Alex Himelfarb has an opinion piece in today's Globe and Mail on the 'anti-tax' sentiment that has grown to play such a dominant role in Canadian politics. Although I agree with his diagnosis, his prescriptions are not so much a plan for countering anti-tax sentiment as they are a symptom of its hold on public discourse.
We have to be smart about taxes and we will all have to carry the burden. The consensus among economists was that cutting the GST was a mistake, and most would also defend the HST. And sooner or later we are going to have to put a price on carbon to share the costs of a new economic and energy paradigm.
What is disappointing is where he goes with this: calls for increased taxes on the rich and on corporations. I don't have a problem with increasing tax rates on high earners, although I have my doubts as to how much revenue they will generate and as to just who will actually pay them ([1],[2]). And although the suggestion that increasing corporate tax rates is harmless because they don't affect employment betrays a fundamental misunderstanding of the issue, that's not the problem here, either.
No, the problem is that Himelfarb proposes the tax measures that are most likely to appease anti-tax sentiment: he suggests increasing taxes that almost everyone thinks they won't have to pay. Increasing taxes on the top 1% doesn't inconvenience the other 99% of the population. And if you're part of the vast majority of voters unfamiliar with the economics of tax incidence, you might be tempted - wrongly - to think that you won't have to pay corporate taxes, either.
Himelfarb does well to remind us that there is no free lunch, but his point is greatly blunted by offering instead lunches at a 99% discount. The real challenge is to persuade voters to accept the responsibility of paying full price for what goverments provide.
This post is a prime example of Autisme-economie.
Posted by: Robert McClelland | October 15, 2011 at 09:18 AM
@Robert McClelland: Huh? Stephen is trying to make the point raising corporate taxes is not a free lunch which is got nothing to do with the article you link to.
Posted by: DavidN | October 15, 2011 at 11:20 AM
Stephen, I only know taxes and macro as a layman, though I do understand incidence, etc. I agree that higher taxes on the rich are not necessarily a good idea, but from a macro perspective is it different when many high earners are effectively public servants or government contractors?
For example, in have-not regions like Manitoba, the Maritimes, and the north, a huge proportion of high earners (e.g., over $200k/yr) are people who work for the government, bill the government, or do contract government work. This would include doctors, executives of provincial crown corporations, many lawyers, most other health care professionals, and companies whose major clients are governments.
I'm not getting into an argument about the rightness or wrongness of getting rich off the taxpayer (yet), but is there a macro difference between taxes on high earners from govt sources, and high earners from private sector sources?
Posted by: Shangwen | October 15, 2011 at 12:33 PM
Shangwen: "I'm not getting into an argument about the rightness or wrongness of getting rich off the taxpayer (yet), but is there a macro difference between taxes on high earners from govt sources, and high earners from private sector sources?"
There is a huge difference between high income earners who can easily shift their income to other jurisdictions or under the table, and high income earners who can't.
Someone whose income comes from government sources can't easily move their income out of the country or evade taxes - unless they do so through some kind of corporate structure i.e. have two companies, one based in Canada, and one based in a low-tax regime like Barbados, and the Barbados company bills the Canadian company for some kind of services, thereby shifting taxable income from Canada to Barbados.
Posted by: Frances Woolley | October 15, 2011 at 12:58 PM
Well, optimal tax theory would tell us to have *lower* taxes on high earners. Because they are the most productive, they should have the strongest incentive to work more. Take that conclusion as you will.
Posted by: Stephen Gordon | October 15, 2011 at 12:58 PM
Economists tend to underestimate how much the anti-GST sentiment has to do with how annoying it is. Folks see it on every bill, and as the last item it's often straw that breaks the camel's back and so becomes the scapegoat for price increases that have nothing to do with the tax at all. It's a good tax but, people might rationally choose to pay higher income taxes for the same services to avoid the annoyance of it all... and that anti-GST sentiment quickly becomes a general anti-tax sentiment.
To reasonably counter antitax sentiment, maybe replace the GST or hide it behind a full price law...
Posted by: LJ Gould | October 15, 2011 at 01:29 PM
Frances has a post on that point.
Posted by: Stephen Gordon | October 15, 2011 at 01:38 PM
Stephen: the "high earners" we are incensed are either beneficiaries of rent ( artists and sports people whose "productivity" is the result of better technical ways of distributing their product) or CEO whose main productivity is about inventing even more devious ways of looting the till of the corporations. For each Steve Jobs, there are ten Bob Nardelli.
Sorry to add,but this comment is well below your average...
The weather is fine in Québec City this afternoon. Care for a walk and discussion?
Posted by: Jacques René Giguère | October 15, 2011 at 02:09 PM
Stephen "Well, optimal tax theory would tell us to have *lower* taxes on high earners."
There is, though, the new thinking on capital taxation sparked by Koherlakota's 2005 Econometrica piece.
I think there's now less certainty about whether or not results from one period models generalize to dynamic situations.
Posted by: Frances Woolley | October 15, 2011 at 02:33 PM
I guess the question is how much of a high earner's wages are really due to high marginal product. I read somewhere recently that c suite compensation is set by benchmarking against peers and not MP. Comp committees all aim to pay at or above the median under the assumption that their managers must be rock stars. Result: upward wage spiral. If a wage spiral for workers must be broken with extreme prejudice, then why not give the c suite the same treatment?
Posted by: Patrick | October 15, 2011 at 04:46 PM
Jacques René: My point wasn't that increasing taxes on high earners is a bad idea.
Posted by: Stephen Gordon | October 15, 2011 at 06:33 PM
I just don't know who to believe about the incidence of corporate taxes. I've read Stephen Gordon's posts. But I've also read pieces such as the following from the Congressional Budget Office:
"as long as countries tend to choose tax rates similar to each other, which appears to be the case, the world becomes like the original closed economy, a model stressed by Harberger, with the burden falling on capital [as opposed to labour, and thereby the corporate tax being progressive]."
