Stiglitz on inequality and unemployment

From Stiglitz, The Price of Inequality (2012), pg. 85:

Moving money from the bottom to the top lowers consumption because higher-income individuals consume a smaller proportion of their income than do lower-income individuals (those at the top save 15 to 25 percent of their income, those at the bottom spend all their income). The result: until and unless something else happens, such as an increase in investment or exports, total demand in the economy will be less that what the economy is capable of supplying - and that means there will be unemployment. In the 1990s that "something else" was the tech bubble; in the first decade of the twentieth-first [sic] century, it was the housing bubble. Now the only recourse is government spending.

Unemployment can be blamed on a deficiency in aggregate demand (the total demand for goods and services in the economy, from consumers, from firms, by governemtn, and by exporters); in some sense, the entire shortfall in aggregate demand - and hence in the U.S. economy - today can be blamed on the extremes of inequality. As we've seen, the top 1 percent of the population earns some 20 percent of its income, a shift of just 5 percentage points to the poor or middle who do not save - so the top 1 percent would still get 15 percent of the nation's income - would increase aggregate demand directly by 1 percentage point. But as that money recirculates, output would actually increase by some 1 1/2 to 2 percentage points. In an economic downturn such as the current one, that would imply a decrease in the unemployment rate of a comparable amount. With unemployment in early 2012 standing at 8.3 percent [the same as today's figure! - Ed.], this kind of a shift in income could have brought the unemployment rate down close to 6.3 percent. A broader redistribution, say, from the top 20 percent to the rest, would have brought down the unemployment further [sic], to a more normal 5 to 6 percent.