A new paper by experts at Wharton and elsewhere has set to rest “widespread concerns” that increased capital investment in equipment is at the cost of worker employment. In the study of tax incentives that boost capital investment in equipment at U.S. firms between 1997 and 2011, the experts found that such investment resulted in matching employment growth, although it did not stimulate wage or productivity growth.
“People have been worried that with corporate tax breaks for investment, we’ve made machines relatively cheaper than workers, and that we will switch from having more workers to having relatively more machines,” said Wharton finance professor Daniel G. Garrett, who co-authored the paper, “Capital Investment and Labor Demand.” The study used bonus depreciation as an example of a tax incentive, or as “a shock to the after-tax cost of physical capital.”
Garrett’s co-authors are E. Mark Curtis of Wake Forest University, Eric C. Ohrn, an economics professor at Grinnell College in Iowa, Juan Carlos Suárez Serrato, economics professor at Stanford Graduate School of Business, and Kevin A. Roberts, a doctoral candidate in economics at Duke University.
“What we show in this paper is the worst concerns of substitution between capital and production labor do not seem to manifest in the U.S. with bonus depreciation, or the acceleration of depreciation expenses,” Garrett continued. “So, at a high policy level, it looks like giving businesses money to invest in new machines doesn’t lead to them to buy machines that replace workers. They buy more machines, but they also hire the same proportion of workers that they were hiring before to work those new machines. Capital and labor are actually complements; they’re not substitutes in production with respect to a change in the cost of physical capital.”
How Bonus Depreciation Works
Bonus depreciation allows plants to deduct capital investments from their taxable income more quickly, lowering the cost of investment. With a 50% bonus, firms that invest in equipment can immediately deduct an additional 50% of their capital costs. The remaining 50% of the costs are deducted according to normal depreciation schedule — usually the Modified Accelerated Cost Recovery System (MACRS). For example, a $1,000 investment would get a $500 bonus depreciation in the first year, plus another $100 on the remaining $500 of the capital cost, assuming it has a useful life of five years, the paper explained.
“At a high policy level, it looks like giving businesses money to invest in new machines doesn’t lead to them to buy machines that replace workers.”— Daniel G. Garrett
By comparing plants that benefit the most from this incentive — those that invest more in equipment that is deducted slowly according to IRS rules — to plants that benefit less, the study isolated investment in equipment that is likely independent of other drivers of capital accumulation.
The paper pointed out that it captured “the decade-long effects of bonus on individual production units,” while prior research had studied only the short-term effects using consolidated firm-level data. In particular, it estimated “novel responses to bonus depreciation,” including the accumulation of capital stocks, plant sales, total factor productivity, labor earnings, overall employment, employment for production and non-production workers, and workforce demographics.
Bonus depreciation is one of the largest incentives for capital investment in U.S. history and has been in nearly continual use since its inception in 2001, the paper noted. After the U.S. introduced bonus depreciation in 2001, several large economies have followed suit, including the U.K., China, Canada, and Poland, it added, citing prior research.
Governments around the world have used accelerated depreciation policies such as bonus depreciation for more than 100 years, the paper noted. They have used it to stimulate business investment, especially to spur defense spending during the First World War and the Second World War, and to replenish industrial capital stocks in the aftermath of these wars, the paper added.
The Reach of a Tax Break
The model used in the study separated the effects of bonus depreciation into substitution and scale effects. In its findings, the study estimated that the scale effect — the increase in the use of all inputs due to lower production costs — accounts for 90% of the employment effects of the policy. As production employment increased by more than the scale effect, the authors concluded that capital and production labor are complements in manufacturing.
According to the study’s estimates, manufacturing plants that used bonus depreciation increased investment flows by 15.8% relative to those that did not do so. It also estimated a relative increase in overall capital of 7.8% between 2001 and 2011. In that same period, plants that benefited more from bonus depreciation had a relative employment increase of 9.5%.
“Capital and labor are actually complements; they’re not substitutes in production with respect to a change in the cost of physical capital.”— Daniel G. Garrett
The study also rejected “a popular rationale” for investment tax incentives that capital investment will raise productivity and workers’ wages. It found that average earnings decreased by an estimated 2.7% at the plants that used bonus depreciation. But those plants hired more “young, less educated, women, Black, and Hispanic workers,” the paper found. “We see that the workers being hired are more from historically disadvantaged groups,” Garrett said. Those workers are typically “non-white, female, young, and [those with] low education,” he added.
“The first high-level policy implication is there’s a real concern that tax incentives for capital investment will lead to capital being relatively cheaper than labor and moving away from labor and towards capital,” Garrett said. “With bonus depreciation, we don’t see that being true. We find that capital and labor, with respect to these tax changes, appear to be complements.”
The second policy implication is that bonus depreciation does not raise average wages or significantly increase productivity at the firms that used that tax break, Garrett continued. “We don’t see increases in what we call total factor productivity — how much a firm is able to manufacture with a given set of inputs. Instead, we really just see the firms scaling up their existing business, so they’re not really doing anything different. We don’t see wages increasing, we don’t see productivity increasing. We do see more workers being hired and we don’t see workers being laid off. But what we do see is this lack of impact on productivity or wages that might come from additional capital being invested.”
Behind all those is a potential bonus for the government’s coffers. The increase in employment that bonus depreciation brings “means that we will get a little bit more personal income tax revenue as well from this type of tax break,” Garrett said.