Lecture 13. Oct. 9 - Ch. 9, part 1

Growth in the U.S. and abroad

We focus on per capita GDP as the operational measure of long-term economic growth, because of its association with increases in the material standard of living. Average growth in the U.S. of 1.7 percent per year (over the past century) translates into a doubling time of about 40 years, but there have been periods of more (1920s, 1960s) and less (1950s, 1974-94) rapid growth (see Parkin's Fig. 9.1; note doubling time estimation -- rule of 72).

Other industrialized countries have been growing more rapidly than the U.S. since 1960, and so have a number of developing countries, mostly in Asia (where formerly large differences in GDP per capita have narrowed considerably).

Sources of growth

Parkin notes that markets, property rights, and monetary exchange are institutions that help to build a well-functioning incentive system, and thereby serve as preconditions to growth. Presently, the former communist countries of Eastern Europe and the Soviet Union are attempting to transform themselves into market economies, in large part with the objective of bringing about sustained and significant economic growth. The first two of these preconditions are proving to be difficult to implement. An additional precondition not mentioned by Parkin but im-portant is a certain degree of political stability. Instability results in a significant risk premium being added on to real interest rates, and the resulting high interest rates inhibit investment and growth.

Saving is a key factor in contributing to growth, because of its role in providing funds for investment in physical capital. Such investment serves to increase the stock of capital per worker, and this contributes directly to increased productivity of workers (cf., notion of an aggregate production function; Y=f(K,L) or y=f(K/L)).

Human capital refers to the acquired skills and knowledge of individuals. Interest by economists in investment in human capital (acquiring productive skills) goes back to the late 1950s/early 1960s, when it became clear that increases in the stock of physical capital, while an important source of economic growth, could not account for all of the observed long-term growth in the U.S. Consideration of the increases that had taken place in schooling helped to account for observed growth.

Investment in human capital may occur in structured settings (formal education, formal on-the-job training programs) or in unstructured fashion (e.g., learning-by-doing). Such investment, like investment in physical capital, contributes directly to increased productivity of workers. In addition, human capital investment serves as a stimulus to technological change.

Technological advances make important contributions to increasing productivity. They may be reflected in changes in the quality of human capital, but they are most evident as embodied in physical capital. Hence, it is not simply the quantity of physical capital per worker that determines productivity, but also the quality (vintage).

Growth accounting

The preceding discussion identifies the key sources of economic growth, but we are also interested in the relative importance of each of these factors (physical and human capital and technological change). Growth accounting seeks to assess the contribution of each source to overall growth.

Productivity is a key concept in considering economic growth. Parkin defines productivity as real GDP per hour of work. However, with a little bit of reflection two things should be clear:

1. Parkin's definition is in fact a definition of labor productivity, and indeed we might alternatively think about productivity in terms of other inputs into the production process (e.g., capital productivity -- real GDP per unit of capital, or land productivity -- real GDP per acre); and

2. the productivity of labor will depend not only on the (human capital) characteristics of that labor, but also on the quantity and quality of capital and perhaps other factors of production that the labor has available to work with in producing goods and services.

The productivity function, which shows the relationship between labor productivity and the amount of capital per hour of work under given technological conditions, is the basis for doing growth accounting. This allows us to divide growth in productivity into two components, each of which we can measure: growth in capital per hour of labor, and technological change (which incorporates human capital as well).

Productivity (real GDP per hour of work) rises as capital per hour of work increases, but at a decreasing rate. This reflects the law of diminishing marginal returns, which states

that as more and more of a variable factor of production (input) is added to one or more fixed factors, there will be increments to output but eventually those increments will get progressively smaller.

As noted above, the productivity function is drawn with technology held constant. Clearly, however, technological improvement can be easily represented on the diagram -- it will

correspond to an upward shifting of the productivity function.

That is, an improvement in technology means that for a given amount of capital per hour of work there will be a higher value of productivity -- real GDP per hour of work (see Parkin's Fig. 9.5).

The last piece of the puzzle required to do growth accounting is the incremental capital-output ratio, or what Parkin refers to as the one third rule. We can treat this as an empirical regularity: on average, with no change in technology, increments in capital per hour of work of three percent are required to increase output per hour of work by one percent.

Beginning with this ratio as our starting point, then, we can allocate growth between increased physical capital (which moves us along the productivity function) and improvements in technology (which raise the productivity function).

For example, suppose that between 1990 and 1996 there was an increase of 12 percent in the capital stock per hour of work and a corresponding increase in real GDP per hour of work of 7 percent. Using the one third rule, we would attribute 4 of the 7 points of growth to the increased capital stock (4 = one third of 12), and the remaining three percentage points would be attributed to improvements in technology (residual approach).

Note how this would be represented graphically:

Applying growth accounting to the U.S. economy

Parkin uses this growth accounting framework to analyze real GDP growth since 1960. He looks at three periods: the rapid growth from 1960-73, the productivity slowdown from 1973-83, and the following period from 1983-94. His results are summarized in the table below and in his Figure 9.7.