If I Were A Poor Black Kid

If by chance you don’t frequent the geekier side of the twitterverse you might have missed the outpouring of wit-in-140 that followed this post by Gene Marks

If I was a poor black kid I would first and most importantly work to make sure I got the best grades possible. I would make it my #1 priority to be able to read sufficiently.   I wouldn’t care if I was a student at the worst public middle school in the worst inner city.  Even the worst have their best.  And the very best students, even at the worst schools, have more opportunities.  Getting good grades is the key to having more options.  With good grades you can choose different, better paths.  If you do poorly in school, particularly in a lousy school, you’re severely limiting the limited opportunities you have.

As it so happens I was a poor black and I built a rational expectations model of the very phenomenon Marks describes.

I look back on it and I see how it was the origin of the Smithian worldview I push today.

The problem is this: You want to build a model of the choices facing a poor black kid in a bad environment. You need to sketch out a decision tree and then turn that into a choice function and then – in my case – simulate the interaction between neighborhoods and student choice on a computer.

Significant insight can be gleaned from the closed form solutions but to really watch the magic you need numerical estimation.

In building this model one thing became glaring clear. The life choice that Mark’s outlines and that is advocated as prudent and reasonable by society is in fact incredibly risky.

I probably can’t convey the view-quakiness of this revelation because its now so entwined with the way I see the world. However, imagine the choice of a poor teenage girl deciding whether or not to have unprotected sex and possibly become pregnant, or to study hard, make good grades and stay in school.

Forget the unprotected sex itself, which we almost all find enticing.

The key is the pregnancy. For a 16 year-old girl regular unprotected sex will result in a full term pregnancy in the modern world with roughly probability one. There is little chance she will die in child birth. Late term miscarriages at her age are rare.

Now, just like any other parent the birth of that child will be the most important event in her life. And, the love of that child will be the most valuable thing she experiences. Some people say that looking back their career was more important than their children, but those people are few and far between.

So, if the girl has unprotected sex she gets right here, right now, the most important and valuable thing in life will happen immediately with PROBABILTY ONE.

Its difficult to get better than that. Waiting at all creates a risk that something will happen to prevent this. Even, if you can be sure it won’t – and many couples find out unfortunately that you can’t be so sure – you still have to discount the time. You have wait for the most valuable thing in your life.

Mark’s would have her set all that aside. Put away time that she will never get back – you must remember that no matter what you will never get these days back – for the chance that supposedly she will go on to college and get some job and meet some guy and then later have a different child under what might be better circumstances.

This is a risk. Taking Marks advice means that you lose a sure shot at the greatest thing in life. It means that you potentially waste time and time is the currency of life. He wants to convince you that the gamble might pay off.

Yet, how is a Bayesian supposed to tackle this problem?

I look around in my neighborhood and by definition none of the folks here have done what Marks suggests. These people are like me. I have no reason to believe that I am different.

What kind of sense would it make for me to take this gamble when no else does? No, it makes more sense to play it safe and take the sure thing.

Now, of course teachers, parents and helpful people like Marks will tell me to do otherwise. Should I believe them?

Not on your life.

By their own admission they want to see me “succeed.” That is, they benefit from my gamble. Yet, they incur none of the risks. They don’t lose time with their child. They don’t risk their fertility. They don’t experience the disutility of social climbing.

Heads they win. Tails I lose.

Listening to them would be nothing short of foolish.

And, so of course the teenage girl does not listen. Not because she is irrational, but because she is rational.

Indeed, strategies to get her to change her mind hinge on coercion or leveraging irrationality. Parents may threaten as poor parents do not have the resources to bribe.

Some teachers will try to convince students that they can do anything if they try. Clearly they can not. Others will significantly downplay the disutility of social climbing. They will cast crossing into the cultural unknown as uplifting, not depressing and possibly deeply lonely. These kind of stories border on outright deception.

Everyone will try to get her to “believe in herself.” This is an attempt to induce Caplan-esque rational irrationality. That is, to attach an emotional preference to a belief about the natural world. This is epistemologically equivalent to the nationalistic fervor that accompanies “America First.”

And, if that doesn’t work some teachers will resort to honest guilt: “We took a chance on you, now you take a chance on yourself.”

However, all of these strategies come down not to encouraging prudence and rationality, because the prudent and rational thing to do is to get pregnant. They instead hinge on emotional appeals to irrationality, noble lies and social, and sometimes physical coercion.

If you were a poor black kid, that’s what you would face.

Down with History

When schools are faced with a budget crunch, as so many are, art teachers and art classes are among the first to go on the chopping block. As the New York Times reports, this appears to be the case in New York City:

For the first time in four years, the number of certified arts teachers in the city’s public schools is declining, according to a report to be released by the Center for Arts Education on Thursday.

In 2009-2010, there were 135 fewer arts teachers in the city schools than in the previous school year, after principals chose to cut positions from their budgets or not replace arts teachers who left. The 5 percent drop puts the number of certified arts teachers working in the schools back to where it was in 2007, when the city first began to survey principals….

…Schools, however, have also cut back greatly on other spending for the arts. Since 2007, when the city began surveying schools’ art education, spending on supplies has dropped to slightly more than $2 million from over $10 million. In a system with more than a million students, that comes to about $2 per student.

This general scenario matches up with other stories I’ve seen. But why should art be on the chopping block before history class? I believe we romanticize history, making it seem practically and ideally more important than it is. People defend history in the gauzy language of citizenship, with appeals that rarely rise above aphorism. “Those who don’t history are bound to repeat it”. This doesn’t hold up in a practical sense though. There’s a reason the phrase isn’t “those who have history as a significant part of their high school curriculum are bound to repeat it”. Being taught history doesn’t make you better voters unless you remember that history. I’m not going to go down the litany of things that huge percentage of Americans incorrectly believe about history, instead I’ll just give one prominent example. How many hundreds of millions of dollars to we spend each year teaching kids about the Civil War, and still 42% of people don’t know we fought it over slavery?

If we want students to know the most important facts and stories about history for the sake of those facts, then having them watch a few documentaries should cover the bases pretty well. As Will Wilkinson pointed out to me, history is something anyone with reading comprehension can teach themselves, in contrast art is a skill that in most cases requires careful instruction. Like I said, even people without reading comprehension can learn most of what they do in high school through documentaries and the History Channel.

Ideally speaking, I don’t think history is the most important subject when it comes to making better voters. You’d do better off to drill students on economics 101 and the basics of the budget. Maybe then we’d wouldn’t have a citizenry convinced that foreign aid is a large portion of our spending, and that understands the downsides of price controls. More importantly schools could stop reinforcing the notion that voting is some moral obligation, and let people know that voting on the basis of uninformed biases is far worse than not voting whatsoever.

There’s also a large cognitive dissonance whereby we view art as being something soft, idealistic, unpractical, and unserious compared to other school subjects. Art is a fun distraction, whereas history is serious business. But of the two art is clearly the more practical real world subject. Many serious, button downed, grown-up careers require artistic skills: architects, marketing, graphic design, engineers, web designers, city planners… the list goes on. Which careers require knowledge of history? Journalists, history teachers, politicians? It seems as though the caricatures of these fields should be exactly the opposite, and that history should be viewed as soft, idealistic, and unpractical, whereas art should be viewed as the hard-nosed practical subject of serious people.

Even in these cases where a career requires knowledge of history, like those above, what they need is 60% confined to modern world history, and the other 40% to modern world history. Where does knowledge of explorers, Mayans, and pilgrims come into play in any career? In art class even when you’re learning how to do stuff you won’t directly use in the future, you’re picking up artistic skills that have wider importance than their immediate application.

For the modern economy, for the betterment of our country, I say down with history, up with art.

Defending KIPP, again

I’ve defended KIPP before from accusations that their performance is due to selection bias and cream skimming. A new paper from Josh Angrist and a bunch of other co-authors provides more evidence in KIPP’s favor. They found that KIPP benefited the weakest students and those with special needs most:

Our results show average reading score gains of about 0.12 standard deviations (hereafter, σ) for each year a student spends at KIPP, with significantly larger gains for special education and LEP students of about 0.3-0.4σ. Students attending KIPP gain an average of 0.35σ per year in math; these effects are slightly larger for LEP and special education students. We also produce separate estimates for students with different levels of baseline (4th grade) scores. The result suggests that effects are largest for those who start out behind their peers. Male and female students gain about equally in math, while boys benefit more than girls in reading. Finally, an examination of Massachusetts Comprehensive Assessment System (MCAS) performance categories (similar to quartiles) shows that  KIPP Lynn boosts achievement primarily by moving students up from the lowest group. Together,  therefore, the findings reported here suggest that KIPP Lynn benefits weak students the most.

The study also provides further evidence against the claim that KIPP’s model is based on kicking out the trouble students. The authors found that students who won the admissions lottery were no more likely to switch schools than those who lost, meaning that controlling for selection bias, KIPP students do not have a higher attrition rate.

Initial skepticism of KIPP’s high achievement results was understandable, and questions about cream skimming, kicking out bad students, and whether the weakest students were benefitting were important to ask. But the evidence continues to come in that KIPP is doing something right, and that skeptics were wrong about why that was. It’s time for skeptics to re-evaluate their positions. I’m hoping that when Diane Ravitch releases the “The Death and Life of the Great American School System” in paperback the section that is skeptical of KIPP is updated to reflect this new evidence.

Is raising a child poorly the ultimate externality?

Lisa Belkin at the New York Times reports on paternalism aimed at making parents more paternalistic:

…other states have already enacted laws aimed at improving parenting. Alaska fines parents for a child’s truancy. In California, a misdemeanor charge can be brought against a parent if the truancy is flagrant enough. California is also the first state to allow judges to order parents to attend parenting classes if their child belongs to a gang.

I’m going to take the lazy route and sidestep the whole issue of whether these types of policy are a worth trying, and just say that probabilistically, I think Belkin is correct:

In the end, then, all these “punish the parents” paradigms will probably take their historical place as just one more shift of the pendulum in the sweep that already includes contradictory certainties like “children are being allowed to grow up too quickly” and “children are being infantilized too long.” Like every other new way of thinking, it will eventually be looked on as a well-intentioned but flawed reflection of a moment in time.

KIPP vs Ravitch

In her 2010 polemic, The Death and Life of the Great American School System, Diane Ravitch has praise and criticism for KIPP charter schools. On the one hand, she recognizes the organization improves scores for students. On the other hand, she credits this success, in part, on the schools ability to kick out hard to educate students:

…KIPP schools often have a high attrition rate. Apparently many students and their parents are unable or unwilling to comply with KIPP’s stringent demands. A 2008 study of KIPP schools in San Francisco’s Bay Area found that 60 percent of the students who started in fifth grade were gone by the end of eigth grade. The students who quit tended to be lower-performing students. The exit of such a large propotion of low-performing students –for whatever reason- makes it difficult to analys the performance of KIPP students in higher grades. In addition, teacher turnover is high at KIPP schools as well as other charter schools, no doubt because of the unusually long hours.  Thus, while the KIPP schools obtain impressive results for the students who remain enrolled for four years, the high levels of student attrition and teacher turnover raise questions about the applicability of the KIPP model to the regular public schools.

Note that teacher turnover in-and-of-itself is considered a problem. I find it baffling to consider success alongside high turnover as evidence of a limitation of the sucees rather than as evidence that turnover is not necessarily a problem.

The italicized portion of her quote is of particular interest, since it is directly contradicted by a study from Mathematica Policy Institute that showed that KIPP improves test scores for students that ever attend KIPP, including those that leave early. This is a direct contradiction of her claim.

