Great post from Don Boudreaux on Ken Arrow and the Virginia election

Here.

I kind of take this point like I take Sraffa's critique of Hayek or certain criticisms of marginalism: it's a good reminder to avoid sloppy talk.

I think we can make statements about the public good, we just have to recognize two things:

1. There is no actual actor that is the collective, and
2. What we identify as the public good is of necessity normative even if it is informed by everyone else's utility function.

So I would feel perfectly comfortable talking about the need for an egalitarian society and potentially legislation to achieve that as being in the public good or even an American ideal. I might use language Don would consider collectivist.

I can make the claim quite legitimately that I come to this conclusion based on some sort of aggregation of what other Americans think. It's a rough, back of the envelope aggregation on my part, but it's legitimately an aggregation of individual preferences.

But it's not some sort of objective, uncontestable aggregation of those preferences. Instead it's an aggregation based on my some of my own values as well (which, in fair divinations of the public spirit, is itself derived from public values and virtues). This isn't a useless way of talking about the public good, but it isn't objective either - not in the same way that my preferences are (they are subjective of course, but I can objectively say "Daniel prefers X").

Similarly when we get an election result it's tough to say that it is the public's will. What we have to recognize is that it is one aggregation of individual preferences, and the aggregation is done according to the dictates of a certain set of democratic institutions.

THAT is a fair, objective claim about what this "public will" is.

As Ken Arrow and Don Boudreaux discuss, it's not legitimate to talk about this as an actual sensible preference set on its own. It isn't.

As individuals and as citizens, what we have to decide is whether those mediating democratic institutions aggregate preferences in a way that we like or not.

I am confident al the people that drone on about the alleged significance of the racist roots of progressivism will repost this IMMEDIATELY

Here.

Field Turing Tests

So you've all heard of Turing tests, but what's probably more important is Field Turing Tests - similar to the distinction between experimental and field tests elsewhere.

A Turing test is useful because it gives you a sense of whether the other person really comprehends what they're arguing with. That's helpful in assessing the quality of the process by which they form arguments, etc.

But it doesn't necessarily mean that they're good interacting with people in the marketplace of ideas. For that, you need a field Turing test. Out in the wild, do people present the ideas they disagree with in a way that would convince someone from the  other side.

This just comes to mind as a result of (what I hope is the end of) a long set of diffuse interactions with Don Boudreaux over monopsony. My concluding thought has been "my God I hope nobody just reads Café Hayek and doesn't read this blog to understand what I think about this issue". There's been some really awful misrepresentation from Don.

I've had similar fears in the last few months about my immigration work - I hope people don't just read stuff at Brookings, ITIF, or from Vivek Wadhwa on our immigration work. If they do they're going to completely miss the point.

The way around that is to be well engaged and a self-promoter of course. I've been in dialogue with people at Brookings and ITIF (Vivek infuriates me too much to talk with, although ITIF has been almost as bad - perhaps that's wrong of me). I've published at EPI, EconLib, and Cato on the issue which I think is very helpful.

But then there are good actors too - people you never have to worry about. David Henderson and Bryan Caplan always represent the other side pretty accurately. Any quibbles with their accounts would be idiosyncratic - not systematic. Bob Murphy is usually great, although I find some of his Krugman posts not just cases I disagree with, but very odd in their reading of Krugman (but we can bracket those off for the sake of this argument). Ryan Murphy and Jonathan Catalan are always great at this among us junior bloggers.

Those are the obvious ones on the libertarian side of things. Cochrane and Williamson are usually quite good on this count too. Horwitz can be OK on more analytical pieces but pretty bad on polemical pieces. This is a huge weak point for Boettke who I think is best read for his brilliant expositions and deep knowledge of his own side. Russ is in the same boat as Don.

Who do you think is especially good at this or especially bad at this on either side?

I think I'm pretty good. I provide accounts of the other side that I know they don't like (for example, the whole point about libertarianism vs. propertarianism and the idea that "liberty movement" is pure euphemism), but I hope I'm clear that this is my take on their position and not the position they are under the impression that they hold. Perhaps making that distinction clear is all that a successful performance in a field Turing test requires.

Instrumental variables get a pass on rigor because economists essentially use them for personal entertainment...

That last line isn't entirely fair. There are some good IV studies out there, but IMO they are the ones that use a bazillion different robustness checks to assure you they're not full of shit.

