The Macroeconomic Consequences Of Infrastructure Investment

Valerie Ramey National Bureau of Economic Research

Date Posted:: October 5, 2020
Is Database:: Database

US infrastructure investment is not effective as short-term stimulus due to time delays & crowding out of private spending. Short-run multipliers are lower than for consumption, & long-run impact is limited by current public capital levels.

Infrastructure investment in the U.S. is not effective as a short-term stimulus due to inherent time delays and the crowding out of private spending. Short-run stimulus multipliers for government investment are lower than those for government consumption, primarily because of time-to-build delays and a higher propensity to crowd out private spending. Quantitative models and empirical studies across OECD countries and U.S. states support these findings. While long-run multipliers depend on the production function elasticity of output to public capital, current levels of public capital relative to GDP are comparable to past high points, suggesting limited immediate impact. Additionally, infrastructure spending often leads to no change or a decline in employment in the initial years, even during periods when interest rates are constrained by the zero lower bound (ZLB).

Though this might be of interest, new NBER from Valerie Ramey looks at the macro effects of government infrastructure investment and finds that it has limited short run effects versus other forms of government consumption due to time lag and crowding out, this is true even when some infrastructure investment has significant positive long runs effects,

On the short term stimulative effect“…First, even when government investment has significant long-run effects, the short run stimulus multipliers are less than those from government consumption in most situations. The two key reasons are (i) the effects of time-to-build delays and (ii) the propensity of government investment to crowd out private spending more than government consumption does.These results are supported by quantitative models, empirical panel studies across OECD countries, time series analysis in the U.S., and cross-state studies. The effects of time-to-spend and time-to-build delays, which appear to be inherent in infrastructure projects, work against the standard New Keynesian mechanisms and lower short-run multipliers. Second, the long-run multipliers on government investment depend critically on both the production function elasticity of output to public capital and on where the economy begins relative to the socially optimal level of public capital. Higher production function elasticity raises multipliers and starting far below the socially optimal level of public capital also raises multipliers….”

Valerie Ramey, “The Macroeconomic Consequences Of Infrastructure Investment,” National Bureau Of Economic Research, July 2020, https://www.nber.org/papers/w27625

Her figure 8 suggest current levels of public capital in terms of GDP are comparable to previous periods of higher investment, “…shows government capital as a percent of GDP from 1929 to 2018. The data are current-cost net stock data on government capital and nominal GDP from the BEA. The figure shows long-run trends for all government capital, nondefense government capital, and transportation capital relative to GDP. All show significant swings over time. The total government capital ratio hit peaks in the 1940s and the mid-1970s. Both the nondefense and transportation government capital ratios hit peaks in the 1930s, the mid-1970s, and in the early 2010s. The ratios have fallen only slight since the early 2010s. Thus, current levels of public capital are comparable to those of some of the past high points….”

Taking a longer view, “…Second, my review and small extension of the empirical literature on the long-run estimates suggests that theaggregate production function elasticity of output to public capital is probably between 0.065 and 0.12,…However, this elasticity is very stylized and does not take into account possible differences in the marginal products of different types of government capital. Some studies find higher estimates for core infrastructure, while others do not. Third, there is both theoretical support and some empirical support for the short-run multiplier on government investment being higher when interest rates are constrained by the zero lower bound (ZLB). The theoretical mechanisms that lead to this effect, however, also imply that at the zero lower financing government spending with distortionary income taxation rather than deficits leads to higher multipliers, a result contrary to most economists’ priors. Fourth, cross-section and panel evidence on U.S. states or counties that focuses on bridge, highway, and road infrastructure spending suggests that the spending leads to either no change or a decline in employment in the first several years, even during ZLB periods. There is no obvious explanation for these puzzling results, though the disruptive effects of construction on existing infrastructure might play a role. In sum, the macroeconomic approach to government investment provides strong support for the long-run benefits of infrastructure spending. However, the same approach raises questions about the suitability of investment in infrastructure and other public capital as a short-run stimulus….”

Inequality and Aggregate Demand

Adrien Auclert Stanford University and Northwestern University

Date Posted:: January 23, 2020
Is Database:: Database

Research by @AdrienAuclert suggests that shifts in income distribution have minimal impact on aggregate consumption, with a MPC gap of < 0.1% btw top 10% and bottom 90% earners.

Current research indicates that the marginal propensity to consume (MPC) between the top 10% of income earners and the bottom 90% is less than 0.1, suggesting that shifts in income distribution have minimal impact on aggregate consumption. For every 1% of income moving from the bottom 90% to the top 10%, aggregate consumption decreases by no more than 0.1% of total income. This small gap in MPCs implies that temporary increases in inequality do not significantly affect output in the short run. However, if inequality stems from rising individual income risk and monetary policy fails to adjust interest rates accordingly, it could lead to a substantial and prolonged economic downturn. While transitory income redistribution may slightly reduce consumption and output, the long-term effects of increased income risk could be more pronounced, potentially leading to larger economic consequences.

New Rognlie looks at income distribution's impact on aggregate demand as of today. As of now they find that the marginal propensity to consume btw rich and poor is quite small: “…average MPC of the top 10% of income earners and that of the bottom 90% of earners is less than 0.1….”

Adrien Auclert and Matthew Rognlie, "Inequality and Aggregate Demand," Stanford University and Northwestern University, January 2020, http://mattrognlie.com/inequad.pdf

“…For temporary increases in inequality, in line with common intuition, we ﬁnd that the key is the relationship between MPCs and income. But although the rich have lower MPCs than the poor, the gap is not large enough for realistic changes in the income distribution to have much effect on aggregate consumption. For instance, in both the data and our calibrated model,the gap between the average MPC of the top 10% of income earners and that of the bottom 90% of earners is less than 0.1. Hence, every additional 1% of overall income shifting from the bottom 90% to the top 10% (a larger-than-usual year on year change; see Piketty and Saez 2003) lowers aggregate consumption by no more than 0.1% of total income….We explore the transmission mechanism of income inequality to output. In the short run, higher inequality reduces output because marginal propensities to consume are negatively correlated with incomes, but this effect is quantitatively small in both the data and our model. In the long run, the output effects of income inequality are small if inequality is caused by rising dispersion in individual ﬁxed effects, but can be large if it is the manifestation of higher individual income risk….We found that transitory income redistribution can lead to declines in both consumption and output, but that this effect is likely small. By contrast, we found that the long-run effect of income inequality, if it involves an increase in idiosyncratic income risk, can potentially be quite large….”

However they find that this could change going forward, “…we generally ﬁndoutput effects that are negative but small, with one notable exception: if inequality is caused by an increase in individual income risk, and monetary policy does not or cannot lower interest rates enough to offset it, then a large, long-lasting slump can ensue…”.