Close

This page had been redirected to a new URL, please update any bookmarks.

Font Size: A A A

macroblog

August 17, 2012

The (Unfortunately?) Consistent Record of the Recovery

In his last two posts (here and here), economist Tim Duy has done some yeoman work displaying and discussing the economic context of monetary policy decisions past and prospective. Though Wednesday's self-titled post "Data Dump" focuses on the incoming data as a set-up to the next meeting of the Federal Open Market Committee (FOMC), what strikes me is the consistency of the broad macroeconomic outcomes over the course of the recovery. Gross domestic product (GDP) growth has pretty clearly clocked in at about 2 percent...


...and, looking through the quarterly ups and downs, payroll employment growth has clearly trended near 150,000 jobs per month after a slower start in 2010:


The inflation picture shows more variation...


...but in my view, that sort of variation is why it makes sense to think in terms of medium-term performance. "Medium-term" is more a measure of art than science, and I would concede the point that the recovery as a whole would be on the shorter end of that time frame. Suffice it to say that the pace of price-level growth over the past two and a half years wouldn't contradict the presumption that inflation is pretty close to the FOMC's stated longer-run objective.

Duy looks at this performance and sees pretty clear evidence of failure:

The economy continues to settle into a path that is not consistent with either part of the Fed's dual mandate. Moreover, there are very real downside risks to even a tepid outlook...

This is frustrating. What in the world is the point of making a big claim to affirm the nature of the dual mandate and then subsequently ignore any forecasts that indicate you have no faith the elements of the dual mandate will be met anytime soon?

That complaint is not really about the inflation part of the mandate, but the employment/growth part of it. But if you are willing to accept that employment growth remains on a pace of 150,000 jobs per month—and I see no clear evidence to the contrary—it is not at all obvious that the pace of the recovery is inconsistent with the FOMC's view of achieving its dual mandate. Here, for example, are the central tendency ranges of the unemployment rate projections from the FOMC's June Summary of Economic Projections (SEP) and the employment growth that would be required to meet those objectives (with some important assumptions, such as the labor force participation rate remaining at the current level).


Here is the important statement of conditionality, as described in the SEP document:

The charts show actual values and projections for three economic variables [GDP growth, the unemployment rate, and PCE inflation] based on FOMC participants' individual assessments of appropriate monetary policy.

Under appropriate policy—which pretty clearly means mandate-consistent outcomes—the majority of FOMC participants don't seem to think that the unemployment rate will improve that quickly. And, to my point, it is not clear that the trend in payroll employment is inconsistent with that pace of improvement.

Of course, individual contributors to the SEP may have different assumptions about things like the labor force participation rate. More importantly, the SEP is silent on what, in each contributor's view, constitutes "appropriate policy."

And I am certainly begging the important issues. Would the economy have achieved even the somewhat unspectacular pace of 2 percent GDP growth, 150,000 jobs per month, and average inflation near the long-run objective absent large-scale asset purchases ("QE2"), forward guidance (statements indicating that policy rates are expected to be exceptionally low through at least late 2014), and maturity extension programs ("Operation Twist")? Does "appropriate policy" imply that more must be done to achieve even the modest progress in the unemployment rate implied in my calculations above? And could we have (looking backward) or can we (looking forward) do even better with an even more aggressive approach, as many Fed critics argue?

Good questions, those.

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

 


August 17, 2012 in Employment, GDP, Inflation | Permalink

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d8341c834f53ef01761748e31a970c

Listed below are links to blogs that reference The (Unfortunately?) Consistent Record of the Recovery:

Comments

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

August 10, 2012

Deflation Probabilities on Our Radar Screen

In the latest Wall Street Journal Economic Forecasting Survey, conducted August 3–6, economists were asked whether they "expect[ed] the Fed to start another round of large-scale bond buying in 2012?" Sixty-three percent answered yes, and 49 percent expected a program would be announced in September, presumably at the end of the next meeting of the Federal Open Market Committee (FOMC) on September 12–13. Obviously this question is of interest to more than just business economists. For example, at his July 17 testimony before the Senate Committee on Banking, Housing, and Urban Affairs, Sen. Mike Crapo asked Fed Chairman Ben Bernanke whether the FOMC should seriously consider more quantitative easing going forward. As part of his response, the Chairman said that "we would certainly want to react against any increase in deflation risk." The entire video exchange can be viewed at the 52–55 minute mark here.

