Kelley and Schmidt explore the relationship between population and per capita output growth rates on international countries based on data gathered from 1960 – 1990.
Summary of Findings
Based on their data, Kelley and Schmidt find no statistically significant relationship between population and per capita output growth rates in the 1960s and 1970s. However in the 1980s, a strong negative relationship exists between these factors. In less developed countries (LDCs) the relationship is more pronounced and negative whereas in developed countries (DCs) the relationship is less negative and sometimes positive. In the 1960s and 1970s, the only factor contributing statistical significance is density, while in the 1980s all factors contribute statistical significance. A potential reason for this negative relationship in the 1980s is that population growth brings about substantial diminishing marginal returns that overshadow the positive affects of technological innovation that normally accompany population increases.
One of the distinguishing characteristics of Kelley and Schmidt’s model in comparison with those before is that they sought to compare the short and long run effects of certain demographic factors individually. An example of this can be seen when comparing the effect of births. The short-term effect of birth is negative since children require resources. The long-term effect of birth is positive because these children become adults and create resources. If one were to combine the two effects, we would see an offsetting result that would incorrectly indicate that birth has no effect on economic growth. Therefore, Kelley and Schmidt’s model contains variables that track the effect of current births and the number of past births.
To better understand the impact of demographic factors on per capita output growth, Kelley and Schmidt use a convergence model, which allows them to dissect the data with regard to level of economic development and to time period. The lagged variable included in the model is a crude birth rate (CBR) 15-year lag. With the addition of this CBR-15 lag variable to the normal CBR variable, Kelley and Schmidt can differentiate between the effects of past and current births on per capita output growth.
In the thirty-year time frame from 1960 – 1990, population growth has a negative effect on economic growth. The increase in the CBR decreases economic growth while the decrease in the CDR increases economic growth, however the impact of the latter is not sufficient to create a positive relationship between population growth and economic growth. When looking at the data with respect to economic development, we see that the negative relationship is stronger in LDCs than DCs.
In the 1960s and 1970s, the population growth rate did not have a large impact on per capita output growth, however there was a strong negative impact in the 1980s. It is believed that the cost of birth increased in the 1980s and thus made the burden of raising children grow. In addition, we see a decline in the impact of mortality (a positively contributing variable), reinforcing the negative relationship found.
“Aggregate Population and Economic Growth Correlations: The Role of the Components of Demographic Change” by Allen C. Kelley and Robert M. Schmidt in Demography, Vol. 2, No. 4, November 1995, pp. 543-555.