Age-size effects in productive efficiency: a second test of the passive learning model

Several studies in developed economies have reported that the rate of growth of small firms decreases with firm age as well as size. The result is consistent with Jovanovic's (1982) version of the passive learning model of competitive selection and is confirmed by data on a random sample of manufacturing firms in Ethiopia. This paper uses the Ethiopian data to test for three other implications of the Jovanovic model. These link age-size effects in firm growth to an underlying distribution of firms by technical efficiency, but have not been investigated empirically before. We find that firstly, the age-size effects detected in the growth of firms in our sample are matched by time-invariant inter-firm differences in technical efficiency. Secondly, there are also age-size effects in efficiency whereby bigger firms are more efficient given age, and older firms are more efficient given size. Thirdly, firm age and firm size mainly proxy for owner human capital and location variables in as far as they explain efficiency scores. In other words, it is not the case that some firms are more efficient than others because they are bigger or older but the other way round some firms are bigger or longer lived than others because they have proved to be more efficient. Advantageous location is a major source of technical efficiency but is not as important as greater entrepreneurial human capital. Among the human capital variables considered, the level of formal education completed has by far the strongest influence on efficiency scores. The owner's access to business networks and his or her ethnicity also have significant effects. On the other hand, there is no evidence that efficiency depends on any one of pre-ownership employment experience, occupational following of parents or prior vocational training.