Tax choices in the OECD

The CBC's vote compass has attracted a certain amount of attention and not a small amount of controversy. I like these sorts of exercises, even if the questions aren't always well-posed. So here are some 'tax compasses'. I'm not going to even try to set up some sort of interactive questionnaire, because everyone's preferred tax rate will be zero across the board (unless there are questions such as 'what tax rate should other people face?'). Instead, I'll use scatter plots to show the choices Canada has made, and put them in context with the other OECD countries. All data are taken from the OECD tax database.

First up is the choice between the GST/VAT rate and the statutory corporate tax rate. (The effective and marginal effective corporate tax rates will vary. But it will still be generally the case that higher statutory rates mean higher effective and marginal effective rates.)

This is the comparison where the economic efficiency results are most clear: consumption taxes are less distortionary than taxes on capital income, so the best pro-growth rax mix will be in the top left-hand corner of this graph:

 

 

I can't count the times people have pointed to Ireland and/or Iceland as 'proof' that the efficiency results af low CIT rates are wrong, so I'm going to digress on this point. Low corporate tax rates are at best a necessary condition, and they are by no means a sufficient condition, for sustained growth. In particular, they do not prevent banking crises. I'm unaware of any explanation of the Irish and Icelandic crises in which having low corporate tax rates was either the culprit or an exacerbating factor.

Anyway, it's interesting to note that the OECD countries who most recently put together a market economy from scratch - the Czech Republic, Hungary, Poland and Slovakia - all chose higher-than-average VAT rates and lower-than-average CIT rates.

It's clear that we are a long, long way from the top left-hand corner. (The GST/HST/QST rate varies by province. I'm using the GST/QST rate for Quebec, because that's where I live.) Before the round of CIT cuts began in 2000, the combined federal and provincial tax corporate income tax rate was 42%, way out on the right-hand edge of the chart. There are four countries that have both lower VAT and higher CIT rates, and 19 with both a higher VAT and lower corporate taxes. The data in the graph are from 2010, that is, before the most recent round of CIT cuts. Once the three-point reduction comes into effect, Canada's rates will be roughly typical of the OECD.

Personal income taxes (PIT) are more complicated. They are typically the the largest source or revenue, and the tradeoffs between efficiency and equity are much more nuanced. The OECD publishes marginal tax rates for those earning 67%, 100%, 133% and 167% of average income in each country. I'm going to look at the numbers for those earning 67% and 167% of the average. In 2009, average income was $43,568, so these rates are for those earning roughly $29,000 and $77,000. Note that those earning $77,000 are not facing the top marginal rate for Canada.

Here are the marginal tax rates; all PIT numbers are for 2009:

The 45^o line traces out combinations where the both types of earners face the same marginal tax rate. In a progressive system, those with higher incomes are taxed at a higher rate. Apart from five countries in which both pay the same rate, all countries are above that line.

In addition to personal income taxes, employees are also obliged to contribute to programs such as employment insurance and social security. The "all-in" rate adds these contributions to the marginal tax rates:

In Canada, EI and CPP/QPP contributions are only required for income levels up to a certain amount (the 2011 thresholds are $44,200 for EI and $48,300 for CPP), and this seems to be the pattern across the OECD. Since high-income earners are typically above the cutoff, their *marginal* contribution out of an extra unit of income is zero. So even though the marginal rates are generally progressive, these cutoffs mean that the all-in rates are less so, and they even decline with income in countries such as Germany and Austria.

Another thing to consider is the fact that employers are typically obliged to contribute to social security programs; these contributions are calculated as a function of wages:

These contribution rates are generally higher for lower-paid workers. Employees usually don't see these contributions, and may consider them to be not their problem. But of course, the economic incidence of employer social security contributions is not necessarily the same as the legal incidence; it turns out that these contributions are largely passed on to the worker in the form of lower wages. It's probably not a coincidence that the most regressive element of the personal income tax system is the part that is not directly visible to workers.

The total tax wedge is the all-in rate plus employer contributions, less transfers received:

 

Canada appears here as a relatively low-tax country, with rates lower than in other places with which we'd usually compare ourselves such as the US, the UK and Australia. If Chrétien's cuts to personal income taxes were reversed, we would be up around where the US is on the graph.

