Viewpoint: Part-legalisation may eradicate bribery
Around the world, petty officials demand bribes to do their jobs.
In Los Angeles, building inspectors get paid under the table to approve permits.
In Delhi, tax officials demand cash fees to give a tax refund.
Indian economist Kaushik Basu calls this harassment bribery and he has an idea for how to stop it.
Basu, who was recently named chief economist for the World Bank, thinks we could eliminate harassment bribery by legalising it or, more precisely, by legalising half of it.
In a controversial 2011 paper, Basu suggested making it legal to give a bribe while keeping it illegal to take a bribe.
I don't know how well Basu's idea would work, but I love the way he thinks.
Hide the act
We've already tried banning bribery and it didn't work, so Basu wants to change the incentives of the bribe giver and the bribe taker.
When it is illegal to either give or take a bribe, both parties are partners in crime.
Their incentive is to hide the act. For the giver, exposing the taker means admitting his guilt, so the giver keeps quiet.
But if giving a bribe is legal, the giver's incentives change dramatically, especially since Basu's proposal includes returning the bribe money to the giver as part of the taker's punishment, which means the giver wants the taker to get caught.
Not only does the giver get the satisfaction of seeing a vexatious bureaucrat punished, he gets his money back into the bargain.
Changing incentives might work where threats of punishment fail.
For me, thinking about incentives is the most interesting thing about economics. I'd go so far as to say that if you're not talking about incentives you're not doing economics at all; you're just doing statistics with dollar signs.
Unfortunately, few economists and very few policymakers think as carefully about incentives as Basu does.
This leads to perverse incentives, policy and laws, which accomplish exactly the opposite of what they were intended to accomplish.
For example, the UN has tried to combat global warming by paying companies to reduce their output of greenhouse gases.
Companies can earn one carbon credit by eliminating one ton of carbon dioxide.
Credits for eliminating other greenhouse gases are awarded by comparing their warming effects to carbon dioxide's.
Eliminating one ton of methane, which is 21 times more potent than carbon dioxide, earns 21 credits.
This plan seems sensible until you look at HFC-23, a waste gas generated during the production of a common coolant.
HFC-23 is 11,000 times more potent than carbon dioxide. Faced with this incentive, companies quickly figured out that they should produce as much of this coolant as possible just so they would be paid to destroy the waste gas.
Today, there are 19 plants worldwide that exist only to produce and destroy HFC-23. The problem is so bad that the European Union has announced that, starting next year, it will no longer accept waste gas credits in its carbon market.
We could avoid these kinds of problems if we followed Basu's example and paid less attention to the intentions of our policies and more attention to the incentives they create.
The opinions expressed are those of the author and are not held by the BBC unless specifically stated.