Raymond Fisman is a super-sleuth who exposes economic gangsters, corrupt leaders and perpetrators of violence. There is a hint of Columbo about him, he of the odd squint and grubby raincoat, although not in sartorial inelegance but in his persistent questioning and rigorous analytical approach. And, surprisingly perhaps, he is not a cop. Mr Fisman is a development economist at Columbia Business School in New York.

Like many a good television sleuth (think Jessica Fletcher or Ellery Queen), he is also an author, even if there the similarity ends. He has written Economic Gangsters, Economic Corruption, Violence and the Poverty of Nations and in it he and co-author Edward Miguel trace the steps of the corrupt using not DNA or forensic science but data and statistics. Theirs is a testament to the truism that when you are looking for clues, follow the money.

The book gives half a dozen examples of how data can be used to find corrupt behaviour, particularly in developing countries. For example, the authors collected data on unpaid fines in New York by diplomats to see which countries had the most corrupt and the most honest representatives. The authors found that their analysis of parking fines bore a strong resemblance to the World Bank's corruption index for countries. Top of the Hall of Shame was a diplomat who averaged two parking violations a day.

It should be noted that there is an oddity here: what is equally amazing is that anyone bothered to pay for parking at all, since diplomats do not have to stump up such fines in New York so can park with abandon throughout the city.

While this may sound like an academic who is having fun with some data, there is a very serious side to Mr Fisman's work. He argues that “there are different kinds of evidence - not just statistics. The nice thing about data is if it is done right, it is relatively easy to see through when people are just trying to use data as a crutch for ideology. It just doesn't lie.”

As the world falls further into recession, Mr Fisman thinks that we need to keep an eye on all the money that is being handed out by the State to support the banks and other parts of industry: “The thing that I am interested in right now is who gets bailout money. The single biggest thing is the trillion-dollar bailout. You'd think we'd be interested in tracking how the money was used pretty carefully?”

The professor argues that there is insufficient collaboration between government and economists - he says that academics could really get their teeth stuck into the big issues if they shared their data more readily. He talks about a former research assistant working in the US's Federal Reserve who has got “better data than others ... This all goes back to the need for connections, the fact that I have someone in the Fed where he can get good data, even though in theory it's all public.”

Although Mr Fisman can turn his hand to most subjects, he is first and foremost a development economist. He is excited by microinsurance - insurance for people with very low incomes. For the academic, it is “the next frontier with potentially the biggest social gains”. He has uncovered bad practices that could be prevented with the help of microinsurance, he claims.

In one of the chapters of the book, the authors show how “witch killings” are related to crop failure. “These killings don't happen at random. Witch murders are overwhelmingly concentrated in years when bad weather and crop failure cause farm incomes to plummet.” The result is that older women are accused of being witches and murdered so that the others have a greater chance of surviving.

Microinsurance may offer a solution to this and many other problems that leave people in poverty. The small-value insurance schemes can be offered for the purposes of providing money that poor people can fall back on when they face an economic shock, a concept called “consumption smoothing”.

Understanding the incentives behind people's motives are at the core of Mr Fisman's approach. Like Columbo, Mr Fisman's incentives are to find out the truth - and, like the crumpled lieutenant, he usually has at least “one more question”.

Explore Management

• MBA