Local Fiscal Multipliers and Fiscal Spillovers in the USA

Abstract

We estimate local fiscal multipliers and spillovers for the USA using a rich dataset based on the US Department of Defense contracts and a variety of outcome variables relating to income and employment. We find strong positive spillovers across locations and industries. Both backward linkages and general equilibrium effects (e.g., income multipliers) contribute to the positive spillovers. Geographical spillovers appear to dissipate fairly quickly with distance. Our evidence points to the relevance of Keynesian-type models that feature excess capacity.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6

Notes

  1. 1.

    There also is some ambiguity regarding their finding that government spending causes a reduction in local investment (Snyder and Welch 2017). Therefore, the effect of government spending on local private activity remains a very open question.

  2. 2.

    While our findings do not directly address the possibility of crowding out in other states, the fact that spillovers, positive or negative, appear to die out rather quickly as distance increases within states suggests that such effects may be small.

  3. 3.

    Recent work has also explored the effects of regional variation in fiscal policy in developing countries. Corbi et al. (2019), for example, document large regional effects of transfer shocks in Brazil.

  4. 4.

    Our dataset is an updated version of the data used in Demyanyk et al. (2019).

  5. 5.

    Apart from salaries and wages, the earnings include bonuses, stock options, profit distributions, and some fringe benefits such as cash value of meals and lodging.

  6. 6.

    According to the US Office of Management and Budget, a core-based statistical area (CBSA) is one or more counties (or equivalents) with an urban center of at least 10,000 people. Counties adjacent to the urban central are tied to the urban center by commuting.

  7. 7.

    For confidentiality reasons, QCEW does not report employment and earnings for an industry/location/period cell that may reveal information about specific firms. As a result, some of these series may be interrupted by spells of missing values.

  8. 8.

    For example, county or city borders and industry definitions are occasionally revised, which can result in large variation of, e.g., industry-level employment that is not related to any kind of fundamental variation in local economic conditions.

  9. 9.

    Interestingly, stock prices started to move before the announcement: there was a sharp decline of Boeing’s stock price and a sharp increase in Lockheed Martin’s in mid September 2001. The average stock price for Boeing in November 2001 was 35% lower than it was in September 2001, while the average stock price for Lockheed Martin increased by 23% over the same period.

  10. 10.

    Denton County had the same per-worker spending as Dallas. Other counties have negligible amounts of contracts awarded by the Department of Defense.

  11. 11.

    When we calculate \( \tilde{G}_{i,\ell ,t}^{\left( 2 \right)} \), we set \( \omega_{i \to i} = 1 \) because we include \( G_{i,\ell ,t} \) as a separate regressor to isolate general equilibirum effects.

  12. 12.

    The term “wealth transfer” here is the government-to-firm analog of the county-to-country use of the term in the international context (e.g., Gourinchas et al. 2012). An alternative term with similar meaning is “non-flow adjustments” (Obstfeld 2012). More generally, we use the term to refer to any exchange or transfer of assets that is not associated with additional production.

  13. 13.

    We experimented with alternative approaches to compute sampling uncertainty in the estimated multipliers, including the Driscoll–Kraay standard error correction. We found that this approach had a mixed impact on calculated standard errors, increasing them in some of our specifications and reducing them in others. We focus on our results based on clustering by state as we do not see a compelling case for going beyond the within-state correction. See Abadie et al. (2017) for further discussion of the process one should follow in determining the appropriate correction for standard errors.

  14. 14.

    To estimate responses of employment to military spending shocks, we use percent change of employment growth as the dependent variable in specification (1) and change in military spending normalized by earnings as the regressor in specification (1).

  15. 15.

    Recall that we distribute outlays over the length of contracts, so that, even if a locality receives only one contract during a period of several quarters or years, there will typically be a change in measured spending over several periods after the initial contract date.

  16. 16.

    While we consider neighboring cities only within states, our findings that the magnitude of spillovers decline sharply with distance suggests that extending the analysis to include cross-state spillovers would have little impact on the pattern.

  17. 17.

    The corresponding state-wide spillover effect on earnings is 0.424 (standard error = 0.109), which is even larger relative to the local multiplier for earnings.

  18. 18.

    The number of observations in each sample varies based on the number of identifiable neighbor cities for each distance category.

  19. 19.

    Recall that cities encompass metropolitan areas, not simply the central cities themselves.

  20. 20.

    Median income of cities in the bottom decile of the size distribution is roughly 1/4 of median income for the sample, while median income of cities in the top size decile is 21 times the sample median.

  21. 21.

