Microfoundations of Economic Growth

Most growth analysis has been primarily a macroeconomic subject with particular emphasis on contribution of capital, education adjusted labor, and total factor productivity to output growth (see Collins and Bosworth 1996, Hu and Khan, 1997, Sarel 1997, Sala-i-Martin 2000, Hall and Jones, 1999, Easterly and Levine 2001). Importance of macroeconomic policies as represented by budget deficits, exchange rate premia, inflation, trade openness and inflow of foreign Investment etc are tagged on in the growth analysis at a macroeconomic level. A few studies have invoked ethnic differences and other exogenous factors to understand cross country differences in total productivity growth and per capita incomes. 

In trying to understand the rapid output growth of East Asian ‘miracle’ countries, Krugman (1994), Young (1995), and others were engaged in an interesting debate on whether capital accumulation or total factor productivity growth best explained the high and sustained output growth of these countries. Their conclusion that capital accumulation was most important was based on macroeconomic data analysis in a factors of production approach to sources of growth. Others have found that the growth of output is strongly correlated with productivity growth in developed and developing economies as reported by Kehoe and Prescott (2002) and Solimano and Soto (2004), and this co-movement appears to be stronger the longer is the time period considered.

These studies provided varying conclusions. Mankiw, Romer and Weil (1992) found that physical and human capital accounted for 80 percent of variation in percapita income across countries, while Klenow and Rodriquez-Clare (1997) emphasize productivity growth as accounting for 90 percent of income per capita variation, while Easterly and Levine (2001) emphatically state that it is not factor accumulation.

Easterly and Levine (2001) document five stylized facts of economic growth – residual or TFP growth rather that factor accumulation accounts for most of the income and growth differences across nations, and that incomes diverges over the long run. Factor accumulation is persistent while growth is not persistent and the growth path of countries exhibits remarkable variation across countries. Economic activity is highly concentrated, with all factors of production flowing to the richest areas. National policies closely associated with long-run economic growth rates. These stylized facts are inconsistent with models with diminishing returns, constant returns to scale, some fixed factor of production, and that highlight the role of factor accumulation.

While growth accounting and growth regressions are important tools, the macroeconomic approach suffers from many shortcomings. In particular, total factor productivity is measured as a residual and as growth is determined by many factors, TFP should not be treated as a measure of technical change. Moreover, accounting framework should not be used for causes of growth as many factors and endogenous to the growth process.

Increasingly researchers are looking at micro-level evidence on total factor productivity at the firm level and other factors (e.g. openness to trade, Schumpeterian creative destruction etc) as determinants of TFP. Apart from capital and labor, variables such as exports, import content, foreign direct investment, etc are introduced to estimate their contribution to TFP.

The growth analysis would have been even more interesting if these researchers looked at microeconomic evidence for the same set of countries for the same time period to re-confirm their results. This would have also enabled them to understand the differences in TFP in each of the East Asian countries and its impact on growth. It is possible that some countries are propelled by physical and human capital while some by TFP. For example, Collins and Bosworth (2003) analyzed data from 84 countries for the period 1960 to 2000 and found that capital accumulation and productivity growth both contribute to output growth. At the global level, they found that the contribution of capital and productivity is roughly equal but there have been substantial variations in their relative contribution across countries and time. 

Thus to understand aggregate growth, we must understand sources of growth (capital accumulation and efficiency) at all levels – firm, industry and the macroeconomy. Two sources of productivity gains drive aggregate (industry and national) efficiency. First, within-firms efficiency gains are possible because of adoption of new and better technologies. Second, the Schumpeterian creative destruction process is possible because of reallocation of resources from less productive firms to more productive ones and through entry and exit of firms. A number of papers report such dynamics amongst firms in accounting for efficiency gains. For example, using plant-level data on Chilean manufacturing firms for the 1980-2001 period, Bergoeing and Rapetto (2006) estimate that reallocation accounted for almost all of total efficiency gains in Chile during that period.

Foster, Haltiwanger, and Krizan (1998) analyze four elements in decomposing productivity growth: (i) a within-plant effect, given by incumbents’ productivity growth weighted by initial output shares; (ii) a between-plant effect, that captures the gains in aggregate productivity coming from the expanding market share of high productivity plants relative to the initial aggregate productivity level; (iii) an entry effect which is the sum of the differences between each entering plant’s productivity and initial aggregate productivity, weighted by its market share; and (iv) an exit effect given by the sum of the differences between each exiting plant’s productivity and initial aggregate productivity, weighted by its market share.

In this paper, Foster et al find that (i) the contribution of reallocation of outputs and inputs from less productive to more productive establishments plays a significant role in accounting for aggregate productivity growth; (ii) for the selected service industries considered, the contribution of net entry (more productive entering establishments displacing less productive exiting establishments) is dominant; (iii) the contribution of net entry to aggregate productivity growth is disproportionate and is increasing in the horizon over which the changes are measured since longer horizon yields greater differentials from selection and learning effects; (iv) the contribution of reallocation to aggregate productivity growth varies over time (e.g. is cyclically sensitive) and industries and is somewhat sensitive to subtle differences in measurement and decomposition methodologies.

The challenge is to estimate and analyze the determinants of output growth both at the micro (firm), meso (industry) and macro (national) levels in a sample of countries. The purpose is to understand the micro-foundations of macroeconomic growth, including the issue of whether there are significant differences in TFP across firms, countries, and regions. One approach is to estimate the sources of growth in the standard Cobb-Douglas production function, and the aim is to estimate the parameters pertaining to the contribution of physical capital, labor and total factor productivity (TFP) to output growth at the firm and national levels. The stylized facts from micro-studies are likely to be different from macro-data based stylized facts of Easterly and Levine. Specifically, we could expect that:

(1) some countries are driven by TFP growth others by capital accumulation and this may depend on stages of development and change over time;
(2) capital accumulation may be important in some industries (say chemical industry) while TFP growth could be important in textiles and garments;
(3) TFP growth is important in countries where manufacturing sector is large (as innovation takes place more in manufacturing and less in agriculture or services);
(4) firms that survive over long run are firms that adopt and adapt new technologies and continuously innovate.