« Abolish the Equity Premium Now! | Main | Who benefits from the oil sands? Who benefits from saying that only Alberta does? »


Feed You can follow this conversation by subscribing to the comment feed for this post.

AFAICS it can't be a Pareto improvement if you assume rational expectations. Quotas exist because either some group lobbied for them or legislators mistakenly believed it was in the public interest, without other inducement. If legislators can't, as a group, be mistaken about the economic consequences, then they must have thought the current system was optimal.

In a somewhat more reasonable vein, we know that there is, in fact, a lobby for quotas. You would also need to assume that all of those paying the lobbyists are mistaken about whether they benefit more by quotas than by their absence and are so certain in their misapprehension that they are willing to spend significant amounts of money to pay a lobbyist. This is highly unlikely.

The situation is like free trade. It is effectively impossible for there to be no losers. Even if you believe claptrap like comparative advantage (try making a complete list of prerequisites for proof. There may be on average net benefit in many cases, but that's not the same thing at all. Comparative advantage just demonstrates where a possible benefit would lie. It may be illusory, unrealizable or offset), that doesn't insure each and every individual in the entire trading process against loss.

Now you might be able to demonstrate Pareto optimality with groups and say no group will, on average of that group, be a net loser, but then you are hostage to your definition of a group, and, of course, optimality would now permit 49.99% of each and every group to be net losers.

Finally, if you assume that the entire economy is a representative agent, then Pareto optimality reduces to societal net benefit. Moreover you have to be able to demonstrate that you can aggregate to that level, meaningfully, which you can't.

Pareto optimality with rational agents requires that by coincidence at the very moment modeled, the change would hurt no one, while previously the benefit was so obscure that a functional majority knew of it, but chose not to make it.

OTOH, proving optimality with irrational agents would seem to require that the modeler have a better (in some sense) understanding of the agents' utility than the agents do. With choice of this better understanding, economics becomes politics, and you still have to demonstrate the validity of your aggregation or use unrealistic assumptions about utility curves

With enough arbitrary simplifying assumptions, you might be able to demonstrate Pareto optimality, but if my cat had wheels it would be a wagon.

Similarly it is more likely you can contrive a potential Pareto improvement, given the additional degrees of freedom potential must have over actual, but it's hard to see how this helps, since the degrees of freedom must have corresponding uncertainties.

I doubt this was exactly the answer you were looking for, but I'm not an economist, and I don't always know the correct set of unreasonable simplifying assumptions required. I just bought Blanchard to help with this. I don't much like it.

The unit price of milk is about 1.5 to 2.0 times higher in Ottawa than Chicago. I think this is mainly due to the import tariffs and supply management policies that are in place in Canada. But I'm not an expert on the details.

Removing the price-distorting policies cannot be a pure pareto improvement because farmers seem to reap economic rents from the status quo arrangement. They would lose those rents under a market based pricing system. The loss to farmers could -- I think -- be made larger or smaller depending on how the farmers acquire rights to sell milk in the first place. But this seems second order to me. Anyways, answers b) and c) seem to be incorrect: dairy farmers, at least, would be made worse off by the policy change.

A shift to a market based policy would probably move the equilibrium price of milk products closer to the the price that prevails in the world market. Consumers would benefit from this change to the extent that Canadian prices really are higher than world prices. The remaining issue is whether the gains to consumers would outweigh the losses to farmers. I think those gains would be pretty substantial in aggregate: there are many more milk drinkers than milk farmers, and this is even more true if we are allowed to count the stream of (discounted) benefits to future generations of consumers. (It's probably too much to ask that we count non-Canadian farmers as beneficiaries. But we probably can imagine that the market based system would improve allocative efficiency and so the reductions in deadweight loss might also be considered beneficial.)

In by book the most correct answer is a). But maybe I'm missing something? Is there some credible reason to doubt that the total gains to consumers is less than the losses to farmers? Or are there other groups that would lose because of the policies?

I'd argue that the best answer here is "a".

