Corporate Tax Incidence and Social Democracy

Over at Worthwhile Canadian Initiative, Stephen Gordon critiques the last federal NDP platform’s reference to “Canada’s wealthiest corporations” on the grounds that people, not corporations, own things.

But as Declan points out in several pithy comments on Stephen’s post, corporations clearly can and do own things. The corporations that own the most valuable things in Canada can quite reasonably be described as “Canada’s wealthiest corporations.”

On the semantic point, I agree with Stephen for a different reason. In a debate about the corporate income tax (CIT), I would instead refer to “profitable corporations” because the tax applies to the flow of profits rather than to the stock of wealth. (References to “wealthy corporations” would be more appropriate in a debate about a corporate capital tax.)

However, there is a more substantive difference of views. The federal NDP sees large corporations as powerful economic institutions with significant latitude to make important decisions. Individual shareholders have very little control over them. Corporations retain substantial profits in excess of real investments or distributions to shareholders.

Stephen sees corporations as a legal convenience through which individuals own things and conduct business. In his debate with Declan, Stephen describes corporations as safes in which households store some of their wealth. The corporate form is of little economic relevance.

It seems to me that there are many different types of corporations. For the local paving company or auto repair shop, Stephen’s view is probably about right. For Suncor or the Royal Bank, the NDP’s interpretation stands.

However, even if one completely accepts Stephen’s view, there are still compelling arguments for the CIT. As I noted regarding the CIT for “small business” and as a couple of commentators noted on Stephen’s post, it is a problem if (wealthy) individuals can indefinitely delay paying personal tax on income by keeping it inside a corporation. By comparison, if I decide to keep my income locked away in a safe, I still pay tax when I earn it.

Stephen offers no solution to this problem. The obvious solution is to tax corporate profits at the same rate as personal income, but also give individual shareholders a tax credit equal to the CIT rate if and when they collect dividends subject to personal income tax. That was more or less the system Canada had until 2000, when the federal government started slashing its CIT rate way below its top personal income tax rate.

Stephen’s concern, succinctly expressed by his post title, is that corporations may be able to shift the tax burden. The CIT may be borne not by (wealthy) shareholders, but by (less affluent) workers and/or consumers.

Even so, it is unclear why he supports cutting the CIT. He advocates a higher GST even though it would mostly be borne by wage-earning households. Yet he opposes the CIT because he thinks that it would mostly be borne by wage-earning households.

Of course, the CIT’s economic incidence is an important question. It has inspired much scholarship with few definitive conclusions.

Stephen typically asserts that this question has been settled beyond a shadow of a doubt. However, it is worth noting that the Australian op-ed he approvingly cites concedes that “the evidence is thin” on how the CIT might affect consumer prices and claims only that “the empirical evidence is stronger” on how the CIT might affect workers’ wages.

A critical point is that other taxes can also be shifted. Consumption taxes are likely to be divided between the buyers and sellers of consumer goods. As Stephen notes elsewhere, rich people may be able to shift the burden of personal taxes for high incomes onto other segments of society.

In a world where the market provides a perfect transmission mechanism, the statutory incidence of taxes is ultimately irrelevant. The economic burden of all taxes will be distributed according to various elasticities of supply and demand.

In such a world, it does not really matter which taxes the state uses to collect revenue. If so, the debate about the CIT is essentially just a debate about more or less government revenue. The social democratic goal would be to raise more revenue, using the CIT and/or other taxes, for redistribution through progressive public expenditures.

Of course, in the real world, market mechanisms are far from perfect. Corporations may be able to shift some CIT, but much of it will come out of after-tax profits. To the extent that the CIT is borne by shareholders, it is a highly progressive tax.

Because a large fraction of Canada’s corporate sector is owned by foreign investors, the CIT is an especially good tax for Canada. To quote a 2004 Finance Canada study, cutting the CIT is “a less cost-effective way to improve well-being” because “some of the CIT cut will accrue to foreigners, who own a substantial portion of the Canadian capital stock.”

As I have argued elsewhere, some of that foreign portion will actually accrue to the U.S. treasury, which taxes American corporations on a worldwide basis. The extent of foreign ownership is a key difference between Canada and the Nordic states that bears critically on the CIT’s desirability.

If we assume limited tax shifting, then the CIT is preferable to other taxes in important ways. If we assume unlimited tax shifting, then the CIT is essentially the same as other taxes. Either way, Canadian social democrats should support a higher CIT.


  • Erin: corporations own things, but people own corporations, so eventually it’s people who own the things that corporations own.

    For example, you say “Corporations retain substantial profits in excess of real investments or distributions to shareholders.” That’s true. But what do they do with those retained profits? If they use it to pay down debt, or buy back shares, that increases the value of the remaining shares, so it goes to the shareholders as capital gains, rather than as dividends.

