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 2013 | Main | October 2013 »

September 26, 2013

The New Normal? Slower R&D Spending

In case you need more to worry about, try this: the pace of research and development (R&D) spending has slowed. The National Science Foundation defines R&D as “creative work undertaken on a systematic basis in order to increase the stock of knowledge” and application of this knowledge toward new applications. (The Bureau of Economic Analysis (BEA) used to treat R&D as an intermediate input in current production. But the latest benchmark revision of the national accounts recorded R&D spending as business investment expenditure. See here for an interesting implication of this change.)

The following chart shows the BEA data on total real private R&D investment spending (purchased or performed on own-account) over the last 50 years, on a year-over-year percent change basis. (For a snapshot of R&D spending across states in 2007, see here.)

Real Spending on Research and Development

Notice the unusually slow pace of R&D spending in recent years. The 50-year average is 4.6 percent. The average over the last 5 years is 1.1 percent. This slower pace of spending has potentially important implications for overall productivity growth, which has also been below historic norms in recent years.

R&D spending is often cited as an important source of productivity growth within a firm, especially in terms of product innovation. But R&D is also an inherently risky endeavor, since the outcome is quite uncertain. So to the extent that economic and policy uncertainty has helped make businesses more cautious in recent years, a slow pace of R&D spending is not surprising. On top of that, the federal funding of R&D activity remains under significant budget pressure. See, for example, here.

So you can add R&D spending to the list of things that seem to be moving more slowly than normal. Or should we think of it as normal?

Photo of John RobertsonBy John Robertson, vice president and senior economist in the Atlanta Fed’s research department

September 26, 2013 in Business Cycles, Capital and Investment, Productivity | Permalink


TrackBack URL for this entry:

Listed below are links to blogs that reference The New Normal? Slower R&D Spending:


As someone who has spent many years in corporate R&D, I think I would advise some caution in interpreting these numbers. My experience is that an enormous amount of corporate R&D spending is simply wasted, essentially through poor management (alternatively just the fact that managing R&D from ideation through to product creation and monetization is really a very difficult task).

So it's possible that a gradual fall in overall R&D expenditure, especially relative to its natural variability, could actually reflect a healthy re-balancing of corporate spending, either through improved research productivity or through a shift towards more product-oriented expenditures. Without a lot more analysis it's difficult to really assess what's going on here.

Posted by: Mark Thomson (@markmthomson) | September 26, 2013 at 05:43 PM

Do these data include expenditures at universities? Maybe it's a low share. But as a public good every state has an incentive to let someone else provide elite higher education (as they're the most mobile geographically) and R&D.

Posted by: mike smitka | October 03, 2013 at 02:03 PM

All over the world the spending on R & D is getting reduced.In educational institution R & D spending getting even lower.

Posted by: Saran | October 18, 2016 at 03:10 AM

R&D is one the critical factor for achieving future growth and maintaining a relevant product in the market. R&D doesn't always guaranteed success commercially. R & D adds more value to an organization( both educational & business ). But organizations may or may not invest funds on R & D because on uncertain results.

Posted by: Priya | October 13, 2017 at 04:05 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

September 23, 2013

The Dynamics of Economic Dynamism

Earlier today, Atlanta Fed President Dennis Lockhart gave a speech at the Creative Leadership Summit of the Louise Blouin Foundation. He posed the questions: Is the economic dynamism of the United States declining? Is America losing its economic mojo? He observed:

“... we see a picture in which fewer firms are expanding, and each expanding firm is adding fewer new jobs on average than in the past. Fewer firms are shrinking, and each is downsizing by less on average. Fewer people are being laid off or are quitting their job, and firms are hiring fewer people. In other words, the employment dynamics of the U.S. economy are slower.

The decline in job creation and destruction was also the theme of this recent macroblog post by Mark Curtis, which featured some pretty nifty dynamic charts of trends in job creation and destruction by industry and geography.

Identifying the policy implications of these slower dynamics requires careful diagnosis of the causal factors underlying the trends. The cutting edge of economic research looking at this issue was featured at the 2013 Comparative Analysis of Enterprise Data Conference hosted last week by the Atlanta Census Research Data Center (ACRDC), which is housed at the Atlanta Fed and directed by one of our senior research economists, Julie Hotchkiss. Through the ACRDC, qualified researchers in Atlanta and around the Southeast can perform statistical analyses on non-public Census microdata.

The agenda and papers presented at the conference are located here. Some of the papers, I think, were particularly relevant to what President Lockhart discussed. A few examples:

Reallocation in the Great Recession: Cleansing or Not?” by Lucia Foster and Cheryl Grim of the Center for Economic Studies at the U.S. Census Bureau and John Haltiwanger at the University of Maryland looked at the so-called “cleansing hypothesis,” in which recessions are not only periods of outsized job creation and destruction, but they are also periods in which the reallocation is especially productivity enhancing. They find that while previous recessions fit this pattern reasonably well, they do not see this kind of activity in the most recent recession. In fact, they find that in the manufacturing sector, the intensity of reallocation fell rather than rose (because of the especially sharp decline in job creation), and the reallocation that did occur was less productivity enhancing than in prior recessions.

How Firms Respond to Business Cycles: The Role of Firm Age and Firm Size,” by Javier Miranda, Teresa Fort, John Haltiwanger and Ron Jarmin, looked at the varying impact of recessions on firms by size and age. They show that young businesses (which are typically small) exhibit very different cyclical dynamics than small/older businesses and are more sensitive to the cycle than larger/older businesses. The paper also explores explanations for the finding that young/small businesses were hit especially hard during the last recession. They identify the collapse in housing prices as a primary culprit, with the decline in job creation at young firms especially pronounced in states with a large drop in housing prices.

As a side note, although not presented at the conference, “The Secular Decline in Business Dynamism in the U.S.,” a new paper by Ryan Decker, John Haltiwanger, Ron Jarmin and Javier Miranda, analyzes the overall secular decline in job reallocation across industries. They find that changes in industry composition (the decline in manufacturing and rise of service industries) are not driving the decline. Instead, the primary driver seems to be the decline in the pace of entrepreneurship and the accompanying decline in the share of young firms in the economy.

Finally, Steve Davis, from the University of Chicago, talked about his joint research with John Haltiwanger, Kyle Handley, Ron Jarmin, Josh Lerner and Javier Miranda on private equity in employment dynamics, Private equity critics claim that leveraged buyouts bring huge job losses. Davis shows that private-equity buyouts are followed by a decline in net employment at these firms relative to controls (similar firms that were not targets of a buyout). However, that net change pales compared with the amount of gross job creation and destruction that typically occurs within the target firm after the buyout. In particular, he finds that in addition to reducing employment at its existing establishments, including by selling some establishments to other firms, jobs are created at new establishments within the firm via acquisition and the opening of new establishments. Moreover, they show that this reallocation is generally productivity enhancing for the firm. Although the data used in the study go only through the mid-2000s, it seems reasonable to infer from the findings that the decline in private equity deals during and since the last recession has contributed to the overall lower level of employment dynamics in this recovery.

The Comparative Analysis of Enterprise Date Conference was an excellent representation of the type of high-quality research being conducted on questions that go to the heart of the cyclical-versus-structural debate about the future course of the U.S. economy. While this is an exciting and important time for researchers in this field, it is troubling to learn that the programs that collect the data used in these types of studies are being trimmed because of federal budget cuts.

Photo of John RobertsonBy John Robertson, vice president and senior economist in the Atlanta Fed’s research department

September 23, 2013 in Economics, Employment, Unemployment | Permalink


TrackBack URL for this entry:

Listed below are links to blogs that reference The Dynamics of Economic Dynamism:


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 17, 2013

The ABCs of LFPR

As the Federal Open Market Committee (FOMC) meets amid much speculation about the next steps for monetary policy, it does so in in the context of an August 2013 Employment Situation report that was generally viewed as a mixed bag. The employment numbers undershot the consensus among most market observers, while the unemployment rate edged down again. But even the drop in the unemployment rate—a cumulative 0.8 percentage point over the past 12 months—failed to impress everyone. Martin Feldstein, for instance:

The official unemployment rate has declined sharply (to 7.3% last month from 10% in October 2009) only because so many people have stopped looking for work or are working part-time.

Part of what Professor Feldstein is referring to, of course, is the labor force participation rate (LFPR), which measures the share of the adult population that is in the labor force. LFPR includes those who are employed and those who are unemployed but looking for a job, but not those who are unemployed and are not looking for a job (which includes retirees and discouraged workers).

We generally refrain from direct commentary about issues related to monetary policy in the time surrounding FOMC meetings. I won't break with that tradition but am more than happy to highlight a resource that can help you draw your own conclusions about all things having to do with the labor market, including the LFPR.

Our Center for Human Capital Studies' Federal Reserve Human Capital Compendium is a collection of Federal Reserve System research published on topics related to employment, unemployment, and workforce development. Our latest update offers several entries that address the LFPR and its implications for the labor market. Two recent additions:

Will a Surge in Labor Force Participation Impede Unemployment Rate Improvement? Researchers at the Richmond Fed concluded that, in the short run, the LFPR and the unemployment rate are negatively correlated. This conclusion is derived from the fact that unemployed participants in the labor force are more likely to leave the labor force than those who are employed. Also, movement from unemployed non-participant to employed participant (basically skipping the unemployed-participant phase) is more likely in an improving labor market. They concluded that movements in the LFPR lag six months behind movements in the unemployment rate.

Cyclical versus Secular: Decomposing the Recent Decline in U.S. Labor Force Participation. Researchers at the Federal Reserve Bank of Boston found that since 2008, the decline in the LFPR largely reflects demographic effects of an aging population. Furthermore, the cyclical response of the LFPR during the latest recession and recovery period has been smaller than expected, so the unemployment rate would have been three-quarters of a percent lower if the LFPR had followed historical norms. They conclude that going forward, the unemployment rate should give an accurate read on labor market conditions and that further cyclical declines in the LFPR are unlikely if the labor market continues to improve.

