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.

« August 2010 | Main | October 2010 »

September 30, 2010

Small businesses are the engine of job growth…or is the story more complicated?

Small businesses typically seek loans to grow their business, hire more workers, or purchase equipment needed to handle more activity. However, business owners have either been unwilling to take on debt or have been unable to find the small business loan opportunities they seek. This message is the one seemingly sent by small business owners who participated in polls such as the NFIB survey and the Atlanta Fed's own surveying efforts highlighted in past macroblog posts (here and here and also on our new Small Business Focus web page).

This week President Obama signed into law a new initiative to try to stimulate borrowing and spending by small businesses. Such policy actions are usually based on the premise that "small businesses are the engine of job growth." However, it is tempting to be skeptical of claims that talk about any large group of individuals or firms as if they are a single, homogeneous unit. Idiosyncratic features such as a firm's industry, location, or age might matter as much as does its size, which would seem to indicate that not all types of small business are equally powerful engines of job growth.

Economic research published last month by John Haltiwanger, Ron Jarmin, and Javier Miranda provides some compelling evidence on the relationship between firm size and job growth. It turns out that the age of a firm is important independent of its size. In particular, the paper finds no systematic relationship between net job growth rates and firm size after controlling for firm age. To quote from the paper's abstract:

"There's been a long, sometimes heated, debate on the role of firm size in employment growth. Despite skepticism in the academic community, the notion that growth is negatively related to firm size remains appealing to policymakers and small business advocates. The widespread and repeated claim from this community is that most new jobs are created by small businesses. … However, our main finding is that once we control for firm age there is no systematic relationship between firm size and growth. Our findings highlight the important role of business startups and young businesses in U.S. job creation. Business startups contribute substantially to both gross and net job creation. In addition, we find an 'up or out' dynamic of young firms. These findings imply that it is critical to control for and understand the role of firm age in explaining U.S. job creation."

This finding doesn't imply that firm size is irrelevant, but size matters mainly because, conditional on survival, young firms grow faster than older firms and tend to be small. In other words, because start-ups tend to be small, most of the truth to the popular perception that small businesses create the most jobs is driven by the contribution of start-ups to net job growth.

Because of the vital role that young firms appear to play in job creation, understanding the various factors that influence business start-up decisions is particularly important. To quote the conclusion of the paper:

"In closing, we think our findings help interpret the popular perception of the role of small businesses as job creators in a manner that is consistent with theories that highlight the role of business formation, experimentation, selection and learning as important features of the U.S. economy. Viewed from this perspective, the role of business startups and young firms is part of an ongoing dynamic of U.S. businesses that needs to be accurately tracked and measured on an ongoing basis. Measuring and understanding the activities of startups and young businesses, the frictions they face, their role in innovation and productivity growth, how they fare in economic downturns and credit crunches all are clearly interesting areas of inquiry given our findings of the important contribution of startups and young businesses."

One of the goals of the Atlanta Fed's new small business web page is to feature data, information sources, and research that helps disentangle the complex contributions of small businesses to economic growth and development. This web page will also highlight the findings from specific initiatives sponsored by the Atlanta Fed.

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

September 30, 2010 in Small Business | Permalink


TrackBack URL for this entry:

Listed below are links to blogs that reference Small businesses are the engine of job growth…or is the story more complicated?:


Also you should document the growth rates of these new winners as they get 'older.'

As with many things in business, there are often lopsided results where a few companies grow tremendously compared to the remainder. Understanding exactly what dynamics are involved in sustaining high growth rates would be worthwhile research.

Posted by: beezer | October 01, 2010 at 11:54 AM

This is a wonderful addition to the factual context of the job-creation problem.

Policies aimed at getting existing businesses to hire may be missing the point; the problem is that entrepreneurs are confused and demoralized and unwilling to take risks in an unstable environment. Someday, I hope to be one of them -- but at the moment I wouldn't go out and start a business unless there were no other way to feed my family.

I have also heard it said that a disproportionate share of small businesses are in the real estate sector, which of course is likely to remain depressed for at least another year.

But I suspect there are, or will soon be, job openings at class action law firms and forensic accounting firms, and that the foreclosure industry is going to have to triple its workforce in order to actually meet bare minimum legal standards in the future!!

Posted by: Wisdom Seeker | October 11, 2010 at 11:50 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

September 10, 2010

Policy may have created the housing bubble, but which policy is to blame?

