The geography of industrial growth in a federal polity
The new political alignments that emerge after 23 May will influence, among other things, the pattern of industrial growth across Indian states. Whoever wins, it is almost certain that regional parties will exercise much power. Rapid industrial growth in their states will no doubt be a priority for them, as also for state governments led by national parties. This is the context in which we need to think about the future geography of industrial growth in a federal polity.
In a conference on Growth and Regional Development in India organised by the Institute of Human Development in New Delhi last week, K.V. Ramaswami presented an interesting paper entitled, ‘Where have all the factories gone: growth and concentration of sub-national manufacturing activity in India’. The paper demonstrates that just six states, Tamil Nadu, Maharashtra, undivided Andhra Pradesh, Gujarat, Uttar Pradesh and Bengal, account for close to two-thirds of all factories in the organised sector (64.3%) and a similar share of workers in the sector (62.2%). These states also account for over 63% of establishments in the informal sector and 59% of workers in the sector. Industrial establishments are thus highly concentrated in these six states that constitute the core of industrial India.
What accounts for such concentration? Paul Krugman’s new work on economic geography explains that differences in transport costs, economies of scale, factor mobility and market size, collectively described as “the economies of agglomeration”, lead to concentrated patterns of industrial location (see Krugman’s ‘Geography and Trade’, Cambridge, The MIT Press, 1991 and ‘The New Economic Geography, Now Middle Aged’ presentation to the Association of American Geographers, 16 April 2010). India is not an exception in this. Industrial agglomeration is the global norm.
However, alongside the concentration of industrial location, another dynamic is at work. The rise and decline of old and new industrial hubs. Though industries were scattered across several states at the time of independence, Kolkata and Mumbai alone accounted for 42% of registered factory employment and 50% of total manufacturing output (C.R. Pathak, ‘Spatial Variation in Urban and Industrial Growth in India, Indian Journal of Regional Science’, vol.vii No. 1, 1975). By 1961, manufacturing industry had spread to 89 industrial districts located mostly in Bengal, Maharashtra, Gujarat, Tamilnadu, Uttar Pradesh and undivided Andhra. At the turn of the century, the same six were still the leading industrialised states, as noted earlier. But Bengal had by now slipped to the fifth rank in share of organised sector factories and employment. Meanwhile, three new states had appeared as new industrial states with rising shares of factories and employment both in the organised and unorganised sectors: i.e., Karnataka, Haryana and Punjab.
This churn of old and new industrial hubs is partly endogenous, a peaking of the agglomeration process. With increasing concentration of industries in a hub, competition drives up the cost of labour and especially the price of land. Crowding, traffic snarls and demand for utilities drive up congestion costs in the core of the conurbation. Beyond a point, the negative “backwash effects” outstrip the positive “spread” effects and deglomeration sets in. Industrial units progressively relocate to the periphery and eventually move to a new location altogether.
To this endogenous process of agglomeration and deglomeration must be added historical and institutional factors, as emphasized by geographers. It is best illustrated by the case of Bengal. As the capital of the British Empire in India, Kolkata and its hinterland emerged a pre-eminent industrial hub in the late 19th and early 20th century, mostly led by colonial managing agencies. Agro-processing industries like jute, mineral based industries such as coal mining and iron and steel, and ancilliary engineering industries linked to the development of railways were the major industries. The shift of the capital to Delhi and political independence triggered Bengal’s gradual decline, which was precipitated by several shocks. The unfortunate freight equalisation policy of the new national government deprived the entire eastern region of its locational comparative advantage in mineral-based industries. Then the collapse of public investment, especially in railways, killed the state’s engineering industry. Bengal was already in relative decline in the 1960s when a third shock, turbulance arising from militant political movements, led to a massive flight of capital from which the state is yet to recover.
Similar stories of industrial decline could be recounted for Uttar Pradesh and Bihar. At the same time, aggressive industrial development programs pursued by state governments, usually supported by a friendly Union government, led to further industrial agglomeration in the old industrial states of Maharashtra, Tamil Nadu and Gujarat, alongside the emergence of new industrial states like Karnataka, Punjab and Haryana.
The political alignments that emerge after 23 May will be an important determinant of the future geography of India’s industrial growth. Along with the dynamics of agglomeration and industrial policies of state governments, a critical third factor will be the role of the Union government. Its spatial allocation of investment in infrastructure and its guidance to public sector financial institutions that dominate the allocation of capital among the states will be important drivers. Is there some way of ensuring that the Union government maintains neutrality, without tilting the playing field in favour of one state or another?
One way of preserving Union government neutrality is to empower the Inter-States Council as a federal institution to ensure that financial resources flow to states transparently in accordance with reasonable criteria. The states can collectively push for its re-invention as a powerful federal body, like the GST Council.
Sudipto Mundle is a Distinguished Fellow at the National Council of Applied Economic Research