As part of the Atlanta Fed's Inflation Project, we regularly update probabilities of deflation in the Consumer Price Index (CPI) estimated from Treasury Inflation-Protected Securities (TIPS) prices, described here and here. The basic idea is that a recently issued 5-year TIPS has less "deflation protection" than a 10-year TIPS maturing about the same date as the 5-year TIPS. The yield spread between the 5-year TIPS and 10-year TIPS can be used to help estimate the probability of deflation.

The most recent (August 8) estimate puts the 5-year probability of deflation from early 2012 to early 2017 at around 15 percent. As seen in the figure below this probability is up slightly from May, but only about half the readings of the 5-year (2010–15) deflation probability seen in the late summer and early fall of 2010 and considerably below readings seen during the height of the financial crisis in late 2008 and early 2009.


It is important to note that these deflation probabilities are estimates based on a relatively simple model that uses a number of assumptions that not everyone may agree on. Jens Christensen, Jose Lopez, and Glenn Rudebusch at the Federal Reserve Bank of San Francisco have built an alternative model for estimating deflation probabilities that also uses TIPS yields. At the time of the publication of their paper in 2011, their model's probabilities were somewhat lower than—but highly correlated with—ours.

The TIPS market has a number of features that make inferring both inflation expectations and deflation probabilities from them tricky. Most notably, there are unknown liquidity differences between TIPS and nominal Treasury securities. A more direct way of estimating deflation probabilities—or in fact the entire probability distribution of future CPI inflation—using so-called inflation caps and floors has recently been explored by economists Yuriy Kitsul and Jonathan Wright. Inflation caps and floors are essentially options on the Consumer Price Index. We used the Kitsul and Wright method for constructing the implied probabilities of (annualized) CPI inflation over the next five years. As seen in the figure below, this method implies a 13 percent probability that inflation will be 0 percent or negative on average over the next five years. This probability is about at the midpoint of the range that prevailed between October 2009 and March 2012. (See figure 3, on page 29 of Kitsul and Wright's working paper.)


As Kitsul and Wright explain, the market for inflation caps and floors is still quite small relative to the TIPS market. So the deflation probabilities from their model should be considered suggestive, as should our own.

What's the takeaway from all this? Well, readings from the financial market indicate the likelihood of a sustained deflation is currently about 15 percent, or a bit less. That's up from earlier in the year, but not nearly as high as in 2010.

Should we be concerned about the prospect for deflation in the years ahead? This is obviously an important policy question. But I'm not a policymaker; I merely put up the numbers for you to consider. And, of course, we will continue to follow these indicators closely—as can you. Our deflation probability estimates are updated every Thursday and posted on our Inflation Project.

Partick Higgins By Patrick Higgins, an economist in the Atlanta Fed's research department


August 10, 2012 in Deflation, Inflation | Permalink

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d8341c834f53ef017617263289970c

Listed below are links to blogs that reference Deflation Probabilities on Our Radar Screen:

Comments

Sir,

What relevance would you give to market demand for government securities? I have generally equated high demand with hedging against perceived deflation risk, even though the data suggest the two are actually diverging. Can bond market signalling be used reliably to determine probability of phase change from inflationary to deflationary environment or vice versa?

Posted by: Ben Johannson | August 11, 2012 at 01:59 AM

Sustained deflation seems unlikely. A far more likely explanation/interpretation (that btw agrees with 1 and 2-yr breakeven-implied TIPS inflation over the June/July time period) is that there is a significant chance that one year of the next five will see deflation. In other words: be very afraid.

Posted by: dwb | August 11, 2012 at 01:32 PM

How could one go about assessing the same calculation using PPI vs. CPI? I am curious if there are any notable differences between the underlying index for inflation as it could potentially have implications for profit margins.