There is little dispute that misguided policy choices led to the housing boom-bust cycle from which we are still recovering. The debate about which policies were most culpable, however, rages on. The latest chapter in this dispute is now available in the proceedings from this year's edition of the Kansas City Fed's Jackson Hole Economic Policy Symposium.

In defense of monetary policy, Charles Bean, Matthias Paustian, Adrian Penalver, and Tim Taylor—all of the Bank of England—write this:

"We argue that while relatively low policy rates compared to past experience contributed to the growth in credit and the rise in house prices in the run-up to the crisis, they played only a modest direct role."

Stanford University's John Taylor (still) isn't buying it:

"Their conclusion differs from mine for several reasons. First, they do not take account of much empirical work completed since the 2007 Jackson Hole conference. For example, Jarocinski and Smets (2008) of the European Central Bank estimated a VAR [vector autoregression] for the United States and found evidence that 'monetary policy has significant effects on housing investment and house prices and that easy monetary policy designed to stave off perceived risks of deflation in 2002-04 has contributed to the boom in the housing market in 2004 and 2005.' In a more recent study focusing directly on deviations from policy rules, Kahn (2010) of the Federal Reserve Bank of Kansas City finds that ‘When the Taylor rule deviations are excluded from the forecasting equation, the bubble in housing prices looks more like a bump.' "

I added the links to the papers cited by Taylor because they are thoughtful challenges by thoughtful people, and they deserve to be considered (though the Jaroconski and Smets article requires some tolerance of relatively sophisticated econometrics). That insightfulness, of course, does not mean they are completely persuasive; I still have my doubts.

Most of you are familiar with this picture of the "Taylor rule" referenced above—which prescribes a funds rate target based on the deviations of output from its potential and the deviation of inflation from a presumed target of 2 percent—compared with the actual path of the policy rate:


Bean et al. make note of a speech from the beginning of the year by Chairman Bernanke in which he in turn notes (among other things) that the period in which policy deviates from this particular Taylor rule is also a period in which the lending standards were dramatically relaxed. To give but one example, data collected by my colleague Kris Gerardi indicate that in Massachusetts the median loan to value ratios (LTVs) for all borrowers rose from 0.82 in 2000 to 0.9 in 2006. For subprime borrowers, LTVs rose during that period from 0.85 to 1.0. The statistical results cited in Taylor's response do not control for such developments, making it difficult to come to a strong causal conclusion.

Second, this observation (from the Bean et al. paper) introduces even more uncertainty regarding the robustness of the chain of events leading from low interest rates to the housing bubble:

"Chart 1 shows that both UK and euro-area policy rates were less noticeably out of line with their respective Taylor benchmarks. That too is striking. Indeed, in the United Kingdom, they were actually above the benchmark for much of the relevant period, even though the United Kingdom saw one of the larger run-ups in debt and house prices during this period. And, in the euro area, countries such as Spain experienced substantial house price booms, while countries such as Germany did not. That need not imply that monetary policy was innocent in the run-up to the crisis…But this is hardly compelling evidence for assigning the central role to monetary policy, suggesting that other factors were more important."

As the Bank of England authors suggest, monetary policy was not necessarily innocent. But at a minimum, it's worth keeping in mind that the monetary policy transmission mechanism is a good bit more complicated than any simple story would indicate.

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

September 10, 2010 in Housing, Monetary Policy | Permalink


TrackBack URL for this entry:

Listed below are links to blogs that reference Policy may have created the housing bubble, but which policy is to blame?:


The Sugarscape work of the Santa Fe Institute, shows that markets generate purely endogenous bubbles and busts. Indeed, so does any market model that assumes that rational arbitrageurs may recognize the limits of fundamental arbitrage and instead bet with market movements -- thus accentuating them rather than dampening them. That kind of self-referential behavior makes markets work like herds, and herds stampede.

In other words, bubbles and busts are market phenomena, pure and simple. Policy may do a better or worse job of PREVENTING them, but the bubble itself is just markets acting the way markets act.

The continuing search for a governmental cause to the failure of private sector markets is deeply disappointing. It indicates that the searchers are still living in their Reaganaut fantasy of a market that can do no wrong and a government that can do no right. That ideology blinded the regulators to the bubble last time, and if it is not discarded, it will deceive them again the next time.

