6  Long run economic growth

Economic growth depends on range of factors, including economic policies, investment in human capital, and innovation; these fall largely outside the scope of this paper. However population ageing may affect economic growth through its potential effects on saving, investment, the stock of capital, and on labour.

6.1  Investment and capital

Most economists agree that capital accumulation affects long-run economic growth. There is considerable disagreement, however, over what kinds of capital investment are most desirable for enhancing economic growth and over what the determinants of capital accumulation are.

One determinant of capital accumulation that is closely associated with population ageing is domestic savings. One school of thought suggests that low domestic savings levels induce high real interest rates[27] and crowd out private investment in physical capital, hence constraining economic growth (Crocombe, et al. 1991). However, this sort of argument does not give full consideration to New Zealand’s position in the world as a small open economy. At the present time, studies of economic growth in small open economies have failed to provide evidence of a link between domestic savings, capital accumulation, and economic growth[28].

The types of investment favoured by ageing populations may adversely affect economic growth. In the previous section we noted the argument that people in old age prefer to invest in fixed return assets such as government debt. This has important implications for growth because it is variable return investments such as investment in equipment that yields the greatest benefit in terms of economic growth (Ahn and Hemmings 2000) (De Long and Summers 1991).

Future investment will also be affected by consumption demand from within the domestic economy and as such growth may be adversely affected by changing consumption patterns in an ageing economy. Arguably the elderly demand fewer manufactured goods than those in the working age population and as such we may see an increase in the proportion of GDP stemming from services. Productivity growth in the service sector is typically lower than that for say manufacturing, which benefits more readily from advances in technology. Thus we may find that growth rates will decline from a decline in total factor productivity stemming from changes in the sectoral output mix induced by population ageing.

Conversely, there is potential for elderly consumption patterns in ageing countries to enhance long-run growth in New Zealand. An increase in demand from developed countries for consumables such as commodities could increase trade volumes or prices and hence increase per capita incomes.

Also, population ageing may increase long-run growth in per capita incomes through a capital-concentration effect. In countries such as New Zealand, where we expect a reduction in the working age population, the level of capital per worker should rise. It is not clear precisely what the implications of this will be, but it may result in increased labour productivity (Ahn and Hemmings 2000)[29].

6.2  Labour

Future changes to labour supply and to labour productivity could have a large impact on New Zealand’s growth path. The projected reduction in the labour force means that there will be a reduction in labour inputs such that, all other things equal, there will be a reduction in per capita GDP.

Using a theoretical growth accounting model we can analyse, albeit rudimentarily, the effects on potential GDP growth rates of a reduction in the labour force. One such model, based on an aggregate production function for the economy, is:

GDP growth rate = (1-α)(labour growth rate)+α(capital growth rate) + total factor productivity

where α is the proportion of GDP that constitutes returns to capital.

In 1996 the Treasury estimated a potential growth rate of 3.3 percent for the New Zealand economy using this simple formula (The Treasury 1996). This was based on estimated annual growth in the capital stock of 2.4 percent, growth in labour supply of 1.4 percent, returns to capital of 0.4, and total factor productivity of 1.5 percent.[30] All other things equal, a reduction in the labour force growth rate of 25 percent would reduce the potential GDP growth rate to 3.1 percent. If the labour force were to actually shrink in the longer term, the potential growth rate would, according to this approach, reduce further. Any such reduction in GDP growth rates, with its commensurate impact on living standards could be offset, however, by an increase in productivity.

Historically, labour productivity has grown at a fairly conservative annual average rate of 1.35 percent between 1960 and 1998 (see Figure 3). In order to sustain future rates of economic growth at levels which offset the effect of rising old-age dependency, productivity growth would need to accelerate.

The creation of a highly skilled workforce through education is one area that may enable labour productivity to keep pace with the declining labour force. Indeed Lucas’s model of endogenous growth suggests that investment in human capital through effective education is fundamental to long run economic growth. If this is the case then it poses very important questions as to how our education system will perform over the next fifty years.

Source: Dalziel and Lattimore 1999 p. 65

The market for education may be affected by population age structural change. An increasing capital to labour ratio indicates that wages will rise in future relative to returns to capital. This has the propensity to affect individual choices to invest in higher education depending on changes to the returns to unskilled labour relative to the returns to skilled labour.

The cost of education may be seen as both the cost of fees, including the cost of capital associated with either debt or equity financed fee payment, and income foregone whilst studying. Thus an increase in the returns to unskilled labour implies an increase in the cost of education whilst a reduction in the cost of capital implies a reduction in the cost of education. Conversely the return on education, seen as the extra income an individual can expect to receive over their lifetime, is dependent on the returns to skilled labour[31]. Therefore if the returns to skilled labour rise relative to the returns to unskilled labour, then we may expect increased investment in education in the economy.

Welch (1979), in a study of the earnings of baby boomers in the United States, finds evidence that an increase in a cohort’s size reduces a cohort’s earnings. Furthermore the study found that as cohort size increased, the differential in earnings between educated and non-educated workers reduced. This suggests that there may indeed be an increase in the returns to skilled labour relative to unskilled, as the working age population shrinks. The corollary of this is that there may be greater investment in education contributing to higher levels of human capital and potentially higher productivity and output growth in the economy.