The States Try Supply-Side Tax Cuts
william gale holds the Arjay and Frances Miller Chair in federal economic policy at the Brookings Institution (firstname.lastname@example.org). aaron krupkin is a research assistant at Brookings (email@example.com).
kim rueben is a senior fellow at the Urban Institute (firstname.lastname@example.org). All three are affiliated with the Urban-Brookings Tax Policy Center.
The authors thank the Laura and John Arnold Foundation and the Ewing Marion Kauffman Foundation for financing the original research cited in this article. The findings are the responsibility of the authors and do not necessarily reflect positions of the Tax Policy Center, the Brookings Institution or the Urban Institute.
Illustrations by Jim Tsinganos
Published October 19, 2015.
Many folks, and from time to time, majorities in Congress, apparently believe that the cure for what ails the economy is lower taxes – in particular, lower tax rates for high-income earners. Now this enthusiasm has spread to state governments that are led by conservatives, offering new tests of a proposition that has generated scant evidence of success elsewhere.
Failure of this idea at the federal level does not necessarily imply that tax cuts would fail to increase output and jobs at the state level. For one thing, lower taxes in one state might lure existing businesses (and jobs) from other states, even if they yield no overall increase in employment or output. But it's also worth noting that the stakes are higher for the states. Washington can finance shortfalls in revenue by selling bonds to the public or by borrowing from the Federal Reserve – in effect, printing money. States are far more constrained by the skepticism of the private credit markets or constitutional prohibitions against deficit finance, or both. Thus, any failure of supply-side economics to work its magic could force punishing cuts in state programs.
Present at the Creation
First, a little history. Ever since the 1970s, when Jude Wanniski (then a Wall Street Journal opinion editor) and Arthur Laffer (then a young college professor) came up with the ideas that are now referred to as supply-side economics, conservative politicians have been unable to resist the siren song of tax cuts for big earners. The claim, of course, is not that high-end tax breaks merely serve the interests of groups that support conservative politicians; rather, it is that the tax cuts would serve the broader public interest by increasing incentives to work and to create (or expand) businesses.
In the extreme versions of supply-side economics that thrived through the early Reagan administration years, supply-siders argued that tax rates were so high that cuts would increase the economic growth rate sufficiently to pay for themselves. After decades in which lower tax rates generated less revenue rather than more, today's supply-siders are more inclined to make less immodest claims; many advocates argue that tax cuts will spur growth that will make up for part of the revenue losses as they create jobs and private income, asserting that the loss of some tax revenue is worth it due to the boost in economic activity. However, some proponents still can't help themselves and lapse into the more hyperbolic claims.
Inside the Beltway
Many conservatives apparently believe that federal tax cuts have borne rich fruit for the country, or, at worst, that the jury is still out. But this seems to be a case of "who are you going to believe – us, or your lying eyes?" At the federal level, there is virtually no evidence that broad-based tax cuts have had a positive effect on growth.
The vaunted Reagan tax cuts in the early 1980s produced a period of average growth, when growth is (appropriately) measured from peak to peak of the business cycle. Indeed, research by Martin Feldstein, President Reagan's former chief economist, and Douglas Elmendorf, the former Congressional Budget Office director, concluded that the 1981 tax cuts had virtually no net impact on growth. Indeed, they found that the recovery in the 1980s could be ascribed wholly to monetary policy. It's also worth noting that they found no evidence that the big 1981 tax cuts induced people to work more.
Apparently, no one claims that the 2001 and 2003 Bush tax cuts stimulated growth. The two enabling acts did have the word "growth" in their titles (the Economic Growth and Tax Relief Reconciliation Act of 2001, and the Jobs and Growth Tax Relief Reconciliation Act of 2003) and slashed tax rates on ordinary income, capital gains, dividends and estates. Yet growth remained sluggish between 2001 and the beginning of the Great Recession in late 2007. Again, the gains that did occur are generally attributed to the Fed's easy-money policy.
But the Reagan and Bush tax cuts, each about 2 percent of GDP, were small potatoes compared to the tax increases during and after World War II, when federal taxes rose by more than 10 percent of GDP. That's right, not 10 percent of taxes, but one-tenth of the entire economy. Income tax rates went up for virtually everyone, and revenues and rates stayed higher for decades. In fact, between 1944 and 1963, the top tax bracket never fell below 90 percent. According to supply-side theory, that should have killed the economy. Instead, according to Nancy Stokey (of the University of Chicago) and Sergio Rebelo (of the Kellogg School), real per capita growth rates differed little from the historical averages.
Tax rates as determinants of long-term growth fare no better in cross-country comparisons. Research by Thomas Piketty (Paris School of Economics), Emmanuel Saez (University of California, Berkeley) and Stefanie Stantcheva (MIT) found no relationship between how a country changed its top marginal tax rate and how rapidly it grew between 1960 and 2010.
