Some of the commenters have put up a "perfectly efficient markets" straw man as an argument against the use of supply and demand to understand today's economy. It is probably my fault for not previously elaborating on this point.
I think there are many distortions in the labor market. See, for example, my A Century of Labor-Leisure Distortions or my Public Policies as Specification Errors. Supply and demand continue to be useful in this setting, as long as you recognize that multiple prices exist in the marketplace.
Suppose, for example, that a distortion existed because of an inefficient intermediary. For example, the price of oil might be $50 in Illinois, but $45 in Texas, because oil comes from the Middle East to Illinois via Texas. We can still talk about supply and demand for oil in Illinois. We just have to be careful that the supply of oil to Illinois embodies more than just the behavior of Middle East oil producers and trans-Atlantic shippers -- it also involves the behavior of the Texas intermediary. Or we might analyze the demand for oil from the Middle East, in which case we have to recognize that it is not just Illinois behavior, but Illinois behavior is intermediated by the Texas middle-men. So the Texas middle man appears on the supply side in one analysis, and on the demand size in the other.
Now back to the labor market. None of my posts refer to "wages" -- that is intentional. I refer to PRODUCTIVITY. This means that a whole bunch of things in the labor market appear on the supply side! That includes everything from sticky wages to employer taxes to (hypothetically -- don't lynch me!) worker laziness. You might say that makes the analysis without content because it has an excessively narrow concept of demand -- it might in principle but in practice it has enabled me to distinguish this recession from several others -- other recessions did have labor demand reductions, even under my narrow definition.
One commenter said that bad employer incentives (I guess an employer tax would fit in that category) have to be considered "demand". That comment is incorrect if the analysis treats, as mine does, the "price" as productivity. In my analysis, a payroll tax owed by employers would properly appear on the supply side of the labor market.
IMPLICATIONS FOR "PRODUCTIVITY WEEK"
Because I have put a variety of behaviors on the supply side of the market, the productivity and employment numbers by themselves do not tell us whether sticky wages, employee preferences, bad working conditions, taxes, or some other factor outside the production process caused the "Labor Supply Shift of 2008." The next phase of the analysis is therefore to investigate some of the specific distortions (get a preview of this here, here, here, here, here, here, here, here, and here.