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 2013 | Main | March 2013 »

February 22, 2013

Nature Abhors an Output Gap

In The Washington Post, Neil Irwin highlights a shortcoming that I know all too well:

Throughout the halting economic recovery that began in 2009, the formal economic projections released by the Congressional Budget Office, White House Council of Economic Advisers, and Federal Reserve have displayed quite a consistent pattern: This year may be one of sluggish growth, they acknowledge. But stronger growth, of perhaps 3.5 percent, is just around the corner, and will arrive next year.

Consider, for example, the Fed's projections in November of 2009. Sure, growth would be slow in 2010, they held. But 2011 growth, they expected, would be 3.4 to 4.5 percent, and 2012 would 3.5 to 4.8 percent growth. The actual levels of growth were 2 percent in 2011 and 1.5 percent in 2012.

What's amazing is that the Fed's newest projections, released in December of 2012, look like they could have been copy and pasted from 2009, just with the years changed: They forecast sluggish growth in 2013, 2.3 to 3 percent, followed by a pickup to 3 to 3.5 percent in 2014 and 3 to 3.7 percent in 2015.

I, for one, am guilty as charged, and feel pretty fortunate that the offense is not a hanging one. In fact I don't think Irwin's indictment is overly harsh, and he is on the right track when he offers up this explanation for the last several years' persistently overly rosy projections:

Economic forecasters tend to look at past experience and extrapolate; in the past, when there has been a recession, the very forces that caused the recession become unwound, sowing the seeds for expansion...

Here is a basic fact about macroeconomic forecasting. The truly powerful driver of forecasts is mean reversion, which is the tendency of models to predict that gross domestic product (GDP) will move toward an average trend over time. This fact holds true whether we are talking about formal statistical analysis or the intuitive judgmental adjustments that all forecasters apply to their formal statistical models.

Forecasters are not completely robotic, of course. Irwin is correct when he says "forecasters tend to look at past experience and extrapolate, but forecasters do leaven past experience with incoming details that alter judgments about what is the mean—the "normal state," if you will—to which the economy will converge. But whatever is that normal state, our models insist that we will converge to it.

Nothing illustrates this property of forecasting reality better than this chart, which supplements the latest economic projections from the Congressional Budget Office:


The potential GDP line in that chart is the level of production that represents the structural path of the economy. Forecasters, no matter where they think that potential GDP line might be, all believe actual GDP will eventually move back to it. "Output gaps"—the shaded area representing the cumulative miss of actual GDP relative to its potential—simply won't last forever. And if that means GDP growth has to accelerate in the future (as it does when GDP today is below its potential)—well, that's just the way it is.

Unfortunately, potential GDP is not so simple to divine. We have to guess (or, more generously, estimate) what it is. That guessing game has been harder than usual over the past several years. Here is the record of the CBO's potential GDP since 2009:


I think this picture is a fairly representative record of how views about the potential level of U.S. GDP has evolved over the past several years. What has not been resolved is the debate over what conclusions should be drawn from persistent overestimates of potential and serial misses to the high side on GDP projections.

Irwin seems to be of two minds. On the one hand he offers very structural-sounding reasons for poor forecasting experience:

... the financial-crisis-induced recession of 2008–2009 was so deep that it had deep-seated effects that go beyond those explained by those traditional relationships. It messed up the workings of the financial system, and banks are still trying to figure out what the new one looks like.

On the other hand, he makes appeals to very traditional explanations tied to deficient spending and insufficient policy stimulus (though even here structural change may be one reason that stimulus has been insufficient):

Breakdowns in the financial system mean that low-interest rate policies from the Fed don't have their usual punch. An overhang of household debt means that consumers hold their wallets more than usual. Federal fiscal stimulus to offset those effects is now long-over...

This much, in any event, is clear: Given any starting point where the level of GDP is below its potential level—that is, given an output gap—forecasts will include a bounce back in GDP growth above its long-run average, at least for a while. That's just the way it works.

If, contrary to conventional wisdom, you believe that the true output gaps are much smaller than suggested in the CBO picture above, you might want to take the under on a bet to whether GDP forecasts will prove too optimistic once again.

Photo of Dave AltigBy Dave Altig, executive vice president and research director of the Atlanta Fed


February 22, 2013 in Forecasts, GDP | Permalink


TrackBack URL for this entry:

Listed below are links to blogs that reference Nature Abhors an Output Gap:


Dave - I generally enjoy your posts, but I am surprised that in this one you pretty much ignored the single biggest...and unresolved...question about econometric modeling in this case. You (not just you, of course, but lots of modelers) assume that GDP is trend-stationary rather than unit root. But the history of the last few years is MUCH easier to describe as 'failing to reject the hypothesis' that GDP growth is unit root, wouldn't you say?

