Some links

- Ryan Murphy has a new Cato publication calling for cutting restrictions on travel visas.

- Bryan Caplan has a new post up on scale economies in marriages. In the comments of the previous post he linked to I suggested an alternative approach that was more consistent with how economists that study household behavior model utility in the family. This one is a little better I think because it gets away from thinking in terms of utility functions and simply rehashes and explores the exercise that statisticians go through when they set poverty thresholds for different family sizes. (Recall the original post also set off a surprisingly contentious discussion about utility vs. preferences on Bob Murphy's blog - surprising in the sense that there should not have been as much disagreement with Bob and my positions on it as there was.)

- Don Boudreaux shares an interesting research effort by a doctoral student of his on the connection between minimum wages and crime here. I have some comments on how to approach such a study. One interesting thing that came up in the comments was a paper by Beauchamp and Chan (2013) on this subject using the NLSY97 (which I had suggested earlier). There is still a research angle for Don's student here, I think. Beauchamp and Chan seem to use the state-level fixed effects approach. We've talked a lot about the short-comings of that sort of analysis, and why Dube et al.'s contiguous county approach is stronger. Unless something has changed in the last couple years you have to access restricted NLSY97 data to get any geographic information on respondents below the level of Census region, so why stop at states if you've got that data?

- Robert Solow responds to Mankiw on inequality (and then Mankiw responds to him) here. Both make very good points, but I thought Solow's was excellent. I especially liked this: "It may be impractical to separate effort from happenstance numerically, but that is not reason to confound them, especially when you are thinking about taxation and redistribution"

Lerman on apprenticeship

Robert Lerman talks about apprenticeship on MSNBC.

My GME panel at the EEA conference in March

Just realized I hadn't announced this previously (we just got funding for the trip from AU yesterday, which reminded me to mention it).

I'm going to be chairing a panel on generalized maximum entropy (GME) methods in economics at the Eastern Economic Association conference in Boston this March. We'll be presenting at 8 am on March 9th, so if you are attending please come by and see it! All of the papers come from research that we did in an information theoretic econometrics seminar this past fall.

Paul Corral and Mungo Terbish will start by presenting their paper on a new STATA command they developed for discrete choice GME models. I believe the ado files are available now, and the paper itself is under review at The Stata Journal. Since they are basically explaining how the program works, this presentation will be a nice introduction to GME methods for audience members that are not familiar with it.

After that Ermengarde Jabir will present work that uses GME methods to look at household asset allocation using Survey of Consumer Finance data. This rounds out the panel nicely because it's an application of GME methods to an actual problem in economics (unlike my paper and Paul and Mungo's paper).

I'll conclude with a discussion of a GME version of propensity score matching (with estimation of the propensity score using a discrete choice GME model and estimation of the average treatment effect in the second stage as usual). I'm specifically interested in whether GME outperforms standard discrete choice estimation of the propensity score in cases of low common support. Eventually I'd like to run this method on LaLonde's (1986) data from his classic test of propensity score matching, but that probably won't be ready by the conference.

$2 million from Templeton to F.A. Hayek Program for Advanced Study in Philosophy, Politics, and Economics at GMU

Here.

My first reaction to it was a feeling of excitement and congratulations. That sort of money can go a long way in a program like this. But it was frustrating and deflated my enthusiasm to see the work of the program described this way in the announcement:

"Boettke, who was profiled in the Wall Street Journal as “the intellectual standard-bearer” for the revival of Hayek’s ideas, continued: “The failures of Keynesian economics to explain the recent economic crisis or to lead a recovery demonstrates the need for an approach that is grounded in the way the world actually works. We think our analytical framework can be that approach.”

Again we have the absurd "battle of the century" mentality. There are smart people at GMU that do interesting work, particularly in this nexus of philosophy, politics, and economics. When I hear something like this, though, I worry about the prospects for good economic science (I'll leave the philosophy and politics to better judges than me) coming out of the effort. If you take the contributions of Keynesianism to economic science to be a "failure", and if you see Hayekian and Keynesian contributions as competing such that in an announcement about the F.A. Hayek Program you have to mention the alleged failures of Keynes I worry about the quality of the output of such a program.

Perhaps this is just Pete Boettke being Pete Boettke. It's the sharp turn in my reaction from excitement for the program to despair at more of this "battle of the century" claptrap as I read through the announcement that made me feel the need to blog about it.

To sum up, it just seems to suggest that we have more stupid food fights to look forward to. Hopefully I'm wrong.

Dave Churvis hoisted from the comments on the minimum wage and racism

I alluded to the fact that the minimum wage/racism connection is a little weird to begin with, and Dave Churvis has an excellent response to that:

"I'm going to go out on a limb here on the third point and give the following hypothesis: the minimum wage *might* have had an effect of crowding out black labor (since black workers obviously typically made less), but this effect would have been minimal at best, and any business owner would have been firmly against something that would crowd out black labor. I say this by analogy to something similar that happened in South Africa during the early 60s. They passed a law that required wages for black workers to be set lower than wages for white workers. The result was a surge in employment for black workers (and remember, this is South Africa in the 60s, one of the most racist environments in history). Why? Because regardless of anyone's built-in prejudices, the fact was that black labor was cheaper, and therefore profits were higher. The law was repealed very quickly, and other, more effective ways of suppressing the black population were implemented.

