Taking Stock
This chapter took a first step towards the analysis of structural changes involved in the process of economic development. Our first step has been relatively modest. The focus has been on the structural changes associated with the shifts in output and employment away 841
from agriculture to manufacturing and to services and with the changes between sectors of different capital intensities.
Section 20.1 focused on demand-side reasons for why growth can be non-balanced. In particular, it incorporated Engel’s law into the basic neoclassical growth model so that households spend a smaller fraction of their budget on agricultural goods as they become richer. This framework is ideally suited for the analysis of the structural changes across broad sectors such as agriculture, manufacturing and services. Section 20.2, on the other hand, turned to supply-side reasons for non-balanced growth, which were first highlighted by Baumol’s (1967) classic paper. However, instead of assuming exogenously-given different rates of technological progress across sectors, this section emphasized how sectoral differences in capital intensity can lead to non-balanced growth. Capital-intensive sectors tend to grow more rapidly as a result of an equi-proportionate increase in the capital-labor ratio. This feature, combined with capital deepening at the economy level, naturally leads to a pattern of non-balanced growth. This type of non-balanced growth may contribute to structural changes across agricultural, manufacturing and service sectors, but is more relevant when we look at sectors differentiated according to their capital intensity. A particular focus of both Sections 20.1 and 20.2 has been to reconcile non-balanced growth at the sectoral level with the pattern of relatively balanced growth at the aggregate. As already noted in Chapter 2, balanced growth need not be taken literally. It is at best an approximation to the growth behavior of advanced economies. Nevertheless, it seems to be a particularly accurate approximation to many features of the growth process, since interest rates and the share of capital income in GDP do appear to have been relatively constant over the past 100 years or more in most advanced economies. It is therefore important to understand how significant reallocation of resources at the sectoral and micro levels can coexist with the more “balanced” behavior at the aggregate. The models in Sections 20.1 and 20.2 suggested some clues about why this may be the case, but the answers provided here should be viewed as preliminary rather than definitive.I also discussed a simple model of the origins of industrialization. This model showed how agricultural productivity might have an important effect on the timing of industrialization, but also demonstrated that the effect of agricultural productivity might depend on whether or not the economy is open to international trade. The origins of industrialization are important because, as discussed in Chapter 1, existing evidence suggests that the timing and nature of industrialization may have important implications for cross-country income differences we observe today, and thus the investigation of the economic development problem might necessitate an analysis of why some countries industrialized early while others were delayed or never started the process of industrialization.
Understanding the sources of the structural changes and how they can be reconciled with the broad patterns of balanced growth in the aggregate, as well as an analysis of the origins
of industrialization, sheds light on both the process of economic growth and the process of economic development. In this sense, the models in this section enrich our understanding of economic growth considerably. And yet, this is only a modest step towards the investigation of the sweeping structural changes emphasized by Kuznets because we have not departed from the neoclassical approach to economic growth.
In particular, Sections 20.1 and 20.2 used generalized versions of the basic neoclassical growth model of Chapter 8, and Section 20.3 used a variant of the Romer (1986) model from Chapter 11.It should be emphasized again that the topics discussed in this chapter, though closely related to the basic neoclassical growth model, are areas of frontier research. We are far from a satisfactory framework for understanding the process of reallocation of capital and labor across sectors, how this changes at different stages of development, and how this remains consistent with relatively balanced aggregate growth and the Kaldor facts. I have therefore not attempted to provide a unified framework that combines the transition from agriculture to industrialization, the demand-side reasons for non-balanced growth and the supply-side forces (even though Exercise 20.17 provides some hints on how this may be achieved). The development of such unified models as well as richer models of non-balanced growth are areas for future research.
20.4.
More on the topic Taking Stock:
- Taking Stock
- Taking Stock
- Taking Stock
- Taking Stock
- Taking Stock
- Taking Stock
- Taking Stock
- Taking Stock
- Taking Stock
- Taking Stock