- http://www.aei.org/docLib/20080404_GravelleandHungerford.pdf
"Adjusting the estimates from the studies to reflect central empirical estimates of key elasticities suggests that capital bears the majority of the corporate tax burden."
- http://www.cbo.gov/ftpdocs/115xx/doc11519/05-2010-Working_Paper-Corp_Tax_Incidence-Review_of_Gen_Eq_Estimates.pdf
I am a law student, not an economist. What ultimately makes me sceptical about Gordon's hypothesis (about the incidence of corporate taxes) is the amount of time that I have spent in tax classes learning about how to use corporations to act as a tax dodge and have people pay tax on what is basically income at the rate for dividends (and shift the realization of that income to a time when they'll pay at a lower rate).
Posted by: slantendicular | October 15, 2011 at 08:11 PM
Well, if theory doesn't give an unequivocal answer, do what I do: look at the data.
Posted by: Stephen Gordon | October 15, 2011 at 08:26 PM
slantendicular: I think what you say is consistent with Stephen's post. The hypothesis is that corporations pass-on increased income taxes by reducing/not increasing worker's wages - not shareholders or management. What you're seeing is people avoiding increased individual income taxes on wages by pretending to be shareholders. It's the flip side of the same coin.
Posted by: Patrick | October 15, 2011 at 08:37 PM
slantendicular is quite right about the true purpose of corporate income taxation. It is meant to be a completeness tool so that corporations cannot be used as as efficiently as tax planning tools. If money received as income is taxed but money held within the corporation is not then the difference is lost as tax revenue. You need to integrate the corporate and personal systems in such as way that there is no tax incentive to leave money within a corporation or to use a corporation for tax planning. Money received should be spent on investment or paid as dividends. So under this reasoning the optimal corporate income tax rate is not zero.
Of course in reality this completeness is not 100% efficient but the point remains.
Sometimes the brutal truth is that in order to get $100 of tax revenue we need to collect $100, deadweight losses be d**ned.
Patrick, LOVE your line about breaking c-suite wages with extreme prejudice!
Posted by: Determinant | October 15, 2011 at 10:29 PM
Everyone thinks they are an expert driver and everyone else is a moron. I wonder if there is a similar syndrom with taxation and services/transfers where everyone thinks they are being robbed blind and getting nothing in return, while everyone else is a welfare queen lapping a the trough.
Posted by: Patrick | October 16, 2011 at 01:17 AM
Patrick: "welfare queen": especially if you know what "queen " means in Southern U.S. White english...
Posted by: Jacques René Giguère | October 16, 2011 at 01:24 AM
slantendicular: That's pretty astute for a law student. I'm familiar with the CBO paper you refer to. They generalize the original Harberger result to show that the incidence of the *average* global CIT is on capital but that deviations from it fall principally on labour. Drop your tax rate below the mean and domestic labour gains but at an even *greater* cost to foreign labour. That's a Harberger triangle.
Stephen: The empirical results are of course all domestic only. Ie. they fail to take into account the impact of a lower CIT on the rest of the world. If everybody cuts their CIT (hey, its a competitive world) all you will get is a lower average rate, and the incidence of the reduction is squarely on capital. What you are saying is equivalent to claiming that one company in a market should slash its prices because it'll drive up their sales, completely ignoring any possible price response of competitors. So when you advocate slashing CIT based on the data, I think you ought to mention that the general equilibrium results are likely to be a lot less rosy. The only theoretical paper in your bibliography, Feldstein 74, finds that despite the use of a model carefully constructed by a conservative economist, the incidence was still principally, though no longer *entirely* on capital.
One other thing. For Canadian companies that extract natural resources the CIT has characteristics of a land value tax. It therefore is far less likely to be subject to the open economy results since land is famously inelastic, and the land tax equally famously efficient and incident on capital.
Posted by: K | October 16, 2011 at 01:44 AM
slantendicular,
But keep in mind that incidence is a dynamic concept. A corporation that shifts the incidence of taxes to its employees or consumers (through higher wages or lower prices) still has an incentive to reduce its tax burden through planning techniques (since its prices and wages are, at any particular time, given. I the tax savings is a one off event, say a transaction, that doesn't change the dynamic equilibirum). And keep in mind as well that the people making the decision to hire the tax lawyers to do corporate tax planning are also wage earners (and therefore have an incentive to drive down the tax bill, even if the savings to the corporation are offset by higher wages - i.e., their bonus).
Also, keep in mind that the type of tax dodge you describe is often used in the owner-manager context, where the distinction between the incidence on shareholders and workers is irrelevant.
Posted by: Bob Smith | October 17, 2011 at 01:32 PM
Would it be arrogant of me to suggest that the majority of Canadians do not understand the economics being discussed here? Possibly. But even if that is true I think it might be beside the point.
Lots of us balk at "paying full price for what goverments provide" because we know that the services provided are of poor quality or are plainly destructive, or because we know we have no franchise. When we notice, for instance, that some of the people in government are rewarded with quite good salaries and other benefits to harass legitimate demonstrators in Toronto, we develop a sense of who it is that those people are working for, and it isn't those of us with small incomes.
Posted by: Bill Bell | October 18, 2011 at 03:37 PM
"There is a huge difference between high income earners who can easily shift their income to other jurisdictions or under the table, and high income earners who can't."
True. Which means that taxes on most university academics could be safely doubled or tripled with no adverse consequences.
Onn the other hand, corporations can and will finds ways to hide income or relocate in order to avoid higher-than-acceptable taxes.
Posted by: richard | October 19, 2011 at 08:22 AM