Meet the Author

Daniel Carroll |

Research Economist

Daniel Carroll

Daniel Carroll is a research economist in the Research Department of the Federal Reserve Bank of Cleveland. His primary research interests are macroeconomics, public finance, and political economy. Currently, he is studying the implications of progressive income taxation for the distributions of wealth and income.

Read full bio

Meet the Author

Samuel B. Chapman |

Author

Samuel B. Chapman

Samuel Chapman is a former research analyst in the Research Department of the Federal Reserve Bank of Cleveland.

Read full bio

11.02.12

Economic Trends

Private Fixed Investment: Not Rebounding as Fast This Time Around

Daniel Carroll and Samuel Chapman

Investment is a key factor influencing economic growth. Investments in factories and machines, houses and computer software, all increase the capital available for production and expand the frontier of goods and services that workers can supply to the economy. As evidence, consider that investment today is strongly, positively correlated with GDP in the future. More than just expanding GDP in the future, growth in the capital stock through investment puts upward force on wages by making workers more productive.

The graph below plots the cross correlations of GDP at a given point in time with private fixed, nonresidential, and residential investments at various lead and lag times relative to it. The horizontal axis represents the distance in quarters from a time-t measurement of GDP, and the vertical axis represents the correlation of each series with GDP at time t. A large positive correlation, such as the ones below, indicates that, on average, when the series is above its trend at that lag or lead date, GDP at time t is also above its trend. This suggests that if investment is currently below its trend, GDP may also be below its trend as well.

[1.29.14: This chart was corrected and reposted.]

In the United States, private fixed investment has averaged about 15.3 percent of GDP over the postwar period; however, more recently it has run below this ratio. It crashed down to 10.5 percent in 2009 and has since hovered around 13 percent. This is unusual since investment is more volatile than income. Typically, investment will fall more than GDP during recessions, and it did in the last recession; but historically it then rebounds just as sharply. This gives the ratio of investment-to-GDP a “V-shape” over recessions. So far the most recent case has not displayed this same pattern. In contrast to previous downturns, investment has been especially slow to recover after this most recent recession.

The primary reason for this is that residential investment is still well below its previous 2006 peak. This is not surprising given the trouble associated with the run-up in housing through 2007, the subsequent decline in home prices, and the tightening of household income and credit. Still, the magnitude of the fall is remarkable. After accounting for inflation, 2012:Q3 residential investment is 58 percent less than it was in 2006:Q1.

On a positive note, real residential investment is up 12.7 percent year-over-year, so the trajectory is improving. Nevertheless, even at this rate, it will still take years to get back to precrisis levels.

Nonresidential investment also crashed during the recession, but unlike residential investment, it followed the historical pattern and rebounded sharply through 2011. From 2010:Q4 to 2012:Q1, real nonresidential investment averaged 8.6 percent growth year-over-year. Recently though, the pace has fallen off. Since the first quarter of this year, real nonresidential investment is only 4.3 percent above its level last year, as headwinds like the European sovereign debt crisis and the fiscal cliff may be causing investors to hold off on capital purchases. Hopefully, low interest rates on mortgages, improvements in household income and credit, and reductions in the uncertainty associated with fiscal headwinds will lead to a turnaround in private fixed investment and a resumption of the V-shape pattern.

[1.29.14: The figure was corrected and replaces the original.]