Posted by: Daniel G | September 12, 2010 at 07:22 PM

One of the strongest arguments against Fed monetary policy, or any federal government policy, as an important housing bubble culprit is the fact that the housing bubble was not a national, or even really regional, phenomena.

Florida, for example, had an intense housing bubble, but Georgia, next door to it, with a similarly growing economic base, did not.

California had an intense housing bubble, but Colorado and Illinois did not.

The policy that really stands out in most of the states that had severe housing bubbles that then collapsed was the presence of non-recourse residential mortgage laws, or laws that were perceived as being non-recourse.

The handful of exceptions, like Arizona and Nevada, probably represent cases of real estate wealth acquired by Californians where there was a non-recourse lending driven housing boom being rolled over to neighboring states where speculative investment and housing for retirees paid for with proceeds of California home sales make up a large share of the market.

Posted by: ohwilleke | September 13, 2010 at 04:38 PM

California had an intense housing bubble, but Colorado and Illinois did not.

I feel like I'm in that scene in Aliens where Ripley says "Did IQs suddenly drop while I was away??"

Bubbles appeared wherever suicide lending was allowed to happen.

Effective interest rates were dropped to 0-1% via 2 and 5 year teaser-rate ARMs and even negative-am products.

Back-end ratios were abandoned with the acceptance of either explicit liar loans or the connivance of mortgage brokers committing fraud to get their commission.

Down payment standards were abandoned with the innovation of the 103% loan.

The 2001-2003 tax cuts didn't hurt in instantly juicing everyone's buying power several hundred dollars a month.

All these inputs coalesced into a bubble run-up 2002-2004 that encouraged and funded REIC insiders to buy as many houses as they could to profit from the bubble trend.

Additionally, the bubble run-up was liberating HUNDREDS of billions of dollars a quarter of home equity credit via HELOC advances and drawdowns, money that was funding both real estate investment and household spending that supported the economic feel-good times that kept the bubble rolling.

I was outside the industry and not plugged in 2003-2005 so I didn't really understand these factors, but once the Casey Serin story (20-something loser able to take out $2.2M in liar loans 2005-2006) broke in late 2006 I understood that we were looking at a several trillion dollar bubble fraud that was perpetrated by the very simple mechanism of abject abandonment of historical lending underwriting standards.

This ain't rocket science! People will borrow as much money as you are willing to lend them!


And lend we did, 2002-2007.

Posted by: Troy | September 13, 2010 at 10:40 PM

The FED's always at fault. If the FED's technical staff had to pass a test of their knowledge they would all fail.

Why? Because it is a scientific fact that economic prognostications are mathematically infallible.

You don't know how, so you fail.

The housing bubble is especially easy to understand and forecast. All the demand drafts drawn on the money creating depository institutions clear through DDs – except those drawn on MSBs, interbank and the U.S. government.

Posted by: flow5 | September 15, 2010 at 03:40 PM

You might look at the recent article by Krugman and Wells. They argue that the foreign savings glut was more central than loose monetary policies of the Fed or the ECB, and more central than lax financial market regulation. That has long been my own view, so it was nice to see it supported from such a respectable source. The reasons they give for their conclusion are the same as those I gave in numerous posts over the last couple of years. For those who argue that it was primarily monetary policy, they need to explain differences within the euro area, as well as Greenspan's 'conundrum' about long term rates staying low after the Fed started to tighten.

Posted by: don | September 15, 2010 at 09:43 PM

Saw an argument from Fama that there were not any bubbles at all.

The savings glut referenced by Krugman has had nothing to do with it. It was low interest rates by the FED, and the backstop of Fannie and Freddie.

Loans could be written with reckless abandon. I recall an article in Barron's around 2003 about the danger of no interest loans. Maybe they should have written about no doc loans too!

Posted by: Jeff | September 18, 2010 at 10:10 PM

It turns out Keynes had it right concerning bubbles of all kinds: When too much money gets into the hands of too few people, their attempts to do anything with their excess cash will necessarily result in some kind of bubble. (If they were to merely put it into savings instruments they would be shooting themselves in the feet by causing interest rates paid to decline.) The fundamental causes of our present predicament really are a replay of the 1920's: a spurt in productivity created increased profits that the controllers of wages and salaries hogged for themselves, while the workers (perhaps expecting that they would surely get their share soon) participated in the general economic boom by using credit to increase present consumption. Keynes's analysis was pragmatic, not moral: Capitalism works best when there is a large, financially healthy middle class, who can consume a lot, save a little, and perhaps put a bit in equities. That's why we need very high marginal tax rates, to use the money collected to subsidize the middle class (transportation, education, housing--dare I say it, health care). While in gangsta capitalism a few people can become wealthy enough to be above personal ruin no matter what, even most wealthy people are better off with a Keynesian approach, as their wealth ultimately depends on a large pool of viable consumers.

