Is inflation targeting really dead?

Harvard's Jeffrey Frankel (hat tip, Mark Thoma) is the latest econ-blogger to cast an admiring gaze in the direction of nominal gross domestic product (GDP) targeting. Frankel's post is titled "The Death of Inflation Targeting," and the demise apparently includes the notion of "flexible targeting." The obituary is somewhat ironic in that at least some of us believe that the U.S. central bank has recently taken a big step in the direction of institutionalizing flexible inflation targeting. Frankel, nonetheless, makes a case for nominal GDP targeting:

"One candidate to succeed IT [inflation targeting] as the preferred nominal monetary-policy anchor has lately received some enthusiastic support in the economic blogosphere: nominal GDP targeting. The idea is not new. It had been a candidate to succeed money-supply targeting in the 1980's, since it did not share the latter's vulnerability to so-called velocity shocks.

"Nominal GDP targeting was not adopted then, but now it is back. Its fans point out that, unlike IT, it would not cause excessive tightening in response to adverse supply shocks. Nominal GDP targeting stabilizes demand—the most that can be asked of monetary policy. An adverse supply shock is automatically divided equally between inflation and real GDP, which is pretty much what a central bank with discretion would do anyway."

That's certainly true, but a nominal GDP target is consistent with a stable inflation or price-level objective only if potential GDP growth is itself stable. Perhaps the argument is that plausible variations in potential GDP are not large enough or persistent enough to be of much concern. But that notion just begs the core question of whether the current output gap is big or small. At least for me, uncertainty about where GDP is relative to its potential remains the key to whether policy should be more or less aggressive.

In another recent blog item (also with a pointer from Mark Thoma), Simon Wren-Lewis offers the opinion that acknowledging uncertainty about size of the output gap actually argues in favor of being "less cautious" about taking an aggressive policy course. The basic idea is familiar. It is a simple matter to raise rates should the Fed overestimate the magnitude of the output gap. But with the short-term policy rates already at zero, it is not so easy to go in the opposite direction should we underestimate the gap.

No argument there. As I pointed out in a May 3 macroblog item, Atlanta Fed President Dennis Lockhart has said the same thing. But, as I argued in that post, this point of view is only half the story. Though I agree that the costs are asymmetric to the downside with respect to the FOMC's employment and growth mandate, they look to me to be asymmetric to the upside with respect to the price stability mandate. And I view with some suspicion the claim that we know how to easily manage policy that turns out to be too aggressive after the fact.

My issues are not merely academic. In an important paper published a decade ago, Anasthsios Orphanides made this assertion:

"Despite the best of intentions, the activist management of the economy during the 1960s and 1970s did not deliver the desired macroeconomic outcomes. Following a brief period of success in achieving reasonable price stability with full employment, starting with the end of 1965 and continuing through the 1970s, the small upward drift in prices that so concerned Burns several years earlier gave way to the Great Inflation. Amazingly, during much of this period, specifically from February 1970 to January 1977, Arthur Burns, who so opposed policies fostering inflation, served as Chairman of the Federal Reserve. How then is this macroeconomic policy failure to be explained? And how can such failures be avoided in the future?...

"The likely policy lapse leading to the Great Inflation …can be simply identified. It was due to the overconfidence with which policymakers believed they could ascertain in real-time the current state of the economy relative to its potential. The willingness to recognize the limitations of our knowledge and lower our stabilization objectives accordingly would be essential if we are to avert such policy disasters in the future."

With this historical observation in hand, it seems a short leap to turn Wren-Lewis's thought experiment on its head. Arguably, the last several years have demonstrated that nonconventional policy actions have been quite successful at short-circuiting the disinflationary spirals that pose the central downside risk when interest rates are near zero. (If you can tolerate a little math, a good exposition of both theory and evidence is provided by Roger Farmer.)

On the opposite side of the ledger, we know little about the conditions that would cause the Fed to lose credibility with respect to its commitment to its inflation goals, and very little about the triggers that would cause inflation expectations to become unanchored. Thus, I think it not difficult to construct a plausible argument about the risks of being wrong about the output gap that is exact opposite of the Wren-Lewis conclusion.

I end up about where I did in my previous post. Flexible inflation targeting, implemented in such a way that the 2 percent long-run inflation target rate exerts an observable gravitational pull over the medium term, feels about right to me. Despite what Frankel seems to believe, I think that idea is far from dead.

By Dave Altig, executive vice president and research director at the Atlanta Fed

Symmetric goals, asymmetric risks

Mark Thoma has been hanging out with my boss or at least was at the same conference, where Thoma had a chance to try out his reporter chops:

"I just got out of a press conference with Dennis Lockhart (Atlanta Fed president) and Charles Evans (Chicago Fed president). I can't say there was any real news, but I did manage to ask a question... I asked whether the 2% inflation target was truly symmetric."

Thoma got an answer, though seemingly not one that left him totally convinced:

"Both insisted that the target is symmetric. However, Lockhart said that we know much more about the effects of inflation than deflation, and that preventing deflation was therefore job number one (which doesn't really answer the question). He didn't explain what he is so afraid of if inflation goes up...

