lower levels of per capita income than observed historically in today’s devel-
oped countries. In the past this historical turning point occurred at per capita
incomes of US$ 10,000; it is now estimated to take place at $3,000 in some
countries (Dasgupta and Singh, 2006). Table 8.2 shows this process has
occurred in both Ghana and Botswana since 2000.
A key argument of the structuralists has been the link between manufactur-
ing growth and export potential. For example, even in the 2000s, despite the
manufacturing sector falling to 15 per cent of GDP in the UK it still accounted
for 60 per cent of total exports. It has been long recognized that services are
less tradable than manufactured goods, since they have to be consumed at the
place and moment of their creation. A haircut, for example, cannot be
produced, stored and then traded at long distance. Technological change,
however, has now facilitated international trade in services such as education,
law, accounting and back-office functions. Trade barriers in services generally
do not take the form of import tariffs, but quantitative restrictions on interna-
tional trade in services are often applied to providers of services rather than
services per se. Doctors, for example, are not generally free to practise their
profession in different countries. Service industries are also sometimes
supported through explicit or implicit subsidies especially in construction,
communications and transport (Banga, 2005). This kind of policy constraint is
tackled by the WTO who in recent years have turned their attention to the liber-
alization of trade in services.
Kaldor placed manufacturing at the centre of his analysis, in terms of its
effects in driving both GDP and productivity growth. The case of India illus-
trates how things have changed. Technological change and greater tradeability
of services means rapid productivity growth in the sector is now possible. TFP
growth in the service sector was 3.9 per cent per annum in India between 1993
and 2004 (Bosworth and Collins, 2008:54). There is strong evidence that serv-
ice-sector growth has had a positive impact on productivity growth in Indian
manufacturing (Banga and Goldar, 2004:13). There are also indications that
agriculture and industry in India have become more intensive in the use of
inputs from the service sector (Sastry et al.,2003). Together these findings
imply that service-sector growth can drive growth elsewhere in the economy.
The conventional theory is that deindustrialization in developed countries is
caused by the internal factors (demand and productivity effects) discussed in
this chapter. Some argue that cheap manufactured imports from developing
countries displacing domestic production and employment are responsible.
This suggests that, rather than being a natural process, deindustrialization can
be facilitated by too much openness to trade, and trade protection may be
necessary to preserve the dynamic industrial heart of an economy. Living stan-
dards and economic dynamism may then really be threatened by Chinese
manufactured exports. Rowthorn and Ramaswamy (1999) used annual data for
18 industrial countries between 1963 and 1994 to test the impact of interna-
tional trade on economic structure. They disagreed with the more pessimistic
view and found that among developed countries, of a total of 350 million jobs,
180 Patterns and Determinants of Economic Growth