The United States, which cut its top rate by over 40 percentage points during that period, grew just over 2 percent annually per capita. Germany and Denmark, which barely changed their top rates at all, experienced about the same growth rate.
The story is much the same when total tax burdens are compared. Over the 1970-2012 period, taxes as a share of GDP were 7 percentage points higher in the rest of the OECD countries (32 percent) than in the United States (25 percent). Yet, per capita annual growth was virtually identical in the rest of the OECD (1.8 percent) as in the United States.
So, there is no reason to believe that tax cuts are an elixir for economic growth. When policymakers count on increased business activity and job creation, as well as revenue to partially offset the initial cost, they are risking serious economic dislocation. There's another problem here, as well. As Piketty and company note, with or without the elusive supply-side effect, high-end tax cuts have exacerbated income inequality.
Despite all of this, tax breaks for high earners are prominently featured in contemporary debate over tax policy. Mitt Romney's 2012 campaign proposals included tax cuts for the "job creators." And don't expect Republican candidates to change their tune for 2016. Marco Rubio (along with Utah's Sen. Mike Lee) has come up with the Economic Growth and Family Fairness Tax Reform Plan; Rand Paul has checked in with his Fair and Flat Tax. While the plans are quite different, both cut taxes at the high end of the income ladder.
The zeal for lowering income tax rates, especially at the top, spread beyond Washington decades ago. In the 1990s, six states cut taxes by more than 10 percent, mostly by enacting significant personal income tax cuts. Their subsequent growth records are mixed. Gross State Product (GSP) did grow faster, on average, in the six tax-cut states combined than in the rest of the country. But that was mostly explained by the explosive growth of the financial sector, which is centered in Connecticut, New Jersey and New York, and was surely more a product of the Clinton-era boom in financial engineering and the dot-com bubble than of state tax policy. On average, the other tax-cut states (Massachusetts, Delaware and Colorado) grew more slowly than the rest of the country. Moreover, employment grew at a faster rate in the rest of the country than in the six tax-cut states.
New Jersey's tax cut was anchored on a 30 percent reduction in personal income taxes from 1994 to 1996. Using county-level data on employment, Robert Reed (University of Canterbury) and Cynthia Rogers (University of Oklahoma) found that New Jersey experienced strong employment growth after the tax cut – but so did counties with similar economic profiles in nearby states that did not have tax cuts. The net effect of the tax cut, measured by the difference in employment gains between New Jersey and the nearby economic regions, was small and statistically insignificant.
Between 2001 and 2007, Arizona, Louisiana, New Mexico, Ohio, Oklahoma and Rhode Island cut personal income taxes. However, there was no discernable impact on economic activity. From 2001 to 2012, these states grew, on average, at virtually the same rate as the rest of the U.S. economy. In fact, extending the measurement period through 2014, four of the six states experienced declines in their shares of total U.S. employment. Two states, New Mexico and Oklahoma, did experience net gains in employment share over the extended period. But a far more plausible explanation centers on the boom in oil and gas fracking, which was hardly related to state income tax policy.
Some states, notably California, Maryland and New York, have retained increases in top marginal income tax rates that were introduced in recent years to address revenue shortfalls. But the supply-side spirit lives on in a number of others, which have cut personal income taxes and corporate income taxes, and provided special breaks for businesses, with the goal of spurring growth.
The most widely reported recent state income tax cut occurred in Kansas in 2012. In the 30 years prior, Kansas' economic growth rate had lagged consistently below the averages for neighboring states and the nation as a whole. Hence Gov. Sam Brownback pressed for a tax cut that would be "like a shot of adrenaline into the heart of the Kansas economy."
Initially, Brownback submitted a revenue-neutral plan that reduced income tax rates across the board and effectively exempted small-business income from all tax. To offset the cost, he proposed to eliminate itemized deductions, make a temporary sales tax increase permanent and cap state spending. But the Legislature passed a bill with quite different provisions. Starting in 2013, the top rate of 6.45 percent was eliminated, the next bracket (income above $15,000) was cut from 6.25 percent to 4.8 percent, and the lowest rate was cut from 3.5 percent to 2.7 percent. Taxes on pass-through income from small businesses were eliminated, reducing the effective state income tax rate on those businesses to zero. Brownback adopted the plan as his own, calling the legislation a "real live experiment" in supply-side economics.
The tax cuts did not produce the hoped-for growth, though, and more revenue was lost than originally anticipated. Fiscal year 2014 revenues were $700 million lower than the previous year – $330 million less than expected – in a period in which most of the American economy was picking up steam. Put in context, these numbers are pretty significant: $330 million represents more than 5 percent of Kansas' government spending from general funds. Responding to the gloomy news, both Moody's and Standard & Poor's reduced Kansas' credit rating.
