Tax Incentives Raised Business Investment

Bonus depreciation appears to have had a powerful effect on the composition of investment. Capital that benefited substantially from the policy -- namely equipment with long tax lives -- saw sharp increases in investment. In contrast, there is no evidence that market prices increased because of the policy.

In Temporary Investment Tax Incentives: Theory with Evidence from Bonus Depreciation (NBER Working Paper No. 12514), authors Christopher House and Matthew Shapiro analyze how temporary changes in taxes affect the incentive to invest. Although their work is motivated by recent changes in tax law, their analysis has general implications for the equilibrium effects of temporary tax incentives.

Small changes in the timing of a firm's purchases of long-lived pieces of equipment have little effect on their value to the firm. For example, how much a machine produces over the next twenty years will be essentially the same whether the machine is installed in late December or early January. On the other hand, if a tax subsidy is available in December but expires in January, then the firm has a strong incentive to install it in December. As a result, powerful incentives to alter the timing of investment in response to temporary tax subsidies exist. These incentives are so strong that, for a sufficiently temporary tax change or a sufficiently long-lived capital good, firms will bid up the purchase price of investment goods by exactly the amount of the subsidy. House and Shapiro use this insight into the effect of temporary investment subsidies to estimate how responsive the quantity of investment is to investment tax subsidies.

The authors estimate the responsiveness of investment and test the theory by examining disaggregated data on investment after the 2002 and 2003 tax bills. These bills provided for temporarily accelerated depreciation -- called bonus depreciation -- that allowed firms to immediately deduct an increased fraction of their investment spending. Under the 2002 bill, firms could immediately deduct 30 percent of investment and then depreciate the remaining 70 percent under the standard depreciation schedule. Under the 2003 bill, the bonus deduction increased to 50 percent. This investment subsidy was explicitly temporary. Only investments made through the end of 2004 qualified for this tax treatment.

Using data on investment expenditures for different types of capital goods, the authors estimate the elasticity of supply for investment-the main parameter determining the size of the response of investment to a temporary tax incentive. The data clearly show that the policy had a stimulative impact on investment in capital that benefited most from bonus depreciation. The authors' estimates indicate that investment reacts strongly to changes in tax policy. Their analysis also suggests that the policy may have increased output by roughly 0.1 percent to 0.2 percent and increased employment by roughly 100,000 to 200,000 jobs. Market prices, on the other hand, showed little if any tendency to increase in the short run.

The authors' general results hold for only the specific circumstance of a temporary change in the cost of purchasing capital goods. Their calculations show that for long-lived durable capital goods, even changes in tax policy that last for multiple years can safely be modeled as temporary. Given the frequency of changes in tax policy, the authors' analysis can be applied to many episodes.

The bonus depreciation allowance, passed in 2002 and then increased in 2003, provides an ideal opportunity to estimate the responsiveness of investment to changes in tax policy. Only investment goods with a tax recovery period less than or equal to 20 years qualify for the bonus depreciation. The theory suggests that there should be a sharp difference in the response of investment spending between the 20-year investment goods and those with more than a 20-year recovery period. In addition, among the qualified investment goods, researchers should observe higher investment spending for goods with higher tax recovery periods. The authors' data support both predictions. Bonus depreciation appears to have had a powerful effect on the composition of investment. Capital that benefited substantially from the policy -- namely equipment with long tax lives -- saw sharp increases in investment. In contrast, there is no evidence that market prices increased because of the policy.

Although the policy expired in 2005, it is not clear whether investment spending returned to normal, as one would predict. This is probably because of the extension of bonus depreciation for certain properties and the increased Section 179 exemption, a tax incentive that shares many of the features of bonus depreciation but, unlike bonus depreciation, was extended beyond the end of 2004.

Because the data indicate that qualified investment goods responded strongly to the tax policy, the estimated elasticity of the supply of investment is quite high. The authors use their estimates to assess the likely aggregate impact of the policy. Because the policy was narrowly focused on a small subset of investment spending, the authors find that it had only modest effects on aggregate employment and output, despite the stark effects on the composition of investment

-- Les Picker