Arnold writes:

So, if the demand for mortgages collapses, all it takes to get back to
2006 levels is for mortgage underwriters to take a 20 percent pay cut?

In a world with no discontinuities, we would not get crazy subprime
lending and sudden sharp drops in demand. The no-discontinuity world is
what classical economists are trained to work with. Too bad it is not
the real world.

The question isn’t whether discontinuities ever exist in the real world.  The question is whether they’re a big deal.  I see no reason to think they are.  Yes, a 20% wage cut probably wouldn’t have saved many jobs in a handful of occupations like mortgage underwriting where demand suddenly dried up.  But what does this have to do with the vast majority of occupations?  What does this even have to do with the vast majority of firms? 

Take Borders.  It spent months hovering on the edge of bankruptcy.  Despite online competition, Borders still had plenty of customers and revenue.  It just couldn’t quite manage to regain profitability.  Why wouldn’t a 5 or 10% reduction in wages have sufficed to save the firm and its workers’ jobs?

When an occupation or industry suddenly vanishes, Arnold can plausibly (though hardly decisively) point to discontinuities.  But when 10% of the workers in an occupation lose their jobs, or 5% of firms in an industry go out of business, continuity isn’t merely a convenient assumption.  It’s a hard fact.