I wrote something about this a couple of years ago to explain the issue to non-economists: https://mikeashton.wordpress.com/2010/10/12/the-root-of-the-problem/

Anyway, otherwise I like your articles. Thanks.

Posted by: Michael Ashton | February 22, 2013 at 08:45 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 15, 2013

Promoting Job Creation: Don't Forget the Old Guys

In a provocative article posted this week, the American Enterprise Institute's James Pethokoukis concludes that the state of entrepreneurship in the United States is, disturbingly, weaker than ever. In particular, Pethokoukis documents a decline in jobs created by establishments less than one year old, a trend that began before the 2001 recession and has continued more or less unabated since. He specifically cites the following symptoms of trouble:

  1. Had small business come out of the recession maintaining just the rate of start-ups generated in 2007, according to McKinsey, the U.S. economy would today have almost 2.5 million more jobs than it does.

  2. There were fewer new firms formed in 2010 and 2011 than during the Great Recession.

  3. The rate of start-up jobs during 2010 and 2011—years that were technically in full recovery—were the lowest on record, according to economist Tim Kane of the Hudson Institute.

That last point appears to be all the more ominous given this observation from Tim Kane:

"...that startups create essentially all net new jobs. Existing employers, it turns out, tend to be net job losers, averaging net losses of 1 million workers per year."

Pethokoukis makes his case with political commentary that we don't endorse and don't find particularly helpful. But we won't argue with his conclusion that more entrepreneurial start-up activity would be a good thing. Nonetheless, we get a little concerned when the conversation jumps from data on net job creation and the role of start-ups and early life-cycle firms, and moves on to policy conclusions that seem to disproportionately focus on that class of businesses specifically.

Here is the source of our concern: Though it is also tempting to lump all "existing employers” into the basket of net job destroyers, there are existing firms that create jobs, and a few are doing so on a very large scale.

Take 2006, for instance. Based on data from the Commerce Department called Business Dynamics Statistics (BDS), new firms (businesses with a payroll that existed in March 2006 but not in March 2005) had about 3.5 million employees. This is the large net job creation by new firms reported by Kane. However, over the same year, expanding firms more than 10 years old added a whopping 11 million jobs—about three times as many jobs as created by new firms. Of course, some older firms were downsizing or closing—contracting mature firms destroyed an estimated 10 million jobs. So the net number of jobs created by older established firms looks somewhat less impressive than the record of those young start-ups. But in the overall picture, were the 11 million jobs created by the expanding older businesses really less important than 3.5 million created by the newbies?

It turns out that older firms also account for a large fraction of the job creation occurring in fast-growing firms, arguably a better characterization of entrepreneurism than newness. We found some compelling evidence reported in recent research by Akbar Sadeghi, James Spletzer, and David Talan. Using data from the U.S. Bureau of Labor Statistics' Business Employment Dynamics (BED), Sadeghi, Spletzer, and Talen find that older firms (those at least 10 years old) accounted for more than 40 percent of the employment created by high-growth firms (those with at least 20 percent annual employment gains between 2008 and 2011). A similar conclusion about the role of older, fast-growing firms is found in this earlier Kauffman Foundation report based on BDS data looking at the 1 percent of fastest-growing firms in the United States.

The point is not that start-up entrepreneurial activity is unimportant. It is vitally important. But in larger terms, we should recognize that all entrepreneurial activity is important, no matter what the age of the firm in which it occurs. Atlanta Fed President Dennis Lockhart highlighted this point in a speech delivered earlier this week at Instituto de Empresas in Madrid, Spain:

My bank's experience in trying to understand the role of small businesses, small-growth businesses, young businesses, and mature-growth businesses in job creation illustrates a key point, I think. In the pursuit of economic growth and increased employment, there is no silver bullet. Rather, the policy community should be pursuing an effective mix of policy elements (with focus in areas such as new business formation, labor rules, and regulatory efficiency, to name a few) that together catalyze a virtuous circle of innovation, growth, and employment.

Certainly, entrepreneurial risk-taking, whether by large, mature businesses or start-ups aimed at becoming growth companies, is part of the solution.

When it comes to promoting job creation, forgetting to throw mature businesses into the mix with start-ups is surely not the path to finding the best policy solutions.

