The Dynamic Effects of Personal and Corporate Income Tax Changes in the United States

Questions & answers

Q1. What is the core econometric innovation, and what problem in the existing narrative-tax literature does it address?

The authors note that existing narrative approaches (e.g., Romer and Romer 2010) “implicitly view” narratively identified tax changes “as mapping one-to-one into the true structural shocks,” even though “in practice there are good reasons to expect that narratively identified shocks suffer from measurement errors as historical records rarely are sufficiently unequivocal that calls of judgment can be avoided.” Their solution “exploits the informational content of narrative measures of exogenous changes in taxes for identification in an SVAR framework” by “imposing the restrictions that narrative measures of exogenous tax changes correlate with latent tax shocks but are orthogonal to other structural shocks” – a proxy-variable (external-instruments) SVAR that, unlike a pure narrative regression, does not require the narrative series to be a noise-free measure of the shock, and, unlike a standard structural VAR à la Blanchard and Perotti (2002), does not require direct assumptions on structural parameters.

Q2. How is the narrative tax-shock series constructed, and what two restrictions does the paper impose to isolate “exogenous” and “unanticipated” changes?

The narrative measure is “developed from Romer and Romer’s (2010) account of changes in federal US tax liabilities which we decompose into changes in personal and corporate income tax liabilities,” using “only… tax changes that Romer and Romer classify as exogenous” (i.e., not motivated by near-term counter-cyclical or deficit-driven concerns). Following Mertens and Ravn (2011b), the authors additionally “exclude legislative tax changes with implementation lags exceeding one quarter to remove anticipated tax changes,” since a tax change households and firms know about in advance has distinct effects from one that changes taxes contemporaneously – the same within-quarter-implementation logic used to separate “shocks” from “news” in this literature more generally.

Q3. What are the paper’s headline output effects, expressed per percentage point of tax-rate change?

“A 1 percentage point cut in the APITR raises real GDP per capita by 1.4 percent on impact and by up to 1.8 percent after three quarters. A 1 percentage point cut in the ACITR raises real GDP per capita on impact by 0.4 percent and by 0.6 percent after one year.” The authors are explicit that these are “large” short-run effects, and that the size difference between personal and corporate tax cuts is itself one of the paper’s central findings, since much of the prior literature analyzed tax policy as a single undifferentiated aggregate.

Q4. Why is a well-defined “multiplier” available for personal tax cuts but not for corporate tax cuts?

“Cuts in personal income taxes lower tax revenues while cuts in corporate taxes have no significant impact on revenues because of a very elastic response of the tax base,” so translating the output effect into “the change in output deriving from a change in tax rates which reduce tax revenues by 1 percent of GDP” – the paper’s multiplier definition – gives “a maximum personal income tax multiplier of 2.5,” while “the corporate income tax multiplier is instead not well defined because we find that changes in corporate income taxes have little impact on tax revenues” (dividing a real output effect by a revenue effect close to zero). This is a case where the underlying effect (GDP rises when ACITR falls) is real and significant, but the specific “multiplier” statistic the paper uses elsewhere cannot be sensibly computed for it.

Q5. What distinct effects do the two tax types have on employment, consumption and investment, and why does this matter for the identification?

“A cut in the APITR raises employment, lowers the unemployment rate, and increases hours worked per worker. A cut in the ACITR, on the other hand, has no immediate impact on either employment or hours per worker. Both cuts in the APITR and in the ACITR increase private sector investment, but only cuts in personal income taxes stimulate private consumption.” The authors read these qualitative differences in sign and timing as themselves evidence that the two tax instruments are being separately and correctly identified, writing that “the differences in the size and signs of the responses to the two types of taxes demonstrate the necessity of discriminating between different types of taxes” – a single aggregate “tax shock” measure would average over two instruments with materially different transmission mechanisms.

Q6. How does the paper quantify the reliability of its own narrative proxies, and what does that measure say?

Under additional measurement-error assumptions, the authors estimate “the reliability matrix of m_t has eigenvalues of 0.30 and 0.69,” implying “that the correlations between the principal components of the narrative tax changes and the true tax shocks are 0.55 and 0.83,” together with R² statistics of “0.22 and 0.38” from regressing the reduced-form tax-rate residuals on the proxies. The authors interpret this as showing the proxies “contain valuable information for the identification of the structural tax shocks,” while the fact that “the reliability matrix has eigenvalues substantially below unity indicates that measurement error is a serious concern in practice” – i.e., the narrative series is useful but is explicitly not being treated as a perfect measure of the true shock, in contrast to how it would be used in a plain narrative regression.

Q7. What checks does the paper report that the identified tax shocks are not confounded with other policy actions?

The authors report finding “no signs of any significant change in government spending or short term nominal interest rates following tax shocks,” and separately verify that government debt “increases significantly at the 95 percent level in the short run after an APITR cut, but does not change significantly after an ACITR cut” – consistent with the differential revenue effects described in Q4 – and that “cuts in one average tax rate lead to increases in the other average tax rate, although neither of these increases is significant,” which the authors take as confirming that “our orthogonalization scheme successfully disentangles the effects of different tax instruments” rather than conflating a personal-tax shock with an offsetting corporate-tax response or vice versa.

Q8. What policy conclusions do the authors draw, and how do they qualify the comparison to spending multipliers?

The authors conclude that, relative to “the currently available evidence on the multipliers associated with US government spending” (citing Ramey’s review), “the federal tax multipliers are likely to be larger than those associated with federal government purchases,” and that “if policy objectives include short run job creation and consumption stimulus, then cuts to personal income taxes are more effective than cuts to corporate profit taxes.” On the revenue side, they note the asymmetry runs the other way for tax increases: “increases in personal income taxes are effective [at raising revenue], but the costs in terms of job and output losses are relatively large,” whereas “increases in corporate profit taxes are not likely to raise significant revenues” at all – so the same personal/corporate distinction that makes personal cuts a more powerful stimulus tool also makes personal tax increases a costlier way to raise revenue.