Phased-In Tax Cuts and Economic Activity

Questions & answers

Q1. What classic public-finance principle does phased-in tax legislation violate, and why does the paper push back against treating that principle as universal?

“Legislating predictable changes in tax rates violates one of the cardinal principles of public finance: Changes in tax rates should be permanent and immediate,” since “the deadweight burden of taxation is a convex function of the tax rate,” so Barro (1979)’s tax-smoothing logic implies tax rates should follow unpredictable random walks. The authors note this is “not universal”: Chamley (1986) and Judd (1985) show that in economies with capital, “the optimal tax rate on capital income must be zero in the steady state,” and “because it is often optimal to tax the initial capital stock heavily, the optimal tax rate on capital income should be phased in” – so standard optimal-tax theory itself calls for phasing in capital-income tax cuts, even while favoring unpredictability for labor-income tax rates.

Q2. What is the paper’s core mechanism for why a delayed labor-income tax cut depresses activity during the delay?

Because forward-looking consumers “experience the income effects of the policy as soon as the tax plan is announced” but the substitution effect on labor supply “is not realized until four years later,” workers face high current taxes relative to the known future rate and so “have an incentive to work less currently while taxes are temporarily high,” an incentive that “will be operative as long as the workers can intertemporally substitute consumption and work.” In the calibrated simulation of a four-year-delayed one-percentage-point labor tax cut, “in the year following the policy announcement, employment is below trend by 0.28 percent, output is below trend by 0.20 percent and investment by 0.95 percent,” with the fall in investment reflecting that “since production falls, the increase in consumption comes at the expense of investment.”

Q3. Why does delaying a capital-income tax cut have the opposite sign effect from delaying a labor-income tax cut?

“In contrast to taxes cuts on labor, delaying capital income tax cuts does not discourage current economic activity,” because “current investment can be stimulated by future tax cuts… since the payoff of current investment largely accrues in the future” and – when depreciation allowances can be written off against current taxable profits – “there is a substantial incentive to move investment spending forward to take advantage of the write off when the tax rate on profits is high.” In the paper’s simulation, delaying a capital tax cut leaves employment “above trend by 0.31 percent” in the year following the announcement. The authors add that this timing pattern is not merely a quirk but has a normative upside: “an immediate tax cut on capital is, for the most part, simply a windfall to existing capital and has no beneficial incentive effects,” whereas a delayed cut “provides incentives for current investment but still taxes existing capital at a high rate.”

Q4. How does the model apply to the actual 2001 and 2003 U.S. tax laws, and what quantitative effect do the authors attribute to the 2003 acceleration?

The 2001 EGTRRA law “called for a scheduled sequence of rate reductions” through 2006 with a 2011 sunset, while the 2003 JGTRRA law made “reductions in income tax rates that were scheduled to occur in 2004 and 2006 under EGTRRA instead [go] into effect immediately” and added temporary cuts to dividend and capital-gains taxes. Comparing simulated activity under the accelerated 2003 schedule to what the model implies would have happened had the original 2001 phase-in continued, the authors find that “in the third quarter of 2003, employment increased by 0.80 percent, GDP increased by 0.52 percent, and investment increased by 1.83 percent,” while “consumption does not respond as much… since moving forward the rate cut has a relatively small effect on after-tax permanent income.”

Q5. What historical narrative do the authors offer linking the model’s mechanism to the slow 2001-2003 recovery and the subsequent pickup, and how carefully do they qualify it?

The authors state that “the phased-in tax cuts enacted in 2001 are a potential explanation of the failure of employment and output to recover rapidly during the period from 2001 to mid-2003,” because the phase-in “provides a powerful incentive for businesses and workers to defer production and employment while at the same time it gives consumers reason to maintain high levels of consumption spending” – weakening investment in particular. They are careful to hedge this as one contributing factor rather than a full explanation: “while there are, of course, many other factors that influenced employment and production decisions, it is likely that the phase-in features of the 2001 tax law contributed to the tepid economic performance of the period.” They note the 2003 law’s passage coincided with real GDP expanding “at a rate of 8.2 percent” in the third quarter, “more than double the highest growth rate in the previous two years,” and suggest “the immediate tax relief included in the 2003 law may have contributed towards the increased pace of economic activity,” again using the hedged “may have.”

Q6. How does the paper’s quantitative conclusion compare to other contemporaneous studies of the 2001 tax law, and what explains the difference?

Gale and Potter (2002) “conclude that the tax law would depress the level of GDP once it is phased in,” whereas House and Shapiro instead “find a modest increase in GDP”; the authors attribute the difference to the fact that “the capital stock increases in our simulations,” driven not only by the direct incentive effect of the capital-tax cut but by “the equilibrium response of the capital stock to the increase in labor supply” – more labor supply pushes down wages and so stimulates capital accumulation, a general-equilibrium channel the authors say Gale and Potter’s partial, long-run-only analysis does not capture. They note Auerbach (2002), using the Auerbach-Kotlikoff (1987) model, reports “a pattern of output effects similar to those shown in our simulations” under the sunset scenario, suggesting some convergence with at least one other structural model.

Q7. What is the paper’s explanation for why phased-in and sunsetting tax law is so common, if it has these real costs?

The authors argue that “federal budget rules mandate reporting the revenue consequences of tax laws over 10-year windows,” which “frames the debate and the description of the legislation,” and that the Byrd rule – which lets senators raise a point of order against tax changes with effects beyond ten years – “effectively requires a supermajority of 60 votes to pass tax reductions that extend beyond 10 years.” Because permanent tax cuts cost more revenue inside any fixed-length budget window than cuts that phase in or sunset within it, and because supermajority thresholds are harder to clear, legislators seeking to pass tax relief with a simple majority have a structural incentive to design cuts that phase in gradually or expire – a procedural explanation for the phase-ins the model’s welfare analysis treats as costly.