Dave AltigBy Dave Altig, executive vice president, and

John RobertsonJohn Robertson, vice president and senior economist, both in the Atlanta Fed's research department

February 15, 2013 in Employment, Labor Markets, Small Business | Permalink


TrackBack URL for this entry:

Listed below are links to blogs that reference Promoting Job Creation: Don't Forget the Old Guys:


I think that to a certain extent the decline in new business formation in 2010-2011 as opposed to 2009 is a bit of a statistical illusion for a couple of reasons.

First, a lot of the start ups in 2008 and 2009 reflect arangements that were first agreed to in 2007 or before--there is a delay between the time when the committment to start a new business is made and when the articles of incorporation are actually filed. Moreover, business cycles follow a bit of a learning curve--back in 2008 and most of 2009, people thought the recession was a lot smaller than it actually was, a misunderstanding that was cleared up with better data in 2010.

Second, there is an extent to which small and new businesses are an inferior good. New businesses typically specialize in smaller orders and attempt to undersell large established suppliers. At the height of the recession, customers downsized, switching from large established providers to small/new competitors in an effort to reduce spending. As the owner of one of these small businesses during the recession, it was definitely my experience that while the type of customers changed in 2008-2009, overall business was still up compared to 2007. That ended when, in 2010, customers started up-sizing again and went to the larger industrial suppliers, so that my business wasn't really hit by the recession until the recovery was in full-swing. If this is a common experience, then the data will show that small/new businesses fared better than large established business in 2008-2009, but then took a hit in 2010-2011.

Posted by: Matthew Martin | February 16, 2013 at 03:27 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 05, 2013

2013 Business Hiring Plans: Employment, Effort, Hours, and Fiscal Uncertainty

How much is fiscal uncertainty holding back hiring? The answer seems to depend on whom you ask. Early in January, the Atlanta Fed spoke to 670 businesses in the Southeast about employment. Conditional on the respondents’ 2013 hiring plans (expand, hold steady, or contract), the following set of charts summarizes the results for how the businesses viewed activity relative to their own interpretation of “normal” along three dimensions: their current employment level, the amount of effort required from their staff per hour, and the average hours worked per employee. These questions were modeled on questions asked in the Atlanta Fed’s December 2012 Business Inflation Expectations Survey. In the following three charts, the green bars represent firms that said they planned to expand employment in 2013. The grey bars represent firms that said they did not plan to change their employment level in 2013, and the red bars represent firms that planned to reduce employment in 2013.

The first chart shows the results for current employment. Regardless of hiring plans over the next 12 months, most firms said they were currently at or below normal employment levels. Those planning on increasing employment over the next 12 months were a bit more likely to say they have already surpassed normal levels of employment than other firms, while those looking to shed employees were very likely to say their employment level is below normal employment levels.


Chart 2 shows that businesses are generally pushing hard along the effort dimension. Firms were quite likely to say that their staff’s effort per hour worked was currently at or above normal,  whether or not they were planning to change employment in 2013.


Chart 3 shows that firms planning to expand were very likely to say that average hours worked were at or above normal (28 percent said hours were above normal, 60 percent about normal), whereas firms planning to contract were more likely to say that hours were at or below normal (48 percent about normal, 39 percent below normal).


Taken together, these results suggest that some firms are approaching the limit of how far they can go along the intensive margins of effort and hours before they have to hire more workers. With effort elevated, as more firms increase average hours worked to above-normal levels, one might expect more hiring to follow.

Each business was also asked how uncertainty about future fiscal policy was affecting its hiring plans. Firms planning to reduce employment tended to cite fiscal uncertainty as having a negative impact on their hiring plans. However, for those firms, hours also tended to be well below normal, so it is unlikely that removing fiscal uncertainty would move many of those firms into expansion mode (although it may help stabilize their outlook).

In contrast, fiscal uncertainty was generally viewed as having less impact by those planning to expand employment and those planning to hold employment levels steady. Presumably, reducing fiscal uncertainty would move some of the firms planning to hold steady into expansion mode, and those planning to expand would do so a bit more. To get some idea of this potential, Chart 4 shows the responses by firms who reported above-normal effort per hour and above-normal average hours worked. About 40 percent of those businesses said that fiscal uncertainty had caused them to scale back their hiring plans.


It is unclear whether eliminating fiscal uncertainty would have a big impact on the hiring plans of these firms. But these results suggest that it sure couldn’t hurt.

John RobertsonBy John Robertson, vice president and senior economist, and

Ellyn TerryEllyn Terry, a senior economic analyst in the Atlanta Fed's research department

February 5, 2013 in Employment, Labor Markets, Small Business | Permalink


TrackBack URL for this entry:

Listed below are links to blogs that reference 2013 Business Hiring Plans: Employment, Effort, Hours, and Fiscal Uncertainty:


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 01, 2013

Half-Full Glasses

Just in case you were inclined to drop the "dismal" from the "dismal science," Northwestern University professor Robert Gordon has been doing his best to talk you out of it. His most recent dose of glumness was offered up in a recent Wall Street Journal article that repeats an argument he has been making for a while now:

The growth of the past century wasn't built on manna from heaven. It resulted in large part from a remarkable set of inventions between 1875 and 1900...

