Just what is it that makes owning a Ferrari so satisfying? Robert Frank, who seems to know his cars, suggests the 'thunk' that the door makes when it locks into position. Ferrari owners might say that the thunk is an indicator of the engineering quality that they have paid such a lot to acquire. In other words, they would argue that their satisfaction derives from the absolute value of the commodity.

Classical economic theory would agree with them. And this, Frank argues, is where it goes wrong. If there were no Ferraris, and Porsches were the most prestigious cars available, then the rich would derive exactly the same satisfaction from the thunk of a Porsche door. Satisfaction arises not from absolute wealth, but from one's position relative to others.

This simple insight, Frank believes, is the key to ameliorating the undesirable consequences of the 'winner-take-all' markets which are becoming an increasingly prominent feature of the economy. These effects are specified in the synoptic subtitle of The Winner-Take-All Society (Free Press, £16.99), which he co-wrote with Philip J. Cook: "How more and more Americans compete for ever fewer and bigger prizes, encouraging economic waste, income inequality, and an impoverished cultural life."

To illustrate the dynamics behind this process, he draws upon an idea central to modern Darwinian theory, that of the 'arms race'. A winner-take-all market, he observes, is something like a polygynous species in which a handful of males enjoy a sexual monopoly. In such species, males go to absurdly wasteful lengths to compete with each other for a place in the tiny minority that reproduces itself. The huge and unwieldy antlers of the elk are one result. Another is the 1,800lb Californian sea lion which last week was reported to have crushed up to 50 females to death each year in the course of mating.

Frank himself looks like a man who knows he doesn't have to join in any arms race: the polo shirt and chinos are suitably Ivy League - he is Professor of Economics, Ethics and Public Policy at Cornell University - but comfortably rumpled; the lean physique is that of a man who plays squash but doesn't have to. No professorial grandeur either; but a infectious enthusiasm for communicating his ideas - which, along with a selection of New Yorker cartoons and an impressive display of familiarity with evolutionary biology, won him an enthusiastic reception at the London School of Economics this week.

The forum for his lecture was the latest of the Darwin Seminar series. These events, which LSE's director John Ashworth has shrewdly predicted will come to be recognised as a major contribution to the development of a new intellectual synthesis, have explored themes ranging from the genetics of pregnancy to warfare in Ancient Greece. Contrary to the popular misconception that Darwinism is intrinsically Hobbesian, a grim vision of war of all against all, contemporary Darwinism is much concerned with co-operation - albeit as a phenomenon which needs to be reconciled with self-interest.

Bringing a Darwinian perspective to bear on economics, Frank has humanized the dismal science. (His interest in biology, he reveals, first arose from his perception of temperamental differences between himself and his adoptive parents.) In his path-breaking book Passions Within Reason: The Strategic Role of the Emotions (Norton, 1988), he pointed out that irrationality often pays. It would be irrational to hit somebody who parks in your parking space, given the likelihood of prosecution, but if you become known as the kind of person who would do that, your neighbours will keep their cars off your patch. So far, so Hobbesian. But Frank went on to note that in situations requiring trust - in other words, most human situations, from business to marriage - excessively self-interested people often fare worse than those with an irrational commitment to abstract ideas of fairness.

In their recent book, Frank and Cook explore another dimension in which the 'rational actor', the stick-figure of classical economics, may be inadequate as a model of how people work. The 'winner-take-all' market, long familiar in the entertainment industries, is one in which a small difference in performance may lead to an astronomical difference in reward. A fraction of a second may separate the winner and the runner-up in a race, yet it is the winning athlete who collects the gold of the merchandising deals. More and more markets, from boardrooms to hospitals, are acquiring a winner-take-all character, in which rewards depend on relative rather than absolute performance, and accrue disproportionately to those at the top. Instead of finding realistic and productive niches - as engineers or teachers, say - people rush to compete for elite positions in law or accountancy. They do so, in part, for the same reason that leads 80 per cent of people to think that they are better than average drivers. It seems to be human nature to overestimate our abilities.

Frank readily acknowledges the benefits of winner-take-all markets. "There are some people who are better than others, and the technology lets us get access to the best people," he observes. "Have you ever heard Kathleen Battle's recording of the Mozart arias? You'll listen to it and you can't imagine that anybody'll possibly do it any better." But, he says, "we could get most of the gains we get from this process at a fraction of the cost".

The solution he offers is a change in the tax system. Frank and Cook's big tax idea is that we should be taxed not on our income, but on our consumption - directly, via our tax returns, rather than indirectly through levies such as VAT. The taxable sum would be what's left after our savings (and personal allowance) have been deducted from our income. Under such a regime, we would avoid tax by hanging on to our money, and we would gain extra income from the interest on these savings. A progressive consumption tax would reduce the effective rewards of superstardom, curb the 'arms race' in which people compete by out-consuming each other, and encourage savings which could be invested in enterprises or social desirable projects. In short, Frank argues, his tax can "create resources virtually out of thin air".

The obvious question is, why has nobody come up with this simple idea before? "The economic establishment doesn't acknowledge that satisfaction depends on position," answers Frank. "All the Adam Smith models assumed that your satisfaction depends on how much you have in absolute terms, not where you stand positionally. That simple one-line amendment to the model makes all this turn out differently." The unamended model underlies the standard justification for 'fat cat' executive salaries in Britain and the US. However, it fails to explain why chief executives in Germany and Japan bother to turn up for work at all. Their pay is 21 and16 times higher than that of their workers, whereas their American counterparts earn 150 times as much. (Positional satisfaction arises from comparisons with our close rivals, Frank argues, rather than with those a great socioeconomic distance away.)

He specifies the effect of his tax on the rich precisely. "You'd go from a $207,000 Ferrari to a $105,000 Porsche 911 Turbo, and with the tax, the Porsche would cost $30 or 40,000 less than the Ferrari still, so that's money you'd have left to put in your savings account."

But what would his tax mean for someone at the level where proposed tax thresholds tend to hover, on £35,000 a year? Frank answers as if his interviewer were Steve Coogan's character Alan Partridge, whose social vision is calibrated in car model numbers. "The progressive consumption tax would generate a ripple effect all the way back down the economic totem pole," he argues. "The Ferrari would drop off the table; the Porsche would fill its niche. What about the people who drove the 911? Well, they can now drive the Nissan 300ZX or the Toyota Supra Turbo. Then there's the BMW Z3 roadster ...  "

As for the person at the £35K mark, "his tax wouldn't go up at all, really," Frank says, sounding like a politician for a moment. "But now that person's able to derive the same subjective satisfaction from owning a car that is cheaper than the car he now drives. The Ford Escort's going to have new lustre!"

"There's a well-known pattern in cognitive psychology called loss aversion," he notes. "People will fight bitterly to keep from losing a pound; they won't fight nearly as hard to gain a pound. My sense is it's related to territoriality in animals. An animal will fight bitterly to keep the territory it has, and won't expend nearly as much energy to acquire new territory. There seems to be something of that heritage in human nature.

"Any time you make a proposal for a change in policy, you inevitably generate some winners and losers. The people that stand to lose in the short run dig in their heels and fight bitterly to oppose the change, whereas those who would benefit are happy about it, but don't work nearly as hard in favour of the change. So you've really got to work hard to change anything ever."

But if the proposal is, as he claims, "unassailable", it will succeed by Darwinian mechanisms of competition. "Eventually some country will do this, and it will work, and others will copy it. We'll look back and say, 'how could we possibly have not stumbled onto this earlier?'"