Superfreakonomics, By Steven D Levitt and Stephen J Dubner

The 'Freakonomics' authors on everything you never knew you wanted to know about sex and terror

Levitt and Dubner, in this "freakquel" to their wildly successful 2005 book Freakonomics, offer another collection of "things you always thought you knew but didn't; and things you never knew you wanted to know but do"." Such as, why it's more likely that you'll die as a drunk pedestrian than a drunk driver, and how monkeys can be taught to use money. So it's great fun. It doesn't quite scale the controversial heights of their first effort, which alleged that the cause of the fall in the crime rate in the 1990s was legalised abortion in the 1970s, which meant the embryonic criminals from the underclass just weren't being born. This time, the authors debunk climate change (but they're not the first to have a go at that) and their big idea, a hosepipe pumping liquefied sulphur dioxide into the stratosphere, is perhaps just a little too freaky even for these intellectual venturers.

Would I recommend this book to an economics teacher? Yes, provided they were comfortable discussing with their students what might be described as "adult themes". Some of us were brought up to understand the laws of supply and demand in terms of how they affected the market for apples, cups of tea, or cars. Not our freakonomists, who instead turn to the market for paid sex in Chicago, then and now, to stimulate the reader. You, like me, may never have realised, for example, quite how much working girls used to earn in the halcyon days of sexual repression. In the Edwardian era, the best brothel in Chicago earned those who rented their bodies in it around $430,000 a year, at today's prices. The sisters who owned the establishment became even more fabulously wealthy. Even the average brothel worker in those days could expect to earn about $76,000 a year, allowing for inflation.

Yet today, the profession pays pitifully small amounts. Still more than the alternatives – such as doing chores for drug gangs or working as hairdressers – but nothing like what these women's grannies were earning. The typical prostitute in Chicago works 13 hours a week, performing 10 sex acts during that period, and earns an hourly wage of approximately $27. Her average weekly take home pay is $350, which includes an average of $20 stolen from clients and an allowance for discounted payment in drugs rather than cash. So it isn't much, considering, and the reason is supply and demand. Especially demand. Many more men can get far more sex these days for free. Meanwhile, supply has risen, especially as women are nowadays more willing to offer oral sex. Once a rare and costly sexual delicacy, it now accounts for some 55 per cent of prostitutes' transactions, at an average of $37.26 a pop.

The good news, if you can call it that, is that economists have demonstrated that pimps offer substantial economic benefits for a modest fee, in sharp contrast to estate agents, who offer little or nothing for an enormous reward. Prostitutes earn more on Saturdays, because clients try the more expensive items on the menu. They are also, in common with the giant drug companies or airlines, able to operate price discrimination – for example charging less to black clients than they do white ones, with Hispanics in between. And so the fascinating economics lesson goes on, and on... but you do feel a bit sordid afterwards.

The authors are rather more uplifting on the subject of terrorism. One of the puzzles they solve is "Why suicide bombers should buy life assurance." The reason is subtle. The best way to deal with terror is not to wait for it to happen but to prevent it, which means tracking the terrorists. You could try, as the West's intelligence agencies have, trawling through billions of bits of electronic "chatter" and cumbersome surveillance. Or you could ask a bank. One British IT specialist was apparently able to construct an algorithm powerful enough to pick out likely terrorists. Factors such as having a Muslim first and second name are powerful indicators, but hardly discriminating. More telling is to examine patterns of usage – large deposits followed by lots of smallish cash withdrawals, few cheques, direct debits or regular payments, no savings accounts and – of course – no life cover. Thus, suicide bombers should take out some life cover to avoid detection by the secret services and their friends in the banks' IT departments. I'm not sure why we're telling the terrorists this secret, though. There must be a freakonomic explanation, but I'm freaked if I can figure it out.