Early Ed Watch - logo

The Trouble with Touting the $10 to $1 Benefit-to-Cost Ratio for Pre-K

Every $1 invested in quality preschool programs saves $10 in reduced expenses down the road. That statistic has been making headlines lately, thanks in part to appearances in major speeches by President Barack Obama and Secretary of Education Arne Duncan. But how accurate is it? And what does it really mean for early education programs?  

A New York Times story last week delivered a prime example of how this statistic can become misleading -- and how it can create unrealistic expectations among policy makers and the public, who may erroneously conclude that we can generate good returns by expanding existing early childhood programs, without first improving their quality. 

The example came in a story about possible use of federal money to rescue preschool programs from recession-based cutbacks. The reporter first provided details about stimulus money for Head Start and Early Head Start and then explained that the Obama Administration's "enthusiasm for early childhood education is based on research showing large paybacks for every dollar spent on the careful nurturing of poor children." Then came that oft-cited statistic, this time in a quote from Arne Duncan, U.S. Secretary of Education: "For every dollar we spend on these programs, we get nearly $10 back in reduced welfare rolls, fewer health care costs and less crime," Mr. Duncan said.

Taken in parts, the story is accurate. But by mentioning Head Start and then moving straight into Duncan's quote, the article leaves readers with the impression that Head Start, or some equivalent, has delivered this 10 to 1 return on investment. Sadly, that's just not the case.

It's not that the numbers have been pulled out of thin air. They are based on sound science from Arthur J. Reynolds and Judy A. Temple of the University of Minnesota who have studied Chicago's Child Parent Centers for years. As we have reported before, the CPCs are high-quality programs based in the public schools that employ well-trained teachers with bachelor's degrees, start with 3-year-olds, align pre-k with what is taught in the elementary school grades, and put a strong emphasis on parental involvement.

Unfortunately, examples of similar preschools are not the norm in your average Head Start or state-based pre-K programs.  

Larry Abramson of National Public Radio is one reporter who took pains to point this out in a March 12 All Things Considered segment titled "Examining Obama's Education Numbers."  He reported that these benefits come from "very well designed" programs, adding that Head Start, generally speaking, has not shown the same payoff. One expert he interviewed contended that getting the data wrong "gives ammunition to skeptics."

We see the same threat. Touting a $10-to-$1 return may be helpful when advocating for more emphasis on quality. But if the numbers are used to advocate for simply doing more of what we've been doing, they will become symbols of exaggeration and failed promises. It's time to get serious about exactly what this cost-benefit ratio means, where it comes from and what high-quality preschool programs actually look like.

First, consider how Reynolds and Temple arrived at the $10 to $1 figure. By looking at how children fared many years after enrolling in the CPC program, they were able to calculate savings in government spending. For example, CPC participants needed an average of 0.7 fewer years in special education services than their non-CPC counterparts, saving money the public school system would otherwise have spent on special education services.

Cost-benefit analyses typically include many other factors, such as arrest and drop-out rates or the impact that pre-k programs have on mothers' earnings by enabling them to work while their child is at a full-day center or preschool. Using such data, studies have shown that the Perry HighScope program in Ypsilanti, Mich., delivered $8.74 in benefits for every dollar spent. The Abecedarian program in North Carolina returned $3.78 per dollar spent. (See page 118 of Reynolds' and Temple's 2008 paper for a helpful comparison chart.) A 2005 RAND study, based on projections of pre-k's impact in California, showed relatively strong benefits too: $2.62 per dollar invested.

Cost-benefit studies have their limitations, particularly when they are making assumptions of savings to come. For example, when researchers make projections about the costs of arrests 10 years from now, they may assume that young adults will be arrested at the same rate they are today. As with any investment scenario, the further away from present time, the trickier it can be to make a solid case for assured returns.

But these analyses can be useful in the aggregate for elevating the dialogue about the benefits of high-quality pre-k--as long as the highest number is not the only one cited. So far, we have evidence from more than a dozen studies that the benefits largely exceed costs. We may be tempted to pinpoint exact numbers, said Steve Barnett, president of the National Institute for Early Education Research, but "the most exact finding is that the cost-benefit ratio was ‘big.' "

What makes for a high-quality program that might achieve these "big" returns? Class sizes of less than 9 children to one adult. Enrollment no later than age 3. Programs that encourage parental involvement. Alignment with and ongoing support through the early elementary years. Competitive salaries for staff.

And having classrooms led by teachers who have earned at least a bachelor's degree. The teacher-degree piece alone is missing in many pre-K and Head Start programs. A 2005 study of pre-k in 11 states showed that three-quarters of lead teachers in state-funded pre-k classrooms had a bachelor's degree or higher. Not until 2013 will half of teachers in Head Start classrooms be required by law to have a B.A.