"Evans, while explicitly agreeing the target was symmetric, made comments that indicated that it may not be. He said the Fed has not done a very good job of communicating its tolerance around the 2 percent target, both up and down, and they need to improve. But if the target is really symmetric, simply saying that (along with the tolerable range) is all that is required. Talking separately about tolerance for over and under-shooting isn't needed."

Evans' and Lockhart's statements stand on their own, but we've collected some information via the Atlanta Fed's Business Inflation Expectations survey that helps me think about the Thoma question. The chart below plots the answers, collected in the February and April surveys, to the following query: "Projecting ahead, to the best of your ability, please assign a percent likelihood to the following changes to unit costs per year over the next five to 10 years."

The question focuses on unit labor costs in order to elicit responses about what businesses may actually be planning for, as opposed to their guesses about a more abstract concept of overall inflation. The question focuses on expectations five to ten years into the future because the inflation goal of the Federal Open Market Committee (FOMC) is explicitly a long-run objective.

The obvious pattern in these survey responses is their asymmetry to the upside. The most probable outcome, according to the respondents, is that long-term costs will rise in a range that includes the FOMC's long-run inflation objective. But they also put an almost 50 percent probability on annual outcomes higher than 3 percent. Less than 20 percent probability is placed on costs rising at rates of less than 1 percent.

This picture is one of asymmetric risks to the inflation outlook, and as such it is an important element in thinking through policy choices. Symmetry in the sense of having an equal distaste for misses on either side of an objective does not necessarily imply symmetry with respect to the risks of meeting that objective.

To begin with, the Fed does have a dual mandate. Disinflation or, in the extreme, deflation has the potential to be problematic for growth and employment when interest rates are very low. The reason, if you buy the analysis, is that, with no room for rates to move lower, a decline in inflation raises the real cost of borrowing. A higher cost of borrowing restrains spending, creating an additional drag on economic activity in already tough circumstances.

The reverse argument has, of course, been made for temporarily tolerating inflation that is somewhat higher than the long-run objective. But even if you aren't quite sold on the wisdom of that approach—and I'll get to that in a bit—it is clear that, with policy rates near zero, misses to the downside on inflation bring risks to the Fed's growth mandate that are not implied by misses to the upside. Hence President Lockhart's comment regarding the importance of preventing deflation.

So why not respond to this asymmetric risk to growth by taking a chance on higher inflation? That question takes us back to the chart above. Taken at face value, the probabilities reported by our survey respondents suggest that the FOMC has been pretty successful in convincing folks that very low rates of inflation or deflation will not be allowed to set in. Perhaps this conviction is not surprising given the relatively aggressive responses of the committee to the disinflation scares of 2003 and 2010.

But the response to asymmetric risks to growth at low inflation rates may have had the effect of inducing asymmetric risks to the upside with respect to Fed's price stability mandate. Again taken at face value, the results of our business inflation expectations survey definitely imply a one-sided bet by businesses on how the FOMC might miss on its inflation objective. That could well explain why one would be so concerned if inflation rises. Just as there are asymmetric risks associated with below-objective inflation when it comes to the Fed's growth and employment mandate, there are asymmetric risks associated with above-objective inflation when it comes to the price stability mandate.

By Dave Altig, executive vice president and research director at the Atlanta Fed

Taking a deeper dive into the definition of inflation

Sometimes our familiarity with a topic gets in the way of our understanding of it. I often think that about inflation. It's widely known that inflation is a condition of rising prices, or what my favorite macroeconomics text defines more precisely in this way: "Inflation is an increase in the overall level of prices in the economy." And while this idea is serviceable for some purposes, it's pretty inadequate if you try to think about how a central bank goes about the business of controlling inflation.

Here's an example. Friday's retail price report for March revealed that prices rose 3.5 percent from February and averaged 3.7 percent (annualized) over the first three months of the year. I think it's pretty clear we have seen an "increase in the overall level of prices in the economy" this year. But how should the Federal Reserve respond to this rise?

I'm not going to offer an answer that question. I'm a policy adviser, not a policy maker. But we need to dig a little deeper into the textbook to get a better understanding of the inflationary process and how a central bank contributes to that process before we offer up an opinion on the matter.

The inadequacy of the textbook definition of inflation is a topic that's been bugging me for a long time. It's silent about the cause of the rise in prices and therefore does not make any distinction about whether the price increase is a corollary of the policies of the central bank or from another source that a central bank can only nominally influence. This distinction is an important one. (For what it's worth, this 1997 article I wrote explains how I think the term "inflation" has come to be synonymous with "price increase." And here's why I think a deeper appreciation of what constitutes "an increase in the overall level of prices" affects how one thinks about the inflationary experience in real time.)

Here's how economist Irving Fisher put it in his 1913 article "The Monetary Side of The Cost of Living Problem":

"If it can be shown, for instance, that today the good things of this world are becoming scarce on the one hand while money and its substitutes are not becoming plentiful it would be reasonable to conclude that the fault lies with goods and not with money. If, on the contrary, it can be shown that money and its substitutes are becoming plentiful and that goods are not becoming scarce, it is reasonable to conclude that the fault lies chiefly on the monetary side."

