The Fed's Approach to Risk Management

“[W]e may well at present be seeing the first stirrings of an increase in the inflation rate--something that we would like to happen.”  Stanley Fischer, Vice Chair of the Federal Reserve Board

The primary task of the central bank is to avert catastrophe, making sure that nothing really bad happens. This risk management approach imparts a natural asymmetry to policymakers’ words and deeds. Sometimes, it calls for bold, aggressive action. Others times, it means cautious plodding. Everyone agrees that 2008 was a clear case of the former. Most Federal Reserve officials argue that the current circumstance exemplifies the latter.

The argument is straightforward: The baseline forecast is for moderate economic growth, with labor markets continuing to tighten and wage gains slowly rising. Consistent with this, the FOMC’s projections show inflation returning to 2% over the next two to three years. But the risks—especially those arising from global factors—remain on the downside. It is easy to conceive of scenarios in which some combination of shocks from China, Europe and Japan cause inflation to fall short of the Fed’s objective and growth to falter. Moreover, conventional monetary policy tools are better at combating inflation that is a tad high than at countering deflation. That is, added to the asymmetry of risks is an asymmetry of policy effectiveness. As a result, it is no surprise that FOMC participants expect only gradual interest rate increases.

Let’s consider each point in a bit more detail, starting with the labor market. The following chart shows three different measures of labor utilization: the standard unemployment rate (labeled “U3”); a broader measure that also includes “marginally attached persons,” who are not currently looking for work but indicate that they want a job, are available for a job and have looked for work sometime in the past 12 months (U5); and a measure of underemployment that adds to U5 those who are employed part-time but would like to work full time (U6).

Alternative measures of labor utilization, 1994-March 2016

Source: FRED.

Reflecting the extensive cyclical progress in labor market normalization, all of these measures have fallen substantially since their 2009 peaks. Moreover, given the recent solid pace of employment growth, these labor measures would have declined further had it not been for the fact that people are entering (or re-entering) the labor force. After peaking at more than 67% in early 2000, the fraction of the working-age population either working or looking for a job fell below 63% in 2013 before starting to rise again. BLS projections suggest that the secular decline will continue: the forecast is for a participation rate of 61% in 2024! But we can have periods of above-trend participation that persist for some time, providing for substantial uncertainty regarding the supply of workers over the next few years.

The net effect of all this is that wage growth has been rising—gradually. The following chart shows both the conventional average hourly earnings index reported by the Bureau of Labor Statistics (BLS), along with a somewhat more sophisticated measure published by the Federal Reserve Bank of Atlanta. This second index, labeled the “wage tracker,” is constructed by carefully matching individuals’ responses across time. Thus, it adjusts for changes in the mix of high- and low-paying jobs, while the BLS index does not. For example, if new employees are hired into lower level jobs, as they often are, this depresses the BLS measure. As a result, the FRB Atlanta index provides a more accurate measure of what is happening to the wages of employing someone to do a specific job.

Measures of wage growth (3-month moving average of 12-month percentage change)

Source: FRED and Federal Reserve Bank of Atlanta.

Looking at the data, we can see that the trend of wage inflation now exceeds 3 percent. Importantly, removing the compositional effects that distort the official index makes it clear that the increase started in late 2013 and has continued thereafter. If compositional distortions fade over time, the combination of wage growth in excess of 3 percent and slow productivity gains (which have averaged only 0.4 percent per year over the past 5 years) already appears consistent with inflation of more than 2 percent over the medium term.

Might the labor market now be tight enough to start pushing wages up even faster, to the 4% levels we saw before the crisis? That remains to be seen, partly because the labor market may be functioning differently than before. For example, when the U6 was at levels around 8 percent, as it was in 2006, a decline to 7 percent was associated with a one-half a percentage point rise in wage growth. Today, it looks like a decline 10 to 9 percent is matched with a pickup of wage inflation less than half as large. This pattern could reflect the limited downward flexibility of wages (see our earlier post), but it also could reflect the lingering caution of workers in demanding higher pay after a sustained episode of extensive underemployment. Regardless, the behavior of wages going forward—how much more quickly they will rise if (as expected) the labor market continues to tighten—is only one of several sources of uncertainty that policymakers face.

Turning to price inflation, matters are far from being out of hand—in either direction. In contrast to the energy-driven sluggishness of headline indexes, our favorite measure of trend price changes, the trimmed mean PCE computed by the Federal Reserve Bank of Dallas, is already rising at a roughly 1.8% annual rate, barely shy of the FOMC’s target of 2%.

So, why the well-advertised caution about raising interest rates? As suggested, the answer is two-fold: (1) the downward bias of risks to the economic and inflation outlook; and (2) the asymmetric effectiveness of conventional policy tools. (You could add a third reason: that the equilibrium level of the funds rate currently appears well below past norms; see our earlier post.)

Starting with the risks, it is much easier to think of shocks that would depress, rather than stimulate, U.S. inflation and growth. Prime among them are the stresses and strains elsewhere in the world. For example, while the base case for China is a moderate slowdown, there is a clear risk of financial disturbances associated with the transition to a more flexible exchange rate and a more open capital market. As we emphasized in a recent post, there is a very real possibility of a sharp RMB depreciation, as domestic Chinese households and firms seek to reallocate their investments abroad amid growth disappointments at home. Added to this is the fact that economic and financial conditions in Europe are far from settled. The problems in Greece have not been solved, while inflation in the euro area remains far below the ECB’s target. The same is true in Japan, where pressure is growing for an even more aggressive monetary policy expansion.

Each of these examples points to the possibility of a new bout of dollar appreciation that would dampen U.S. inflation and growth. Indeed, these risks already may be weighing on U.S. long-term inflation expectations. Five-year forward, five-year (CPI) inflation expectations based on bond prices, which averaged 2½ percent from 2010 to 2014, have declined steadily over the past two years and are below 2 percent today. This is consistent with PCE price inflation expectations of less than 1½%.

Finally, as the U.S. central bank focuses on securing its inflation objective over the medium term, while sustaining moderate growth and high employment, FOMC members are acutely aware of policy’s starting point: a near-zero interest rate and a very large balance sheet. We doubt that any Fed official is enthusiastic about a scenario that would compel them to restart their massive asset purchase programs. And, while Chair Yellen did not rule out negative rates when she spoke at the press conference last month, it is also very unlikely that anyone relishes the idea of testing precisely how low rates can go (see our earlier post). Finally, if the Fed’s 2% inflation objective is really symmetric—a point that was emphasized in the January revision of the FOMC’s “Statement on Longer-Run Goals and Monetary Policy Strategy”—then policymakers should be just as willing to overshoot as to undershoot the target.

All of this leads us to agree with Federal Reserve Bank of Chicago President Charles Evans. In a speech last week, President Evans argued that a risk management approach means that policy should move gradually; above all, ensuring that inflation doesn’t fall persistently and unemployment doesn’t rise significantly. Put differently, after 8 years of worrying about the collapse of the financial system, high unemployment and the possibility of deflation, what’s so bad about having inflation temporarily exceed the target a bit, if it is then easier to secure that target over the medium term? This is a problem that the most of the FOMC would love to have because they believe they have the tools to manage it.