Some Questions Regarding the Diamond and Zodrow Modeling of Camp’s Tax Plan
March 17, 2014
John Diamond and George Zodrow were recently commissioned by the Business Roundtable to model the dynamic effects of House Ways and Means Chairman Dave Camp’s discussion draft. The study is interesting for a number of reasons, not least because John Diamond is also the principal architect of one of the macroeconomic models used by the Joint Committee on Taxation (JCT) – the overlapping generations model (OLG). As such, it provides additional insight into the official JCT analysis of the Camp plan, and introduces some new questions.
First, Diamond and Zodrow are not strictly modeling the Camp discussion draft, but rather a more economically beneficial variation on it. As they describe on page 7:
The government must balance its budget in each period, after taking into account enough borrowing to maintain a constant debt-to-GDP ratio. During the five-year phase-in of the reduction in the corporate tax rate under the Camp discussion draft, we assume that government transfers are adjusted to meet the government budget constraint; after the phase-in period, the corporate tax rate adjusts endogenously to balance the federal government budget, so that dynamic revenue increases are offset by further reduction in the corporate tax rate.
Plowing dynamic revenue into a lower corporate tax rate pushes the rate down to 19.9 percent by year 10. This is not part of Camp’s plan, and the inclusion of it provides a big boost to growth. As well, Camp says nothing about transfer payments. It is not clear how this affects the results.
Second, Diamond and Zodrow find the Camp plan (as modeled) would grow the capital stock in every period, while the JCT OLG model finds the Camp plan would shrink the capital stock in the last half of the 10 year budget window. Certainly, the lower corporate tax rate modeled by Diamond and Zodrow could explain much of this difference. However, in the first five years the two models are assuming the same corporate tax rate trajectory, yet the Diamond and Zodrow model apparently finds larger positive effects on capital (0.5 percent growth in year 5 versus zero or 0.2 percent growth in the first five years according to JCT).
Diamond and Zodrow attribute much of the growth in the capital stock, as well as the growth in GDP, not only to Camp’s lower standard corporate tax rate but also Camp’s “patent box” rate of 15 percent applied to intangible income:
These gains are partly attributable to a reallocation of firm-specific capital (FSK) to the United States in response to a relatively more favorable investment environment due to the reduction in the corporate income tax rate to 15 percent for foreign sales and the elimination of deferral on foreign income related to FSK (holding foreign taxes constant). For example, FSK is 16.7 percent higher 5 years after the enactment of reform, 23.5 percent higher after 10 years and in the long run, relative to the initial steady state. Domestic investment in ordinary capital follows a path similar to FSK (the return of FSK to the US makes domestic factors more productive), as it increases by 6.5 percent 10 years after reform, and by 6.8 percent 20 years after reform and in the long run.
Certainly, patents and other intangibles can be shifted across borders relatively easily, and Camp’s proposal could create those incentives, due to the lower tax rate on domestically sited intangible income and the elimination of deferral resulting in a higher tax rate on foreign sited intangible income. However, this appears to work mainly through the tax increase on foreign intangible income, based on the JCT analysis. Therefore, the cost of this capital goes up and investment goes down. Instead, Diamond and Zodrow find that U.S. workers and ordinary capital would be more productive under Camp’s plan due to patents and other intangibles sited in the U.S. rather than in the foreign subsidiaries of U.S. multinational companies. It is unclear how Apple workers or Apple machines, for example, would be significantly more productive if Apple patents were sited in the U.S. Are the patents somehow inaccessible currently by the U.S. parent company?
As with the JCT analysis, the Diamond and Zodrow study provides insufficient detail to resolve these and other issues. What is needed is a cost of capital measure, or rather multiple cost of capital measures based on a) the type of capital: ordinary physical capital versus firm-specific capital, b) the location of the capital: domestic versus foreign, and c) the type of firm ownership: C corporation versus S corporation and other pass-through entity types. The JCT at least provides one measure of the cost of capital, presumably an overall business cost of capital for domestic physical capital. The JCT concludes that the Camp proposal would raise this cost of capital and therefore reduce the domestic capital stock. Since Diamond and Zodrow provide no such measure of the cost of capital, it is unclear how they come to the conclusion that the capital stock would grow under the Camp plan.
The bottom line is that all of these dynamic analyses of the Camp plan are important for understanding the actual effects, but they are only useful if they provide a sufficient amount of information to understand exactly what is being modeled and what the overall effects are.
Follow William McBride on Twitter