Are all price increases inflation? Should central banks use their tools to fight against any rise in costs? The answers to these questions have important implications for how we measure inflation in the short run versus the long run, the horizon over which a central bank ought to be held accountable for achieving price stability, and how monetary policies should be calibrated in the face of rising prices.

In that spirit, the Atlanta Fed has developed an animated video that provides additional perspective on the issue of inflation—specifically, what sorts of price increases are part of the inflationary process that is under the control of the central bank, and which are not. This video is the first in a new series that covers economic issues and the work of the Fed. The video is part of the Atlanta Fed's new feature called "The Fed Explained," which includes a range of new and existing content. We hope these videos stimulate conversation on a range of topics important to the Federal Reserve. The format is targeted to the general public as well as to students and teachers, and we hope it provides an engaging supplement to the study of the Federal Reserve.

Let us know.

Mike Bryan, vice president and senior economist at the Atlanta Fed

Why we debate

It's been a while since we featured one of my favorite charts—a "bubble graph" comparing average monthly job changes during this recovery with average changes during the previous recovery, sector by sector.

(enlarge)

If you try, it isn't too hard to see in this chart a picture of a labor market that is very close to "normalized," excepting a few sectors that are experiencing longer-term structural issues. First, most sectors—that is, most of the bubbles in the chart—lie above the horizontal zero axis, meaning that they are now in positive growth territory for this recovery. Second, most sector bubbles are aligning along the 45-degree line, meaning jobs in these areas are expanding (or in the case of the information sector, contracting) at about the same pace as they were before the "Great Recession." Third, the exceptions are exactly what we would expect—employment in the construction, financial activities, and government sectors continues to fall, and the manufacturing sector (a job-shedder for quite some time) is growing slightly.

For the skeptics, I below offer a familiar chart, which traces the level of total employment pre- and post-December 2009, compared with the average path of pre- and post-recession employment for the previous five downturns:

(enlarge)

We are now more than 16 quarters past the beginning of the recession that began in the fourth quarter of 2007, and total employment is still 4 percent lower than it was at the beginning of the downturn. In the previous five recessions by the time 16 quarters had passed, employment had increased by about 6 percent. Even in the worst case, indicated by the lower edge of the gray shaded area, employment growth was flat—and that observation is qualified by the fact that the recovery from the 1980 recession was interrupted by the 1981–82 recession.

This unhappy comparison is not driven by the construction, financial activities, and government sectors. In the area of professional and business services, which has logged the largest average monthly employment gains in the current recovery, the number of jobs still sits 2.7 percent below the level at the outset of the last recession, as the chart below shows.

(enlarge)

Total private-sector jobs in education and health services, which never actually contracted during the recession, nonetheless remain abnormally low in historical context.

(enlarge)

In these charts lies the crux of some very basic disagreements about the appropriate course of policy. The last three graphs draw a clear picture of labor markets that are underperforming by historical standards—a position that I take to be the conventional wisdom. An argument against following that conventional wisdom centers on the question of whether historical standards represent the appropriate yardstick today. In other words, is the correct reference point the level of employment or the pace of improvement in the labor market from a permanently lower level? For the proponents of the latter view, the bubble chart might very well look like a return to normal, despite the fact that employment has not returned to prerecession levels.

One way to adjudicate the debate, in theory, is to rely on the trajectory of inflation. If there remains a significant amount of slack in labor markets, as the conventional interpretation of things suggest, there ought to be consistent downward pressure on prices. The case for consistent downward pressure on prices is not so obvious. Measured inflation appears to moving in the direction of the Federal Open Market Committee's 2 percent long-term objective.

(enlarge)

Also, the Atlanta Fed's own monthly survey of business inflation expectations, which surveys a panel of businesses from our Reserve Bank district, indicates that this inflation number (shown in our March release from earlier this week) is in line with what private-sector decision makers anticipate:

"Survey respondents indicated that, on average, they expect unit costs to rise 2.0 percent over the next 12 months. That number is up from 1.9 percent in February and comparable to recent year-ahead inflation forecasts of private economists. Firms also reported that their unit costs had risen 1.8 percent compared to this time last year, which is unchanged from their assessment in February. Inflation uncertainty, as measured by the average respondent's variance, declined from 2.8 percent in February to 2.4 percent in March, the lowest variance since the survey was launched in October 2011."

Does that settle it? Not quite. There may not be much evidence of building disinflationary pressure, but neither is there building evidence of an inflationary push that you would expect to see if the economy were bumping up against capacity constraints. Obviously, the story isn't over yet.

By Dave Altig, senior vice president and research director at the Atlanta Fed

 

Weighing the risks to the inflation outlook: Two views

The Federal Reserve Bank of Atlanta's Survey of Business Inflation Expectations released earlier today showed a continuation of rather modest expectations for unit cost pressures over the coming 12 months. In February, our panel of firms reported a 1.9 percent average expected rise in unit costs over the coming year, still within the very narrow 1.8 percent to 2 percent range the group has been reporting over the past five months.