The top PIT tax is typically a hotly-contested point of contention: what tax rate should the highest earners face? The upsides of high top rates are obvious; the downsides are less so:

• The revenues generated by increasing the rates for top earners are smaller than you might expect.
• When you take behavioural responses into account - and you should - increasing taxes at the top end may have unintended consequences for the distribution of after-tax income.

That said, here is a graph of the all-in tax rates faced by the top earners against the income levels - expressed as a multiple of average income - at which the top rate applies:

Canada is sort of in the middle of the pack here, although if the top rate were increased and were applied at lower incomes, we would probably still be in the range of a typical OECD country.

One of the factors constraining PIT rates in Canada is the fact that there are several countries - especially English-speaking countries - where high earners might move to escape higher PIT rates. What surprises me in these graphs is that Canadian top PIT rates are at or below those of other English-speaking countries.

One way of summarizing Canada's choices relative to those of other rich countries is that if we were a 'typical' OECD country, we'd have

• a higher GST/HST/QST,
• corporate taxes around where they will be in 2012, and
• higher personal income taxes.

 

 

Comments

[laurent sauve]

Excellent post, very informative.

[Shangwen]

Terrific post. Another point about taxing high earners is that their incomes vary a lot more, since the very rich earn the money from investments, business deals, and big contracts. One of the problems that has best California is that, as a state where the top 1% apparently pay 50% of all PIT, the recession brought about a massive drop in state revenue from that source, which might not have happened had the burden been less unequal.

[Leo]

Anyway, it's interesting to note that the OECD countries who most recently put together a market economy from scratch - the Czech Republic, Hungary, Poland and Slovakia - all chose higher-than-average VAT rates and lower-than-average CIT rates.

Could it possibly be because they were getting advice on how to put together a market economy from economists such as yourself, and so chose exactly what you recommend because it was the only recommendation made?

[Mark_dowling]

The Corp Tax axis on graph 1 doesn't read right to me. Ireland looks way further to the left than 12.5pc. Which OECD table was used, please?

[Stephen Gordon]

No, it's 12.5%. Looks right to me.

And Leo, yes, I like to think so.

[Paul Friesen]

Very interesting.

[Determinant]

Extremely interesting. I would not have guessed that Canada has lower taxes once you add up all the "tax-like" levies. The fact that we end up lower than the US is really surprising. Especially as we fund a universal health system which means a lot at the personal and retail level. It's the one government program that really says "I'm getting something for the taxes I pay."

[K]

Great stuff! As always, thanks for all the analysis. On the unintended consequences of high income taxes:

I pay the top marginal rate. My plan for retirement is based on continuing my current rate of consumption until I die. I have it all worked out pretty carefully. The more money I have, the earlier I will retire. My consumption is extremely inelastic as a function of expectation of future annual current income: upwards, since my overriding purpose is to maximize my retirement time and I simply don't need more stuff, but especially downwards as I cannot tolerate any decrease in consumption :-( . So if my wealth decreases, I *will* work longer. It is a painful truth that if you raise my tax rate I will feel forced to work more, not less. I suspect that you would have to raise it to 85-95% before I would seriously contemplate just giving up, accepting a huge decrease in permanent income, and "going Galt". Here's another way to look at it: If my marginal tax rate had been zero until now, I'd quit working. Right now, at the age of 42.

If you do the math, I'm pretty sure it's the most entrepreneurial workers, the ones who take the greatest risk, and who expect the highest wealth growth rate who would be least disincentivized by a high tax rate, since they are the ones who would still be able to affect a significant change in their personal wealth by working. On the other hand, those with high wealth, but low outlook for future income would be most disincentivized - they will simply prefer to consume their wealth.

[Canadianpoliticaleconomy.wordpress.com]

Determinant:

Not to in anyway dismiss the analysis Stephen has presented but part of the reason that Caanadian PIT taxes appear to be comparable to American ones is that this analysis is done at the INDIVIUAL PIT level. On that basis, Canadian personal income taxes certainly are fairly comparable.

