Kaushik Basu named as World Bank chief economist

Annie Lowrey has the scoop.  Excerpt:

A paper he released last year caused a bit of a stir: While advising the Indian government, Mr. Basu argued that countries could reduce the incidence of “harassment bribes” – e.g., “I’ll approve this home renovation project for you for a small fee…” – by making it legal to give a bribe, though not to receive one.

“This will cause a sharp decline in the incidence of bribery,” Mr. Basu said. “After the act of bribery is committed, the interests of the bribe giver and the bribe taker will be at divergence. The bribe giver will be willing to cooperate in getting the bribe taker caught. Knowing that this will happen, the bribe taker will be deterred from taking a bribe.” (Mr. Basu notably argued against giving an amnesty for past incidents of bribery.)

Mr. Basu is to start at the bank Oct. 1. In the meantime, you can follow him on Twitter.

And do not forget:

He is also the creator of the two-player board game Dui-doku.  He is also a Woody Allen fan.


Why on earth was bribe-giving illegal in the first place? I mean, I can understand prohibiting giving bribes where it would advantage you over other applicants, but giving a bribe for something that you should be getting anyway for free? The bribe-giver is doing nothing wrong.

So as not to give the public official an incentive to do something they should not have done (and would not have done in the absence of the bribe).

+1 Game theory 101. But, a bit difficult for the man on the street to understand.

Also, a bit impractical. The bribe giver would probably expose themselves, not to suits by the government, but to suits by a disadvantaged business that was injured by the bribe giver. Also, a little impractical for US firms, as the Foreign Corrupt Practices Act would still apply.

Which explains why bribery is so rare in the developed world even though it too is illegal there.

"He is also a Woody Allen fan." As a clarinetist, I assume?

"After the act of bribery is committed, the interests of the bribe giver and the bribe taker will be at divergence"

Doesn't the bribe-giver have an interest in having the bribe-taker remain in the office from which he can perform the service the bribe was intended to facilitate? And, thinking ahead, to have a known-reliably-bribable civil servant remain in that office for future transactions? Because in a culture where this sort of thing is a problem, the replacement for a crooked civil servant won't be an honest civil servant, but another crooked civil servant. One who, knowing what you did to his predecessor, won't take your bribes at all and you won't be able to get future building permits (or whatever) at any price.

Or maybe you'll be able to get building permits for free, and for e.g. firetrap SRO hotels likely to kill dozens of innocent victims, by simply threatening to falsely accuse the civil servant (honest or otherwise) of having taken your bribe. The corrup civil servants who have hit me up for bribes have all been pretty good about making sure they had plausible deniability in place should I report them, so has Basu said anything about what standard of proof his proposal would involve? I'm having a hard time seeing how this is implemented in a genuinely positive way; at "best" you wind up with a way of making sure civil servants who do take bribes remain reliably bought.

Good points. Depends on whether a one shot or repeated game. Maybe give an award to the recipient just as you do in antitrust cases where a firm does in and discloses an unknown conspiracy to get thei execs off the hook and avoid fines. Look up Antitrust Div amnesty program in google for an example.

I don't understand why the bribe giver would cooperate in getting the bribe taker caught -- what is the benefit to the giver? Maybe there is a benefit if the government started paying bribe givers who get bribe takers caught. Then it seems like the government is just outsourcing sting operations, although maybe that would be more effective than the alternative. (?)

Basu taught a popular game theory course to undergrads at Cornell. It was the less theoretical, more interesting (and more popular) version. I never had him but apparently people like him as an instructor. Here is his RateMyProfessors.com page: http://www.ratemyprofessors.com/ShowRatings.jsp?tid=188471

Let's say that I am a private-citizen, clean government advocate, and I want to expose bribe takers, even if it costs me money to do so (such people do exist, believe it or not). I cannot presently perform the simplest method for doing so- running a sting, as you put it, since it is illegal for me to offer the bribe in first place, regardless of what my motivation might have been.

Also, it will be marginally easier to get witnesses/participants to come forward if bribe giving is not actually illegal.

You could also run stings or reward employees for disclosing attempts. Or, you could reduce discretion of certain levels of employees to grant or deny licenses, etc., Or, you could move permitting and other transactions online where the potential briber would now be dealing with either a machine or a randomnized someone on the other end of the computer line. Randomnization also works, because you never know who you are dealing with because their is no history.

