Skip to main content

The return to firm investments in human capital (Almeida & Carneiro 2008)

  • Findings

    See findings section of this profile.

    Evidence Rating

    Low Causal Evidence

Review Guidelines

Citation

Almeida, R., & Carneiro, P. (2008). The return to firm investments in human capital. World Bank Social Protection and Labor Discussion Paper No. 0822.

Highlights

  • The study’s objective was to examine the returns to firms’ investments in training for their workers.
  • The authors used data from an annual survey conducted by the Portuguese Ministry of Employment, which is a mandatory survey for all Portuguese firms with at least 100 employees, to determine the internal rate of return to the firm of an additional hour of training per employee.
  • The study found that increasing the amount of training per employee by 10 hours per year led to an increase in current productivity of 0.6 percent.
  • The quality of causal evidence presented in this report is low because the instruments used in the estimation approach were weak, as described by the authors. This means we are not confident that the estimated effects are attributable to the training provided to the employees; other factors are likely to have contributed.

Features of the Study

The authors analyzed data from an annual survey conducted by the Portuguese Ministry of Employment. Completing the survey was mandatory for all Portuguese firms with at least 100 employees. Specifically, the data included a panel of 1,500 manufacturing firms in Portugal from 1995 through 1999, for a total of 5,501 firm-year observations.

Using generalized method of moments, the authors estimated a structural model in which the parameter of interest was the internal rate of return to the firm of providing an additional hour of training per employee. To estimate the components of the model, they used a first-difference instrumental variables approach that controlled for firms’ unobservable and time-invariant characteristics. To account for possible correlation between input choices and transitory productivity or cost shocks, the authors used instruments to statistically account for current differences in inputs using lagged values of inputs.

Findings

  • The authors’ preferred model found that increasing the amount of training per employee by 10 hours per year led to an increase in current productivity of 0.6 percent. Forgone productivity accounted for less than 25 percent of the total cost of training.

Considerations for Interpreting the Findings

The authors used a sophisticated statistical method to estimate the returns to training. This method involved the use of instrumental variables to account for possible correlations between firms’ input choices and productivity or cost shocks. Although the authors did not report the results of a test of instrument strength directly, which is required for such an approach to receive a moderate causal evidence rating, the report acknowledged that the instruments the authors used were weak. Therefore, the study cannot meet the criteria for a moderate causal evidence rating.

Causal Evidence Rating

The quality of causal evidence presented in this report is low because the instruments used in the estimation approach were weak. This means we are not confident that the estimated effects are attributable to the training provided to the employees; other factors are likely to have contributed.

Additional Sources

Almeida, R., & Carneiro, P. (2006). The return to firm investments in human capital. World Bank Policy Research Working Paper 3851.

Reviewed by CLEAR

September 2015

Topic Area