This narrow time frame saw the introduction of running water and indoor plumbing, the greatest event in the history of female liberation, as women were freed from carrying literally tons of water each year. The telephone, phonograph, motion picture and radio also sprang into existence. The period after World War II saw another great spurt of invention, with the development of television, air conditioning, the jet plane and the interstate highway system…

Innovation continues apace today, and many of those developing and funding new technologies recoil with disbelief at my suggestion that we have left behind the era of truly important changes in our standard of living…

Gordon goes on to explain why he thinks potential growth-enhancing developments such as advances in healthcare, leaps in energy-production technologies, and 3-D printing are just not up to late-19th-century snuff in their capacity to better the lot of the average citizen. To paraphrase, your great-granddaddy's inventions beat the stuffing out of yours.

There has been a lot of commentary about Professor Gordon's body of work—just a few examples from the blogosphere include Paul Krugman, John Cochrane, Free Exchange (at The Economist), Gary Becker, and Thomas Edsall (who includes commentary from a collection of first-rate economists). Most of these posts note the current-day maladies that Gordon offers up to furrow the brow of the growth optimists. Among these are the following:

And inequality in America will continue to grow, driven by poor educational outcomes at the bottom and the rewards of globalization at the top, as American CEOs reap the benefits of multinational sales to emerging markets. From 1993 to 2008, income growth among the bottom 99% of earners was 0.5 points slower than the economy's overall growth rate.

Serious considerations, to be sure, but there is actually a chance that some of the "headwinds" that Gordon emphasizes are signs that something really big is afoot. In fact, Gordon's headwinds remind me of this passage, from a paper by economists Jeremy Greenwood and Mehmet Yorukoglu published about 15 years ago:

A simple story is told here that connects the rate of technological progress to the level of income inequality and productivity growth. The idea is this. Imagine that a leap in the state of technology occurs and that this jump is incarnated in new machines, such as information technologies. Suppose that the adoption of new technologies involves a significant cost in terms of learning and that skilled labor has an advantage at learning. Then the advance in technology will be associated with an increase in the demand for skill needed to implement it. Hence the skill premium will rise and income inequality will widen. In the early phases the new technologies may not be operated efficiently due to a dearth of experience. Productivity growth may appear to stall as the economy undertakes the (unmeasured) investment in knowledge needed to get the new technologies running closer to their full potential. The coincidence of rapid technological change, widening inequality, and a slowdown in productivity growth is not without precedence in economic history.

Greenwood and Yorukoglu go on to assess, in detail, how durable-goods prices, inequality, and productivity actually behaved in the first and second industrial revolutions. They conclude that game-changing technologies have, in history, been initially associated with falling capital prices, rising inequality, and falling productivity. Here is a representative chart, depicting the period (which was rich with technological advance) leading up to Gordon's (undeniably) golden age:

Source: "1974," Jeremy Greenwood and Mehmet Yorukoglu,
Carnegie-Rochester Conference Series on Public Policy, 46, 1997

Greenwood and Yorukoglu conclude their study with this pointed question:

Plunging prices for new technologies, a surge in wage inequality, and a slump in the advance of labor productivity - could all this be the hallmark of the dawn of an industrial revolution? Just as the steam engine shook 18th-century England, and electricity rattled 19th-century America, are information technologies now rocking the 20th-century economy?

I don't know (and nobody knows) if the dark-before-the-dawn possibility described by Greenwood and Yorukoglu is the apt analogy for where the U.S. (and global) economy sits today. (Update: Clark Nardinelli also discussed this notion.) But I will bet you there was some commentator writing in 1870 who sounded an awful lot like Professor Gordon.

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

February 1, 2013 in Economics, Productivity, Technology | Permalink


TrackBack URL for this entry:

Listed below are links to blogs that reference Half-Full Glasses:


Thank you. It is always much easier to see what can go wrong than to see what can go right.

Posted by: Douglas Lee | February 02, 2013 at 11:30 AM

Dr. Altig. A great post. Thanks for the historical reference which will be useful in my research. I just strongly disagree with Gordon. Betting against a new technological wave with a very large positive supply (and demand)shock is fool's gold. My personal bet is a truly radical breakthrough in energy.

Posted by: Steve Bannister | February 02, 2013 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