One other solution: India could outsource its permitting, etc. to the US which has a civil service corruption.

That'll be the day

should read: low level of civil service corruption.

I am not sure what to make of Prof. Basu, but here is a startling sentence from one of his papers:

"The aim of this paper is to show, by using a more complex reasoning, that it
is indeed possible to manipulate the market so as to make a profit by deliberately
making the exchange rate fluctuate. All one needs is a deep pocket, moderate
intelligence, and unfussiness about moral scruples. The purpose of this short
paper is to show exactly how the art of market manipulation works."

The paper is written in a faux-scientific style, and I lost interest very quickly.
Are his views on "market manipulation" widely shared in the Economics community?
I assume Prof. Basu has never tried to implement his "model", but I wonder if we could
persuade him to set aside his "moral scruples" for just a brief period, try the "model" and report back?
(I assume he satisfies the other two requirements...)

So what he's saying is that it's possible to make a small fortune out of a large one?

The paper I referred to in the previous message: http://www.kaushikbasu.org/Currency%20manipulation.pdf

This can be done in a different way: paying and receiving bribes is illegal, but whoever goes first to the authorities gets out free.

Also, entrapment should be legal: the multinational corporation official who thinks he can get his way with money can be taped, and ditto for the corrupt official who asks for money for doing what he ought to do anyway or for not doing what he ought to do, or for doing what he ought not to do.

The corrupt official or businessman can choose to be subject to blackmail by whomever has him on tape, or can go to the authorities without fear of punishment.

If you make it legal only for one side, you generate a perverse incentive for blackmail. The main thing is that incentives become divergent.

Caught on tape doing what? Saying "Haha! I will not give you a building permit unless you pay me $1000, which you will now place one bill at a time in my outstreteched palm!"?

Because in the real world, at least in places where there is any real possibility of being arrested for taking bribes, what the corrupt official actually says is something like, "Your application for a bulding permit has too many flaws for me to allow it; in its present form it would endanger the community. I know a guy who works as a consultant for people like you; first-rate structural engineer, charges $200/hour but he can get you on the right track in an afternoon..."

Now prove beyond a reasonable doubt that the corrupt official is engaged in a criminal conspiracy with the "engineering consultant" and that the two of them have just solicited a bribe. I'll even spot you a tape of the engineering consultant's sales pitch, where he promises a 100% success rate in helping people get building permits. Or are we dispensing with the whole proof-beyond-a-reasonable-doubt thing in our quest for an honest civil service?

Bribery is a serious problem in so many economies, not because nobody before Basu ever came up with the simple, obvious solution for fixing it, but because this is a genuinely hard problem and the simple, obvious solutions don't work.

From his Wikipedia page:

"His entry into government in 2009 was a new experience. In an interview to the Bengali magazine, Desh, he said his earlier experience of government, when he was setting up CDE, was not a happy one. Letters and phone calls were met with no response. In desperation he went to see the then Finance Minister, Dr. Manmohan Singh. Many bureaucrats saw him waiting to meet the Minister. Thereafter their behavior changed so markedly that he had to toy with the idea of writing to the Minister to visit him periodically. He would, however, not disturb the Minister, just sit in his waiting area and go away."

So, in today's shocking development, one academic praises another academic's foolishness, even though the latter is manifestly untethered to reality.

One commenter has already made the point that the bribe-giver is not going to turn in the bribe-taker because the briber has now effectively developed a client in the bribe-taker. To be clear, as the bribe-giver, I now have found a chink in the bureaucracy that I can ply with money and get the results I want. Am I just going to give that up?

Here's another reason why the bribe-giver won't turn in the bribe-taker: he (i.e. the bribe-giver) will quickly develop a reputation as a whistle-blower. It's reasonable to think he can expect retaliation from other corrupt bureaucrats. It's also reasonable to think he can never bribe again. If the culture of corruption is dominant (as is the maintained assumption in even Basu's story), it is dominant for the bribe-giver to not turn in the bribe-taker ex-post.

Maybe it eliminates one-shot opportunistic bribery. For example, a speeding ticket in a town that's hundreds of miles from your hometown, so that it is common knowledge that the game is one-shot. But even in that example, the deleterious reputational effect of whistle blowing (to the whistle-blowing bribe-giver) applies.

Keep patting yourselves on the back, academics.....

Comments for this post are closed