Radical tax policy ideas - or just bad ones?

In a recent article in Policy Options, Professor Tom Kent proposed bewildering variety of tax policy suggestions. Most are, in my opinion, bad ideas. His most radical one is to abolish the corporate income tax and the dividend tax credit.

Now the corporate income tax would not disappear entirely. It would be replaced with a 29 percent federal tax on dividends paid out to "external" shareholders, that is, shareholders who do not pay Canadian income taxes.

The problem with this idea is that it is impossible for a corporation to know whether or not a shareholder is a resident of Canada for income tax purposes. How would this be implemented? Professor Kent suggests charging this external shareholders tax on those who do not have a Canadian social insurance number, on the presumption that any shareholder with a social insurance number is liable for Canadian personal tax. That is just not true - there are lots of people with Canadian social insurance numbers who are not resident in Canada for income tax purposes.

[Amendment added] And in any event, corporations are currently required to deduct "Part XIII" tax at a rate of 25 percent - or less if specified by the terms of an international tax treaty - from dividends paid to shareholders who are not resident of Canada for income tax purposes. So not only would Professor Kent's proposal here not be much different from the status quo, it would require renegotiation of dozens of international tax treaties.

Moreover, Professor Kent's proposal suggests a lack of knowledge of the purpose of Canada's corporate income tax.

Under the current system, Canada's corporate income tax acts as a withholding tax. When a corporation makes a profit, it pays corporate income tax. But the government doesn't keep that money forever and ever - it just withholds it for a time.

When a corporation pays out its profits to shareholders in the form of dividends, shareholders get a dividend tax credit - which credits the individual with the taxes the corporation paid on her behalf. 

Here's a simple stylized example of how it works:

Assume a corporate income tax rate of 20%, personal income tax rate 40%

With the corporate income tax, a company earns $125, pays tax of $25, and then has $100 in after-tax profit to distribute to shareholders:

    company earns                                       $125

    pays tax                                                  <$25>                        

    distributes dividends                              $100

When the shareholder calculates her personal income tax, the dividends are "grossed up" - basically a calculation to go from the dividends paid out to the pre-tax corporate profits. The personal income tax is applied on the grossed up amount, then a dividend tax credit - equal to the amount of corporate tax paid - is applied.  

dividends grossed up 25%      $125

personal income tax 40%      <$50>

dividend tax credit               $25

personal income tax paid      <$25>

The really cool thing is that $125 of corporate income is, in the end, taxed at exactly 40%. $25 gets paid in corporate income tax, $25 in personal income taxes, for a total of $50. (In real life, it isn't always this neat, sometimes the dividend tax credit over-compensates for corporate income taxes paid, sometimes it under-compensates).

Why do it the way that we do right now, instead of the way that Professor Kent proposes? Like I said, it's very hard to identify which shareholders are pay Canadian income taxes and which ones don't.

There's another reason too.

If corporate income doesn't trigger a tax liability until it is distributed to shareholders, corporations have an incentive to retain earnings. 

Remember that taxes deferred are taxes avoided: If income is not taxed until received by shareholders, a corporation can re-invest pre-tax earnings. This saves taxes like an RRSP does.

Why is this a bad idea? Suppose that Old Tech Co makes $100 in profits. Professor Kent's proposal is in place, which means Old Tech Co can retain those earnings and pay no tax. Old Tech Co can reinvest the earnings in a new venture that will generate a rate of return of 5%.

Novo Tech Co also has a new venture that it needs investment funds for. Novo Tech Co's venture will generate a rate of return of 7%. 

Which is better for the shareholder, Old Tech Co paying out dividends - and triggering a shareholder liability - so that the shareholder can take the profits and invest in Novo Tech? Or Old Tech retaining the earnings?

Old Tech: $100 invested at 5% return - $5 payoff.

Novo Tech: $100 dividends paid, $40 personal income tax paid by shareholder, $60 to invest. $60*0.07=$4.2 payoff.

Even though Novo Tech is offering the more productive investment opportunity, if there is no corporate income tax, the shareholder has an incentive to keep their money in Old Tech!

So there are some very serious problems with Professor Kent's corporate income tax proposals.

He has some other radical tax policy ideas. 

He suggests replacing the GST with an expenditure tax, and implement it by estimating expenditures as "after-tax income" less net assets acquired, that is, assets minus liabilities.  

At the same time, I think Professor Kent would like to restrict RRSPs. At least, Professor Kent does not directly advocate eliminating RRSPs, but he seems very unenthusiastic about them:

It [the RRSP] has morphed into a bonanza for people with incomes high enough for them...The program has become an easy, risk-free road for people with good earnings to retire as millionaires.

The first expenditure taxation idea is a non-starter. Any tax that requires the valuation of assets is problematic, as assets are typically difficult and costly to value - there is a reason why income taxes are much more common than wealth taxes.

As my last post pointed out, an expenditure tax is equivalent to allowing people to make unrestricted RRSP contributions. If you like expenditure taxes, RRSPs and TFSAs are your friends - they're the easiest way of implementing progressive expenditure taxes. So these two tax policy proposals are completely inconsistent.

Professor Kent also proposes taxing capital gains equally with salary and interest income - so a $100 capital gain would be taxed at the same rate as $100 in interest income. 

Inflation anyone?

Capital gains on housing would also - after some exemption of $500,000 or $1 million - be taxed. I worry that such a scheme would be thin end of the wedge for advocates of mortgage interest tax relief, and the net result would be a serious erosion of the tax base.

Perhaps I'm just too cranky, too set in my ways, too blinded by preaching tax orthodoxy for years and years. Perhaps I'm just not open enough to new ideas.

But one of Professor Kent's ideas struck me as worth exploring: reinstating an inheritance tax. Oddly enough, that's one of his ideas that hasn't been picked up in the National Post.  

 

Comments

[Stephen Gordon]

One thing that I've wondered about is the 'dual income tax' regime, where investment income is taxed at a lower rate than wage income. A lower investment income tax rate would make it less costly for investors to reallocate their investments from one firm to another. Apparently this is what they have in the Nordic countries, where corporate taxes are lower, and personal taxes are higher. (For the self-employed, they impute some sort of a rate of return on what is deemed to be a capital investment.)

[K]

Why do we have different tax rates for interest and capital gains income? Like you said, inflation eats into capital gains. But it eats equally into interest. And interest bearing instruments can be restructured into capital gains generating ones. What gives?

[Frances Woolley]

Stephen - dunno. In an optimal tax framework it makes sense to tax more elastic things (like capital) less than relatively inelastic things (like labour supply) - which would argue for a dual income tax type model.

K -

- yes, logically we should real interest, not nominal interest. And with real returns falling, and a risk of future inflation, this could start to become a serious issue in five or 10 years time. At some point I want crank through some examples comparing RRSPs and TFSAs in an inflationary environment.

- the other reason for less than 100% taxation of capital gains, which I didn't want to get into, is as follows: capital gains on stocks arise because companies build assets by retaining earnings. Retained earnings have already had corporate tax paid on them. So capital gains have already had corporate tax paid on them.

But it's not quite as easy to show as the argument for giving people dividend tax credits.

[Kevin Milligan]

I think anyone who pats himself on the back for 'outside the box' thinking must first demonstrate that he understands why the existing box was built.

[Robert McClelland]

Another tax idea that favours the wealthy. Why do economists think the people in the most heavily subsidized tax bracket need more breaks. I'm telling you, implement a progressive user fee system and you'll see just how much the wealthy leech off the middle class.

[Frances Woolley]

Kevin - I feel a bit badly beating up on an 88-year old - I'll be lucky if I'm half as lucid in my golden years. But if everyone takes this attitude, then the ideas start entering into the policy debate - as with Bill Watson's op-ed in the National Post - and start taking root in people's imagination.

Robert - if you read Kent's article, you might not find it as pro-business as you think. What do you think of his suggestion of reinstating an inheritance tax?

As for user fees - yes, getting rid of subsidies to the major orchestras etc would take a whack at the wealthy, and personally I'd love to see more user fees on roads (but how progressive? that would involve a creepy amount of personal tracking). But do you think that getting the rich to pay for their own health care would do a lot to enhance the base of political support for our current health care system?

[Robert McClelland]

What do you think of his suggestion of reinstating an inheritance tax?