    The evidence of positive spillovers and size-dependent multipliers implies that locations benefit from proximity to locations that receive DOD spending. An open question is whether the inverse is true. Does a recipient city benefit from proximity to other cities? In other words, are multipliers higher in cities that are isolated or in cities surrounded by economic activity? Proximity could be expected to lead to higher or lower multipliers. Subcontractors in nearby cities may compete with subcontractors in locations that receive the DOD spending, thereby dampening local income effects. But nearby cities may also provide necessary inputs into production of local contractors, especially when recipient locations do not produce such inputs locally. Nearby cities may also increase local multipliers through consumption effects, with DOD spending in a recipient location inducing higher spending in a nearby city, which may have a second-round positive effect on the recipient location. To explore the effects of proximity, we derived a city-specific measure of nearby economic activity by summing income (discounted by distance) across other cities. We found that conditional on city size multipliers appear to be slightly larger in the presence of nearby economic activity. We conjecture that a more detailed exploration of this issue would be a fruitful avenue for future research.

  22. 22.

    Although there is no obvious scaling to use in computing general equilibrium effects, this lack of scaling for the variable should be kept in mind when comparing the coefficients for this spillover effect and the one based on backward production linkages. A one-dollar shock to the production linkage variable would have a one-dollar spillover if the affected industry satisfies all of the increased input demand of the shocked industry. But, a one-dollar shock’s effect on other industries simply depends on the shock’s overall general equilibrium effects.

  23. 23.

    The only evidence of negative spillovers is in the effect of spending 100–150 miles away on local same-industry GDP (column 5, Panel B).

  24. 24.

    Our body of evidence suggests that the economy features slack (and hence private activity is not crowded out by public purchases) on average over the business cycle. A natural follow-up question is whether multipliers are higher during periods of higher slack. Even if multipliers are slack-dependent, detecting this dependence in our setting is challenging due to the fact that slack is endogenous to government spending, especially at the annual frequency: if government spending is effective, then measured slack is low. The literature using subnational variation generally finds state-dependent multiplies although there is variation across studies (see Chodorow-Reich 2019). Consistent with this literature, we found that multipliers are substantially higher for cities experiencing unemployment above the 25th percentile level. We did not find evidence that the strength of multipliers increases much as slack increases beyond the 25th percentile (Appendix Table 8). One interpretation of these results is that multipliers are high when there is some slack, but beyond that the amount of slack does not matter (perhaps short-run aggregate supply curves are relatively flat and rapidly steepen only as the economy approaches capacity). The fact that the threshold level of unemployment is around the 25th percentile suggests that multipliers on average are high and do not lead to crowding out. This is consistent with our industry-level results that on average (across the business cycle) there is no systematic evidence of crowding out.

  25. 25.

    Murphy (2017), for example, predicts that in the presence of excess capacity, temporary spending shocks by a rich trading partner can have persistent and large effects on income for both the rich agent and its poorer trading partners.

References

  1. Abadie, Alberto, Susan Athey, Guido W. Imbens, and Jeffrey Wooldridge. 2017. When Should You Adjust Standard Errors for Clustering? NBER Working Paper 24003.

  2. Auerbach, Alan J., and Yuriy Gorodnichenko. 2012. Measuring the Output Responses to Fiscal Policy. American Economic Journal: Economic Policy 4(2): 1–27.

    Google Scholar 

  3. Auerbach, Alan J., and Yuriy Gorodnichenko. 2013. Output Spillovers from Fiscal Policy. American Economic Review 103(2): 141–146.

    Article  Google Scholar 

  4. Auerbach, Alan J., and Yuriy Gorodnichenko. 2016. Effects of Fiscal Shocks in a Globalized World. IMF Economic Review 64(1): 177–215.

    Article  Google Scholar 

  5. Auerbach, Alan J., and Yuriy Gorodnichenko. 2017. Fiscal Stimulus and Fiscal Sustainability. In Federal Reserve Bank of Kansas City, Fostering a Dynamic Global Economy, 217–270.

  6. Auerbach, Alan J., Yuriy Gorodnichenko, and Daniel Murphy. 2019. Macroeconomic Frameworks. NBER Working Paper 26365.

  7. Autor, David H., David Dorn, and Gordon H. Hanson. 2016. The China Shock: Learning from Labor-Market Adjustment to Large Changes in Trade. Annual Review of Economics 8(1): 205–240.

    Article  Google Scholar 

  8. Bartik, Timothy. 1991. Who Benefits from State and Local Economic Development Policies?. Kalamazoo: W.E. Upjohn Institute.

    Google Scholar 

  9. Blanchard, Olivier, Christopher J. Erceg, and Jesper Lindé. 2016. Jump-Starting the Euro Area Recovery: Would a Rise in Core Fiscal Spending Help the Periphery? NBER Macroeconomics Annual 2016: 103–182.

    Google Scholar 

  10. Chodorow-Reich, Gabriel. 2019. Geographic Cross-Sectional Fiscal Spending Multipliers: What Have We Learned? American Economic Journal: Economic Policy 11(2): 1–34.

    Google Scholar 

  11. Chodorow-Reich, Gabriel, Laura Feiveson, Zachary Liscow, and William Gui Woolston. 2012. Does State Fiscal Relief During Recessions Increase Employment? Evidence from the American Recovery and Reinvestment Act. American Economic Journal: Economic Policy 4(3): 118–145.

