Skip to:

It's The Thought That Counts

In an editorial for JAMA Pediatrics, Health Policy and Management Professor Fred Zimmerman discusses how inclusion-based incentives for new parents can improve child development. 
Monday, January 27, 2014

In 1949 the government of Finland began sending boxes to the parents of every newborn child in the country.1 The boxes have been given to every expectant mother in Finland since and now include clothes, diapers, a bib, and other essential supplies. The baby can even sleep in the box itself. The purpose of distributing these boxes is to encourage pregnant women to visit a physician for prenatal care, and it works.2 Finnish mothers seek timely prenatal care at rates exceeding 97%, among the highest rates in Europe.3 Of course, there are other incentives and facilitators besides Finland’s maternity box, but these incentives are typically less generous in Finland than elsewhere.2 Finland’s approach is, however, the most concrete. The arrival of a box filled with baby clothes carries a powerfully tangible sign that the baby is both real and a welcome member of society.

The SEED for Oklahoma Kids (SEED OK) approach also produces benefits, perhaps in a similar way. As the article by Huang et al4 in this issue of JAMA Pediatrics shows, the distribution of $1000 529 accounts in children’s names promotes improved social-emotional development at age 4 years, presumably because of greater parental attention to their children. One thousand dollars is a nice, round number, and large enough to focus the mind. But how important is the actual amount?

This policy derives from a literature that suggests that asset holding can improve child outcomes, in part by changing the attitudes and behaviors of parents.5 While there is an obvious positive association between assets and child outcomes, the empirical evidence testing whether giving poor people assets improves their children’s outcomes is promising, but somewhat mixed and not yet fully persuasive.6 Researchers and policy makers will accordingly be extremely excited to see a rigorous test of this approach in a real-world setting.

One of the many advantages of a randomized design is to focus attention on the intervention. Here what is involved is not only a transfer of assets, but also regular add-on gifts and the annual account statement. This distinction is important: if only the assets matter, then adding more would presumably produce a larger effect; but if the mechanism of action is instead through the reminders and add-ons, or through the signal of inclusion that a major child-based transfer implies, then the specific amount is less important, and the program could have a larger effect by enhancing these other inclusion-based components.

It is striking that so few people—only 15%—take advantage of the offer of an additional $100 in the form of a parent-owned 529 account. So while $1000 seems to matter, $100 seems not worth the trouble. What’s going on here? Of course, there are some barriers to signing up for the additional $100, but even so, with such a small proportion of people electing to accept the free $100—and even fewer electing to accept the matching amount by investing their own money—it is clear that real money is being left on the table. Perhaps assets are not the only mode of action after all.

The SEED OK plan has an effect on child outcomes only among those who are at risk. Children of parents with incomes greater than 200% of the poverty line experienced no benefit, significant or otherwise. It may be that there was a floor effect in that children of middle- and upper-class parents had few enough social-emotional problems to make further reduction in these problems quite difficult. It may also have been that the primary mode of operation of the intervention was not through financial incentives, but rather through the social solidarity communicated through the $1000 529 plan and other gifts—a message that middle-class parents already get by other means. Participants reported that the program helped them feel “a whole lot better” and gave them hope for their children. In America, money talks, and there is no better message to send with money than that a child is valued.

In Finland the emphasis is on babyhood—colorful clothes whose designs change from year to year, a teething ring, and a picture book—and this gift seems to celebrate infancy as a stage in itself. The Oklahoma plan, by contrast, reflects the American cultural emphasis on money and professional success.

Yet not all children will in fact go to college, nor would either individuals or the American economy be well served if they did. Nationally only about 60% of children will attend college,7 and only about 25% complete college.8

In this context it is striking that the SEED OK program is so narrowly focused around a future outcome that will largely elude many of the participants, including so many of those for whom the intervention seems to work most effectively. One can’t help wondering how parental interest in their children will evolve as some children do not excel in school over time. Will the effect of the program be diminished or even reversed?

The Washington State Institute for Public Policy has produced a report reviewing the costs and benefits of well-studied interventions in early childhood.9 They find 7 programs in which the social benefits exceed the costs by a wide margin, including 4 in which the benefits to the taxpayers alone significantly exceed the costs. With additional replication of the results presented by Huang et al, some form of inclusion gift for infants is likely to join this impressive list of evidence-based and cost-saving programs.

Yet this study leaves many important questions unanswered, such as what the method of action really is, how long the effects will last, whether they are concentrated in one subpopulation or are more general, how they can be enhanced and strengthened, and whether there are other beneficial effects of the program beyond social-emotional development. Far more work needs to be done. Yet here may be where the study makes one of its most important contributions, for it demonstrates the enormous value of subjecting social policy to the rigors of randomized trials.

Randomizing benefits, especially when the true effects of those benefits are uncertain, would result in far more rapid improvements in social policy and social well-being at far less cost than the current practice of universally implementing policies that are politically popular, and only sometimes supported by strong evidence, much of it derived from efficacy, not effectiveness, research.

At minimum what is known now, or at least strongly suggested by this well-conducted trial, is that something about giving parents a $1000 savings plan with a few other goodies seems to help their children to develop well. While the mechanism is not yet clear, the general principle is established: parents respond well to a formal welcome of their child. This response can be observed in the United States as in Finland and works whether the welcome is in the form of money for future education or clothes for the baby.

As one recipient of the Finnish box put it, “This felt to me like evidence that someone cared, someone wanted our baby to have a good start in life. It strengthens that feeling that we are all in this together.”1 Sometimes, indeed, it’s the thought that counts.

Read the editorial in JAMA Pediatrics