Inflation Expectations: How Credibility Pays Off

Monetary policymakers always worry about inflation expectations. They can’t directly observe what households and business anticipate for the future path of prices, so they construct estimates from market prices and surveys. Why do they care so much? The reason is simple: keeping inflation expectations low and stable is the first step to keeping inflation low and stable. It also makes the economy more resilient in the face of adverse shocks.

The good news is that U.S. long-term inflation expectations currently are stable and have been for some time. To see this, let’s start with a market-based measure of inflation expectations: the “break-even” inflation rate. The break-even rate is the difference between the yield on a standard nominal bond, one with payments that are in fixed-dollar amounts and at a fixed coupon rate over its entire lifetime, and the yield on an index bond of the same maturity that compensates the owner for inflation. In the United States, for example, we can compare a Treasury bond yield with the yield on a TIPS (Treasury Inflation-Protected Securities) issue with the same maturity.  The principal amount of TIPS adjusts with inflation measured by the consumer price index (CPI), so the yield that you see quoted is a real yield [with one exception: when paid out, the TIPS principal is not reduced by deflation below the original principal]. By subtracting this real yield from the nominal yield on bonds of the same maturity, we get a market-determined estimate of expected CPI inflation. (This uses the well-known Fisher equation.)

Given the complexities of the bond market, it takes some work to make the nominal rates directly comparable to the real rates, but researchers at the Federal Reserve Board have done these technical adjustments for us.  (Their paper describing the statistical methodology is here, and the daily data itself is in a spreadsheet here.) We find two of their series particularly illuminating: expected inflation over the coming five years, and expectations for the five years after that. The first one gives us a sense of medium-term inflation expectations, while the second, what is known as the “five-year forward five-year inflation expectation” (that’s a mouthful), indicates what market participants think about the long-term trend in inflation.

Two things stand out about these expectations measures, both of which are plotted in the chart below from the beginning of 1999 to the middle of 2014 (the entirety of the data that are available).  First, medium-term inflation expectations (the black line) fluctuate with actual inflation (the blue line).  The correlation is roughly 0.5.  Second, long-term inflation expectations (the red line) are fairly stable, with a standard deviation less than one-quarter that of actual inflation and less than one-half that of medium-term inflation expectations. Not only that, but the level of long-term inflation expectations is currently right around 2½%.

Source: FRB and authors' calculations.

What should we make of these market measures? Investors – who have a considerable financial stake in what inflation turns out to be – are relatively sanguine about both the medium- and long-term inflation trends in the United States. They don’t see either a dramatic rise or a significant fall in inflation on the horizon, near or far.

Survey data point in the same direction. The Federal Reserve Bank of Philadelphia’s survey of professional forecasters shows their long-term inflation expectations have been remarkably stable in the range of 2% to 2½% (see chart). Even the dispersion of their forecasts (the interquartile range) is currently in line with its historical norm (0.5 percentage points), despite heightened uncertainty about U.S. economic prospects. Similarly, the University of Michigan’s survey of consumers shows that household expectations for five-year inflation are very stable, with a standard deviation less than one fourth that of their one-year inflation expectations (see Table 33 here).

Survey of Professional Forecasters: 10-Year Inflation Expectations

Source: Federal Reserve Bank of Philadelphia.

Why are long-run inflation expectations so stable? The most likely reason is the credibility of the Federal Reserve’s commitment to price stability. The Fed aims at a rise in the price index of personal consumption expenditures (PCE) of 2% annually over the long run. As we pointed out in an earlier post, from 1990 to 2007, PCE inflation averaged 2.1%.  For various technical reasons, CPI inflation averages nearly ½ percentage point higher than PCE inflation over time. (From 1990 to 2007, the average difference was 0.4% per year.) That means that the current break-even expectation for long-term CPI inflation of 2.4% is consistent with the Fed meeting its 2% objective for PCE inflation. So, one reading of these expectations data is that investors and others believe that the FOMC will do what it takes to hit (or get close to) its stated inflation objective.

The Fed could cite many advantages of stable inflation expectations. One notable benefit today is the impact on private spending following a big negative shock, like a financial crisis. If inflation expectations were to fall precipitously in a crisis, this would drive up estimates of the real interest rate, encouraging households and businesses to postpone spending. The resulting decline would depress the economy further, validating the downward pressure on inflation. This could set off a downward spiral in which inflation expectations would fall further, driving output down even further, and so on.

To see the point, compare the inflation and growth experience during and after the Great Recession that began in 2007 with the experience of the Great Depression that began in 1929. Despite the weakness of the U.S. economy after 2007, neither inflation expectations nor inflation fell as much as many economic models calibrated to recent data predicted it would. That stability probably reflects, at least in part, confidence that the Federal Reserve would make good on its commitment to price stability. Back in the 1920s and 1930s, the Fed made no such commitment: while we have no good proxies in that era for inflation expectations, we know that the consumer price level plunged by nearly 9 percent annually from 1930 to 1933. The result was widespread bankruptcies as households and firms became unable to pay off their debts, and a fall in real GDP of roughly 30% (nominal GDP went down by more than half). As one of us has described in detail, it’s a good bet that people expected prices to fall, making real interest rates extremely high in the Great Depression. By contrast, during the Great Recession, real interest rates were low, or even negative, providing an important impetus for recovery.

Stable inflation expectations alone aren’t sufficient to avoid business cycles, even deep ones. But they help keep deep recessions from becoming historic depressions.