Negative value added: education edition

By Eric Crampton 09/09/2014


Remember the story of the West German auto exec who, after touring the Trabant plant, wept because the value of the steel and other inputs going in exceeded the value of the car coming out? The Trabant plant was destroying value.

That’s how I felt on reading the New Zealand Qualifications Authority standard on economics. Or at least this bit. They are making the students dumber by teaching them things that are not so.

Here’s the particularly offensive bit:

3.  Market failure refers to situations when a market fails to deliver an efficient or equitable outcome. Efficiency occurs when Social Marginal Cost equals Social Marginal Benefit. Equity occurs if a situation or outcome is considered to be fair. The different market failures relate to: consumption externalities, production externalities, public goods, imperfect information, inequitable income distribution. 

4.  Government interventions refer to interventions in a market by central or local government. For example, these may include, for each market failure, a selection from:

  • subsidies, taxes, regulations, property rights and government provision (consumption externalities) 
  • subsidies, taxes, regulations, property rights and government provision (production externalities) 
  • government provision (public goods) 
  • regulation (imperfect information) 
  • progressive taxes, welfare benefits, collective provision and minimum wage (inequitable income distribution). 

This just isn’t so!

Pick up any reasonable intermediate microeconomics text. Market failure occurs when the conditions underlying the first welfare theorem fail to hold. That doesn’t necessarily mean that there’s anything the government can or should do to fix things, but it opens up the possibility that there could be interventions that improve outcomes. The kinds of things in the first four bullet points are pretty standard textbook analysis, though we’d need to caution that the failures are necessary but not sufficient basis for intervention.

The last bullet point is right out. Draw the standard Edgeworth Box and derive the contract curve. Any of the infinite number of points on that curve are consistent with market efficiency (the First Welfare Theorem), but only some of the points will be consistent with particular views of equity.

If the government decides that one point on the curve is preferred, for equity reasons, to the one that would be achieved by markets, it can push things in that direction through appropriate redistributive measures. That’s the Second Welfare Theorem. But we would never describe equity-based interventions as correcting a market failure. If we were on the contract curve, where marginal rates of substitution and marginal rates of transformation all lined up nicely, but we just didn’t like that point for some reason, that just isn’t a market failure. And I would have given a failing grade to a student who would have described it as such. Heck, I ran multichoice questions specifically checking against this kind of fallacy. NZQA is teaching students how to get a failing mark on intermediate micro exam questions.

Even worse, were we to take the Second Welfare Theorem as justification for equity-related interventions, the Theorem does require that least-inefficient means be used to get to the preferred point on the contract curve: typically, lump-sum redistribution of initial endowments. In the real world, you aim for the least distortionary taxes available: maybe land value taxation, maybe progressive consumption taxes. But you sure as heck don’t go for minimum wages. That’s a good way of ensuring that you never get back to the contract curve, or that you stay farther away from it than you could have using alternative measures like wage subsidies (or, to be consistent with theory, lump-sum transfers).

Why does this matter? There are decent theoretical reasons for interventions addressing First Welfare Theorem considerations. They’re sciency. And, if designed properly with appropriate side-payments, they can in theory make at least one person better off while making nobody worse off. But there just aren’t such justifications for equity-based policies – they’re based on aesthetic considerations. It would be like having an engineering class move on from efficiency considerations in engine design to say that it’s also an efficiency failure if the engine is painted blue instead of red.

Somebody please fix NZQA so that the economics standard isn’t teaching kids how to fail their intermediate micro exams?

Update: in comments, Granite26 very correctly notes:

In your red vs blue engine example, it might be more correct to say ‘uses too much fuel regardless of the efficiency with which it is converted to horsepower’. We should all be riding Mopeds, you know

Update 2: Paul Walker points me to the source on the Trabant story.