As she does throughout the book, Ravitch drives the point home with a rhetorical flourish befitting of a speech at an NEA pep rally, lamenting the unfair advantages that charter schools have, and how easy that makes it for them compared to the underdog public schools:

Regular public schools must accept everyone who applies, including the students who leave KIPP schools.   They can’t throw out the kids who do not work hard or the kids who have many absences or the kids who are disrespectful or the kids whose parents are absent or inattentive. They have to find ways to educate even those students who don’t want to be there. That’s the dilemma of public education.

Ravitch creates the image of KIPP schools taking better students from public schools, and simply kicking out bad students, sending them back into the public school system. This negative model of charter success is an important theme in the book. However, another recent study by Mathematica Policy Institute shows that her claims here are also false. They found that students leave KIPP schools at the same rate as they do for nearby public schools. In fact, for black and hispanic students, the attrition rates for KIPP were lower.

Ravitch also credit’s the lottery admissions for KIPP’s success. Her argument is that

“Like other successful charter schools, KIPP admits students by lottery; by definition, only the most motivated families apply for a slot. Charters with lotteries tend to attract the best students in poor neighborhoods, leaving the public schools in the same neighborhoods worse off because they have lost some of their top-performing students. They also tend to enroll fewer of the students with high needs – English-language learners and those needing special educaiton.”

This complaint puzzles me. Ravitch once was a supporter of charter schools. But if lotteries are “by definition” going to cream skim and advantage charter schools, how did she ever support them? Her argument here is definitional, and not a matter of data. When criticized for changing positions on education reform Ravitch likes to quote Keynes who, perhaps apocryphally, said “When the facts change, I change my mind. What do you do, sir?”, but have the definitions changed as well?

Furthermore, the Mathematica study found that KIPP did not admit the “best students”. On average, KIPP entrants were “not more advantaged than other students in their communities, as  measured by poverty and prior achievement levels”. For instance, 84% of students who attended the sample of KIPP schools qualified for a free lunch program, compared to 64% in KIPP host districts, and 72% in the elementary schools that send any children to KIPP. They enroll more minorities, and they enroll students with lower test scores than the district average and the same as the average for the public schools that KIPP students came from.

They do find that KIPP tends to enroll less ESL students and students with disabilities, but this is not the same as admitting the “best students”. Ravitch clearly agrees with this, as you can see in the quote above she includes the complaint that KIPP doesn’t admit enough ESL and disabled students as distinct from and in addition to the complaint about only letting in “the best students”.

These results are important because if Ravitch claims that KIPP cream skimming higher acheiving students makes public schools worse off, then the fact that they take lower performing students must make public schools worse off. This doesn’t just remove a complaint about charters then, it actually represents a benefit to the public school system. This positive impact of more charters and choice on public schools is reinforced by studies that have shown public schools improve in response to more competition, an entire vein of literature ignored in Ravitch’s book.

These are clear examples of the facts disproving Ravitch’s claims. Will future editions of her book correct this? Will she call attention to this fact and publicly reverse her opinion of KIPP?

How do charter schools impact outcomes other than test scores?

A common, although misleading, refrain heard from education reform critics is that charter schools on average do no better on standardized tests than public schools. Less commonly discussed is the impact on other, non-test outcomes. A paper (working version) in the most recent Journal of Labor Economics by Kevin Booker and Brian Gill from Mathematica, Tim Sass from Florida State, and Ron Zimmer from Vanderbilt, looks at how attending a charter schools affects the probability of graduating high school and attending college for a sample of students from Chicago and Florida.

An important question in the literature is the extent to which selection bias is a problem. Do charter school students have different outcomes than public school students because the charters educate them somehow better or worse, or is it because people who decide to go to charter schools are systematically different than those who don’t? In order to control for this they look at the sub-group of students who attended a charter middle school, and compare outcomes for those who went onto a charter high school to those who went on to a public high school. The idea is that selection bias shouldn’t be an issue between those who are already attending a charter school, because as charter school attendees they should be similar in terms of whatever unobservable variables lead to charter attendance.

They find that in both Florida and Chicago, attending a charter high school increased graduation and college attendance rates. In Chicago, students were 7% more likely to graduate from high school if they attended a charter, in Florida it was 12% to 15%. Charters increased the probability of attending college by 8% to 10% in both Florida and Chicago. The authors argue that these results are consistent with the studies on the effects of attending a Catholic school,  and a recent D.C. voucher experiment, both of which have been shown to improve educational attainment.  Consistent with the literature on the affects of charter attendance on test scores, they find that the impact in Florida is stronger in urban areas.

These are important results to keep in mind. Given education reform critics skepticism of standardized testing, I expect results like this will receive more weight for them than studies that look only at test scores.

How do you protect old teachers?

Under much debate these days is LIFO, or last-in-first-out. Its the policy used frequently by public schools to determine who gets laid off by simply going with whoever was hired the most recently. Critics of this policy point out it’s obvious shortcoming: that the youngest teachers are by no means the worst, and that if you could fire one older, ineffective, more expensive teacher, you could save two younger teachers’ jobs. Defenders of LIFO first bristle at the notion that there are plenty of old ineffective teachers sitting around who can be clearly and uncontroversially  identified and removed. They also point out that any alternatives to LIFO will allow administrators to cut costs by shedding older workers, and that young workers will be much less eager to go into a career where job security is poor and an inverse function of age. After all, who wants to know that the longer you work somewhere the more likely to be fired?

But the question to ask is why is it that administrators would want to fire older teachers? There’s two possible explanations: administrators don’t care about quality, or the cost their spending isn’t worth the quality. While the former may be true in some cases, whether it is or not, the latter is likely to be true almost all the time. As Michael Petrilli points out in Education Next, teacher pay scales continually rise with experience even though studies show that only the first few years of experience increases teacher effectiveness. So for example, even though a teacher with 25 years experience is no worse than a teacher with 10 years experience, he or she gets paid a lot more. Consider the following chart from economist Jacob Vigdor, which shows how average teacher pay grades compare to doctors and lawyers.

In these other jobs pay rises rapidly with experience, whereas teacher pay rises much more slowly and don’t peak until age 55, even though the benefits of experience all occur within the first five years. As Petrilli points out, if teacher pay rose in a commensurate fashion with quality enhancing experience, then it would increase quickly in the first five years, and then flatten out, which by making pay proportional to quality would remove all of the incentive to lay off older teachers. This is not to make the case for lower pay for teachers overall, in fact you could do this whole thing in a revenue neutral way.

So there are two big benefits to a flatter pay scale:  it doesn’t distort incentives in a way that creates an older then optimal teacher workforce, and it allows schools to replace LIFO with something better without worrying about effective older teachers being targeted. In addition, if we’re concerned that too many young teachers are leaving the field too quickly (and almost half do within the first 6 years), then a pay scale that increases more quickly is a way to encourage them to stay.

When schools are forced to adhere to a system where costs are so divergent from benefits, you can expect unintended consequences. What can result is a pile of inefficient policies, each trying to offset the unintended consequences of the next. That certainly seems to be the case here.

Education by the states

It is often said by education reform critics that if you look at U.S. education by state, some states actually excel in international comparisons. The following charts are from a paper by Hanushek, Peterson, and Woessman on how U.S. States perform compared to other nations by looking the percent of students that  scored advanced on 2006 PISA math exams. Massachusetts clearly outperforms all the other states, but is still outdone by 14 countries, 11 of which perform better by a statistically significant margin. The U.S. overall does much poorer:

I thought this chart served as a useful reminder that all states have room for improvement, and it’s not true that some of them can simply pat themselves on the back, say “great job!”, and turn their backs to education reform.

Centrally planned education

In the room for debate on “Why Blame the Teachers?” a high school teacher offers this:

“…college graduates are not going to be attracted to a profession that only encourages short stints. The majority of teachers did not choose their profession because of the vacation time, or the salary, or because they thought it would be easy. They chose teaching because they wanted to make a difference in children’s lives in the long term. Those teachers that entered the profession for the more mundane reasons don’t actually stay for very long.”

This very well may be true. This means that either teacher quality will suffer as a result of getting rid of LIFO or school will have to pay teachers more to compensate them for the now riskier profession. On the other hand, a little bit of this may be worth it. If this means on the rare occasion when layoffs happen they go to worst 5% of teachers (worst obviously measured with some error) then it may be worth the slight drop in overall teacher quality that results. After all, the net effect on teacher quality will be the decrease in quality from risk aversion but also the increase in quality by trying to get rid of the worst instead of just the newest teachers. But the fact of the matter is, whether LIFO or some other layoff policy is best may very well depend on local labor markets and other factors. Sure, if you have to pick one policy for all schools, LIFO is probably not the way to go. But the need to mandate policies like this at broad levels is one of the problem with the public education system: it does not allow flexibility and adaptation.

The real problem isn’t deciding which layoff policy to use. After all, the right one today may be the wrong in five years. And the right one in Town A may be the wrong one in Town B 30 miles away. The real problem is assuring that the people who run schools have the freedom to choose policies that best cater to their specific local labor market, and that they have the incentives to do so in a way that maximizes the educational value of taxpayers money. If schools are allowed freedom and accountability, then if a school gets rid of LIFO but the best employees value it, then the other school down the road can adopt it and lure talent away, and if the loss in talent causes the schools quality to suffer then they will be pressured to adopt it again.

It may also be the case that with the threat of change in the air a school system that promises stability will be able to lure the most talented teachers, and that a school that can promise tenure for life, LIFO, and no standardized testing will outperform all the rest in some places.

The problem, I think, comes because we can not provide local monopolies like school districts freedom without accountability. Yet without freedom, we must centrally plan. The education reform movement is hard at work finding good ways to measure success, and determine universal and objective rules for how people should be hired and fired, and figuring out what kinds of certifications should be required, etc. The other camp, (I don’t know what to call them but surely they don’t wish to be called anti-reform) scoffs at how difficult this proves and instead thinks we should have… um, smaller classrooms? Actually I’m not really sure what the alternative is other than to convince people of the limitations of education and talk them into applauding the system that we have.

While I applaud the work of the reformers and think it is a drastic improvement over the status quo, I am increasingly worried they may not be able to find the perfect set of rules that allows centrally planned education systems to succeed at the varied and difficult tasks we require of it, and that anti-reform critics may be, in part, correct. Unfortunately for anti-reformers, this does not mean we simply learn to love the status quo, but instead we need the freedom and experimentation that markets and choice provide. This doesn’t necessitate a fully privatized education system, and can include charters, public school choice, and vouchers. It’s also important to remember that, as Rick Hess emphasizes, choice by itself also does not guarantee success. But competition can allow reformers and anti-reformers to test their policy recommendations out and see what works, and it can allow space where the right rules for the right places can evolve.

I hope the reformers succeed in developing better measures and policies so that public education can thrive. I sincerely do. In the meantime, we should be continuing to push for more choice and competition, and work hard to understand the conditions under which that can be successful.

The market for teachers

The debate over public sector unions has been lacking in economic arguments, and the justifications argued on unions behalf have not been of the kind one usually finds in these debates among labor economists. The two main economic arguments for unions are to provide workers with voice, and to counteract monoposonies. My main concern here is the latter, but I’ll briefly address both with regard to teachers unions in particular.

With respect to the voice function of unions, a standard reply of labor economists is that the National Labor Reform Act should be changed to allow more non-union types of voice. For instance, the act prevents  firms from supporting any type of labor organization. That means a firm is unable to organize a non-union way for  workers to voice complaints, settle disputes, negotiate over working conditions. As Barry Hirsch argues, these parts of the NLRA, while not without legitimate purposes,  “restrict development of nonunion vehicles for employer-employee, cooperation and productivity-enhancing voice.” Canada, in contrast, does not bar employer-initiated or supported labor groups. And every progressive loves Canada, right?

