The Atlanta Fed's macroblog provides commentary and analysis on economic topics including monetary policy, macroeconomic developments, inflation, labor economics, and financial issues.

Authors for macroblog are Dave Altig, John Robertson, and other Atlanta Fed economists and researchers.

« January 2012 | Main | March 2012 »

February 22, 2012

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.

Expected change in unit costs over next 12 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.

Distribution of respondent expectations for unit costs over next 12 months and next 5 to 10 years

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 Mike Bryan, vice president and senior economist,

Laurel Graefe Laurel Graefe, economic policy analysis specialist, and

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

February 22, 2012 in Business Inflation Expectations, Data Releases, Inflation | Permalink


TrackBack URL for this entry:

Listed below are links to blogs that reference Weighing the risks to the inflation outlook: Two views:


First of all, I came across this blog and had to blink a couple of times. Had no idea that any entity within the FED system actually had its own blog. What a concept - excellent from a transparency perspective. As a Brit, I wish the BOE would do something similar. It sounds like overall the worries are deflation in the short-term, inflation longer-term. As a non-economist, non-central banker, I wonder if the 5-10 year higher inflation estimate has to do with a worry that some of the money from QE might leak out into the economy, or is it something different? Probably a stupid question, but every time one reads about QE one sees these overwrought articles that inevitably talk about "hyperinflation" or the like, which I do know enough of to understand this is stupid. BTW, for anyone interested, the Daily Telegraph notes that the BOE now owns ONE THIRD of all Gilts outstanding - is this a normal situation for a central bank?

Posted by: investment in farmland | February 22, 2012 at 04:12 PM

is it obvious that a realized uptick in ULCs would be associated with higher inflation rather than lower profits?


Posted by: john | February 23, 2012 at 08:54 AM

How valid are these panel forecasts? How much history do you have of them?

Posted by: GregL | February 26, 2012 at 08:14 AM

Post a comment

Comments are moderated and will not appear until the moderator has approved them.

If you have a TypeKey or TypePad account, please Sign in

February 16, 2012

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:

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.)

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.

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


February 16, 2012 in Federal Reserve and Monetary Policy, Forecasts, Inflation, Monetary Policy | Permalink


TrackBack URL for this entry:

Listed below are links to blogs that reference How are we doing?:


I'm just wondering, how will inflation react to our overall economy with the recent 9 month high increase of crude prices? Just a thought.

Posted by: Johnny | February 21, 2012 at 02:57 PM

Post a comment

Comments are moderated and will not appear until the moderator has approved them.

If you have a TypeKey or TypePad account, please Sign in

February 10, 2012

Reading the bump in inventories

Yesterday's wholesale trade report, with its positive surprise in December inventory accumulation, has estimates of fourth quarter gross domestic product (GDP) on the rise again. For the advance GDP release, the U.S. Bureau of Economic Analysis assumed that the book value of merchant wholesale inventories rose by $17 billion (at a seasonally adjusted annual rate, or SAAR) in December. The wholesale trade report suggests the book value instead may have risen by $56 billion SAAR. Our own calculations suggest fourth quarter GDP may be revised up from 2.8 percent to around 3.1 percent. A piece of that revision comes from positive sales activity, which would appear to be an unambiguous plus.

The inventory piece is trickier. Forecasters have a tendency—because the statistics have a tendency—to take a larger-than-expected inventory buildup in one quarter out of growth estimates for the next quarter. The implication in present tense is, of course, that 2012 may start out on the slow side as the fourth quarter inventory swell is run off.

That's not how we see it. Our current read is that it is better to think of the fourth quarter inventory buildup as a payback from a decumulation in the third quarter. Here's a look at overall inventory changes over the recent past, broken down into their various industrial components:

If you look hard, you will see that, though the fourth quarter inventory rise was broad-based, the third to fourth quarter change in wholesale inventories was particularly notable. In fact, the wholesale inventory picture in the back half of 2011 was dominated by a fairly large decumulation of nondurable goods inventories in the third quarter, a decline that was reversed in the last three months of the year:

In the background of those details are some pretty nonthreatening-looking inventory-sales ratios:

So, consider two stories that might frame thinking about the role of inventories in GDP growth in the first quarter or first half of this year. One story is inventory-inflated growth in the fourth quarter of 2011, to be followed by payback in the form of a drag on production in the first quarter (or so) of 2012. Another story is that the drag actually emerged in the third quarter of last year, providing a little extra juice in the fourth quarter, with no particular consequences for the current-year growth trajectory.

Right now, it looks to us like the latter story might be the right one. Of course, that doesn't mean there aren't significant risks to the outlook for domestic production, and hence inventories. For instance, although today's report on international trade in December was relatively benign in terms of fourth quarter GDP revisions, it did show a substantial further weakening in exports to the euro zone. Weaker demand from Europe will weigh on U.S. export growth. The big unknown is how weak that demand will get.

