Some theories of the business cycle say that booms and recessions are caused by shocks to the supply of key inputs: a wave of new high-tech ideas, a rise in oil prices, big changes in labor law.  In supply-side stories, it’s easy to understand why a recession happens: You just can’t make as much stuff.  Other “multipliers” or “propagation mechanisms” may amplify the effect but the first shock is to the supply side.  

Other theories of the business cycle focus on shocks to the demand for goods and services: People get cautious, firms get worried, governments get terrified, so they hold back on spending.  After hearing a demand-side story anyone who has taken a few weeks of economics should ask, “Why don’t businesses just cut prices and sell the same amount?”  In response there are many tales of how it takes a while for businesses to cut prices even though there’s lots of unsold stuff on the shelves—stories of price rigidity, of sticky prices.  In these demand-side stories–whether Monetarist, New Classical, or sticky-price Keynesian–stuff goes unsold even though workers are willing and able to supply inputs, even though firms stand ready to meet orders for goods.  If firms and workers produced the stuff, it just wouldn’t sell: Prices are stuck too high. 
I think both the supply side and demand side stories have a lot to be said for them, and real life is likely a mix of both plus some of neither, but that’s not what I’m here to write about today.  Instead, I’m here to remind you that one popular brand of Keynesianism–sticky wage Keynesianism–should be lumped in with the supply side theories of recessions rather with the demand side theories. Sticky wage Keynesianism says workers hate wage cuts so much that businesses dare not cut wages, even in a competitive market.  There are a lot of stories about why wages don’t fall during a recession revolving around fairness norms but I’m here to talk about the effects of that wage rigidity not the causes.  

In sticky wage Keynesianism, demand for goods falls because of, say, bad news about the stock market. Since businesses want to sell stuff, rational individual firms cut their prices in response to bad news.  But if firms cut their prices while keeping worker wages fixed, firms find workers more expensive than before–workers become more expensive in terms of goods.  Because workers–a key input–become more expensive, firms rationally decide to produce less stuff.  

It’s basic microeconomics: If the price of a key input rises, you’re quite likely to produce less output.  That’s the heart of sticky-wage Keynesianism.  
You can see now why it’s a supply-side theory: If, magically, some extra output showed up in a company’s inventory, in a sticky-wage Keynesian world it would sell and probably sell quite quickly because end users are starved for output: there’s no failure of “effective demand” for final goods in a sticky-wage Keynesian world.  The reason there’s so little output during a recession according to sticky-wage Keynesians is because high wages make output too expensive to produce. 

Sounds like a supply-side theory to me.  

Yes, it’s a supply-side story that can be fixed by raising the demand for goods–by printing money, by ordering more government goods–but these solutions are ultimately Rube Goldberg devices for raising the price level so that workers become cheaper so they can go back to producing output.  The core economic story in sticky-wage Keynesianism is about firms that are rationally deciding to produce less output not because of a fall in demand for goods but because of a rise in costs.  

I think supply-side failures are important to the business cycle, and the sticky-wage Keynesian channel is one of those supply-side failures.  But I’m no business cycle monocauser: I suspect there’s some straightforward demand failure going on during recessions as well.  

For discussion: There are no overstocked shelves in a simple sticky-wage Keynesian world.  Do you think we live in something a lot like that world?