Accounting for Inflation Ambiguity

New research reveals that bond investors extract a premium not only for risk but also for uncertainty.
June 25, 2009
Print this page

In the late 1970s, inflation began to run rampant, and Federal Reserve chair Paul Volcker lead the central bank into an era of monetary-policy experimentation designed to curtail the rapid rise in prices. Volcker made a number of moves in quick succession, targeting money growth, increasing the federal funds rate 3 percentage points in March 1980 and then cutting it by more than 7 percentage points between April and July of that year. The annual inflation rate peaked at 13.5 percent in 1980; by the end of 1983, it had returned to a relatively tranquil 3.2 percent.

In that the Fed’s policy experiments had the intended effect, the period of experimentation from 1979 to 1983 was successful. But it was a wild ride for investors as the Fed’s actions, coupled with the looming questions of whether inflation would retreat and for how long, stoked investor uncertainty. U.S. bond prices and yields fluctuated dramatically, and bond investors eschewed what they viewed as riskier long-term bonds, driving down prices. Instead, investors flocked to the safe haven of short-term bonds, driving those prices up and yields down.

To understand how inflation might move and how it shapes bond prices and yields — their term structure — economists try to model future inflation. Term structure is typically attributed to three factors: expected inflation, expected output growth (the increase in goods and services produced), and the inflation premium.

The inflation premium has typically been thought to reflect the reduced price and increased yield an investor typically requires to offset the risk associated with investing in bonds. For example, when future inflation is expected to remain stable or to increase at a slow, steady rate, an investor might be willing to pay 60 cents now for a 10-year bond that will pay one dollar in 10 years; should the outlook on inflation become more volatile, the price of long-term bonds might shrink from 60 to 50 cents — a premium of 10 cents, or almost 17 percent — to offset the risk posed by volatility.

But, Professor Maxim Ulrich says, “projections of current term structure models don’t account for the fact that it is impossible for investors to know the true statistical distribution of future inflation. At best, inflation models reflect approximations about that distribution.” This suggested to Ulrich that risk might not be the only factor contributing to the inflation premium.

Ulrich theorized that during times when inflation was very volatile, such as the experimental era of the early 1980s, investors would require a premium not only to offset inflation risk but also to offset inflation ambiguity and, in particular, ambiguity resulting from the possibility that the Fed might not want to fight inflation. An ambiguity premium might exist distinct from (and perhaps be hidden within) the inflation risk premium.

To test this theory, Ulrich built a nominal term structure model that accounted for this suspected ambiguity premium, in part by introducing investor uncertainty about the true statistical distribution of inflation into calculations of bond pricing. Previous models assumed that the statistical distribution of inflation is known; in Ulrich’s model, the investor is uncertain about the future inflation process, which more closely reflects reality. Ulrich plugged in U.S. data over a 30-year period and found that bond prices generated by his model closely correlated with actual bond prices over the same period.

The results confirmed that during times of high inflation uncertainty an inflation-ambiguity premium comes into play and can be distinguished from the inflation risk premium. During the inflation scare period of 1983–84, U.S. long-term bond rates increased dramatically, although inflation remained constant at 4 percent. Ulrich’s model attributes the increase in bond rates to a rising inflation ambiguity premium, a finding consistent with the generally accepted perception that inflation scares are periods of high inflation uncertainty and strong skepticism about the Fed’s future inflation policy.

Ulrich was also able to show that the inflation-ambiguity premium disappeared when inflation uncertainty dropped during the Great Moderation — a period beginning in the mid-1980s marking a decline in the volatility of a number of major economic indicators. Overall, the presence of the inflation-ambiguity premium suggests that uncertainty must be taken into account as a factor distinct from risk when considering the term structure of bonds. The ambiguity premium also helps to explain the flight to quality — investors prefer to hold short-term Treasuries instead of long-term Treasuries during periods of high ambiguity to offset uncertainty.

The Fed could exert some influence over the ambiguity premium and effectively keep it in check, Ulrich suggests. One way to achieve this would be to adopt inflation targeting, setting explicit inflation goals as a way to keep inflation from rising or falling too abruptly and taking steps to try to insure that inflation stays within those set bounds. “When central banks can credibly commit to such goals,” Ulrich says, “ambiguity tends to remain very low because when the Fed embraces an official policy, investors are assured that the bank will take a variety of actions designed to keep the inflation rate in that target range.” While the exact rate of future inflation might remain unpredictable, the explicit commitment tells investors that inflation is likely to remain within a predictable range.

The same is true of a monetary-policy rule called the Taylor rule, which provides guidelines for how much the nominal interest rate should change in response to GDP and inflation. The Fed has not explicitly adopted the rule but frequently acts in accordance with its prescriptions.

To complicate matters, the Fed’s apparent hesitation to adopt inflation targeting and the Taylor rule has not stopped it from taking many actions that aim to keep the annual inflation rate in the 2 to 3 percent range. But unless the Fed makes an explicit commitment to officially adopt inflation targeting or the Taylor rule, investors will remain uncertain about which actions the Fed can reasonably be expected to take. That lack of certainty, especially when financial markets are volatile, is likely to ensure the persistence of an ambiguity premium in bond prices.

Ulrich’s work supports the common notion that if the Fed increased its efforts to communicate the rules and factors that constitute its inflation-control efforts, then investors’ inflation uncertainty would decrease and consequently help to suppress the inflation-ambiguity premium. In the meantime, Ulrich says, it’s best for investors to acknowledge the ambiguity premium when assessing bond prices and expected yields.

Maxim Ulrich is assistant professor of finance and economics at Columbia Business School.


Maxim Ulrich

Maxim Ulrich was a Columbia Business School faculty member from 2008 to 2010.