factor of a country's likelihood to default. Advanced economies’ debt is predominantly denominated in domestic currency. Conversely, in 2005 the share of sovereign debt denominated in or indexed to foreign currency stands at 42 per cent in the emerging economies. In the emerging economies that defaulted over the past two decades, foreign currency averaged at 63 per cent the year prior to default, with Argentina's foreign currency debt standing as high as 99 per cent of total debt (Cottarelli et al. Citation2010: 15).
28. The warning that Member States in a currency union without lender of last resort might face liquidity issues can already be found in the Delors Report (Citation1989: 20, see also Dyson and Featherstone Citation1999: 669–73).
29. Despite the fact that the so-called Maastricht criteria, which lay out the conditions for EMU membership, ‘were massively relaxed when the current Member States faced their entrance exams in 1998’ (De Grauwe Citation2009).
30. A related explanation for muted government responsiveness can be found in Wyplosz's post-Maastricht fatigue (2006).
31. As an additional test we created an interaction term, multiplying both dependent variables and future Eurozone Member States status. The interaction term was not significant in any of the time periods and country groups analysed. The lack of a seal-of-approval effect is potentially due to the coverage of our data. With only EU and future EU Member States, theoretically the entire sample could be covered under such a seal of approval. To tease out the different membership discounts both a larger time-period starting already in the 1980s and the inclusion of non-EU countries as control group would be required.
32. As of 2012 these are (year of entry): Bulgaria (2007), Czech Republic (2004), Estonia (2004), Hungary (2004), Latvia (2004), Lithuania (2004), Romania (2007), Slovakia (2004) and Slovenia (2004).
33. This is further confirmed by the inclusion of interaction terms with a post-communist country dummy and both dependent variables which remains insignificant throughout the different specifications.
34. Not doing so would create bias as interest payments are already taken as a dependent variable. Furthermore, interest rate payments, while to a large extent dependent on a country's economic outlook and debt sustainability, can be subject to fluctuations that are out of the hands of national policy-makers. According to Fedelino et al. (Citation2009: 1) ‘interest payments are often kept separate because their movements, while “automatic” in the sense of not generally reflecting discretionary fiscal policy actions, may not be necessarily correlated with cyclical output changes’.
35. So doing requires two inputs: first, the cyclical position of the economy as measured by the output gap (the distance between actual and potential output); second, the responsiveness of the budget balance to the economic cycle as expressed by budget elasticities. See Larch and Salto (Citation2005) for a detailed description.
36. Wren-Lewis (Citation2013) goes on step further and argues in favour of market discipline formally replacing the SGP: fiscal governance, so his argument goes, can and should reside at the national level, where it can focus on both national stabilisation and the control of public debt.
37. The external constraints imposed by globalisation or European integration (Hay and Rosamond Citation2002) are frequently evoked as cause and justification for painful and/or unpopular social and economic reforms. For example, in the context of monetary union, Talani (Citation2003) argues that Italy's policy-makers deliberately imported disciplinary neoliberalism to push through domestic reforms (see also Dyson and Featherstone Citation1996).