The recent controversy with regard to computation of the GDP raises some issues. Is GDP always the appropriate indicator to measure the economic activity of a country or a region? Why is it important for us to have a geographical perspective on this aggregate measure of economic activity? Economic activity clearly is much more intense near oceans, or, if inland, along navigable rivers where transportation by ship is feasible.
The best example of capitalising on the geography of a place for furthering economic activity is China’s special economic zones, which India is now trying to emulate. The 2009 World Development Report highlights the fact that since the 1990s, millions of Chinese workers have migrated to get closer to economic opportunity concentrated along the coast.
John Gallup, Jeffrey Sachs and Andrew Mellinger in 1999 introduced the concept of ‘GDP density’ , calculated by multiplying GDP per capita by the number of people per square kilometre. Defined in this way, GDP density is a measure of economic activity by area. One of the original purposes for deriving this measure was to study the role of geography in economic development.
As described by Sachs, et al, one finding is that the great majority of the poorest countries lie in the geographical tropics , the area between the Tropic of Cancer and the Tropic of Capricorn. In contrast , most of the richest countries lie in the temperate zones as well as along coastal areas. Reasons for these differences are discussed by the authors.
Taking this geographical measure, we find that the GDP density in China presents a declining trend from the southeast to the northwest of China. In fact, China is divided into five grades of GDP density. While the GDP density for India, based on 1999 data, was roughly $117.204 (based on an exchange rate of $1= . 46.37), we found that there are no recent estimates of GDP density at the subnational level for Indian states.
Based on data from the Central Statistical Organisation , we computed and examined GDP density for all Indian states in constant 1999-00 prices from 1999-00 to 2009-10 , and compared this with GDP per capita for the states over the same period. The findings are interesting . If we were to take GDP per capita , the bottom states were Madhya Pradesh , Uttar Pradesh and Bihar (both in 1999-00 and 2006-07 ).
Also, West Bengal is at 16 and 17 respectively (based on 1999-00 and 2006-07 GDP per capita). However, if we were to take our new geographical measure, GDP density, into account, MP,UPandBiharareinthetop15states (their ranking being eight, 12 and 14, respectively). West Bengal’s rank moves up to six (both in terms of its 1999-00 and 2006-07 GDP density).
Madhya Pradesh’s GDP density is higher than India’s national average, being $310.402 (in 1999, in 1999-00 constant prices). So is Uttar Pradesh’s ($156.186 in 1999 in 1999-00 constant prices) when compared with the national average for that year (Bihar’s is slightly lower than the national average for that year, being $114.911).
This implies that these states have a smaller area in relation to their economic activity. While Goa and Delhi are the top ones in terms of GDP per capita, Delhi, Chandigarh and Puducherry are the richest in terms of GDP per square kilometre (GDP density). Thus, Goa moves out of the top three league when GDP density is taken into account.
The above implies that GDP density is a much more important measure of economic activity for the poorer states. This is so because the data shows that while the areas are relatively poor, economic activity in these states is quite high in relation to their geographical area . This is important to know because of the implications for service provision, along with others. Given density varies across regions much more than GDP per capita, it tends to have a larger effect on income per unit area — the most important variable in determining the feasibility of public network access.
Take the example of telecom. Economies of density are an important characteristic that defines cost per line in the provision of telecom. ‘Cherry-picking ’ high average revenue per user (ARPU) in select high density locations is a strategy successfully adopted by private competitive fixed-line telecom service providers in the country to reduce breakeven time from six-seven years down to one-two years.
The current categories of telecom circles of A, B and C in India are based on revenue potential for telecom services (with circle A having the most revenue potential and circle C, the least). This classification is the basis of varying licence rates (10%, 8% and 6% of adjusted gross revenue) and varying reserve prices (. 320, . 120 and . 30 crore, respectively) for the recently concluded 3G auction.
However, a look at the GDP density for these states shows some surprising findings. The category B circles of Haryana , Kerala and Punjab lead in GDP density over all the category A circles of Andhra Pradesh, Gujarat, Karnataka, Maharashtra (including Mumbai) and Tamil Nadu.
This implies that the potential for recovering the cost of service provision in these ‘poor revenue potential’ states is better than what is perceived to be the case. Based on this, a relook at the categorisation of circles as envisioned in the basic, cellular and unified access service licence guidelines , might be needed.
Thus far, cross-country econometric models of growth have focused only on GDP per capita as a variable to be explained. But we believe that geography continues to matter importantly for economic development along with economic and political institutions. From an analytical point of view, we believe that geographical considerations should be reintroduced into econometric and theoretical studies of cross-country economic growth, which so far have almost completely neglected geographical themes.