Keynes, in one of his wiser moments, left us with the simple advice that "ideas matter." Their influence lingers, long after they are current -- indeed, even after they have been discredited. "Practical men," he wrote, "who believe themselves to be quite exempt from any intellectual influence, are usually the slaves of some defunct economist." As, for example, Keynes.

The paradox of our times is that even as it has become a commonplace that "Keynes is dead," Keynes lives. His specific prescription, of deficit spending as the balance wheel in the business cycle, may no longer be much in favour -- as Preston Manning likes to say, if deficits were the cure for unemployment, everyone in Canada would have three jobs by now -- but the broader notion, that unemployment and economic growth are best approached in the aggregate, as mere functions of aggregate demand, remains fixed in the public consciousness.

To the problem of a surplus supply in the market for labour, that is, the answer most commonly proposed is still to boost the demand, not just for labour, but for everything: to stimulate overall economic growth, in hopes thereby of increasing employment. Yet it is equally commonly believed that this is impossible. And rightly so: if there is anything that the last 30 years should have taught us, it is that there is no straight-line function connecting increased demand to higher growth to higher employment. Any amount of demand stimulus is theoretically consistent with any amount of growth, which is in turn compatible with any amount of unemployment.

We stumble about for a solution to chronic 10 per cent unemployment, then, in a weird game of blind man's bluff, rejecting Keynesian remedies yet unable to think in anything but Keynesian terms. There remained a certain coherence to the debate so long as it centred on whether or not the Bank of Canada should take up the slack in demand, i.e. to use the monetary lever, rather than the fiscal -- even if the acceptance that "money matters" was itself a long-forgotten victory for Keynes's arch-foes, the monetarists. But now that that, too, has been more or less put aside, what we are left with is a very odd assortment of joblots, patches and schemes.

The most Keynesian-sounding of these is the infrastructure program. To listen to Liberal rhetoric during the last election, you'd have thought the year was 1933, not 1993. To be fair, the Red Book only claimed the $6-billion program would create 100,000 short-term jobs, which would hardly make a dent in the overall unemployment numbers even if it were true, but who has time to read these days? What people saw on TV was the Liberal leader promising a massive public works program, vintage R. B. Bennett, to stimulate the economy and create jobs.

As a traditional, Keynesian bit of pump-priming, financed from increased public borrowing, this would be as debatable as ever. But carried out, as it was, in concert with a massive reduction in federal and provincial deficits, the program dissolves into incoherence. Whatever hydraulic effect might be imagined to flow from increased government spending on roads and bridges -- and bocce courts and stadium skyboxes, but never mind -- was fully offset, and then some, by the reduction in spending on health care and education.

Indeed, even had the infrastructure program resulted in a net increase in spending, the effect would be less to create jobs than to trade them, for the jobs created building roads and bridges -- and cultural centres and ski resorts, but never mind again -- only equal the jobs destroyed by the taxes that financed it, or the investment diverted from other industries and firms, less fortunate or less well-connected than the program's beneficiaries. And of course, as the Auditor General's recent report points out, a third or more of spending on the program went to projects that would have proceeded in any event.

Remarkably enough, even the government's own studies confirm the program's irrelevance. An internal report carried out by a professor of engineering at the University of Toronto for the federal Canada Infrastructure Works Office, which an accompanying press release claims describes the program as "successful" (apparently, none of the bridges fell down) in fact has this to say: "improvements in the economy, as measured by changes in GDP and total investment, are estimated to be comparable to the debt reduction alternative" -- comparable, that is, to not spending the money at all.

In other words, the program had essentially the same short-run impact as if it had never happened; "longer-term impacts would clearly favour the debt reduction case." But then, as someone once said, in the long run we are all out of office.