Ah yes, the fig leaf suggestion that people can point to in order to claim the suggestions aren't as pro-business as you think. As for what I think, it's an absurd idea. Taxes were already presumably paid on that wealth and a simple transference of wealth should not be subject to taxation a second time; nor would it be under a user fee system as income of any kind would no longer be taxed.

...would take a whack at the wealthy...

This system of taxation isn't meant to take a whack at the wealthy. It's intended to make people pay for what they use; it just so happens that the wealthy use a larger share of government services than anyone else.

that would involve a creepy amount of personal tracking

I don't think so. It's fairly evident who uses what. The biggest obstacle would be getting the dopey masses to realize who is using what. Here's a perfect example.

...the government's new focus would be on fighting the sale of contraband cigarettes.

Who predominantly benefits from fighting the sale of contraband cigarettes. Well this is a no brainer since they actually brag about it.

Tobacco giant Philip Morris International, appeared to confirm those numbers, telling shareholders the sale of legal cigarettes in Canada was up 4.2 percent "mainly reflecting government enforcement measures to reduce contraband sales."

So everyone is paying to enrich Philip Morris. Yippee! Under a progressive user fee system, Philip Morris and all the other tobacco companies would be picking up the tab for this instead of leeching off the middle class.

But do you think that getting the rich to pay for their own health care would do a lot to enhance the base of political support for our current health care system?

It's not just the rich who would pay. Everyone would according to their means. And yes, paying only for what you use would enhance support for not only our current health care system but everything else as well. People do grumble about paying their non government bills but they also recognize they are doing so because they've chosen to use a service or buy a product.

[Bob Smith]

Frances,

I don't think that it would be that difficult for Canadian corporations to identify their non-resident shareholders. After all, they ALREADY do that in order to withhold the existing withholding tax on dividends (amongst other things) paid to non-residents (in theory 25%, but generally reduced to 5-15% under an applicable tax treaty). Since, corporation are liable for failing to withhold you had better believe that they can indentify which shareholder are non-residents.

So, in that respect, Professor Kent's proposal is not particularly radical, it really would only involve an increase in the statutory withholding rate to 29%. The practical problem is that Canada's (and most developed countries) have implemented a web of tax treaties premised on a standard 15% withholding tax rate (or 5% where the shareholder holds more than 10% of the shares of the payer), and that rate has been integrated into our domestic tax systems (for example, in Canada, you only receive full recognition for foreign withholding tax up to 15%). I suspect most of our treaty partners would be unimpressed if we were to override our treaty obligations (which, in theory we could do) to reduce the benefits received by their citizens under the relevant treaty.

I'm also not sure I agree with your points on the retained earnings. After all, investors who want to switch their money from Old Tech to Nuvo Tech just need to sell their shares of Old Tech.

I do agree with your other comments however. If you want to tax expenditures (i.e. consumption) the only practical way to do it is to provide deduction for savings otherwise as you suggest, you'd not only have to monitor the value of assets, but also borrowing. Similarly, there are compelling practical reasons against taxing capital gains on housing, not least of which is that it'll discourage labour mobility (as people can't afford to leave their current house - a problem facing our friends in the South, albeit for different reasons). There's also the practical political reason against doing it, namely that the party that did it would be run out of Ottawa on a rail (and, needless to say, would never get elected in any of Vancouver, Calgary or Toronto).

[Frances Woolley]

Bob, I was hoping that you'd comment on this.

Now I know why I always do corporate income tax in the last week of class and hope that I don't get to it. Too much tax law type stuff to learn and remember.

Yes, the Part XIII tax on dividend payments to residents really does more or less what the Kent proposal suggests.

However if corporations retain earnings, and the profits are distributed to shareholders in the form of capital gains, the problem of taxing corporate income received by foreign shareholders isn't solved by either the Kent proposal or Part XIII taxes.

"I'm also not sure I agree with your points on the retained earnings. After all, investors who want to switch their money from Old Tech to Nuvo Tech just need to sell their shares of Old Tech."

By doing so would trigger a tax liability (capital gains tax) and again they would have less than $100 to invest, so again Old Tech looks better than paying capital gains tax and then investing in Novo Tech (the impacts would be slightly smaller, however, because of the preferential treatment of capital gains income.

