Since my last post on the Anna Hazare led anti-corruption movement, I've found myself thinking about the issue more seriously that ever. This post too, however, is no less ambivalent than the last one. This would perhaps be expected for the matter at hand is indeed complicated. To be sure, this is not a desirable situation. It is expected of all thinking citizens to hold well-reasoned views on matters of public importance. But reason had led me afar, into a land of unanswered questions, albeit with some insight.
So why does corruption persist? To understand this, we need to understand why corruption occurs in the first place. Corruption can be viewed, as mentioned before, as a principal agent incentive problem. In a similar spirit, it may be seen as a Coasean bargaining transaction between say a bureaucrat and a private agent. (This particular interpretation lends weight to the efficiency thesis. Even if the efficiency idea were to go through, it would not mean that there are no distributional consequences). Through one of the above routes, or through any other, once the society converges to a 'corrupt equilibrium', why is it not able to deviate? Precisely because it gets 'locked into' an equilibrium. There are several explanations to how this lock-in is sustained. The textbook explanation in Economics is that the elaborate system of state licences and permits gives rise to rent seeking behaviour by private agents, which sustains corruption. The sociologists explain it differently. They attribute persistent social behavior to 'social norms' that become inherent in a society and bootstrap themselves. This may well turn out to be a tautological explanation though, especially to the economist (or even the analytical sociologist). Recent developments in the theory of repeated games have sought to develop a coherent theory of social norms, with agent based models, evolutionary game theory, bounded rationality models, institutional economics, putting in their bit. The body of literature is wide, yet it is undoubtedly work in progress. (Perhaps more on this at a later date). Another way of looking at the tenacity of corruption is to view it as a coordination problem, and this is possibly more productive from our point of view. Essentially, there are several equilibrium behaviors a society can stabilize at several equilibria (and there is no conclusive theory as to which equilibrium will be chosen) but once the stabilization happens, deviation from the equilibrium becomes difficult. The equilibria thus attained are frequency dependent equilibria i.e. the expected gain from a certain behavior (say corruption) depends on how many people are (believed to be) engaging in that behaviour (how many people are corrupt). There can be several variations on this idea, for instance, Andvig and Moene (1990) paper which assumes that the expected punishment for corruption, if detected, falls as corruption becomes more rampant. But the essence is unchanged. Jean Tirole's (1996) papers adopts an approach in a similar spirit, but provides greater insight. Tirole uses the idea of collective reputation, nicely explained (and exemplified) by Rajiv Sethi. The broad idea is that in an overlapping generations framework, the younger generation inherits the bad reputation of the older generation and because actions may be imperfectly monitored, they have no incentive to act in a virtuous manner. The implication derived by Tirole in the paper (in respect of corrupt behavior) is that while a sudden increase in corruption levels might be long lasting, a one shot fall in corruption may not have substantive effects. This, in some sense, undermines those who make a case for Anna's protest by claiming that the movement has helped create mass awareness. However, another emerging branch of game theory offers them some respite. Epistemic game theory tries to construct game theory bottom up from foundational issues about beliefs and knowledge. In this tradition, though inching away from core theory, Michael Chwee has a beautiful little book, Rational Ritual, Culture, Coordination and Common Knowledge, (available here for free download and reviewed here). The main thesis of the book is that people can coordinate without communication simply by forming common knowledge of each others intentions, and, in traditional societies, rituals are an effective way of ensuring common knowledge of certain beliefs. The idea is that if I know that you will act in a certain way, and I know that you know that I know that you will act in a certain way and so on, and there are mutual gains to be had through cooperation (say both of us adopting fair means), then without communicating individually, we can converge on the same (good) focal point. Rituals ensure that people can see each other at common ceremonies, and they can see each other seeing the same things that they are seeing. This helps them coordinate. In a sense, we can view the countrywide anti-corruption get together as a modern day (though on a much larger scale, and much weaker) ritual, manifesting itself not only on the streets but also on social networking sites.
The point nonetheless is that we cannot be sure that such a bill once passed will be able to make a significant difference to prevalent corruption. This does not offer us reason to be pessimistic. But if we are to significantly alter a system, we must be aware of potential consequences.
