Monday, May 11, 2009

Why welfare programs fail to yield expected results?

Armchair critics and academicians have consistently blamed governments and bureaucracies for the apparent lack of success of our anti-poverty and welfare programs. While, a large part of the blame is deserved, they are not alone in shouldering responsibility for the failures.

It is incorrect to colour all the failings of the government with the paint of corruption and inefficiency. Such explanations, while attractive in its simplicity and salience, fail to appreciate the extent of challenges faced in implementing and administering such massive and complex programs. It also prevents the identification of the actual reasons that come in the way of successful implementation of such programs and effective delivery of welfare assistance.

From the process of beneficiary selection, through transfer of funds across levels of administration, actual delivery of funds or benefits to the individuals, and reconciliation of accounts, there are leakages at all levels. The context in which these programs are implemented are immense and complex, unlike in any other country. The sheer numbers of people being targeted and the geographical area it covers is staggering. The administering mechanism is spectacularly complex with massive numbers of officials at multiple levels, spread all over the country. The absence of a reliable beneficiary database and formally institutionalized delivery channels (like bank accounts etc) compounds the problem.

I have tried to list out a few structural impediments that come in the way of effective implementation of such anti-poverty programs.

1. The size and reach of the program means that administering them requires monitoring large numbers of beneficiaries. The challenge for officials, especially at the lower levels, is two-fold - spatial (geographical jurisdiction) and numerical (numbers of beneficiaries). Both the jurisdictional area and the numbers of beneficiaries being administered are too large for one official to directly monitor. To compound the problems, the officials at the cutting-edge have neither the capability (in terms of qualifications and skills) nor the logistics and resources to effectively monitor their subject.

2. The number of levels of administration and the officials involved in administering such programs are numerous. The cutting-edge functionary, who is responsible for direct disbursement of the welfare assistance, is supervised by at least three to four levels of functionaries, even within the district itself. The multiplicity of levels increases the probability of leakages and misappropriation, besides dilution of the intensity of monitoring.

3. Successful implementation of programs, especially on a large canvas, requires trust at all levels, between all stakeholders - administrators, beneficiaries, politicians etc. Iron-clad rules and regulations cannot substitute for the absence of trust, just as market-based transactions are more likely to succeed in a framework which is under-pinned by social capital. In a milieu where trust is absent or at best at a premium, the opportunities for agents to game the system for structural and institutional weaknesses increases exponentially.

4. The monetary incentives to deviate are considerable at all levels. The scale of the programs means that even small extents of pilferage or siphoning off, aggregates to considerable sums at every level, enough to sustain a rent-seeking network. In other words, even for welfare programs which confer minimal benefits, the sheer numbers of beneficiaries ensures that the pie is large enough to satisfy large numbers of rent-seekers.

5. Though political pressure is a constant across all levels, it is stiflingly so at the lower levels. The field functionaries, being mostly from the same areas, are more vulnerable to succumbing to such pressures. Most often, local caste and community networks and political equations, entraps the field functionaries in the net of the local power clique. Transfers and postings are another important reason that makes government officials beholden to political representatives.

6. The rent-seeking gets institutionalized with the effective co-option of all local stakeholders, including even the beneficiaries, into the corruption network. Every one benefits from the low-level equilibrium, where benefits gets shared across the entire rent-seeking chain. The alternative being denial of the benefits, the beneficiary prefers to take whatever comes his way. The voice of the actual beneficiary, who loses out on his rightful share of the benefits, is too marginal to be heard, drowned out by the rent-seekers along the chain. In feudal or semi-feudal societies, the demand side pressures that can exercise societal vigilance on the implementation of such programs is virtually absent.

7. It is difficult, even impossible, to quantify the outcomes of many welfare programs, especially those not involving financial or physical handouts. For example, the outcomes of school education and primary health care are too diffused to be captured in quantified formats, though its processes can be. Even where the data is acaptured, it accuracy is of questionable nature. Wherever the benefits are handed out, the only way to ensure its actual delivery is field verification of each beneficiary for both actual delivery and quality of the benefit delivered.

8. The mechanism to punish those caught gaming the system are very weak and vulnerable to being itself gamed! The deterrent against being punished for rent-seeking is very minimal, and in some cases virtually absent. Criminals and corrupt officials take shelter behind the bureaucrateuse of "due process of law", mandated to be followed before any disciplinary punishment is finalized. Here too, the large numbers of officials being administered becomes a major challenge. At any point of time, the typical department has too many disciplinary cases on its plate to effectively administer and dispense justice on any of them. All this results in endless delays and the final closure of the cases for evidences disappearing or receding with time and other causes associated with elapse in time.

As can be appreciated, administering welfare programs poses principal-agent problems on a monumental scale. In many respects, supervising a welfare program in a country of India's size, is the mother of all principal-agent problems. Given the socio-political context, numbers of beneficiaries, extent of area, it is well nigh impossible for the principals in the government (at different levels - state and center) to effectively monitor the activities of the massive network of official agents. Further, in view of all the aforementioned, the possibilities of gaming the system are substantial.

In the coming weeks, I will try to elaborate on a few or all of these with illustrative examples from real world experience with such programs.

No comments: