The Economic Recovery Continues, but State Finances Remain Weak

Fiscal conditions for states were somewhat mixed in the 2014 fiscal year as state general fund revenue growth declined due to the impact of the federal fiscal cliff, while total state spending growth accelerated due to increased federal Medicaid funds from the Affordable Care Act. The number of states making midyear budget cuts remained low and states maintained stable rainy day fund levels. In the 2015 fiscal year, states are expecting both revenue and spending to grow slowly, but below the historical rate of growth. It is likely that budget proposals for the 2016 fiscal year and beyond will remain mostly cautious with limited spending growth.

  Download the Article in PDF / E-Reader Compatible Format

About the Author

Donald J. Boyd is a senior fellow at the Rockefeller Institute of Government and the former director of the Institute’s State and Local Government Finance research group. Boyd has over three decades of experience analyzing state and local fiscal issues, and has written or coauthored many of the program’s reports on the fiscal climate in the 50 states. His previous positions include executive director of the State Budget Crisis Task Force created by former Federal Reserve Board Chairman Paul Volcker and former New York Lieutenant Governor Richard Ravitch, director of the economic and revenue staff for the New York State Division of the Budget, and director of the tax staff for the New York State Assembly Ways and Means Committee. Boyd holds a Ph.D. in managerial economics from Rensselaer Polytechnic Institute. Lucy Dadayan is a senior policy analyst at the Rockefeller Institute and contributed to the work of the State Budget Crisis Task Force throughout its first phase. Dadayan holds a Ph.D in Informatics from the University of Albany.

The national economic recovery has been underway for six years, yet some states have announced budget shortfalls recently and several are even contemplating tax increases. Why are state finances so weak in the midst of recovery and what does it mean for state policy choices? This economic recovery has been slower than past recoveries. Employment is one of the most important economic indicators for state budget officials. It is a broad measure of the overall economy, and it plays a major role in determining wages subject to state income taxes and in supporting consumer purchases subject to state sales taxes. Employment has been recovering continuously since late 2010 with occasional slowing and acceleration.

While the extended employment recovery has been welcome news for states, it is slow by historical standards. Seven years after the start of the recession, employment is only 2 percent above its prior peak, compared to 3.7 percent at this point after the 2001 recession, and more than 12 percent for each of the three prior recessions.1

Consumer expenditures are important to state sales taxes, but the consumption recovery has been slower than the employment recovery relative to past recessions. Seven years after the recession’s start, inflation-adjusted consumption is only 10 percent above its prior peak, compared to more than 20 percent for each of four major previous recessions.

The national data mask growth in some states and decline in others. Employment ranges from 30 percent above the recession’s start in North Dakota—which benefited from an oil boom since gone bust—to 4.4 percent below in Nevada, which was devastated by the recession but is now recovering, although it hasn’t reached its prior peak. Employment in 12 states remains below the prerecession peak of seven years ago. 

The tax revenue recovery is slower than previous tax recoveries. The weak economic recovery has caused tax revenue to be weak by historical standards. Seven years after the start of the recession, inflationadjusted state government tax revenue is only 5 percent above pre-recession revenue, whereas in four major preceding recoveries, inflation-adjusted state tax revenue by this point ranged from 15 to 25 percent above pre-recession revenue.2 

The corporate income tax, which plays a minor role in most states’ revenue structures, is more than 15 percent below its 2007 level. The sales tax has been the weakest of the major taxes, reflecting the slow growth in consumption. The personal income tax, while stronger than the sales tax, has been held back by a huge decline in capital gains, which has recovered only partially. Other taxes have grown substantially, reflecting legislated increases in taxes on cigarettes, motor fuel and other items. Figure F shows the path of revenue by tax type since the start of the recession.

State income taxes have been affected acutely by capital gains, which do not appear in traditional measures of the economy. Between 2007 and 2009, capital gains fell by 72 percent. They have since more than doubled, gyrating substantially in 2012 and 2013 as taxpayers moved money between tax years in response to expected and actual changes in federal tax rates.3 Despite increases in 2010 through 2014, capital gains remain about a third below its 2007 peak, contributing to the relatively low level of the personal income tax.

