Economic growth can either be achieved by increasing the factor inputs to production, such as capital and labor, or by increasing productivity (i.e., the efficiency by which factors of production are used to generate economic output). Without productivity growth, long-run output growth cannot be maintained with limited or depletable resource inputs; as a result, complex societies become increasingly vulnerable (Tainter, 1988). Changes between inputs and outputs are usually analyzed by drawing upon the production function approach pioneered by Tinbergen (1942) and Solow (1957). Yet, empirical analyses (e.g., Denison, 1962, 1985) quickly identified that quality and composition of factor inputs are more important in explaining long-run output growth than merely the quantitative growth in available factor inputs. For instance, at first sight population growth might be considered as central for economic growth, because it increases the labor force. Upon closer examination, however, institutional and social factors that govern working-time regulation, female workforce participation, and above all the qualification of the workforce (education) have been more important determinants of long-run economic growth (Denison, 1962, 1985) than simple growth in the numbers of the potential workforce (usually calculated as the population in the age bracket 15 to 65 years). Another puzzling finding of Solow (1957) is that, even when changes in quality and composition of factors of production are accounted for, increases in per capita economic output (productivity) remain largely unexplained, a "residual" in the analysis remains unclear (for a review, see Griliches, 1996). The "residual," is usually ascribed to "advances in knowledge and technology" which, unlike capital and labor, cannot be measured directly. However, it might also be the result of other influences, which potentially include growing contributions to the economy by non-market or under-priced natural resources. Thus, considerable measurement and interpretative uncertainties remain in the explanation of productivity growth.
New approaches and models extended the neoclassic growth model (e.g., Romer, 1986; Lucas, 1988; Grossmann and Helpman, 1991, 1993). In these, increases in human capital through education and the importance of technological innovation via directed activity (research and development (R&D)) complement more traditional approaches, which represents a return to the earlier work of Schumpeter (1943), Kuznets (1958), Nelson et al. (1967), and Landes (1969).
Neoclassic economic growth theory embraces as a general principle the notion that long-term per capita income growth rate is independent of population growth rate. Thus, a rapidly growing population should not necessarily slow down a countries' economic development. Blanchet (1991) summarizes the country-level data. Prior to 1980, the overwhelming majority of studies showed no significant correlation between population growth and economic growth (National Research Council, 1986). Recent correlation studies, however, suggest a statistically significant, but weak, inverse relationship for the 1970s and 1980s, despite no correlation being established previously (Blanchet, 1991). As noted in Section 3.2, the reverse effect of income growth on demographics is much clearer.
Population aging is another consideration advanced as having significant influence on economic growth rates. Reductions in workforce availability and excessive social security and pension expenditures are cited as possible drivers. Section 3.2 above concluded that evidence for a strong negative impact is rather elusive. Two additional points deserve consideration. First, population aging is not necessarily the best indicator for workforce availability, because while the percentage of the elderly, in particular those of retirement age, increases, the proportion of younger people (of pre-work or -career age) decreases. As a result, the percentage of the working age population (age 15 to 65 years) in the total population changes less dramatically, even in scenarios of pronounced aging. For instance, in the IIASA low population scenario (7 billion world population by 2100) discussed in Section 3.2, the percentage of age categories 15 to 65 years changes from 62% in 1995 to 54% by 2100. This percentage falls to 48% in the regions with the highest population aging (Lutz et al., 1996).
A second point is that these demographic variables only indicate potential workforce numbers. Actual gainfully employed workforce numbers are influenced by additional important variables - unemployment levels, female workforce participation rates, and finally working time. The importance of these variables can be illustrated by a few statistics. Currently, about 40 million people are unemployed in the OECD countries (UNDP, 1997). The female workforce participation ratios vary enormously, from about 10% to 48% of the workforce (as in Saudi Arabia and Sweden, respectively; UNDP, 1997), and have been changing dramatically over time. For the US, for instance, female workforce participation rates increased from 17% in 1890 (US DOC, 1975) to 45% in 1990 (UNDP, 1997). Similar dramatic long-term changes have occurred in the number of working hours in all industrial countries. Compared to the mid-19th century, the number of average working hours has declined from about 3000 to about 1500 (Maddison, 1995; Ausubel and Gr�bler, 1995). However, in most OECD countries the trend in working time reductions has slowed to a halt since the early 1980s (Marchand, 1992).
Thus, unless the rather implausible assumption is made that with population aging all these other important determinants of labor input remain unchanged, the impacts of aging are likely to be compensated by corresponding changes in these variables (e.g. greater female workforce participation, earlier retirement, etc.). Finally, it must be reiterated that qualitative labor force characteristics, most notably education, are a more important determinant for long-run productivity and hence economic growth than mere workforce numbers.
The importance of social and institutional changes to provide conditions that enabled the acceleration of the Industrial Revolution is widely acknowledged (Rosenberg and Birdzell, 1986, 1990). Rostow (1990) and Landes (1969) identify many social and cultural factors in the "preconditions for economic acceleration" and in the process of economic development.
The importance of institutions and stable social environments is also increasingly discussed in the literature concerned with current economic growth (World Bank, 1991, 1998a). Barro (1997), and Barro and Sala-I-Martin (1995) report a statistically significant relationship between rule-of-law and democracy indices with per capita GDP growth. Law enforcement and legal rights are important indicators for human development in their own right, but enforceable legal contracts are equally important for markets to function. Other socio-institutional factors have been identified that are important to productivity and economic growth: education is mentioned above. Income inequality (and resultant social tensions) also appears to correlate negatively with economic development (World Bank, 1998a; Maddison, 1995).
Strong parallels run between social, institutional, and technological changes (Gr�bler, 1998a; OECD, 1998a). In particular, many features common to the processes of evolution in biologic organisms have been found (e.g., Teilhard de Chardin, 1959; Hayek, 1967; Matthews, 1984; Dawkins, 1986; Michaelis, 1997c). Thus, to understand these processes would involve:
Many aspects of the processes of technical change (e.g., its unpredictability and the importance of mechanisms such as path-dependence and "lock-in") also apply to social change.
It is obviously difficult to evaluate the role of social, cultural, and institutional changes in economic and technical development. Whereas the monetary and technological aspects of change are often measurable and can be observed on a relatively "objective" basis, social, cultural, and institutional processes are hard to measure and often subjective. They tend to involve personal interactions among people, sometimes large numbers of people, over long periods.
Nonetheless, these factors must be taken into account in the scenarios. The SRES approach to develop qualitative scenario "storylines" that provide an overall framework and background for quantitative scenario assumptions and model runs can be considered a particularly valuable strength. Storylines allow these issues to be addressed explicitly, even if current knowledge does not allow social, cultural, and institutional factors to be treated in a rigid, quantitative (not to mention deterministic) way.
Other reports in this collection |