The second economic argument for unions is to help counteract the power of monopsonies. In the public sector, this is certainly a plausible complaint. Given that governments are often the only ones providing some services, they may have large degrees of labor market power, also called oligopsony power. While few may be explicitly making this economic argument, the common concerns about teacher powerlessness in the face of fickle administrators certainly reflects this problem. In a well-functioning and competitive market, administrators would face repercussions for firing good teachers. For one thing, parents would be upset, and competitive enterprises suffer consequences when customers are unhappy. In addition, competitive labor markets would mean teachers would demand a risk premium to work at firms where they could be fired for political or illegitmate reasons. When there is no competitive check on a school, neither of these repercussions is as important.

Thus if we are concerned about the oligopsony power of schools, and we want teachers to be treated as other comparable professionals are treated, then one possible solution is to make schools more competitive by allowing more school choice, either through charters, vouchers, or simple choice among public schools. This way, if an administrator has a reputation for firing good teachers, there are consequences. This will not prevent unjust firings from happening ever, as no system could, but it will make it less likely. After all, we don’t worry too much that about good accountants getting fired, do we?

On the other hand, the lack of competition for government, and schools in particular, also provides the opportunity to earn so-called economic rents just as lack of competition does in the private sector. These rents may be shared in an organization among administrators and teachers, meaning that less competition means higher pay for teachers. So while more competition can mean more labor market power relative to administration, it may also theoretically mean lower pay.

In order to shed some empirical light on this issue, a recent paper (I can’t find an ungated version) by Lori Taylor at Texas A&M looked at teacher pay in Texas. Importantly, Texas is a right-to-work state so collective bargaining does not confound her analysis. What she found was that in a handful of very uncompetitive markets, teacher salaries actually would decrease as a result of more competition, meaning that they were sharing in some of the economic rents. In most markets however, oligopsony power meant that increasing competition among schools would actually increase teacher salaries. Here is how Taylor summarizes her results:

More than 88% of the teachers with less than 20 yr of experience would benefit from increased competition. Seventy-nine percent of the highly experienced teachers would also benefit. Only 2% of beginning teachers, 5% of experienced teachers, and 6% of highly experienced teachers could expect increased competition to lower their pay.Profits and Rent-Sharing.

To provide one example, she highlights the Houston, Dallas, and San Antonio areas, home to more than 70% of the charter schools in Texas. Prior to the existence of these charter schools, a 1% increase in competition would increase teacher salaries by 2.5%. Given these increases in salaries, it seems likely that non-wage power of teachers is increased as well, meaning it’s harder to fire a good teacher for bad reasons.

Despite the theoretical ambiguities discussed above, these really are common sense results: the more labor market opportunities you have, the less power your bosses have over you, and the more competition there will be to hire the best teachers. The only place this isn’t going to be the case is areas where teachers are earning rents and schools are very uncompetitive. In these areas teachers are already likely to be overpaid, and the other benefits from competition, like better student outcomes, are likely to be the greatest. Long story short: more competition helps the teachers who need it.

Educational Reform vs. Financial Reform

Arnold Kling asks why economists think they can improve educational outcomes by firing the worst teachers (a position he labels E) but none of us think that we could improve financial performance by firing the worst financial managers (a proposition he labels F).

He concludes

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.

On a basic level the difference , somewhat reflected in Arnolds third option, is that economists believe that we know up front who the best or worst teachers are going to be over the next five years. Its the teachers who were the best or worst over the last five years. That is, economists believe individual teacher quality explains a huge fraction of the variation in student outcomes.

However, economists don’t believe that we can pick the best or worst financial managers over the next five years, just by looking at the best or worst from the preceding five years. That is, economists tend to think luck is the primary determinant of financial returns.

Now, admittedly Warren Buffet is an embarrassing counterexample for the luck hypothesis. On the other hand Merton and Scholes really looked like the smartest guys in the room until Long Term Capital blew-up in everyone’s faces. Moreover, Buffet’s legendary status makes him the exception that confirms the rule.

More deeply, economists argue that the most sensible way to invest is to simply buy a weighted average of all investment vehicles based on your risk tolerance.

Yet, few of us believe that the most sensible education is simply to take a weighted average of all currently used curricula and methods.

Education is hard

Freddie has been complaining about the Atlantic and other media organization’s coverage of higher education, in particular he thinks they’re too quick to blame the institutions. In contrast, he presents this quote from Tyler Cowen:

“In contrast to earlier in the twentieth century, who today is the marginal student thrown into the college environment? It is someone who cannot write a clear English sentence, perhaps cannot read well, and cannot perform all the functions of basic arithmetic. About one third of the college students today will drop out, a marked rise since the 1960s, when the figure was only one in five. At the two hundred schools with the worst graduation rates, only 26 percent of the students will finish. The typical individual in these schools– much less the marginal individual– is someone who struggled in high school and never was properly prepared. It also may be the student who, whatever his or her underlying talent level may be, comes from a broken and possibly tragic home environment and simply is not ready to take advantage of college.

Educating many of these students is possible, it is desirable, and we should do more of it, but it is not like grabbing low-hanging fruit. It’s a long, tough slog with difficult obstacles along the way and highly uncertain returns.”

This, he rightly says, is “sober, it is measured, and it is self-limiting”, therefore he claims it would never be published by the Atlantic. I think Tyler’s point is an important one, and I think Freddie is correct to praise Tyler, but I’m not sure the Atlantic deserves the ire he directs at it. In fact one of the first stories that came to my mind when I read this blog post from Freddie is an excellent story from the Atlantic called “In the Basement of the Ivory Tower”. I didn’t go searching for a single story to disprove Freddie, it’s just that when I think about the plight of the marginal college student and his or her educator, I think of this story.

Here is one typical passage that conveys an honest, sober, appraisal that of the type I think Freddie is looking for:

The colleges and the students and I are bobbing up and down in a great wave of societal forces—social optimism on a large scale, the sense of college as both a universal right and a need, financial necessity on the part of the colleges and the students alike, the desire to maintain high academic standards while admitting marginal students—that have coalesced into a mini-tsunami of difficulty. No one has drawn up the flowchart and seen that, although more-widespread college admission is a bonanza for the colleges and nice for the students and makes the entire United States of America feel rather pleased with itself, there is one point of irreconcilable conflict in the system, and that is the moment when the adjunct instructor, who by the nature of his job teaches the worst students, must ink the F on that first writing assignment.

The professor’s portrait of the students is a sympathetic one, almost heartbreakingly, which in turn lets you empathize with the professor. The story even demonstrates self-awareness of the criticisms Freddie is making, seen here in the author’s struggle over whether to fail a middle aged woman who returned to school to better herself but was clearly not capable of college level work:

I thought briefly of passing Ms. L., of slipping her the old gentlewoman’s C-minus. But I couldn’t do it. It wouldn’t be fair to the other students. By passing Ms. L., I would be eroding the standards of the school for which I worked. Besides, I nurse a healthy ration of paranoia. What if she were a plant from The New York Times doing a story on the declining standards of the nation’s colleges? In my mind’s eye, the front page of a newspaper spun madly, as in old movies, coming to rest to reveal a damning headline:

THIS IS A C?

Illiterate Mess Garners ‘Average’ Grade

Adjunct Says Student ‘Needed’ to Pass, ‘Tried Hard’

This story acknowledges the difficulties higher education institutions face when it comes to marginal students, and it recognizes the friction between this challenge and the American ideal of college education, an ideal the author himself holds. Importantly, yet dishearteningly, he doesn’t provide any easy answers. I know Freddie is working on a wider critique of the Atlantic, but I think this piece fits pretty squarely with what he’s looking for.

The Value of Good Teachers, ctd.

Tom Levenson at Balloon Juice has a lengthy reply to my article on the value of good teachers that accuses me of neither understanding nor reading the paper at hand, and drawing incorrect conclusions from it. He’s wrong.

It’s ironic that Tom is accusing me of not reading a paper based on a blog post of mine that he has clearly not read, but has instead relied on the two paragraphs of it quoted by Conor Friedersdorf at The Daily Dish. Tom says I wouldn’t be as surprised by the results of the study if I were familiar with the literature:

That jolt would be a little less if you read—hell not the primary literature—but, say, just blog posts written by the New York Times’ best economics writer, David Leonhardt, who in the distant obscurity of … oh, this summer, reported on another study that looked at the impact of a good kindergarten teacher on future earnings and other social outcomes.

Anyone who is stunned by the specific result or the broader claim that educational outcomes have an impact on economic success simply hasn’t been paying attention…

Like Tom, I think it helps to put research in the context of existing literature, which is why I included this in my original post:

You can find similar results in the work of Raj Chetty, which suggests that good kindergarten teachers are worth $320,000.

If you follow that link you’ll see it takes you to that same David Leonhardt article, and had Tom actually read my post he would have seen this.

Unlike Tom, and despite having already read David Leonhardt, I do find Hanushek’s results “shocking”.  For one thing, Hanushek runs through a large range of plausible values for the parameters determining the value of better teachers and even with very large knowledge depreciation rates he still finds a value of $150,000 for a teacher in the 75th percentile with a class size of 20.

In addition, the net present value of over $100 trillion that would result from replacing the bottom replacing the worst 5-8% of teachers with average teachers would be, as Hanushek calls it, an “astounding” improvement in welfare. This is especially important given that, as Tom points out, we don’t yet have a good idea how to design performance pay in a way that ensures higher teacher quality. Hanushek argues that this difficulty raises the importance of the replace-the-worst-teachers policy:

The foregoing analysis has also implicitly suggested an alternative approach to simple performance pay that could be more cost effective.  If there is an accurate screen on teacher effectiveness, many of the properties of a performance pay scheme can be achieved by eliminating low performing teachers and paying the remaining teachers higher but relatively flat salaries.

So I’m sorry if Tom is not impressed or surprised by these results. Given the considerable media attention this paper has received, including bloggers at the New York Times, the Huffington Post, Education Next, the National Review, the Atlantic Wire, and the Daily Dish, I’m going to have to say that a lot of people disagree with Tom.

As an aside to all of this, I want to add that I don’t think those who prefer a system that consists mostly of public schools should be happy about the difficulty of finding good ways to structure performance pay and use value-added measures to identify effective teachers. The harder it is for social scientists to identify clear ways to improve the education system through bottom down reforms the more likely it becomes that more market based, bottom up reforms are both necessary and popularly demanded. I think this will mean things like vouchers, charters, and more creative reforms like this proposal Reihan Salam highlights from Rick Hess and Olivia Meeks:

Wisconsin would do well to start exploring a new model at the high school level. It ought to continue insisting that schools provide the 11 core credits, amounting to about 55% of the high school curriculum, but then rewrite the funding formula so that the per pupil allocation currently delivered to school districts is broken into two pieces: 55% to fund “core” mandated instruction and 45% deposited in a virtual Educational Spending Account (ESA) created for each child. Parents would have a choice. They could direct those ESA dollars to their child’s school and simply enroll their child in the usual manner, or they could use them to procure instruction from other state-approved providers.

If we could drastically improve the public school system simply through better pay structure and value-added measures then I think the public school system as it stands would require much less change than it’ll get if those relatively simple changes fail. I don’t think those who prefer a system that consists mostly of public schools will be happy with what we get when simpler reforms fail, because I believe the end result will be a more privatized system than would otherwise be necessary if we could centrally plan our way to a great education system.

For the record, I recognize that this is a very speculative claim, and there is a lot of room for reasonable disagreement here.

The Value of Good Teachers

From a new NBER working paper by Eric Hanushek comes this shocking abstract:

A teacher one standard deviation above the mean effectiveness annually generates marginal gains of over $400,000 in present value of student future earnings with a class size of 20 and proportionately higher with larger class sizes. Alternatively, replacing the bottom 5-8 percent of teachers with average teachers could move the U.S. near the top of international math and science rankings with a present value of $100 trillion.