In 2012, approximately three-quarters of net capital gains were claimed by the one-quarter of 1 percent of tax filers who had adjusted gross income of $1 million or more.4 Thus, virtually all capital gains are taxed at the highest rates unless a state has a tax-rate preference for gains; decisions by relatively few taxpayers can have a large impact on state tax revenue. This volatility has been a source of frustration to many state revenue forecasters and a cause of budget forecasting errors.5 

State fortunes have varied, reflecting differences in economies, tax structures and tax policy choices. Inflation-adjusted tax revenue is up by 2.1 percent in the median state in the seven years since the start of the recession, but it ranges from a near-tripling in North Dakota to a decline of 57 percent in Alaska. North Dakota revenue was driven upward by its oil boom, while declines in Alaskan oil production and cuts in petroleum taxes drove Alaska tax revenue down. Inflation-adjusted tax revenue remains lower than its level at the start of the recession in 21 states, with Southeastern states faring particularly poorly (Figure H). It is not easy for elected officials to bring budgets in line with revenue that is lower than it was seven years ago. 

Little support for tax increases

Although the Great Recession drove state tax revenue down more than any other recession since the Great Depression, it did not lead to the largest tax increases. Table A shows state government legislated tax changes for boom and bust periods of the past 25 years, adjusted for inflation.6 States raised taxes by $33 billion in the five years of greatest response to this recession, 38 percent less than the $54 billion raised in response to the 1990 recession.

The recent increases were greater than the $24 billion raised after the much milder and briefer 2001 recession, but that doesn’t tell the full story. Very few states raised taxes significantly in the recent recession, with the top three—California, Illinois and New York—accounting for 81 percent of the total, compared to 50 percent for the top three responses to the 1991 recession. The next 47 states raised taxes by $12.1 billion in the 2001 response, versus only $6.4 billion in the 2007 response. Thus, large tax increases actually were more common after the mild 2001 recession than after the deep 2007 recession. This declining reliance on large tax increases is shown in Figure I. Elected officials and the electorate have become extremely reluctant to rely on tax increases to close budget gaps. Public opinion polls and recent state government behavior do suggest that Americans are more likely to support taxes for transportation purposes.7

Spending pressures continue, but their character is changing. A sea change in state and local government spending The U.S. Bureau of Economic Analysis publishes estimates of aggregate state and local government spending on social benefits, approximately 80 percent of which is Medicaid; consumption, which includes spending for services such as teachers, police, firefighters and administrators; and investment, including construction spending on infrastructure, buildings and other investments. Investment data include gross investment expenditures and net investment after allowing for “capital consumption” to reflect the fact that assets generally are used up over time (e.g., roads and bridges deteriorate in quality). The bureau does not regularly break state and local government expenditure data into state government spending and local government spending. Figure J shows inflation-adjusted social benefit, consumption and investment spending from 2000 through 2014, relative to the fourth quarter of 2007 when the recession began.8 Real gross investment fell by 15 percent since the start of the recession and net investment declined more than 45 percent. Social benefit spending has risen by about 25 percent, while consumption spending on many of the bread-and-butter services of government has declined slightly. This is a dramatic change from the previous five years. Much of the recent decline in investment spending appears to reflect a decline in school building construction (see next section) rather than traditional infrastructure investment. Spending on transportation and water infrastructure, a large majority of infrastructure spending, has been relatively stable at about 2.5 percent of gross domestic product since the early 1980s.9  

Nonetheless, the recent sharp shift in investment expenditures indicates a very substantial change in state and local government spending. Sharp declines in construction spending, particularly for school buildings. According to U.S. Census Bureau data, inflation-adjusted state and local government construction spending declined by $48 billion at annual rates—15.6 percent—between the fourth quarters of 2007 and 2014. Spending declined, or increased only negligibly, in every category other than bridges and mass transit (Table B).

Spending on primary and secondary education buildings accounted for more than half of the decline. This may reflect changing demographics. Growth in the number of primary and secondary education pupils slowed from 0.8 percent annually in 1996–2006 to 0.1 percent annually in 2006–13. High school enrollment also has declined for five consecutive years. The National Center for Education Statistics projects that the number of pupils will rise 0.6 percent annually from 2015 through 2023.10

