Among the more reliable indicators of an impending recession is the onset of a rash of press clippings on the death of the business cycle. An article in a recent issue of The New Republic drily records the title of a 1969 academic conference -- Is the Business Cycle Obsolete? -- before going on to reach that very conclusion. It isn't just that the economy is on a roll, and shows no signs of slowing. Officials at the U.S. Treasury, it reports, now believe that "through correct policies, governments are able not only to stem contractions once they begin, but they can also extend expansions indefinitely." Business Week dips its toe in the same optimistic waters ("How Long Can This Last?"), but gives more credit to the private sector. "If companies really have mastered the art of improving productivity," it is bold enough to predict, "there's no reason to expect the bad old days of slow growth and rising inflation to return any time soon." Indeed, as far as Wired magazine is concerned, the party's just begun ("we're facing 25 years of prosperity, freedom and a better environment for the whole world"). It's not clear what's behind The Long Boom, as the magazine calls it, but it seems to have something to do with the Internet.

In principle, the optimists are right -- or at least, they're not wrong. Certainly there is no inevitability to the business cycle, much less the sort of metronomic regularity the term would suggest. Contrary to a much-used anthropomorphic metaphor, the economy does not simply "tire itself out" after so many years of expansion. It can still be thrown into recession by a chance event or "shock" -- war in the Persian Gulf, say, or a financial crisis in Mexico -- but the mere passage of time is no guarantee of the sort of speculative excess in asset markets that is often blamed for past recessions.

Indeed, rather than inquire whether the business cycle is dead, we should more be inclined to ask why recessions should occur at all. The long-run rate of growth of output is determined mainly by the pace of advances in productivity. While productivity growth might be expected to fluctuate somewhat from year to year, this is nothing like the sort of wild swings in output that have been typical of recent decades.

The truth is that recessions, certainly of the modern variety, are almost always policy-induced: that is, they are the product not of time or chance but of government action. They are almost always policy-induced because they almost always follow hard upon an outbreak of inflation. And inflation is always the result of government policy: namely, a too-fast expansion of the money supply.

Why did governments in the modern era pursue inflationary policies?

Precisely to compensate for lagging growth. Seduced by the rapid increase in productivity in the decades after the Second World War, policy makers came to look upon real growth in excess of 4 per cent annually as the norm.

What is more, the new theories of macroeconomic policy seemed to offer governments the means to correct for any slowdown in growth, from whatever cause. At the worst, it would come at the cost of a little inflation, a relationship made famous in the Phillips Curve.

This made sense to everyone, until they actually tried it. The result was 30 years of rising inflation, accompanied by slower and slower rates of economic growth -- and punctated by painful recessions. The confidence of an earlier generation that the business cycle had been tamed was indeed hubris, based as it was on the belief that governments could put a floor under real growth rates. But what if, instead of trying to prop growth up, governments concentrated on keeping inflation down? That would not preclude "shock" recessions. But it would at least rule out the policy-induced kind, which is to say, almost all of them.

Optimism about inflation is in fact behind much of the current round of eulogies for the business cycle. Where the "it's different this time" school goes wrong is in ascribing inflation's quiescence to external factors, whether the latent productivity gains we are now starting to realize from the information revolution or the heightened competitive pressures of a global marketplace.

Rising productivity does not in itself spell the end of inflation: with enough monetary stimulus it would still be possible to whip prices into a froth.

Faster productivity growth only means that the economy can absorb that much more money into circulation, and therefore support higher rates of spending and incomes, without giving rise to inflation.

At bottom, inflation remains a consequence of policy. If there is reason for optimism, it is not that inflation is so low, but that it is so low after being so high: we have, it may be hoped, learned our lesson. That is what makes today's sub-2 per cent inflation different than the same rate in 1962. Today, the wounds of our thirty-year struggle with inflation are still fresh. Then, they could not even be imagined.