Prof. Avinash Dixit’s suggestion to curb corruption (based on Lin Ostrom’s insights)

Prof Avinash Dixit has this interesting paper on resolving corruption in India (and elsewhere). He is to discuss the paper at IGIDR in early Feb as well.

His suggestion is very different from Kaushik Basu’s who wanted different incentives  for bribe taker and bribe giver. But am sure like Basu’s suggestion this should also bring some interest and discussion.

Prof. Dixit uses the community approach on lines of Lin Ostrom. Here corporates ostracize those companies which are found engaging in corruption. And this ostracisation is going to happen on terms agreed mutually by the industry.

I will briefly describe this work and derive some implications for controlling corruption. Begin with Greif. Consider a community whose members have frequent  occasions to interact with other members for business purposes. Many of these deals turn into prisoners’ dilemmas: after the fact, one or both parties to the  agreement have the temptation to renege and extract greater benefit for oneself at  considerable cost to the other. If the formal legal system is working well, they can write a contract whose terms forbid such cheating, and expect to get the contract enforced in the event of a dispute. But if the formal legal system is corrupt, biased,  inexperienced, or defective in some other way, they need an alternative contract  enforcement mechanism. Otherwise fear of the partner’s cheating may deter the  people from entering into the deals in the first place. Many mutually beneficial  opportunities will go unexploited, and business will grind to a halt. Therefore the business community as a whole has a strong motive to establish and sustain a highquality contract enforcement institution, even though (in fact because) each member faces a temptation after the fact to evade enforcement.

… Thus my proposal is for the business community to take matters in its own hands, and establish a norm of conduct and sanctions for its violation, that will deter  any member from giving a bribe. This institution operates entirely on the supply side of would-be corrupt deals with any government officials who demand bribes. It  can operate on its own, regardless of any punishment the law may have on its books  for taking (or giving) bribes, but in practice the two institutions — formal laws of the  state and private ordering by the business community — can coexist, and the  deterrence each provides can reinforce the other.

Ostrom’s studies and meta-analyses led her to identify several conditions for successful self-governance of common pool resources. “In all self-organized systems,  we found that users had created boundary rules for determining who could use the resource, choice rules related to the allocation of the flow of resource units, and active forms of monitoring and local sanctioning of rule breakers.” 8 In other words,  the rules, and to whom they apply, must be clear and clearly understood, and  violation of rules must be monitored, and effective sanctions applied, by the  participants themselves.


He says in India this can be done via various business chambers:

India has several well-established business associations. Confederation of Indian Industry, Federation of Indian Chambers of Commerce and Industry and Indian Merchants’ Chamber are probably the largest and most comprehensive; there are also local and industry-specific groups, and networking associations like local chapters of Rotary and Lions Clubs. These associations have web sites, publications, and events for networking. These provide an excellent basis for news to disseminate quickly. There is also a good basis for insider knowledge; I am sure the capabilities  of members and non-members are sufficiently well known to other businesspeople  in their local communities, that if someone wins a government contract the others  will quickly figure out whether the winner has plausible ability to win on merit. Any suspicion can be investigated formally or informally. The associations can even infiltrate informants in government departments, who can discover and report on suspected corrupt practices. Leaders of the community can then hold internal hearings to come to a judgment and issue verdicts of sanctions against miscreants.

Internal industry-based tribunals have some great advantages over formal  courts of law. They have expertise to evaluate information, therefore they can use  some items that would not meet the standards of evidence in a criminal court.9 And their ultimate sanction of ostracism, which puts the miscreant out of business  altogether, can be an even better deterrent than the fines a court would levy. The value of such private ordering has been amply proved for private commercial contract disputes.10 Its use for exposing and punishing corruption would be a bold and potentially valuable extension.

Hmm. This is interesting.

This isn’t entirely new. Industry leaders did it in the past and should do it now:

To get such a system started, the business elite must take a lead. Again a  historical precedent proves instructive. At one time there was no legal requirement  for firms to publish any financial statements even if they sold shares to the public. In 1898, Elbert Gary and J. Pierpont Morgan voluntarily started issuing quarterly reports for their Federal Steel, in pursuit of their belief that “corporations issuing publicly traded securities should account for their financial performance.” Only later did the progressive movement make this, together with external auditing, into a legal requirement. Economic theory supports the idea: the largest players are the  ones who take the lead in private provision of public benefits. In the context of  corruption, the pioneers who refuse to bribe and refuse to have any business  dealings with anyone who bribes will suffer some loss of business until the norm  takes hold. But the largest businesses are in the best position to suffer the  temporary loss, and also have the most to gain from establishment of a cleaner  system.

Fascinating. So the quarterly earnings released by corporates was basically because of JP Morgan…

Just a comment here. Chambers of commerce are run by industries themselves. So why should they bell themselves? Gulzar in a recent fab post questioned the corporate leadership in India.  Most corporate leaders  suggestions to improve Indian economy are basically for corporate India. He questions corporate leaders in India for being too self-oriented and never talking about things that matter to all Indians. In such a scenario why will they be interested in curbing corruption??