That's the good news. Now for some (potentially) bad news. In a special question this month, we asked the panel to weigh in on their expectations for annual unit cost increases over the longer term—specifically, the next 5 to 10 years. The group's expectation was a percentage point higher, at 2.9 percent.

The reason for the higher expectation for unit costs over the longer term can be seen in the following chart, which compares how the group assigns probabilities to unit cost changes over the next 12 months to how they judge these probabilities over the longer term.

In both instances, the Atlanta Fed's Business Inflation Expectations panel of firms puts the greatest likelihood that unit costs will rise in the 1 percent to 3 percent range—in a range that matches the Federal Open Market Committee's longer-term inflation objective.

But how does the group assess the risks around that increase? Over the short term, the panel sees a higher likelihood that unit costs may fall short of the 1 percent to 3 percent range. Specifically, the group sees a 36 percent chance that unit costs will rise less than 1 percent compared against only a 26 percent chance that they will rise above 3 percent. Yet when sizing up the next 5 to 10 years, the group sees only a 15 percent chance that unit costs will rise less than 1 percent per year compared with a 46 percent chance that costs will rise by more than 3 percent.

What our panel of firms appears to be telling us is that the risks to the inflation outlook—in both the near term and longer term—aren't particularly balanced. In the near term, they weigh the inflation risks more heavily to the downside. But looking over the next 5 to 10 years, the panel sees the inflation risks leaning decidedly to the upside.

What we can't tell from these data is whether the panel's assessment of the inflation risks is different today than it was before. After all, this is the first time we've asked the question, but you can bet it won't be the last.

Mike Bryan, vice president and senior economist,

Laurel Graefe, economic policy analysis specialist, and

Nicholas Parker, economic research analyst, all with the Atlanta Fed

How are we doing?

Near the beginning of the minutes of the January meeting of the Federal Open Market Committee (FOMC), released yesterday, you'll find a reiteration of the FOMC's historic decision explicitly endorsing a numerical definition of long-run price stability:

"The inflation rate over the longer run is primarily determined by monetary policy, and hence the Committee has the ability to specify a longer-run goal for inflation. The Committee judges that inflation at the rate of 2 percent, as measured by the annual change in the price index for personal consumption expenditures, is most consistent over the longer run with the Federal Reserve's statutory mandate."

The minutes include the motivating force behind this decision:

"The Chairman noted that the proposed statement did not represent a change in the Committee's policy approach. Instead, the statement was intended to help enhance the transparency, accountability, and effectiveness of monetary policy."

In a speech given Tuesday at New College in Sarasota, Fla., our local participant in this decision—Atlanta Fed President Dennis Lockhart—provided his interpretation of this numerical inflation objective:

"The 2 percent inflation target is an aid to understanding how the FOMC will react to developments in the economy within an overarching approach that can be called 'flexible inflation targeting.'

"The word 'flexible' describes and qualifies the committee's exercise of judgment in reaction to adverse developments. The word 'flexible' also reflects the principle that it is not always feasible or desirable to hit the target in the short run. Short-lived shocks to the economy can temporarily move measured inflation well away from the 2 percent target."

The thinking behind that statement can be clearly seen in the following chart, which illustrates the volatility of annualized inflation rates as the horizon extends from one to five years:

(enlarge)

As the chart shows, volatility noticeably declines as the horizon extends beyond one year to two years, and a similar decline occurs as we move from a two- to five-year horizon. This picture is exactly the type you would expect if inflation were subject to temporary ups and downs that dissipate over time. In his speech, President Lockhart offered his thoughts on the policy meaning of an inflation process that has this characteristic:

"Consider last year's energy and commodity price increases. Those cost pressures pushed up inflation in the early part of the year. Then, as expected, their influence dissipated as the year progressed.

"Had the FOMC tightened monetary policy early last year in response to the inflation threat, we might have compromised progress on growth and employment to no particular benefit with respect to our inflation mandate…

"As I see it, this is a recent, real-world example of a balanced approach in action. It illustrates the idea of flexible inflation targeting."

Of course, that particular example might ring somewhat hollow if the record suggested that the FOMC just got lucky this time around. A criticism that emerged in the aftermath of the inflation target announcement was not so much that it was flexible per se, but that in its focus on an undefined long-run, it is essentially an empty commitment. That opinion was offered in a Financial Times article penned by Lorenzo Bini Smaghi:

"The first [question about the FOMC's definition of price stability] relates to the time horizon over which the Fed is supposed to achieve price stability, namely the long-run. This differs from most other central banks in advanced economies, where price stability is targeted over a horizon of two to three years…

"Monetary policy produces its effects with lags of one to three years. This is the period over which the central bank should be held accountable."

That thought was echoed on The Economist's Free Exchange blog:

"According to the Fed's projections, it hits its target—2% inflation—over the long term. Mr Bini Smaghi's point is that it doesn't make much sense to judge current Fed actions against a long-run inflation projection.

