Cost-benefit analysis key to rational policymaking

Every desirable social intervention cannot be taken up by the government because of budgetary constraints. So, policy-makers prioritise among competing interventions on the basis of their ‘gut feeling’ or perceived political dividends, administrative convenience and pressure group considerations. The most glaringly faulty examples of such prioritisation are found in the crores of rupees spent on numerous hoardings and newspaper and TV ads, with smiling faces of the prime minister or chief minister boasting of their government’s achievements.

Can we think of any better way? In theory, social benefit-cost analysis can provide inputs on the basis of which a more rational choice can be made. The ratio of social benefits to costs can be calculated for all competing projects/interventions. If the social benefit-cost ratio (BCR) is greater than one, it is a desirable project in itself. But since we cannot accommodate all the projects having BCR greater than one within the budget constraint, we rank projects according to their respective BCRs and go on ticking from the top till the budget is exhausted. A higher social value may also be given to a rupee received or paid by the ‘poor’ relative to others. This exercise would give the biggest bang per rupee spent.

Theoretically, this is a neat idea. What about in practice? No doubt, measuring social costs and benefits, and hence computing BCRs, is a tricky business, unlike private profitability in investment projects. In social infrastructure projects like health, education, energy and environment, social costs and benefits are somewhat subjective.

One needs to first identify all the costs and benefits for the society as a whole, then quantify and finally value those by appropriate prices (in some cases, no market price to measure social value or cost exists. For instance, lives saved). None of this can be done with great accuracy or unanimity. Nonetheless, a probable range of values can be arrived at. Here, the basic guiding principle is that we should try to measure the right things, even if imperfectly, rather than measuring the wrong things perfectly.

The sceptics would also say that irrespective of what economists may think of rational intervention, eventually the preferences of elected politicians would determine what interventions are undertaken.

Notwithstanding these problems, in recent years, an important beginning has been made in pushing this idea among policymakers in India and several other countries (like neighbouring Bangladesh).

The Copenhagen Consensus, an international think-tank, in collaboration with Tata Trusts, has undertaken social benefit-cost studies for hundreds of policy interventions that are already being pursued or are being contemplated, after detailed consultations with sectoral experts, government officials, civil society organisations, field-level workers and eminent economists. This project, known as the India Consensus, is already underway for Rajasthan and Andhra Pradesh, where the governments of these two states have actively participated in the ongoing deliberations. There are indications that several other states are also interested in undertaking such studies in their states.

Simple interventions

Even if unanimity of opinion is hard to find on the exact values of BCRs for many projects, it is interesting to note that some simple interventions have been found to be occupying the top ranks in the priority list, by the unanimous opinion of experts. For example, in Rajasthan, where I had first-hand knowledge of the exercise being a member of the panel of economists, the treatment and active case-finding of tuberculosis gets the highest ranking.

Apart from the benefits in terms of lives saved of the patients treated, there are significant additional benefits in terms of potential lives saved of others who would have been affected by TB through the contagion effect, especially among the poor, who live in congested, unhygienic conditions. The cost of treatment needed, for only a few months at a stretch for complete cure, on the other hand, is low. The problem is that cases of TB are not detected early or treatments are not completed or proper treatment is not provided as TB, resistant to early generation drugs, is not diagnosed properly. So, BCR is found to be very high here.

Similarly, some simple things like educating people about the benefits of hand-washing and breastfeeding, providing nutrition supplements and iron tablets to women, holding immunisation camps in lagging districts and community mobilisation against domestic violence can buy a lot of benefits at low price. In a sense, these are low-hanging fruits which do not require any complex or costly infrastructure but can generate a lot of benefits quite easily. In agriculture, e-mandis to reduce the high margins of middlemen, certified seed production and building warehouses, storage and cold chains have very high BCRs. At the other extreme, favourite projects of politicians like farm-loan waivers or soil health cards or extending solar power to more rural homes or more free trade warehousing zones have been found to have very low BCRs.

So, though not quite exact in terms of yielding unique numbers, these exercises have turned out to be useful in deflating some wrong notions of social value associated with many projects. Hopefully, with greater and wider acceptability of these additional inputs, policymaking in India, specially at the state levels, would move towards a more rational mix of projects.

Even if a few low BCR interventions (say, those with BCR less than 2) give way to a few high value ones (say, with BCR higher than 20), roughly Rs 18 additional social benefits would be generated by the re-allocation of just one rupee in the government budget. This would be a big achievement, by any standard.

--

(The author is a former professor of Economics, IIM, Calcutta)