There are really two important claims here. I think progressives tend to be very pleased with claims like the first one, which is that teachers have a very high value. You can find similar results in the work of Raj Chetty, which suggests that good kindergarten teachers are worth $320,000. If this is true then the marginal benefit of teaching skill -or quality, if you want to think of it that way- is far below the marginal cost, and therefore we should increase wages to draw more talented teachers.

However, the second claim is just as important and is suggested by, although not a necessary condition of, the first: if good teachers are very valuable, then bad teachers are very costly. This means we should be willing to pay more for good teachers, but it also increases the benefit of getting rid of bad teachers and ensuring we have a system that can do that.  After all, every dollar spent on a bad teacher has the high opportunity cost of good teachers.

Findings like this tell us that we should place even less relative value on teacher well-being for it’s own sake (which is separate from teacher well-being to the extent that it improves outcomes) when considering reforms. I think this is something that some progressives aren’t as happy to hear, especially with regard to using the teaching profession as a middle class jobs program.

Overall though these results reinforce one fact that progressives and conservatives should agree on: this is a really important issue.

The Incentive to Think Like An Economist

I – and I believe many of my fellow economics professors – seek not to get our introductory students to learn specific facts or techniques but to begin to “think like an economist.”

I want them to abandon wishful thinking, good vs. evil analogies, just-so-stories and general ad hocery, in favor of treating human behavior as if it stemmed from some (perhaps unknown or even unknowable) set of systematic principles. In particular we are big on the notion that people respond to incentives.

Students it turns out are people. While “thinking like an economist” is often a bitter pill, this semester it went down much easier – and I think I see why. No less than half of the final papers were written on either immigration in general or the DREAM act in particular. A typical paragraph

Immigrants not only join the circular flow of the economy as labor, but also as consumers. They spend money on goods and services which results in firms having more revenue. Higher revenues tend to increase firm production to meet the higher demand of consumers. This increased production most likely would result in more jobs as firms expand to meet the needs of consumers. On a very basic level, this explains why immigration would only serve to expand the economy.

Unlike our discussions of taxes, rent control and the minimum wage “thinking like an economist” gives them in edge in an argument they already want to make: that the US should be more welcoming to immigrants.

While it is encouraging to see the tools taken up so easily, it is a warning that we cannot be sure students or the public generally has abandoned ad hocery when the systematic explanation does in fact suit their immediate needs.

Reihan vs Ravitch

Reihan Salam has a criticism of education policy expert Diane Ravitch that I will nominate for Blog Post of the Quarter, at least. Reihan charges Ravitch of making crude simplifications and showing overall poor analysis. This is something I’ve written about before, and quite frankly I’ve been waiting for someone with more subject knowledge than me to write exactly what Reihan has written. While there are many points Ravitch makes and Reihan dissects, the central disagreement is this: Ravitch says that the real problem with education is poverty, Reihan says that while poverty is a real problem and it has an impact on educational outcomes, there are a lot of other things we can improve in our education system that will make real differences.

To take but one example, here is Reihan countering the claim that if public schools had the same classroom sizes as they do in the Harlem Children Zone they would do as well or better:

…on the class size point, note that Shanghai, the PISA outlier this year, finds that the average class size in Shanghai is 35. That is, students in Shanghai are achieving the best educational results in the world with a teacher-student ratio of 1:35, not the 1:7.5 that Ravitch cites as the source of the success of HCZ. One has to assume that the push for smaller class sizes has helped dilute the teacher talent pool in the United States. This doesn’t mean that larger class sizes are necessarily the right answer. But it does at least suggest that Ravitch’s analytical framework is decidedly imperfect.

That Ravitch’s “analytical framework is decidedly imperfect” is, I think, the key takeaway from Reihan’s piece.

I’d like to counter one common point that Reihan quotes Ravitch making:

Instead we’re creating a revolving door where we say if you’re no good, you’re out and let’s bring in Teach For America. They’ll send in 8,000 kids to stay for two years and then they’re gone. This is no way to build a profession.

A 2008 study by Morgaen Donaldson on Teach for America has some useful numbers on this subject. Contra Ravitch, 61% of TFA recruits are teachers for longer than the required 2 years, and 24% stay teachers for at least 6 years. While this may seem low, remember that 40-50% of all teachers leave the profession within the first 5-6 years. In addition, 15% of teachers in low income schools leave those schools annually.

These number show that, as Ravitch well agrees, the status quo for attrition in public schools is not so great. So to point to tenure choices of TFA teachers as unbecoming of the profession, when those teachers are actually less than twice as likely to leave the profession in the first 6 years relative to all public school teachers, is an exaggeration.

There is a lot more in Reihan’s piece and you should really read the whole thing if you care about education reform. I’d like to see a back and forth between Reihan and Dana Goldstein on Ravitch and her analysis of the education system.

Teach for American and Goldman

Dana Goldstein seems bothered about a new program where teachers from Teach for America work at Goldman Sachs over their summers. Her concern is that this will lure TFA teachers from the profession since a starting position at Goldman pays $60,000 plus bonuses, which is more than double what a starting teacher makes in a poor district. But will this harm retention?

A 2008 study by Morgaen Donaldson looked at “whether, when, and why” TFA teachers leave the profession. One of their findings was that science teachers that majored in science in college were more likely to leave the profession than science teachers who didn’t. Donaldson finds this to be consistent with previous literature showing that science majors are more likely to leave the profession than humanities majors due to the higher salaries they can get in science, technology, and engineering careers. This, the author argues, shows that teachers can be pulled out of a teaching career as well as pushed out.

On the other hand, the ability to intern at Goldman Sachs may increase the lure of enrolling in TFA. There may be potentially talented teachers that would enroll in TFA but choose work at an investment company instead who might now be tempted by this opportunity.

Then again, retention is considered more of a problem for TFA than supply of applicants. Indeed, a frequent criticism of the program is that too many TFA teachers leave the profession, which means it’s potential is extremely limited. But I think the reality here is frequently exaggerated. First, keep in mind that 40-50% of all teachers leave the profession within the first 5-6 years. In addition, 15% of teachers in low income schools leave those schools annually. So the status quo is not so great in this regard.

In comparison, 61% of TFA recruits are teachers for longer than the required 2 years, and 24% stay teachers for at least 6 years. So TFA teachers are less than twice as likely to leave the profession in the first 6 years relative to all public school teachers, which I think is much closer than most would believe. I suspect these rates would be even closer when compared to public school teachers in low-income areas rather than all public school teachers.

The program is only for 20 internships at Goldman right now, so it’s nothing to worry about in either case. But I would argue that the willingness to experiment and break from the traditional mold of teacher education is what made TFA, and they should be encouraged to continue trying different things.

An important assumption

I wanted to make one additional point about the CREDO study I mentioned in the previous post, but I’m putting it in this separate post because it is a more boring econometric question that I don’t want to bog down the other more obvious issues with.

There is an important assumption the study makes and people should ask themselves how believable it is. The assumption is this:

Take a group of, say, six students from the public school system who have the same grade level, gender, race, income, special ed and english language learner status, and previous achievement test scores. One of these students decides to go to a charter school, and the other five do not. It is assumed that the one who chose to leave would have performed the same as the other five students had he stayed in the public school.

The presumption here is that there is no systematic difference between these students that is not captured by these variables. This begs the question, if these students are the exact same, then why did one decide to go to a charter school and the other five didn’t? Obviously there are some unobserved (to this study) variables that explain the decision to attend the charter school. Is it believable that these unobserved variables are uncorrelated with the student’s performance in school?

To illustrate the assumption even more starkly, imagine that the child goes to a virtual charter school. Is it believable that the kid whose parents pull him out of school to have him go to school on the internet is not different than the comparison kids? The importance of unobserved variables here suggests to me that these results will be biased against charter schools.

I’m not saying the assumption the authors have made is necessarily false, but simply that people who tout this study should understand that this assumption underlies the results, and they should ask themselves whether in another context they would consider it believable. I would argue that had the results of the study shown that on average charter schools outperformed public schools, some charter school critics would be dismissing the study on the basis that charters are cream skimming on unobservable variables, e.g. the charters are accepting systematically smarter, better performing students. Then again, perhaps I am biased and would be less skeptical of the assumption in this case.

My best guess is that whether or not this assumption holds depends, and that in some schools or states it does and in others it doesn’t. In Ohio, where there are a large number of virtual schools, I’m guessing it doesn’t. In New York, where charters are generally high quality and whether or not a student gets into onel contains a large randomization component, the assumption probably does hold. In fact, CREDO was able to replicate the results of a study based on randomization using their matching technique, which suggests the assumption does hold in New York.

So take it as you will, but understand that this assumption is there, and it’s important.

Making Charter Schools Work

In the most recent National Affairs, Frederick Hess has a very important piece on education reform that everyone should read. His thesis is simple: school choice proponents have focused too much on simply getting more choice in education, and they should be emphasizing the importance of competition. This has led to exaggerated beliefs about what choice alone can accomplish. Here is how he puts it:

The biggest mistake pro-market school reformers have made can thus be put simply: They have mistaken choice for competition. The conviction that school choice constitutes, by itself, a market solution has too often led reformers to skip past the hard work necessary to take advantage of the opportunities that choice-based reform can provide. Choice is merely part of the market equation; equally crucial are the requirements that market conditions permit high-quality or cost-effective suppliers to flourish, that regulation not smother new entrants, and that rules not require inefficient practices or subsidize also-rans.

This is absolutely correct, and as he points out, nobody would argue that choice by itself would constitute sufficient or desirable deregulation in any other government monopolized industry. So following his advice, what can we say about when charter schools work and when they don’t?

Ironically, I think there are a lot of ignored lessons in the charter school study which has produced the number one charter school talking point of the day, “charter schools perform no better on average than public schools”. Read any newspaper article on charters and this fact is highly likely to be there. As I’ve argued before, there are a lot of important truths lying beneath the surface of this quote that I think most people who say it miss. In the study this quote comes from, in some states charter schools perform way better than public schools. Should not the primary lesson from this be that some states have it right, and those that don’t should learn from them? The authors of the study this fact comes from note this fact, pointing out that “results vary strongly by state and are shown to be influenced in significant ways by several characteristics of state charter school policies”

They found three lessons about policies that affect performance:

1. States that have limits on the number of charter schools that can operate performed worse than those that had no caps

2. States with multiple entities that can authorize charters did worse

3. States that allowed appeals to be made in the application or renewal of charter school authorization did better than those that did not

The impacts are significant as well. The study does not say which states do and do not have charter caps but the impact is large enough that removing the cap would make charters not underperform public schools in all but Texas, New Mexico (for math), and Ohio (for math). Again, it’s not clear which states have the cap so we can’t say removing it would actually accomplish this, but the impact of the effect is large enough that it could.

Another important policy implication of this study is that charter schools do outperform public schools nationally for poor students and english language learners. So anytime someone provides you with the quote that “charter schools perform no better on average than public schools”, the response should always be this quote from the study:

In our nationally  pooled  sample, two subgroups fare better in charters than in the traditional system: students in poverty and ELL students.

For a lot of other relevant information about the CREDO study I have a much longer more detailed post here.

School gardens and progressive values

There were many comments here and over at Kevin Drums’ blog in response to my previous post on school gardens and progressive values. I think much of the criticism reflects a misreading of (and in some cases clearly not reading whatsoever) what I wrote, which in turn probably reflects a lack of clarity on my part. So let me try and respond to some criticisms and clear up some confusion.

Much of the criticism stemmed from a belief that I was arguing something like the following:

Low income people don’t like gardening, don’t garden, and/or shouldn’t garden

This is not what I said. Gardening is obviously a hobby that is enjoyed by people of all income levels. My point here is that as a strategy for increasing vegetable consumption for low-income families or, for that matter, anyone who works a lot, home gardening has very little potential. Obviously, some blue collar workers do grow gardens in window boxes, and some live in single-family homes with yards where they can have larger gardens. But given the amount by which Americans are falling short of their daily recommended vegetable intake, window boxes and backyard gardens for families who have the free time, energy, and desire to maintain them are not going to get us very far.

The problem here isn’t just with gardening as a solution, as Alice Waters’ and her organization clearly sees them as just part of the solution, but that gardens represents a broader slow food philosophy that underlies the entire movement. This focus on slow food is where progressive values get in the way of practical solutions.

For instance, I’ve argued that it’s important to focus on ways of making vegetables cheap, easy, and delicious.  In contrast, supporters of the slow food movement, and some commenters,  seem to believe that low-income and working people have a lot of extra free time to spend on gardening, food preparation, and frequent trips to the store for fresh vegetables. Quite frankly I never expected to see so many people claim that low-income  people have a lot of free time on their hands; judging by the responses I got it would seem Americans are suffering from a glut of free time. I believe this presumption is unpractical and problematic.

Slow food is a luxury which many low-income and working people simply won’t be willing or able to make time for. While it’s okay for schools to teach kids to the ideas of slow food as a small part of a broader healthy schools program, a practical solution must also focus on fact, cheap, easy, and delicious vegetables.  The mission of Alice Waters’ Chez Panisse Foundation goes completely against this idea:

Our mass consumer culture has created an unprecedented crisis of diet-related disease among our nation’s youth… Not only are children eating unhealthy food, they are absorbing the values that go with it: the notions that food should be fast, cheap, and easy; that abundance is permanent; that it’s ok to waste.

For those that would defend the local/fresh/organic focus by arguing this it’s really just as cheap, fast, and easy as other vegetables, keep in mind that this organization thinks those qualities are negative, and to be avoided.

The other thing to note from that part of the mission statement is that it doesn’t represent universally shared values, but the progressive values of the slow food movement. Tying a healthy foods movement to progressive values like this will limit its success in parts of the country outside liberal urban areas. While it’s reasonable to show kids that food can be more enjoyable if you embrace slow food, pushing slow food as a more prerogative is not.

For instance, the idea that local and organic foods are great is not a fact or universal value, but a progressive value. Many parents disagree, and it’s completely reasonable to believe that eating local foods for the sake of local foods is wasteful and foolish, and that specialization and economies of scale mean that farms should be industrial and located wherever they can be grown most efficiently. Many parents won’t want to spend their tax dollars buying local, organically grown food at a premium. The majority of consumers have certainly expressed this preference.

I’m not arguing that schools shouldn’t necessarily serve any local, organic, or fresh vegetables. But rather that these things are useful only to the extent that they are an effective means to a desirable end. Do they make kids healthier, or cost less, or help them form lifelong preferences for vegetables?  To the extent they do, then they should be used.

For local and organic foods, I’m skeptical that they are useful means to desirable ends, and therefore skeptical that much if any money should be spent on it. To the extent that the goal of using organic is that it’s healthier, then I would argue that schools shouldn’t spend money on it, since it’s not any healthier. To the extent that the goal of using local is to support local farmers, then I would also argue that schools shouldn’t spend any money on it, since charity for farmers isn’t a desirable objective for schools.

The problem is that the mission of these organizations is to make local, fresh, and organic an ends in-and-of themselves. It doesn’t matter if buying 10% more organic foods won’t make the kids eat healthier; children must be taught that organic is good. It doesn’t matter if only serving students fresh vegetables means they won’t eat frozen vegetables; they must learn that only fresh, local vegetables are good.

If you don’t believe that pushing local, fresh, and organic are objectives of the organization then you should read their websites and statements. In their food procurement criteria list, Waters’ organization includes these requirement:

• Local. The average meal travels 1,500 miles before it gets to our plates. Find local farmers, ranchers, and dairies from which to buy directly
• Organic or sustainably produced. Buy from farms that take care of the land.

In a statement before Congress, the executive director of Chez Panisse foundation made the argument for local foods explicit:

Buying and eating locally is a very simple concept that could have a huge impact on the environment if big public systems like schools districts, cities, parks and hospitals and private businesses all began to do it.  Imagine the way that we could stimulate local economies and reduce food miles by simply choosing to eat what is in season and buying locally from sustainable farms?

It’s impossible to make the case that getting the schools to buy foods from local farmers or those that “take care of the land” is simply in students best interest and not mainly about promoting a particular set of values. Asking schools to spend their money to benefit local farmers is egregious, and certainly not a universally shared value.

It is also telling that one of their strategies to deal with the higher expensive of organic foods is not to purchase organic to the maximum extent useful, but the “maximum extent possible”.

It is clear that progressive values are the focus of these programs, and this is at the expense of practical lessons, like how to make frozen vegetables taste good. This is extremely unfortunate, because frozen, out of season vegetables from far away are as important and deserving a part of a nutritious diet as local, fresh vegetables. Yet Waters’ organization actively works to completely remove frozen vegetables from school lunches.

If you think healthy school lunches and school gardens are good, you should agree that these organizations pushing for them need to remove the emphasis on progressive values and focus more on practical solutions. Slow food may be useful part of a healthy schools program as a means to an end, but pushing those values for their own sake should not be the objective, and certainly should not come at the expense of more practical lessons.

How progressive ideology is holding back the healthy schools movement

If integrating a school garden into curriculum can help teach kids subject matter better and get them to eat healthier, then I’m all for it. Likewise, I think improving school lunches and making them healthier are something worth spending money on. People like TV chef Jaime Oliver and school garden maven Alice Waters who are working to push these issues into mainstream deserve praise. Unfortunately, it seems that these genuinely useful policies and programs are being bogged down with wasteful progressive ideas.

Case in point is this paragraph from a recent Atlantic piece on Alice Waters:

…Waters recruited chef Ann Cooper (a.k.a. the Renegade Lunch Lady) to revamp what was on the school lunch menus in Berkeley, which then reflected typical school-lunch fare. As Director of Nutrition Services, Cooper banned processed foods and started making everything from scratch. She sought local produce, dairy, and bread, and, as much as possible, organic foods, too.

The first problem here is teaching kids to spend any time or money on organic foods, or spending public funds on such things. This may be good for the earth, but as several  comprehensive literature reviews have shown, organic foods aren’t any healthier. Here’s liberal wonk and foodie Ezra Klein summing up the evidence:

The most recent data on this come from a massive literature review commissioned by Britain’s Food Safety Agency (their version of our FDA, essentially) and conducted by Britain’s London School of Hygiene and Tropical Medicine. They concluded that a “systematic review of literature over 50 years finds no evidence for superior nutritional content of organic produce.”

Local is useful so long as it means more fresh, as fresh foods deliver more nutrition than frozen. But local for the sake of local is the kind of thing you worry about when you’ve got time and money to spend on luxuries; it’s not an important value to instill in kids, and especially not poor kids.

Waters and her organization are touting a new study showing that school gardens get kids to eat more vegetables. This isn’t surprising, but how much does it impact their lives once they graduate? Are future blue collar workers really going to take the time to grow themselves vegetable gardens in window boxes outside their apartments? A lot of working people, like Megan McArdle and Matt Yglesias, frequently don’t have time for fresh vegetables. Like Matt, many people have to teach themselves late in life how to make quick delicious snacks out of frozen vegetables. This would be a much more valuable lesson for poor kids then how to select the freshest kale at your local organic farmers market, or even more ridiculously, how to grow your own.

From every description of these programs I’ve read they have an obsession with local, fresh, organic, and growing your own food. The obsession should be on quick, easy, delicious, and inexpensive. These sets of descriptors are damn near antonyms.

If you can get kids to eat and prefer frozen vegetables then you’ve got a sustainable improvement in diet and nutrition. If you get them to like fresh organic vegetables they’ve grown in the garden or bought at the farmers market, then you’ve temporarily instilled in them the tastes of upper middle class people with enough time and money on their hands for such luxuries.

If people like Alice Waters and Jaime Oliver want wider support for heathy schools movements they need to purge them of the wasteful upper-class liberal obsession over local, fresh, and organic foods, and instead focus them on practical and sustainable lessons like how to prepare frozen vegetables cheaply, quickly, and deliciously.

How do you increase support for more education spending? Don’t let people know how much we spend now

Matt Yglesias made the case recently that we should increase teachers’ salaries in order to raise teacher quality. He provides this graph showing teacher salary compared to per capita GDP in different countries:

Unfortunately for Matt, if he wants to increase teacher salaries the last thing he should be doing is telling people how much teachers make. A 2009 survey by Education Next and Harvard’s Program on Education Policy and Governance surveyed 3,000 people and split them into two groups. In the first group they found that 56% supported increasing teachers’ salaries and 46% supported increased education spending. The second group was first told the average teacher salary in their state and the average spending per pupil and then they asked them the same questions. Support for more teacher pay fell from 56% to 40% and support for more education spending fell from 46% to 38%. Apparently a majority of people favor higher teacher pay, but not when they know how much teachers currently make.

In reality the responses would probably be different if they were shown Matt’s graph rather than just told how much we spend. But the fact that people support higher education spending less when they know more doesn’t bode well for those who want to increase it.

Bryan Caplan: Socialist

I’m only kidding, but Bryan has responded to my challenge for him to provide a conceivable story where more vouchers and charter schools lead to an education system that is more “statist” than the status quo, and I’m afraid his reply raises more questions than it answers. Here is his story:

Once vouchers create a massive industry that is almost entirely dependent on vouchers, the industry incessantly propagandizes and lobbies for ever-larger subsidies?  Public schools, teachers’ unions, etc. already do this, of course.  But I’m worried that the private sector’s public and government relations would be slicker and more energetic.

This is not an implausible story. But it sounds like a ringing endorsement for pushing for more choice. If the only blowback that occurs comes after reform has been more successful than anyone would dream, then I would say Bryan’s precautions about pushing for choice are not important here. This is like telling a kid it’s not worth trying to be an astronaut when he grows because he’ll be bored on the moon. “What? So the only possible downside to trying occurs once I’m on the moon and have succeeded beyond my wildest dreams?”

Even if Bryan’s right it’s not too much blowback if you ask me. If you don’t educate your youth properly you end up paying for it later with higher crime rates, lower human capital, and expensive government efforts to fix these things. So I wouldn’t mind spending too much money on an effective school system rather than just the right amount of money on an ineffective one.

In the end Bryan recognizes his “blowback” scenario would be an improvement over the current system and summarizes his position, thusly:

Given a choice between choice and the status quo, I’d still probably choose choice.  But given a choice between choice and austerity, austerity’s the way to go.

I’m glad to see Bryan agreeing at least that more choice and free market reforms are good things, which is a healthy step back from his suspicion that “‘constructive’ free-market reforms like Social Security privatization, school choice, Medicare vouchers, etc. are largely a waste of libertarians’ political capital.” If austerity isn’t on the table and choice is much more likely to succeed, than it’s not a waste of political capital to push for something that Bryan agrees is better than the status quo and won’t result in any blowback until it has succeeded far beyond any reformer’s dreams.

An interesting implication is that if Bryan prefers austerity over choice then surely he must be willing to buy austerity at the cost of choice, which sounds more like Ezra Klein that Bryan Caplan and, yes, just a little bit like a socialist. Would Bryan be willing to accept a completely government run disbursement of food for the poor in exchange for a 10% decrease in food stamps? Shouldn’t he also support a public option for health care? Would he sacrifice all existing charter schools, which only account for 3% of students in the U.S., for a 3% decrease in public school spending? In what programs would he be willing to trade less choice for more austerity?

Overall I think Bryan needs to walk his claim back even further and state that sometimes austerity is better than choice, but sometimes choice is better than austerity.

