Arnold Kling  

Education vs. Money Management

PRINT
What I'm Reading... Some Health Care Tidbits...

Consider the following two propositions, one in education (E) and one in finance (F).

(E) One way to improve education would be to get rid of the bottom 10 percent of teachers and to try to replicate more widely the techniques used by the top 10 percent of teachers.

(F) One way to improve financial markets would be to get rid of the bottom 10 percent of money managers and to try to replicate more widely the techniques used by the top 10 percent of money managers.

What is interesting is that I know many economists who believe (E), although they may be skeptical that the best teaching can be replicated. However, I do not know any economists who believe (F). Certainly, no suggestion like it was ever made while economists were discussing financial reform. What accounts for the difference between the belief in (E) but not in (F)? Some possibilities:

1. (E) is true beyond all doubt, and (F) is false beyond all doubt.

2. (E) and (F) are both true in some sense, but the market has already solved the problem of pushing out bad money managers and replicating as far as possible the techniques of the top money managers. In education, the absence of market forces is the problem.

3. They are both true, but public policy needs to worry much more about education than finance. However, consider: could you have prevented the financial crisis if in 2003 you had gotten rid of some of the worst performers from All the Devils are Here and instead replicated the heroes of The Greatest Trade Ever (also found in The Big Short)?

4. Both (E) and (F) are false, but economists have much softer priors about (E). That is, economists are strongly inclined to believe that outstanding performance in money management is luck. As a result, when someone claims to find something like (F), you can be sure that a significant effort will be expended on research designed to disprove that finding. On the other hand, economists would, if anything, like to believe (E), so that when someone claims to find something like (E), relatively little effort is expended trying to disprove that finding.

I put most of my money on (4).



COMMENTS (30 to date)
Eric Falkenstein writes:

The bottom 10 percent of money managers lose their jobs at much higher rates. Fund inflows are very convex (like a call option), but the fund manager, especially new ones, who severely underperforms over 12 months has a hazard rate significantly above that of average performers. A new fund manager, with lifetime performance below average, basically has to 'cure' that to start selling. Within hedge funds, underpeformers are dropped, outperformers get more money.

I suppose you are thinking about some other mechanism. I think the above is efficient, and don't find that confusing. Indeed, 44WallStreet was famous for being an anomaly, a fund that consistently underperformed--but it stood out because it was an anomaly.

Silas Barta writes:

How about:

5) It's fundamentally impossible for "everyone to beat the market", but possible for all students to improve test scores.

Re-adapting to what the top money-managers would do changes distribution, not efficiency. Now, if you expand money managers to include e.g. venture-capitalists, who make genuine efficiency improvements...

Tracy W writes:

Another option - financial market's problems are more about contagion, situations where individually rational reasons can leave to overall bad outcomes, say bank runs - if no one withdrew their money the bank would have been fine, but because everyone was worried that the bank might fail, they withdrew their money, so the bank failed. Eliminating the bottom 10 percent of performers doesn't avoid that problem.

It's hard to think of similar stories in education.

I'm also puzzled as to why you think that (4) is a stronger contender than (2). What test do you think could distinguish between your four options? I find myself interested in taking the other side of your offer to put your money where your mouth is on this one (though I'm worried about the noisiness of the potential signal).

Finch writes:

Both are true, but the way to identify bad money managers is to look at their fees, not their performance. :)

Even if you ignore fees, regardless of what one believes about extremely good performance, one might expect bad performance to be persistent because it's partially caused by things like transaction costs and poor tax management that eat away performance without ever violating the efficient market hypothesis. These things don't necessarily show up as fees.

A dude writes:

Simple really... Money management results measured in relative performance is a zero sum game (slightly negative if you include the fees). Bottom 10% are just as valuable to risk-reward discovery that leads to efficient resource allocation as top 10%.

Education is hopefully not a zero sum game.

If you look at education only as a market for credentials, both industries are selling illusions about the future. But education is much more than that.

mark writes:

I suppose the difference is that people believe, right or wrong, that a bad teacher is a bad teacher at all times and under all conditions. A perma-bull money manager can at least be right when the economy is going strong. I do find it funny when I hear so many people talk about firing the bottom ten percent of teachers as if they could identify them and those teachers are always in the bottom ten percent. We certainly don't think that is the case with baseball players. Which kind of teachers are in the bottom ten percent? If inexperience is a factor constantly firing people doesn't seem like the solution.

Julien Couvreur writes:

I believe that 2 is the case (E is a monopoly, while F is not), with caveats. Both have unique challenges in terms of measuring performance and incentives. Also F has many anti-competitive interventions.

Floccina writes:

After reading the first few line I though that you were going to say "No one advocates paying more money to money manager to improve finances but many advocate paying more to teachers to improve schools".