Thanks for your time and efforts.

Posted by: Danny | August 15, 2012 at 10:43 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

June 25, 2012

Do falling commodity prices imply disinflation ahead?

Cost pressures at the manufacturing level appear to be easing—at least, so say the manufacturers in our Business Inflation Expectations survey. In June, manufacturers reported that unit costs were up only 1.3 percent over the last 12 months, a full percentage point below their assessment at the end of last year. Retailers, on the other hand, report unit cost increases of 2.1 percent, down a bit from May, but 0.3 percentage points higher than in December.

We put a special question to our panel in June that may shed a little light on these patterns. When we asked firms to tell us what has been driving their unit costs over the past 12 months, manufacturers saw considerably less pressure coming from their cost of materials compared with other firms. Perhaps this discovery isn't very surprising. After all, commodity prices have been falling pretty sharply of late, and these costs are especially influential to manufacturers' assessment of the cost environment. (Indeed, in response to a special question we asked our panel in March, manufacturers ranked materials costs as the number-one influence on their pricing decisions.)

Does the fall in commodity prices mean we can expect a pass-through of these lower costs to consumers?

Perhaps. There's certainly a strong intuitive appeal to the "pipeline" theory of inflation. Here's the idea as described by the Bank of England (BOE):

"Consumer prices…can be thought of as the end of a 'pipeline' of costs and prices. The final price will be made up of many different components of cost as well as the retailer's profit or margin… Prices at one stage of the pipeline become costs for the next stage…"

But economists who have looked down the inflation pipeline haven't found flows, but rather trickles. Years ago, Todd Clark of the Cleveland Fed put it this way while he was at the Kansas City Fed: "the empirical evidence… shows the production chain only weakly links consumer prices to producer prices."

So the "inflation pipeline" theory isn't that simple, as the BOE goes on to explain:

"The [pipeline] idea is a simplification… Prices are determined by the interaction of supply and demand. If the cost of raw materials rises, for example, producers or retailers might accept lower profit margins rather than raise their prices. They are more likely to do this if demand is weak or because of competition. The degree of competition in markets can affect how much cost increases are passed on to consumers."

Investigations into what might be obstructing the flows through the inflation pipeline have taken several approaches, including the one suggested by the BOE above: Firms may vary their markups (or margins) to damp the influence of costs on prices as they pass from one stage of production to the next. This idea has become a cause célèbre in macroeconomics and a key element of something called the New Keynesian Phillips Curve.

And so we've been keeping our eyes on how our panel assesses their margins, and we note something pretty striking. That is, margins are rising, but primarily for retailers. Indeed, as our panel sees it, retail margins are getting pretty close to returning to normal. Manufacturers, however, still see their margins as well below normal.

Expanding margins, then, may slow the flow of falling commodity prices through the inflation pipeline. Manufacturers may take the fall in commodity prices as an opportunity to improve their woeful margins. And if they do pass these cost savings on down the production chain, it still might not hit consumers' wallets if retailers continue to increase their margins. (Based on our survey, that's what seems to have been going on lately, anyhow.)

For other insights from the June Business Inflation Expectation survey, see the Inflation Project on our website.

Mike BryanBy Mike Bryan, vice president and senior economist,

Laurel GraefeLaurel Graefe, economic policy analysis specialist, and

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

June 25, 2012 in Business Inflation Expectations, Inflation, Pricing | Permalink

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d8341c834f53ef017742b578d8970d

Listed below are links to blogs that reference Do falling commodity prices imply disinflation ahead?:

Comments

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

June 13, 2012

The armchair Fed historian

I enjoy researching economic history—some of my work with Steve Quinn on early central banking is here, here, and here.

The only problem with historical research, though, is that it tends to involve some real work—long hours spent with dusty archival volumes, consumption of lots of coffee and antihistamines, and a steady hand on the digital camera.

Just recently—and somewhat belatedly—I became aware of a Google application (thanks to Benjamin Guilbert's blog) that lets would-be economic historians breeze over the rough stuff and do some interesting research from their own computer keyboards. The application is called Ngrams, and here's how it works.

Basically, Ngrams counts occurrences of words in books that have been scanned by Google into its Google Books database. It then plots out the frequencies of these words as annual time series. These plots can then be used to measure how interest in a topic varies over time—to construct "cultural histories."

There are some limitations to this technique, mostly related to unavoidable issues in Google's database. For example, the dataset I chose to work with covers only English-language publications and stops in mid-2009. (See the Ngrams website for more detailed information.)

The six charts below represent a first attempt to use Ngrams to delve into the cultural history of the Federal Reserve.

Question 1: How popular is the Federal Reserve as a discussion topic compared with other central banks?

  • Search terms: Bank of England, Federal Reserve, Reichsbank, Bundesbank, Bank of Japan
  • Time period: 1900–2008


  • My interpretation: almost from its beginning in 1913, the Federal Reserve has been the primary focus of English-language writing on central banks.

Question: 2 The Fed was founded as a means to counteract banking panics. What has been the impact of the Fed on the discussion of panics?

  • Search term: bank panic
  • Time period: 1866–2008


  • My interpretation: that bank panics were widely discussed in the wake of three National Banking Era panics in 1873, 1893, and 1907, no surprise. Interest in bank panics peaked following the widespread bank failures of the early 1930s. This topic became less popular after World War II, but interest reawakened with the numerous savings and loan failures of the 1980s and early 1990s.

Question 3: One of the early policy goals of the Fed was to improve the efficiency of the check payment system. When did use of checks become the norm for ordinary Americans?

  • Search terms: pay envelope, pay check, paycheck
  • Time period: 19002008


  • My interpretation: in 1920, most people did not have checking accounts and were paid in envelopes stuffed with cash. By 1960, most households had checking accounts and were paid by "pay check," later contracted to "paycheck."

Question 4. What has been the impact of the Fed on people's concerns about inflation and unemployment?

  • Search terms: unemployment, inflation
  • Time period: 1900–2008


  • My interpretation: interest in unemployment shot up during the Great Depression, fell back in the postwar years, but resurged in the 1970s. Discussion of unemployment then falls steadily to the end of the sample in 2008. Inflation was rarely discussed until the United States left the gold standard in 1933. Interest in inflation remained below unemployment until inflation began to accelerate in the 1970s. Since about 1980, interest in these two topics has been almost identical.

Question 5: In the mind of the public, which policy goal should the Fed be most concerned with: price stability, financial stability, or employment?

  • Search terms: price stability, financial stability, Phillips curve (as an imperfect proxy for "employment"; note that the original article by William Phillips appeared in 1958)
  • Time period: 1900–2008


  • My interpretation: financial stability was paramount until after the 1951 Treasury-Fed Accord. Price stability then takes center stage until the turn of the 21st century but by 2008 had converged with financial stability. Interest in the Phillips curve seems to have peaked in the early 1980s.

Question 6: What has been the impact of two "big ideas" on monetary policy, proposed by Robert E. Lucas (1976) and John B. Taylor (1993)?

  • Search terms: Lucas critique, Taylor rule
  • Time period: 1970–2008


  • My interpretation: in his 1976 paper, Lucas argued that there were limits on the usefulness of statistical relationships (the Phillips curve in particular) in monetary policymaking. Partly in response to the Lucas critique, Taylor in 1993 proposed that central banks follow a simple rule in setting short-term interest rates. Interestingly, discussion of the Lucas critique peaked around the time of the publication of Taylor's paper. Interest in the Taylor rule was still growing at the end of the sample in 2008.

You may or may not agree with the choice of search terms or the interpretations of the search results, but you are welcome to conduct your own historical research with the same application—all from the comfort of your armchair, no digital camera required. We'll have more of these cultural histories to share in later posts.