    Now, it’s possible that corporations can be “captured” by people other than their shareholders, like their managers for example (or their unions?). In which case the retained profits of the corporation may end up in the pockets of those who have captured control, rather than the pockets of the legal owners (the shareholders). But they are still people, and people’s pockets.

    One other point: “In such a world, it does not really matter which taxes the state uses to collect revenue.”

    I don’t think that’s correct. If you tax smokes, rather than booze, there will be less smoking and more boozing. And those who supply and demand smokes will lose; while those who supply and demand booze will gain.

    Back to corporations: capitalist corporations are lenders’ cooperatives, as opposed to workers’ cooperatives or consumer cooperatives. My own argument against corporate profit tax is that the tax system should be neutral with respect to the type of economic organisation people choose to coordinate their economic activity.

  • Thanks for your comments. I suggest that a major corporation like Suncor is more than just a free association of lenders.

    Anyway, neutrality between different forms of economic organization supports my position. If I have an unincorporated business, I pay personal income tax on my profits in the year that they arise.

    But if I have an incorporated business with no CIT, I may indefinitely defer paying tax on my profits. A CIT (with a dividend tax credit) treats corporations the same as unincorporated enterprises.

    In commenting on Stephen’s post, you appeared to acknowledge tax deferral as an important problem, “But corporate profit tax seems to be not the best way of handling it.” What alternative solution do you propose?

  • Emergent properties. Institutions have characteristics of their own. You cannot reduce a corporation to a collection of shareholders or even of individual managers, any more than you can reduce a polity to a group of independent individuals. There is such a thing as society, whatever Maggie Thatcher thought. Yes, there are individuals in it, but the legal framework of the corporation (limited liability, legal quasi-personhood etc.), the institutional relationships among the individuals, and the longevity of the corporation as an institution, with rule-sets and information that continue as individuals come and go, create emergent properties that aren’t accounted for by the desires of shareholders or individual corrupt executives.

    Similarly, although it is true that an amoeba is ultimately a collection of subatomic particles, it is not generally possible to usefully describe the behaviour of an amoeba in terms of particle physics. Living things and even their biochemical building blocks have emergent properties which don’t have that much to do with the behaviour of individual electrons in particle accelerators, even though there are certainly electrons involved.

    So sure, “eventually” it’s people who own the things that corporations own. But for practical purposes, no they don’t. They don’t generally even *know* what the corporations own–and when you consider that corporations can own (pieces of) corporations, which can own corporations . . . ultimately such daisy chains generally end in people (although I’d be willing to bet there are corporations which, completely unknown to the employees, are owned by some shares in a Swiss safety deposit box, which are in turn owned by someone who was shot by the mob 20 years ago and didn’t have a will), but how much does this matter for practical purposes? In a very real sense it is the corporation that makes the money, the corporation as instanced by its ever-changing employee base that decides what to do with the money.

  • Erin: I acknowledge that the tax deferral problem is a genuine problem; I’m not sure how important (“big”) a problem it is.

    In principle, I think the correct theoretical solution is to tax individuals on capital gains on share prices *annually*, rather than on *realised* capital gains. Not sure how practical this would be.

    I can’t quite wrap my head around the “CIT plus dividend tax credit” as a solution. But my guess is that it only duplicates the above solution if all shares are held by Canadian taxpayers, who all face the same marginal tax rate. And how it works for shares held inside vs outside an RSP/TFSA (when we want to allow tax deferral) puzzles me even more. Plus, the tax system out to be neutral between dividends and capital gains from owning shares, and I don’t think that is.

    Purple Library Guy: Funnily enough, last night I was working on a post trying to articulate the logic behind the “emergent properties of institutions” argument. I can dream up a model in which an institution does have “interests” apart from any of the people whose interactions socially construct that institution, and that it is *logically* possible to tax the institution without taxing any of those people. But the sort of model needed to generate those results would have to be so weird I’m not sure anyone would believe it. Plus, I think it would mean that people would not construct those institutions in the first place. Essentially, it would be a model in which all corporate profits are rents that get dissipated in rent-seeking, so you can tax corporate profits without taxing any individual.

    Philosophically fascinating (to me); economically less interesting (I think). But I will probably post it in a day or two anyway, just for the hell of it. (And for intellectual honesty, to recognise that my opponents might not be totally out on left field as a matter of logic.)

  • In principle, I think the correct theoretical solution is to tax individuals on capital gains on share prices *annually*, rather than on *realised* capital gains.

    That is an interesting option. Applying it to Canadian shareholders would be conceivable (if possibly impractical). However, how could we make the foreign shareholders of Canadian corporations pay Canadian tax on their capital gains?