Where American personal income taxes are lower however is at the family unit level. Americans can file either as individuals, as head of household or as joint tax filers and it is here that the Americans do have a tax advantage, though as you rightly note, they get less for those taxes and not simply just universal health care.

It's been a while since I've looked at the stats in detail or the OECD website - they moved things around a bit - but the following illustrates the difference:

A single unmarried earner at 100% of average income:
Canada - 30.62%
US - 29.36%

A single income earner, married couple, 2 children with 100% of average income:
Canada - 17.45%
US - 13.72%

[Determinant]

We've been over that. Americans when they file as a family have to pay at 167% of their marginal rate. It works out to a wash.

If you don't have a family it doesn't matter anyway.

K:

If you are paying the top marginal rate after a retirement savings vehicle deduction you are extremely fortunate. Who cares what you invest in, it's just a large tax advantage available to you.

[K]

Determinant: what I invest in? I was only commenting on the impact of the tax rate on my choice of work vs leisure. Nothing about tax shelters. I don't think what I said had any investment implications. And, yes I'm lucky.

[Determinant]

Since RRSP's, Individual Pension Plans and other retirement vehicles are tax deductions, your contributions to these vehicles will not be impacted by increasing tax rates. In fact they are helped by tax increases because of the way a deduction works.

Therefore I fail to see how your ability to retire is negatively impacted by higher marginal rates. I also fail to see the basis of your complaint that you cannot retire immediately since you pay substantial taxes, unless your income is what the income tax system refers to as "unearned income" which doesn't generate retirement contribution allowances.

[K]

Determinant: I really wasn't complaining. On the contrary, I was making the point, as strongly as possible, that lower income tax rates discourage, rather than encourage productivity. Cut my taxes and I'll work less, not more. Though I'd love to pay less tax I have *no* problem with my current rate of taxation. You really misread my comment - probably it was poorly written.

[Determinant]

Ah. My point, in reply to your point, was that wealth accumulation for retirement, due to its unique treatment in the Income Tax Act, has a different calculation framework than simple saving vs. spending of after-tax income. Income tax rates have a much different impact here, due to the treatment unique to retirement savings provisions in law.

Perhaps I was misled by the word "retirement" which has a specific meaning in financial circles.

[westslope]

K:

How much time and energy do you spend avoiding taxes? There is some theory and evidence that high marginal rates encourage greater tax avoidance effort.

Anyhow, a great way to tax you leisure-loving lifestylers is through a high VAT.

[K]

Westslope: I pay an accountant but have very little leeway on taxes (not a business owner). The only form of "evasion" I would ever consider would be to permanently leave the country. But taxes would have to rise a lot. Social/family costs too high.

[Declan]

Thanks for the charts Stephen,

So looking at the first chart, no corporate tax - massive financial crisis, no vat = massive fiscal crisis, best stick with the baby bear approach to things...

[Stephen Gordon]

no corporate tax - massive financial crisis

Um, I really hope you're kidding. Especially after I went through such lengths explaining how policy analysis is done.

[Mark_dowling]

I figured out the axis thing - apparently the vertical is at 10pc not 0, should have spotted that, but it doesn't explain why Chile at 17pc isn't to the left of Iceland at 18pc. Transposition? I would point out that setting the CT axis start at 10 not 0 makes Ireland's CT rate look ridiculously low. If I wanted to see that kind of spin I'd call Berlin or Paris.

[Stephen Gordon]

About the axes: call it a stylistic preference: I try to avoid empty spaces where I can.

As for Iceland, you're right. I had made a graph for 2009 data and updated it for 2010. Along the way, I missed Iceland's increase from 15% to 18%. Dang.

[Declan]

Mostly just teasing Stephen, although I do think the lack of a VAT in the U.S. reflects the challenges of developing a rational tax policy there which is a factor in their fiscal crisis, and I think the low corporate tax rates in Ir/celand are partly a reflection of a push for modernization/liberalization/deregulation that was a primary cause of their financial crises (more so in Iceland, Ireland was somewhat like the tech bubble in the sense of being a real leap forward that just got out of hand towards the end)