Subir Gokarn: Describe to prescribe

One has to be conscious of the conflicts of interest when the same institution is serving both public and private interests.

Of the three institutions I have worked in, two, the Indira Gandhi Institute of Development Research and CRISIL, were established during the 1980s. The third, the National Council of Applied Economic Research (NCAER), where I spent three years, celebrates its golden jubilee this year. At some point, the contrasts and similarities between public and private, academic and corporate and old and new organisations may merit a column or two. This one, however, is a reflection on the contemporary significance of an institution like the NCAER, both with reference to the value that it creates and the challenges that it faces.

Any institution that has survived for half a century needs to be given some credit for resilience and adaptability. The NCAER certainly showed this in the early 1990s, when its virtually unconditional budgetary support from the Ministry of Finance was cut and it had to find ways to generate resources from elsewhere. Faced with this challenge, it exploited its capability for large household surveys to provide what remains a definitive database of consumer spending patterns. This database was really the first to quantify the emerging Indian middle class. In an economic environment which was irresistibly transiting from being a sellers' to a buyers' marketplace, it was a valuable input into the strategising of the several new producers that entered the market after the abolition of industrial licensing.

However, this swing towards meeting private sector requirements for economic research, while demonstrating the adaptability of the institution, also highlights the essential dilemma facing institutions of this nature. To comprehend this, one has to go back prior to the 1990s, when much of the NCAER's activities focused on the evaluation of public programmes based on field survey methods. The floor which I occupied in the building had several cupboards stacked with reports from the sixties and seventies relating to one or the other scheme being assessed for its impact in one or the other district.

In some form or the other, this activity still goes on. Despite the profile that its consumer surveys and macroeconomic forecasting have created for the institution, much of its revenue still comes from projects carried out for various government agencies. These also typically involve evaluations of programmes and policies. But, apart from the occasional report, a recent example being the study of petroleum product subsidies, these do not seem to play the role that they probably deserve to in the wider public debate on economic and social policy.

This, to my mind, is a loss and it brings me to the essential issue of contemporary relevance. As policy is increasingly driven by the need to make objective, visible improvements in the quality of life for citizens, there is an obvious need for continuous tracking of the impact of specific policies on this critical indicator.

But, given that there are several forces at work simultaneously, some emerging from policy, others from fundamental changes taking place in the environment, simple field observations are not enough. In order to measure the effects of specific policies, some understanding of the inter-linkages that can either reinforce or vitiate their impact is necessary. This calls for a combination of modelling and field surveys, a hybrid that the NCAER successfully implemented in a wide variety of applications.

Providing a factual and transparent analytical foundation to a whole range of policy debates is an absolutely essential function in today's environment and, having established itself in that space, even if in a somewhat different context, there is no reason why it cannot continue to play that same role in the future. But, here, one has to be conscious of the possible conflicts of interest that arise when the same institution is serving both public and private interests.

Besides being a disseminator of consumer data, the NCAER also generates revenue from assignments from the private sector, which essentially involve developing analytical arguments for or against a policy change. Of course, lobbying based on analysis is both a legitimate and a potentially efficiency-enhancing activity within our system of governance. But, it requires great efforts to keep the private interest and the public interest separate.

For the former, such research is useful only when it serves their objectives, whereas, for the latter, it is necessary to reveal both positive and negative findings. Imagine a situation in which a study commissioned by a private party turns up analytically sound findings that indicate a policy change contrary to the sponsor's demands. The commercial contract will obviously dictate that such findings not be made public, at least for a considerable length of time. The public interest, on the other hand, would require that these findings enter the policy debate as soon as possible.

While it is possible to think of governance structures that will manage such conflicts, public trust in the credibility of a policy research institution is probably going to be enhanced by a complete absence of even potential conflicts of interest. The short point is that there is a definite need for a policy research institution that provides the description for the prescription. But its value is entirely dependent on its credibility. In this respect, public funding may be preferable to private, the former, of course, being subject to careful external oversight.

However, that positioning itself raises two fundamental challenges""of people and processes. First, even without the massive increases that salaries in the knowledge sector have seen in recent times, it stands to reason that knowledge workers will be paid more when their employers can appropriate the full benefits of their services. Publicly-funded institutions will cease to offer anything meaningful if they cannot find the necessary human capital. They cannot obviously compete directly on compensation, but they must find viable alternatives.

Second, even with the right skills, relevance and value are directly proportional to speed and responsiveness. Otherwise, the research becomes, in the truest sense, academic. How to bring facts and analysis into play within a timeframe in which public and policy-making attention is focused on the issue is the question. Some combination of off-the-shelf current data and models and the ability to tweak them quickly in response to questions being raised will obviously be necessary, but implementing this in even a narrowly defined range of applications is enormously difficult.

I can only wish my former colleagues well as they ponder these issues. They will derive some comfort from the fact that the NCAER has fifty years of institutional experience to help it deal with these and future challenges.

The author is chief economist, CRISIL. The views here are personal