Dalton McGuinty’s minority government in Ontario has agreed to the NDP’s new 2% tax on income over $500,000 to get the budget passed. However, according to Scott Stinson this “is effectively a tax increase of 3.12% because it will be imposed before a high-income surtax that already exists.” McGuinty expects this new tax to bring in an extra $470 million per year, and I’m reminded of the Laffer curve as a reason why this added revenue may not fully materialize.

It’s important to consider the possible secondary effects of any change. We tend to take for granted that increasing tax rates will produce more revenues, but the simple example of the Laffer curve shows that this isn’t always true. If the tax rate is 0%, tax revenues will be zero. But if the tax rate is 100%, tax revenues will also be zero because nobody would bother to work. So, somewhere between 0% and 100% tax rates, tax revenues stop increasing as the tax rate increases.

Getting back to the new tax increase, the obvious secondary effect is that some high-income Ontarians will leave the province. I have no idea how many will leave, but the number who stop paying taxes in Ontario due to this tax increase will not be zero.

Let’s focus on those whose income is $1 million per year. The new tax aims to collect an additional 3.12% of the amount over $500,000, or $15,600. If a million-dollar earner would have paid $400,000 in income taxes, then having just one of these people leave the province will wipe out the added revenue from $400,000/$15,600 = 26 others who choose to stay. So, the new tax will be revenue-neutral on this group if only about 4% of them choose to leave Ontario.

Trying to pile extra taxes on high earners is a tricky game, especially when they can easily choose to settle in a new province or even a new country.

You need to be careful with that argument. The ratio of leavers to stayers (26 in your example) is inversely proportional to the tax rate introduced. What if the proposed increase was 0.001%?

ReplyDelete@Patrick: I think the argument works exactly the same. If the tax increase is 2000 times lower, then it takes 2000 times fewer people who leave to cancel the effect of the tax increase. This points out the folly of making tiny but visible tax increases.

ReplyDelete@Michael: that's a good point, but I still think this analysis requires a healthy grain of salt, or else you'd arrive at the conclusion that your best bet is a 100% tax on the wealthy, and that a 0% tax increase would be worst of all because it would be impossible to make up the loss from just one person leaving. This line of reasoning falls prey to exactly the same kind fallacy that the Laffer curve addresses.

ReplyDelete@Patrick: I'm not sure I understand your reasoning. With a 100% tax on the wealthy, the proportion that would need to leave to cause a breakeven in taxes collected would actually be less than 100%. But 100% would leave (or stop bothering to work beyond the point where the 100% tax kicks in), and so the 100% tax on the wealthy would lead to lower tax revenues. In the 0% case, presumably 0% would choose to leave and the effect on tax revenue would be zero. For this 3.12% tax increase, the breakeven point is close to 4%. It seems likely that less than 4% will leave (but I'm not certain of this) and so tax revenues will increase. My main point is that some wealthy people will leave and tax revenues will not go up by as much as hoped.

ReplyDelete@Michael: Sorry if I wasn't clear. Let me plug the 100% numbers into your line of reasoning to show where it goes.

ReplyDelete"Let’s focus on those whose income is $1 million per year. The new tax aims to collect an additional 100% of the amount over $500,000, or $500,000. If a million-dollar earner would have paid $400,000 in income taxes, then having just one of these people leave the province will wipe out the added revenue from $400,000/$500,000 = 0.8 others who choose to stay."

Maximizing the tax rate maximizes the denominator on that last fraction, thereby minimizing the apparent damage. That leads to the false conclusion that higher tax rates are better here.

I hope that was clearer.

Sorry, I should have finished by saying that the number who leave is inextricably linked to the magnitude of the tax increase, and it's not at all clear to me that the number of people who would relocate as a result of paying 3% more in taxes would be distinguishable from random noise.

ReplyDeleteHowever, at a high level, you are absolutely right that the revenue from a tax increase is not equal to [tax increase * income subject to tax increase], and in fact could be qualitatively different (as the 100% example demonstrates).

@Patrick: OK, I think I see what you mean. The way I expressed it could cause some people to misinterpret the results. The same thing seems to be happening with the dividend stocks discussion :-)

ReplyDeleteI am confident that a some number of people who were planning to move to Calgary anyway will loudly say that they're packing up because of this tax, and they'll get media attention.

ReplyDeleteI am also confident that no one is going to decide to move to Calgary because of this new tax.

I'm not sure where I stand on this new tax. I generally think our tax rates are high enough, and I'm certainly not interested in paying more. But I hate the "tax flight" argument, because finding real world examples of it are difficult. New York, Boston, LA & SanFran are all massively high tax jurisdictions relative to near neighbours. New York City has it's own income tax! And yet rich people live in, and continue to move to those cities. Why is that? Perhaps it's because tax rates actually aren't very high on the list of factors real people use to make real life decision about where they live.

If someone making $500k/year and living in Toronto is so cost sensitive that a 2% (or 3.2%, or whatever it is) is the tipping point, they'd had left ages ago and moved somewhere with cheaper real estate, where they'd have saved far more money than the tax differential.

@Anonymous: Ther are simply too many people in Ontario to think that nobody at all will move to Calgary for tax reasons. However, if we take it as a given that less than 4% of those affected will leave (or even much less than 4% as you suggest), this still means that the added tax revenues will be less than expected, which is my main point about secondary effects.

ReplyDeleteI'm with anonymous on this one. I don't buy the case of the rich leaving due to a 2% increase. Just think how much it would cost just to sell your house here. If people haven't already fled from McGuinty's taxland, another 2% is almost expected. Now the Buffett tax which I see logic in (30% minimum, no exemptions), will make the rich flee, possibly to Canada.

ReplyDelete@Chris B: It's actually a 3.12% tax increase, and keep in mind that we're talking about people who earn enough to buy a new house every year.

ReplyDeleteI'm surprised that in not only the article, but in all the comments that take issue with it, the Laffer curve is presumed to be something factual, empirical. It's not. Before you reply to me, quick go Google the "neo Laffer curve". It is folly, and falls into the category of appeals to common sense, in my opinion.

ReplyDeletehttp://en.wikipedia.org/wiki/File:Neo-Laffer_curve.svg

@Anonymous: I'm not sure what your point is. I agree that many factors enter into people's willingness to pay taxes. So, the Laffer curve's shape is too simple to reflect the real world. But the end points are correct: both 0% and 100% tax rates will generate zero tax revenue. Also, every increase in tax rate will cause some number of people to choose to earn less or leave. This secondary effect must be considered when projecting expected tax revenues.

ReplyDeleteThat referencing the Laffer curve in a debate about this tax is analogous to referencing Phrenology in a discussion about investor behaviour. People will leave and come to Ontario, taxes collected will fall short or exceed expectations, and all of it will be due to complex system interactions, of which this tax will be a tiny input variable into, and that which cannot be modeled even remotely accurately by the "Laffer curve".

ReplyDelete@Anonymous: Even if it is difficult to separate out the effects of different forces, to deny that tax rates affect people's decisions about working and where they choose to live is incorrect. We may not know precisely what effect a 3.12% tax increase on high earners will have, but it makes no sense to conclude that it will have no effect. But this is what we are doing if we just look at last year's tax returns and calculate the expected increase in tax revenues.

ReplyDelete