In a column earlier this year1 I explored the promise and the potential pitfalls of the government's 'Make in India' programme. There I noted a possible tension between the government's goals for manufacturing value-added and its goals for manufacturing employment. While the government's ambitions are focused on improving the share of manufacturing in Gross Domestic Product (GDP) through improvements on the supply side, I noted that there was also a need for clear thinking on the demand, both domestic and foreign, for India's manufactured production.
I would like to return to both these themes, stimulated by work on the US economy that has recently come to my attention2. I do so recognising that there are important differences between the circumstances of a large rich economy such as the United States and a much poorer India. But there are also some intrinsic characteristics of manufactured goods that the Indian debate needs to take more into account than has been the case.
While the US debate on manufacturing is about many things, including stagnant wages, income inequality, offshoring, the impact of trade and of trade pacts, there are two characteristics of interest for India. The first is the ability of manufacturing to sustain increases in productivity over the very long run. The second is that demand for manufactured goods by final consumers is relatively price inelastic.
In the US these two features have interacted to reduce both the relative price of manufactured goods (when compared with services) and the share of final consumption expenditure allocated to manufactures. The outcome is that US manufacturing is following the path previously trodden by agriculture: a highly productive sector limited by the demand for its output, and employing a diminishing fraction of the labour force.
Before turning to the implications of these trends for Indian policy, I would like to highlight some of the remarkable stylised facts embedded in the Lawrence-Edwards paper. As always, the value of US illustrations is the depth and quality of the data over the long sweep, and the fact that the US is a relatively undistorted market economy allowing underlying trends to express themselves clearly.
Measured in constant 2005 dollars, the share of manufacturing in US output (i.e. value added) in 2005 was the same as in 1947, at around 17.5 per cent of GDP. While 'industry' is a broader concept than manufacturing, US 'deindustrialisation' is an artefact of relative price trends and not of physical output, even acknowledging the massive changes in the types of manufactured goods produced by the US economy over these sixty years. As the share of physical output has remained level, the share of employment in manufacturing has declined steadily, from 27.7 per cent of US nonfarm employment in 1961 to 8.9 per cent in 2010.
Lawrence and Edwards note, "most remarkable is the persistence and stability of this declining relationship, irrespective of the changing developments in international trade flows, the size of the trade deficit and other factors…a forecaster in 1980, knowing that in 2010 total nonfarm employment would be 129 million would have predicted manufacturing employment in 2010 to be within 25,000 of its level of 11.534 million".
Improvement in labour productivity in manufacturing is thus sustained and inexorable: between 1960 and 2007, labour productivity in manufacturing increased by 1.5 per cent per year more rapidly than labour productivity in the economy as a whole, a substantial difference to be sustained over such a long period. This in turn reflected relentless competitive pressure in manufacturing, both domestic and through trade, but also the impact of automation and of research and development: the sector, accounts for the vast bulk of research and development spending.
In light of the scale of the US and trade exposure of the manufacturing sector, higher sustained productivity growth in manufacturing eventually leads to a fall in the relative price of manufactured goods versus services. This then brings us to the demand for manufactured goods. If demand is fixed, then improvement in labour productivity clearly results in declining manufacturing employment. The one surprise is that this striking productivity performance has been accompanied by stagnant real wages in some parts of manufacturing. The reasons for this remains highly debated. One interpretation put forward by Robert Lawrence one of the authors cited above, is that this reflects what he calls 'labour-augmenting' technical change. In effect, innovation has added to the effective supply of labour.
What do these US results imply for India's aspirations in manufacturing, and indeed for the future of manufacturing globally? The US experience reinforces findings by other scholars, such as Dani Rodrik, that national employment in manufacturing is peaking at lower levels of income than in the past. Equally, the hangover of overcapacity in Chinese manufacturing is an object lesson in the consequences of a stress on supply without concern for demand. So in the choice between competitiveness and employment as a goal for Indian policy I would clearly opt for competitiveness.
Among the various policy tools available to stimulate competitiveness, trade policy remains crucial and pleas for protection in the interest of developing national champions should be stoutly resisted. Real exchange rates matter, but the key tool here should be rebalancing the macro policy mix, tightening fiscal policy to allow monetary policy to ease. Fortunately the government and the Reserve Bank seem in agreement on this script.
There are two additional points that I would make. The first is that, along with the US and China, India could be a big enough force in global manufacturing to shift the relative price of manufactures against itself. This is an additional reason to be judicious in the attention given to manufacturing. Second, the global future belongs to services. India is fortunate in having a vibrant and increasingly internationally competitive services sector. Policy toward services is at least as important as policy toward manufacturing. In this regard, domestic deregulation should be supported by participation in the WTO services talks.
The writer is chief economist, Royal Dutch Shell. Views are personal
1 'Yes, make for India'. Business Standard January 30, 2015
2 US employment deindustrialization: Insights from history and the international experience'. Robert Z.Lawrence and Lawrence Edwards. Policy brief PB13-27. October 2013. Peterson Institute for International Economics, Washington