But much more information on the LFPR and other topics including wages and earnings, outsourcing, and productivity is available. If you're looking for something to do while you await the FOMC's decision, one option is building a little human capital of your own with our Human Capital Compendium.

Photo of Whitney MancusoBy Whitney Mancuso, a senior economic analyst in the Atlanta Fed's research department

September 17, 2013 in Employment, Labor Markets | Permalink


TrackBack URL for this entry:

Listed below are links to blogs that reference The ABCs of LFPR:


"Researchers at the Federal Reserve Bank of Boston found that since 2008, the decline in the LFPR largely reflects demographic effects of an aging population" Is this a joke? Take a look at CR's graph of the participation rate of those in the 25-54 year-bracket. The rather sudden and steep decline since 2008 isn't oldsters retiring. What am I missing here?

Posted by: Fred | September 26, 2013 at 04:44 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

September 13, 2013

Job Reallocation over Time: Decomposing the Decline

One of the primary ways an economy expands is by quickly reallocating resources to the places where they are most productive. If new and productive firms are able to quickly grow and unproductive firms can quickly shrink, then the economy as a whole will experience faster growth and the many benefits (such as lower unemployment and higher wages) that are associated with that growth. Certain individuals may experience unemployment spells from this reallocation, but economists, starting with Joseph Schumpeter, have found that reallocation is associated with economic growth and wage growth, particularly for young workers.

Recently, a number of prominent economists such as John Haltiwanger have expressed concern that falling reallocation rates in the United States are a major contributor to the slow economic recovery. One simple way to quantify the speed of reallocation is to examine the job creation rate—defined as the number of new jobs in expanding firms divided by the total number of jobs in the economy—and the destruction rate, defined likewise but using the number of jobs lost by contracting firms. Chart 1 plots both the creation and the destruction rates of the U.S. economy starting in 1977. These measures track each other closely with creation rates exceeding destruction rates during periods of economic growth and vice versa during recessions. The most recent recession saw a particularly sharp decline in job creation (you can highlight the creation rate by clicking on the line), but it is clear this decline is part of a larger trend that far predates the current period. A decline in these rates could indicate less innovation or less labor market flexibility, both of which are likely to retard economic growth. Feel free to explore the measures for yourself using the figure’s interactivity.

To better understand these important trends we create a common variable called reallocation, which is defined as total jobs created plus total jobs destroyed, divided by total jobs in the economy. This formula creates one measure that describes how quickly jobs are moving from shrinking firms to expanding firms. Using data from the U.S. Census Bureau’s Business Dynamic Statistics, we examine differences in this variable across sectors and across states. Furthermore, using some basic data visualization tools, we can see how reallocation has evolved over time across these dimensions.

Chart 2 plots reallocation rates by industry from 1977 to 2011. The plot highlights the reallocation rate for all industries, but you can also select or deselect any industry to more clearly view how it has changed over time. Scrolling over the lines allows you to view the exact rates by industry in any time period. A few interesting patterns emerge. First, sectors have different levels of job reallocation in the cross section. Manufacturing stands out as having particularly low reallocation rates, probably the result of the large fixed-cost capital requirements required in production. Second, not all industries experienced sharp declines during this period. If you highlight the finance, insurance, and real estate sector, it is evident that reallocation rates actually increased for this sector until the most recent recession. Retail and construction, on the other hand, have experienced steady and significant declines during the past 35 years.

Chart 3 maps reallocation rates across states for the year 1977. This figure provides us with a cross sectional view of geographical differences in reallocation rates. States with the highest reallocation rates are dark brown, and states with the lowest rates are light brown. You can click through the years to visually capture how these rates have changed overtime for each state. Compare the color of the map in 1977 with the color in 2011. Scroll the mouse over any state to view that state’s reallocation rate in the particular year.

As with industries, states display clear cross sectional differences in their reallocation rates. The highest rates are found in western states, Florida, and Texas, and the lowest are in the Midwest. Scrolling through the years shows that the decline in reallocation rates is common to the entire country.

Overall, these figures display a stark trend. The economy is reallocating jobs at much slower rates than 20 or even 10 years ago, and this decline is, with only a few exceptions, common across states and industries. Economists are just now starting to explore the causes of this trend, and a single, compelling explanation has yet to emerge. But some explanation is clearly in order and clearly important for economic policymakers, monetary and otherwise.

By Mark Curtis, a visiting scholar in the Atlanta Fed's research department

Please note that the charts and maps in this post were updated and improved on November 27, 2013.

September 13, 2013 in Economics, Employment, Labor Markets | Permalink


TrackBack URL for this entry:

Listed below are links to blogs that reference Job Reallocation over Time: Decomposing the Decline:


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