« Gender-Based Analysis: A Guide for Economists | Main | Health Spending Numbers: An Update on the Long-Term »


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

I find this post rather odd. The purpose of regulation is to deal with market failure. Regulation is suppose to address some kind of externality, an information asymmetry, an incomplete market, prohibitive transaction costs, free-ridership, something that the market cannot internalize on its own. The notion that one would even ask about racial or cultural specific regulation to achieve a normative reallocation suggests a fundamental disconnect from economic theory. The purpose of regulation is not to find a new equilibrium different from what the Welfare Theorems would imply, but to create an economic environment in which the Welfare Theorems work! At least that's how it's suppose to go...

If there are people in society who are hard done by, whether they belong to an identifiable group or not, we can help them through direct subsidies. If we feel that some group should get more, fine - let's advocate for it, let's debate the levels, and then let's provide an on target direct subsidy separate from the tax code and separate from regulation.

The problem with special group regulation is the incredible capture pressure. Regulatory capture will be off the scales with such policy. Public choice theory, eat your heart out on that one!

Is the regulation explicitly targeted at certain ethnic groups with certain skin pigmentations or is it targeted towards all Canadians? It was already mandated in 1965 which suggests it's targeted at all Canadians.

There is medical evidence that links vit D deficiency to certain serious diseases.

Perhaps the regulation is designed to make sure that all Canadians are protected (Canada being comparatively sun radiation deficient?), the aim being to reduce the future public health bill.

Henry: "Is the regulation explicitly targeted at certain ethnic groups with certain skin pigmentations or is it targeted towards all Canadians?"

The regulation applies to all milk sold in Canada, but it was introduced in 1965, which is probably the point in time where the percentage of people in the population who were lactose intolerant was lower than it has been at any time before or since. So it's not in any sense explicitly targeted at certain ethnic groups. However the optimal level of vitamin D fortification, and the best way to deliver that fortification, is likely to depend upon the ethnic and cultural composition of the society (and, in a country like Canada, also on the geographical distribution of the population north/south).

Avon, vitamin D fortification was introduced to protect children. Vitamin D is required for healthy bone and other types of growth. Past experience has shown that, absent fortification, not all parents will ensure that their children receive adequate levels of vitamin D, either through exposure to sunlight, eating cod liver oil, or in other ways.

Markets require fully informed economic agents to achieve efficient outcomes. Children cannot be assumed to be fully informed and rational economic agents.

There's your market failure.

OK, I am not General Jack Ripper from Dr. Strangelove: “Do you realize that fluoridation is the most monstrously conceived and dangerous communist plot we have ever had to face?...” ;)

The vitamin D thing is a bit of a stretch. People are informed. Your argument is that some large fraction of the population will choose to ignore it, do the wrong thing and harm their kids. Maybe. But if we are going to go down that route – that some parents aren't very good – I don't think we would start with insufficient vitamin D consumption. But fine, vitamin fortification of food (it's more than just milk) is not a big deal.

Let's be sensible with regulation. Let's try to insulate it from capture as best as possible by making it specific, limited, and transparent. Use it for bona fide market failures. The idea that we should create some weird hybrid of regulation based on ethnic, racial or other special interests just screams capture. The next thing we'll see at city council is a new taxi regulation banning Uber based on the culture and race of most taxi drivers.

I don't think the suggestion is that regulation should be based on "ethnic, racial or other special interests". The issue is that the intended effect will not be uniform across those at the intended margin (children), because some cultural norms will weaken its effect; in this case, that's the volume of fresh milk consumed.

Frances, I think the problem may be narrower though: if you take SE Asia as the example, children in those groups may consume less milk if their families are recent immigrants, AND the children weren't born in Canada, AND the parents are lower on the scale of education and income. OTOH, I know plenty of Thai and Cambodian immigrant families whose kids are pretty typical in their fresh milk consumption. My point is that assumptions about the relevance of culture are usually too broad, and other factors may matter more. In this regard I agree with Avon.

As an alternate example, policies that support or encourage going up against authority (like reporting child abuse, whistleblower legislation, reporting domestic violence) will also be taken up by people in ways that may vary significantly with culture or ethnicity. Admittedly those wouldn't be the only variables.


While reading this post, I couldn't help thinking about how it provides an compelling counter-example to an earlier post of yours "why culture is a lousy explanation" (http://worthwhile.typepad.com/worthwhile_canadian_initi/2013/03/why-culture-is-a-lousy-explanation.html).

If we were looking at the cause of vitamin D deficiency in Canada, culture would provide a pretty good explanation for differences between groups. Nor is it fair to say that saying culture is the explanation has only only "trivial predictive value". Granted, in this case, we know about the link between milk consumption and vitamin D deficiency, so culture doesn't add much. But if we were coming into the situation blind and were trying to figure out why vitamin D deficiency was higher among, say, aboriginals or East Asians, then among Europeans or Northern Indians, coding for culture might be a useful tool.

Shangwen: " My point is that assumptions about the relevance of culture are usually too broad, and other factors may matter more. In this regard I agree with Avon."

Sharp and to the point as always, Shangwen! Status of Women Canada has an on-line gender-based analysis course. One of the examples in this course is thinking about ways of increasing young women's activity levels. The great insights offered by gender-based analysis: try activities with music! And less ones that involve individual performance in front of peers! It may well be true (I've just bought myself some fancy noise cancelling headphones so I can go to the gym and listen to my music, not the electro-pop and rap favoured by the Carleton gym staff). But is this what feminist boils down to: "let's all do aerobics!"? Really???

So thinking about race and ethnicity without promoting stereotypes and being overly essentialist is really really hard.

Can you be more precise about a regulation you're thinking about with regards to something like e.g. whistleblower legislation? Develop that a little more fully? There's a sincere interest on the part of some folks in government about this, and a desire to truly understand the issues - so any insights you have would be most welcome.

Bob - hey, the advantage of totally contradicting myself is that there's a chance you'll will agree with me at least some of the time.

Frances, whistleblower legislation came to mind because it makes enormous assumptions about what will motivate employees (notionally, low- to mid-level employees of a public organization) to rat out major misconduct higher up. These assumptions include believing that public organizations are service-oriented and transparent, unlikely to succumb to the pressures that public choice theorists (and Avon) talk about, and most importantly, in a classical liberal sense, that that's the right way for them to operate.

If you operate from the perspective of, say, a very tribal culture, or one where the state is adored, or where it is acceptable for government bodies to be overtly self-serving (e.g., China), then the kinds of things considered whistle-blowable don't look that bad. They may even look like great opportunities, perhaps for your children or your poker-addicted brother-in-law, or even ways out of poverty and insecurity. Why destroy that? Too many policy writers think that skepticism about the public service is due to exhaustion or cynicism ("Disengagement! The Harper legacy!!"). There are other ways of looking things, and plenty of Canadian examples to support those alternate views.

For example--and without going too far for fear of public lynching--what do you think the outcome would be of a series of whistleblower disclosures against Indian reservations with chronic inability to deliver on the use of funds for safe housing and clean water? How likely are such disclosures? Maybe it would be good, but maybe not.

Frances, you know what they say about consistency being the hobgoblin of small minds.

"The purpose of regulation is to deal with market failure."

Says who? There are all sorts of laws justified with all kinds of reasoning.

"Regulation is suppose to address some kind of externality, an information asymmetry, an incomplete market, prohibitive transaction costs, free-ridership, something that the market cannot internalize on its own."

Pick a random law that some people who are not economists support and ask them why they support it. The chances they'll give one of these as a reason are slim.

"Regulation is suppose to address some kind of externality…"

I wonder if we have any regulation that addresses property rights or insults…

Well Australia is considering putting a special case for Aboriginals into the Constitution, thus ensuring perpetual divisiveness. That would be a case where it matters to regulatory policy for no other reason than because the policy is to ensure that it matters.

Am I cheating on that one? Not entirely what you were looking for?

Tel: yes, that is cheating - the idea is to find something non-obvious. The context here is that I was being asked by someone who was responsible for ensuring that, e.g., regulations didn't have unintended impacts on Aboriginals, but wasn't sure what kind of ways in which those unintended impacts could arise.


Economics is not obvious and it cannot be understood by introspection alone. Economics is as technical as anything you will find in physics or mathematics. I find it funny: if you ask the layperson what her opinion is about quantum gravity, she will answer that she knows nothing about it. But if you ask her about financial regulation, she will give you a laundry list of things that the government should do. It's silly – both problems are equally hard.

In a pristine economy with no exernalities, no information asymmetry, and everyone earns their marginal product, there is little for the government to do in the economy. Of course there are laws, and the rule of law (the state will provide security, etc.), by there would be little need for regulation – the Welfare Theorems would hold. If the resulting distribution is normatively distasteful, we can redistribute endowments and income (that is, just give money to the people who you think need help) and rely on the Welfare Theorems to do the rest. The result would be exactly the same as what a benevolent social planner would choose.

The real world is not a pristine economy and it is the departure from it that dictates the amount of and need for regulation. Regulation must not only address the market failure in question (the failure of the Welfare Theorems) it must also not open the floodgates to capture. In many cases, the attempt to correct a small amount of market failure leads to regulation that makes the situation far worse. With capture, the regulation stops being a helpful mechanism, it starts to be used by the regulated to entrench special interests, limit competition, and create monopoly powers.

Financial regulation is a great example. The idea is that there is an information asymmetry when it comes to investing, so we need regulation. People hand over part of their income to an investment firm to save for retirement. Since few people know or understand how investing works and since few people can actually monitor what their investment advisor is doing, the government creates regulation in the investing industry. The government demands that industry participants hold special licenses and satisfy reporting requirements. It sounds reasonable, but the industry participants quickly capture the regulation for their own benefit – usually to restrict competition. In fact, the industry itself becomes the champion of advancing more licensing requirements because it restricts the number of potential competitors and creates more monopoly power. This is how capture looks – the special interest groups present new regulation as protecting the public, the layperson has little interest in understanding the details (but likes the tone of the sales pitch that someone is “thinking of the children”, etc.), and the special interests laugh all the way to the bank. In Canada, that capture process results in management expense ratios for equity mutual funds north of 2%. I'm not arguing for no financial regulation, I am arguing for transparent, limited, and hyper-targeted regulation to prevent capture as much as possible. Regulation will always involve tradeoffs, let's ensure that the nasty stuff like capture is limited.


Here's one – CMHC is a giant debt subsidy machine. We regulate our banks to ensure that people's real credit rating doesn't get reflected in their mortgage interest rates. There are cultures in Canada who do not borrow. Since the debt subsidy drives up house prices, it puts the non-borrowing cultures at a disadvantage.

You miss my point, Avon. With quantum physics, there Is no normativity involved. Same with economics. But that is entirely false when it comes to policy. Someone does not need to understand economics to make policy recommedatiosns, depending on the policy in question.

You also didn't get my last point. Why is an economy with externalities not "pristine"? Presumably your "pristine" economy would have private property, but that itself is an externality.


Private property is not an example of an externality. An economy with externalities will not achieve Pareto efficiency. That is why we call it not “pristine”. Recall the Welfare Theorems: 1) All competitive markets lead to Pareto efficient outcomes; 2) All Pareto efficient equilibria can be reached by a competitive market. There are some ifs that make the theorems work (perfect information, no externalities, etc, look them up if you're interested). Since a benevolent social planner will seek a Pareto efficient outcome, if the Welfare Theorems hold, the social planner only needs to rearrange endowments and let the market do its thing. In reality, perfectly competitive markets do not exist. The only questions are: “How much do they depart from perfect competition?” And “Is there a government policy that can address the market failure without creating large second order distortions that become first order in their own right?”

Policy must be informed by economics – you are using government force to arrange an outcome. You are forcing people to do things they might not otherwise choose. You better understand what you're doing before you start pointing guns at people.

Upcoming regulation: “Thou shalt use the word “ethnicity” instead of “race” because PC rule No.17 states that thou shalt not use one syllable where four will do.” Ha ha.

The context here is that I was being asked by someone who was responsible for ensuring that, e.g., regulations didn't have unintended impacts on Aboriginals, but wasn't sure what kind of ways in which those unintended impacts could arise.
You want me to make a presumption of good faith? Smokes! That is a tough one.

Hopefully you don't mind if I just fake it for the purpose of intellectual exercise? Feel free to suspend disbelief if that helps any.

OK, let's suppose all employers (for no particular reason) are racist A-holes.

Thou shalt use the word “ethnicity” instead of “race” ...

Ooopse, sorry, I meant to say ethnicisit A-holes. And let's suppose there are two different, errr, ethnicities which we will call "green" and "puce". So in a free market those bigoted employers (the bastards) are willing to pay puce people 25c per hour more than green people (shocked I tell you). Admittedly, 25c isn't a whole lot (over sufficient time it might even diminish), but it's enough to be annoying (enough for grievance mongering, and that wins votes, and votes are gooood).

Oh wait! This isn't a free market, we have a minimum wage here... so Mr Government says all unskilled workers (regardless of, errr, ethnicity) must earn exactly $10 per hour. Employers then hire a whole lot of puce people, and largely skip past the green people, because 25c is still 25c.

To review: getting paid 25c less than the guy next to you sure is annoying; but getting paid nothing and sitting out of work, while other guys do get hired, that really stinks. Well golly now I do have a grievance.

How to take a small prejudice and turn it into a "Really Big Thing (TM)" ... awww shucks, do I have to keep maintaining this obviously fake pretence of good faith? You all know where this ends.

"You want me to make a presumption of good faith? Smokes! That is a tough one."

Yes, I do. This whole blog exists because Nick and Stephen and the rest of the gang are people of good faith, and 95% of our commentators are also people of good faith.

I'm going to close the comments on this post.

The comments to this entry are closed.

Search this site

  • Google

Blog powered by Typepad