Estimating the Production Function for Human Capital: Results from a Randomized Controlled Trial in Colombia

EGC Research Summary, January 2020

In a dusty courtyard, two women show a young girl a toy.
Home visitors instructed parents in engaging children in educational play.

An intervention in Colombia aimed at encouraging low-income families to stimulate their toddlers with play and involve them in household activities improves cognitive and socio-emotional skills development, finds a new study led by EGC faculty affiliates Orazio Attanasio and Costas Meghir.

The positive effects of the program were primarily explained by parents’ increased investment in learning materials and quality time spent with the children as a result of the program, not the weekly intervention on its own – a finding which underscores the importance of promoting early and continuous investment in children’s development.

Estimating the Production Function for Human Capital: Results from a Randomized Controlled Trial in Colombia

Orazio Attanasio, Sarah Cattan, Emla Fitzsimons; Costas Meghir, and Marta Rubio-Codina, American Economic Review, January 2020, Pages 48-85

Results at a glance

  • Positive effects of early-childhood programs on cognitive and socio-emotional skills have been shown in research around the world, however this study is the first to build a model to explain why the program is effective.
  • Researchers found that the increases in child cognitive and socio-emotional skills as a result of the program could be almost entirely explained by parents’ increased investment in learning materials such as toys and books encouraged by the program. Thus, the visits by home visitors aimed to increase the quality of mother-child interactions by themselves benefited the children through the channel of increased parental investments in quality time and materials. 
  • Micronutrients supplements had no effect on children’s cognitive development, while the stimulation visits did. This highlights that nutrition at that age window (1.5-3.5 on average) was not the main hindrance to cognitive development in this middle-income country context.
  • The study designed and evaluated a child stimulation program for beneficiaries of a Colombian conditional cash transfer program. The study trained local women to visit families with children ages 12-24 months at the start for two years.It taught mothers about creating learning experiences out of household activities for their children, in addition to providing micronutrient supplements. Covering approximately 1,500 children, this is one of the largest samples ever that examined the effects of programs aimed at stimulating children’s development. 
  • Two years after the program ended, parents who received visits from the home visitors decreased their investments to the same level as parents who did not receive the intervention. This highlights the necessity of encouraging parents to sustain their investments in their children and to follow up with further intervention.

Researchers increasingly agree that life outcomes are rooted in early childhood development, which has led to many middle- and low-income governments around the world developing social programs targeting young children. These programs typically involve home visits where parents are taught ways they can create learning experiences for their children out of everyday routines and resources. While studies have generally found these programs to be effective in improving child development, the actual mechanisms through which the programs work have remained unexplored. This paper is among the first to develop a model that explains the factors affecting parents’ investment choices. 

The evaluation was based on a clustered randomized controlled trial that randomly assigned 96 municipalities to one of four groups: a control group, a group that received the stimulation intervention delivered by home visits, a nutritional supplementation intervention, and a group that received both. 

The interventions, and in particular the stimulation interventions, were designed to be both scalable and sustainable by using the infrastructure of a preexisting welfare program. In particular, the stimulation intervention was delivered by the elected representatives of the recipients of the Colombian Conditional Cash Transfer, which targets the poorest 20 percent of the population. These representatives were trained in delivering weekly home visits to the mother of children aged between 12 and 24 months, with the aim to teach mothers more effective ways of engaging with their children. 

Parents increased their material and quality-time investments in their children as a result of the intervention.  

Parents who participated in the program increased their investments in both learning materials and in quality-time spent with their children. When Attanasio, Meghir (and co-authors) modeled this increased investment on child cognitive ability and socio-emotional skills, they found that material investments were the main explanation for children’s improved development. The important point here is that it was a change in parental behavior, rather than the visits themselves, that improved children’s development. The primary effect of the intervention was to encourage parents to provide more books and toys for their children, which in turn improved child development. Future early childhood development programs should keep this in mind when designing their interventions. 

Micronutrient supplements had no impact on child cognitive ability. 

Unlike in similar studies in other countries, the researchers found the micronutrient supplement to have no effect on child cognitive development. This finding suggests that child development among the target population, and for the specific age group considered, was not significantly hindered by nutritional deficiencies, which has been shown to be the case in other contexts, especially those that targeted malnourished children or children with especially low cognitive ability. The target population in this study was children from poorer families in a middle-income country, which indicates the importance of tailoring early childhood developments to different development contexts. 

Parents invest more when they receive outside assistance, not less.

A key finding was that parents invested more in their children, not less, as a result of their participation in the early childhood development program. This helps resolve a debate of whether outside assistance “crowds out” parental investment, i.e. causes parents to reduce their own investment because of the greater external help, or “crowds in” parental investment by pushing parents to increase their involvement and contributions. The paper firmly supports the latter, which means that government programs should not worry about displacing parents’ own efforts. 

Sustained intervention might be necessary to encourage the continuation of parental investment.

Two years after the program ended, parents who were randomly assigned to receive the intervention no longer demonstrated greater investment in their children than parents who did not receive the intervention. Parents reduced their investment at the same time when investment would be more productive. Because cognition skills are self-productive, meaning that investments in children with higher baseline skills induce greater improvements than children with lower baseline skills, continued investment has the potential to be very meaningful in child development. 

This study is part of a series of randomized trials Attanasio and Meghir are conducting in different countries exploring the effects of early childhood interventions on the development of human capital. 


Research Summary by Lisa Qian