So I am inclined to believe that, while it's feasible that there may have been racist intentions behind early minimum wage legislation, any effects that disproportionately happened to black workers would have been absolutely DWARFED by the negative effects of the general racism present in society as a whole. This is not an easy hypothesis to test, but I believe it could be done."

I think this is dead on. Anyone that knows anything about how old school racial caste systems worked in the South knows that white power brokers didn't want blacks unemployed. That's the last thing they wanted. Hell, for a couple hundred years these guys imposed a maximum wage of... errrr... zero. Even in the twentieth century, their goal was not a bunch of unemployed black people.

The goal of Southern racism was (and is) to keep blacks in "their place", not to keep them out of work.

Three quick minimum wage memes to push back on...

So as the minimum wage blog discussions drag on I've run across three memes (the original "meme", not the pictures) that I want to push back on a little (although one I need a little help with... it's more of an inkling right now).

1. Privileging theory.

The first is the tendency to say that when empirical evidence contradicts solid economic theory, you have to dismiss the empirical evidence. I have seen this put out there with varying degrees of strenuousness, of course. In any case, my view is that this way lies madness. You might be very curious about an empirical result for which there is no theoretical explanation (although in this case, of course, there is). You might want to think about how you might be getting that result. In my dissertation I'm coming up with a funny finding - a labor demand subsidy is resulting in reduced employment but increased earnings. Strange. My working hypothesis (that I'm going to test) is that part-time jobs converted to relatively fewer full time jobs without necessarily changing total hours worked (you have to offer a full time job to get the subsidy). I have no obligation to take my result at face value. But I don't just dismiss it because it doesn't at first blush match how I think labor markets work.

The moment we start privileging theory over empirics like that, our theories are going to start getting unmoored from reality and we're going to have doctrines and dogma rather than theory. Science is an interaction between theory and observation - and a constant revision and testing of theory to ensure that it is a story that best explains the world we experience. You figure that out by going and collecting data on the world we experience.

2. Empiricism as a popularity contest

I've also seen references to the fact that X number of empirical studies support the view that the minimum wage has disemployment effects. Who cares? You don't do meta-analyses by counting up studies. You organize and assess the studies by the quality of the methods they use. Any undergrad STATA monkey can run a state-level fixed effects model. That data is readily available and the model is very straightforward. So I don't care if there are a million variants of a state-level fixed effects model telling me that the minimum wage has disemployment effects. I mean, it's nice to have confirmation that that model works out that way I guess but when you use similar methods on similar data, you tend to get similar results. The important question is whether that's the right specification. The quasi-experimental studies may be outnumbered (I don't know this for sure, but I suspect it's true, particularly if you start trolling the lower tier journals [the original "trolling", not the thing Ryan Murphy does]), but the point is they are better. That's not to say you can't criticize them. Bob Murphy recently raised some important criticisms of how the quasi-experimental studies use time trends, which you may or may not be convinced by. But by almost any econometric standard the identification of the quasi-experimental studies is stronger than the fixed effects studies.

This is very similar to the fiscal multiplier debate or the immigration debate (in the empirical economics literature). These debates are not about counting studies - they are about dueling methods. If you run fiscal multipliers with certain identification techniques, you tend to get a certain magnitude of results. If you run immigration impact on native studies with Card's methods you tend to get different results from if you run it with Borjas's methods. Knowing this, nothing could matter less than how many of each type of study was run and published. What matters - once we establish the basic distribution of point estimates - is which specification we think is the best one. We might still disagree on that, but you don't argue that point by counting studies.

3. Racism and the minimum wage (this is the one I need a little help with).

I've heard a lot that the original proponents of the minimum wage wanted to push blacks out of the labor market. It's a little hard to parse... it requires the assumption that blacks are less productive than whites and that in the early twentieth century people needed the minimum wage as an excuse to discriminate against blacks. But it always seemed plausible to me simply in the sense that lots of progressives at the time were racists and progressives favored minimum wages. (And you can wave your hands over the fact that opponents of the minimum wage at the time were pretty racist too, I guess.)

The other day for completely unrelated reasons I was reading a portion of Bruce Schulman's book From Cotton Belt to Sunbelt, on the economic development of the South from the 30s to the 80s. He was briefly discussing the minimum wage legislation in the 30s and he noted that the proponents were advocates of black Southerners and that the opponents were the defenders of Jim Crow and racial terror in the South.

I doubt it's a clean story. You've probably got a mix. I am very concerned now that people who argue this point are cherry-picking cases and that there's no real solid correlation. But I do need help on this - does anyone know anymore details? My advice is to be cautious about this meme.

