Today Obama announced a largely symbolic gesture to freeze federal employee pay for 2 years. That's all fine and good, but I've been thinking about this as a potential broader policy too, particularly in relation to some post-Keyensian economists who are calling for dramatic 'employment' tool(s). They are in favor of a tool that basically negates the need to try and effect unemployment indirectly with interest rates and spending, and replacing it with a tool whereby the government would hire droves of people during recessions (kind of like a 'New Deal' instrument). I've mentioned before I'm not in favor of such an idea in large part because it sounds needlessly bureaucratic, full of moral hazard, and could theoretically permanently increase the size of government to a dramatic and largely inefficient level.
However, what if the instrument were not employment levels but the actual wage levels - private wages - the government could control temporarily. It wouldn't be bureaucratic since the government could pronounce say that, until such time X, Y, Z (sufficient employment/GDP growth has been attained), job-earned wages (including bonuses) are to be kept frozen across the board for EVERYONE. This in effect would be a way around the New-Keynesian 'sticky wage / sticky price' problem. According to classical theory if wages were allowed to freely adjust in recessions, employment would not fall as hard. Economists point out that wages often don't fall downward (or even stagnate - many keep growing) due to union contracts, social pressure/norms, maintaining a sense of morale, etc. But what if the government had all the power to set wages during extreme downturns? That would take the stigma off of private employers.
It wouldn't be without issues: timing would still be important. Rules vs. discretion would be as well. I would imagine such a program wouldn't work (politically or otherwise) if it was left to discretion, but if certain rules were enacted as law, it might stand a chance. What the rule is is best left to economists more mathematically inclined that myself. I suspect tying it positively to some level of inflation would be useful such that the real wage keeps a level of stability. Unions wouldn't be particularly happy about governments temporarily over-riding their contracts, etc. But, I think some of them would come around when they saw that the alternative was significant unemployment. Perhaps the biggest issue is the same big issue with all forms of fiscal/government-mandated stimulus - it all starts with politics, not economics. That is to say, economics cannot be separated from its political cousin in the real world. Devising a program rule that would withstand changes in political climate or parties should be paramount - but how one would go about doing that is a bit beyond me at this point....
At best, it could ease unemployment in recessions - create a more direct target for unemployment. At worst, it could provide evidence for or against the very underpinnings of New Keynesian economics.