Thanks again for your comments.

[Bob Smith]

Frances,

It's always a pleasure. Do you know how many blogs there are on the internet with intelilligent discussions (i.e., other than the "they're too high, they're too low" variety) of corporate tax reform?

Having finally had an opportunity to read Kent's entire proposal, I've got to say I'm disappointed. It's quite clear that his grasp of the Canadian tax system, as it currently stands, is quite limited, which certainly undermines the credibility of his proposals.

For example, he suggests that the RRSP regime "has morphed into a bonanza for people with incomes high enough for them to save up to $25,000 annually". I'm not sure where he comes up with that figure, but since the RRSP limit in 2010 is $22,000 (to be indexed to inflation in the future), I'd suggest that no one has an income high enough to save $25,000 a year in an RRSP (granted, you can carry-forward unused RRSP room, but I doubt that's what Kent had in mind).

As for his statement that RRSPs' have "become an easy, risk-free road for people with good earnings to retire as millionaires", I'd suggest that's indicative of his just being out to lunch. That might be an accurate description of some of the more generous defined benefit public sector pension (of which he is, in light of his long years of distinguished public service, no doubt a beneficiary and which are typically far more generous than any private pension plan or RRSP), but not of the RRSP system. Certainly, I doubt many of the boomers who watched ther "easy", "risk-free" RSSPs cut in half in the fall of 2008 would agree with that assertion and few people who are dependent on their RRSPs to retire have much hope of retiring as millionaires (at least in current dollar terms).

Moreover, a number of his purportedly radical proposals are, already, the law. For example, at one point (in the middle column of page 78) he suggests that Canada should impose a departure tax on the assets of ("wealthy people") who leave Canada. He is, perhaps, unware that Canada already does precisely that (under existing section 128.1(4) of the Income Tax Act which replaced a previous provision to that effect in 1993). Similarly, his suggestion that taxpayers can shift money to offshore tax havens (with the help of clever tax lawyers and accountants) is a remnant of an earlier era when such transactions were possible. That hasn't been the case for a long time (and indeed, the CRA recently won a long-awaited offshore trust case, which they're going to use to go after a whole laundry list of offshore trusts). These days the Income Tax Act is so filled with anti-avoidance rules that they invariably threaten to trap perfectly innocent offshore investments (for example, a number of proposed rules dealing with off-shore trusts threatened to trap non-taxable Canadian pension plans until they were amended).

Similarly, his suggestion that "benefits, from meals and club fees to travel by executive jet, should be fully costed and taxed to the corporation at rates comparable to those on similar spending by individuals", is somewhat odd. First, if its his position that these are benefits to the executives that would be taxable (under his regime) if they paid for them directly, why would you shift the cost to the poor shareholders? That seems to be dinging the shareholders twice, once when their executives use corporate resources to featherbed their lot(assuming these are not legitimate expenses to earn income, as many of them likely are) and once when they get taxed on it. Wouldn't it make more sense (at least in his framework) that those benefits should be taxed in the hands of the corporate executives who receive the benefit? Indeed, that is generally the EXISTING regime, where benefits such as club fees, meals and other perks are taxed in the hands of the recipient where they aren't incurred for the benefit of the employer (for that very reason, I turned down my employers' offer to pay for a fitness club membersship - it might be free, but i won't use it and I don't want to pay tax on it).

You are right that neither part XIII tax nor the Kent proposal deal with gains realized through capital gains (since capital gains on the sale of Canadian publicly listed companies, and since March, most private companies are generally exempt for non-residents). I suppose the flip side is that Canadians generally receive the same treatment in our tax treaty countries, so the net effect for the fisc is probably not huge (certainly not relative to the compliance cost on having to withhold on capital gains from publicly traded securities).

I'll concede your point on the retained earnings trap. I had thought that the higher after-tax return in the companies in the Kent world would offset the higher shareholder level tax, but I worked through the math, and I'm wrong.

Cheers

[rcp]

Just a word on inheritance tax. When a taxpayer dies there is a "deemed disposition on death", triggering capital gains tax on asset appreciation. This was introduced when capital gains tax was introduced in the early 70's, and is the reason that we don't have an inheritance tax - the deemed disposition on death replaced it.