And this for the following reasons:
I. As you've defined a potential and an actual Pareto improvement, they're mutually exclusive (I think this definition is correct.). Something cannot be a potential Pareto improvement and an actual one if the former creates winners AND losers, whilst the latter does not. This throws out option "c"

II. As there is not enough information in the question to decide whether the quota has resulted in an increase in the producer surplus compared to the situation without the quota, it is impossible to call it an actual Pareto improvement. This throws out option "b"

III. The question does state that the quantity is restricted and this tells me that "d" can be thrown out. For that the current quantity is restricted means that the sum of consumer and producer surplus will increase without it as quantity increases. It is therefore at least a potential Pareto improvement.

Contra III one could argue, what if the quota is not binding? In that circumstance there would be no potential Pareto improvement. I don't think that's a correct counter-argument, because production is defined as contributing to the utility of individuals. Therefore if it is restricted, the utility of individuals is restricted and I think it is then reasonable to read restricted as to mean a binding restriction.

As to your last question:

"Or just throw the question out entirely as it's far too difficult for third year students?"

it depends ;). How many questions do you have in the exam and what is the subject material covered? It might have been unreasonable to have an exam consisting of only this question, but if precise knowledge of the definitions is part of the subject matter and there are plenty of other questions covering other areas of the subject matter I'd say that the question is a reasonable one to ask.

Best though is to check whether as the number of correct answers increases it is more likely that a student gave the correct answer to this question. That way you have some indication whether it was too difficult and whether it was well-understood. The slope and the intercept of a regression should tell you that I reckon?

Frances, as you've written about this multiple times, I'd be really interested in a post on the history of these quotas -- is this a holdover from Depression era policies?

It is 'a' although the following is an actual Pareto improvement:

1) Create a trading market for quotas, to price the quotas.
2) Using bond purchases, buy back all the quota on the market (slowly) and rent it out.
3) Once all quota is bought, eliminate quota, pay off the price of buying it by taxing milk by less than the amount of the price distortion.

I thiiink this is correct but am open to criticism! I think the strongest criticism is that some farmers may not want to give up their quota at any price.

Since I've never taken an econ course I think this question must not be very hard, but I could be wrong.

When I wrote the question I thought, like just about everyone else here, that the answer was obviously (a). Then I realized that I hadn't specified the form that the elimination could take. An elimination along the lines suggested by John would, I think, be an actual Pareto improvement. Probably. It also occurred to me that there may be long-run dynamic effects along the lines of: Canadian market is flooded with imported milk. Canadian dairy farms go out of business, all production capacity is lost. Something happens to make us wish we had that production capacity e.g. US imposes massive tax on all exports of milk. Can one be *sure* with no more information that is given in this question that it is a potential Pareto improvement?

Do the quotas compensate farmers for the costs of the disparity in animal welfare standards between here and the US. (Personally I place Canadian standards as higher, this does however raise the cost per litre.)

Frances: I'm not convinced that this story about "flooding" --> disappearance of Canadian dairy sector --> American milk export tax would be such a serious worry. The whole thing seems unlikely to me. But even if it happened, Canadian consumers would be paying the same inflated prices that they are right now. The current system, which keeps prices high, is basically equivalent to the worst case American export tax scenario except for the fact that under the American export tax scenario we could escape the problem in the long term because we could start some new Canadian dairy farms.

I lean to (a) as the fact that quotas exist as a rent means that farmers gain by them and there would likely be at least one loser unless the quota-elimination scheme was implemented very carefully. John has demonstrated such a careful improvement, but it is a low-probability result.

So (a) or (c) seem to be correct. Perhaps (c) could get a slightly higher mark.

Frances, as you've written about this multiple times, I'd be really interested in a post on the history of these quotas -- is this a holdover from Depression era policies?

See here: http://www.cbc.ca/news/politics/story/2012/01/04/pol-supply-management-trade.html

The milk quotas took shape in the 1970's but the first case of Supply Management was the Canadian Wheat Board, formed in 1935.

The Dairy Industry has always been very political. Until 1949 margarine was illegal in Canada under the Criminal Code. The Supreme Court struck down this law as an invalid exercise of the federal criminal law power and invoked the "no discrimination against manufactured products in interprovincial trade" clause. As a result it is impossible in Canada to talk about the division of powers and their use without talking about margarine.

The dairy industry in concentrated in Quebec and Ontario, Quebec has a very large dairy sector. As the House of Commons has a rural bias in seats, dairy protection is a political winner.

