How StatsCan wants you to pay taxes till you drop
That’s not quite fair. Or rather, it’s not at all fair. Statistics Canada’s employees presumably have their own views on whether taxes should be higher or lower. It wouldn’t be surprising if, as government employees, many favoured higher. But StatsCan doesn’t, nor should it, have an official view on the question.
On the other hand, it has just published a research paper that seems likely to have at least a small influence on the always-simmering debate about how high taxes should be. The paper carries the unsexy title Intra-household labour income responses to changes in tax rates among older workers and it’s by Derek Massacar, an economist working in the agency’s Social Analysis and Modelling Division.
It’s the kind of research that’s a little breathtaking to someone trained in the dawn of the digital age when data involved time series of 20 or 30 years or cross-sections of, at best, a few hundred people. This study looks at the responses of no fewer than 527,286 Canadians, representing a 20 per cent sample of the relevantly-aged population, to the 2011 introduction of income-splitting for seniors paying income tax on their pension income.
No reflection on Mr. Massacar’s work, but I don’t recommend reading the paper unless you’re interested in the finer points of the econometrics. I’m neither an econometrician nor a labour economist but he does seem to address the sorts of methodological concerns my McGill colleagues always ask about when visiting labour economists present their econometric work. So I have no reason to doubt his results, which are that:
• Older workers appear to respond to changes in their tax rates. That’s good news for us seniors. Beyond a certain age, it’s gratifying to be told you’re responding in any way.
• They tend to respond to changes in their average rate of tax—i.e., their taxes as a share of their total income—though not so much their marginal rate of tax—i.e., what they pay on the next dollar of income. This is at least a little perplexing as much economic theory says people should respond to events at the margin. Massacar suggests the fact that they don’t is consistent with other work that finds that when incentive systems are complicated—as modern taxes regimes certainly are—people focus on the average rather than the margin. That’s yet another argument, as if another were needed, for simplifying the tax system.
• For every one per cent increase in their after-tax income, these older workers “are 1.6 to 2.3 percentage points less likely to be employed, on average.”
• In the same way, a one per cent increase in people’s after-tax incomes induces a 1.7 to 1.8 percentage point reduction in their spouses’ likelihood of being employed, on average.
You see where I’m going with this. The paper’s results are that if you reduce older workers’ taxes, thus letting them keep more of their incomes, they tend to drop out of the labour force.
The thrust of a lot of policy these days is to keep people working, in part so the labour force won’t decline and thus bring down potential output, and in part so tax revenues will stay buoyant and enable full funding of all the social programs we older folks consume more than our younger co-citizens. (In Quebec, if we keep working beyond 65, we get a letter of thanks from the minister of… I’m not quite sure which ministry but we do get a letter, which is a rare thing of course. How often do governments say thank you to taxpayers?)
If in fact your goal as a government is to keep people working, how do you go about that? The answer, this study implies, is to keep their taxes up. If raising their after-tax income encourages them and their spouses to drop out of the workforce, then lowering it—by raising their taxes—presumably will have the opposite effect. The paper doesn’t actually recommend that. What it does say, much more neutrally, is that “the results indicate that the tax code is a viable policy lever for influencing the employment and labour income decisions of older workers.”
My own view is that we really shouldn’t be trying to lever the employment and labour income decisions of older workers—or of any workers, for that matter. We should let them compare what they earn from working with how much they like or (at the margin) dislike working and then make their own decisions about what’s best for them. If we’re so concerned about the implications of their decisions for social programs, well, maybe that’s an argument for making them bear more of the costs of those programs themselves, rather than collectively, so they factor such concerns into their decision-making.