Photo of Will RoberdsBy Will Roberds, research economist and senior policy adviser at the Atlanta Fed



June 13, 2012 in Employment, Federal Reserve and Monetary Policy, Inflation, Monetary Policy | Permalink

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d8341c834f53ef0163067399a0970d

Listed below are links to blogs that reference The armchair Fed historian:

Comments

That is a fantastic tool! Thanks for sharing. I especially found your search on the Lucas critique vs. Taylor rule surprising.

Posted by: Miraj Patel | June 13, 2012 at 02:24 PM

great post!

Posted by: dwb | June 13, 2012 at 06:51 PM

This is indeed a cool post. Glad that you shared this. thanks!

Posted by: business consulting | June 14, 2012 at 01:28 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

May 17, 2012

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.

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

May 17, 2012 in Deflation, Economic Growth and Development, Federal Reserve and Monetary Policy, Inflation, Monetary Policy | Permalink

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d8341c834f53ef0168eb937883970c

Listed below are links to blogs that reference Is inflation targeting really dead?:

Comments

"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."

Well, sort of. First of all, the Fed's forecasts certainly look like 2% PCE is an upper limit not an average, most commentators rightly called foul after the last quarterly set of forecasts. There is no de-facto difference right now between strict inflation targeting with a 2% ceiling and what the fed is doing. ther is no "flexible" in the target, and the range of full employment forecasts is 5-7%, we are still well above that yet there is no tolerance for higher inflation. So no I would not say "we took a big step forward."

In fact, nominal expectations have collapsed (see beckworth and the Evans new paper) so i would say the Fed has zero credibility on the unemployment side of the mandate. unsurprisingly, confidence and planned expenditures have collapsed. Its not a structural thing.

TIPS markets are screaming: epic fail.

Second, the fed cannot promise to control import prices (and indeed, tightening in response to a supply shock is textbook bad macro). The Fed should only focus on the price of domestically produced goods (gdp deflator). The Fed should have payed more attention to the gdp deflator and less to the PCE during the 2003-2008 period. Cheap imports depressed the PCE early, then expensive oil pushed it up in 2008. Policy would have been tighter, earlier.

At Sept 16th 2008 meeting, despite declining employment, tight credit, high mortgage delinquencies, housing market is recession since 2006, and declining GDP deflator, the Fed was concerned about inflation - not the clearly weakening economy (because oil and commodities prices were high). 3 days later Lehman collapsed and 3 weeks later they eased.

If you think the Fed follows a "balanced Taylor Rule" using output and inflation (use the GDP deflator as i said above) thats just ngdp targeting, except that the Fed promises to correct its own errors over a 5 year period so that the average works out.

Also, another aspect of ngdp targeting is that it prevents a debt-deflation spiral that happened in 2008 (and i think this is what prevented the 1990 real estate bubble from becoming worse in 1990s).

So, yes, IT is dead. RIP and good riddance.

Posted by: dwb | May 17, 2012 at 11:34 PM

thanks for reading comments, this is an interesting debate. Just a couple points as I forced myself to go back and reread Orphanides paper.

1) He used the GDP deflator. I view the conclusions as applied to the Fed framework with suspicion there since the Fed targets PCE and we know they sent very different signals during 2008;

2)The Taylor rule performs worse under imperfect information about the output gap (see figure 9); in fact ngdp targeting still is better under perfect information than strict inflation targeting. This is consistent with McCallum's 1998 results as well as i recall.

3)His ngdp rule is a *growth rule* not a path rule; Sumner and Beckworth propose a path rule (i.e. the Fed promises to correct errors so that the 5 year average (say) is on a target path. Important difference.

4)The chief criticism is that we do not know what potential output is, therefore do not know what to set the path to. But we can observe the trend GDP deflator and adjust the path as needed to be consistent.

I think that if you were to compare Sumner/Beckworth ngdp path level targeting to gdp deflator path level targeting, or inflation targeting using the GDP deflator, then there would only be a very mild difference.

The crucial differences are: the response to supply shocks (the Fed can only control domestic goods prices); and all the theory and evidence that ngdp targeting avoids debt-deflationary spirals like we saw in 2008. Again, compare the response to the housing real estate in the early 90s (yes, we had a housing crisis then too!).