Posted by: tall cotton | September 23, 2010 at 08:48 AM

First, permit me to admit that I am out of my league and should not have the temerity to comment. But, having observed the real estate market for decades, it is not possible to describe what the United States has just been through in the lead up to the Great Recession as a mere market bubble. Nothing like this has been seen before. Rather, the "savings glut" appears to be the best explanation for the unusual market phenomena. In the end, some person or entity must have the money to loan to borrowers to pay ever-increasing large amounts for the housing stock. In essence, it was inflation within a particular asset. Would it be fair to say that, de facto, the Chinese have, by virtue of their surplus in the form of immense U.S. bond holdings, become a second "central bank" for the U.S. economy, able to infuse money supply as suggested by Krugman and Wells? I agree that in some ways, to quote "Troy" it is not "rocket science." There were horrific lending practices. But from whence came the funds to endlessly lend to whomever asked for money? And perhaps it was also a combination of many of the factors identified by the comentors that caused this endless flow of loanable funds to drive prices to their absurd levels, well beyond the means of ordinary income to simply sustain the debt.

Posted by: mme | September 30, 2010 at 11:40 PM

The RE bubble formed because the FED under Greenspan decided that investment banks rather than the central bank should provide liquidity via expansion of an unregulated derivatives market as well as the erosion of banking capital and reserve requirements (as instituted under the Basel accords). The adoption of adjustable rate mortgages, the abrogation of Glass-Steagall, and the excesses of the Financial Modernization Act completed the gutting of any management and regulation of the financial system.

The FED is primarily concerned with the profitability of the banking industry, not the stability and well being of the general economy. As such, it will continue to support increased credit and higher debt/GDP levels until no sector of the economy can service its debt load regardless of the market manipulation exercised.

Posted by: MarkS | October 06, 2010 at 12:37 PM

In my opinion the reason was alluded to in your post, THE GROWTH OF CREDIT


Total debt / gdp has been rising for years and exceeded the peaks of the 1929 crash many years ago.
The rising debt/ gdp ratio manifested itself with a housing bubble.
If you are looking for why housing, employment and other economic indicators are not recovering look to the debt / gdp ratio. The economy is saturated with debt and consumers are concentrating on reducing their debt levels.

Posted by: Greg Merrill | October 12, 2010 at 05:07 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

September 03, 2010

Optimism…pessimism…and a bit of perspective

Here's how I'm tempted to summarize today's release of the August employment report from the U.S. Bureau of Labor Statistics: more of the same. That theme fits nicely with comments this morning from Atlanta Fed President Dennis Lockhart, in a speech at East Tennessee State University. Here he calls for a little perspective:

"Some commentators are reading recent economic data as suggesting the onset of a second recession and deflationary cycle. Quite naturally, business people and consumers aren't sure what to believe.

"At the last meeting of the Federal Open Market Committee (FOMC) in Washington, the committee made a decision that has been widely interpreted as signaling declining confidence in the strength and sustainability of the recovery….

"In my remarks today, I will provide a less alarmist interpretation of recent economic information and the Fed's recent policy decision. I will argue that, generally speaking, there was too much optimism in the early months and quarters of the recovery and now there may be excessive pessimism."

One point is that recoveries are not generally linear affairs:

"Growth at the end of last year and early part of this year was stronger than I anticipated while economic activity in the second and third quarters seems weaker than I expected.

"But such ups and downs are not unusual during a recovery. A little history: following the 2001 recession, gross domestic product (GDP) grew at the annualized rate of 3.5 percent in early 2002. Growth then decelerated to about 2 percent for the next two quarters then fell to almost zero in the fourth quarter. Entering 2003, growth edged up to a little over 1.5 percent and then accelerated from there to a sustained period of relatively strong growth for two years."

Here's a look at a little more history:


Even in the rapid-growth, pre-1990 recoveries, there was generally a quarter or two of growth that underperformed. In the first three months of 1971, the first full quarter after the 1969–70 recession, growth came scorching out of the gate at 11.5 percent. But that was followed by growth rates of 2.29, 3.23, and 1.12 percent. Though the early expansion after the inflation-breaking 1981–82 contraction was robust throughout, the 1973–75 recession softened noticeably in the second year of expansion, with quarterly growth falling just below 2 percent at one point.

But the better benchmarks will likely prove to be the slower-growth, low-employment recoveries post-1990. In addition to the 2001 experience noted by President Lockhart, the expansion that followed the 1990–91 recession stumbled along with quarterly growth rates of 2.7, 1.69, and 1.58 percent, before picking up to above-potential growth rates. Despite that, the eighth quarter after that recession's end clocked in at an anemic 0.75 percent.

What is more important is that there is a reasonably good explanation for why we might have hit a soft patch:

"Looking at the 2009–2010 recovery, it seems clear that some of the early strength was promoted by policies that pulled forward spending from the second and third quarters of this year. The recent sharp decline in housing-related indicators following the expiration of homebuyer tax credits is the most obvious example of this effect."

Comparing monthly home sales patterns with year-over-year performance really does illustrate the point:



Essentially, President Lockhart's is a simple message: don't ignore the short-term data, but be careful with getting too carried away with it as well.

"Simply stated, I was expecting a relatively modest recovery, a pattern typical following the kind of financial crisis we experienced….

"Melding all this mixed information, my basic view of the economy has not changed, but my perception of risks has shifted somewhat to the downside.

"It was this perspective—a perspective I'd characterize as moderate optimism tempered by acknowledgement of weaker conditions and greater downside risk—that I carried into the last FOMC meeting on August 10."

And with respect to that meeting, here is the main policy point:

"At the last meeting there were two important considerations as I saw it. First, as already discussed, some economic data came in weaker than expected, shifting the balance of risks to slower growth in the near term and further disinflation. Second, the Fed's holdings of MBS were projected to decline faster than previously thought because lower rates were generating heavy mortgage prepayments and refinancings.

"So, in the context of a softening economy, the FOMC was confronted with the prospect of unintended withdrawal of support for the recovery through a decline in the level of liquidity provided to the economy….

"That is how I interpret the decision announced following the August meeting—a small tactical change designed to preserve the level of liquidity provided to the system. I supported the committee's decision, but I do not view it as a fundamental change of outlook or strategy. I do not believe this change necessarily heralds the beginning of a period of further expansion of the Fed's balance sheet. Nor do I think the decision precludes a return to a policy of allowing the balance sheet to shrink on its own.

"I think the decision has been over-interpreted in some quarters."

I'll close with that thought by President Lockhart. Have a nice, long holiday weekend.

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

September 3, 2010 in Federal Reserve and Monetary Policy, Monetary Policy | Permalink


TrackBack URL for this entry:

Listed below are links to blogs that reference Optimism…pessimism…and a bit of perspective:


GSE mortgage originations doubled between 2001 and 2003, with most of the jump coming in 2003. This direct impact of Fed policy on the transmission channel may have been responsible for rescuing growth from the late 2002 slowdown. Subsequent to this mortgage boom "kick-start", the shadow banking system was the major engine of credit growth fueling the 2004-2007 recovery.

Further, the 2008/2009 cumulative stimulus was much greater, as a % of gdp, than in 2002. Therefore, absent further stimulus, the economy has a much greater fiscal headwind in 2010/2011 than it did in 2003.

I could go on: the drag from long-term unemployed is much greater now; households face a prolonged period of delevering; the recession is more widespread globally than in 2003; the shadow banking system is paralyzed; etc.

An excess of pessimism, or a closer examination of the differences between this recession and previous ones?

Posted by: David Pearson | September 05, 2010 at 11:28 AM

Past data tell exactly what you say. Nevertheless I still wonder what will be the engine driving a strong recovery in 2011 or 2012.

Posted by: Ignacio | September 06, 2010 at 09:59 AM

" the shadow banking system was the major engine of credit growth fueling the 2004-2007 recovery"

Can you really call the massive misallocation of capital in 2004-7 a "recovery"? I suppose in purely GDP terms you can, but it would be really nice to have some other form of economic measure that indicates how well the economy is doing in terms of sustainable ("real", if you like) growth. With such a measure 2004-7 would appear pretty bad, but today we might be doing all right.

Posted by: John Smith | September 17, 2010 at 09:31 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

Google Search

Recent Posts



Powered by TypePad