Most economists would agree that rapid manufacturing-led growth is what India needs and must seek to promote, says Ajit K Ghose.
The prime minister's 'Make in India' campaign signals a key objective of the present government's economic policy: promotion of rapid manufacturing-led growth.
This means not just rapid growth of manufacturing, but also a lead role for manufacturing in India's growth process.
It does not call for discouragement or lowering of services growth; it calls for services growth to be pulled by manufacturing growth and not vice versa.
Most economists would agree that rapid manufacturing-led growth is what India needs and must seek to promote.
Though the country has had rapid services-led growth for a decade or so, economists have tended to view this as fortuitous and temporary.
Historically, manufacturing has led the growth process at early stages of development not just in today's developed countries of Europe and North America but also in late-developers of Asia such as Japan, South Korea and Taiwan, and most recently in the People's Republic of China.
And no country in the world has achieved high-income status without developing manufacturing to a point where it accounts for a high share (around 30 per cent) of GDP.
The economic logic of this historically observed pattern of development is also well understood.
On the supply side, manufacturing production is characterised by increasing returns to scale and rapid productivity growth.
Manufacturing can also stimulate non-manufacturing activities significantly.
On the demand side, the income elasticity of demand is higher for manufactures than for non-manufactures at relatively low levels of per capita income. And manufactures are eminently tradable, so that external demand can play an important role in supporting manufacturing growth.
In the case of India, there also are two specific reasons why manufacturing-led growth must be the order of the day.
The first is that economic growth cannot in fact be rapid if it is not manufacturing-led.
The spell of rapid services-led growth has come to an end and the reason why it has come to an end is also the reason why it cannot be revived.
Services-led growth ended up generating a major macroeconomic imbalance: while manufactures figured much more prominently in domestic consumption and investment, the economy produced far more services than manufactures.
The sustainability of services-led growth, therefore, required services exports to finance imports of manufactures to meet the shortfall in domestic production.
This was quite impossible, and the inevitable balance of payment difficulties ultimately halted the growth process itself.
Any attempt to revive services-led growth will fail for the same reason; severe balance of payments difficulties will re-emerge.
It is manufacturing-led growth that can begin to restore the balance between domestic absorption and domestic production so that the external balance remains manageable.
The second reason is that India has a mass of surplus low-skilled labour that must be moved to productive employment in non-agriculture.
This is the employment challenge, which must be met if economic growth is to be inclusive.
Both historical experience and India's own past experience suggest that only manufacturing is capable of absorbing this mass of low-skilled labour into productive employment.
Services in India have been significantly less labour-intensive than manufacturing; in 2011-12, services accounted for 55 per cent of GDP but only 32 per cent of employment, while manufacturing accounted for 15 per cent of GDP and 13 per cent of employment.
More relevant is the fact that organised manufacturing has been far more employment-intensive than organised services; in 2011-12, organised manufacturing accounted for 10 per cent of output and five per cent of total employment in the economy, while organised services accounted for 27 per cent of output and just nine per cent of employment.
Moreover, organised services employed mainly high-skilled (those with above-higher-secondary education) labour, while organised manufacturing employed mainly low- and medium-skilled labour.
In 2011-12, the share of high-skilled workers of all those employed was around 66 per cent in organised services and around 32 per cent in organised manufacturing.
This view runs contrary to that expressed in the latest, 2014-15 Economic Survey (Chapter 7), and a comment is warranted.
In the Survey, organised (or registered) manufacturing is shown to be as skill-intensive as services in the aggregate (i.e., organised and unorganised services taken together) and far less skill-intensive than those identified as the dynamic services (again organised and unorganised taken together) such as "financial services and insurance" and "real estate, renting and business services".
Yet, in conclusion, it is argued, incorrectly in my view, that organised manufacturing and the dynamic services do not differ that much in terms of skill-intensity.
It is often argued, correctly, that manufacturing cannot possibly play the same role in development in the twenty-first century as it did in earlier times.
The peak share of manufacturing in GDP in China (32.5 per cent in 2005) was not all that different from that in Japan (36 per cent in 1970); but the peak share was reached at a much higher level of per capita GDP in Japan than in China.
It is also argued, again correctly, that technological change has progressively lowered the capacity of manufacturing to pull low-skilled labour out of agriculture and other low-productivity activities.
The peak share of manufacturing in GDP, for example, was associated with a much higher share of manufacturing in employment in Japan (27 per cent in 1970) than in China (16 per cent in 2005).
But these arguments do not tell us that India should not go for manufacturing-led growth.
They do not tell us that it would not be worthwhile to increase the share of organised manufacturing in GDP from 10 per cent to 30 per cent and thus to increase its share in employment from five per cent to 15 per cent.
What they do tell us is that India should not expect manufacturing-led growth to quickly transform it into a developed country or to quickly absorb the mass of low-skilled labour into productive employment.
But high manufacturing growth still holds the best prospect for maximising both economic growth and growth of productive employment of low-skilled labour.
High manufacturing growth, moreover, should be associated with high growth of construction and of certain services such as transportation and trade, which too are intensive in low-skilled labour.
Ajit K Ghose is Honorary Professor at the Institute for Human Development, New Delhi