And he is right here. Just a few months ago, corruption was the central issue. Leaders talked about it as it affected their wealth and stock prices declining. And now no one talks about it as stock prices have moved up. Has corruption disappeared suddenly? There is clearly a lot of crony relation between business and govt. As long as profits move up we are not talking about lack of governance etc kind of issues…

Prof. Dixit  goes on to discuss how government can play a role. He says government can support this mechanism by offering a single office for whatever services and permissions are needed:

The government can support and reinforce the norm-and-sanction based, supply-side governance of the business community in more positive and direct ways, using measures that work on the demand side as well as the supply side of corruption. Much has been said and written about detection and punishment. Let me offer an idea that has been advocated for other reasons, but has a useful role to  play in reducing the harmful effects of corruption: one-stop shopping. Whenever  any economic activity such as setting up a new corporation or construction of a new  plant requires multiple licenses or government actions, they should all be channeled  through one office; once an investor or a firm has this office’s clearance, no other  agency or official in any capacity should be able to hinder it further.13 This arrangement, which exists in some cities and states in the U.S. and in some countries like Indonesia, has obvious advantages of saving administrative and  transaction costs.

He points to examples from industrial organisation economics:

Its side-benefit in dealing with corruption arises by analogy with  the idea in industrial economics that mutually omplementary goods are better sold  as a package by a monopolist than separately by oligopolists. As an example,  consider hardware and software, both of which are needed by computer users. 

Suppose the two items are sold separately. If the hardware seller firm raises its  price, that makes the hardware-software package more expensive and therefore reduces the demand for it. This hurts the sales of software and therefore reduces the profit of the software firm. The same is true for the software firm’s pricing decision.  Each firm has no reason to take into account this harmful effect on the other;  therefore both raise prices too far. If both items were sold by the same firm, it would recognize the interactive effect on its overall profit; in economists’ jargon it would  “internalize the externality.” It would set prices lower, raising the quantity it sells of  both items. This would benefit not only the consumers, who would get their  computers and software at a lower price, but also the firm, which would make more total profit; after all, that is the reason why it lowered the price.

Hmm.. So complementary products should be sold by monopolist  and substitutes by oligopolists:

Now apply this idea to the context of corruption when several licenses are  needed to start a business or construct a plant. If each license is handled by a  separate corrupt office, each demands a high bribe, ignoring the negative side-effect  on the take of the others. In a one-stop shop where this effect is internalized, each applicant would pay a lower total bribe for the package of licenses. That would  lower the overall cost of entering the business or constructing the plant, thereby  increasing the number of applicants and the level of economic activity. Incidentally  it would also increase the total amount paid as bribes by all the applicants together,  so this is one reform that should not encounter opposition from the officialdom!

Just as complements are better sold by a monopolist, substitutes are better  sold (from the point of view of the consumer or of efficiency in the economy as a  whole) by separate oligopolists, because they will compete fiercely with each other.  For example, computerization of the Indian railway ticketing system has given  travelers the ability to buy tickets from anywhere, and the bribes that had to be paid  at the ticket window have disappeared. In our context, there could be two or  more  distinct offices, each a one-stop-shop empowered to issue the whole package of   licenses. Their competition can keep down the cost of the package to the consumer, that is, lower the bribe, perhaps reduce it close to zero.

He says one should not just dismiss this suggestion and work on it:

I have not offered a fully fledged scheme ready to roll out tomorrow; far from  it. I have sketched a basic idea and pointed out some problems it must overcome;  there are surely other difficulties that further study will reveal. But I believe that  instant dismissals — “It will never work in India; the country is just too complex; the  problem is just too rampant.” — are mistaken. At the minimum, a good case exists  for continuing to think further and deeper, to examine whether and how the difficulties can be countered, ameliorated, or overcome. Allow me to offer this as a  research program, or even throw this down as a challenge, to the Indian business  community and to academic researchers in Indian business schools, law schools,  and economics departments.

Worth a thesis for so many departments. Corruption anyways is an interdisc area and scholars from different fields can latch on Dixit challenge. There are so many perspectives here…

Quite a read.

Though, much of it looks theoretical. Prof. Dixit assumes that large number of corporates in India are clean. Which is just not true. Looking at how resources have been distributed in India’s high growth phase, we have clearly moved from a licence raj to rent-seeking raj. There is a wide understanding between corporates and government on several issues and people are fooled. Doing business rankings tells you the story. We are way behind on most parameters. And India is growing at such levels despite such poor doing business rankings, does suggest of the cronyism in the economy.

So it is unlikely the leaders will move here. The revolution could perhaps come from the bottom where Small enterprises push the chambers to take actions on these lines. We have seen this in recent times where people have pushed for reforms and protested to political leaders. Similar things could be attempted in the economy/corporate sphere as well..Most revolutions are from driven from lower rungs and not higher rungs..

One Response to “Prof. Avinash Dixit’s suggestion to curb corruption (based on Lin Ostrom’s insights)”

  1. derek Says:

    I’ve read Prof Dixit’s article and find it redundant. He didn’t use any real life examples of why some governments (e.g. Singapore, Hong Kong) can fight corruption effectively, and the importance of institutional arrangements and transaction costs (i.e. institutional economics). I remember one economist said that ‘government intervention is the hotbed of corruption’. Thus, less government intervention in markets, and the government only defines & enforce property rights can effectively reduce chunks of corruption.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: