From December 2007 through September 2008, we committed $537 billion. That is not money spent, but it is money put at risk by guaranteeing various financial instruments, and by loaning to banks that could not otherwise get loans.
But in September 2008 the Fed went nuts, nearly doubling the commitment to $1097 billion, or around $5500 per taxpayer. (There are around 200M taxpayers, right?)
And then in October, the FDIC joined the Fed, and, according to the NYT, together they upped the commitment to $5069 billion. I am not following the NYT very well here, however, since it appears that $1600 billion of this refers to the size of the commercial paper pool, and not the size of the projected government purchases within that pool. So it's not like the government is actually selling $5 trillion of T bills.
It appears the Fed has become the world's underwriter. The commercial paper thing appears pretty transparent, for instance. The Fed sells T bills at 1% interest, and buys commercial paper at 5 to 10% interest. So long as the default rate is lower than the spread, the Fed makes money. Given the size of the money flow here, it is possible that the Fed could either make or lose amounts similar to the size of the national debt over the next couple of years.
Of course the problem is that someone at the Fed has to decide what rate they want for paper from which companies. Since the decisions required are vast -- they have to price the entire commercial paper market -- one presumes the same people are doing this that just presided over the credit default swap implosion. So, it seems like we might be more likely to headed for the "likely to double the national debt" outcome and less likely to end up at the "paid back the national debt" outcome.
This is Macroeconomics, for real. Wow. It really makes you wish there were a way to get off this train.