"…the Fed doesn't necessarily run into problems of inconsistency if it projects inflation above 2% 1 or 2 years from now—a timeframe over which markets readily understand this group of policymakers to have control—while maintaining the long-run 2% goal."

The second part of The Economist comments move the conversation to an operational middle ground between an inflexible commitment to a target in the very short run and a promise that provides little discipline because its attainment remains out in a perpetually undefined future.

In particular, think about monitoring policy performance against a stated inflation objective over some "medium-term" horizon. "Medium-term" is itself a term of art, but I find it attractive to think about a three- to five-year horizon. Given the continuous arrival of shocks to the economy, uncertainties about the timing of policy effects, and the desirability of trading off precise control over inflation against the risks of destabilizing influences on real economic growth, I think it is still unrealistic and unwise to expect that an inflation target will be hit precisely even over a medium-run horizon. This is the reason, I believe, that an exact point target for inflation is relegated to the long run. But I think it is realistic, and wise, to expect realized average inflation to fall within a reasonable tolerance range about a long-run target over something like a three-to-five-year medium-term horizon.

People can disagree about what constitutes a reasonable tolerance range, but one option that I find sensible would be along the lines of the average volatility of medium-term inflation (calculated over a period in which inflation outcomes were deemed to be acceptable, which I've chosen to be the period since the mid-1990s). With this in mind, the following chart plots realized inflation over three-, four-, and five-year horizons. (For reference, the chart highlights the 2 percent target with upper and lower limits that are plus and minus 1 percentage point.)

(enlarge)

The plus or minus 1 percentage point threshold in the above graph is somewhat above the standard deviation of medium-term outcomes shown in my earlier chart, so one might want to tighten up the bounds. But if you are willing to accept that it's close to your definition of tolerable deviation, the record does support the position that, over the past two decades or so, the Fed has delivered on the its now-explicit long-term objective, or taken sufficient to steps to correct matters when it wasn't.

Looking forward, if the midpoint of FOMC participants' most recent inflation projections comes to pass, the four- to five-year averages would remain near the long-run objective, with the three-year average moving away from its recent flirtation with the lower end of my hypothetical tolerance range.

I'm not saying that the above chart alone defines "appropriate policy"; performance against the other half of the dual mandate is obviously relevant. But I think it provides at least one way to think about what success looks like, and a sensible metric for whether the Fed is delivering on its long-run promise.

By Dave Altig, senior vice president and research director at the Atlanta Fed

 

Keeping an eye on inflation

Where's inflation heading? Well, here's what the minutes of the December meeting of the Federal Open Market Committee (FOMC) had to say on the subject:

"Participants observed that inflation had moderated in recent months as the effects of the earlier run-up in commodity prices subsided . . . many participants judged that the moderate expansion in economic activity that they were projecting . . . would be consistent with subdued inflation going forward."

But not all FOMC meeting participants viewed these trends with equanimity:

"Indeed, some expressed the concern that, with the persistence of considerable resource slack, inflation might run below mandate consistent levels for some time."

According to Reuters, San Francisco Fed President John Williams said it this way:

"The data so far on the inflation front are confirming my view that inflation is ebbing and moving to be too low, and that is an important driver of my thinking about policy."

But as you might expect, some see the inflation risks weighing a bit on the other side of the scale. Again, from the December FOMC meeting minutes:

"Some participants were concerned that inflation could rise as the recovery continued . . . A few participants argued that maintaining a highly accommodative stance of monetary policy over the medium run would erode the stability of inflation expectations."

In fact, Philadelphia Fed President Charles Plosser had this to say in a speech earlier this week:

"I do anticipate that with many commodity prices now leveling off or falling, and inflation expectations relatively stable, inflation will moderate in the near term . . .

"But as a policymaker, my focus is less on the near term and more on the medium term. Looking further ahead, I believe we must monitor the inflation situation very carefully, particularly in this environment of very accommodative monetary policy. Inflation most often develops gradually, and if monetary policy waits too long to respond, it can be very costly to correct. Measures of slack such as the unemployment rate are often thought to prevent inflation from rising. But that did not turn out to be true in the 1970s. Thus, we need to proceed with caution as to the degree of monetary accommodation we supply to the economy."

What doesn't seem to be in dispute is that monitoring the data for any sign that the inflation trend is shifting—either higher or lower—is probably a good idea. And there are a lot of data to watch. In a speech last year to the Calhoun County Chamber of Commerce, Atlanta Fed President Dennis Lockhart had this to say about reading the inflation data:

"To achieve price stability, policymakers must detect inflation in its early stages before it is firmly established, especially in the psychology of consumers and businesses. This early detection is a challenge because inflation is not easily measured in the short term with any precision. No single price statistic enjoys a sufficient vantage point from which to assess inflation in the short term. With imperfect tools, inflation is more easily monitored than precisely measured."

The research department of the Federal Reserve Bank of Atlanta has taken pretty seriously the task of monitoring inflation developments. Where there are gaps in our information, we've been working to fill them with data, and we've aggregated it all into one place: the Inflation Project web page.