Eventually what I have ended up with is a set of open questions. There are pointers in both directions, but they confuse rather than clarify. If we are to be thinking individuals, however, then, I suspect, it becomes impossible for us to avoid such quagmire. To form strong opinions, and hold on to them, might, for all one knows, require some indoctrination. Or else we shall remain guilty of ambivalence. However, as Joan Robinson once said, "The purpose of studying economics is not to acquire a set of ready-made answers to economic questions, but to learn how to avoid being deceived by economists." Similarly, any exercise in constructively thinking about a social issue has value to it perhaps not in being able to provide the right answers, but in being able to see through a more conventional outlook.
 |
| Photo Credit |
It has been asserted that a group of individuals self-professedly holding high moral ground cannot rightfully claim represent the myriad interest of civil society. Some have doubted Anna's intentions altogether. These are questions worth asking. At the same time, it cannot be denied that the government has itself been using arbitrary, high handed means to overcome the situation. These too make for unworthy constitutional practices. An excellent editorial by Pratap Bhanu Mehta in today's Indian Express succinctly puts together the contradictions at hand and argues for a 'fine balance' that a constitutional democracy deserves.
I wish to pick up from where I left off in the last post. It takes an elementary knowledge of 'matters of state', and a modicum of common sense, to appreciate that a democracy 'idealized' is different from a democracy 'realized'. In practice, there are several frictions inherent in the workings of a constitutional democracy that hinder its smooth functioning. Specifically, the bureaucratic hierarchy instituted by the system provides ample scope for corrupt practices, broadly defined as the use of public office for private gain. The idea that decentralization, or delegation of authority by the principal (the authority of the public vested through the constitution in the state) to agents (the functionaries of state, including the bureaucracy, the executive, the legislators, even the judiciary), in the presence of imperfect monitoring of actions by the principal, creates perverse incentives for side-payments between the agents and the beneficiaries of state action. This is definitely not a happy situation. But it is inbuilt into a democratic system. Surely, checks and balances can be placed. They tend to be costly and the state functionaries lack incentives to install them in place. The first cry for the Lokpal bill was a clearly demand for such checks and balances (as were successful movements such as the RTI which did become an Act). By coming out in the open once again though, what Anna seems to be pushing for implicitly, perhaps unknowingly, is an overthrow of parliamentary democracy as we understand it today. He seems to be advocating a 'civil-society authoritarianism' where the civil society dictates a point wise mandate to the government. But who forms the civil society? The millions of protesters who have come together in support of his movement. This is a potential onset of anarchy. May be I am exaggerating this a bit too much, but it is clear that by defying the proceedings of the Parliament, he is hitting out at fundamental principles of democratic government. The desires of a certain section of middle class for Anna to stay within the restraints of parliamentary framework may be interpreted as their 'conservative bourgeois' mindset, much like that of the moderates who formed the INC in 1885 (now that such analogies are being indiscriminately being drawn). I wish to take no sides. I feel I am too young, and too unwise, to yet indoctrinate myself . If democracy has to suffer a blow for the want of a better system let it. (Here is a fairly radical viewpoint on the issue. An exceedingly good piece on Ambedkar's distaste for Gandhian ideals and methods is here. It forms an interesting counterpoint.) My contention is this: As a leader of a movement that has instantly gathered a large following, Anna has a higher moral responsibility. If his agenda is marked with such radicalism, the people taking part in the movement need to know this. Or the movement becomes self defeating. I take nothing away from the conscience and good sense of the Indian public when I say that 'they' need to know it. I myself am a part of 'them'. But the aggregated beliefs of such a large multitude of people tend to lose their individual shades of grey, and the 'wisdom of the crowds' is instead turned into a 'foolishness of the mob'. This is where an entire program might become misdirected without any particular person's intention.
At the same time, I do doubt the extent to which passing of a law like the Jan Lokpal Bill can go forth to tackle the more deep seated phenomenon of corruption. If it cannot, then is all the noise pointless? To be sure, corruption is not new to indian polity. In perhaps the earliest foreboding of the principal agent kind of problem, Kautilya writes in the Arthashastra:
Just as it is impossible not to taste the honey (or the poison) that finds itself at the tip of the tongue, so it is impossible for a government servant not to eat up, at least, a bit of king's revenue. Just as fish moving under water cannot possibly be found out either as drinking water or not drinking, so government servants employed in the government work cannot be found out while taking money (for themselves).In fact, the UCLA economist Deepak Lal (though I personally do not subscribe to some of his views) has gone on to argue in his book, The Hindu Equilibrium, that some sort of expropriation of resources from the public has always been undertaken by the ruling class (in one form or the other) and it is so deep rooted that it may never be removed. The society is so structured that it may continue to do good despite such expropriation. (Note this is different from the efficiency theories of corruption which say that in a second best world with policy induced distortions corruption might well be a corrective factor. It is has also been argued that the opportunity for bribe taking helps the public authorities obtain a signal as to who has the greatest marginal willingness to pay for a certain public benefit being provided by the authority (whosoever is willing to pay the highest bribe) and hence deliver the benefit to the person who values it the most, improving the efficiency of the system. There are several counterarguments to these views, but we would be digressing. The point here is not to examine whether corruption is efficient.)
