Network News

X My Profile
View More Activity

'It’s just a question of using the same dollars wisely'

By Dylan Matthews

James Heckman is the Henry Schultz Distinguished Service Professor of Economics at the University of Chicago, and shared the 2000 Nobel Memorial Prize in Economics. Much of his recent work has focused on the economic gains to funding early childhood education. In particular, he has studied the Perry Project, which provided preschool in Ypsilanti, Mich., for a group of 3-year-old children chosen at random from 1962 to 1967. A lightly edited transcript of our conversation, on Perry and related topics, follows.

One of the most interesting things about your research, particularly the Perry preschool intervention and your work on that, is that care that’s very expensive in the short run -- the numbers I saw was that Perry was $10,000 a year in current dollars -- still yield enough benefits in the long run to overwhelm the costs. Is there some correlation, in that if you’re paying more you’re likely to get higher-quality care, which will yield more dividends?

Be careful. You say it’s expensive. First of all, it’s more like $7,000 to $8,000 a year, in real terms. Maybe $8,500. And if you compare with the amount of money spent in the budget, per capita, for secondary school education in public schools, it gets pretty close to that. The national average is pretty close to that. So it’s not extraordinarily expensive.

If you compare it to something like Head Start, it is expensive, but the structure is very different. There are other programs that are much more expensive, but that’s not one of them, relatively speaking. I would put it in context, because we’re spending a lot more on that.

The really important thing to understand is that you want to look at what the rate of return is. The rate of return is somewhere between 6 to 10 percent per annum. So no matter what the costs are, it’s going to earn a rate of return that says for each dollar invested, I give six cents back for the life of the child.

So you put in those $8,000, you’re going to get multipliers of more than 100 at age 65 in real dollar terms. I once did a calculation, it’s whatever 1.06 is to the 60th power. It’s a huge number, I promise you. I think it’s very shortsighted to talk only about the costs. You want to look at the returns.

Is there anything that differentiates programs based on rate of return? Are there certain types of programs where you see greater percent per annum numbers and certain ones where you see less?

I was looking at Perry. But I do believe the structure is similar for other programs. We’re in the middle of evaluating those other programs. For example, I haven’t yet done a study of the Nurse-Family Partnership Program.

Those people are a little bit too young to really make a strong case one way or other, but we are in the middle of conducting an evaluation of the so-called ABC project that was conducted in North Carolina some 30 years ago. There are a lot of studies that are underway.

If you were to ask what’s the best possible structure, I can’t tell you for sure. What I can say is there are a lot of different structures out there, and that some of them seem very promising. Some have long-term evaluations, like Perry. ABC will have one shortly. So it’s to be determined. It’s a subject very much under development.

One the main takeaways from some of the work you’ve done on this is on what, exactly, the value provided by these programs is, and one of the points you make is not that it helps students develop intellectually, though it helps with that, so much as that it helps these non-cognitive skills. Can you talk a little about what goes into that and why that’s helpful later on?

Look, what did Perry do? Perry took these kids in for 2 ½ hours a day, five days a week, and then had a home visit that went with that, once a week. They did two things. They talked to parents, something about parenting and encouraging the child, but they also taught the child a lot of things that had to do with coming in, showing up, and then what they call a “plan, do, review” sequence. So you plan what you’re going to do that day, you do it and then you review it.

So what you’re learning is self-discipline, to stay on task, you’re learning social relationships, because you’re doing this assessment collectively, and you’re building a set of life skills that turn out to be important. So we looked at what the consequences were of these changes early in life for the child. And we see that those patterns are there.

It leads to less aggression, more socialization, what sometimes psychologists call externalizing behavior, and it promotes a lot of productivity down the line. So you’re changing the character of these children.

A lot of them are intuitive, but what are some of the ways these non-cognitive skills can show up, say, in the workforce or later on in life?

Showing up on time for a job, for example, or learning the importance of getting on with others. Or self control and controlling your anger, respecting the rights of others, various kinds of aspects that you think of as character formation, which we typically take for granted, really shouldn’t be taken for granted for large groups of people in the society. And we can address those deficits with these kinds of programs.