On the Inflation Project, we now report a sticky-price CPI statistic calculated from consumer price index data using only those components whose prices are slow to change. Joint research with the Cleveland Fed has shown this measure to be helpful when thinking about inflation expectations. Using Treasury Inflation-Protected Securities data, we now produce a weekly measure of the probability of a sustained deflation. And come January 27, we'll begin reporting the results of a monthly survey of business inflation expectations that examines firms' price-setting environment and the pricing pressures they face. From the responses, we'll generate a monthly measure of respondents' year-ahead unit cost expectations.

But of course, there are already a lot of data to keep an eye on. To make it a little easier to gain some perspective, we're also unveiling our inflation dashboard. The dashboard provides a platform for visualizing some of the data we commonly monitor to keep abreast of emerging inflation developments. It tracks 30 data series grouped into six major categories—retail prices, inflation expectations, labor costs, producer prices, material and commodity costs, and money and credit.

Our data and the inflation dashboard are available on the Inflation Project web page. Let us know what you think.

Mike Bryan, vice president and senior economist,

Laurel Graefe, economic policy analysis specialist, and

Nicholas Parker, economic research analyst, all with the Atlanta Fed

Core cuts both ways

With the six-month average of annualized headline inflation running just over 5 percent, this Wednesday's consumer price index (CPI) report looms a little larger than usual. While it is dangerous to predict such things, there is every reason to believe that the measured increase in CPI inflation for May could be quite low. And there is every reason to believe that this softness will persist into June.

The reason is quite simple. Movements in gasoline and fuel prices really do push around the headline inflation number, and at long last it looks like that movement is in the downward direction. Here's the relevant picture:

Let's assume that, annualized, CPI excluding food and energy rises 2.1 percent, as it has so far this year, and food and nongasoline energy prices each rise at 5 percent. Then when you plug in gasoline prices already realized for May and EIA predictions for June, you get a 0.4 percent rise in the headline CPI in May and a 0.7 percent decline in June (both rates annualized).

Despite some probable relief on the headline inflation number, I remain aware of what that relief means and what it does not. Earlier in the year, Atlanta Fed president Dennis Lockhart had this to say:

"I want to contrast inflation to the cost of living. In casual language, we often interpret a rise in the cost of living as inflation. They are not the same thing. Cost-of-living increases are a result of increases in individual prices relative to other prices and especially relative to income. These relative price movements reflect supply and demand conditions and idiosyncratic influences in the various markets for goods and services…

"… The Fed, like every other central bank, is powerless to prevent fluctuations in the cost of living and increases of individual prices. We do not produce oil. Nor do we grow food or provide health care. We cannot prevent the next oil shock, or drought, or a strike somewhere—events that cause prices of certain goods to rise and change your cost of living."

Two points, then. First, even if things evolve as the chart above suggests, the level of gasoline prices will remain relatively high by recent standards. Low inflation readings for the next couple of months would therefore leave the cost-of-living high by recent standards, a fact that is not lost on us here at the Atlanta Fed.

Second, as President Lockhart's comments reveal, we were reluctant to attribute the run-up in fuel prices to monetary policy. And I imagine we will be equally reluctant to credit monetary policy with any relief from that trend.

In fact, I will fearlessly predict that, should our guess here about headline inflation in the next couple of months be proved accurate, we will point to core inflation measures as reason to look through some very low inflation readings. See these comments from President Lockhart's most recent speech for some additional perspective:

"Are the recent outsized increases in headline inflation the best signals of the inflation trend going forward? Or are other statistics—like core inflation or measures of inflation expectations—yielding a truer picture of what lies ahead?

"The answer matters a lot. And it certainly weighs heavily on my thinking. I would not hesitate to support an exit from our current policy stance if I believed that the headline inflation number of the past six months is really indicative of the underlying trend inflation rate. I don't believe this to be the case. And I am wary of tightening monetary policy in the face of quite ambiguous economic circumstances unless doing so is absolutely necessary to meet the FOMC's price stability mandate."

And here's why it matters, quoting President Lockhart from an interview with Reuters last week:

"In the interview, Lockhart said maintaining an easy Fed policy stance should ensure the moderate U.S. recovery does not fall off the rails."

But…

"How high would the bar be for further Fed easing?

"It would take 'a significant deterioration as reflected in the overall economy, a set of deflationary signals and also unemployment numbers that rise dramatically. Last fall (when the Fed launched QE2), by some measures, we were seeing declining inflation expectations that were headed in a pretty negative direction, and that was happening pretty rapidly. We were in a disinflationary environment. So the risk of deflation was plausible. We acted and the situation has turned around. That was, at least in my way of thinking, very central to supporting the policy. We don't have anything remotely like a deflationary risk at the moment, short of a shock of some kind.' "

If what I suggest above—falling headline inflation, stable core inflation—comes to pass in the near term, don't expect me to start ringing the disinflationary bell.

This isn't the last word on the usefulness of core inflation statistics, and the debate is certain to rage on (see here and here, for recent installments). But I do hope that this post is remembered the next time the Federal Reserve is accused of hiding inflationary pressures behind the rhetoric of core.

