Which country has suffered from inflation lately?

Unbundling of the various influencing factors in the recent shifts in breakeven inflation rates

(Excerpt from pages 10-11 of the BIS Quarterly Review, March 2009)

Breakeven inflation rates, i.e. the yield differentials between nominal and indexed bonds, have fluctuated extraordinarily strongly in recent months; they sank to record lows before picking up again in early 2009. For example, the 10-year breakeven rate for the US fell to almost zero towards the end of 2008, after having previously remained at around 2.5% for several years (Graph A, left). A similar, albeit less pronounced, development was observed for the breakeven rates in the euro area (Graph A, center).

Against this background, the question arises to what extent these recent fluctuations can be classified as actual changes in inflation expectations. Breakeven inflation rates have long been used as an indicator of market inflation expectations over the bond time horizon. However, the large price swings that were recorded in numerous markets - including those for nominal and indexed bonds - during the financial crisis naturally reflected other influences in addition to fundamental factors. Overall, while falling inflation expectations may have contributed to the decline in breakeven rates, a significant part of the decline was likely attributable to other factors, including liquidity and market influences.

In principle, breakeven inflation rates can be broken down into four main components: i) expected inflation, ii) inflation risk premium, iii) liquidity premium and iv) market-specific factors.1 The relative importance of these components can vary over time, depending on the respective framework conditions in the economy and on the financial markets.

For example, the role of the first component can be assessed using other indicators of inflation expectations, such as forecast data from surveys. According to the "Survey of Professional Forecasters" (SPF) conducted by the Federal Reserve Bank of Philadelphia, expectations for US inflation on an annualized quarterly basis fell from 2.9% in Q3 2008 to 0.8% in 1st quarter of 2009; in the longer term, over a time horizon of 10 years, the expected inflation rates only fell by 0.1% to 2.5% in the same period. but for long-term (five-year time horizon) inflation expectations, a corresponding survey by the ECB also showed a decline from just 0.1% to 1.9% in the period from 3rd quarter 2008 to 1st quarter 2009. These results suggest that Average inflation expectations for the next few years have changed little, despite rapidly falling short-term expectations. However, given the massive shocks that have hit the economy young has recently been shaken, some observers have expressed doubts about the plausibility of consistently stable average inflation expectations over a long time horizon.

As for the inflation risk premium, recent estimates suggest that this component is generally relatively small and quite stable.2 If so, it is unlikely that the inflation risk premium played a major role in the observed changes in break-even inflation rates. However, some estimates also contain indications that inflation risk premiums are positively correlated with the inflation rate. As a result, in line with falling breakeven rates, the recent fall in inflation may have resulted in a lower inflation risk premium. However, intuitively, it seems plausible that the inflation risk premium may have risen as inflation has become more volatile and there is some uncertainty about the possible price effects of recent monetary policy measures.

The (broadly defined) liquidity premiums, on the other hand, have in all probability played an important role in the development of breakeven rates. During the market turbulence, there was a massive flight into liquidity, which caused the demand for nominal government bonds to rise sharply and which is likely to have led to a negative premium in this segment. In other words, this pushed nominal yields to extremely low levels, which in turn led to strong downward pressure on breakeven rates. In addition, since the markets for inflation-indexed bonds are much less liquid than the markets for nominal bonds, investors in indexed bonds have more difficulty in closing positions in these markets quickly at the prevailing market prices. Under normal circumstances, this factor typically creates a relatively small liquidity premium on inflation-linked bonds. However, as liquidity risks increased, this premium has likely increased significantly, and aversion to that risk grew as the crisis deepened in the second half of 2008. As a result, the yield on indexed bonds could in turn have increased compared to the yield on nominal bonds, which may have increased the downward pressure on breakeven rates.

Closely linked to these liquidity effects and in some cases hardly distinguishable from them are market-specific factors that apparently also played a significant role in the recent development of breakeven rates. One such factor is, for example, selling pressure from leveraged investors who have had to liquidate their positions in inflation-indexed bonds under adverse market conditions; as a result, real yields rose and breakeven rates fell.3

Data from the inflation swap markets can shed light on the significance of these effects. An inflation swap is a derivative instrument that is similar to an ordinary interest rate swap. However, a fixed interest rate is not exchanged for a variable interest rate linked to a short-term interest rate, but rather the variable payment in the inflation swap is linked to an inflation measure, usually the inflation rate that has accrued during the term of the swap. The fixed part of the inflation swap thus provides a direct breakeven "inflation price" that is unaffected by any different liquidity conditions on the markets for nominal or inflation-indexed bonds or by a flight into liquidity.4

After the difference between the prices for 10-year inflation swaps and the corresponding breakeven rates for bonds had been stable in recent years, it widened considerably towards the end of 2008 (Graph A, left and center). This indicates that the aforementioned liquidity and market-specific effects played an important role in the breakeven rates of the bond markets. Nevertheless, the inflation swap rates also fell significantly towards the end of 2008. This move is in line with expectations for lower inflation, but is also likely to be attributed to hedging break-even positions in the bond markets. The decline in breakeven rates was partially offset in early 2009, possibly because investors wanted to take advantage of breakeven inflation levels that were viewed as overly depressed.

Finally, it may be worth taking a look at the breakeven rates for the more distant future. For example, the 5-year forward rate in 5 years' time is often seen as more meaningful for long-term inflation expectations than, for example, the 10-year break-even rate, since - at least in principle - it should not be influenced by short-term inflation expectations. Such forward breakeven rates have become much more volatile in recent months, but overall there is no clear change in their level (chart A, right). The lack of such a change could be an indication that longer-term inflation expectations have remained broadly stable. This, in turn, agrees with the view that central banks' credibility in terms of their determination to maintain price stability has not suffered from the recent rapid rate cuts.

1 Break-even inflation rates can be influenced by other factors, including seasonal factors and so-called carry effects. The former are well-known seasonal fluctuations in consumer prices; these influence the price of bonds that are linked to non-seasonally adjusted consumer price indices. Carry effects relate to longer-lasting changes in consumer prices, such as those that are related to the development of oil prices. They are known to affect current inflation, while inflation-indexed bonds are linked to an index of prices several months old. However, these effects tend to be relevant mainly for short terms, around up to two years.
2 S. P. Hördahl, "The inflation risk premium in the term structure of interest rates", BIS Quarterly Report, September 2008, pp. 23-38, and the references contained therein.
3 Another market-specific factor is the value of the embedded deflation floor, which has increased recently for many inflation-linked bonds, particularly for newly issued bonds that are close to the floor. Increased values ​​for the deflation floor mean higher prices for the indexed bonds concerned, which results in lower real yields and thus higher breakeven rates. This latter effect would not explain the recently observed decline in breakeven rates. In addition, when calculating the real returns on zero coupon bonds and break-even inflation rates, recently issued indexed bonds are excluded. Therefore, the deflation floor is likely to have played a negligible role in the breakeven inflation data.
4 Of course, this does not mean that inflation swaps remain unaffected by any market-specific factors, e.g. also hedging effects. In addition, the inflation swap markets are typically less liquid than the bond markets.