Declines in most spending other than Medicaid.
State governments cut spending in most functional areas. Table C shows inflation-adjusted state government expenditures in 2008 before the federal stimulus package took effect, five years later in 2013 (latest available year), and five years earlier. The third from right column shows the dollar change in spending between 2008 and 2013, reflecting policy choices states made in response to the recession. The “Exhibit” block at the bottom shows selected groupings of expenditures included above. Total general expenditures rose $190 billion —13.3 percent—between 2003 and 2008, and then slowed as states responded to the recession, rising $60.8 billion—3.7 percent—between 2008 and 2013. Spending on higher education, medical vendor payments—similar in concept to Medicaid—and health and hospitals rose by $118.4 billion, or 17.1 percent, between 2008 and 2013; all other spending declined by $57.6 billion, or 6.2 percent. Pension contributions, while small relative to state budgets in aggregate, accounted for substantial spending growth in both periods.11

Spending on higher education includes spending from state support and from tuition and fees. Even though state governments cut their direct support in response to the recession, public higher education institutions raised tuition and fee revenue to help meet increased demand.12, 13 

Medicaid expenditures were driven upward primarily by recession-related increases in enrollment.14 Medicaid is largely an entitlement and as a practical matter, states needed to fund the spending at the expense of cuts elsewhere or higher taxes. States had substantial assistance early in the recession from the federal stimulus package. States cut inflation-adjusted spending in almost every other significant area, despite growth in student enrollment and populations served by government programs.

States cut employment as they reduced spending. According to data from the U.S. Bureau of Labor Statistics, in the past five and a half years, states have cut non-education employment by 5.8 percent, compared with an increase in past recoveries of 18.7 percent for the typical, or median, recovery.15

More-detailed Census Bureau data show that between 2009 and 2013, states increased employment in higher education and health, but cut employment in other areas by 6.5 percent. Almost every major area was cut. The largest cuts were to corrections at 10.4 percent, which may reflect changes in workload. The state prisoner population peaked in 2009, then fell 3.4 percent by 2013, reflecting changes in policies for sentencing, parole and probation, among other things.16 Special circumstances in many states. A surprising number of states face special circumstances, most of which are contributing to fiscal stress. Some states face several of the stresses described below.

Connecticut, Kansas and New Jersey are struggling with the aftermath of well-publicized income tax revenue shortfalls at the end of the 2014 fiscal year that threw their 2015 budgets out of balance. Efforts to balance these budgets relied disproportionately on non-recurring revenue, in turn creating difficulties for their 2016 budgets. 

Kansas cut taxes sharply in 2012 and 2013. In addition to the April 2014 income tax shortfall, it has had several rounds of subsequent significant revenue shortfalls. It faces difficult choices about the extent to which it should cut spending or modify elements of the tax cuts.

Illinois, Kentucky, New Jersey and Pennsylvania are struggling to accommodate increasing pension contributions that are required due to investment shortfalls and years of contribution underpayments.17 Falling oil prices are a threat to the finances of several oil producing states, either directly through their impact on severance taxes, or through their impact on the broader state economy. Alaska, Louisiana, New Mexico and Oklahoma are addressing revenue shortfalls or slowing revenue growth related to the drop in the price of oil. While North Dakota and Texas are also oil-rich states, the falling oil prices have had less impact on these states. Texas has a wider and more diverse economy and is not solely reliant on the energy sector. And in North Dakota, the general fund is well insulated from the fluctuations of oil price and production. By law, only $300 million of oil tax revenue is allowed to flow through to the general fund. The rest of the oil and gas tax collections are distributed to oil producing counties and tribes, as well as to several constitutional and statutory funds. 

Looking ahead

State finances are tied closely to the economy. Major economic forecasters expect that the national economy will continue to improve throughout 2015 and 2016; inflation will be low, but gradually rising, and interest rates will rise slightly.18 The Congressional Budget Office’s January 2015 forecast is in line with this consensus (see Table D).19 This forecast suggests relatively slow growth in tax revenue. Income taxes generally grow faster than the economy. As incomes rise due to inflation and productivity increases, more income is taxed at higher rates. Capital gains and other financial market income can alter income tax growth substantially, but CBO forecasts that capital gains will grow only slightly faster than income in 2015, and more slowly in later years.20 

Retirement income—including federally taxable portions of pensions, Social Security income and IRA distributions—is another potential source of faster growth. It has been driven upward by an aging population and, in the case of IRA distributions, growth in financial assets. Between 2007 and 2012 wages grew by 7 percent, but these income sources grew by 35 percent.21 However, most states exempt some of this federally taxable income in an apparent effort to lure seemingly footloose senior citizens, despite the absence of evidence that state income tax breaks affect elderly migration.22, 23 Illinois, arguably the most fiscally stressed state, exempts all retirement income from tax, including IRA distributions.24

General sales taxes have been growing more slowly than the economy for more than 40 years.25 This is a result of the difficulty in taxing services and of collecting tax on Internet sales, among other things. Slow sales tax growth is likely to continue. Selective sales taxes and licenses—such as taxes on cigarettes, motor fuel and alcohol—usually are based on the quantity of the good sold (for example, 15 cents per gallon of gas or $3 per pack of cigarettes), rather than on the sales price. Two of the largest selective sales taxes, motor fuel taxes and cigarette taxes, have long-term downward trends. Fuel economy improvements have led to declines in gasoline consumption, while cigarette consumption has declined in response to government efforts to reduce smoking. In aggregate, these taxes generally decline except when increased legislatively. Income, general sales, and selective sales and license taxes account for nearly 90 percent of state taxes. The current economic environment suggests that slow growth is likely for these taxes. 

The slow economic recovery, a fall-off in capital gains and a reluctance to raise taxes have combined to depress state tax revenue compared to past recoveries. States also have had to make room for Medicaid spending driven by recession-induced enrollment increases. The result has been years of cuts in infrastructure spending, government employment, education and other areas, which in turn appear to have created pent-up demand  pressure to restore some cuts, and reluctance to cut spending much further. These pressures have been exacerbated by special circumstances in many states—circumstances that mostly increase fiscal stress. Several  states need to increase pension contributions significantly to pay now for services delivered in the past, driven by years of shirking and by investment income shortfalls. A few states are struggling to live within the confines of large tax cuts. Other states face pressures to find revenue to fund transportation. Many oil producing states are finding their economies out of phase with the national recovery, facing new revenue shortfalls while most other states’ revenue has stabilized. 

The net result of these forces has been budget shortfalls in several states and fiscal stress in many. Looking forward, mainstream economic forecasts for 2015 for the U.S. as a whole call for low inflation, with nominal income growth in the 4 to 5 percent range. Absent booming financial markets or other special factors, income and sales tax revenue growth for the U.S. is unlikely to fall far outside the 4 to 5 percent range—on the high side in the case of income taxes and on the low side in the case of sales taxes. This is not likely to be enough to restore spending cuts, fund infrastructure expansion, pay for Medicaid growth and cover the costs of past promises. States face difficult choices in the midst of growth. 


1 The recovery from the 2001 recession was historically weak but, employment did get more than 4 percent above its prior peak. Seven years after the start of that recession, employment was again heading downward, reflecting the onset of the Great Recession. The graph does not show the 1981 recession separately; rather, it is included in the period after the 1980 recession.

2 We use the four-quarter moving average of tax revenue, adjusted for inflation with the gross domestic product price index from the U.S. Bureau of Economic Analysis. As with our analysis of economic recoveries, we treat the 1981 recession as part of the 1980 recession. 

3 We have written about these gyrations and the incentives that caused them in several state revenue reports during the relevant time period, available at

4 Authors’ calculations from Table 2. Individual Income and Tax Data, by State and Size of Adjusted Gross Income, Tax Year 2012, Statistics of Income, Internal Revenue Service,
12in54cm.xlsx, and

5 “States’ Revenue Estimating: Cracks in the Crystal Ball” (Pew Center on the States and Rockefeller Institute of Government, March 2011). Donald J. Boyd and Lucy Dadayan, State Tax Revenue Forecasting Accuracy (The Nelson A. Rockefeller Institute of Government, September 2014), Managing Volatile Tax Collections in State Revenue Forecasts (The Pew Charitable Trusts and the Nelson A. Rockefeller Institute of Government, March

6 The table is based on data reported in the Fiscal Survey of the States, compiled annually by the National Association of State Budget Officers and the National Governors Association, supplemented with information from state sources for selected large actions taken outside of the survey periods. We have divided years into episodes of response to boom and bust based upon clear patterns in the data. We adjusted for inflation using the gross domestic product price index from the U.S. Bureau of Economic Analysis.