Update: The May CPI report is in, and once again the facts trump arithmetic. We got the flip in the headline/core measures right; the rest, not so much.

Update: Here's one way to construct a synthetic CPI:

CPIHeadline(T) is the calculated value of the seasonally adjusted CPI in month T > December 2010. CPIHeadline(Dec2010) is the reported seasonally adjusted headline CPI for December 2010. CPIFood, CPICore, CPIGas, CPIOtherFuel, and CPIHHenergy are the seasonally adjusted component level CPI indexes for the above mentioned components. Note that this formula will not exactly replicate the January through April 2011 headline CPI since the components are not granular enough. When using the above formula to compute CPIHeadline(Apr11) and CPIHeadline(May11), we compute one-month headline inflation in May 2011 as the percent difference between the values of CPIHeadline(Apr11) and CPIHeadline(May11), both calculated with the above formula (i.e., do not use the reported value of CPIHeadline(Apr11)). In our calculations we used the same inflation assumptions for "other motor fuels" and "household energy" (both increase at a 5 percent annualized rate in May and June). By Patrick Higgins, an economist at the Atlanta Fed

By Dave Altig, senior vice president and research director at the Atlanta Fed

Should we even read the monthly inflation report? Maybe not. Then again...

In a recent issue of Economic Synopsis, our colleague Dan Thornton of the St. Louis Fed questions the usefulness of the traditional core inflation statistics—the consumer price index (CPI), or the personal consumption expenditure price index that strips out food and energy costs. Specifically, Dan asks whether the core inflation statistic is a better predictor of future inflation over the medium term (say, the next two or three years), than the headline inflation statistic. His conclusion is that:

"[F]or the most recent period, there is no compelling evidence that core inflation is a better predictor of future headline inflation over the medium term."

But Dan also invites the following:

"[I]n the interest of greater transparency and to allow the public to better understand its focus on core measures, the FOMC [Federal Open Market Committee] should provide evidence of the superior forecasting performance of the core measure it uses."

Well, of course neither writer of this blog post is on the FOMC, and equally obvious is the fact that we don't speak for anyone who is. Moreover, we're not very big fans of the traditional core measures, and we much prefer trimmed-mean estimators of inflation when thinking about recent price behavior.

Nevertheless, we'd like to attempt an answer to Dan's call, even if it wasn't aimed at us.

Here's the experiment run by Dan: He used the past 36-month trend in the traditional core inflation measure and the ordinary headline inflation measure and tested which one most accurately predicted the next 36 months of headline inflation. He found that they're about the same. A similar look at 24-month trends yielded a similar result.

The upshot of these experiments can be seen in the figure below (which is a figure of our construction, not his).

(enlarge)

The chart shows how accurately we can predict headline CPI inflation over the next three years using only headline CPI price data or, alternatively, using only core CPI price data. The essence of the conclusion reached in the Economic Synopsis is summarized within the shaded box. The forecast accuracy of the two- and three-year trends of the core CPI price measure doesn't seem to be a significant improvement to the plain-vanilla headline CPI.

But we wonder whether the contribution of the core inflation statistic is being accurately reflected in this experiment. For us, the power of a core inflation measure—whether it be the traditional ex-food and energy measure, or some more statistical construct like the trimmed-mean estimators—can't be seen by comparing data trends of this sort. The volatility of an inflation statistic, what we would characterize as "noise," dissipates rather quickly, generally within a few months (although for food and energy, it could play out over a longer period of time, we understand).

At issue is how much the most recent month's or quarter's inflation data should inform one's thinking about the future path of inflation. Implicit in the experiments reported above is that they shouldn't—well, only as much as the most recent monthly or quarterly data influence the trend of the past two or three years.

It may be that the most recent monthly or even quarterly data are so noisy that they have nothing useful to contribute to our perception of the future inflation trend. But then again, an experiment that assumes there is no useful information in the most recent inflation data does not necessarily make it so.

We'd like to call your attention to the remainder of the figure above, where we ask the question, what happens if you try to predict headline CPI inflation over the next three years using only the most recent price data? For example, what if we restrict ourselves to looking only at the most recent month's CPI report? What we see is that the core inflation statistic provides a much improved prediction of the future inflation trend compared to the headline measure. Specifically, forecast accuracy is improved by nearly 50 percent if you use the core inflation measure. (For you wonks, the root mean square error, or RMSE, of the core CPI prediction is about 1.4 percent, compared with a RMSE of 2.7 percent for headline CPI inflation.)

Now consider the behavior of CPI prices over the past three months. How informative of the future inflation trend are these prices? Well, the accuracy of the headline inflation statistic improves relative to the one-month percent change because averaging the data over time in this way necessarily reduces the transitory fluctuations in the data. But again, the three-month core CPI price statistic provides a much better prediction of future headline inflation than does the three-month trend in the ordinary CPI statistic. In other words, if you're wondering what the past-three months of data tell you about developing inflation pressure, you're much better off considering the core statistic than you are the headline number.

