Why are apparently good workers unemployed for so long in recessions? One channel that Tyler has promoted is ZMP: Some workers are almost completely unproductive, so employers can them rather than have them eat up all the pizza on Pizza Fridays.  

One might debate the pros and cons of the idea.  And of course it’s not to be taken literally: Perhaps as in Sargent and Ljunqvist’s story about high European unemployment the “ZMP” workers just lose half of their productivity due to, say, job-specific technological change.  As Patsy says in Holy Grail, only a model
But I’m not here to debate the existence of ZMP: I’m here to note that mainstream macroeconomics already believes in ZMP.  Any story of recessions that relies heavily on the story that “stuff doesn’t get sold because businesses don’t cut their prices enough” is a story of ZMP workers.  If price rigidity is important–and at least some evidence suggests that it is–then when total dollar spending falls, some of the hit to spending shows up as a fall in the quantity of output and not just as a fall in the price of output.  

In a sticky-price slump, there’s less real demand for goods so output really is “demand-constrained.”  And if output can’t be stored then low demand for output means that it’s only valuable to have as many workers as you need to meet demand.  People used to by 1000 burgers a day but now they’re only buying 900?  The restaurant manager knows how to produce 900 burgers and it’s not by paying workers to just stand around.  Especially when the machines and equipment to make the burgers are already sitting right there, it’s obvious which input you’ll cut back on: work hours. It takes fewer labor hours to make fewer burgers, end of story. Any extra labor hours are just, well, ZMP.  

At this point, the only way to destroy the simple sticky-price ZMP story is to say that unemployed workers will cut their wage demands so low that the restaurant manager decides to spread the fixed amount of work across more workers.  But if it costs a fixed amount to keep each individual employee on the rolls—due to paperwork costs, health care, management attention–then spreading the work around becomes really expensive (Eli has more on the related topic of labor nonconvexities). The hourly wage would have to fall a lot to make work-spreading reasonable, so low that we’re getting back to, well, ZMP. 

Keynesians, New Keynesians, and Monetarists alike use price rigidity as part of their toolkit to explain why a fall in nominal spending turns into a decline in real spending.  And lurking in the background of their stories is the idea that if real spending falls, so do the number of labor hours needed to make the stuff that people are buying.  Any extra work hours are, well, ZMP.