The higher education system of the future

Via Denis Dutton, Camille Paglia offers a criticism of todays college education and appeal to a more job centered approach focused on trades:

Jobs, and the preparation of students for them, should be front and center in the thinking of educators. The idea that college is a contemplative realm of humanistic inquiry, removed from vulgar material needs, is nonsense. The humanities have been gutted by four decades of pretentious postmodernist theory and insular identity politics. They bear little relationship to the liberal arts of broad perspective and profound erudition that I was lucky enough to experience in college in the 1960s.

I have to say I am drawn into argument by her criticism of the humanities and their ills, e.g. “pretentious postmodernist theory and insular identity politics”, and any suggestion that begins with gutting this from the higher education system will catch my ear. However, I don’t know how much of a solution she has offered.

She wants us to “revalorize the trades”, and make sure that “every four-year college or university should forge a reciprocal relationship with regional trade schools”. I agree these are good things, but ultimately the problem lies with the incentives and constraints these institutions face, not with mission statements or relationships.

Until we know exactly why it is that universities aren’t already operating with an “obligation to think in practical terms about the destinies of their charges” we won’t know how to make them be better. Surely parents and students desires and freedom of choice should incentivize them to be effective already. I don’t think imploring them to be different nor a school’s acceptance of a new mission to do so will suffice. I believe the problem is deeper than administrators and professors knowing how to be a good university.

Value-Added: the best among bad measures

A big story today is a report out by the Economic Policy Institute that criticizes value-added scores for teacher performance evaluations. Kevin Carey at The Quick and the Ed puts this challenge to the authors of the paper regarding how much weight should be placed on these measures in teacher evaluations:

The Economic Policy Institute’s new brief, which details the many concerns with and limitations to current value-added measures, says that 50% is “unwise.” However, despite EPI’s litany of concerns with value-added, the authors, who include Diane Ravitch, Helen Ladd, and Linda Darling-Hammond, conclude that: “Used with caution, value-added modeling can add useful information to comprehensive analyses of student progress and can help support stronger inferences about the influences of teachers, schools, and programs on student growth.”

But if 50% is unwise, what is EPI’s number? The paper doesn’t specify and calls for experimentation among districts. Experimentation is good. But I’d also like to see EPI’s authors and other value-added critics put their best number on the table. I doubt they will, though, because for many, that number is very close to 0%. And defending that number would be much more difficult than pointing out the flaws in value-added.

Kevin also discusses the shortcomings of value-added in a broader perspective:

Value-added measures of teacher effectiveness are not all that great…. But, and this is an enormous caveat, everything else we currently use is worse. A teacher’s years of experience, their education credentials, their certification status, the prestige of their college or their college GPA, even in-class observations. None of these measures does as good of a job at predicting a student’s academic growth as a teacher’s value-added score. Yet, we continue to use these poor proxies for quality at the same we have such passionate fights about measures of actual performance.

Carey’s two points together highlight an important question: if the weakness of the connection between value-added scores and teacher effectiveness means we should place a low weight, say 10%, on those scores for teacher compensation, than what does that tell us about the weight that should be placed on seniority and credentials? The answer has to be much less than 10%, since the evidence suggests those are far worse measures. Test scores representing 10%, and everything else that is currently used to decide pay representing less than that would be a significant improvement over the status quo. That would just mean a pay scale that is about 80% flat.

You know what, maybe you should just be reading The Quick and the Ed instead of me.

Education as consumption

Over at Econlog, Bill Dickens is trying to convince Bryan Caplan that signaling does not explain the majority of the value of higher education. Two of his reasons why education is productive is that is has a value as a consumption good, and as consumption capital:

2. Education is a consumption good. This should be self explanatory. At the margin school may be work, but infra-marginally at least some (if not most) people actually enjoy the reading, the lectures, the homework, etc.

3. Education is not just investment in work capital, its also an investment in consumption capital and social capital. I feel much more  at home in the world due to the fact I understand certain cultural references… The  shared culture produced by the education experience expands our common  language with a lot of meaning, and that produces huge network  externalities. Knowing history does help me do my job, but it is much  more important that it allows me to make analogies that will be  understood by acquaintances.

As an explanation for why people value college, this has some appeal. As an explanation for why college has a social value, I think it’s a pretty weak defense. Grant for a moment that it is entirely factually correct, is there any reason why this should be subsidized?

For the first thing this is a terribly regressive subsidy, primarily benefitting people with above average ability and wealth. Second, if the goal is to increase “social capital” for consumption purposes this is probably the least efficient way to do it. The money would be better spent subsidizing high-minded TV shows that make audiences more literate and cultures, or providing grants for creating and broadcasting informative documentaries or books that are catered towards people who normally wouldn’t watch them or read them. You would almost certainly generate more consumption capital and welfare by providing free subscriptions to the New Yorker ($40) for 175 households than a year in college ($7,020) for one person, and it would cost the exact same.

I’m not defending the signaling theory, Bill Dickens’ theory, or any other theory of education as a matter of fact. But proponents of more education investment should not look to Dickens’ criticisms of the signaling theory education, because even if he is right education is still way oversubsidized.

Ideas the near future will scoff at

The L.A. Times investigation into standardized test scores is amazing, and you should definitely be reading everything they have. The Times deserves lots of praise, especially for it’s humane but honest treatment of the teachers, for whom this must be a stressful and, for many, shameful ordeal. Consider this teacher:

Even at Third Street Elementary in Hancock Park, one of the most well-regarded schools in the district, Karen Caruso stands out for her dedication and professional accomplishments.

A teacher since 1984, she was one of the first in the district to be certified by the National Board for Professional Teaching Standards. In her spare time, she attends professional development workshops and teaches future teachers at UCLA….

Third Street Principal Suzie Oh described Caruso as one of her most effective teachers.

But seven years of student test scores suggest otherwise.

In the Times analysis, Caruso, who teaches third grade, ranked among the bottom 10% of elementary school teachers in boosting students’ test scores.

This is clearly not a lazily tenured teacher, but someone who was trying in earnest and working hard at being good at their job. She was not being protected by the union and the school district’s decision to hide these scores, she was being done harm and is one of the victims here. Good teachers want this information to make them better teachers, which you can see in her response to the news of her poor scores:

Still, Caruso said the numbers were important and, like several other teachers interviewed, wondered why she hadn’t been shown such data before by anyone in the district.

“For better or worse,” she said, “testing and teacher effectiveness are going to be linked.… If my student test scores show I’m an ineffective teacher, I’d like to know what contributes to it. What do I need to do to bring my average up?”

When you read these stories about teachers across the hall from one another, with the same population of students from the same socioeconomic background, yet performing so vastly different from each other I find it hard not to reflect on a dying notion in educational reform: that the problem is that teachers need is more resources and smaller classrooms. In the not-so-distant future it will unanimously be understood that this was a completely wrongheaded idea, and we will wonder how it could ever have been believed. It will be like price controls in the 70s, or the idea that the Soviet Union would outperform the U.S. The not-so-distant future will wonder how anyone ever thought education could be reformed when performance and pay were so disconnected, and when nobody -not even the teachers themselves- knew who was succeeding and who was failing.

Smaller classrooms and more resources without better information and incentives would be like trying to save a sinking ship by filling up the gas tank.

Methodological Note: Teacher Salaries

Thankfully my professional responsibilities forbid me from publically weighing in on the issue of teacher salaries. So, in response to this Andrew Biggs post I’ll just make a few methodological notes. Andrew’s basic question is

I’ve touched on these questions many times elsewhere, so here let’s just run a simple example: of school teachers in Krugman’s home state of New Jersey. Are they driving Cadillacs—or, at least, could they afford to be?

To get at this he compares teachers SAT scores to the average SAT scores and the income from a two teacher household with the average household income in New Jersey and the average income of Cadillac buyers.

So, you don’t want to use SAT scores and then compare to average statewide income because SAT takers are not a representative sample. A fair number of people drop out of high school, especially in the cohorts of older teachers. Of those who stay typically only those planning to go to college take the SAT.

What you want are ASVAB scores.

Then you want an estimate of teacher household income. Obviously, a fair number of teachers will be singles as will a fair number of New Jersey residents and Cadillac buyers for that matter. Thus, you don’t want to just compare two teachers with the state average. As I am sure AEI is aware, household income would rise significantly if average adults per household were closer to two.

Additionally, its iffy to draw intuition from a tradeable good like Cadillacs. Places with higher costs of living in general are going to see consumption shifted towards tradables. It might not mean the people are on net richer if they have a Caddy but a 350 sq ft apartment.

On the other hand, you might want to focus on non-monetary benefits like job security. You also might think about how collective bargaining affects wages in New Jersey in general and whether or not that is optimal.

Lastly, the question is, are we getting anything for our money. If the marginal benefit of a good teacher is 300K then we are getting the deal of a lifetime. If its 30K then we are getting the shaft. Estimates from private schools might help here but if there are significant social costs to bad teachers that will be lost.

Technorati Tags: andrew biggs,sq ft apartment,average household income,asvab scores,teacher salaries,methodological notes,average sat scores,monetary benefits,jersey residents,target,professional responsibilities,job security,cadillacs,representative sample,state of new jersey,cohorts,average income,collective bargaining,aei,caddy

Words of Wisdom

Despite what Brad Delong says in the comments, I think that Adam’s interpretation of the difference between the writing styles of Matthew Yglesias and Paul Krugman is very accurate.

Hoisted from the comments over at Scott Sumner’s blog, we get this little bit of wisdom:

And the irony is that Yglesias’s approach actually prepares his readers for battle much better than Krugman’s. If I debate a Krugman reader, I have no problem picking apart their arguments. They’ve been told that conservatives are morons who lack any good arguments. They don’t know what they have missed. It’s like shooting apples in a barrel.

How true this is. Krugman regularly makes blunt points that — to argue them with success — need to be backed up with subtle modifications, asides, and caveats. Unfortunately (as I said in the comments), Krugman tends to leave these types of things out (at least in his columns, and it is understandable with the limitations of the format)…or, he buries them underneath a strong point that supports his claim.

So when Krugman’s less economically sophisticated NYT crowd reads, “We’re in a liquidity trap! The laws of economics don’t apply! Only fiscal stimulus can save us!”; most probably don’t end up going to his blog to read about how a monetary policy of inflation targeting is the first best solution…and they don’t head over to the PKArchive to read about how Krugman reformed his (Keynes’) model of liquidity traps to include rational expectations. But they certainly do know that Republicans (and conservatives in general) are 1. evil, 2. racist, and 3. stupid.

Not to say that there is anything wrong with what he does…I write like that too, sometimes. I don’t even wholly disagree with him about conservatives. However, it hands your audience just the bare bones of the argument that you’re making, and that weakens their position (even while strengthening yours with numbers). In contrast, Yglesias often goes out of his way to add meat to the bones of his argument (while he’s making it). It sets his audience up in a better position.

Would the world be a better place if everyone wrote with the thoughtfulness and respect of Tyler Cowen? I think so.

Update: Mark Thoma points out that I should have worded things more subtly. I often write hastily, so it is my fault…and he is correct, so here is an update to his criticisms on from my Facebook page:

1. Paul Krugman’s argument for fiscal and monetary policy in a liquidity trap (as I understand it) is that first, a monetary policy of inflation targeting is the optimal solution. Not only would this be the optimal solution, but it would increase the multiplier effects of fiscal policy, so we should use both to combat a deficit in aggregate demand. If the commitment to these policies is credible, then with expectations of an increase in the future price level, inflation targeting will cause the current AD curve to shift to the right. Since SRAS is fairly flat in recessions, a subtle increase in the price level will produce a large jump in real output, and a negligible rise in inflation. Taken together with fiscal policy, which can be much more finely-targeted (I note that Krugman uses the “opportunity costs of borrowing at low rates argument”), not only can we boost output, but we can also achieve various social investments at a bargain, as well. However, since it seems politically impossible to go after inflation (and even I blame conservatives for this), then our (only?) next-best option when in a liquidity trap is fiscal policy.*

Now why, in Paul’s model, do the normal rules of economics not apply? I think it’s the residual effect of an intense focus on the interest rate as monetary transmission mechanism. Indeed, the standard NK model uses no money at all — just movements in *the* interest rate. You can, of course, assume money in the parameters, but it makes little difference in the world of bonds or consumption goods. Indeed, the NK models can predict wild things happening at the ZLB. Of course, in the papers** Paul cites regarding these bizzare phenomenon; there are all kinds of the “subtle modifications, asides, and caveats” that I alluded to earlier.

Here is what I believe happens to the central bank reaction function

2. The “74-year old theory” comment has been retracted. Krugman does, indeed, use modern NK models to justify his policy stances. And he does, indeed, build models on his blog. I explicitly didn’t compare his blog to his columns. I realize that the two are different vehicles, as said as much.

I take issue with Krugman’s delivery. In offering an opinion, you can say something to the effect of, “There are quite a few models out there, some are garbage, some are well worth exploring…but I personally believe that Keynes got it mostly right in saying…” This is not what Krugman does. I’m pointing out that I think it is a better way to make a point.

3. Mark offers a warning to Yglesias readers:

Finally, a note of caution for Ygleisias readers. He hasn’t been consistent in his policy recommendations and at times, without knowing it, has taken contradictory positions. I don’t think is very helpful to readers even if it is couched in language you happen to approve of (the problem is that he doesn’t seem to fully understand the modern monetary transmission mechanisms).

---

*This paragraph written in my own language, not necessarily Krugman’s.
**This, for example.

Obscurity

Bryan Caplan likes “trembling hand perfect” equilibriums, but identifies the concept as an obscure academic concept…and indeed it is. Then he offers a challenge:

My challenge for other econ bloggers: Name yours. What’s your favorite obscure-but-genuinely-enlightening academic economic concept?

I have written about mine fairly extensively.

The G-R Conditions of Wealth-Creating Economic Activity

1. Irreversibility: All economic transactions that create value are thermodynamically irreversible.
2. Entropy: All transactions that create value reduce entropy locally within the economy and increase entropy globally.
3. Fitness: All transactions that create value produce goods/services that are fit for human purposes.

The Smart Money

Why can’t “the market” get education right?

Adam has a series of posts on the market for education. His latest details a paper by Jim Heckman showing that the availability of GEDs might actual make students worse off by leading them to substitute towards a seemingly worthless product.

In addition, there is no shortage of hand wringing over what appears to be the growing short comings of the for-profit education sector.

We have excellent private schools at both K12 and higher education levels but overwhelmingly they are not for profit.

To make matters worse it seems that the “best schools” do everything wrong. Take Harvard. One gets a job as a Harvard educator not by being talented at or even particularly interested in education. Furthermore, Harvard turns down the overwhelming majority of potential customers and actively discourages millions of others from even attempting to buy its product.

Yet, surprisingly it doesn’t attempt to use this market constriction to wring the highest possible prices from its students (yes Harvard could charge much much more and still fill every class). Instead, it goes out of its way to attempt to make “price not a concern.”

What way is this to run a business? Yet, it seems to be a crackerjack way of running an educational institution. What gives?

Some economists will tell you that its all farce. Harvard doesn’t actually educate you see. It selects. By only allowing in the certain students it is stacking the deck in favor of amazing outcomes. How can your graduates fail when they came in with such stellar test scores?

But there are at least two problems.

1) Economists who study education consistently find that institutions like Harvard do seem to be doing something useful. In instances where random luck determines whether or not a particular student goes on to University, going on seems to make a big difference. Perhaps, this is success by osmosis but it seems difficult to reject out of hand the possibility that the students might just be learning something

2) If the students aren’t learning anything then this poses no less of a challenge to market efficiency. Maybe university is a signal that you can hack four years of tough work. But, then why isn’t hacking four years of tough work, as say a lowly paid or even volunteer probationary worker an even stronger signal.

Maybe Harvard is really just a master selector. Picking out the best and brightest. If a student whose ultimate entrance is determined by luck still gets the resume glow of having been picked by Harvard and hence the subsequent success.  But, then is all of the pomp, the vast research apparatus and the still high, if not as high as possible, tuition bill really necessary. I am sure that some enterprising chaps could provide an excellent search and selection tool for far less than that. Call it Google Worker.

 

While I’ve heard dozens of theories to explain what the heck is going on in the educational universe, none of them seem really satisfying. Education seem to be important, perhaps crucial to the function of capitalist institutions. Yet, capitalism, that is for profit enterprise, just can’t seem to get education right.

The teacher bailout needs sticks, not just carrots

The jobs bill in congress includes a $23 billion dollar provision that’s being called “the teacher bailout”, to prevent school districts from laying off 100,000 to 300,000 teachers. There’s certainly something to be said for trimming the fat, and I’m sure every school district in the country spends hundreds of thousands of dollars on stuff that could, and should be cut from the budget. Nevertheless, Derek Thompson at the Atlantic Business points out that the good kinds of cuts are not necessarily the kinds of cuts we would get:

Smart education reform includes clear incentives for administrators to control costs and teachers to demonstrate achievement against a reasonable baseline. But we don’t want schools firing teachers willy nilly in the fog of deep budget cuts that could wipe out programs based on their cost rather than their effectiveness.

This certainly seems to be what you see with school districts cutting budgets during this recession. You don’t hear about school districts whose budget cuts involve finally firing some of the old, ineffective, teachers who everyone knows shouldn’t be teaching, or taking back some of the overly generous salaries commanded by 8th grade history teachers because they have PhDs they don’t need. No, what you see is the most recently hired teachers being fired, and whole departments like music and art being cut. And many State laws ensure that future cuts are not going to get any better:

Fifteen states, including New York and California, now operate under union-backed state laws mandating that seniority, or “last in/first out,” determines layoffs. These quality-blind layoffs could force a new generation of teachers, like those recruited by Teach for America, out of classrooms in the coming months

I understand the desire to force school district towards efficiency, but it does not seem like the most inefficient costs are being cut as a result of budget shortfalls. However, that doesn’t mean that this is not an opportunity for reform. One way to get more reform and efficiency out of this is to go ahead with a teacher bailout, but model it after Obama’s successful Race to the Top program. States could compete for a piece of the bailout by getting rid of the first come/first fire laws, making it easier to get rid of bad teachers, and implementing other reforms. Given what RTT was able to achieve with $4.3 billion, $23 billion would likely buy some pretty significant reforms.

Legislators would be wise to consider what Obama said in his State of the Union: “Instead of funding the status quo, we only invest in reform”.

Foreclosing on your college degree, ctd.

Over at The Economist, Ryan Avent has a thoughtful rejoinder to my criticisms of his post on the value of a college education. I don’t think Ryan and I are too far apart in terms of both assessing the the facts of the situation and what to do about it, but rather wish to emphasize different things. I want to emphasize that on the margin a lot of people are probably going to college who would be better served doing otherwise, and much of our money and theirs is being wasted. Ryan wants to emphasize that in the long-run, both individuals and society in general benefit when the population is more educated, and so we should work to get people more educated. I think there is truth to both of these and they are not necessarily contradictory ideas. However, I do want to elaborate on some points of disagreement.

Ryan disagrees with my analogy that housing and education are alike in that they are large, leveraged, illiquid, and speculative investments. His first reason is cost: median housing cost is $218,000, whereas the average education cost is $50,000. But here he misses opportunity costs. If you go to school you not only have to pay tuition, but you forgo the earnings and experience you would have got by working instead. For four years of school and an income of -let’s be conservative- $15,000 a year, this means forgone wages of $60,000. That puts the cost at $110,000. In addition, while the full opportunity cost of not buying a house is a complicated question, some type of rental or housing cost is going be incurred no matter what, so the $218,000 is certainly too high. If rental costs would have been 50% of housing costs (good luck!) then you’re at the same cost for housing and education.

He also argues that the benefits of education accrue over a lifetime, whereas the benefits of housing only accrue while you live in the house. I think this actually favors housing over education as a more liquid investment. The benefit of housing services can accrue to the owner over the life of the building, or the owner can rent that flow of services, or can sell the house for the net present value of the future housing services, in essence taking all the benefits today. In contrast, the benefits to education can only accrue over the whole of an individual’s life. There’s no way to sell the net present value of the benefits of the investment, it is what you would call in irreversible investment.

Lastly on this point, Ryan concludes that “an investment in higher education is unambiguously more conducive to mobility than an investment in a home”. There are multiple dimensions upon which one can be immobile. Having a highly specialized PhD in one method of repairing a particular kind of windmill only found in Holland restricts you occupationally and geographically. Having a highly specialized PhD in repairing a particular kind of windmill found all over the world restricts you occupationally. Owning a property with negative equity in Holland restricts geographically. None of these restrictions are insurmountable of course, but only at a significant cost. If you’re willing to lose the investment value of your education, then a college degree doesn’t restrict you to any industry. If you’re willing to lose the moving costs and a potential equity loss from a poorly timed market exit, than a housing investment doesn’t restrict you to a geographic location. Both types of investments can put you in a large debt hole, and both can cash constrain you. Both are risky and speculative. And both have their own type of illiquidity problem. And it seems to me that there are many people in this country every year who make bad investments in both.

Some things we probably agree on: I believe Goldin and Katz, who Avent cites, that education is one key to slowing or ameliorating inequality. I also believe that having an educated population is important to long-run growth. But importantly, I also believe the work of Erik Hanushek that shows that it is not years of education that matters, but quality of education. The following chart from a recent paper of his shows that the relationship between standardized test scores and economic growth is much stronger than the relationship between years of schooling and growth.

To quote Hanushek, “A country benefits from asking its students to remain in school for a longer period of time only if the students are learning something  as a consequence.”

I suspect Ryan would agree with this. Where he and I might differ is that I think we need to first make sure that marginal college students are getting their dollars worth and increasing their human capital, and then discuss how to send more people to college. If individuals are not getting their dollars worth on the investment, then that to me is a good sign it is not the best human capital investment we could make.  I am not an expert on this, but I’m not convinced that the twin studies, clever instrumental variables, and other complex econometric techniques that attempt to tease out returns to education while controlling for other investments and innate ability persuasively tell us that expanding college enrollment today will generate positive returns for the individuals or society.

If greater subsidies that expanded college enrollment just swells the ranks of the University of Phoenix does that really pass cost benefit? Is there a marginal dollar better spent on this than on improving high school education? And are these schools actually improving human capital, or are they just a signal of ability that employers will pay a premium for? What matters for economic growth and for individuals is actually increasing human capital. I don’t think expanding the ranks of college students without first reforming the bottom rung of college education is likely to be the best way to do it, in fact I think some of the time and money many individuals and our government currently spend on college could better increase human capital and overall well-being by being put to use elsewhere. Postal workers with college degrees come to mind.

Don’t go to college

Some facts:

Perhaps no more than half of those who began a four-year bachelor’s degree program in the fall of 2006 will get that degree within six years…

For college students who ranked among the bottom quarter of their high school classes, the numbers are even more stark: 80 percent will probably never get a bachelor’s degree or even a two-year associate’s degree.

Of the 30 jobs projected to grow at the fastest rate over the next decade in the United States, only seven typically require a bachelor’s degree…

And one thought provoking quote:

Professor Vedder likes to ask why 15 percent of mail carriers have bachelor’s degrees, according to a 1999 federal study. “Some of them could have bought a house for what they spent on their education,” he said.

There is much more in this article on increasing calls for alternatives to four-year colleges. This highlights the puzzle of why more alternative means of accreditation and training haven’t arisen. What’s the market failure? There are some obvious possibilities including informational problems. But the large degree of government intervention, especially in the low-end of the college market,  does suggest that part of the explanation may lie with policy and not market failure. After all, the large majority of revenues for poorly performing low-end schools like the University of Phoenix come from federal student loans.

