Inflation protection: one index doesn’t fit all

Will inflation spoil the ECB’s 10-year anniversary party? While the ECB has succeeded in keeping inflation close to two percent for almost nine years since its inception, at the start of 2008 inflation crept above three percent and has remained there since. This level is clearly incompatible with the ECB’s price stability objective. Moreover, inflationary pressures from energy and food markets show few signs of abating. A prudent monetary policy response would focus on preempting second-round effects via wage formation and inflationary expectations. A firm interest rate hike by the ECB is, however, hampered by worries about financial sector fragility and the potential consequences of the credit crisis for the real economy.

Uncertainty about the inflation outlook and the monetary policy response is reflected in the market for index-linked bonds. Demand for these bonds has surged this year, pushing so-called breakeven inflation rates to their highest level since years, as investors seek protection from soaring oil and commodity prices. Market participants complain about the limited supply of indexed linked bonds, which results in high prices and low real yields. In the euro area the supply is also of the wrong kind. Inflation protection predominantly comes in the form of plain-vanilla bonds indexed to a euro area consumer price index. These bonds are not tailored to the needs of users based in individual countries and fall short of offering full inflation protection. Nor do these bonds have very attractive diversification properties. If European governments would issue more bonds linked to national inflation rates, investors would be better able to lock in real-value certainty and the bond market would be enriched by a varied supply of index-linked flavors.

Index-linked bonds (ILBs) exist because they benefit issuers as well as investors. From the issuer perspective, ILBs may reduce borrowing costs and remove the incentive to create inflation. For investors, the main reason to invest in ILBs is to reduce inflation risk. This appeals most to investors for whom real value uncertainty is a major concern (e.g. pension funds). ILBs also show stable real returns compared to conventional bonds and equities. Unanticipated inflation shocks generate real return volatility in conventional bonds, but not in ILBs. This makes this asset class interesting from a portfolio perspective.

ILBs constitute a small but growing segment of the European bond market. On the European continent, ILBs have been introduced by the French state in 1998. In 2003 the governments of Greece, Italy and Austria followed. Germany issued its first ILB in 2006. While France has issued both ILBs linked to French inflation and ILBs linked to eurozone inflation, recent issuers have chosen to link their ILBs exclusively to a European index (the harmonized index of consumer prices excluding tobacco, abbreviated as: €HICPx). The share of ILBs linked to €HICPx has overtaken the share of ILBs linked to national price indices. This development is puzzling, if one considers that:

a) people consume locally (and the national consumer price index is thus a better proxy for the erosion of their purchasing power than a euro area price index) and

b) national inflation rates within the euro area are far from synchronized.

Is it really true that inflation uncertainty is still a national experience, ten years after monetary unification? After the introduction of the euro, the cross-country variation in the inflation rates has in any case not fallen quickly. Inflation differentials have also been quite persistent (much more so than in the United States). As a consequence, the consumer price indices of euro area countries have fanned out and seem to bend together only slowly, if at all (see the graph).

image002_53.gif

Over the period 1998-2007 the purchasing power of €1000 has diminished from €858 in Germany to a low of €707 in Ireland. Even if over time inflation rates would converge to the euro area average, there is no guarantee that the accumulation of past inflation deviations will be compensated by opposite deviations in the years to come. In that sense, national inflation shocks still entail considerable long-run price-level uncertainty even if the inflation rates themselves would have a tendency to converge.

Price level divergence is much more of a problem in the euro area than in the United States, the latter being a better integrated currency union with greater price/wage flexibility and higher mobility of production factors. The persistence of euro area inflation differentials is exacerbated by the macroeconomic adjustment process. With a uniform nominal interest rate, domestic real interest rates will be lower in high inflation regions, discouraging savings and stimulating consumption and investment. Compared to a monetary policy that is conducted nationally via a Taylor-type interest rate rule, within a monetary union the real interest rate channel no longer acts as a brake on the cycle but instead may accelerate regional economic developments. This effect may be further amplified by wealth effects, as low real interest rates inflate housing prices (Spain and Ireland being the obvious examples here). The remaining countervailing force is the appreciation of the real exchange rate. However, the elimination of nominal exchange rates has reduced the speed with which this variable adjusts. This implies that a strong cross-country synchronization of inflation cycles cannot be taken for granted inside the euro area.

Inflation uncertainty is likely to be a bigger problem for smaller euro area countries. ECB interest rate setting is based on aggregate euro area data. Because of their low weight, inflation shocks in small countries have a minor effect on euro area data. Large countries have greater influence on ECB policy. While investors residing in small EMU countries are faced with increased inflation uncertainty, their hedging opportunities are currently inadequate. This suggests that there may be a demand for inflation-protected securities aimed at investors in individual countries.

The data illustrate the usefulness of nationally indexed ILBs in hedging inflation risk. Nationally indexed ILBs will also contribute more in terms of portfolio diversification than euro-indexed ILBs. If the safeguards in the Maastricht Treaty do their work, the ECB will be able to keep inflation close to 2%. When market expectations center around this number, the return differential between euro-indexed ILBs and conventional bonds will be small and hence the added value of euro-indexed ILBs in an investment portfolio will be low. The sole remaining use of euro-indexed ILBs is as an insurance policy against a neglect of duty by the ECB. In a non-optimal currency area like EMU, national inflation shocks will continue to generate imperfect correlations between nationally indexed ILBs and conventional bonds. As a result, nationally indexed ILBs will be able to add much more value in terms of portfolio diversification than euro-indexed bonds.

There is one drawback. The superiority of nationally indexed bonds is compromised by their lower liquidity. Linking to €HICPx has increased the attractiveness of ILBs to foreign investors, leading to an extension of the investor base, a higher tradability and a lower liquidity premium. The €HICPx-linked market has also benefitted from market incompleteness. Investors seeking inflation protection where no nationally indexed ILBs exist (e.g. Dutch pension funds), will prefer euro-indexed bonds to ILBs indexed to another member state’s price index. With few countries issuing nationally indexed ILBs, this automatically creates a larger market and thus better liquidity for euro-indexed ILBs. This effect is self-reinforcing.

So the puzzle is that out of liquidity considerations a market has developed in an index-linked instrument which doesn’t fully satisfy the hedging needs of European investors and is least suited to add value in a diversified portfolio. Only in the special case that investors do not trust the ECB, euro-indexed bonds would be the right instrument to turn to. If the need for portfolio diversification or the wish to lock in real value certainty drives ILB demand, investing in nationally indexed bonds would make better sense.

2 Responses to "Inflation protection: one index doesn’t fit all"

  1. Guest   June 25, 2008 at 10:30 am

    Low liquidity in linkers will be just temporary after the popularization of global linker index funds as an alternative to the self-defeating use of commodities investment as an ‘inflation hedge’.