May 24, 2011
Is The Core Inflation Concept Rotten?
St. Louis Fed President James Bullard recently attacked the concept of core inflation, which excludes volatile food and energy prices. The "core is rotten," he argued in a speech last week. "One popular argument for focusing on core inflation is that core inflation is a good predictor of future headline inflation," he said. "I think this is wrongheaded, as well as wrong."
It’s fair to be skeptical of core inflation as a flawless predictor of future headline inflation. Nothing rises to that standard. It’s also prudent to recognize that there’s more than one way to measure core inflation. But to reject the idea entirely is going too far because it requires dismissing several decades of research and the historical record. Bullard obviously disagrees, as do other analysts, but such opinion is hardly airtight.
Still, it’s easy to see why there's so much fuss about core vs. headline inflation. After all, the man on the street must routinely pay for food and energy and so the idea that inflation should be measured without these essential commodities sounds crazy. Indeed, Bullard comes close to espousing a political view in rejecting core. "One immediate benefit of dropping the emphasis on core inflation would be to reconnect the Fed with households and businesses who know price changes when they see them," he noted. Reconnect? It's slightly dangerous to suggest that central banks should start considering what's popular rather than what's economically logical. Recognizing what’s practical for consumers isn’t necessarily enlightened when it comes to designing monetary policy.
Bullard’s charges against core inflation are hardly an obscure academic point since the Federal Reserve does pay quite a bit of attention to the notion of core for setting monetary policy. So when the president of one of the banks in the Fed system publicly dismisses the idea, it’s big news. The problem is that the basis for using core inflation didn’t arise in a vacuum. The focus on core has evolved via years of research that suggests that this narrower gauge inflation is a better predictor of headline inflation for the medium- to long-term future. A few of the many examples in the literature that favors core include:
"A Review of Core Inflation and an Evaluation of Its Measures" by Robert Rich and Charles Steindel (NY Fed staff report, 2005).
"A Simple Adaptive Measure of Core Inflation" by Timothy Cogley (Journal of Money, Credit and Banking, Feb. 2002)
"Comparing Measures of Core Inflation" by Todd Clark (Kansas City Fed Economic Review, 2nd Quarter 2001)
Bullard gave a passing nod to the body of research that supports core's role in monetary policy, but then he largely dismissed it on the basis of a few recent papers that suggest otherwise. Again, that’s a sign that there can and should be serious debate about the value of core, but it’s premature to reject it out of hand. Former Fed Governor Frederic Mishkin explained the reasoning a few years ago:
It does indeed make sense for central banks to emphasize headline inflation when determining the appropriate stance of monetary policy over the medium run, but policymakers also are right to emphasize core inflation when deciding how to adjust policy from meeting to meeting. Why? Because what central bankers are truly concerned with--both for the purposes of internal deliberations and for communications with the public--is the underlying rate of inflation going forward, and core inflation can be a useful proxy for that rate. Thus, focusing on core inflation can help prevent a central bank from responding too strongly to transitory movements in inflation.
Another former Fed governor, Larry Meyers, who now heads up Macroeconomic Advisers, made a similar case recently in a piece for the New York Times:
There are two fundamental measures of inflation: overall (or "headline") inflation and "core" inflation, which excludes food and energy prices because they are very volatile and mostly transitory and as a result don’t necessarily reflect underlying inflation trends. A central objective of the Fed’s monetary policy is price stability, defined as a low, steady rate of overall inflation. So are rising food and gas prices a sign that the Fed is falling down on the job?
The answer is no. There is very little that the Fed can do to control today’s inflation, whether core or headline. What the Fed does influence is inflation a year or two down the road, which is why it needs to look to the future, not overreact to the present.
The most significant question for the Fed, then, is whether overall or core inflation right now is a more reliable gauge of where headline inflation will be next year. And the data unequivocally tell us that core inflation better predicts overall inflation tomorrow.
But talk is cheap when it comes to inflation. Far more persuasive are the numbers, and on that score there's still plenty of support for taking core seriously. As Marcus Nunes recently reminded, reviewing the long sweep of history between core and headline inflation should give us pause before throwing out the former in favor of the latter. Consider that during the inflationary surge of the 1970s, core inflation and headline inflation indices were rising in tandem. This was a true inflationary episode. As Nunes notes, "What stands out is the fact that during the 'Great Inflation' of the 1970s, the ‘sticky components’ became unstuck! That´s what we mean by inflation: a continuing rise in all prices." This is quite clear in the chart below, which compares the unadjusted annual rates of headline consumer price inflation (blue line) with its core reading (red line) from 1960 to the present.
This time there’s more of a divergence between headline and core. In recent years, headline inflation has popped up when energy and/or food prices spiked. But as the trend in core inflation suggests, these pops were temporary and not necessarily indicative of an inflation problem per se, at least not to a degree on par with the 1970s. Nonetheless, if you think that headline inflation is a better measure of pricing pressures, there’s a strong case for raising interest rates now, a point that Bullard seems to be emphasizing.
On the other hand, the argument for being cautious on tightening if only headline inflation is bound up with the expectation that inflation will remain contained until (or if) core inflation starts to rise as well. For the moment, however, there’s still some debate if the latter condition applies. Indeed, the economy is still weak on a number of fronts, starting with the high jobless rate, and so the outlook for inflation is questionable.
Ultimately, the numbers will have the last word. Although the relatively low core readings provide a level of comfort that inflation isn’t poised to take off, there’s always the possibility that it could be different this time. To be fair, core readings are only one metric, and Bullard’s certainly right that we should be skeptical of any one data point. That’s still no excuse to ignore core. There was nothing in Bullard's speech that provides a smoking gun for abandoning a narrow view of inflation for short-term monetary policy adjustments.
Meanwhile, when we look to other measures of inflation risk, there’s additional evidence that prices aren't like to surge in the foreseeable future. The Treasury market’s inflation forecast, for instance, is now falling again. As of yesterday, the implied inflation forecast based on the yield spread between the nominal and inflation-indexed 10-year Treasuries was 2.32%, down about 30 basis points from a month ago and roughly equal to the market's prediction in February. The market can be wrong, of course, but you need a good reason to question the crowd's collective pricing.
It all boils down to whether you believe that food and energy prices will continue rising for years to come without a sustained downturn. That may be fate, and it’s a possibility that no one should dismiss lightly. But it’s a prediction. History suggests we should be cautious in extrapolating recent price trends as a fact for the long run future. If we haven’t learned that lesson at this late date, there’s not much hope for thinking clearly.
The next few months may prove decisive on deciding if core is misleading us. But for the moment, the jury’s out. Core inflation is hardly a silver bullet, but it’s debatable if it’s rotten.
Posted by jp at May 24, 2011 6:35 AM
I don't know about core inflation, but I think the Fed missed the whole run up in housing by measuring "equivalent rent" as the proxy for residential house prices. That measure is plain stupid.
Posted by: William McGurn at May 24, 2011 1:34 PM