Asking whether ‘corporations are people’ is not a helpful question – Part I: the incidence of corporation tax
ROMNEY: We have to make sure that the promises we make — and Social Security, Medicaid, and Medicare — are promises we can keep. And there are various ways of doing that. One is, we could raise taxes on people.
AUDIENCE MEMBER: Corporations!
ROMNEY: Corporations are people, my friend. We can raise taxes on —
AUDIENCE MEMBER: No, they’re not!
ROMNEY: Of course they are. Everything corporations earn also goes to people.
ROMNEY: Where do you think it goes?
AUDIENCE MEMBER: It goes into their pockets!
ROMNEY: Whose pockets? Whose pockets? People’s pockets! Human beings, my friend.
(H/T Brad Plumer)
There is currently a legal dispute running in the US over the ballot for the Republican South Carolina Presidential Primary. Comedian Stephen Colbert, or rather, Colbert’s Political Action Committee Colbert Super PAC is attempting to reverse a court decision whereby a question where voters are asked to choose whether ‘Corporations are people’ or ‘Only people are people’ was removed from the ballot. I’m not going to get into the semantics as to whether it is or is not correct to say that corporations are people, because I believe it is a trivial linguistic distraction* from two vitally important real issues – first, who actually pays corporation tax (the subject of this post), and secondly whether corporations should be allowed to make campaign finance contributions (the subject of a follow-up post I am still working on).
Who pays corporation tax?
The short answer is, I’m not sure. To say that corporations pay corporation tax isn’t very interesting. Mitt Romney in the above quote is entirely correct – just by levying a tax on something does not in and of itself tell you who pays for it, and all taxes are paid by people. To take a silly example, it’s ludicrous to think a tax on cigarettes is paid by cigarettes. A tax on cigarettes is paid by some combination of the people who buy cigarettes, the owners of the companies that make/sell cigarettes and the workers of the company who make/sell cigarettes.
In Econ 101 (or, in my case, in IB Economics at school), you are taught that the ‘incidence’ of a tax is determined by the relative elasticities (i.e. responsiveness of quantity to changes in price and vice-versa) of demand and supply. To get the gist of why this is true, think for a second of what happens with cigarette tax. Cigarettes are addictive, and hence demand for them is ‘price inelastic’ – that is to say, the amount of cigarettes people demand is not very responsive to price. For simplicity, let’s posit that demand is ‘perfectly price inelastic’, which means that price has no effect on the number of cigarettes bought. Let’s say the government then imposes a £1 tax on each pack. If you are in the business of selling cigarettes, you know that people will buy x cigarettes at any price. So, you will raise the price of cigarettes as high as competition allows. Cut a longer story short, all businesses that sell cigarettes will raise the price of cigarettes by £1**. The consumers pay the tax. If instead the supply of cigarettes was perfectly price inelastic, the entire tax would be borne by the supply side. In the real world where all supply and demand curves are not perfectly price elastic or inelastic, the tax is paid by some combination of the two.
In the case of corporation tax (i.e. a tax on profit), who pays the tax will be either the buyers of whatever the business sells (if their demand for the product is price inelastic, the corporation can pass on the cost of the tax to them), the owners of the business (the relevant elasticities here is whether they can remove their capital to invest in something else where they won’t have to pay the tax, or to opt to consume instead of invest) and the workers of the business (the relevant elasticities here are over the capital owners demand for labour and the workers willingness to supply labour. Essentially, the bargaining power of labour versus capital).
One case that can be made that insofar as the tax falls on the corporation rather than its consumers is that whatever part is paid by the business is likely to fall on the workers because capital has bargaining power. Capital can go abroad to where taxes are lower, which is considerably more difficult for labour, or it can be consumed instead of invested. But capital isn’t perfectly mobile, so capital will pay some of the tax. How much I really have no idea, because I don’t know all that much about the relevant elasticities. I used to find the capital mobility argument quite persuasive, and concluded that corporation tax was probably not nearly as progressive as most people thought it was (not to mention the fact that if you are worried about increasing bargaining power of labour over capital, then you should be worried about the incidence of corporation tax shifting more towards labour). However, this morning I read an excellent paper from the non-partisan Tax Policy Center (again, H/T to Brad Plumer) which has affected my views somewhat. Near the beginning, it affirms the severe difficulties in figuring whether the tax incidence falls on labour or capital
Until recently, researchers built models of varying complexity to predict the impact of the corporate tax on the equilibrium allocation of capital and to measure the effect of the consequent change in capital allocation on wages, output prices, and return to capital. Researchers then used the models’ results to draw conclusions about the relative burden falling on capital versus labor. These theoretical papers tended to reach very different conclusions about the burden of the corporate tax depending on model structure and underlying assumptions.
In the past several years, researchers have attempted to estimate the incidence of the corporate income tax empirically. The primary approach used in empirical papers has been to examine cross-country variation in the corporate tax rate over time and measure subsequent changes in wage rates. All of the recent empirical papers have found that corporate taxes lead to depressed wages, but critics have questioned the validity of the empirical methodology. To date, there remains little, if any, consensus about who bears the burden of the corporate tax.
However, the author (Benjamin Harris) points out that capital and labour income are highly correlated – those with higher incomes also have higher capital incomes and vice-versa. Therefore, whether the tax is a tax on labour or capital is somewhat irrelevant to its progressivity. Even a flat percentage tax on labour is still ‘progressive’*** (although that is not the exact point Harris made, it remains true). Low-income workers would still pay some corporation tax if a portion of the incidence falls on labour, but most of corporation tax is probably paid by rich people.
However, the number of people who realise that poor people also probably pay corporation tax is pretty small. If you are persuaded, welcome to the minority. Reality is almost always more complicated than you initially suppose. Furthermore, insofar as one might advocate for raising or cutting the corporation tax, what one would in essence be advocating is for some change in the mix of taxation on income, capital gains and consumption, but you don’t know how much of each or who really pays for it. As a big fan of tax transparency (for essentially Rawlsian reasons, as the incidence of taxation on persons is highly relevant to its justness), I’d probably count this as mitigating in favour of having more of those kinds of taxes (for each of which it is relatively easier to figure out incidence) rather than a corporation tax. But that reason is probably not decisive.