10/20/2020 | News release | Distributed by Public on 10/20/2020 06:31
The COVID-19 pandemic has placed great strains on the budgets of state and local governments. In order to understand these strains and their potential consequences, it is useful to examine some of the fiscal challenges that state and local governments have faced during past economic downturns. Using these experiences as a baseline, we can analyze how the challenges currently facing state and local governments are likely to play out. The current downturn is likely to mirror certain aspects of historical downturns, but it will undoubtedly differ in important respects.
What's at Stake?
State and local governments are significant players in the U.S economy. At the beginning of 2020, they employed approximately 20 million workers, which represented 13 percent of total U.S. employment. In 2017, their expenditures totaled $3.1 trillion or approximately 15 percent of U.S. gross domestic product. Local governments spent just over half of this amount, with their largest expenditure shares going to K-12 (40 percent), health care (10 percent), and police services (6 percent). State governments spent a modestly smaller sum of money, with the largest shares devoted to public welfare (43 percent), higher education (18 percent), and health services (9 percent).
Deep recessions cast long shadows on state and local government tax revenues. The Great Recession is a case in point. From 2007 to 2010, taxes collected by state and local governments declined 5 percent - from $1.52 trillion to $1.44 trillion. Tax revenues did not fully recover to pre-recession levels until 2013.
Revenue drops of this magnitude have substantial effects on those in our society who are least well off. State and local governments are typically required by state laws to balance their budgets regularly, so when their revenues fall, they often have no recourse but to cut spending. When this belt tightening happens, state and local governments almost always end up paring down on essential services, including education, health care, and social welfare programs. Research tells us that these kinds of funding disruptions create long-lasting damage to individuals' earnings.
Cyclically-Sensitive Tax Revenues
State and local government tax revenues are tied to macroeconomic outcomes. For both levels of government, tax revenues increase substantially during economic booms, such as the one we observed between 2003 and 2007. (See charts, below.) But state tax revenues are more cyclically sensitive than local tax revenues, mostly because states are relatively more reliant on income and sales tax revenues, which are highly correlated with macroeconomic activity. Indeed, state tax revenues declined substantially during the recessions of 2001 and 2008-09 (denoted by the shaded areas in the charts). Local tax revenues tend to be less sensitive to economic downturns, and they tend to react to recessions with an 18-24 month lag. This somewhat muted and delayed response partly reflects local governments' relatively heavy reliance on property tax revenues, which are based on assessed property values that tend to be adjusted with a delay. As a result, local government revenues began to decline only toward the end of the 2008-09 recession and continued to decline well after the recession was over.
During the Great Recession, the decline in tax revenues experienced by state governments was just about fully offset by increased federal intergovernmental grants (IG). For local governments, however, total IG funding actually declined when it was most needed as local tax receipts declined in the aftermath of the recession. By 2017, total IG support for local governments was no higher than it was prior to the Great Recession, and it appears that local governments have become increasingly reliant on their own sources of revenue.
The current downturn's effect on local revenues is likely to vary by locality - with some cities being more vulnerable to the coronavirus shock than others. One would expect that local revenue would fall more sharply, and perhaps for a longer time, in cities that are relatively reliant on cyclically-sensitive revenue sources and in cities that have relatively large shares of employment in high-risk industries, such as food service, hospitality, and retail trade.
The chart below plots data for 139 U.S. cities, including many of the largest cities in the Fifth Federal Reserve District.1 For each city (represented by a dot), the share of general fund revenues from cyclically-sensitive sources (mostly sales and income taxes) is shown on the horizontal axis, while the city's exposure to high-risk industries in terms of employment is shown on the vertical axis. A key takeaway is that the cities of the Fifth District are not particularly reliant on cyclically-sensitive sources of tax revenue. Among Fifth District cities, Baltimore, Md., and Newport News, Va., have the most cyclically-sensitive tax revenues, but their sensitivity is still modest relative to many U.S. cities.
Another takeaway is that Fifth District cities' exposure to high-risk industries ranges between 10 percent and 20 percent, which is similar to the range seen for the full sample of 139 U.S. cities. Durham, N.C., is the least exposed in the Fifth District, while Charlotte, N.C., and Greensboro, N.C., are the most exposed. In the cases of Charlotte and Greensboro, the difficulties posed by their exposure to high-risk industries is likely to be mitigated somewhat by their nonreliance on cyclically-sensitive revenue sources.
So what can we expect for local tax revenue? For most local governments, budgets for the 2020 fiscal year only capture a few months of the COVID-19 pandemic recession. Current estimates from the National League of Cities project that revenues will not grow in 2020. Local government budgets already anticipate a large decline in sales tax revenue (11 percent) and a more modest decline in income tax revenue (3.4 percent). However, property taxes are projected to increase 1.9 percent in 2020. As expected based on past economic cycles, any effect that the crisis will have on property taxes is likely to be seen with a lag.
Next year's budgets are expected to more fully capture the fiscal effects of the COVID-19 pandemic. Cities are forecast to see a 13 percent decline in revenues in 2021, although it is not yet clear how commercial property tax revenues will be affected by near-term business closures and longer-term changes in the demand for office space. Moreover, it is uncertain how the expiration of eviction moratoriums will affect the value of multifamily housing units or how the crisis will ultimately affect the value of owner-occupied residential property.
Forced Spending Cuts
States and local governments tend to respond to revenue declines by cutting social services and canceling or postponing capital projects. And these austerity measures tend to involve payroll reductions through layoffs, furloughs, and hiring freezes. Following the Great Recession, for example, we saw large cutbacks in local public employment - which did not recover to pre-recession levels until 2019. As a result of the COVID-19 pandemic, we are now seeing large declines in local public employment, but the changes are more abrupt and on a larger scale than during past recessions.
Evidence from the previous recession also suggests that overall education funding will decline sharply due to COVID-19. Public schools are mainly funded by state and local governments, in almost equal proportions. As revenues fall, resources devoted to education will likely decline as well.
Based on the experiences of the Great Recession and its aftermath, the impact of economic downturns on school budgets may be long-lasting. Spending per student in elementary and secondary schools started to decline in 2009 and did not fully recover to pre-recession levels until 2016.
Educational funding disruptions of this kind matter. Recent research that has examined the effects of the decline in educational spending during the Great Recession has found that cohorts that were exposed to the cuts had worse educational outcomes. Also, these cuts appear to have contributed to an increase in the test score gap between black and white students.
Other work offers some hints on the possible effects of a different source of educational disruption: school lockdowns. This research finds that less instruction time due to fewer school days tends to adversely affect the development of cognitive skills, lower students' performance in standardized tests, and ultimately lower their future earnings. The literature also indicates that parental inputs and online learning cannot fully replace face-to-face instruction and that this problem is more severe for students from disadvantaged backgrounds.
Quantifying the effects of educational disruptions, the evidence suggests that when spending goes down by $1,000 per student, several bad things happen: Average test scores in math and reading fall by 3.9 percent of a standard deviation, the test score gap between black and white students increases by 6 percent, and the college-going rate declines by 2.6 percent.2 Another study found that a 12-week school closure (similar to the one we experienced recently) could reduce test scores in math by 9 percent of a standard deviation.3
Socioeconomic gaps in access to online learning resources also increased during the COVID-19 pandemic. A recent study that used Google search data found that search intensity rose twice as much in areas with above-median socioeconomic status compared to other areas, and the increase in search intensity was lower in areas with more rural schools and black students.4
New Evidence on Fiscal Multipliers
So what can fiscal policy accomplish in this kind of environment? A recent strand of literature in economics attempts to estimate the size of fiscal multipliers by focusing on the local effects of fiscal policy. Based on this research, economists have estimated that the cost of creating a local job could be between $22,000 and $75,000, depending on the type of policy adopted to spend the money.5 These estimates of the cost per job created are lower than the results of previous studies. Consequently, they could change the cost-benefit analysis of various public policy options - perhaps tipping the scales in favor of increased fiscal spending.
Additional studies may also alter the cost-benefit analysis of state and local expenditures. For example, recent research has found that local spending shocks tend to affect nearby localities through what are called 'spatial spillover' effects, which would tend to increase the societal benefits associated with certain expenditures. And other research has found that fiscal multipliers are substantially larger in cities with higher unemployment and higher consumer debt-to-income ratios.
Relevant Now and for the Long Run
State and local governments are responsible for a growing share of public service provision in the United States. The fact that these governments are limited in their ability to fully smooth expenditure when revenue drops means that, in the absence of sizable federal transfers, a large amount of local government services will be disrupted by the current economic downturn. State and local spending is critical for delivering human capital, but a key feature of human capital is that skills beget skills. Consequently, disruptions now are likely to cause significant long-term value losses. The considerable impact of state and local government spending on everyday lives makes it clear that it is not only relevant for the recovery, but also for the long run.
Have a question or comment about this article? We'd love to hear from you!
Views expressed are those of the authors and do not necessarily reflect those of the Federal Reserve Bank of Richmond or the Federal Reserve System.
Pagano, Michael A., and Christiana K. McFarland. 'When will your city feel the fiscal impact of COVID-19?' Brookings Institute: The Avenue, March 31, 2020.
Jackson, C. Kirabo, Cora Wigger, and Heyu Xiong. 'Do School Spending Cuts Matter? Evidence from The Great Recession.' National Bureau of Economic Research Working Paper No. 24203, January 2018.
Lavy, Victor. 'Do Diﬀerences in Schools' Instruction Time Explain International Achievement Gaps? Evidence from Developed and Developing Countries.'Economic Journal, Nov. 2015, vol. 125, no. 588, pp. F397-F424.
Bacher-Hicks, Andrew, Joshua Goodman, and Christine Mulhern. 'Inequality in Household Adaptation to Schooling Shocks: Covid-Induced Online Learning Engagement in Real Time.' National Bureau of Economic Research Working Paper No. 27555, July 2020.
Bartik, Timothy J. 'New Evidence on State Fiscal Multipliers: Implications for State Policies.' Upjohn Institute Working Paper No. 17-275, July 2017.