So why does corruption persist? To understand this, we need to understand why corruption occurs in the first place. Corruption can be viewed, as mentioned before, as a principal agent incentive problem. In a similar spirit, it may be seen as a Coasean bargaining transaction between say a bureaucrat and a private agent. (This particular interpretation lends weight to the efficiency thesis. Even if the efficiency idea were to go through, it would not mean that there are no distributional consequences). Through one of the above routes, or through any other, once the society converges to a 'corrupt equilibrium', why is it not able to deviate? Precisely because it gets 'locked into' an equilibrium. There are several explanations to how this lock-in is sustained. The textbook explanation in Economics is that the elaborate system of state licences and permits gives rise to rent seeking behaviour by private agents, which sustains corruption. The sociologists explain it differently. They attribute persistent social behavior to 'social norms' that become inherent in a society and bootstrap themselves. This may well turn out to be a tautological explanation though, especially to the economist (or even the analytical sociologist). Recent developments in the theory of repeated games have sought to develop a coherent theory of social norms, with agent based models, evolutionary game theory, bounded rationality models, institutional economics, putting in their bit. The body of literature is wide, yet it is undoubtedly work in progress. (Perhaps more on this at a later date). Another way of looking at the tenacity of corruption is to view it as a coordination problem, and this is possibly more productive from our point of view. Essentially, there are several equilibrium behaviors a society can stabilize at several equilibria (and there is no conclusive theory as to which equilibrium will be chosen) but once the stabilization happens, deviation from the equilibrium becomes difficult. The equilibria thus attained are frequency dependent equilibria i.e. the expected gain from a certain behavior (say corruption) depends on how many people are (believed to be) engaging in that behaviour (how many people are corrupt). There can be several variations on this idea, for instance, Andvig and Moene (1990) paper which assumes that the expected punishment for corruption, if detected, falls as corruption becomes more rampant. But the essence is unchanged. Jean Tirole's (1996) papers adopts an approach in a similar spirit, but provides greater insight. Tirole uses the idea of collective reputation, nicely explained (and exemplified) by Rajiv Sethi. The broad idea is that in an overlapping generations framework, the younger generation inherits the bad reputation of the older generation and because actions may be imperfectly monitored, they have no incentive to act in a virtuous manner. The implication derived by Tirole in the paper (in respect of corrupt behavior) is that while a sudden increase in corruption levels might be long lasting, a one shot fall in corruption may not have substantive effects. This, in some sense, undermines those who make a case for Anna's protest by claiming that the movement has helped create mass awareness. However, another emerging branch of game theory offers them some respite. Epistemic game theory tries to construct game theory bottom up from foundational issues about beliefs and knowledge. In this tradition, though inching away from core theory, Michael Chwee has a beautiful little book, Rational Ritual, Culture, Coordination and Common Knowledge, (available here for free download and reviewed here). The main thesis of the book is that people can coordinate without communication simply by forming common knowledge of each others intentions, and, in traditional societies, rituals are an effective way of ensuring common knowledge of certain beliefs. The idea is that if I know that you will act in a certain way, and I know that you know that I know that you will act in a certain way and so on, and there are mutual gains to be had through cooperation (say both of us adopting fair means), then without communicating individually, we can converge on the same (good) focal point. Rituals ensure that people can see each other at common ceremonies, and they can see each other seeing the same things that they are seeing. This helps them coordinate. In a sense, we can view the countrywide anti-corruption get together as a modern day (though on a much larger scale, and much weaker) ritual, manifesting itself not only on the streets but also on social networking sites.
The point nonetheless is that we cannot be sure that such a bill once passed will be able to make a significant difference to prevalent corruption. This does not offer us reason to be pessimistic. But if we are to significantly alter a system, we must be aware of potential consequences.
Eventually what I have ended up with is a set of open questions. There are pointers in both directions, but they confuse rather than clarify. If we are to be thinking individuals, however, then, I suspect, it becomes impossible for us to avoid such quagmire. To form strong opinions, and hold on to them, might, for all one knows, require some indoctrination. Or else we shall remain guilty of ambivalence. However, as Joan Robinson once said, "The purpose of studying economics is not to acquire a set of ready-made answers to economic questions, but to learn how to avoid being deceived by economists." Similarly, any exercise in constructively thinking about a social issue has value to it perhaps not in being able to provide the right answers, but in being able to see through a more conventional outlook.
 |
| Photo Credit |
