Friday, October 12, 2007

How Not to Solve Inequality Problems

Ezra Klein has produced an amazingly stupid post on inequality, and referenced an equally dumb assertion that education can't solve inequality problems.

As I have mentioned before, I respect Klein's opinion on a wide variety of topics, even when I disagree with it. This one seems so wildly far off the reservation that I wonder if I'm missing something.

If inequality really were a function of higher returns to education -- and, sorry folks, but it's not -- we wouldn't respond by demanding everybody get dumber. We'd change the tax code around a bit, institute some progressive social programs, put seed money into asset programs for the poor and working class, and so forth.

Let's look at a quick-and-dirty decomposition of the American job market. There are:

  • Professional jobs. These require at least 4-year bachelor's degree and a high IQ. They're expensive to obtain but, once obtained, have nearly unlimited job security and chances for advancement. Furthermore, in the globalized economy, there is effectively no limit to the number of professionals that the economy will absorb. I'm sure there's a limit up there somewhere, but we're nowhere near satisfying the global demand. Indeed, the more professionals we inject into the economy, the faster it seems to grow.

    Clearly, education produces lots of returns here.

  • Skilled, geographically localized jobs. These jobs require some training (1-2 years) and have the characteristic that they can't be outsourced. Plumbers, HVAC techs, network and PC installers, lab techs, etc. These jobs are not quite as secure and they have a definite wage ceiling on them, but a family with two skilled workers in it is going to be solidly middle class.

    Education is providing excellent returns here but, unlike the professional jobs, the market can't absorb an unlimited pool of labor. The job pool grows as the economy grows, but if we trained every available worker for one of these jobs, they wouldn't all be employed.

  • Skilled, outsourceable jobs. Here's where things start to go off the rails. These jobs require the same level of training but they're going overseas because skilled foreign workers can do them just as well. CPAs, web-masters, graphic artists, and a whole slew of office workers fall into this category.

    This is an area where education can be misapplied. It's not that you can't compete in this market, but you have to have an edge. That edge can't be provided by education.

  • Semi-skilled jobs. Think factory workers. You probably need a high school education and a little luck to get these jobs, but they're going offshore at a huge rate. These folks are going to be unemployed and are likely to be pushed into unskilled jobs.

    Again, other than simple welfare redistribution, is there any solution for these workers other than to get them education necessary to get a skilled job? I think not. However, we're once again bumping up against the fact that the economy only has a certain capacity for skilled labor.

  • Unskilled jobs. Retail. Low-end construction. Pure service workers. There is, oddly, more capacity in the labor market for these jobs, but there's absolutely no way to put upward prices on wages except by statute.

If you take a look at this taxonomy, you can see why inequality is occurring. The top end of the scale is infinitely elastic and utterly secure, while the bottom end grows but is utterly price-inelastic. And the middle is shrinking due to outsourcing.

Indeed, you sort of see this with the constant invocation of education. Saying this is about skills -- as if hedge fund managers have been studying since 1690 -- is a way of justifying it. Saying we need more and better education is a way of doing nothing, as that's much too grant a task to undertake as a response to inequality and, in any case, it wouldn't work. College graduates have seen their wages stagnate for the last six years. This is about distribution. And distribution can be rejiggered without digging deep into the everyday workings of the economy.

I guess I just justified it above--bad RadicalModerate. But there's some real rot in Klein's statements here. First of all, let's not kid ourselves: he's talking about simple income redistribution. He wants to tax the top end of the scale and disburse the proceeds to the bottom end of the scale, in perpetuity. His justification for this appears to be that some people make more money than they deserve.

Some questions: First, How equal will be equal enough? If the top end continues to run away from everybody else (which they will, for the reasons I've stated above), will we prevent them from running away? Are we going to institute a 100% tax bracket? Second, is this going to be the new status quo? A pure welfare state? Is Klein really so indifferent to the causes that he's not willing to try to engineer a self-sustaining fix?

Of course, he might be right. Given that there's limited economic capacity in the middle and unlimited capacity at the top, things may just get more and more unequal, especially for those at the bottom. Of course, the dynamic shifts when the rest of the world reaches something like equilibrium with us and the wage disparities go away. But that's probably thirty or forty years in the future.

Meanwhile, the only way to really screw things up is to cap the top end. And that's the only way to "solve" inequality. The proper goal is not an equal system. The proper goal is a society where the vast majority can live a decent life. All else is the most poisonous kind of class-warfare.


Sergey said...

At first place, why we should be so obsessed with inequality? Is it really a problem for stability of society, as it was 100 years ago, or it is not? As I see it, when poverty is only relative, but not absolute, inequality is accepted without much fuss. People adapt fast and began to see it as a natural and fair thing.

TheRadicalModerate said...


I pretty much agree. The goal is to make sure everybody's living a decent life, not to make everybody "equal." However, it's hard not to come to the conclusion that globalization is putting some pressure on portions of the skilled and all of the semi-skilled labor markets. Help with health care and job training (with financial support during job training) seem to be reasonable policies, if you can find a way to keep them from running out of control.

Sergey said...

