One important class of macroeconomic models comprises the two-sector models that emphasise the distinction between internationally traded and non-traded goods. The literature utilising these models has been mainly theoretical, with the possible exception of the Scandinavian model of inflation. One reason for the lack of empirical work could have been the fact that categories such as ‘tradeables’ and ‘non-tradeables’ do not lend themselves to direct measurement. However, this should not stop us from drawing the macroeconomic implications of such models, and testing them using aggregate data. This paper puts a variant of such models to the test, using aggregate time series for the United Kingdom. The model that is put forward treats the traded goods sector as fundamentally competitive, and the non-traded goods sector as oligopolistic. The focus is on the determination of the relative price of traded to non-traded goods, the equivalent of international competitiveness in these models, and its role in aggregate fluctuations. The model is ‘real’, and refers to the short run. Monetary factors have been de-emphasised. On the other hand, factors such as government expenditure, and the relative price of oil are given a prominent role. The econometric estimates suggest that at a general level the model is quite successful in accounting for the UK business cycle. At a ‘deeper’ level, however, the main problem is the implausibly low estimate of one crucial parameter, namely the share of tradeables in GDP. This takes a rather low value, especially in systems estimation. The point estimates of most of the other parameters are quite plausible and the overidentifying and cross-equations restrictions of the model are not rejected at conventional significance levels.
The Economic Journal