In addition, these schools are by far the top recipients of Pell grants, with the University of Phoenix receiving $656.9 million in the 2008-2009 school year alone.

With this level of intervention, policy has a large degree of responsibility for the state of the market. And that state does not appear healthy.

What’s wrong with Texas and Ohio?

There was a longish piece in the New York Times yesterday about the quality of charter schools across the country. The crux of the article was that charters vary greatly in policy, and running a successful charter is harder than it looks. Matt Yglesias’ takeaway is this:

…educating poor children is a difficult task. What some people don’t get, however, is that while demographics matters a great deal so does school quality. You can see this in traditional public schools where poor kids in New York and Boston do much better than poor kids in Washington, DC. And you also see it in charter schools, where a minority are excellent but most are not excellent.

Dana Goldstein comments as well.

When people say that charters do worse nationally on average than regular schools they are talking about the CREDO study. My question is -and maybe someday I’ll have time to dig in answer it myself- what do the national results say when you exclude Texas and Ohio? I’ve written before about how you would expect Texas and Ohio to perform poorly, since the former’s original charter law encouraged the establishment of charters for disadvantaged children , and the latter has a large number of “Virtual Schools” that are essentially online schools. Thus I would argue it is unfair to pool these states in with traditional charters.

These states, along with Arizona, Florida, Minnesota, New Mexico, are two of six where charters perform worse than public schools. Given that they likely make up a decent chunk of the underperforming charters, I think it’s a good guess that the national numbers would improve greatly if they were left out.

Another thing worth noting is that while charter school critics are quick to cite the results of the CREDO study that reflect poorly on charters, they are not so quick to support the study’s findings that a cap on the number of charter schools appears to decrease student performance. If you believe their empirical model is good, then shouldn’t you be calling for states to remove this cap?

All this said, the article does an overall admirable job of attempting to grapple with the issues and is worth reading, but read Goldstein and Yglesias as well.

Maybe some education reform is too radical

With education reform I tend to be a fan of what Chester Finn, the editor of Education Next, calls “blowing up the system”. That said, I think Gail Collins criticism of the reforms that Florida tried to pass are worth considering:

Can I digress, people, and say that while it’s important to make teachers accountable, telling them their jobs could hinge on their students’ grades on one test is a terrible idea? The women and men who go into teaching tend, as a group, to be both extremely dedicated and extremely risk-averse. The stability of their profession is a very important part of its draw. You do not want to make this an anything-can-happen occupation, unless you are prepared to compensate them like hedge fund traders.

It may be that we do want to pay teachers like hedge fund traders, but there are 3.5 million teachers in the country right now, and whatever reforms we put into place must consider not only the stock of teachers you want, but the stock you currently have. Given that the job selects for risk aversion, injecting too much employment risk into the job, especially too fast, is probably going to do more harm than good.

A Rocky Start for the School of the Future

Eductationnext has an update on Microsoft’s School of the Future located in Philadelphia. The school opened a few years ago with a untraditional image of what a school should be. It would involve incorporating technology into education, including giving students laptops, project based learning, teachers working in teams, and the “course of study would be dynamic, interdisciplinary, and driven by their interests”. The school was designed and launched by Microsoft, but is a collaborative effort with, and operates within, the School District of Philadelphia. This means that the school was not paid for by Microsoft, so that if it was successful it would be scaleable by being able to operate within school districts budgets. However, there seems to be a problem executing the vision of the SOF within the restrictions of the districts school system:

For starters, the school district’s computer couldn’t accept SOF’s narrative-style report cards, which evaluated students’ proficiency in the core competencies rather than giving them traditional numeric grades in individual subjects…. By the middle of year three, the district had pressured the school to begin using its core curriculum and, like other neighborhood high schools, administer biweekly benchmark tests based on it. Two periods a week were set aside for mini-projects.

Even without the problems of working within the district, the school’s brand new model of education is creating difficulty for their traditional model of students:

The students, almost all African American, more than 80 percent of whom qualify for free or reduced-price lunch, came with skill levels all over the map; a majority read at a 5th-grade level or below. Used to worksheets, paper-and-pencil tests, and being asked to regurgitate information, many weren’t prepared to take control of their own learning. Some thrived on the project-based, interdisciplinary, and technology-rich model, and were finally able to connect to the purpose of school; others simply found it bewildering.

And as others have found, the technology-heavy model of education has problems of it’s own:

“I would spend 30 or 60 minutes of a period deleting games from the computer,” lamented one teacher. Students would be instant messaging and checking emails during class. “When you’re exhausted because you’ve been telling kids to stop playing Halo all day, you’re not actually teaching them literature or skills or the content that they need to drive their own learning.”

So is the project doomed? It doesn’t sound like it. The students perform no better on standardized tests, but appear to be getting a more enriching educational experience that is tailored to their interests. In a country with a 30% dropout rate, just getting kids to like being in school more is a good thing if it can lower that number. Also, one thing that is also clear is that the school is evolving, adapting, and learning to fulfill it’s mission within the constraints of the school system. It may yet hit it’s stride. However, this school does provide another lesson in the difficulty of implementing radical reforms within the existing system.

Children are unfortunately more complicated than simple economic models predict

I recall reading somewhere that in this culture we praise kids for getting good grades instead of praising them for working hard, when we should be doing the opposite. The simple economic response to this is that grades are more observable than effort, and so you should reward grades otherwise you may just get a lot of ineffective or half-hearted effort. Imagine, for example, paying a business for how much work they put into making something instead of how much stuff they actually made; you’d obviously get less output. Roland Fryer has a new NBER paper with evidence that the simple economic view is wrong when it comes to children and education:

In stark contrast to simple economic models, our results suggest that student incentives increase achievement when the rewards are given for inputs to the educational production function, but incentives tied to output are not effective.

His explanation for this phenomenon sounds amusingly economist-y, but makes sense:

Qualitative data suggest that incentives for inputs may be more effective because students do not know the educational production function, and thus have little clue how to turn their excitement about rewards into achievement.

I can imagine this problem is discussed by many parents and academically struggling children across the country.

Dad: I just don’t understand the problem, Jimmy. You seem to want to get good grades, but you keep failing.

Jimmy: Dad, mom… I don’t know my production function! I’m sorry!

Mom: Dammit Greg, I told he didn’t know his production function!  You know Jimmy, you’re uncle Arthur didn’t know his production function when he was you’re age either, and he turned out just fine.

The two takeaways from this evidence are that 1) when designed correctly, financial incentives for students do work, and 2) empirical evidence matters for education reforms designed to impact children’s behavior.

Simpson’s Paradox and the NAEP

Chad Aldeman at the excellent Quick and the Ed explains how improvements in the NAEP reading scores are masked by looking at overall scores because of Simpson’s Paradox. To put it simply, this means that the you can see progress when looking at group-by-group numbers, but then see no progress overall due to relative group sample sizes. In the context of the NAEP, what this means is that overall scores show no progress because of changing demographics, and that looking within demographic groups progress is obvious.

Here are the overall long-term NAEP scores which show very little progress:

And here are the same scores broken down by race/ethnicity, which paint quite a different picture:

Chad sums up these graphs:

Each group has actually made greater gains over time than the overall total. White students increase 11 points, one more than the national average. Black students scored 23 points higher, and Hispanic students were scoring 24 points higher in 2008 than they were in 1975 despite quadrupling in size. In other words, the white-black and white-Hispanic gaps are closing and every group is scoring higher, but the national score is showing more modest improvements because of demographic changes.

How Do Charter Schools Perform Nationally?

Diane Ravitch has become the education reform movement’s Benedict Arnold. After turning against the accountability and school choice reform movement she once staunchly promoted,  she has been hitting the op-ed pages and is now taking part in a debate at TNR. Unlike some critics, both reform and anti-reform, I won’t begrudge her changing her mind. I think it took a lot of guts to do a public about-face as she has, and I think her willingness to do so lends to her credibility rather than detracting from it.

That said, I want to draw attention to one piece of evidence she that she keeps bringing up, which is that test scores are lower at charter schools on a national level:

The only major national evaluation of charter schools was carried out by Stanford economist Margaret Raymond… Her group found that compared to regular public schools, 17% of charters got higher test scores, 46% had gains that were no different than their public counterparts, and 37% were significantly worse.

I’ve said before that this surprises me, and is in contrast with much of the evidence I’ve seen, so I decided to dig up the study that she refers to and take a closer look.

The first thing to note is that the negative results are a based on a pooled national average, which I’m not sure is all that informative. As the authors and Ravitch both say, the performance of charters relative to traditional public schools varies widely in quality by state:

In five states ‐ Arkansas, Colorado (Denver), Illinois (Chicago), Louisiana, and Missouri ‐ charter school students experienced significantly larger growth… In six states — Arizona, Florida, Minnesota, New Mexico, Ohio, and Texas — charter school students experienced lower learning gains… In four states — California, District of Columbia, Georgia, and North Carolina — the results were mixed or no different…

Some of these low performing states probably have good policy-based reasons that they are low performing. In fact, the study found that “results vary strongly by state and are shown to be influenced in significant ways by several characteristics of state charter school policies”. For example, they found that a cap on the number of charter schools has a negative impact on student performance.  The lessons I would take from this are that policies matter for charter quality, the poor performing states should learn from the good, and this may mean removing the cap on the number of charter schools. The only thing this evidence cuts against is Ravitch’s straw-man argument that charters are a magic bullet that will fix everything no matter what.

Given the high variability in state performance, it’s important to take a more detailed look at state policies. Here are how Ohio and Texas did in the Stanford study:

Pretty terrible, and among the worst in the study. But a recent RAND study had this to say about Ohio’s negative charter school performance:

The dramatically lower estimated performance of Ohio’s K-entry charter schools appears to be attributable not to grade level per se but to virtual charter schools that use technology to deliver education to students in their homes. Virtual schools constitute a large part of the enrollment of K-entry charter schools in Ohio, and students have significantly and substantially lower achievement gains while attending virtual charter schools than they experience in TPSs.

Allowing kids to be home-schooled at “virtual schools” on the internet decreased their academic performance? Gee, that’s a big surprise. I don’t know if schools like this were included in the study Ravitch cites, but this should be a firm reminder of the problem of looking at charter schools as a class without controlling for schools like this.

Another problem state is Texas, about which the RAND study had to this say:

Relative to local averages, prior achievement levels of charter entrants were particularly low in Texas, which could be attributable (at least in part) to the success of the provision in the state’s original charter law encouraging the establishment of charter schools for disadvantaged students.

These pre-existing differences in students again should caution against high level studies, and studies that do not carefully control for selection bias.

I’d be willing to guess that if this national study had shown that charter schools outperformed public schools that reform critics would be protesting that the charters are probably just cream skimming, and the results can’t be trusted. And they’d have a point. Controlling for unobserved differences is really tricky. I haven’t looked closely at how this study does it, but no matter what you do you’re going to have a concern about this problem. This exactly why randomized studies are so important, and why we can have much higher confidence in their results.

One very good recent example is a study based on randomized school assignment in New York City that found drastically positive results for charter students:

On average, a student who attended a charter school for all of grades kindergarten through eight would close about 86 percent of the “Scarsdale-Harlem achievement gap” in math and 66 percent of the achievement gap in English… a lotteried-out student who stayed in the traditional public schools for all of grades kindergarten through eight would stay on grade level but would not close the “Scarsdale-Harlem achievement gap” by much.

Replicating the results of NYC and preventing outcomes like Ohio should be what people who care about students should be talking about and focusing on, not talking points about national averages that don’t address most of the important questions.

I’m not saying the study Ravitch cites tells us nothing. I just think if this the compelling study she’s going to use as a talking point and continue to cite to as an important result then I’m not optimistic about her ability to draw relevent conclusions from empirical evidence.