This past Tuesday, we had our first public session of the year for Year 28 (!) of the NYU Tax Policy Colloquium. Our guest was Bill Gale of the Brookings Institution, discussing his co-authored (with Samuel Thorpe) paper Rethinking the Corporate Income Tax: The Role of Rent-Sharing.
A couple of preliminary side notes: Public sessions will be available through Zoom while they are live. We had some technical issues with this at the session, which I am hoping will be cleared up in the future. There are also some issues regarding remote participation in live vs. hybrid classes that I am trying to get my hands around.
Second, we had a post-session dinner with 9 attendees (students plus colleagues at NYU and elsewhere), for the first time in 3 years. What a relief to be lurching towards post-pandemic normality, even if (given Covid's continued threat) we are not entirely there yet.
Okay, onto the paper itself. It raised a bunch of interesting issues, and I will offer a brief run-through here regarding the main ones and my thoughts about them.
1) Background - What the US corporate rate should be is a big political issue these days. The 2017 act lowered it from 35% to 21%. But it likely would have been raised somewhat, either this year or last, by the Democratic majority in Congress if not for Manchin's and Sinema's opposition. If the Democrats keep the House and add at least 2 Senate seats in the 2022 elections, it is plausible that the corporate rate will indeed be raised next year. So there is a real-time public policy debate going on here.
There is also a rich and ongoing economic literature regarding the incidence of the corporate tax. I see this paper as fitting into what I would call "Stage 5" of this debate.
Stage 1: As in the classic Harberger 1962 paper, the view that saving and investment are inelastic in a closed economy supports the conclusion that the incidence of the corporate tax falls on all investors, suggesting that it is progressive.
Stage 2: A shift to the view that the US has a small open economy in which investment capital can easily exit leads to the view that the incidence of the corporate tax falls on local resource owners (including those who would supply labor), suggesting that it is not so progressive.
Stage 3: The view that corporate profits consist largely of excess returns that are reasonably viewed as rents supports viewing the corporate tax as borne by shareholders, suggesting that the corporate tax is progressive (as well as fairly efficient).
Stage 4: The view that rents are to a significant degree shared with workers, whose wages rise with the profits that result from earning rents, arguably suggests that the tax might not be so progressive after all, at least if one thinks about progressivity in standard "capital versus labor" terms.
Stage 5: Here is where the paper comes in. Relying on empirical evidence that excess returns are mainly shared with high-income labor, and basing vertical distributional rankings on ECI (expanded cash income), the paper concludes that the corporate tax is actually fairly progressive after all, and indeed comparably so to where the incidence / progressivity literature was at Stage 3.
The core results are provided in Table 2 at page 36 of the paper. Choosing the values that the literature arguably suggests are most plausible (regarding both the % of rents that are shared, and who gets them as between higher-wage and lower-wage workers), the corporate tax is arguably slightly more progressive than prior Brookings models (with rents but not rent-sharing) had shown, albeit slightly less progressive as to the top 1%.
This supports taking a favorable view of increasing the corporate tax rate, at least within the range that Democrats appear to be contemplating, and indeed all the more so if one views the excess returns as rents that can efficiently be taxed. (More on that below.) Even the slight estimated decline in progressivity as to the top 1% arguably calls merely for using different instruments to reach that cadre, if one's normative views support doing so.
Before moving on to other issues, I'll just mention one concern about the data. Even if one agrees that economic income is the right vertical rankings metric, ECI inevitably falls short of it, in particular by reason of its excluding unrealized appreciation. This could conceivably become a significant concern, upon a fuller evaluation than one can give it in a law school colloquium session.
E.g., consider that owner-employees' ECI depends on their W-2 wages, plus various other things but not unrealized appreciation generally. This could be an especial concern when the likes of a Jeff Bezos or an Elon Musk or a Larry Ellison really doesn't care how much he pays himself, since to a degree (despite less than 100% ownership) it is like the left pocket sending funds to the right pocket.
I am not entirely sure how stock option grants fit into ECI. My guess is that they might enter into it when they become includable for tax purposes. But stock option tax planning often results in their being taken into income at far less than what one might think is actually true value as discernible to insiders at the time.
To be sure, an owner-employee who underpays herself given her benefit from stock appreciation in effect causes the Brookings model to classify the underpaid earnings in entity-level capital income, which it apportions mainly to high-income individuals given its empirical assumptions. But at this point it occurs to me - and the relevance of this point requires further reflection than I have had time to give it - that rents are shared very unequally among shareholders.
This is the familiar "early bird" point. Suppose we agree that Amazon and Apple are earning huge excess returns that we might or might not call rents. Should you and I immediately start buying lots of Amazon and Apple stock? Not necessarily, because the expected future returns are built into the stock price. It's the early movers, the people who had shares before these firms' outsized success became general public knowledge, who captured huge excess returns relative to what they paid for the stock. (Leaving aside for now the important "sweat equity" point that they may have invested labor for which they were fully compensated at the value that became manifest ex post - I will get to this in a bit.
Is the "early movers" point an important one that might lead corporate tax incidence models, even without rent-sharing, to underestimate the incidence of the corporate tax. Suppose that the early bird windfalls go especially to people who turn out (because of those returns) to be higher in the distribution than shareholders generally. I'll just note this question for now and move on, albeit in my next post rather than this one.