First, some background. For decades Congress has employed “bonus depreciation,” a provision in the tax code that allows companies making investments in machinery and equipment to write off a greater share of expenses for tax purposes sooner than would be the case with normal, longer depreciation schedules.
Because of the time value of money, bonus depreciation—or as is the case with the most recent Congressional tax reform legislation, first-year expensing—makes the after-tax rate of return on investment in machinery higher than it would be otherwise, therefore leading to greater investment.
This is important, not so much to get more machines in the economy, but to get more newer and more productive andinnovative machines. As Nobel Prize winning economist Paul Romer has written, the channel of new capital investment is one of the major ways that innovation diffuses through the marketplace. For example, Moore’s law improvement in semiconductors is only an engineering marvel if companies don’t buy new computers and other machines with the latest chips in them.
Judging on Job Creation
So, the first question to ask is: Has accelerated depreciation had its intended effect of spurring more capital investment? And here the study, like virtually all others before it, finds that the answer is a definitive yes. The authors “find significant and persistent increases in the capital stock.” This should be the end of the story, but the authors seem committed to show that this is not enough.
But few, if any, economists justify longer-term depreciation policies, like the hopefully permanent first year expensing in the most recent tax reform bill, on job creation grounds. The reason is simple. Job creation during non-recessionary periods is a function of two factors: changes in the labor force and Federal Reserve policy. If more people are entering the labor force (more people turning 18 than turning 65, for example) and if the Fed is not tightening the money supply, the employed workforce will be growing. Tax policy that spurs more growth of jobs in a period of full employment will not create net new jobs; it will simply stimulate inflation. So judging bonus depreciation on the metric of job creation, rather than productivity spurring capital investment, is completely misleading.
Even with this caveat, though, the authors find that in the counties where firms took most advantage of bonus deprecation (because they had on average longer-lived assets that benefited more from a shorter depreciation schedule) job growth was 1.9% higher than the baseline of no bonus depreciation.
Wage Effects Misleading
The authors go on to criticize bonus depreciation because they find that it had no positive impact on wages. But this is an even more misleading metric than job creation. Wages largely are not set on the basis of an individual firm’s or even industry’s productivity; they are set on the basis of labor market supply and demand factors. Just because one firm raises productivity does not mean that they can or should raise wages. The workers they are competing for has not changed just because one firm is now more productive. They will likely continue to pay the same wages they paid before. However, because of competitive product markets, firms do respond to productivity increases by passing on a significant share of those savings to consumers in the form of lower prices. This is why the prices of computers and cars (two industries that have seen strong productivity gains) have fallen more than the costs of health care and higher education (two industries with stagnant productivity). And these lower-relative prices from higher productivity that is in turn enabled by new capital equipment is the motor force of higher U.S. living standards, just as they have been since the founding of the Republic.
As such, when the authors look to wage growth to evaluate accelerated depreciation policy, they are making two mistakes. First, they wrongly assume that all or even most of the benefits of productivity get passed on in the form of higher wages. And when they find little change in wages from bonus depreciation, they erroneously conclude that bonus depreciation had no impact on productivity. This is not only wrong – if there were productivity benefits, the effect should be seen in prices, which they don’t measure – it is illogical, as it is not clear why firms would buy new machinery if not to be able to boost productivity (or quality).
Finally, the authors imply that there is no economic rationale for bonus depreciation. As one of the authors told a Washington Post reporter, “if you believe that the market is doing things correctly, the rate at which we invest in these things should not be accelerated artificially.” This is a growing narrative of many who (incorrectly) fear that new technology will lead to massive job loss.
In fact, there is considerable evidence that the market, when it comes to capital equipment investment, is not doing things correctly and that the social welfare-maximizing policy does in fact provide incentives like first-year expensing. For example, Jonathan Temple finds externalities from capital investment that the investing firm does not benefit from. Likewise, Xavier Sala-i Martin finds that both equipment and non-equipment investment (e.g., buildings) are strongly and positively related to growth, but that equipment investment has about four times the effect on growth as non-equipment investment. Kenneth Judd finds that imperfect competition in intermediate capital goods, because innovation is concentrated there, implies that the price is higher than marginal cost. Therefore, he argues there should be greater subsidy for goods with prices significantly higher than marginal costs, and these are more likely to be equipment than structures.
Studies at the industry and firm level have also found compelling evidence of capital equipment spillovers, particularly in information technology (hardware, software, and telecommunications). Van Ark finds that the spillovers from investment in new capital equipment are larger than the size of the benefits accrued by the investing firm. Lorin Hitt finds that the spillovers from firms’ investments in IT are “significant and almost as large in size as the effects of their own IT investment.” In other words, firms capture on average only about half the total societal benefits from their investments in IT, suggesting that the current level of IT investment is significantly less than societally optimal and that policies like first year expensing correct a serious market failure.
The authors conclude:
Our results are immediately relevant for policy makers concerned with job creation and wage growth. If the estimated trends in the substitution between capital and labor persist, then incentives for capital accumulation in the TCJA will likely have small effects on employment and wage growth and may induce investment in labor-replacing capital. Policy makers looking to stimulate labor markets with these tax incentives for capital accumulation should proceed with extreme caution.
In fact, policy makers should proceed with extreme caution in following the policy recommendations from this study. Given both that the U.S. economy has been in an unprecedented productivity growth slump for over a decade and the massive baby boom retirement wave to come, the economy will desperately need faster productivity growth if we are to have any hope of increasing after-tax wage growth faster than minimal levels. The last thing policy makers should do is to reduce the incentive for companies to invest in new machinery and equipment. Instead, given that the first-year expensing provisions are set to expire automatically at the end of 2022, one of the best things Congress could do to ensure strong future growth would be to make that provision permanent.
Robert D. Atkinson is president of the Information Technology and Innovation Foundation, a nonprofit think tank whose mission is to formulate, evaluate and promote policy solutions that accelerate innovation and boost productivity to spur growth, opportunity and progress.