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


February 10, 2012 in Data Releases, Economic Growth and Development, Forecasts | Permalink


TrackBack URL for this entry:

Listed below are links to blogs that reference Reading the bump in inventories:


Post a comment

Comments are moderated and will not appear until the moderator has approved them.

If you have a TypeKey or TypePad account, please Sign in

February 06, 2012

Employment: Some good news, some bad news

Comparisons can be useful in determining where the economy is at any given point in time, and today's Employment Situation report from the U.S. Bureau of Labor Statistics provides another opportunity to do just that. According to that report, the U.S. economy added 243,000 nonfarm payroll jobs in January 2012. But total nonfarm employment is still 5.6 million lower than at the start of the last recession (December 2007).

For additional comparisons, more than 1 million fewer people filed initial unemployment insurance claims during the last week of January 2012 than during the last week of January 2009 (at the height of the recession). However, 55,000 more people filed initial claims during the last week of January 2012 than during the last week of January 2007 (before the recession started).

When examining layoffs, more than 400,000 fewer workers were laid off or discharged during November 2011 (according to the most recent data) than during the height of the recession in November 2008. Nonetheless, there were roughly a million fewer job openings and hires than during November 2007 (before the recession started).

Nominal average hourly earnings of wage and salary workers were about two dollars (9.6 percent) higher in January 2012 than around the start of the recession (January 2008). Nonetheless, real average hourly earnings (controlling for inflation) were only eight cents (0.8 percent) higher over the same time period.

Considering everything, there is both good news and bad news in the labor market today.

The median three-digit NAICS (North American Industry Classification System) industry lost 7 percent of its jobs during the most recent recession. In other words, half of the industries lost less than 7 percent of their jobs and half of the industries lost more than 7 percent of their jobs. Industries faring the worst (those in the 75th percentile of job losses) shed 13 percent of their jobs. And what might be considered "fortunate" industries (those in the 25th percentile of job losses) saw only 3 percent of their jobs disappear over this time period.

Chart 1 compares the year-over-year employment growth among industries that experienced below-median and above-median job loss, as well as those industries in the 75th and 25th percentiles of job loss. That chart also shows another potential bit of good news—that is, in spite of the dramatic differences in job losses, all four categories of industries are currently adding jobs at about the same rate. And that overall job growth, at about 0.24 percent per month, is roughly the same as the average monthly job growth seen between 1993 and 2000 and exceeds the average monthly growth before the recession, from 2004 through 2007.

Still, there is more bad news. At the current rate of growth, those industries that experienced above-median job loss during the recession will not regain prerecession employment levels until the end of 2015.

Chart 2 illustrates the employment level for those industries with above median job losses along with projections of employment based on three assumptions of monthly employment growth. To even recover before the end of 2013 the jobs that they lost during the recession, these industries as a group would need to experience extraordinary employment growth.

Does the projected labored employment recovery among these particularly hard-hit industries suggest there are more serious structural impediments to the efficient operation of the labor market today than there were after the previous two recessions? Several posts on this blog (here and here, for example) have addressed this question of structural change without coming to a definitive answer. Returning to chart 1 gives us yet another opportunity to speculate on this point.

Note that the category in chart 1 into which each three-digit industry is placed (above-median or below-median job loss, etc.) is based on job losses between January 2008 and June 2009. Plotting the annual growth rates back to 1990 illustrates that the industries that were hardest hit during the most recent recession were also those with the greatest job losses during the previous two recessions. So there appears to be nothing special about these industries that led to their suffering during the most recent recession.

Additionally, the pattern of recovery of these hardest-hit industries is similar to that experienced after the previous two recessions. Like before, the worst performing industries (those with job losses in the 75th percentile) are adding jobs at a faster rate than industries that did not suffer as much. Industries in the 75th percentile of job losses added jobs in 2011 at an average monthly rate of 0.22 percent; industries with below-median losses added jobs at an average monthly rate of 0.12 percent. This analysis does not suggest to me that unique structural features of this recession or recovery are holding employment growth back—it appears that the culprit is simply the extraordinarily deep hole the economy, and thus the job market, fell into this time around. The bad news, then, is that time may be the only answer for those industries to fully recover.

Julie Hotchkiss By Julie Hotchkiss, a research economist and policy adviser in the Atlanta Fed's research department


February 6, 2012 in Data Releases, Economic Growth and Development, Employment, Labor Markets | Permalink


TrackBack URL for this entry:

Listed below are links to blogs that reference Employment: Some good news, some bad news:


what do you make of the spin on the numbers. some are saying that the labor participation rate is understated, and with a traditional number we would be looking at close to 11% unemployment.

Posted by: Jeff Carter | February 05, 2012 at 02:26 PM

Post a comment

Comments are moderated and will not appear until the moderator has approved them.

If you have a TypeKey or TypePad account, please Sign in

Google Search

Recent Posts



Powered by TypePad