7 Asha W. Agrawal and Hilary Nixon, “What Do Americans Think about Federal Transportation Tax Options? Results from Year Five of a National Survey,” Mineta Transportation Institute, no. Report 12–36 (June 2014),

8 The data are adjusted for inflation using price indexes specific to each category. Price indexes were chosen based on conversations with staff of the BEA.

9 Much of the decline in real state and local government infrastructure spending in the late 2000s and early 2010s was not a decline in “effort” by state and local governments, but rather reflected increases in the prices of goods and services needed to build and maintain infrastructure. (See Public Spending on Transportation and Water Infrastructure, 1956 to 2014 (Congressional Budget Office,March 2015).) According to CBO, real transportation and water infrastructure capital spending fell 23 percent from 2003 to 2014 when adjusted for inflation using input price indexes. However, when adjusted using the GDP price index, real infrastructure spending actually rose by 11 percent (authors’ calculations). Put differently, state and local
governments were trying harder, but getting less for their money due to input price increases. 

10 Authors’ analysis of data in Table 203.10 of the 2013 Digest of Education Statistics, National Center for Education Statistics.

11 Pension contributions vary greatly around the country. In general they are a larger share of local government budgets than state budgets, and in some states they are increasingly important.

12 Spending on higher education includes spending from tuition and fees as well as from direct support. 

13 Michael Mitchell, Vincent Palacios, and Michael Leachman, “States Are Still Funding Higher Education below Pre-Recession Levels,” Center on Budget and Policy Priorities, 2014,

14 Rachel Garfield et al., Trends in Medicaid Spending Leading up to ACA Implementation (Kaiser Commission on Medicaid and the Uninsured, February 2015), http://

15 Authors’ analysis of Current Employment Statistics from the U.S. Bureau of Labor Statistics. 

16 For prisoner statistics, see E. Ann Carson, Prisoners in 2013 (U.S. Bureau of Justice Statistics, September 2014).

17 For pension contribution underpayments see Chris Mier, Twelfth Annual Public Pension Funding Review (Loop Capital Markets, September 2014).

18 See and for surveys of major forecasters, and see for the latest forecasts available at this writing.

19 We use the CBO forecast because it is a well-regarded high-quality forecast developed from an internally consistent model, and because CBO publishes details that are useful for analysis of tax revenue. It is not necessarily likely to be more or less accurate than other forecasts. The Wall Street Journal and the Philadelphia Federal Reserve Bank surveys report averages of major forecasts, and thus their reported numbers are not produced by a single internally consistent model. This makes them less useful for revenue analysis than the CBO forecast.

20 For useful discussions of these issues see D.J. Boyd and L. Dadayan, “Revenue Declines Less Severe, But States’ Fiscal Crisis Is Far From Over,” State Revenue Report (The Nelson A. Rockefeller Institute of Government, April 2010); David L. Sjoquist, Andrew V. Stephenson, and Sally Wallace, “The Impact of Tax Revenue from Capital Gains Realizations on State Income Tax Revenue and Budget Conditions,” Public Budgeting and Finance, Winter 2011; and Norton Francis and Sarah Gault, Federal Tax Policy Uncertainty and State Revenue Estimates, State
and Local Finance Initiative (Urban Institute, March 2015), tax-policy-uncertaint....

21 Authors’ calculations from Table 2. Individual Income and Tax Data, by State and Size of Adjusted Gross Income, Statistics of Income, Internal Revenue Service, various
years, and 54cmcsv.csv

22 Karen Smith Conway and Jonathan C. Rork, “No Country for Old Men (or Women)—Do State Tax Policies Drive Away the Elderly?,” National Tax Journal 65, no. 2 (2012): 313–56.

23 Karen Smith Conway and Jonathan C. Rork, “State Income Tax Preferences for the Elderly,” Andrew Young School of Policy Studies Research Paper Series, no. 07–20 (2007), Ron Snell, State Personal Income Taxes on Pensionsand Retirement Income: Tax Year 2010 (National Conference of State egislatures, February 2011), fiscal/taxonpensions2011.pdf. Note that exemptions for IRA distributions appear to be less common than exemptions for pension or Social Security income.

24 Jeffrey R. Brown, Including Retirement Income in the Illinois Income Tax Base, Illinois Budget Policy Toolbox (University of Illinois Institute of Government and Public Affairs, February 27, 2014). 

25 John L. Mikesell, “The Disappearing Retail Sales Tax,” State Tax Notes, March 5, 2012.