Industrial development and economic growth: Implications for poverty reduction and income

2. The role of structural change in economic growth
The current understanding of economic growth is largely based on the neo-
classical growth model developed by Robert Solow (1956). In the Solow
model, capital accumulation is a major factor contributing to economic
growth. Productivity growth – measured as an increase in output per worker
– results from increases in the amount of capital per worker, or capital accu-
mulation (e.g. Fagerberg 1994). Capital deepening will continue until the
* Pellervo Economic Research Institute, Helsinki, Finland.

---

296
Industrial Development for the 21st Century
economy reaches its steady state – a point at which net investments grow at
the same rate as the labour force and the capital-labour ratio remains con-
stant. The further the economy is below its steady state, the faster it should
grow (see e.g. Jones 1998). In the steady state, all per capita income growth
is due to exogenous technological change. The rate of technological process
is assumed to be constant and not impacted by economic incentives. Several
authors have found that capital and labour actually explain only a fraction of
output growth and that allowing for the quality of the labour force (human
capital) only partially reduces the unexplained growth – or Solow residual.
Endogenous growth theory, initiated by Romer (1986, 1990) and Lucas
(1988), focuses on explaining the Solow residual. Technological change
becomes endogenous to the model and is a result of the allocative choices of
economic agents (see Aghion and Howitt 1998, Veloso and Soto 2001).
Technological progress is driven by R&D activities which in turn are fuelled
by private firms’ aim to profit from inventions. Unlike other production
inputs, ideas and knowledge are nonrivalrous (see Romer 1990). Moreover,
new knowledge can augment the productivity of existing knowledge, yield-
ing increasing returns to scale. Because of this, the marginal productivity of
capital does not decline with increasing GDP per capita, and incomes need
not converge across countries. 
Technological change and innovations are essential sources of structural
change. In Schumpeter’s view, innovations lead to “creative destruction”, a
process whereby sectors and firms associated with old technologies decline
and new sectors and firms emerge and grow (see Verspagen, 2000). More
productive and profitable sectors and firms displace less productive and less
profitable ones and aggregate productivity in the economy increases.
Technological change is thus at the very centre of modern economic growth.
Based on the observation that, beginning with the Industrial Revolution,
technological change took place mainly in the manufacturing sector, authors
like Kaldor (1970) and Cornwall (1977) have asserted that the expansion of
this sector is a driving force for economic growth (see Verspagen, 2000).
Moreover, Cornwall (1976, 1977) saw technological change in certain man-
ufacturing sectors as a driving force for productivity growth in several other
sectors.
1
Syrquin (1986) observes that, when overall growth accelerates, man-
ufacturing typically leads the way and grows faster than other sectors. At low
income levels, the share of manufacturing in GDP is, however, low and its
immediate contribution to aggregate growth minor. When manufacturing
increases its output share – often as a response to changes in domestic
demand and in comparative advantage – faster sectoral growth noticeably
raises the aggregate growth rates of output and labour productivity. 
In developed countries, research and development (R&D) activities are
the main driver of technological change. This is not, however, the only
mechanism of technological change. Firms and individual employees learn
by doing, increasing output and productivity even if technology or inputs

---

297
Industrial development and economic growth
remain unchanged (see e.g. Arrow 1962). As R&D activities in developing
countries are relatively limited and countries are far from the technological
frontier, international technology diffusion is essential for productivity
growth. International economic relations, especially international trade but
also foreign direct investment, are important channels of technology transfer
and increased productivity growth. However, technology diffusion can only
be efficient if the level of human resources is high enough, incentives for
technological improvement are strong, and institutions are relatively well-
functioning.
One of the driving forces for structural change is the change in domes-
tic and international demand. At relatively low income levels, individuals
spend a significant part of their income on food. As income rises, this share
tends to decline, whereas demand for manufactures rises. Similarly, as
income rises further, demand for manufactures increases at diminishing rates,
whereas demand for services rises rapidly. Changes in demand will also
change sectoral employment and output shares and impact the economy’s
labour productivity. Furthermore, trade has an impact on countries’ special-
ization patterns and on the rate of industrialization or structural change
within industries. Under an open trade regime, countries tend to specialize
in the production of commodities for which they have a comparative advan-
tage and import commodities which are relatively expensive to produce
domestically. Trade openness is also likely to bring foreign investment into
the country. This is often vital, and especially so at early stages of develop-
ment. It is also likely to increase productivity as domestic companies are fac-
ing external competition. 
However, the composition of foreign trade matters as well as the open-
ness of trade (e.g. Amable, 2000; also, Rodrik in this volume). Moreover, spe-
cialization in itself does not necessarily lead to higher growth rates. This is
most evident in the case of developing countries dependent on exports of pri-
mary products. As real international prices of non-oil commodities have
trended downward over time and are subject to sizeable short-term fluctua-
tions, specialization in primary production seldom promotes sustained eco-
nomic growth.