That article I cited does make an interesting point though that instead of paying for farm bailouts in "free markets" we just pay a steady price and don't give dairy subsidies out of taxes.

When I come across this problem in marking, my approach is to
1) accept the answer that best suits the sophistication level expected of the students. (In this case, that means that I'd accept "a" with no further ado.)
2) also accept alternative answers provided that the student is able to make a reasonably convincing case to me on why it should be accepted. Therefore, if a student wants to argue that some scenario might arise to make us regret our choice, I'd accept that answer....but the student has to make the case for it.

Determinant - I would say that (c) is unlikely to be the correct answer for the reasons people have given already - if it really was possible to find an actual Pareto improvement that resulted in eliminated milk quotas, we wouldn't still have quotas. Sure, one could make a case for (c) along the lines suggested by John, but it seems unlikely.

The article you mentioned is interesting, and it is a good example of the reasons people might choose (d) rather than (a). Someone comments at the end of that article:

"Doing away with the boards will lower prices - perhaps drastically. It will also drive up the cost of healthcare. Probably by a LOT more than any savings realised,"

The idea being that milk imported from the US or elsewhere may not be as healthy as good, ol' Canadian milk, due to differential production standards - regulations around use of growth hormones, frequency of health and safety inspections, etc. Consequently, elimination of milk quotas would be neither an actual nor a potential Pareto improvement.

Simon, it's multiple choice. I've looked at a few papers, and about 90% answered A, so I've decided to include the question and accept A as the answer. WIth a question that allowed writing I would take the approach you suggest, absolutely.

Determinant - I would say that (c) is unlikely to be the correct answer for the reasons people have given already - if it really was possible to find an actual Pareto improvement that resulted in eliminated milk quotas, we wouldn't still have quotas. Sure, one could make a case for (c) along the lines suggested by John, but it seems unlikely.

Huh? I thought you said in a previous marking thread you didn't like appealing to history.

So (c) has to get some mention.

Second, that statement is a classic fallacy of denying the antecedent. It was so strong the fallacy smacked me in the face. Besides, if you want to look at history, the Federal Government was granted exclusive jurisdiction over Divorce in 1867, but didn't enact anything under that power until the Divorce Act, 1968. The only divorce legislation before then was individual private Acts to grant a specific couple a divorce. Politics can be deadlocked for a long, long time.

I would answer a. But then I would be assuming that the milk quotas would be abolished without compensation. Trouble is, some students may approach this question with different implicit assumptions about what the policy would be -- whether or not quota holders would be compensated.

Safest policy: any student who objects to your saying a is the correct answer is given the opportunity to submit a one paragraph explanation why some other answer is better.

I reckon this question would have been better as a True, false, uncertain, Explain, just to winkle out any such implicit assumptions.

Nick, I like your suggestion, thanks.

Consider it a Zen koan.

The answer is Moo.


Canada has a comparative advantage in milk. Cows are basically a northern species that produces its own heat ( baby in amanger anyone). U.S. dairy production is concentrated in nnorther states or the higher altitudes. Poultry is a tropical thing and is adapted to Alabama, ot here.
So, Québec has been a dairy prvince since the 1880's when rail transport and refrigeration made it possible to ship large quantities. So much so that St-Hyacinthe was the third largest banking center in the late 19th century, after Montréal and Québec City. see Banking en français by Ronald Rudin

In the 60's, productivity in agriculture exploded all over the developped world, while food remained a low price-elasticity product. Given the low tranferability of capital in face of declining revenue. Canada, like Europe choose the quota route. Of course, in Canada, it turned out as "QC favored again." The parochialism of the chattering class, combined with its usual bias.
The Wheat Board came from crashing prices in the Depression and the depredations of wheat traders. Now that the Board is gone, I hope the Western farmers survive the tender altruistic mercies of ADM,Cargill and Bunge.
Grain didn't go the quota route because of the potential for export ( wheat,unlike milk, is both cheap to ship and a low income staple.) So we could sell it to the succesful central planners in Moscow.The kind-hearted Tory farmers never hesitated to feed those poor starving Commies...