    Google Scholar 

  12. Clemens, Jeffrey, and Stephen Miran. 2012. Fiscal Policy Multipliers on Subnational Government Spending. American Economic Journal: Economic Policy 4(2): 46–68.

    Google Scholar 

  13. Cohen, Lauren, Joshua Coval, and Christopher Malloy. 2011. Do Powerful Politicians Cause Corporate Downsizing? Journal of Political Economy 119(6): 1015–1060.

    Article  Google Scholar 

  14. Corbi, Raphael, Elias Papaioannou, and Paolo Surico. 2019. Regional Transfer Multipliers. Review of Economic Studies 86(5): 1901–1934.

    Article  Google Scholar 

  15. Demena, Binyam A., and Peter A.G. van Bergeijk. 2017. A Meta-Analysis of FDI and Productivity Spillovers in Developing Countries. Journal of Economic Surveys 31(2): 546–571.

    Article  Google Scholar 

  16. Demyanyk, Yuliya, Elena Loutskina, and Daniel Murphy. 2019. Fiscal Stimulus and Consumer Debt. Review of Economics and Statistics 101(4): 728–741.

    Article  Google Scholar 

  17. Dupor, Bill, and Peter B. McCrory. 2018. A Cup Runneth Over: Fiscal Policy Spillovers from the 2009 Recovery Act. Economic Journal 128(611): 1476–1508.

    Article  Google Scholar 

  18. Gorodnichenko, Yuriy, Jan Svejnar, and Katherine Terrell. 2014. When Does FDI Have Positive Spillovers? Evidence from 17 Transition Market Economies. Journal of Comparative Economics 42(4): 954–969.

    Article  Google Scholar 

  19. Gourinchas, Pierre-Olivier, Helene Rey, and Kai Truempler. 2012. The Financial Crisis and the Geography of Wealth Transfers. Journal of International Economics 88(2): 266–283.

    Article  Google Scholar 

  20. Hall, Robert E. 2009. By How Much Does GDP Rise If the Government Buys More Output? Brookings Papers on Economic Activity 2009(2): 183–231.

    Article  Google Scholar 

  21. Javorcik, Beata Smarzynska. 2004. Does Foreign Direct Investment Increase the Productivity of Domestic Firms? In Search of Spillovers through Backward Linkages. American Economic Review 94(3): 605–627.

    Article  Google Scholar 

  22. Leeper, Eric M., Nora Traum, and Todd B. Walker. 2017. Clearing Up the Fiscal Multiplier Morass. American Economic Review 107(8): 2409–2454.

    Article  Google Scholar 

  23. Michaillat, Pascal, and Emmanuel Saez. 2015. Aggregate Demand, Idle Time, and Unemployment. Quarterly Journal of Economics 130(2): 507–569.

    Article  Google Scholar 

  24. Miyamoto, Wataru, Thuy Lan Nguyen, and Viacheslav Sheremirov. 2019. The Effects of Government Spending on Real Exchange Rates: Evidence from Military Spending Panel Data. Jounal of International Economics 116: 144–157

    Article  Google Scholar 

  25. Murphy, Daniel P. 2015. How Can Government Spending Stimulate Consumption? Review of Economic Dynamics 18(3): 551–574.

    Article  Google Scholar 

  26. Murphy, Daniel P. 2017. Excess Capacity in a Fixed-Cost Economy. European Economic Review 91(1): 245–260.

    Article  Google Scholar 

  27. Murphy, Daniel P., and Kieran Walsh. 2018. Government Spending and Interest Rates. New York: Mimeo.

    Google Scholar 

  28. Nakamura, Emi, and Jón Steinsson. 2014. Fiscal Stimulus in a Monetary Union: Evidence from U.S. Regions. American Economic Review 104(3): 753–792.

    Article  Google Scholar 

  29. Obstfeld, Maurice. 2012. Does the Current Account Still Matter? American Economic Review 102(3): 1–23.

    Article  Google Scholar 

  30. Ramey, Valerie A. 2011. Identifying Government Spending Shocks: It’s All in the Timing. Quarterly Journal of Economics 126(1): 1–50.

    Article  Google Scholar 

  31. Snyder, Jason Alan, and Ivo Welch. 2017. Do Powerful Politicians Really Cause Corporate Downsizing? Journal of Political Economy 125(6): 2225–2231.

    Article  Google Scholar 

  32. Suárez Serrato, Juan Carlos, and Philippe Wingender, 2016. Estimating Local Fiscal Multipliers. NBER Working Paper 22425.

Download references

Author information

Affiliations

Authors

Corresponding author

Correspondence to Alan Auerbach.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendix

Appendix

See Table 8.

Table 8 State dependence of local multipliers

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Auerbach, A., Gorodnichenko, Y. & Murphy, D. Local Fiscal Multipliers and Fiscal Spillovers in the USA. IMF Econ Rev 68, 195–229 (2020). https://doi.org/10.1057/s41308-019-00102-3

Download citation

JEL Classifications

  • E62
  • H5