Various writes:

Both E and F are probably relatively false because, ex ante, it is difficult to identify the best and worst 10%. However, market forces are at work to attempt to implement F. In other words, the market is essentially firing the worst 10% of money managers (or at least what it believes are the worst 10%) and promoting the best 10% (or what appear to be the best 10%). No such mechanism exists for teachers for grades K - 12. Although imperfect, probably the single most powerful mechanism for improving education would be the introduction of genuine market forces, i.e., school choice and/or teacher choice, with the money going to only those schools and teachers in demand. Significant inefficiencies would likely exist in such a system, just as they exist in the market for money managers, but the quality of schools and teachers would likely improve nonetheless.

John Thacker writes:
I do find it funny when I hear so many people talk about firing the bottom ten percent of teachers as if they could identify them and those teachers are always in the bottom ten percent. We certainly don't think that is the case with baseball players.

Huh, mark? We certainly do think that that is the case with baseball players. People who study baseball stats make very precise allowances for age and experience and predict with a great deal of accuracy how a baseball player's stats will translate from the minors to the majors and how they'll develop.

To the degree that this point is true, mark, you need to explain how teachers are not at all like baseball players.

David N writes:

Thoughts on E: Bad teachers have almost nothing to do with the poor state of education today. There have always been bad teachers and always will be. You've all had at least one. A motivated student will survive one or two bad teachers. The percentage of teachers that through incompetence or malice are actaully harmful to students is likely far less than 10%. If you factor out income or education level of parents, how much of student performance is actaully dependent on teacher quality?

Thoughts on F: EMH teaches us that this year's top decile money manager has a fair chance to be next year's bottom. The percentage of money managers that actually have positive alpha is likely far less than 10%. To whatever extent success depends on asymmetric investing or trading skill, the participation of the dumb tenth increases the returns of the smart tenth.

I'd love to know what Jack Welch thinks of propositions E and F.

John Thacker writes:

I actually know a lot of educational reformers who reject (E) but adopt (E') as follows:

(E'): It is unreasonable to expect that the vast majority of teachers will be able to replicate the techniques of the small number of highly skilled natural teachers, and indeed for them to develop individual approaches to teaching like a craft. One way to improve education is to force the lower 90% of teachers (or even all) to follow more regimented techniques that will produce better results for the average teacher, as demonstrated by study upon study. (Direct Instruction, Success For All, some others.)

(E') is even more hated by teachers than (E). It's very odd. Teachers have an approach to the profession as though it were an artistic craft full of unique techniques, perhaps passed from mentor to apprentice. However, they strongly reject the sort of wildly varying pay scales and star system that such an approach would suggest, instead arguing for standard pay scales of the sort associated with regimented unskilled labor-- the very type of labor and teaching techniques that they also reject.

Either teaching has a individual skill component that cannot be easily replicated, in which case merit pay of some form (not necessarily test score based) is appropriate, or completely regimented pay scales are appropriate, in which case teachers are indistinguishable and it is reasonable to force them to adopt similar methods.

Chris Stucchio writes:

2 is clearly the case, mainly due to the fact that in investing, performance multiplies influence.

Any given teacher can have, at most, 2-3x the influence that any other teacher has (say one teacher has 15 students, another 45). No matter how great a teacher is, he/she can only influence 45 students at time.

George Soros and William Gross both have a influence literally millions of times greater than mine. The bottom 10% of money managers are broke, and have no influence on the market.

(You actually do see this effect with one type of teaching, namely PhD advising. The best advisers get the most students.)

Dave Schuler writes:

To amplify a bit on Various's comment above, which I think is on the right track, is there a widely accepted metric for performance in education? Is there an equivalent to "teaching to the test" in finance?

John Thacker writes:

Most of the educational reformers that I know do *not* try to spread the techniques of the top 10% of teachers. They try to spread techniques that can be used by median and even bad teachers and get better results, as study after study shows.

However, teachers remain stuck in the contradiction. They want to teach and be treated as though it's a craft, while being paid as though it's unskilled labor (and all teachers are equally skilled.)

I could be convinced either way by the evidence, that teacher quality either does or does not matter. But I'd like to approach it consistently.

Frank Howland writes:

I don't think that bad money managers were fundamental to the financial crisis. Sure, there were lots of stupid people out there managing lots of money, but poor monetary policy, bad governance at large firms, deregulation and poor regulation (allowing too big to fail, high leverage, etc.), and poor housing policy are to blame. Bad governance allows bad money managers to exert undue influence and poor regulation gives large firms employing very smart money managers the incentives to take risks which are bad for the financial system.

David N writes:

High leverage is not a failure of regulation. If you're suicidal the laws against it are not an effective deterrent.

Floccina writes:

I am making this second post to point out:

The amazing truth about PISA scores: USA beats Western Europe, ties with Asia. from Super-Economy by Tino

What I have learned recently and want to share with you is that once we correct (even crudely) for demography in the 2009 PISA scores, American students outperform Western Europe by significant margins and tie with Asian students. Jump to the graphs if you don't want to read my boring set-up and methodology.


http://super-economy.blogspot.com/2010/12/amazing-truth-about-pisa-scores-usa.html

Also John Thacker makes some great points especially about direct instruction.

