European policymakers are confronting a heightened crisis characterised by a perverse and seemingly intractable interplay between sovereign debt pressures and financial-sector fragilities (Wolff 2011). Three questions arise:
- How did the crisis reach this stage?
- How should we think of policy options when the financial sector and sovereign weaknesses threaten to pull each other down?
- What policy interventions have the best chance to succeed at this time?
To shed light on these questions, we analyse the correlation structure between weekly changes in sovereign spreads and banks’ equity valuations in a panel of Eurozone countries (Mody and Sandri 2011).
The phases of the crisis
Our econometric results document the changing nature of the crisis across several different phases (Figure 1).
- Before the subprime crisis, sovereign spreads (the risk premia over German bunds) were virtually indistinguishable across Eurozone countries.
- From the subprime crisis in to the Bear Stearns rescue (July 2007–March 2008), spreads rose moderately.
This phase saw an intensification of financial stress in the US. In our statistical work, we capture this with the credit default swap on US banks. There was, however, as yet very little differentiation across EZ countries. At this point, the US crisis was creating generalised financial pressures worldwide.
Figure 1. Financial-sector equity valuations and sovereign spreads (Eurozone average)
- After the rescue of Bear Stearns, spreads rose and became differentiated across countries, reflecting the heterogeneity in domestic financial problems.
This was in response to the deterioration of domestic financial conditions, captured in our analysis by losses in the stock-market valuations of the financial sector relative to the overall market.
Even though Bear Sterns was a US bank, its rescue signalled the severity of the crisis and especially governments’ intentions to support financial institutions (Reinhart 2011). This interpretation, we believe, spilled over to Europe.
- Our econometric analysis reveals that sovereign spreads in this phase increased with a lag of two to three weeks following the downgrading of banks’ stocks, suggesting that markets were absorbing the implications of banks’ financial health for the future state of public finances.
The lag may reflect the difficulty of the assessment. It required an understanding of how banks’ losses would weaken sovereigns’ fiscal positions directly through bailout costs and indirectly by reducing credit supply and GDP growth. It also required forming expectations about the effectiveness of public support in supporting credit flows and avoiding contagion.
Our results also show:
- The nation-level differentiation in spreads reflected not only the heterogeneity in domestic financial problems, but also differences in the level of public debt and growth potential.
An equal loss in the stock-market valuation of the financial sector translated into a larger increase in sovereign spreads for countries with a higher fiscal burden and lower growth potential.
- The problems entered a new phase – becoming a full-blown crisis – with the nationalisation of Anglo Irish in January 2009.
The relevance of Anglo is, at first, not obvious, since it was a small bank in a relatively small country. However, the data quite robustly suggest a break at this point. It is possible that the large fiscal costs as a share of Ireland’s GDP associated with this rescue raised serious concerns about fiscal sustainability. Suddenly, the ability of the sovereigns to support the financial sector came into question. The worrying news a few months later about Greece’s fiscal imbalances confirmed that the Eurozone crisis had evolved from a banking crisis into a sovereign crisis, as documented for other countries by Reinhart and Rogoff (2011).
The sovereign and banking problems start to move in lockstep
The changing nature of the crisis is reflected in the econometric analysis through a change in the autocorrelation structure between sovereign spreads and banks’ equity valuations. Stock-market losses on financial firms no longer precede the increase in sovereign spreads; rather, the correlation is now contemporaneous. While in the months preceding, financial-sector shocks had led to concerns about sovereign-debt sustainability, now shocks to fiscal sustainability also became operative by weakening the prospects of banks. It was as if a fiscal shock signalled more limited ability of the sovereign to support banks and, as such, an increase in sovereign spreads began to be associated with increased market pressure on banks. This was particularly the case since increases in sovereign spreads created valuation losses on banks’ holdings of government bonds. At this point, the sovereign and banks were joined at the hip, with their respective weaknesses threatening to reinforce each other.
A negative feedback loop between banks and the sovereign creates the risk of multiple equilibria, where small shocks can get amplified into larger unsustainable positions. Markets do not have the capacity to stop the slide down this slippery slope. Thus, while in the pre-subprime crisis phase, markets misjudged by ignoring all risk (and creating an unwarranted convergence of spreads), recently, markets have been party to pushing banks’ and sovereigns’ financial conditions into deeper and self-fulfilling distress.
Financial markets fostered the very imbalances that they are so ferociously punishing now. The resulting destabilisation has had grave consequences. Some may argue that the destabilisation was inherent in the design of the Eurozone. That may well be the judgement of history, but clearly the markets have greatly amplified it. As plans are wrought for the future of the Eurozone, much thought is being rightly given to fiscal integration and fiscal transfers. But it would be a mistake to forget that a weak banking system can have far-reaching costs, which no amount of fiscal rectitude can eventually handle.
A lesson of our paper is that banking fragilities are not primarily to be sought in their cross-border dimensions but even the domestic negative feedback loops from banks to sovereigns and back can be potent. As such, a robust banking sector will be central to a more stable Eurozone.
Once the crisis begins, the choices become increasingly unpleasant. In retrospect, the nature of the crisis prior to Anglo Irish was simple, being mostly driven by problems in the financial sector. Since sovereigns, though under increasing pressure, still enjoyed market confidence, the provision of public support to banks was viewed favourably by the market and from an economic perspective also offered the best potential to halt the crisis from spiralling in an uncontrolled manner.
The policy advice in such conditions is to act early and decisively. Actions within the Eurozone to support the banks helped maintain credit supply and, to the extent banks are leveraged, bolstering their capital positions created the potential for leveraging public finances for growth. Obviously, this process had its limits, as future events have harshly revealed. The appropriate extent and form of public guarantees was hard to gauge, and, looking back, perhaps the extent of the support was excessive in some cases. The winding down of Anglo Irish, for example, would have been preferable to its nationalisation, not least to avoid concerns about moral hazard. But given that banks’ fragility was the key determinant of the crisis and that many governments had enough fiscal space, the prompt and resolute support of the financial sector was warranted.
Policy options are now much more constrained. Investors are questioning the fiscal sustainability of several countries and this is creating additional pressure on those banks that are directly exposed to sovereign bonds. Clearly the ability of such countries to support their financial sectors is compromised. A more precise policy approach is needed. It is important to keep in mind that the payoffs from strengthening banks’ balance-sheet can still be large and, therefore, fiscal support is merited. But a more resolute strategy for winding down banks is also needed.
Where banking models have failed or proven fragile, a credible approach to phasing down such banking operations is needed. At the start of the crisis, bank resolution powers were relatively limited in most European economies. A lesson learned from the crisis is the need for such powers, which now are more widespread. In this regard time is of the essence, as the crisis has once again demonstrated that delays can be extremely costly. Prompt action may still have the potential to turn around the current vicious cycle between sovereign downgrades and banks’ losses.
Editor’s note: The views expressed here are those of the authors and not necessarily those of the IMF, its management, or its Executive Board.
Mody, Ashoka and Damiano Sandri, 2011, “The Eurozone Crisis: How Banks and Sovereigns Came to Be Joined at the Hip”. Paper presented at the 54th Economic Policy Panel Meeting in Warsaw, 27-28 October 2011.
Reinhart, Vincent, 2011, “A Year of Living Dangerously: The Management of the Financial Crisis in 2008”, Journal of Economic Perspectives 25(1): 71-90.
Reinhart Carmen M. and Kenneth S. Rogoff, 2011, “From Financial Crash to Debt Crisis”, American Economic Review 101(5):1676-1706.
Wolff, Guntram, 2011, “Is the recent bank stress really driven by the sovereign debt crisis?”, VoxEU.org, 30 October.