It is crucial that monetary policymakers know what expectations the public has for inflation rates over the short and long term. One reason is that inflation expectations help policymakers to gauge the public’s perception of the central bank’s commitment to maintaining a low and stable rate of inflation. Even more important is the fact that long-term inflation expectations—if they are stable—often mute short-term movements in inflation. So keeping long-term inflation expectations well-contained is desirable.
Inflation expectations, despite their importance, are notoriously difficult to measure. One well-known measure comes from the University of Michigan’s Survey of Consumers. Households are asked what they expect inflation will be on average over the next year (short-term expectations) as well as in the next 5 to 10 years (long-term expectations). According to this measure, mean short-term inflation expectations have crept up more than 50 basis points since the beginning of 2007. Over the past month alone they increased 50 basis points, although part of that increase served to erase some of the improvement that had occurred over the summer. Longer-term expectations have remained fairly steady, not showing any discernible movement over the year.
Survey measures are greeted with skepticism in some quarters. Since they do not reflect market transactions, responses do not necessarily reflect what market participants are truly expecting to happen. Furthermore, while there is evidence to suggest that changes in survey measures may be informative, the levels themselves are clearly not very accurate. It is doubtful that people really expect inflation to average 3.5 percent over the next 5 to 10 years. For that reason, many simply report the median versus the mean response from survey participants. This pushes down the average inflation rate that people expect over the next 5 to 10 years to 3 percent. Even this seems high, but it does not strain credulity as much.
As an alternative, many economists have looked to the TIPS market to get a truer sense of what markets are expecting inflation to average. In theory, the yields on two different kinds of Treasury securities—nominal Treasury notes and Treasury inflation-protected securities (TIPS)—can be used to calculate a market-based estimate of expected inflation. Nominal Treasury notes earn a fixed nominal rate of interest on a fixed amount of principal, whereas the principal of TIPS is adjusted for inflation. Because the return on nominal Treasuries is vulnerable to inflation, it compensates investors for inflation over the time they hold the bonds. In principle, one ought to be able to simply subtract the real yield on TIPS from the nominal yield of Treasury notes of the same maturity to derive expected inflation.
Using this method, one can derive inflation expectations over the next 5, 7, 10, and 20 years. This measure suggests that expected inflation in the next 5 or 7 years is around 2.25 percent. Over the next 10 years it is slightly higher, 2.4 percent, but in 20 years, it jumps to 2.6 percent.
Is it really realistic that market participants expect inflation to average a full one-half percentage point more over the next 20 years than in the next 7 years? Perhaps, but the fact that expectations for inflation in 20 years consistently run above those for other time frames suggests that something else may be going on. The answer is that TIPS estimates of inflation expectations are also imperfect. There are two factors that cause TIPS to be a biased predictor of expected inflation: an inflation-risk premium and a liquidity premium. To make matters more difficult, these biases likely go in different directions.
The existence of inflation risk suggests that the TIPS measure of expected inflation likely overestimates actual expected inflation. Nominal securities must compensate investors for the risk that inflation will change and affect the securities’ returns but by definition, a TIPS real return is constant. Because nominal securities include an inflation-risk premium but TIPS do not, the real return on TIPS will be less than the average return on nominal bonds. Over short periods of time the compensation for inflation risk will be pretty small, but over long periods like 20 years it can be quite substantial. Studies suggest that because of inflation risk, even the 10-year TIPS-derived measure will overestimate actual expected inflation by 50 to 100 basis points. Although this bias may not be constant over very long periods of time, monthly movements in the bias are probably not too important.
In contrast, TIPS returns contain a premium to compensate investors for liquidity risk. While the TIPS market is deep, it is probably less liquid than the market for nominal Treasury securities. Because of this relative liquidity difference, a TIPS real return should be greater than the real return on nominal government securities. As a result, TIPS-derived expected inflation will underestimate actual expected inflation. This difference, while typically small, can be important during periods in which there are severe liquidity concerns. While potentially important, the bias due to liquidity risk is more difficult to correct for than the bias due to inflation risk because it is likely not constant over time.
One measure of the market’s liquidity concerns is the difference in the yields on nominal Treasuries in the primary and secondary markets. (The primary market refers to bonds bought directly from the Treasury at auction, and the secondary market refers to bonds bought from other investors.) This difference, although small, can pick up broader liquidity concerns in the market. For example, during the Asian crisis this measure increased little, from less than 10 basis points to 15–20 basis points. The Russian default crisis saw a much more dramatic increase, with the difference in yields increasing nearly 30 basis points over the months of the crisis. Recently, this measure has increased 10 basis points, similar to the magnitude of the increase during the Asian crisis.
A 2004 study used this measure of the liquidity premium to correct TIPS 10-year expected inflation for liquidity risk. The study was largely a statistical exercise and, if correct, has pretty alarming implications for today. Given recent financial turmoil, liquidity-adjusted inflation expectations have increased nearly 50 basis points. According to this measure, market participants are expecting inflation to average nearly 3 percent over the next 10 years.
The liquidity correction used in this study, however, almost assuredly overcorrects. This is because the study was done when the TIPS market was a lot less liquid and deep than it is today. For example, volume in the market has doubled since then.
While the recent increase in TIPS-derived adjusted inflation expectations deserves our attention, it is still not clear that long-term inflation expectations have truly increased 50 basis points in the past two months.
A better measure of inflation expectations might be to use the separate 7-year and 10-year TIPS-derived inflation expectations to construct a measure of expectations for the combined period—7 to 10 years out. The problem of the liquidity premium is probably minor since liquidity concerns for 7- and 10-year TIPS are probably negligible. There is undoubtedly more inflation risk over 10 years than over 7, but this probably only biases up by a small degree the measure of inflation expectations 7–10 years out.
Unfortunately, this measure shows the same qualitative pattern as the 10-year liquidity-adjusted measure. According to the 7–10 year measure, there has been a marked increase since early summer in long-term inflation expectations of nearly 50 basis points.
Both the liquidity-adjusted measure of long-term inflation expectations and the measure for 7–10 years out suggest that inflation expectations may have crept up in response to the latest federal funds rate cuts and the probability of possible future rate cuts. Of course, these moves in the funds rate may be appropriate considering the recent financial turmoil caused by the housing market correction, and the possible specter of a recession.