There has been a lot of talk scattered across the tubes about consumers "deleveraging" as in actually reducing the amount of debt they owe. (Deleveraging is not really the appropriate term for consumers because they don't "lever up" as financial companies do in an attempt to control assets far in excess of what their own capital could achieve. But I digress.) The Flow of Funds Report (Z.1) showed that consumers had stopped accumulating new mortgage debt, but that consumer debt was still increasing. A more current data set (G.19) shows that accumulation of consumer debt has stopped. (Note that this data set is in nominal dollars so the YoY numbers are NOT adjusted for inflation.)
Revolving credit (mostly credit cards) is actually being paid down in both nominal and real terms. Non-revolving credit (mostly car loans) is still increasing slightly in nominal terms and is probably decreasing in real terms.
I'm less familiar with this data set as of now, so I don't know how far consumer debt "should" be paid down. Eyeballing the chart suggests that debt levels of 10% of GDP for non-revolving credit and 5% for revolving credit might be good targets. But since revolving credit is usually very-high interest credit card debt, that target should probably be lower. If the targets were 10% and 3% of GDP, that would mean consumer credit would fall by 5% of GDP, or $700B. That translates to about 7% of 2008 consumption - which is not to say the debt would all be paid off in one year. But even spread over 4 years the hit to the economy would be huge. Unfortunately, it has to happen at some point.