Why should we believe the market this time?

Paul De Grauwe 07 February 2009

a

A

Spreads of sovereign debt within the eurozone have increased dramatically during the last few months. Figure 1 shows the evidence. The governments of Greece and Ireland now pay an interest rate on their debt that exceeds the German government bond rate by more than 200 basis points, while the governments of Spain, Portugal and Italy have to pay more than 100 basis points extra.

Figure 1.

Source: ECB, and FT for Jan 09

Since all these bonds are expressed in the same currency, the euro, these spreads reflect a pure default risk (assuming that the German bonds are risk-free). Thus the market now takes the view that the governments of these countries will default on their debt in the foreseeable future.

How reasonable is such a forecast? Answering that question remains difficult. The fact that none of the Western European countries have defaulted on their debt in the last sixty years does not preclude that they will do so in the next fifty years. It is difficult to understand, however, why the market (and the rating agencies) forecast a default of the Spanish government debt, while they do not forecast trouble for the UK, which has a debt build up similar to Spain’s and a more serious banking problem.

Flight to safety

My hypothesis is that the widening bond spreads within the eurozone are the result of panic in the financial markets. The panic that followed the banking crises has led investors into a stampede away from private debt into assets that are deemed safe. These are mainly government bonds of a few countries. The US, Germany, and possibly France are a few of these countries that, for some strange reason, have been singled out as supplying safety. Other countries do not profit from the same “panic flight to safety”. This is shown in Figure 2, which presents the levels of the government bond rates in the eurozone. The German government bond rate has declined by close to 100 basis points. Surely it cannot be the case that holding German bonds in February 2009 is significantly less risky than a year earlier. But Germany was singled out by the market as the country that provided safety. France also benefited from this, though less so. With the exception of Greece and Ireland, countries kept their bond rates more or less unchanged (compared to a year ago), suggesting that these countries were bypassed by panicky investors. Only Greece and Ireland saw their bond rates increase significantly over the last year, suggesting that the increased spreads of these countries are not only due to panic.

Figure 2.

Source: ECB, and FT for Jan 09

Panic is distorting

Panic is never a good guide for risk analysis. It distorts prices and gives wrong incentives. During the decade preceding the eruption of the financial crisis in August 2007, rating agencies and market participants, gripped by euphoria, systematically underestimated the risk inherent in a wide range of financial assets. Today, the panic that has gripped them leads to an equally distorted view of the risks involved. Private debt is being dumped in favour of government debt of just a few countries. How these countries are selected is unclear. This selection mechanism is the result of an emotional reaction that leads market participants to believe that “it is inconceivable that serious countries like the US, Germany, or France would ever default on their debt” while other countries are deemed to be to be capable of doing such a bad thing. These emotional reactions create distortions that affect economic activity in profound ways.

The interest rate spreads faced by Southern European countries and Ireland are giving those governments incentives to reduce their efforts to stabilise their economies. Extra spending which leads to higher deficits are punished by a higher interest cost, discouraging these countries from stimulating their economies. No such penalties are imposed on Germany and France.

In addition, these spreads create a perception of future default crises and impending fiscal doom. This impacts the effectiveness of budgetary policies. We know that fears of future default crises reinforce the “non-Keynesian” effects of fiscal policies, i.e. when agents fear such future crises, they are more likely to react to budgetary stimuli by increasing their savings (see the well-known paper of Giavazzi and Pagano 1996)). As a result, budgetary stimulus packages lose their effectiveness.

The penalties imposed by increasing spreads on Southern European countries and Ireland also create negative externalities. For example, the rescue of banks in these countries is more expensive than in the rest of the eurozone, making it more difficult to resolve the banking crisis in these countries. This is likely to lead to further weakening of economic activity in these countries with possible feedback again on the banking system, government budget deficits, and ratings applied by the rating agencies.

How to dampen these distortions and externalities?

It will remain difficult to prevent cycles of euphoria and panic. Authorities can, however, attempt to offset the distorting effects these cycles of risk perceptions produce. The ECB, in particular, has the tools to do just that. Here is my proposal. As the ECB will be forced very soon to engage in quantitative easing, it will be buying long-term assets – in particular, government bonds. It should privilege the buying of Irish, Greek, Spanish, and Italian government bonds. In doing so, it would increase the price of these bonds and reduce their yields. Such a quantitative easing would tend to reduce the spreads in government bonds in the eurozone and eliminate the resulting distortions and externalities. It would also make it possible to stimulate the economies of all eurozone member countries, benefiting the whole area.

Clearly this proposal will meet with much resistance. We now increasingly hear the story that the market is punishing profligate governments, and rightly so. But why should we trust the market’s judgment any more now than five years ago when the same market (orchestrated by rating agencies) completely misjudged the risks that existed in the financial markets? When the market was so wrong in the past, why should we believe it now and accept its verdict? We should not. The spreads that have been created within the eurozone are largely the result of panic. We can and should correct panic-induced distortions and externalities.

That being said, it remains the case that Spain, Ireland, Italy and Greece face important adjustment problems. These countries have experienced relatively high wage and price inflation in the last ten years, leading to a loss of competitiveness, as shown in Figure 3. Part of the higher domestic price and wage inflation may be due to the Balassa-Samuelson effect, but this is unlikely to be the case in all these countries. Spain and Italy, for example, have experienced sluggish productivity growth. It follows that these countries will have to go through a period of adjustment redressing their competitive position, very much like Germany did during the first five years of the eurozone. This will be a painful process. The overriding problem these countries face today, however, is the same as everywhere else, i.e. how to counter the deflationary effects of the financial crisis. There is no reason why these countries should be singled out for punishment. That would only make it more difficult to adjust to the deflationary dynamics. It would also not help them in regaining their competitive positions.

Figure 3. Relative unit labour costs in Eurozone

Source: European Commission, European Economy

Reference

Giavazzi, Francesco and Marco Pagano (1996): “Non-Keynesian effects of fiscal policy changes: international evidence and the Swedish experienceSwedish Economic Policy Review, May.

a

A

Topics:  International finance Macroeconomic policy

Tags:  eurozone, sovereign debt, spreads, government bonds, flight to safety, financial panic

Professor of international economics, London School of Economics, and former member of the Belgian parliament.