Tax Shifting

Earlier this week, the Globe and Mail’s Economy Lab published a piece by Stephen Gordon arguing that high income and corporate taxes won’t generate much revenue.  Gordon used used the metaphor of Jean-Baptiste Colbert’s (finance minister to the Louis XIV, the “Sun King”) that the art of taxation was like plucking feathers from a goose: “ obtain the largest amount of feathers with the least possible amount of hissing.”

It’s a quaint and vivid description, but it’s not appropriate.  Money is fungible—it can be shifted around with relative ease—while feathers in a goose are not.

And it’s this shifting around of money between forms of income and between countries that explains a lot of the “elasticity of taxable income”.

Globe columnist Neil Reynolds also frequently uses the Laffer curve argument—when tax rates are lowered, revenues sometimes increase—and attributes this to the unleashing of entrepreneurial activity.   But most of this is due to the magic of tax accountants rather than the magic of the marketplace.

Studies found that after top personal income tax rates were cut in the United States in 1986, most of the increase in reported individual income of high-income households was due to timing and particularly shifting of income—for example, from the corporate tax base to the individual tax base—and not from additional labor supply.

The opposite has happened in Canada.  With federal corporate income tax rates at 15%, almost half the top federal personal rate of 29%, it’s profitable for those with the means to channel and retain income through a corporation instead of through personal income.  Viola! It appears corporate revenue has increased as a result of tax cuts, but much of it is merely shifting of income.

Money isn’t just fluid, it’s also got a sense of gravity, seeking out lower tax rates.  This is why a fundamental principle of good tax policy is that income from different sources should be taxed at relatively similar rates and loopholes should be minimized.  “A buck is a buck”, as the Carter commission put it forty years ago.   Unfortunately, this principle was too often ignored as governments became infatuated with supply-side economics.   As a result, much effort goes in tax shifting and tax accounting, which may profit a few but are deadweight economic losses to society.

Of course there are limits to how high tax rates can go.  If taxes are too high, opportunities are available, and taxes are considered unfair, avoidance will become widespread.   Inequitable tax rates and turning a blind eye to tax havens undermines the credibility of a fair tax system, encouraging more avoidance.

A recent paper by Nobel-prize winning economist Peter Diamond and Emmanuel Saez, The Case for a Progressive Tax, acknowledges the phenomenon of tax shifting, distinguishing between elasticity of taxable income and elasticity of economic activity.   Using data from the United States they estimate an optimal marginal tax rate for the top 1 percent in the 48 to 76 per cent range, higher than the top marginal rates in much of Canada.

But just who are these 1 percent or 0.1 percent who have acquired an outsized share of income gains over the past decade?   If tax rates are increased, will they really start working much less?   Will millionaire hockey players play fewer or lower quality games?

As Laura D’Angelo Tyson, Clinton’s former chief economic advisor, and  Paul Krugman recently pointed out, a majority of the 1% and about 70 percent of the top 0.1 percent are finance professionals, executives, lawyers or in real estate“: corporate suits”, essentially.

If higher tax rates lead to less resources going into high profit mergers and acquisitions on Bay Street and more into productive investments on Industrial Avenue, and less money going into CEOs salaries and more into workers’ wages, is that really a bad thing?


  • I always understood Colbert’s quote, which I used on this blog a couple of years ago, to be about the politics of taxation rather than the elasticity of different tax bases.

  • The Diamond and Saez paper is interesting, as much for what it does not prove as what it does prove.

    1) The 48% to 76% top marginal tax rate range depends critically on use of a single-period model (p. 175). Implicitly this assumes labour will not move to jurisdictions with lower taxes. This is empirically false, as the number of U.K. rock stars tax-exiled to the U.S. and Switzerland in the 70’s proved.

    2) The 48% to 76% top marginal rate range is derived for labour income only. Pages 177-183 of the paper are devoted to arguments that capital income should be taxed, despite the Chamley-Judd and Atkinson-Stiglitz results, but come up with no concrete numbers. They conclude (on page 183):

    “The bottom line is that uncertain future earnings opportunities argue against zero taxation of capital income…”

    but this is obviously a very weak assertion. They certainly don’t justify anything like a 48%-76% marginal rate for capital income.

    Kenneth Carter said “a buck is a buck”, but he was an accountant, not an economist. As the Diamond and Saez paper makes clear when read carefully, to economists a buck is not always a buck.

Leave a Reply

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