So take that as poetic license and shift attention to the first part of the title.

I'm doing some work right now for a class paper (and hopefully a publication some time in the future) on within-study tests of propensity score matching. Similar papers are out there for other methods but I'm working with PSM. As anyone that's worked with the method before knows, it's decent enough but very dicey. You have to really know what kinds of selection mechanisms are in play to make it useful because it's simply not a solution to the problem we worry about most: selection on unobservables. It ONLY works for selection on observables that we're not good at modeling. In a panel framework you can make the case that selection on unobservables is accounted for by using past data (i.e. - I don't see my unobservables but I match on wages at t-1 and since I'm interested in wages at t+1, all the unobservables I think affect wages at t+1 are also affecting wages at t-1 in the same way, so by matching on t-1 I'm OK).

We've learned much of this about PSM from random assignment studies that use PSM to test a result with real world data (lots of this can be shown with simulation, but simulation isn't necessarily relevant to outcomes in the real world).

This is true of most quasi-experimental methods. There's usually a literature on within-study tests of the method comparing to random assignment. This is a great way to learn about the limits of our quasi-experimental options, and that's always good to know especially when we only have quasi-experimental options).

As far as I know, though, there's no such within-study test of IV estimators. There's lots of within-study comparisons to OLS but that's a different exercise entirely. If they're both biased then who cares if it's close to OLS or not?

I'm not entirely sure why this is, except that IVs aren't generally used for evaluations where you might have random assignment. Instead they're used for more fundamental scientific questions. But at least in the returns to education literature you'd think this could be possible.

One more reason to have a healthy skepticism of IVs.

Quick note on Mario Rizzo and food stamps

Mario Rizzo argues here that foodstamps are becoming a new entitlement, highlighting the fact that their use has not declined with unemployment, suggesting that they are not offering a counter-cyclical stabilizer the way Keynesians suggest.

This is a perfect example of where it's important to use employment to population ratios rather than unemployment rates. If you use the latter you probably think we've been going through a slow but steady recovery. If you use the former you think we've been virtually at a standstill for the last couple years.

And note that's the behavior of food stamp participation - it's growing, but the growth has leveled off as declines in employment have leveled off. Think about what you're measuring. The unemployment rate is an important indicator but if you're concerned with economic hardship more generally - or with the volume of employment (the Keynesian concern) there are probably better statistics particularly in a recovery like this one.

As Rizzo notes, there are changes associated with benefit take-up. That might imply a leveling up of food stamps (and if we like the eligibility rules now and after this recession is over, what's the real problem with robust take-up?). But as far as the counter-cyclical question he raises, I don't see any reason to believe it's not just as counter-cyclical as ever. That's because he and I seem to have different perspectives on where in the cycle we are exactly.

I find his discussion at the end about the self-generating recovery nature of Keynesian policies very odd. But then again, he thinks we've had a lot of stimulus and I don't see much evidence for that after 2009/early 2010. That's on top of the fact that liquidity traps and financial crises are going to be especially hard to climb out of even in the best policy environments.

HET class blog post of interest

We talked about the early monetarists on Friday and I posted this question. It's pretty straightforward for most readers of this blog, but perhaps some will get something out of it. Still always good to remember:

Accounting identities/Tautologies vs. Behavioral laws

Posted by Daniel Kuehn at Sunday, November 3, 2013 4:59:15 AM EST

In class we went briefly over how the quantity theory of money is just an accounting identity, and how Wicksell, Fisher, and Keynes pointed out that it's dangerous to draw conclusions from accounting identities without having specific economic theories about the behavior of the variables in the identity.

Another identity you're all familiar with (that we'll work with in studying Keynes in the next couple weeks) is the national income equation:

Y = C + I + G + NX

If I were to present as my argument "The national income equation shows us that if you increase G, government spending, you will increase Y, national income, so policymakers should increase spending in a recession", is this a good or a bad argument for fiscal policy? More importantly (because obviously I wouldn't be asking the question if it was a good argument!): why is or isn't it a good argument, and if it isn't a good argument what would a good argument look like?

One more point on perfect competition (and why I think Don Boudreaux has been so confusing)

Textbook perfect competition really boils down to two things: (1.) nobody has market power, and (2.) nobody can earn an economic profit.

