Benefit‐Cost Analysis of Surfaced Roads in the Eastern Rice Region of India

In a recent study, Easter, Abel, and Norton attempted to measure the contribution of various inputs to total output in the eastern rice region (ERR) of India.1 They included both the traditional inputs such as land, labor, and fertilizer as well as nontraditional inputs such as irrigation, technology, environmental factors, and infrastructure. They estimated production functions in log form using district level data and found that the production elasticity on surfaced roads was highly significant and very stable under alternative equation specifications. The present research note uses their estimated production elasticity on surfaced roads (0.208) in estimating a benefit-cost measure of public investment in surfaced roads in the ERR. In this paper, a model is set up to estimate the benefits, then costs are estimated, and, finally, the estimated benefits and costs are joined together in a benefitcost ratio. According to the Easter-Abel-Norton (EAN) study: "Surfaced roads appear to be important in explaining productivity differences among districts. . . The absence of roads in a heavy rainfall area such as the rice region has the effect of raising input prices paid by farmers and lowering output prices received by them due to higher transportation costs." Following this, the benefits of public investment in surfaced roads may be assessed as follows. Consider figure 1, which depicts the basic economic interrelationships between the rural producing and consuming sector, the marketing sector, and the nonrural consuming sector. Figure 1(a) describes the supply (S) and demand (D) curves for the rural or food surplus area. Figure 1(b) describes the excess supply (ES) curve from this area and a demand (DU) curve for the urban or food deficit area. Figure 1(c) describes the supply (SM) and demand (DM) curves for marketing services, where the supply of marketing services are assumed to be perfectly elastic over the observed range of services supplied. This is thought to be reasonable, in the absence of congestion costs, following some earlier research (for example, Ruttan).2 Using a perfectly competitive model, the last mentioned curve is a vertical subtraction of the ES from the DU curve.