Economic Growth and the Balance-of-Payments Constraint

By way of introduction to the demand orientated theories, it is useful to briefly recapitulate the conclusions of the ‘old’ neoclassical growth theory, pioneered by Solow. The 1970s saw a proliferation of theoretical neoclassical one sector growth models based on the aggregate production function, although a few models were constructed with separate consumption and capital goods sectors. Under the usual neoclassical assumptions there is always the full and efficient utilisation of all resources. That is to say, countries are assumed to exhibit both allocative and producer efficiency. Resources are optimally distributed between firms so that no redistribution could lead to a net increase in output. Firms are also technically efficient and costs are always minimised; there is no X-inefficiency in Leibenstein’s (1966) sense of the term. In the steady state, the growth of the capital stock is equal to the growth of output.

We outline in this chapter the concept of the balance-of-payments equilibrium growth rate, as originally developed by Thirlwall (1979, 1982a), and then show its empirical application to a wide range of developed and less developed countries, drawing also on the recent work of Bairam (1988, 1990). The two original papers by Thirlwall are presented here basically in their original form, but with some modifications.

We showed in the previous chapter that the slow growth of national output (particularly for the UK) has been associated with a very slow rate of export growth relative to other countries. This resulted in a massive decline in the UK’s share of world trade in manufactures from just over 20 per cent in 1950 to just under 8 per cent in 1988. It is difficult to argue that this unfavourable trend has been the result of continually declining price and cost competitiveness because, for long periods over the postwar years, measures of price and cost competitiveness moved in Britain’s favour. Figure 4.1 shows movements in relative producer prices and unit labour costs from 1963 to 1989.

As we saw in Chapter 2, the post-Keynesian view of economic growth denies that the postwar economic performance of the majority of the advanced countries has been seriously constrained by the growth of factor supplies. Even during the ‘long boom’ of 1950–73, when the growth of output of the advanced countries in aggregate was double that of the historic norm, labour shortages were never a limiting factor. There was either sufficient disguised unemployment in the non-manufacturing sectors or enough immigration to satisfy the demand for labour (Kindleberger, 1967). The rate of capital accumulation is never a long-run constraint on economic growth as investment is as much a result of the expansion of output as its cause (Kaldor, 1970). The evidence in support of these conditions has been summarised in Chapter 2.

There are two main reasons for believing in the importance of exportled growth; one uncontroversial, the other more contentious. The first is that export growth can lift a balance-of-payments constraint on demand and therefore permit faster growth if factor supplies are available to be utilised. The second is that export growth may create a virtuous circle of growth by virtue of the link between output growth and productivity growth. A number of models of this genre have appeared in the literature in recent years (e.g. Lamfalussy, 1963; Beckerman, 1962; Kaldor, 1970; Dixon and Thirlwall, 1975a) and they are surveyed later. There is a problem with these models, however, and that is they do not incorporate an explicit balance-of-payments equilibrium condition or constraint, which means that the equilibrium growth rate specified may be inconsistent with the long-run requirement of payments balance.² The implicit assumption seems to be that provided it is exports that are the engine of growth, as distinct from domestic demand, the balance of payments will look after itself. Indeed, it is assumed in some models that the initial export growth and trade surplus generates such favourable responses in the economy that the balance-of-payments surplus actually grows.

In this chapter we use the ideas of Kaldor, already introduced in Chapter 7, to consider the role of the Verdoorn relationship in a model of ‘circular and cumulative causation’. Kaldor was a longstanding critic of the application of neoclassical modes of thought to the analysis of economic growth and development. He supported Myrdal’s (1957) notion of vicious circles of success and failure (the principle of ‘cumulative causation’), and attacked the predictions of neoclassical theory that regional (national) growth rate differences would tend to narrow with trade and the free mobility of the factors of production. The essence of the argument is that once a region gains a growth advantage it will tend to sustain that advantage through the process of increasing returns that growth itself induces — the Verdoorn effect. Kaldor first articulated his theory in a purely verbal way in a lecture given to the Scottish Economic Society in 1970 (Ka’dor, 1970). We discussed the structure of the model in the previous chapter. Here we shall use the model to consider such questions as: the role of the Verdoorn effect in contributing to regional growth rate differences; whether regional growth rate differences will tend to narrow or diverge through time; and how policies of regional ‘devaluation’ can raise a region’s growth rate (if at all).

In this chapter we consider some of the applications to which our balance-of-payments constrained growth model can be put — other than the analysis of growth within or between countries and regions. First, we use the model to consider the effects on growth of the removal of tariff barriers within Customs Unions. Secondly, we show that in a closed economy model of industry and agriculture, the equilibrium level of industrial output is determined by the Harrod trade multiplier result. Thirdly, we see how the model can be used for forecasting the rate of growth of output consistent with balance-of-payments equilibrium given assumptions about the growth of world income and changes in the nominal and/or real exchange rate.

Up until now, the evidence for a country being balance-of-payments constrained has been obtained by examining the extent to which the balance-of-payments equilibrium growth rate approximates to the actual growth rate of that country. Using data from highly aggregated export and import demand functions, we concluded in Chapter 3 that for the vast majority of countries there was no significant difference between the two growth rates. From this we deduced that relative prices played an insignificant role in determining the growth of imports and exports (or, more implausibly, any effect was just compensated by offsetting capital flows). Additional support for our argument is provided by the fact that the relative price terms in the estimation of import and export demand functions are either small or statistically insignificant, or both (e.g. Houthakker and Magee, 1969). Nevertheless, while these results are instructive, the estimation of such highly aggregated export and import demand functions has been criticised because of, for example, the unreliability of the data and the problem of simultaneity (Morgan, 1970). Moreover, it is not possible to test directly the significance of the inflationary feedback from a currency depreciation to domestic costs (i.e. the real wage resistance hypothesis). The estimates of the price elasticities are also insufficiently precise to permit us to determine with any degree of confidence to what extent (if at all) the Marshall-Lerner conditions are satisfied. On the other hand, as we have seen, changes in relative prices are a relatively unimportant determinant of changes in export and import market shares, compared with disparities in non-price competitiveness.