Obviously, your work stresses that these programs aren’t very expensive in the long term.

If I tell you there’s a minimum cost of $8,000, $9,000 to build a device, and that device will pay off 6 to 10 percent per annum every year after, most would consider that a pretty good return. It’s better than the rate of return on equity between 1945 and 2008, so it’s a much higher rate of return than we’re typically used to getting. As an economist, you want to look at opportunity costs and what the benefits are, and you compare the costs relative to the benefits. And the costs are far less than the benefits.

But even if you have this great device that has this amazing rate of return, you still need the initial capital, and you still need a way to invest initially. That’s where a lot of states run into problems with this. Is there anything that looks promising to you as a way of providing that, of getting the process up and running?

There’s a sense in which states are spending a huge amount of money on worthless programs. So, for example, we know that one of the payoffs of these programs is reduced incarceration. So if the state took a long haul, instead of just thinking about what will balance the budget next year or this year, and thought about how they’re going to be solvent in forty years, this is going to dramatically improve financial solvency.

Prison is costing $50-60,000 a year to house a prisoner in a state prison. We see dramatic changes in crime, which is one of the reasons the rate of return is so high. So in this kind of analysis you actually see a huge payoff. The long-term fiscal burden is alleviated. That’s point one.

The second part of it is that the structure even in the short run, the burden on the schools, on special education and a bunch of other burdens that we typically think of as associated with problem children and so forth, these get substantially reduced. So there is a payoff. In some sense, the underlying premise is wrong to think that it’s just cost. It’s cost and return.

My hope was the Obama administration would actually think more broadly about these things. They’ve been under such a gun because of the problem with the economy in the last couple of years that they haven’t been able to do it. But the whole architecture of these programs is producing what should be long-term productivity.

Look, President Eisenhower built the highway system. President Obama could build the child production system if he wanted to. It would have a much higher payoff than a lot of the programs that are currently there. If you do a cost/benefit analysis of the rate of return for job training, if you talk about early convict rehabilitation programs or literacy training for adults, the rates of return on those programs are generally quite low, very low. It’s just a question of using the same dollars wisely.

Last week there was a great op-ed piece in the New York Times by Paul Tough. He pointed out that we’re spending billions, $8.2 billion a year on Head Start, and Head Start is not a very effective program.

If you had an enriched version of Head Start and invested the same amount of money, you’d get much higher payout in the long run. Each of these programs has a political barnacle connected with it. People are promoting it because they see some advantage, but at the same time there’s really no value in those programs. The point is it’s not a question of raising new money, it’s a question of using existing money wisely.

What specifically would you want to have changed? What do you mean by a more efficient program than Head Start, a more effective program?

Well, higher-quality people teaching in the program. Higher standards, better facilities and a more rigorous curriculum, and focusing on developing soft skills and training character for children. Head Start should be much better in terms of targeting children and producing an effective child, instead of just processing a bunch of kids.

You mentioned being disappointed with Obama on this, and if I’m not mistaken you consulted a bit with his campaign on these issues.

I did, on this very issue.

If somehow the economy gets to a place where they can move on to other issues, what would a good first step, federally, be, in moving toward high-quality early education?

What you do is move beyond the Harlem Children’s Zone focus that seems to have gripped the administration. That’s fine, but it hasn’t really been evaluated in any serious way yet and it’s not clear that’s the answer.

The key idea is to encourage more experimentation across a broader range of projects, targeting a larger range of people and providing a refocus of what these programs are all about, which is teaching aspects of self-confidence and teaching these soft skills which are typically ignored in a lot of social and political life.

So I think they could play a great leadership role. It’s not a question of just raising money, but they could certainly raise money as well. They could certainly put a lot more resources into this, but could also try to create a greater emphasis.

Instead of putting money into community colleges and putting money into high schools, to actually understand that there should be a major reemphasis toward the early years and producing children who are going to be productive.

It really would require intellectual leadership. He could do it with a stroke of a pen or a few speeches. He’s started. I don’t want to quarrel with him, and he’s got a lot of other things on his plate, so I’m not faulting him. But I do think the structure has been a little disappointing, to put it mildly.