    I take your point that the CIT goes too far the other way because foreign shareholders cannot benefit from our dividend tax credit. However, as noted in my post, taxing foreign shareholders (and the U.S. treasury) is one of the chief benefits (for Canadians) of Canada levying a CIT.

    It strikes me that a necessary condition for abolishing or cutting the CIT is to either conclude that tax deferral is not an important problem or be confident that you have a practical solution to it. So, in as collegial a manner as possible, I am throwing down the gauntlet for you to meet that condition.

    Also, I look forward to your forthcoming post on corporations as institutions. J. K. Galbraith wrote a classic book on the subject: The New Industrial State (1967).

    He argued that much modern economic theory misleadingly assumes the “entrepreneurial firm” of Eighteenth Century fame. You and Stephen seem to be asserting that large modern corporations are essentially the same as such firms.

  • I wouldn’t dare pick up that gauntlet Erin 😉 !

    Actually, I’m not sure it’s a fight that economists like you and me would be any good at. There are some questions in tax policy that are best left to hard-nosed accountants who know all the dodges and practical questions; and we are just hopelessly naive innocents in comparison.

    I confess I wasn’t even aware of the deferral problem until Frances explained it to me about a year back. And I promptly forgot it again until I saw her comment on Stephen’s post!

    But my guess is that most people (not you, and not Purple Library Guy above) who advocate a CIT don’t have any awareness of the conceptual/logical problems of arguing that you can tax corporations without taxing people. The slogan “People before profits” betrays the same sort of conceptual problems (or, worse, is a dog-whistle of even worse underlying sentiments – that capitalists aren’t people too). Mrs Thatcher, by the way, should I think be interpreted as merely pointing out that same sort of “category mistake” — that society was not something separate from the interactions of the individuals who comprised it.

    I have to remind myself that my employer, Carleton University, is also a corporation. I think yours is too? My working theory is that universities are typically corporations whose ownership has been captured by the “insiders” — the tenured profs, like me. Rather like Yugoslavian worker cooperatives, in practice. A truly general theory of corporations ought to be applicable to universities and unions as well.

  • I think the tax deferral problem is overstated. The same problem occurs (theoretically) with mutual fund companies, and we deal with it by imposing distribution requirements under the Income Tax Act. Similar but less stringent requirements could be imposed on active businesses.

    It’s also not obvious that Canadian social democrats should be in favour of higher corporate taxes: see, for example


    Lower corporate taxes relative to (for example) the EU could in fact be beneficial to Canadian workers.

  • Nick why would you want a truly general theory of corporations? Once you get to that level of generality it will be rather without content. Think of Williamsonian theory of the firm: an institution to manage ex post facto change. On that definition so is a church, a household and my brain.

    I think you are retro-projecting for the Iron Lady; ie., that methodological individualists like yourself are not guilty of a fallacy of decomposition / composition. I am pretty sure madam Thatcher truly thinks that society is merely the sum of its parts and at best a de-facto whole and not an integral whole. Although to understand the difference between your compositional position on society, the Iron Lady’s and my own would require a much longer ontological conversation. Care to dust off your Blaug?

    I do not agree that the slogan “people before profits” is a solipsism as you suggest. Unless of course you want to argue there is no such thing as not for profit sector. Surely you know the difference between the C.D Howe institute and Proctor and Gamble.

    Further, when the “slogan people before profit” is used, it is used in a manner that does not have a perfectly competitive, neoclassical theory of the Firm in mind. Nor does it take it for granted that the firm, or its putative “owners,” have a right to all the value added produced there from.

    With some irony, that slogan agrees with your take because it makes no distinction between the corporation and its legal owners.

  • He advocates a higher GST even though it would mostly be borne by wage-earning households. Yet he opposes the CIT because he thinks that it would mostly be borne by wage-earning households.

    You have heard my views on these topics enough times to know very well that this is a misrepresentation.

    – Every time I advocate increasing the GST, I add the condition that low-income households be compensated for their lost buying power.

    – The reasons for lowering the CIT are based on economic efficiency.

    And your last sentence is a non sequitur: advocating lower CIT rates is not inconsistent with social democracy. As you know very well, there are many examples of successful social democracies that have CIT rates lower than that of Canada.

  • Thanks for jumping in.

    Every time I advocate increasing the GST, I add the condition that low-income households be compensated for their lost buying power.

    Indeed. Similarly, every time the NDP has advocated not cutting the CIT, it has also been advocating increased financial support for low-income households.

    I have never denied that some social democracies have low CIT rates. My last sentence urges Canadian social democrats to support a higher CIT. The paragraph above it identifies a critically important difference between Canada and the Nordic social democracies.

Leave a Reply

Your email address will not be published. Required fields are marked *