RDP 8501: Neoclassical Theory and Australian Business Investment: A Reappraisal 6. Concluding Comments

The model of Section 4 is, to the best of our knowledge, one of the first empirical applications of the theoretical literature on multiple-output growth models.^[37] It is a complete model of investment behaviour, and it integrates production decisions and portfolio decisions. Although there is plenty of room left for improvement,^[38] our empirical results are rather encouraging: the model is well behaved (once expectations are exogenised), the parameter estimates are plausible and in line with prior expectations, and the goodness of fit of the individual equations is highly satisfactory. Yet the approach that we have followed is unmistakenly neoclassical. It thus appears that the Australian facts are not in conflict with neoclassical theory.^[39]

Neoclassical investment theory and the portfolio approach are sometimes viewed as competing theories of investment behaviour.^[40] Our approach shows that this need not be the case. The two approaches merely focus on different aspects of the capital accumulation process (usership and ownership), and they are like the two sides of a coin. In fact, we have argued, there is a third dimension to it, since capital goods must also be produced. Moreover, all three aspects must be considered simultaneously if one wants to account for the endogeneity of the price and the rental price of capital.

The production model of Section 4 could be generalised to allow For more inputs and more outputs. Of particular relevance for an open economy like Australia would be the inclusion of imports and exports. Imports can be viewed as an input to the technology, and exports are an additional output. The treatment of imports as intermediate inputs is analytically convenient,^[41] and it is justified by the fact that Australia imports many non-finished products. Moreover, most imported finished goods are still subject to domestic landing, transportation and retail charges before reaching final demand so that a significant proportion of the final price tag is accounted for by domestic value added. This is true for consumption goods as well as for investment goods. Australia imports many capital goods from overseas, but for analytical purposes one can view these goods as flowing through the domestic production sector and being combined with domestic capital and labour services in the process. The supply of investment goods would therefore become a function of the price of imports, and one would expect an increase in the price of imports to reduce investment, if Australia's output of investment goods is indeed import intensive. Similarly the supply of investment goods becomes a function of export prices, if exports are viewed as an additional output of the production sector.^[42]

The above argument can be linked with the debate on the so-called resources boom.^[43] Australia is a large exporter of mining products. We can safely assume that the capital requirements of the mining sector are relatively high. It then follows that an increase in the price of mining products (and in the price of exports to a large extent) results in an increase in the rental price of capital and consequently in the price of capital goods. This tends to increase the supply of investment goods and imports of machinery. The resources boom can be interpreted in this light: the relative price of exports did go up following sharp increases in world commodity prices in 1979/80, and no doubt even larger increases were anticipated. Australian investment and imports rose, but the trend suddenly reversed itself with the collapse of world commodity prices, and the resources boom failed to materialise, at least for the time being.

Of course the portfolio model of Section 4 could be expanded as well, with the inclusion of liabilities and additional assets. Of special interest would be bank advances and foreign liabilities;^[44] this would enable us to bring financing issues into the analysis. The question of foreign borrowings is of course also closely linked to the debate on the resources boom. Investment in the mining sector taking place to a large extent in anticipation of increased exports, and Australia being a relatively small country, much of the financing must come from overseas. The effects of the resources boom on our integrated model of portfolio and production behaviour are therefore manifold. The portfolio aspects involve increases in holdings of capital largely offset by increased foreign liabilities. The production effects involve increases in the demand for imports and in the output of investment goods resulting over time in a larger input of capital services and additional exports.^[45]

One question that often arises concerns the employment effect of investment. A number of different views have been expressed in this area, with some commentators insisting on the job creation effect of investment and growth, and others more concerned with the job displacement effect of capital accumulation.^[46] By treating employment as exogenous our model can only bring a partial answer to these questions. However, the estimates reported in Table 4 suggest that labour requirements of investment are relatively large, hence an increase in investment will tend to be accompanied by an increase in the (inverse) demand for labour.^[47] Furthermore, the subsequent increase in the stock of capital, other things equal, will increase the marginal product of labour. Of course, one can argue that the entire debate is a red herring. Full employment is consistent with any output mix. What is required for full employment is that the economy operates on the production possibility frontier; this can be achieved even if all output is consumed, that is if investment is nil. Moreoever, one can argue that, in the long run, investment has little to do with economic growth. Elementary growth theory teaches us that the steady-state growth rate is independent of the savings ratio.^[48] Hence the conclusion that taxation can do little to affect investment in a durable way should not cause policy makers any undue distress.