Here's another observation we'd like to make: The most recent three-month trend in the core CPI inflation measure appears to be a more accurate predictor of future inflation than the 12-month headline CPI trend. Moreover, the three-month trend in the core measure is roughly as accurate as its longer-term trends. This observation suggests that paying attention to the core measure may allow you to spot changes in the inflation trend much more quickly than using headline alone.

Again, to be clear, we aren't endorsing the core inflation statistic. We're fans of trimmed-mean estimators and think they do an even better job of informing thinking about what the most recent price data tell us about the likely future path of inflation. (As evidence, we included in the chart above the same forecasting results for the median CPI.) We only want to make one simple point—the usefulness of a core inflation measure is best seen in the monthly and quarterly intervals that span FOMC meetings, not in the two- or three-year trends which are, by construction, largely silent about the most recent data.

By Mike Bryan, a vice president in research at the Federal Reserve Bank of Atlanta, and Brent Meyer, a senior economic analyst at the Federal Reserve Bank of Cleveland

 

Monetary policy exit: Is a bad bank the solution?

It is not difficult to find people who espouse the following belief:

"One of the Fed's recent errors was increasing the money supply by buying more than $1 trillion of mortgage-backed securities as part of its 'quantitative easing' policy. Its hefty balance sheet now threatens to finance further inflationary increases in the money supply. How can it be unwound in an orderly way?"

I dare say that I hear that criticism, in one form or another, nearly every day. So the claim above might only merit notice here because it comes from a Wall Street Journal op-ed penned by Professor Allan Meltzer, the eminent monetary macroeconomist and chronicler of Federal Reserve history. But what really distinguishes the critique is that it comes with a fairly novel and creative way of resolving the perceived problem:

"One idea is for the Fed to create its own version of a 'bad bank.' The Fed should promptly put the $180 billion of its long-term government debt and more than $1 trillion of its mortgage-backed securities into a separate entity. The long-term government debt and mortgage-backed securities would be the new bank’s assets. (The $1 trillion in Fed-created 'excess' bank reserves as a result of quantitative easing would become the liabilities of the bad bank.)

"The Fed would make a commitment not to sell any of the bad bank's mortgage-backed securities and Treasurys until they mature. Almost half of the Fed's currently held assets, more than $1 trillion, have 10 or more years until maturity, so all of them would be off the table as far as financing inflation during the gradual economic recovery. As the mortgages mature and are paid off, the bad bank's assets decline. The reduction in the bad bank's assets means that its liabilities, the excess reserves, would also decline—though that would be years away. Letting the market know precisely when the mortgage-backed securities would be sold makes the adjustment to the future elimination of excess reserves manageable."

Generally speaking, the Meltzer strategy offers what I perceive to be two critical criteria for a viable exit plan. One is that the winding down of the mortgage-backed securities (MBS) and long-term Treasury securities on the Fed's balance sheet should be conducted in a way that avoids market disruption and distortion as much as possible. The second is, of course, that the excess reserves held in the banking system—the liability side of the Federal Reserve’s balance sheet—have to be removed or "locked up" as needed to avoid an inflationary expansion of broad money and credit.

It should not go unmentioned that these criteria are also features of exit plans that have been sketched by Federal Reserve officials—by Chairman Bernanke last year and by Philadelphia Fed President Charles Plosser more recently, for example. What distinguishes the Meltzer plan is less in approach and more in timing.

In fact, the Meltzer approach is similar in spirit to the Fed's Term Deposit Facility that became operational last summer. The simplified description of this facility is that it involves the Federal Reserve temporarily taking onto its own accounts the excess reserves of banks. If you want to call that account a "bad bank," you're getting close to the Meltzer plan.

Expanding just a bit, in a term deposit facility the private banking system deposits its excess reserves—an asset of private banks—with the Fed. In exchange, banks receive an asset in the form of interest-bearing "accounts," the analog of a time deposit that you might purchase from your bank. Just as you cannot spend the funds you put into a time deposit you own, banks cannot use funds deposited in the Term Deposit Facility to support credit expansion (at least not until the term of the deposit expires).

Which brings me to a point that I don't quite follow about the Meltzer plan: If reserve assets are removed from the banking system, what are the corresponding offsets on the balance sheets of private banks? My guess is you end up with something like term deposits or their economic equivalent—nonnegotiable sterilization bonds, for instance. And if you match the maturities of those deposits with the maturities of the MBS and long-term security portfolio, it becomes pretty clear that the debate is really less about tactics and more about some pretty familiar, but difficult, issues: When is it time to stand pat on policy, and when is it time to reverse course?

On this conversation, I will, as I often do, give the last word to Dennis Lockhart, our Bank's president, who spoke earlier today in Knoxville, Tenn.:

"My view of the future permits a degree of patience as regards monetary policy. There is still a halting and fragile quality to the economy. I think the process of restoration of full economic strength with higher employment continues to require support. That said, planning for an eventual change of course is completely appropriate as long as public discussion about planning deliberations and the plan itself don't create premature expectations of tightening."

By Dave Altig
Senior vice president and research director at the Atlanta Fed