[extraneous facebook-related counts removed from quote--Econlib Ed.]

Sridhar Loke writes:

Arnold,
This is not a correct comparison. It is a teacher's job to positively influence the education. A money manager's job is to "predict", intelligently as some people may assume, the outcomes of players. Money managers don't influence the performance of a company whose stock or debt they buy. Their job is to pick winners, and a teacher's job is to create winners.

Dan Weber writes:

What the "top money managers this year" do is probably dangerous for others to reproduce. They probably took risks that happened to pay off this year.

But there are teachers who consistently succeed year after year. Reason's old article on "Stand And Deliver" has been making the rounds again, and Escalante really did consistently perform well. He ended up being punished instead of rewarded for his success, though.

Dave writes:

The bottom 10% of money managers help other money managers. They are the suckers that the better traders profit from trading with. "Dumb money." There is no such dynamic in education. Poorly educating inner city children in DC does not help students at Exeter. Maybe it does in a "more highly skewed relative income inequality" type of way, but not with their learning. In fact, it may even make the Exeter kids ultimately poorer, because the poorly educated kids won't become as productive later in life to expand the economic pie.

mark writes:

Somebody here can make a heck of a lot more money consulting for a Major league baseball team then reading Econolog. If your a top 100 prospect in baseball there is about a 30% you will be a legitimate success in the Major leagues and about a 70% chance you will be some kind of bust. Hitters who can hit very well in the minors can probably hit in the majors but beyond that projections are far from a "great deal of accuracy". I don't doubt anyone can draw distictions from the top 10 percent of teachers and the bottom 10 percent of teachers but as a practical matter I was more interested in the difference between the bottom 10% and the bottom 20%. I don't think it is an easy call but I'm not a teacher or a pither for that matter.

Seth writes:

Luck is a market force.

John Thacker writes:
If your a top 100 prospect in baseball there is about a 30% you will be a legitimate success in the Major leagues and about a 70% chance you will be some kind of bust. Hitters who can hit very well in the minors can probably hit in the majors but beyond that projections are far from a "great deal of accuracy".

But there's absolutely is completely different treatment of players who are in the top 10% of prospects, or top 10% in the minors, or top 10% in the majors, and players in the bottom 10%. It has an enormous impact on their pay, career prospects, and makes an enormous difference as to their expected career.

We certainly do think that we can identify the bottom 10% of baseball prospects in the minors and we don't think that they'll get better and we cut them.

Baseball players are not treated like teachers. Some are treated as stars, and paid like it.

Either individual teachers have a dramatic effect, and stars should be paid like stars, or individual teachers don't have a dramatic effect, they're all interchangeable and they shouldn't mind using lecture notes and regimented methods of learning derived from someone else.

John Thacker writes:
If your a top 100 prospect in baseball there is about a 30% you will be a legitimate success in the Major leagues and about a 70% chance you will be some kind of bust.

And do you think that those chances of success are significantly lower if you're a less than top 100 prospect in baseball?

If so, you've just vitiated your own argument, mark.

Psychohistorian writes:

There is good reason to suspect the market for money managers is efficient enough that their performances are largely random. This is emphatically not the case for the market for teachers, who are not paid by their beneficiaries and who often have an essentially captive market, no ability for clients to switch teachers, and incredibly strong job protections that are largely competence-blind.

The Efficient Markets Hypothesis requires something vaguely resembling an efficient market. There isn't much reason to expect that to exist for teachers.

q writes:

the comparison isn't quite correct.

EMH would imply, say, that if i took a random bunch of stocks i'd do about as well as the market as a whole and beat half the funds. but if i stood in front of a classroom and said and did random stuff, i'd be among the worst of the teachers.

perhaps, though, futures/betting markets on teachers (or on their students' future income or some other handicapped metric of success) could provide a decent way of judging teacher performance prior to their students maturing. maybe robin hanson could invent such a market.

Steve Sailer writes:

Teaching is less like being a baseball player and more like being a baseball manager. Do we have surefire statistical ways to tell which are the best baseball managers and which are the worst?

Granite26 writes:

For E: why are you bundling getting rid of the bottom with duplicating the top? They solve two different problems. Getting rid of the bottom implies teachers can't be taught, duplicating the top implies they can. It's a contradiction.

Also, I like the zero sum game theory for F. Any reason it's not true?

Jeff writes:

The comment by 'A Dude' is obviously correct. So much so that I find it hard to believe that it didn't occur to Kling before he made this post. These "say something obviously wrong to try to provoke a discussion" kinds of posts are what I find most annoying about many of Tyler Cowen's posts at Marginal Revolution. Please, Arnold, don't go down that route.

Comments for this entry have been closed
Return to top