2017 NYU Tax Policy Colloquium, week 1: Lily Batchelder's "Accounting for Behavioral Considerations in Business Tax Reform: The Case of Expensing"

Yesterday we had our first meeting of the 2017 NYU Tax Policy Colloquium. This is Year 22 of the long-running series, so we're ahead of Survivor in longevity, albeit behind it in the numbers both for separate seasons (they've had 34) and separate episodes. But we're probably ahead in screen time (so to speak), as our sessions are 110 minutes long and theirs are usually just 45 minutes or so, leaving aside commercials. Then again, we don't have "Reunion Episodes" after the last session.

On a more serious note, I'm glad to be co-teaching with Rosanne Altshuler, and to have had the above-titled paper by Lily Batchelder at our first session. (BTW, the paper has just been posted on SSRN, and is available here.)  I might also mention here - indeed I evidently am mentioning here - that next year and presumably thereafter I will be co-teaching the colloquium with Lily.

Here are 5 thoughts that I had in relation to the paper, although I was not this week's lead discussant:

1) The paper rightly uses a budget-neutral framework for assessing alternatives. It compares a corporate tax system with economic depreciation and a lower rate, to one with expensing but a higher rate. (The higher rate makes the two alternatives budget-neutral.) This is the right framework because the federal government faces an overall budget constraint, making it useful to decide which of these two alternatives is better for the same bucks. Plus, there may also be an underlying political decision regarding how much the business or corporate sector should pay.

The paper, rather than asking which of these two choices is better overall (note: it also discusses using an investment tax credit in lieu of expensing), asks which would lead to greater U.S. investment. The standard view is that this would be expensing. But the paper notes that public companies whose managers care only about currently reported book income, rather than about the companies' true long-term tax burdens, will be wholly unmoved by expensing, for the simple reason that accounting rules completely ignore its effect, by reason of their treating the time value of money, with respect to when one pays taxes, as zero (!!!).

This is great stuff and an important contribution. But the current policymakers in Washington aren't necessarily doing tradeoffs under an overall budget constraint. There may be some limits in the back of their minds, or else they would simply make the corporate rate zero, but I think these are optical and have nothing to do with actual numbers (which they will finagle if they need to, never mind the Senate's Byrd Rule). So I suspect that, in their framework, such as it is, a lower rate and expensing are viewed additively, not as alternatives involving tradeoffs.

2) The paper makes some good points about transition issues. E.g., why lower the rate that corporate investments made in the past will make in the future? A second transition issue is: Why give companies a windfall from having deducted accelerated depreciation against a higher rate in the past, and then getting to include the resulting income at a higher rate in the future? The Reagan Treasury, in their 1985 "Treasury II" tax reform plan that eventually gave rise to the Tax Reform Act of 1986, tried to address this issue through what they called "depreciation recapture." In principle, both of these issues might be worth addressing if we lower the corporate rate, but don't hold your breath and expect this to happen.

3) After noting that the accounting rules deny companies any credit for the genuine economic benefit that they receive by reason of lowering their tax burdens in present value via expensing or accelerated depreciation, the paper further explains that investors would not easily be able to figure out the truth for themselves via things that are discernible from the financial statements. This leads to the question: Why don't companies try to communicate this to investors somehow? It seems as if they "should," and yet apparently they don't.

4) Suppose we agree that, because accounting treatment dwards economic reality in shaping both company behavior and investor perceptions, the tax system fails to generate the positive investment response that it "should" get by reason of allowing acceleration or expensing. Does this suggest that there are further opportunities the tax rules could try to exploit? E.g., suppose tax prepayment were required, way in advance, in zero-interest-bearing accounts, and that the accounting rules treated these as irrelevant until the tax liabilities actually accrued. Then the government would be gaining present value tax revenue, and (in the pure case) the companies would be acting as if they hadn't lost anything in present value. The example is deliberately fanciful and indeed unrealistic - but the point it makes is that, insofas as the accounting rules induce companies to ignore time value considerations with respect to their tax liabilities, there might be opportunity for policymakers to make greater use of the lacuna than merely by using economic depreciation in lieu of expensing.

5) As per my prior blog post describing a (sort of) new corporate tax reform idea, one might conceivably want to tax the normal rate of return, when earned by companies, at a lower rate than the tax rate on rents or extra-normal returns. One way to do this might be by allowing companies interest on basis, but taxing that interest at a lower rate than the corporate rate generally, The paper pushes towards taxing the normal rate of return at the same rate as rents, but one might nonetheless end up in an intermediate position.

UPDATE: Lily blogs at Tax Vox about her paper here.