CBO on labor supply effects of the ACA

Here. Estimate is 2 million lost FTE jobs by 2017, 2.5 million by 2024.

I imagine in the near future we'll have sharper minds than mine discussing what to think of this. Just seemed worth noting.

Productivity growth for fast food workers

The other day I wrote (emphasis on the bold):

"I don't think that's the end of the story, though, because if you want to know whether the minimum wage is binding you really do have to think about trends in productivity, because that's what ought to determine labor demand. EPI has also done excellent work tracking what the minimum wage would look like if it had grown since the late sixties at the rate of productivity growth. Now productivity growth as it is measured is NOT the same thing as marginal productivity that we use in theory, but I think it's a fair proxy to consider when we're ballparking this sort of thing. And of course we don't have productivity information for minimum wage workers (although BLS now breaks it out by industry - perhaps someone should take a look at productivity growth in typical minimum wage industries)."

It appears the same day I wrote that, someone did take a look at food services productivity growth. The result?:

"Taking a longer view, from 1987 to 2012 the same BLS data show that worker productivity in the food service sector rose by an average of 0.6 percent per year. In limited service restaurants, the gains were slightly lower, only averaging 0.5 percent per year. Meanwhile, unit labor costs have risen by an average of 3.6 percent. Over this period the minimum wage has risen from $3.35 to $7.25 per hour which is an average annual increase of 3.1 percent. In other words, at least in food service, the minimum wage has risen at a rate five or six times as fast as justified by the gains in worker productivity."

Now at the end there he is doing the average nominal growth rate. Average annual real growth rate is about 0.96%*. So it's not quite as dire as the author suggests, but it should move your priors towards the idea that a 39% nominal increase to $10.10 is not a "modest" increase by the standards of the empirical literature and guesses at what the binding level would be (we still don't know that, of course - this data just gives you less room to make that argument).

* So I say "about" because I'm using Larry Mishel's real minimum wage values for 1989 and 2011 rather than the author's nominal values for 1987 and 2012, because I can't find immediately at hand real numbers for those years.

Help the Mises Institute of Canada Blog out, Bob Murphy!

I forgot to mention that Bob is also going to be coming out with a post on the Mises Canada blog on the minimum wage... and they really need his thoughtful suspicions of the quasi-experimental literature!

Currently they have a post up on the minimum wage seeking out "the world-view behind minimum wage advocacy". I'll give you a hint... it starts with "M" and ends with "arxism"!!!!

Now I would be the last to argue that most supporters of the minimum wage ground their views in a thoughtful consideration of what theoretical and empirical economics has to say (although some do). But do we really have to reach for Marx to talk about this? I don't think they're supporting it because of Marxist theories of power and property either.

Hurry Bob! They need you!

Bob Murphy on the Minimum Wage at EconLib

His long-awaited post is here (well, long-awaited by me because I've been discussing this a lot with him lately!). David Henderson's summary is here.

The post comes in two parts - first arguing that even if what I'll call the quasi-experimental studies (Dube's work and related papers) are right it does not mean raising the good idea. Bob presents familiar, and I think strong, arguments about the fact that the increase to $10.10 is not "modest" by the standards of the study, and second that even if employment doesn't decline, employment for disadvantaged workers might.

Both I think are essentially right, although I probably wouldn't make the point about a modest increase quite as strongly as Bob does. As I wrote up recently, the proposed nominal increase is certainly larger than what is typically studied in these studies, but that is only half of the equation. You also have to consider changes in labor demand, and productivity statistics strongly indicate that the point where the minimum wage is binding has probably been increasing faster than the minimum wage for quite awhile. Forget the fact that labor productivity as measured by the BLS isn't exactly marginal productivity - if one serves as a decent proxy for the other, the faster growth of productivity than the real minimum wage decade after decade seems notable if we are talking about whether the increase is a "modest" one or not. So I think Bob makes an important point here, but it's only half of the point that ought to be made.

The second half discusses whether the quasi-experimental studies should be trusted at all. We've been over this ground a lot recently. You know I think that:

1. The contiguous county sample is ESSENTIAL.
2. Meer and West have a strong critique (although it would be strange to think of when it might apply in the real world), but...
3. Dube seems to have done precisely what I thought would be the right response - using trends from the pre-period.

What's nice is that Bob's piece brings up Neumark and Salas, which I haven't discussed here. They show that alternative time trend specifications reverse some of these results (although I don't know if this is with a contiguous counties sample - Bob, do you know?). I don't know the paper well - my one questions is whether there is an overfitting issue - basically the Meer and West critique could very conceivably apply to Neumark and Salas.

That's just thinking off the top of my head - curious what you all think of Neumark and Salas.

I think Bob is right to treat these as open questions (both the scientific question and the policy question), but I think on the scientific question the quasi-experimental literature is quite strong. From my perspective, identification and eliminating bias in the estimate is the primary question - so the contiguous county studies carry a lot of weight with me.