Do you want to make a final point?

I would just emphasize a couple of things. The first is economic productivity. The second is that the argument that has typically been made for these programs in the past has been that it’s the right thing to do. It’s an emotional argument or a political argument. That may be or may not be true depending on your politics.

But the case for it is also a family-values argument. It’s an argument about recognizing that the American family is in deep trouble, that a lot of kids are not getting the kind of environments that other kids get, and that that environment makes a big difference, and that the structure of the whole program of American society is oriented too much around support for remediation and much less around prevention. That’s true for health, and it’s true for education.

This kind of program has much broader implications. It’s a health program, and it is an education program. It’s also a crime program. If you recognize the multiple benefits that come from promoting both the cognitive and soft skills of these children, especially disadvantaged young children who are being produced at much greater numbers now, because the American family is more in shambles than probably ever before, at least in recent time, the last 40-50 years.

So I think that there’s a strong economic productivity argument to be made. The societal impact is much, much greater than is typically viewed. People have to get out of these very narrow policy conceptions, out of the idea that this is the province of the Education Department or HHS. It’s a way to improve quality of life across a number of dimensions. It means people can live bigger lives, greater lives.

By Dylan Matthews  |  August 25, 2010; 10:46 AM ET
Save & Share:  Send E-mail   Facebook   Twitter   Digg   Yahoo Buzz   StumbleUpon   Technorati   Google Buzz   Previous: How prisons make bad citizens
Next: Research Desk: Where would unemployment and GDP growth be without the stimulus?


Very good interview, Dylan. Heckman was very forthcoming

Posted by: bdballard | August 25, 2010 11:50 AM | Report abuse

Yes, agreed. Good interview. I'd probably put the bulk of it behind a click-through, though. It does take up a lot of real estate on the front page...

Posted by: MosBen | August 25, 2010 11:53 AM | Report abuse

Heckman is impressive and presents an excellent case.

Finding no-brainer investments is very difficult. Here we have a no-brainer investment that is also provides a badly needed moral/ethical good. We should have the courage to take it and run.

While we're on the topic, California is fast approaching the point where we will spend more on incarceration than on schooling for children. Can you get a clearer indictment of society than that?

The 2000 census #'s had 19M kids under 5 years old. So it's roughly 10M kids that might be of pre-school age. 10M x $8,500 = $8.5B/year.

Posted by: BHeffernan1 | August 25, 2010 11:57 AM | Report abuse

"The 2000 census #'s had 19M kids under 5 years old. So it's roughly 10M kids that might be of pre-school age. 10M x $8,500 = $8.5B/year."


I calculate $85 billion, not $8.5 billion. Not that this makes a difference in the investment decision if the ROI is really that strong. I'd also agree with you that we spend(waste) way too much money incarcerating people, particularly for victimless crimes, money that can be put to a higher and better use.

Posted by: justin84 | August 25, 2010 1:04 PM | Report abuse

What bothers me about this issue is that people see it as so binary. Someone does this, therefore we must lock them up. If we're at a point where we're incarcerating 10% of the population, we're doing something wrong. Maybe we need to take a look at the things for which we've made incarceration a punishment. Maybe some of them are based on principles which we've grown out of. Maybe we took that position in an era when it was cheaper to lock people up and we had the room, so we went ahead with it, but can't afford it now. Maybe we're not doing enough to ensure that some crimes are avoided to begin with, or that former criminals are reintegrated into society such that they are better able to avoid committing the same or new crimes.

There's nothing holy about our penal code. It was made up as we went along, and there's nothing wrong with deciding to change it if its current form is causing us problems.

Posted by: MosBen | August 25, 2010 2:00 PM | Report abuse

"Instead of putting money into community colleges and putting money into high schools, to actually understand that there should be a major reemphasis toward the early years and producing children who are going to be productive."

So what, we just give up on older kids? Anybody over 12 is a lost cause? No retraining for you, unemployed worker, better suck it up and walk it off? The idea that you could do such a thing is just as unrealistic as the current system. You need a balance.

