Cash for Clunkers, the one-off housing credits, and the hiring for the census all boosted the economy and put dollars into it to increase AD. However, all those programs have ended. The car industry is still in a mess. There is still a 40+ month supply of houses and new home construction is very low to non-existent. Unemployment remains above 9% and would be much higher without birth/death adjustments arbitrarily placed on the figures. (One of the B/D adjustments is to automatically fudge in an increase in housing construction and hospitality jobs during the late spring and summer. Both of these increases account for almost all of the job growth in recent months. Data show that housing construction is very low and the hospitality sector is enduring a very rough season, meaning those B/D adjustments are misplaced in the jobs environment of 2010.)

What the increase in G has accomplished is, following a temporary statistical boost, no net gain to GDP, an increase in the federal government’s share of total USA indebtedness, and an increase in USG interest payment obligations. The net impact of the increased spending was zero.

That doesn’t mean the government should cut all programs: that’s a recipe for a disaster of a different sort. But so-called stimulus programs will stimulate the economy only as long as they are in existence. We saw the same thing in Depression-era programs. If the fundamentals of the economy remain troubled, as they do in the wake of severe asset devaluation recessions, no amount of stimulus spending will get the economy back on its feet permanently. Put another way, a true Keynesian solution means regulating the economy to prevent it from melting down in the first place, thereby keeping the fundamentals in the banking and finance sector sound enough to provide a foundation for further recovery. In those cases, government stimulus can help reduce the impact of a short recession.