Empirical Evidence on Taxes and Growth: A Response to CBPP
February 21, 2014
Earlier this week, the Center for Budget and Policy Priorities (CBPP) published a response to my review of empirical studies on taxes and economic growth, which was published over a year ago. This was my conclusion at the time:
While there are a variety of methods and data sources, the results consistently point to significant negative effects of taxes on economic growth even after controlling for various other factors such as government spending, business cycle conditions, and monetary policy. In this review of the literature, I find twenty-six such studies going back to 1983, and all but three of those studies, and every study in the last fifteen years, find a negative effect of taxes on growth. Of those studies that distinguish between types of taxes, corporate income taxes are found to be most harmful, followed by personal income taxes, consumption taxes and property taxes.
CBPP objects to this conclusion, stating:
A 2012 Tax Foundation report asserted that “nearly every empirical study of taxes and economic growth published in a peer-reviewed academic journal finds that tax increases harm economic growth.” The report cited 26 studies (19 on the impact of federal or national taxes on national growth and seven on the effects of state taxes on state growth), claiming that 23 of them find that taxes have a “negative” effect on economic growth, while the other three find a “neutral” effect. A previous CBPP analysis found that the Tax Foundation misrepresented the findings of three of the seven state-level studies it cited. This analysis looks in detail at the 19 national-level studies and finds:
- The Tax Foundation mischaracterized, exaggerated, or selectively described the findings of six of those 19. When one adds to these six studies the three state-level studies that the Tax Foundation misrepresented and the three studies that the Tax Foundation correctly identified as showing a “neutral” effect of taxes on growth, 12 of the 26 studies that the Tax Foundation cites do not support its flat assertion that tax increases harm growth.
- The Tax Foundation’s review omitted dozens of relevant studies published in major journals or edited compilations since 2000, many of which conclude that levels of taxation have little if any impact on economic growth or that adverse impacts are limited to particular taxes or time periods.
- The Tax Foundation’s assertion of a growing “consensus among experts” that taxes harm growth is false. In fact, studies that the Tax Foundation cited, as well as others that it omitted, explicitly note the lack of academic consensus.
None of CBPP’s claims are substantiated.
For example, CBPP labeled it “misleading” to restate the finding by OECD economists that progressivity of income taxes is associated with slower economic growth, but there is no other way to interpret this statement from the OECD economists (emphasis added in bold):
Personal income taxes are seen as more harmful to growth than consumption taxes for three reasons. First, they are generally progressive, with marginal tax rates (which discourage growth) that are higher than their average rates (which generate government revenues). This means that they discourage growth more per unit of tax revenue than consumption taxes, which are generally flat rate and not (or not very) progressive. Second, they typically tax the return to savings (interest or dividends) in addition to taxing the income from which savings are made, thus discouraging savings. While this second effect may not harm the growth of publicly quoted companies that can raise funds overseas, it may reduce the growth financing for small and medium-sized companies (especially those that rely on the funds of family and friends). Third, high income tax and social security contributions on low-wage workers can lead to people choosing to stay on social benefits rather than work (Brewer et al., 2010).
And here is the abstract from the OECD working paper in which the specific progressivity results are discussed (emphasis added in bold):
This paper examines the relationship between tax structures and economic growth by entering indicators of the tax structure into a set of panel growth regressions for 21 OECD countries, in which both the accumulation of physical and human capital are accounted for. The results of the analysis suggest that income taxes are generally associated with lower economic growth than taxes on consumption and property. More precisely, the findings allow the establishment of a ranking of tax instruments with respect to their relationship to economic growth. Property taxes, and particularly recurrent taxes on immovable property, seem to be the most growth-friendly, followed by consumption taxes and then by personal income taxes. Corporate income taxes appear to have the most negative effect on GDP per capita. These findings suggest that a revenue-neutral growth-oriented tax reform would be to shift part of the revenue base towards recurrent property and consumption taxes and away from income taxes, especially corporate taxes. There is also evidence of a negative relationship between the progressivity of personal income taxes and growth. All of the results are robust to a number of different specifications, including controlling for other determinants of economic growth and instrumenting tax indicators.
CBPP also claims I hid the nuance of the results, that spending matters, too. Of course the spending matters, which is why I did another review focused on that. The tax studies I referenced explicitly account for spending, and they generally find that income taxes harm growth, while productive public investment, such as building roads, helps growth. In the hypothetical situation where income taxes are used strictly to pay for public investments, most studies conclude it is probably close to a wash. However, one of the studies CBPP claims I misinterpreted actually finds a small negative effect on growth from even this hypothetical scenario. The authors of the paper, Gemmell, Kneller, and Sanz, write: “simulating a 1% of GDP tax increase simultaneously with a 1% productive expenditure increase suggests that this is mildly growth-retarding initially such that even after 20 years GDP remains around 0.5% lower than otherwise.”
The thing is in reality the federal government spends only a small fraction of its budget on public investments, such as roads and airports, and instead spends most of the budget on transfer payments, such as social security and healthcare. Transfer payments are unproductive and even harmful to economic growth, according to most studies. So in practice, income taxes mainly go to transfer payments, and this deal is a clear economic loser, according to the IMF and most academic economists.
As for CBPP’s claim that there are more studies out there than those I cited, I agree. I never claimed otherwise. My approach was to find a large sample of the most important studies ever done, where important means published in the most prestigious peer-reviewed academic journals or by the most widely esteemed international organizations, such as the OECD and the IMF. Any statistics class teaches you 26 is usually a sufficiently large sample from which to draw many conclusions about the underlying population with a high degree of certainty.
As for my sampling process, when I found an article in a top journal, such as the American Economic Review, I followed up on all the references therein, because important articles tend to reference other important articles. Particularly if there was more than a passing reference, or if many studies referred to the same study, I included it. In this way I built up a sample of 26 studies going back to 1983. Never did I exclude any study I found referenced in this way. Through this process, I figured I found all the really important studies and probably most of the rest.
I have since found a few other studies I didn’t know about, and the results do not change my conclusion that the vast majority of studies (roughly 90 percent) find taxes harm growth, particularly income taxes. CBPP cites only three studies that I did not include. All three are news to me. One is a comment on a previous study, which I didn’t find referenced elsewhere. Another finds mixed results: individual income taxes reduce growth but corporate income taxes do not. The third study finds “increasing the size of the public sector (through current expenditures and direct taxes) improves the distribution of income at the expense of economic growth.” This seems to support my conclusion.
CBPP also writes: “A recent academic survey of the literature on government size and growth listed 16 cross-country panel data studies published in peer-reviewed journals between 1999 and 2011. The Tax Foundation study cited just four of those.”
I’m surprised I cited four, since that survey doesn’t deal with taxes and growth, rather the elasticity of taxable income.
The bottom line is this critique doesn’t add up. Why would I falsify the conclusions of published reports, when anyone (with access to academic journals) can check my work? Why did it take CBPP over a year to concoct a rebuttal if it were that easy?
Rather, a more plausible explanation is that CBPP has no case and must instead resort to filling a few pages with noise so as to create confusion. I’m afraid this is just politics, and not economics. Instead of clarifying the issue, the CBPP study brings new confusion to an issue on which the economics is settled: taxes negatively affect economic growth.
Follow William McBride on Twitter