This isn't to say that Head Start in its current iteration makes no difference. The program is helping children in many ways. The Impact Study of 2005 shows Head Start children are better prepared for kindergarten, achieve gains in pre-reading skills and are more likely to attend college than their non-Head Start counterparts, who typically attended childcare centers or were cared for in relatives' homes. The gains are not as high as those from the Perry or CPC models, but they may be enough to deliver more than $1 in benefit for every $1 invested, according to a 2007 policy report. Keep in mind, too, that a Westat study shows Head Start centers score higher on measures of classroom quality than child-care centers, and because Head Start offers services like dental care and vaccination screenings, it can provide long-term health benefits that can lead to public savings.

Many state-funded preschool programs don't emphasize quality enough either -- in fact some have far fewer positive quality characteristics than Head Start nationally. The 2008 NIEER report card shows that just 27 states require pre-k teachers to have a B.A. And the majority of preschool children in publicly-funded programs live in states that, according to NIEER, do not spend enough money per child to deliver a high-quality experience.

It is possible, however, to scale up preschool programs that meet more rigorous standards. In Oklahoma, for example, where Head Start centers collaborate with the state-funded preschool program, children going through the Tulsa pre-k program have experienced large gains on tests of cognitive ability, according to research by William Gormley of Georgetown University. These gains are comparable to those of children participating in Chicago's CPC program. That suggests that Oklahoma's pre-k initiative may deliver long-term benefits for children similar to those found in the CPC research, although it's too early to tell, since the children in the Oklahoma study are still in school.  

When it comes to scaling up high-quality pre-k programs, "Oklahoma gives us a shot," Barnett said.

Our wish is to see all programs -- whether Head Start, state-funded pre-k, or child-care settings -- evolve into such high-quality services that they deliver the kind of long-term benefits we see in these more intensive programs.

That is why we want to make sure that the rhetoric about preschool does not lead to lazy policy. To generate returns on investment, we must not simply increase the numbers of children enrolled in existing early education programs. We need to also raise the quality of these programs to the level needed to generate long-term results. Failure to do so could undermine support for early education. Five years from now, people will want to start seeing some of those promised returns on early childhood investments. But they won't show up unless children are getting the kind of experiences that produced positive outcomes in high-quality model pre-k programs.

"The risk is that people will implement the weaker program thinking they are going to get the same return, and they are not," Barnett said.

Or, as Larry Schweinhart of Perry HighScope said during a presentation at the Society for Research in Child Development's meeting earlier this month:  If we don't get serious about enhancing quality in early childhood programs, the result will be a big disappointment. "Are we missing the promise?" he asked. "Are we baiting and switching?"

(Graphic created using photo by Flickr user Miranda4Flicker under the Creative Commons license.)


Classroom Quality

Thank you for pointing out the good quality of Head Start classrooms. In doing so, you referenced a report with late 1990's data. The good quality of Head Start classrooms has also been observed to be good more recently. (See page 3-13 of the 2005 Head Start Impact Study report at

Perry Economic Impact

Nores et al (2005) estimated the return of the Perry Project as $12.90 for every dollar spent.

Head Start Equals Quality

I wanted to say that I found the article very interesting and I would agree with the premise that it makes sense for Head Start programs to move toward BA teachers. And they are moving in that direction as is required by the new reauthorization act.

However, I do think that the New America Foundation seems to almost be obsessed with whether or not teachers have degrees and touting the benefits as though they should be the answer to all of our problems. I think a more comprehensive evaluation of Head Start is necessary--one that looks at the benefits to both the child and the parent and includes a real evaluation of all the services provided to the family. This could include emergency shelter assistance or health services. For example, what are the savings to taxpayers when family advocates find families stable affordable housing as opposed to living a shelter or out of car or what is saved when a parent obtains a GED or develops a new outlook on life so they go to school or work. What does that save taxpayers? What about early detection of health problems? How much does that save taxpayers?

In any event I think we sometimes get too caught up in the discussion of the BA degree and not look at the bigger picture and the incredible benefits Head Start provides to the entire family. A closer examination of what the program provides would demonstrate much larger savings than what the authors would have us believe. Oh and by the way if the BA degree is the essential ingredient to all the ills with education why do the public schools in poor neighbors repeartedly fail where Head Start programs succeed?

Defining quality

Your point that high-return early childhood education programs need to be high-quality is well-taken.

However, I'm not sure that everything you list as defining high-quality is backed by research. For example, the post lists "Enrollment no later than age 3" as essential to high-quality. Yet CPC, one of the programs cited, only had half their kids enrolled for two years. The others were only enrolled at age 4, for one year. Furthermore, Professor Reynolds's own research found that although returns were somewhat higher for children enrolled for 2 years, they weren't doubled. Generally the differences between 1 year and 2 years of CPC were not statistically significant.

In addition, the Oklahoma program you list later on is a one-year program at age 4.

On the whole, I think the research suggests that preschool programs that start at age 4 can have high returns.