I should make clear from the outset that wage controls are not my preferred method of enhancing the Bank of Canada's credibility in the fight against inflation. A return to monetary targets, and bigger cuts in the deficit would be the means of choice. But given the reluctance of the relevant authorities to consider either of these, we are forced to improvise.

Wage controls are usually advanced as a complement to monetary loosening; certainly that is what the Canadian Manufacturers' Association had in mind last month when it called for a public sector restraint program similar to the ''6 & 5'' plan of the early 1980s. In the naive form, they are imagined as an alternative to monetary restraint in the task of containing inflation. This is muddled, however, since inflation is not caused by rising wages, or by any other cost factor, but is, by definition, a monetary phenomenon.

The more sophisticated rationale begins from the understanding that inflation will indeed accelerate if monetary policy is eased. Allowing prices to rise, while keeping nominal wages in check, would bring about a reduction in real wages; lower labor costs would allow employment and output to increase. The Macdonald Commission, for one, was an advocate of this approach.

Apart from the moral and political questions involved in filching workers' wages in this manner, several difficulties present themselves. Either controls are temporary, or they are permanent. If temporary, then the pressures built up by monetary expansion will cause wages to balloon the minute controls are lifted. If permanent, then the distortions in the labor market accompanying controls will accumulate to the point of breakdown. Whichever, inflation will carry on all the while, creating its own distortions.

But wage controls that accompany a monetary tightening are a different matter. Here the intent is not to produce a job-creating cut in real wages by allowing prices to run ahead, but to prevent a job-destroying rise in real wages that might result from holding prices back. If Bank of Canada Governor John Crow is serious about cutting inflation to zero - and, since that's his job, he ought to be - then current annual wage increases of 5% to 6% are a real concern.

The Canadian labor market is encrusted with enough ''rigidities'' - union scale, nationwide bargaining, pay equity, minimum wages, and so forth - that one may doubt whether wages would adjust downward fast enough in the event of disinflation. Liberalizing labor markets is the long-term prescription, but in the short term this again is unlikely. We needn't call them wage controls: How about a Zero-inflation Adjustment Program (Slogan: ''ZAP!  You're frozen'')?

The point of controls, then, is not to serve as a substitute for monetary discipline, but as its complement. Rising wages are a concern not because they increase inflation, but because they raise the costs of reducing it. And so far as controlling inflation means heavy losses in employment and output, that in turn impairs the Bank's credibility in the international investment community, which will reckon with some justification that this would prove more than the governor's political masters could bear - assuming the governor himself could. Betting on higher inflation, they will demand higher interest rates. If they're right, we get higher inflation. If they're wrong - if the governor proves tougher on inflation than they thought - we get a recession, since real rates would be that much higher.

Such are the paradoxes of monetary policy when expectations are taken into account: Only the demonstrated willingness to contemplate a recession can ultimately forestall one. But pledges to conquer inflation whatever the costs are a whole lot more convincing if the costs themselves are not that great.

Tight-money wage controls, unlike their loose-money counterparts, could be temporary without losing their effect, since in a tight-money environment there would not be the buildup of demand pressures that might produce a wage explosion after decontrol. Once prices were stabilized, controls could - and should - be lifted.

More ambitiously, wages could be reduced more sharply during the transition than inflation, in an attempt to capture some of the employment gains associated with loose-money wage controls: instead of 6 & 5, perhaps 2 & 1. With any luck, having reached zero inflation, there would be less impetus for labor afterward to catch up.

If wage controls are justified, then there is no sense applying them only to the public sector: Make them across the board. But if controls are abhorrent, then let's prove we're serious about fighting inflation in some other way. We cannot simply tighten policy and hope for the best.