Arguably as important, Kansas failed to keep up with the region's pace of job growth. Kansas' cumulative job growth rate from January 2011 through March 2014 was 3.4 percent – 39 months in which the United States registered 5.5 percent job growth and employment in neighboring Colorado, Oklahoma and Nebraska grew by 8.2 percent, 5.6 percent and 4 percent, respectively. Even if we focus on the latter half of that period, beginning in January 2013, which was after the tax cut was enacted, Kansas still falls behind. The state's job growth clocked in at 1.4 percent, compared to the U.S. average of 2 percent. Thus, it seems as if the "live experiment" failed to produce the projected economic gains.
Under the threat of large spending cuts, especially for schools, the Kansas Legislature raised taxes in June. However, instead of reversing the income tax changes and the provision that eliminated tax liability for many businesses, Kansas increased its regressive sales tax to 6.5 percent (from 6.15 percent) and increased the regressive excise tax on cigarettes by 50 cents a pack. This suggests yet another way in which the experiment failed: once progressive income taxes are cut, the political path of least resistance for reversing the revenue losses is to raise indirect taxes that are usually regressive.
Over the past few years, other states have also tried cutting income taxes, partially offsetting the revenue losses by raising their sales taxes. Since 2011, Gov. Scott Walker of Wisconsin (who was, as of mid-summer, one of the leading candidates for the Republican presidential nomination) has signed a collection of tax cuts that reduced revenue by roughly $2 billion – a remarkable loss for a state with general-fund spending of about $15 billion. However, the expected job growth and budget surpluses in Wisconsin did not arrive. Gov. Walker recently signed a budget that cut spending on higher education, roads and scientific research.
In Louisiana, many politicians thought that the economic boom and revenue gains after Hurricane Katrina would stand the test of time. The state took the opportunity to cut income taxes to the tune of $700 million, adding to the woes of a budget that is vulnerable to the business cycle and heavily dependent on royalties from oil production. The $1 billion budget surplus in 2007-08 has morphed into a projected deficit of $1.6 billion for the current fiscal year. In response, Gov. Bobby Jindal froze public-employee wages, scaled back tax expenditures, increased the cigarette tax by 50 cents per pack and raised motor vehicle fees.
In April 2014, Oklahoma's governor, Mary Fallin, chose a slightly different approach. She signed a bill that would reduce the state's top marginal personal income tax rate to 5 percent from 5.25 percent by 2016, but only if state revenue projections were larger than expected. Because this trigger was based on projections, it was met in December 2014 and, despite a budget shortfall, the tax cut automatically went into effect. A second cut, to 4.85 percent, will go into effect after 2016, again provided revenue projections exceed targets. However, with oil and natural gas prices in decline, it is unclear whether the second round will occur.
While it is too soon to understand the full effects of the recent state income tax cuts, the examples above hardly strengthen the case for supply-side tax cuts. Moreover, these ambiguous state experiences tend to reinforce the conclusions from a voluminous academic literature. Recent studies have generated almost every conceivable finding: tax cuts raise, reduce or have no clear effect on growth. In addition, the effects of changing different taxes – income, corporate, property and sales – vary dramatically within and across studies.
A variety of methodological factors complicate interpretation of these findings: the econometric studies used different dependent variables, analyzed different time periods, employed alternative measures of tax revenues or rates or both, included different measures of government spending, controlled for different independent variables and used different control groups and identification methods. Additionally, balanced-budget requirements imply that revenues and spending should be closely correlated, making it especially difficult to study the independent influence of taxes or spending.
In our own recent research, we worked to be fair to both sides, making fresh estimates of how state tax policy affects economic growth and entrepreneurial activity. We used a framework that in prior research had led investigators to the conclusion that lower taxes do stimulate growth, extending the sample period and performing a variety of tests to see how sensitive the results were to plausible changes in the statistical models. We found that neither tax revenues nor top marginal income tax rates bear any stable relation to economic growth rates across states and over time.
Consistent with these findings, we also concluded that neither marginal tax rates nor tax revenues consistently affect employment. And while the rate of firm formation is negatively affected by top income tax rates, these effects are small. The gist: there is no guarantee – and there should not even be a presumption – that cutting state income taxes will boost economic growth.
At the core of supply-side economics is Arthur Laffer's back-of-the napkin curve illustrating the undeniable reality that, at some point, higher tax rates will lead to lower revenues as well as fewer jobs and slower growth. But this does not imply there are many real-world examples of tax rates so high that cutting them would have much impact on jobs or growth. That has been amply demonstrated at the national level, where tax cuts have eroded revenue without discernable effect on economic activity.
The states have no good reasons to believe that tax cuts will bring the desired manna. Yet some continue to erode their tax bases in the name of business growth in an era in which few states can afford to cut critical services (that businesses care about) ranging from education to infrastructure repair. Some ideas live on and on, no matter how much evidence accumulates against them. States that accept them as gospel anyway do so at their peril.