On September 11, 2001, Mohammed Atta boarded American Airlines Flight 11 at Logan Airport in Boston and sat down in seat 8D. Not long afterwards, the plane took off on the fateful journey that left thousands of Americans dead and changed the world.

How much do you think it would have cost to persuade Atta to change his plans, get off the jet and go and enjoy a hearty breakfast instead of martyrdom? What would he have taken as a bribe? A tenner? A thousand dollars? A million?

The question seems ridiculous, even offensive. But it is simply a very crude version of a remarkably common theory – the idea that we can defeat terrorism with money. And among those with some sympathy for this notion is our Prime Minister.

Gordon Brown likes his big ideas, bless him, and to use them to batter his audiences into submission with his Shock and Bore speaking style. So naturally he has a big idea for the War on Terror. It’s economic development through international action. This is not his only policy, of course, but it is, if you like, chef Gordon’s signature dish. A Brown address on Iraq, the Middle East or domestic security is not complete without some mention of economic policy.

The notion is that by reducing poverty, by increasing the economic stake that citizens have in the success of their society, by combating illiteracy, you help to stop terrorism before it gets started. It is also a classic new Labour idea. It provides a third way between neocon aggression and liberal complacency. Tough on terrorism, tough on the causes of terrorism, that’s Gordon Brown.

Which leaves this question – is the Prime Minister right? Is he right that economic development would combat terrorism and improve security? And we can ask the question sharply because it is possible to answer it sharply.

And the answer is no. No, he isn’t right.

One way of responding to the Prime Minister is as I did in the beginning of this article. Simply by asking whether financial wellbeing was on the minds of the terrorists as they committed suicide, it is possible to cast doubt on the idea that you could have bought them off by making their living conditions somewhat better.

Fortunately, we can do far better than this. For the Prime Minister’s idea is that rare thing in politics – a testable hypothesis. Poverty, the financial stake that individuals have in society and rates of literacy are all measurable. So are the extent, the individual circumstances and country of origin of terrorist activity. If the Brown big idea is correct, then the relationships between these variables are predictable.

The tendency to be a terrorist should go up as personal wealth goes down; terrorists will tend to come from poorer places rather than wealthier ones; and societies with low literacy rates will tend to produce more terrorists than those with higher literacy.

So, when you look at the data, what do you find? This week sees the publication of an invaluable little book by Alan Krueger, a professor of economics at Princeton. What Makes a Terrorist uses standard tools of economics and statistical analysis to get at the truth about terrorism.

Krueger started by studying hate crimes in Germany. He regards hate crimes as a close cousin of terrorist activity, but likely to be more spontaneous and less well organised. He thought, therefore, that they were a good way of looking at individual behaviour before organisations became involved.

He studied different regions of the country and found that within them there was an inverse relationship between unemployment and ethnic violence – where there was less unemployment there was more hate crime. It seemed also that hate crime was less likely where standards of law enforcement were tougher.

He then created a tight definition of terrorism – “premeditated, politically motivated violence . . . perpetrated by substate organisations and individuals with the intent of influencing an audience beyond the immediate victims” – and began to look at different terrorist groups. Suicide bombers from Hamas and Palestinian Islamic Jihad were less than half as likely to come from families below the poverty line as an average Palestinian. Less than 15 per cent of Palestinians have more than a high-school degree, while almost 60 per cent of the suicide bombers did.

Krueger also studied Hezbollah. He found that “members of Hezbollah were better educated than the relevant segment of the Lebanese population, less likely to come from impoverished families”. He also provides a survey of public opinion that shows that support for terrorism is associated with better education, literacy and personal circumstances.

That’s all very well, but perhaps the terrorists are just a vanguard, a politicised elite reflecting the dire economies in which they live. The computer says no.

A study of the country of origin of terrorists reveals that lower-income countries are no more likely to produce terrorists and, crucially for the Brown hypothesis, “we find no significant impact of GDP growth on the frequency of international terrorism”. Furthermore “neither the overall literacy rate nor the female literacy rate appeared to have any effect”.

None of this means that you shouldn’t try to improve living conditions in other countries. If it works, it’s moral, it’s right. But it is not, as the Prime Minister argues, a counter-terrorism strategy.

What would be? Krueger finds one familiar fact in all his numbers. Countries with fewer civil liberties tend to produce more terrorists. But that’s the neocon contention. And we don’t want to start up with them again, now do we?

Times Recommends

• A dose of defeat is just the medicine
• Protecting ourselves from Eurocreep
• After the summer of LUV, it'll still hurt