Posted by: dwb | May 19, 2012 at 02:31 PM

Another reason NGDP level targeting trumps inflation targeting: it would not allow expectations of nominal income growth to collapse.

http://macromarketmusings.blogspot.com/2012/05/dereliction-of-duty.html

Posted by: Anon1 | May 19, 2012 at 10:00 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

May 03, 2012

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."

Distribution of Respondent Expectations for Unit Costs

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.

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

May 3, 2012 in Business Inflation Expectations, Federal Reserve and Monetary Policy, Inflation, Monetary Policy | Permalink

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d8341c834f53ef01676612c071970b

Listed below are links to blogs that reference Symmetric goals, asymmetric risks:

Comments

1. wheres the equivalent chart for unemployment or jobs? there is a dual mandate you know.
2. its not just about the miss, but the costs. how does 3% inflation compare to 8% uemployment?
3. do we know whether this is biased or not over the long term, or how this sample compares with a representative slice of gdp?

Posted by: dwb | May 03, 2012 at 03:36 PM

i looked at the data and I'm pretty skeptical this makes the point inflation risks are skewed. Its based on ~160 respondents, many of whom don't answer all the questions and some of whom only answered 5%+. The data series only goes back 2 years so its impossible to assess bias (but over the last two years the mean is stable at ~2%). There seems to be a break 1/2 through the series where 5%+ was added.

There is always going to be some distribution of relative unit costs: some industries are doing really well and some not (for example several of the >5% respondents are in the legal and professional services, not representative of the whole economy). Raising inflation might move the <-1% and -1 and 1% category into the 1 to 3% category, without impacting the rest - raising the mean only somewhat.

also, its dangerous to generalize individual business results to the overall economy, when unemployment is 8% and wages are sticky: There is a large pool of workers stuck at 0% wage growth, See here: http://www.frbsf.org/publications/economics/letter/2012/el2012-10.html

Wages are ultimately ~70% of gdp, which means higher demand will flow back into wages. Unit costs could go up or down depending on productivity.

A broader, better, forward-looking assessment of inflation is here: http://www.bloomberg.com/quote/USGGBE03:IND/chart

Posted by: dwb | May 04, 2012 at 07:48 AM

i cannot find an update of this CPI diffusion index below (seems like a really useful metric but i cannot find it on the inflation dashboard). Seems to me that what you really want to do is compare the probabilities in the survey relative to the distribution of changes ordinarily seen in the CPI, to see if its really that skewed.

http://macroblog.typepad.com/macroblog/2010/04/disinflation-is-it-all-housing-we-think-notand-were-not-alone.html

Posted by: dwb | May 04, 2012 at 12:20 PM

But what would this chart look like if unit labor costs were "deflated" by expected productivity gains across the same ten years? The modal expectation falls below 2%.

Posted by: Bo Parker | May 04, 2012 at 02:14 PM

From Nicholas Parker, Economic Research Analyst: On http://www.frbatlanta.org/research/inflationproject/dashboard/ if you click on the "Retail Prices" category, you will see its underlying series, including the Consumer Price Index (CPI). Currently, the CPI data reflect the most recent release (April, containing the March data), and the next update is scheduled for May 15.

Posted by: Webmaster | May 04, 2012 at 03:51 PM

Consumer or commodity price inflation and asset price inflation are two different animals. It is disturbing to see the primitive understanding of this point at the Fed and elsewhere.

It is deflation in asset prices, particularly real working plant and equipment, facilities and real estate(as opposed to financial assets) that starves the economy of the investment it needs. One only has to look at the paucity of real investment we are now experiencing to see that this has not been avoided.

The fact that we have had zero interest rates for, what, four years and have seen nothing more than low single digits illustrates there is something missing from the guvna's analysis.

Posted by: demandside | May 15, 2012 at 01:11 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

April 16, 2012

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

April 16, 2012 in Africa, Inflation, Monetary Policy | Permalink

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d8341c834f53ef01676537e008970b

Listed below are links to blogs that reference Taking a deeper dive into the definition of inflation:

Comments

Inflation should be correctly defined as the rise in relative prices as a result of an increase in the general money supply. I am not sure if there is a good way to measure this definition but it is important to be properly understood.
The reason is because what needs to be defined is the "bad" price increases. If goods are more efficiently produced in a manner that would bring down prices this is "good". But if those prices are not allowed to drop and are offset due to an increase in the general amount of money, this is still considered inflation "bad" because it removes purchasing power from what would normally be there if intervention in the money supply was not induced even though the price has not changed. Similarly to the situation of a supply glut where a good is scarce; the resulting price increase can occur irrespective of monetary increases and is considered "good". This is because it signals the market to produce more of these goods and reduce demand. This should not be entered into the inflation equation. Therefore, inflation should only be measure as the "relative" prices increase due to the change in money supply.

Posted by: Darryl Jones | April 17, 2012 at 09:48 AM

How about Mish Shedlock's definition of inflation and deflation:

--Inflation is a net increase in money supply and credit.

--Deflation is a net decrease in money supply and credit.

I think he tries to argue that inflation itself is a byproduct of increases in the money supply and credit. Not sure Mish is right, but he is thinking somewhat outside the box.

Posted by: farmland investment | April 17, 2012 at 05:14 PM

I agree totally Mike.

In fact I think inflation can also be understood better simply by looking at its impact on behaviour and attitude to money.

During the pre-crisis years money was plentiful. It was easily available because of easy credit conditions and loose monetary policy. As a consequence people lost respect for money. People both borrowed and lent it stupidly. People spent money today that they would ordinarily have deferred spending until later. These conditions are, to me, inflationary regardless of what price indices told us (although we all know that prices of assets not recorded in the various inflation measures rose considerably).

Now what do we have?

Money is scarce. People have regained respect for money. Banks aren't lending it. Households want to save more. Companies aren't investing. People are deferring today's consumption until later. This, to me, is what deflation is, regardless of what the price indices are telling us.

Today's price rises are squeezing disposable income but workers know that the fragility of the economy can not tolerate much upward wage pressure. Can you really have the type of inflation that destroys competitiveness in these conditions?

Posted by: Peter C | April 18, 2012 at 05:23 AM

Inflation is a net increase in money supply and credit.

Posted by: sunglasses hut | April 19, 2012 at 09:20 AM

If the Fed doesn't know what the definition of inflation is or how to measure it, the Fed should not exist. A surgeon who does not know the outcome of removing one organ and adjusting another should not operate. If the definition of inflation has changed from Friedman's definition to the idea of price increases, why is a rising stock market not considered inflation? This article conjours up a view in my head of a medeival sorcerer mixing frog necks in a bubbling pot casting spells and looking for evidence of success.

Posted by: Bob | April 21, 2012 at 07:06 AM

personally i ask myself this all the time, until i finally hit on Ed Dolan's definition of inflation:

If you could choose between shopping on line today at today’s prices, or buying from a mail order catalog of the past at past prices, what items, if any, would you buy from the past?

with the internet, you can find all sorts of examples (like the $150 bike, $30 helmet, and $25 lock in the 1988 sears catalog can be found in *todays* sears online catalog for the same price).

Its nearly impossible to quantify substitution and quality improvements (the laptop i am writing this on sells new for 50% of what i bought it for 3 years ago). oh, and changing baskets of goods as we retire and consume more health care.

But, making it concrete helps understand what we are talking about. some things, like oil and cars have gone up (except that cars are so much more fuel efficient now, has the all-in price really increased?).

and by the way, that bike has not changed in price sine 1988, but wages have (productivity!) so no wonder I bought a much more expensive bike!

in other words, the more deeply you think about it, the more ephemeral it becomes.

http://www.economonitor.com/dolanecon/2012/03/08/finally-proof-real-proof-not-just-data-of-what-inflation-has-done-to-our-economy/

Posted by: dwb | April 24, 2012 at 05:15 PM

The video was helpful, but I felt like a child watching a cartoon. A little humiliating. The content was good though.

Posted by: Anthony T | April 24, 2012 at 10:09 PM

It may be time to tackle or at least examine inflation from different vantage points within the economy. How about 4 different perspectives: those in poverty/low income families, the middle class, the upper class, and industry. I wonder if aggregating prices with the gross assumption that each price impacts all sections of society equally causes distortions in policy makers perception of what is really going on. I must simply ask should/would Central Bank monetary policy change if we found out that from the middle class perspective inflation was 5,6, 7% or more but in aggregate inflationary pressures were only 2.5%? Maybe someone can reading this blog can provide some guidence but intuitively I would think the relative importance of inflationary pressures decline as income increases. If this were true AND aggregation of price data were shown to skew/smooth inflation data thereby masking the impact of inflation on the middle class and below...Central Bank policy would most likely be out of whack for the majority of people. Just wondering...

Posted by: Danny | April 27, 2012 at 10:29 AM

Today's price rises are squeezing disposable income but workers know that the fragility of the economy can not tolerate much upward wage pressure. Can you really have the type of inflation that destroys competitiveness in these conditions?

Posted by: sunglasses hut | April 29, 2012 at 11:44 PM

As a practical matter, in my working life, The Fed has always defined inflation as increasing wages for ordinary idiots. Thus it has never been "inflationary" for bank executives to quadruple and quadruple again their rake, nor has it been counted as "inflation" when the price of oil went through the roof. And when the price of literal roofs (and the homes beneath them) went absolutely bonkers, from 2002--2006? Not inflation.

That was called "a dynamic and prosperous economy."

My wages have been flat, or nearly so, since I began my working life in 1989. My wage was frozen in 2008, reduced in 2009-2010, restored to its 2008 level last year and remains there now. Adjusted to the official CPI I am paid nine percent less today than I was in 2004.

Ergo: no inflation!

Hence: that insipid video from the Atlanta Fed.

Posted by: Edward Ericson Jr. | April 30, 2012 at 02:09 PM

Consumer prices are not overall prices. In this case they reflect the rise in commodities that the Fed itself has engineered with the various QE's. Overall prices would include labor, perhaps housing and assets. Rental rates are included, and their rise is really from the misery in homeownership.

What is clear is the rise in aggregate prices is not caused by a wage-price spiral or anything close to nearing limits on capacity. Ah, but the magis at the fed are ready to raise rates to stall the economy that is already moribund. (Not that raising rates would not be a good idea for other purposes.)

Posted by: Demandside | April 30, 2012 at 11:19 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

March 23, 2012

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.


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:


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.


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


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.


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.

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

 

March 23, 2012 in Deflation, Employment, Federal Reserve and Monetary Policy, Inflation, Labor Markets, Monetary Policy | Permalink

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d8341c834f53ef01630330bd3a970d

Listed below are links to blogs that reference Why we debate:

Comments

I like your Employment chart. The bubbles giving the labor group size is a nice addition.

On thing on inflation: I picture inflation as a tug-of-war between the money supply and economic growth. Normally, I would say both teams have 1 person on each side of the rope. However, in present times, we have had extended periods of very low interest rates pumping inflation up while a slow economy is tugging it down. The teams are much bigger in this case, 10 on each side. When one side eventually falls in the mud, the consequences will be much greater due to the higher tension.

Inflation may be moderate for now, but we are balanced on a knife edge.

Posted by: BTN | March 26, 2012 at 12:41 PM

Inflation can be a headache in these situations. Employment must be high to counter such effects.

Posted by: Dallas Real Estate | March 30, 2012 at 05:23 PM

If one assumes that there has a been a structural shift in the labor market in the past twenty five years, what do the average charts look like for the past three cycles (as opposed to five)?

Posted by: dickens | April 03, 2012 at 01:09 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 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

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d8341c834f53ef016762cad49e970b

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

Comments

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?

http://www.econbrowser.com/archives/2012/02/the_2012_econom.html

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

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d8341c834f53ef0168e7773525970c

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

Comments

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