What we’ve had is a sharp increase in the desired private surplus at any given level of GDP, due to a combination of higher personal saving and reduced investment demand. This is shown as an upward shift in the private-surplus curve. In the 1930s the public sector was very small. As a result, GDP basically had to shrink enough to keep the private-sector surplus equal to zero; hence the fall in GDP labeled “Great Depression”. This time around, the fall in GDP didn’t have to be as large, because falling GDP led to rising deficits, which absorbed some of the rise in the private surplus. Hence the smaller fall in GDP labeled “Great Recession.” What Hatzius is saying is that the initial shock — the surge in desired private surplus — was if anything larger this time than it was in the 1930s. This says that absent the absorbing role of budget deficits, we would have had a full Great Depression experience. What we’re actually having is awful, but not that awful — and it’s all because of the rise in deficits. Deficits, in other words, saved the world.
If we consult table 1.1.6 of BEA’s NIPA tables (real gross domestic product and its components in 2000$), we see very little change in either real GDP or consumption but an almost $450 billion decline in investment from 2007QI to 2009QI. Government purchases rose by approximately $90 billion with the rest of the offset coming from an improvement in net exports. However, real exports fell by over $35 billion so one of the saving graces was the fall in imports. A positive marginal propensity to import tends to reduce the Keynesian multiplier and hence works as another form of automatic stabilizer.