If  you are criticizing someone who is simply saying that most people have market power and lots of people can earn an economic profit, then don't be surprised when people think you are presenting a naïve perfect competition view in contrast to the way competition works in the real world.

Against "perfect competition"

On the comment thread of a post by David Henderson, Don Boudreaux shares a great blurb for Alan Manning's book "Monopsony in Motion" that drives home an important point:

"What happens if an employer cuts wages by one cent? Much of labor economics is built on the assumption that all the workers will quit immediately. Here, Alan Manning mounts a systematic challenge to the standard model of perfect competition. Monopsony in Motion stands apart by analyzing labor markets from the real-world perspective that employers have significant market (or monopsony) power over their workers. Arguing that this power derives from frictions in the labor market that make it time-consuming and costly for workers to change jobs, Manning re-examines much of labor economics based on this alternative and equally plausible assumption."

If you've read the book, you know that Manning hits this idea of perfect competition hard in the first chapter, pointing out how broad the use of it is in textbook economics and how little time is spent with monopsony cases (of course, quite a bit more time is spent with monopoly - he's thinking in terms of labor texts).

The conversation with Don has been confusing to say the least. On David's thread he was reading this as Manning criticizing markets. At the time that struck me as Don confusing blackboard perfect competition with markets, and I told him as much. Now he's got a post on his own blog saying precisely the opposite - contrasting perfect competition with market competition in the real world.

So now he seems on board with Manning and me, but he still seems to think he's not on board.

So I'm hopelessly confused by Don now, but the fact remains that it's critical to draw this distinction between perfect competition like you learn in Econ 101 and market competition, which is full of frictions and irregularities and which unfolds over time as a part of a process of social interaction. In the market you are always dealing with a lot of market power - monopolistic and monopsonistic. Entrepreneurs (among employers and employees, I should emphasize) spend their time doing two things - (1.) figuring out how to compete away the market power of others, and (2.) figuring out how to gin up some market power for themselves. Any time you do something to make yourself indispensable you're generating market power for yourself. You have monopoly rents over what you have to offer, to say nothing of the power over the information about yourself in your brain. And we use that power all the time in the real world.

So the real world is very much Alan Manning's world. Markets are a process where people figure things out and develop new advantages for themselves. It's not textbook perfect competition.

The best that can be said for perfect competition is that it can offer a useful simplification in a bigger model in some cases. Old Keynesian theory worked that way. It used perfect competition assumptions because it was easy, others were assuming it, and the focus was elsewhere. Both Post Keynesian and New Keynesian analysis has dropped those assumptions because we can do better. But the assumptions served their purpose for a time (and they're still useful when you're introducing people to Keynesian ideas). There are presumably other examples like this.

But never mistake the occasional pedagogical value of perfect competition for a picture of the real world.

Some links

- David Henderson makes much the same point I did earlier about Don Boudreaux's weak case against monopsony in minimum wage labor markets.

- Jared Bernstein on the problem of the coincidence of market failure and government failure. Surprisingly, a lot of people see government failure and suggest we should just institutionalize it rather than try our best to do something about it.

- Cash for Clunkers is not good at creating jobs. This should surprise no one, I hope. Blinder's justification is that it gave the auto industry a boost at a very rough time. I suppose that's plausible, but it's awful fiscal stimulus.

Professional whereabouts/updates

Interesting stuff going on.

First, both my Review of Black Political Economy article with Marla McDaniel and my Journal of Economic Behavior and Organization article with Signe-Mary McKernan and Caroline Ratcliffe are out as of this week. They've been online for a little while, but it's always nice to see the issue come out.

Second, I'll be at two conferences this week. I'll be presenting my job creation tax credit work at the Society for Government Economists meeting on Wednesday. The plan is to revamp this paper in the spring to get it in its final form as a dissertation chapter. Then on Friday I'll be chairing and presenting on a high skill immigration panel at the Association for Public Policy Analysis and Management conference. The first conference is relatively small, but the latter is the major public policy scholarly association. I'll be talking about the F-1, the OPT, and the issues around connections between school and work for foreign students with particular reference to the H-1B.

Finally, an interesting development is that I've picked up some contract work from the National Academy of Engineering looking at the engineering technicians and technologists labor market. Most of the work will be in the spring - should be interesting.