I want to set aside the first question and focus on the second question.

2. Was the CMHC right to impose a monthly quota of $350 million on individual banks as a way of implementing the total cap of $85 billion?

I learn from the CBC that the Canadian Mortgage and Housing Corporation has imposed a quota of $350 million per month per individual bank (or other lender) on the amount of mortgage-backed securities it will guarantee. Presumably the CMHC did this because the Federal government wanted to put a cap of $85 billion on the total amount of new mortgage-backed securities the CMHC (and thus, indirectly, the government and the taxpayer) would guarantee in 2013.

If it looked like the $85 billion annual cap would be exceeded (as seems likely given sales of $66 billion in the first seven months) clearly CMHC needed to do something. But why impose a quota? Why not raise the insurance premium instead?

Any economist will recognise this as the classic question of quotas vs tariffs. Or quotas vs taxes. If the government thinks the total amount of some activity is too high, it can impose a quota on that activity or else impose a tax on that activity. Imports are one example: the government can reduce imports either by a quota on imports or by a tax (tariff) on imports. Pollution is a second example: the government can reduce pollution either by a quota on the amount of pollution or by a tax (Pigou tax) on each unit of pollution. Tradable quotas (tradable emissions permits) are a third intermediate policy that are partly like a quota and partly like a tax.

Here's the simple diagram we use to show the similarities and differences between quotas and tariffs:







With no quota or tariff, the free market equilibrium is where the supply and demand curves intersect, at a price Pfm and quantity Qfm.

With a quota of Qq, the supply curve now becomes the vertical red line (strictly, it's a kinked supply curve that follows first the blue line, and then the red line), and the equilibrium price is now Pb.

But the same effect could be attained by a tariff (tax) per unit equal to the height of the green line. Buyers pay Pb including the tax; sellers get Ps net of tax; and the quantity is Qq, just like with the quota.

If you choose the right size of tax, you get exactly the same quantity Qq with a tax as with a quota.

Buyers pay exactly the same price Pb with a tax as with a quota.

So what's the difference? The difference is who gets the brown shaded area in the diagram.

With a tax, the government gets that shaded area as tax revenue, because that area equals the height of the green line tax per unit, times the length which is quantity sold.

With a quota, that shaded area goes to whoever is given the quotas, because each unit quota is worth the difference between Pb (the price you can sell the good at) minus Ps (what sellers would be just willing to accept to produce one more unit).

That's the main message of this post. And it's a simple message. By imposing quotas on banks, as opposed to raising insurance premiums to hit the same target quantity of CMHC guaranteed mortgage-backed securities, the banks were better off, and the government/taxpayer was worse off.

Or, we could re-frame the same message, and say that if CMHC had raised insurance premiums, as opposed to imposing quotas, the government/taxpayer would be better off, and the banks would be worse off.

Relative to the status quo, with neither quota nor increased premiums, we cannot say whether the banks are better off or worse off with a quota. It depends on the size of the quota, and on the elasticities of demand and supply. With a large quota, that only slightly reduces the quantity of mortgage-backed securities guaranteed by CMHC, banks will be better off relative to the status quo. With a small quota, that reduces the quantity a lot, banks will be worse off relative to the status quo. (I would need to draw a lot more diagrams to explain why this paragraph is true.)

My guess is that this is a large enough quota that banks will be better off than without the quota. So I don't think banks will complain. But that's just a guess. But banks would definitely be better off with a quota than if CMHC had raised insurance premiums enough to get the same total quantity of guaranteed mortgage-backed securities. And that's not a guess. So banks definitely won't complain about the quota if they think the alternative is an increase in insurance premiums.

This is all just intro textbook stuff. There may well be more to it than this, of course. But this is a good place to start.

A few additional observations:

1. If CMHC must announce insurance premiums in advance, and cannot change them quickly without notice, and doesn't know where the demand and supply curves will be, it might be hard for CMHC to hit an exact quantity target by raising insurance premiums. It won't know how much to raise premiums. Maybe that's why CMHC chose quotas over raising premiums.

2. On the other hand, one peculiar feature of the CMHC policy is that each individual bank (or lender) gets the same $350 million per month quota. That looks like a big incentive to set up a new bank, maybe just as a dummy split-off from an existing bank? Or is it restricted to existing banks? It's also very strange that small banks and big banks (based on past lending) get exactly the same quota. Apart from fairness this doesn't seem to be efficient, because the bigger banks would value a marginal quota more than the small banks, which is why quotas should be tradable. My guess is that, if the quotas are not in fact directly tradable, a lot of clever finance guys are currently figuring out how to trade them indirectly, and I have a very strong hunch those clever finance guys will figure out a way to do it indirectly, but I hate to think what that might do to systemic problems of financial stability.

3. Going back to my first question, which I said I would avoid: if the Federal government thinks that the risks to the taxpayer are too costly given the insurance premiums, why not just raise the damned insurance premiums?

4. Maybe it's because the average voter simply doesn't understand the quota vs tariff question, and thinks that taxes raise prices to the consumer (correct) but that quotas don't (incorrect). Just like the carbon tax.

5. Or maybe there's some sort of adverse selection/moral hazard problem that would get worse with raising insurance premiums but wouldn't get worse with quotas? Is this an application of the theory of credit-rationing? Dunno.

This is all just off the top of my head, plus intro micro. Over to you guys.

[On a personal note, I keep wondering if I left my brain on a farm in England or on the shores of Lake Huron. It's been hard getting it to work on economics again after nearly a month away. It feels weird to be blogging again.]