Posted by: wayward_va | August 25, 2010 2:44 PM | Report abuse

Whoops! Thanks justin84.

Yes, if ROI is there it should still be done. In theory. But it's easier to find $8.5B than $85B. (And of course it need not be federal-only dollars, but it's still a lot).

The prison spend really bothers me as I have kids entering underfunded schools now. The 'prison without walls' concept is very appealing.

Posted by: BHeffernan1 | August 25, 2010 3:42 PM | Report abuse

BTW, I have two kids, one in kindergarten and one in preschool. Our preschool is a co-op, so we work at the school and it costs about $4K/year (CA) for 3 days/week, 1/2 days. High end preschools here cost $9K/year. What struck me initially about Dylan's comment about it being very expensive was that he probably doesn't know the cost of preschool yet given his age. $8500 is expensive but not insane for an intensive, 5-day a week program.

If you have the money, your kids go to preschool. The ones that don't have the money suffer (as does the rest of society in the long run).

Posted by: BHeffernan1 | August 25, 2010 4:03 PM | Report abuse

This part is missing something big:

"Obviously, your work stresses that these programs aren’t very expensive in the long term.
If I tell you there’s a minimum cost of $8,000, $9,000 to build a device, and that device will pay off 6 to 10 percent per annum every year after, most would consider that a pretty good return. It’s better than the rate of return on equity between 1945 and 2008,"

He left out the part where you don't start getting the money for about 20 years, and it only lasts for about 40 years (and doesn't compound) $9000 * 6% * 40 = $21600

"So you put in those $8,000, you’re going to get multipliers of more than 100 at age 65 in real dollar terms. I once did a calculation, it’s whatever 1.06 is to the 60th power. It’s a huge number, I promise you. I think it’s very shortsighted to talk only about the costs. You want to look at the returns."

1.06 to the 60th power is about 33 (a huge number!). 33 * 9000 = 29700. However, I would say, most people don't work 60 years.

It's nonsense that this guy is trying to make all of these numbers claims when he's so obviously distorting things or just making glaring errors.

Overall though, the premise that we should be putting more money into taking kids away from stupid parents that can't read before school starts is a good one. It's certainly better than the continued subsidies we use to drive up the price of university education.

Posted by: staticvars | August 25, 2010 4:10 PM | Report abuse

@staticvars, good post but it may not be that bad. The link to Heckman's work given yesterday (, said that 65% of the ROI (which he gives as $8.70 per $, close to your $9) are due to reduced levels of crime. So most of the benefit doesn't come from higher pay over a career.

They would calculate savings from crime reduction by assigning a number to the social cost of crime and taking a portion of it based on the reduced likelihood of incurring that cost. Agreed, he's making his case sound as good as he can, but I don't think it's clear he's distorting.

I mean the guy is a Nobel prize winner in economics. (But to be fair, this post is from someone who was not successful in multiplying two numbers together above).

Posted by: BHeffernan1 | August 25, 2010 4:35 PM | Report abuse

Didn't I hear something about an ounce of prevention being worth a pound of cure?

Why have we forgotten that? The way we spend our money is cheap, it isn't fiscally conservative, it's stupid!

Readers of this blog would also be very interested in the proven results of the home visitation program he links to. I think it's state-wide in Minnesota.

Posted by: lroberts1 | August 25, 2010 8:49 PM | Report abuse

For more information about Prof. Heckman's ideas and to see him speak, check out

Posted by: Michael_Hoffman | August 26, 2010 3:59 PM | Report abuse

Post a Comment

We encourage users to analyze, comment on and even challenge's articles, blogs, reviews and multimedia features.

User reviews and comments that include profanity or personal attacks or other inappropriate comments or material will be removed from the site. Additionally, entries that are unsigned or contain "signatures" by someone other than the actual author will be removed. Finally, we will take steps to block users who violate any of our posting standards, terms of use or privacy policies or any other policies governing this site. Please review the full rules governing commentaries and discussions.

characters remaining

RSS Feed
Subscribe to The Post

© 2010 The Washington Post Company