There is a problem in US that almost every reasonable government aid program tends to go out of control and get misused by those who want to make advantage of it at everybody else expense. The worst abusers of these programs are politicians, who actually buy votes paying for them by taxpayers money.

Sergey said...

Another problem is the higher level of bureaucracy involved the less is efficiency and accountability. It is much more easy to loot federal budget than a state or municipal one. When a school board decides to give free lunches to poor students from school funds, you can be sure that this aid will be properly used. When the same decision is made by a city counsel, chances of abuse are higher. And national-wide program certainly will be squandered.

TheRadicalModerate said...


Now we're down to hard-core philosophy of government. If you could get free-market solutions to all social problems, that would be wonderful. And I completely agree with you the the Law of Unintended Consequences reigns supreme with any government-run program. Not only are such programs not very efficient (at least to start with) and sometimes corrupt, but they cause effects that you never thought about.

So: If we both agree that inequality isn't a problem, and we both agree that government remedies are difficult and expensive, how do we differ?

First, while I am a rabid free-trader and am completely convinced globalization is a) good and b) unavoidable, I do think that it causes a significant social disruption when it puts a big chunk of the middle class out of work. I simply don't see a free-market solution to that problem. For the foreseeable future, labor is going to be cheaper overseas than it is in the US. Workers in commodity jobs, even skilled commodity jobs, are going to have problems. That means that those workers are going to need to be re-trained so they can get jobs that are immune to globalization forces.

Education is moderately expensive in and of itself, but the main cost in re-training somebody is that they can't work and get trained at the same time. Can you have a part-time job and go to school simultaneously? Well, yeah, but you can't raise a family on a part-time job and you can't do well in your training--especially if it's intensive--without committing most of your time to the training.

So retraining is very difficult without some sort of assistance. My son is currently in this situation. He didn't go to college, joined the Army, and got married. Now he has three little kids and, since leaving the Army, few marketable skills. He's lucky; I can partially support him and his GI Bill benefits will allow him to spend two years in school getting decent skills. Most people aren't that lucky.

So you need a realistic, robust re-training program, with financial support during retraining. It's expensive. It's going to be exploited by some people. Politicians will lard it up with pork. It will need to be fine-tuned over time to make it somewhat more efficient.

But let's look at the alternative: If you get a bigger and bigger constituency of families that used to be solidly middle-class but are now barely hanging on, they'll get angrier and angrier. Ultimately, that empowers fairly extreme left-wing politicians, who certainly advocate for programs like this. Unfortunately, they also wish to impose all kinds of other methods of wealth redistribution and government control of many of the institutions of modern society. Giving power to left-wing extremists is as bad as giving it to right-wing extremists.

So the moderate solution is best: Provide support for re-training. I think that providing reasonable healthcare reform is important for the same reasons (see my posts from earlier this month, if you're interested).

Sergey said...

This re-training program looks alike New Deal in many aspects. Creating a large class of dissatisfied voters is a disaster for any democracy, it empowers demagogues and can undermine the system, as we seen in Weimar Germany. The Great Depression could do the same to US democracy, so FDR actually saved the country with a massive government intervention. It is not obvious to me yet that globalization can result in as big destabilization of society as Great Depression, but some measures along these lines can be justified. It seems to me now, that one of the programs to create many localized, not outsourceable jobs would be federal program building dozens of nuclear energy plants, with federal funding and lots of local private subcontractors. Private investors could not support this switch to atomic energy because of very high capital costs and long period of recoupment. Re-training should be a part of this program, that can be sold to public as alternative to Kyoto. In effect, it will make US independent from ME oil and will be a more devastating blow to terrorism than several wars in ME - and cost much less.

TheRadicalModerate said...


I could go for a WPA-like on-the-job training program for some things, but it's hard to do training for skilled jobs--a lot of them require classwork or at least off-the-job training. You can't really be a network installer or a power systems tech without learning enough to do the real job. Similarly, I really don't want the guy who's on his first construction job pouring the concrete containment for a nuke.

See my previous comments for reservations about how deeply the government should bet on nukes but some bets and subsidies are certainly in order.

BTW, since you can do math better than me, I wonder if you've been following what's going on with the Polywell fusion stuff. I give this a fairly low probability of working but the late Dr. Bussard had a pretty decent track record and the cost to confirm his previous experiments is modest. Even the cost for a full-scale prototype is only about a tenth what the US spends on ITER in a year. More resources here, here, and here.

Sergey said...

This does not seem to me promissing. Of course, this is only intuitive assessment, I could not make accurate calculations from this information, but hot plasma is devilishly unstable system. Many dosen thousand scientists worked on Tokamak design for half a century, and every new, larger machine always shown new type of instabilities: their number seems infinite. This is like to design a new supersonic jet by a team of four. Tokamak still is the most promissing design. My university friend works as leading scientist of ITER project, he is very optimistic, but, honestly, I do not expect much from it. For near future (30-40 years) advanced fission reactors are more sure bet. Concerns about used nuclear fuel safe burial are greatly exaggerated, their volume is very small. While their radioactivity is horrific, it diminish fast, and after several decades only tiny amount of it is left. And we can safely put them away for centuries.