Frances: would your anxieties in pedagogical docimology be assuaged if you came to the annual Convention of The Cegep Professors Pedagogical Association at Lake Leamy this week? ;-)
Hope the weather in the National Capital will better than it was today.

Hmm. This is a very tough one. Besides John's scenario there is still a chance that even with higher production the actual producers of diary products could end up being better off. There are real world stories about things like this that I had not studied in much detail, but supposedly deregulation did miracles for agriculture (and diary production) in New Zaeland. For instance being able to produce large quantities of cheaper milk (compared to quota regulated markets) may attract manufacturers of processed food that use dried milk in their products (like sweets and such) actually increasing profits of farmers. I think what you had in mind was not only abolishing production quotas but abolishing import tariffs. Now this is something I would be more confident to select as potential Pareto improvement.

Nick: For a policy is Potential Pareto improvement the compensation do not have to be actually paid out (see the original post). You can just assume that people who are harmed by the policy could be theoretically compensated.

J.V. - the New Zealand and Australian experiences of deregulation seem to have been quite different, I don't know why.

Jacques - The weather forecast looks better for Tuesday onwards, are you in town all week?

A successful separatist campaign riding on anger resulting from elimination of milk quotas is another 'eliminating quotas in not a Pareto improvement' scenario.


Cows live in any climate that is not too hot or too cold. But they were first domesticated in Turkey.

The ideal place to raise them is where land is cheap but close enough to populated areas to ship the milk there. Obviously transportation is important.

Possibly there is an advantage to colder places in that they need to spend less on refrigeration? Those places would need to pretty cold. Rather, the cold is something to overcome by feeding the cattle a bit more rather than an advantage.

In the U.S. the largest milk producing state is California. After that you have rural Pennsylvania, for historical reasons (it was close to the populated eastern seaboard) and then some colder northern states -- the empty ones where land is cheap, mostly, but don't forget Florida is a big milk player. The world's largest producer is India.

I can't possibly imagine any natural advantage that Quebec may have for milk production -- perhaps it is close to waterways or other means of cheap transport to nearby cities? Or maybe it is one of those accidents of geography where one industry happens to cluster together.

This is not to say that you Canadians need to get rid of milk quotas, but if Quebec really has an advantage in milk production, you should be benefiting from it by drinking lots of milk, and it should be cheap. Switzerland has the highest rate of per capita chocolate consumption, but I'm not sure what swiss chocolate prices are vis-a-vis the rest of the EU. They claim to have a competitive advantage in chocolate because the grass is sweeter; like wine is supposed to be better in France because the soil is more poetic.

JV: Yep. I would say that this policy would be a potential but not an actual Pareto Improvement.

The answer to every question in economics is "it depends". So I'd pick "e".

In fairness, you've left a loophole, asking people for the answer that "best" describes that policy. It may well be that the might be circumstances in which another answer is correct, but if I were looking for the "best" (or perhaps, most likely) description of that policy, (a)'s the "right" answer.

"A successful separatist campaign riding on anger resulting from elimination of milk quotas is another 'eliminating quotas in not a Pareto improvement' scenario."

Says who? Quebecers might be happy to be freed from the imperial oppression of "les Anglais". And "les Anglais" would get cheap milk and be spared the cost of "oppressing" Quebec (by subsidizing their "distinct" society - a curiously incidious form of oppression). A win/win scenario, non?

Bob: "A win/win scenario, non?"

I'll let Jacques Rene speak to that one!

Frances: yep. Here till Saturday.
rsj: Turkey, apart from the Mediterranean coast, is a rather high altitude cold country. Same as people thinking that Korea is a tropical country...

Bob Smith:Essentially,each province produces its own milk. And there is no payment from the feds to farmers. There are transfers from consumers to producers within each province yes.
If you think that milk quotas of which they are provincial ones) have any bearing on
sovereignty, you'reading Maclean's too much...

Without knowing anything about Canadian milk prices and consumption, the answer that jumps out is "a." Hearken back to examples of butter mountains in the EU and so on: how could the answer be otherwise? I'd allow for multiple answers if the logic shows the student can walk through Pareto optimality and understands the concept, which is probably what you want pedagogically. Throwing the question away seems like a cop out.

The comments to this entry are closed.

Search this site

  • Google

Blog powered by Typepad