Not many things have emerged from the quagmire of US Congress recently which have produced a truly pleasant surprise. But could the troubled asset relief programme - better known as Tarp - turn out to be one?
That is the question hanging in the air this week, as the US government kicks off the process of selling the 92 per cent stake in AIG it acquired in 2008, during the crisis. At first glance, this week's news from AIG does not look particularly encouraging. The share price of the giant insurance group has fallen 40 per cent this year, forcing the Treasury to slash the size of its planned offering.
But while this is disappointing, it is probably just a speed-bump on a rather striking road. Taken as a whole, the results from Tarp have been dramatically better than cynics (including myself) ever imagined. That is something that European leaders would do well to ponder, as they mull their own, costly banking mess.
The numbers are striking. Back in 2008, when Tarp was first launched, $700 billion was earmarked for the programme. Some doubted whether $700 billion would even be enough and predicted the imminent arrival of Tarp II.
But what actually appeared was Tarp-lite: the Treasury subsequently declared that it "only" needed $475 billion, not $700 billion, to save the system, and now expects to get (almost) all of this money back. Right now, for example, the official accounts show that the Tarp has spent $245 billion supporting banks, $151 billion bailing out AIG and the auto groups, $1 billion supporting the housing market, and a further $17 billion on credit support programmes. However, $296 billion has already been recouped from banks, via asset sales and the repayment of capital injections.
And when other asset sales are completed, the Treasury expects to recoup almost everything, apart from a $40 billion-odd fund earmarked for housing. Moreover, since it expects to make profits from other government financial support programmes, such as the Fed's purchase of assets, the overall tally from the crisis- fighting measures is now projected to be a $24 billion profit - at least in the Treasury accounts.
Now, it must be stressed that this "accounting" excludes some key things. It does not capture the wider damage to the economy from the financial crisis. Nor does it show what might happen to banks - and Tarp - if asset prices fall again (and remember those assets have been inflated by the ultra loose monetary policy.) The subtle but significant "cost" of moral hazard is ignored, and some losses may have been shuffled into opaque corners.
But even allowing for those caveats, the result is still amazing - at least by historical standards. Three years ago the International Monetary Fund examined the 40 global banking crises that occurred in recent decades, and concluded that "fiscal costs, net of recoveries, associated with crisis management ... [were] about 13.3 percent of gross domestic product on average and can be as high as 55.1 percent," with an average recovery rate of "18.2 percent of gross fiscal costs".
The US Savings & Loans crisis, for example, delivered a fiscal hit of 3 percent. This time, however, the bill is under 1 percent.
So what are the lessons? One is that it pays to be big: America has found it easier to support its banks than, say, Ireland, because they are a smaller proportion of GDP. Another lesson is that it pays to be both decisive and dramatic with policy announcements - but flexible in implementation. The fact that America announced Tarp with such fanfare in 2008 helped to calm markets, particularly since it was followed by stress tests.
However, once the Tarp pot was in place, American officials kept experimenting with different ideas about how to spend that cash. What is perhaps most interesting for students of financial history, for example, is that the US did not embark on asset auctions (as it did with the S&L) mess, or attempt to create "bad" banks; instead, it pursued a general reflation, and encouraged banks to remove bad assets themselves and raise capital. As such, this seems more successful than some observers (including myself) expected.
But perhaps the biggest political lesson is about America's political economy. US officials like to claim today that Tarp shows the value of being "resolute" in banking crises. However, the dirty secret is that America procrastinated shamefully for a year before Tarp was produced. And even then, Tarp was initially rejected by Congress, before a market crash shocked politicians into a deal.
This has important implications for the fiscal debate. Tarp shows that America can break political gridlock. Better still, this can sometimes produce better than expected results. But, the dismal truth is that it took a full-blown market crisis to spark the crucial policy switch.
If America is ever going to produce a proper, bipartisan deal on its debt, in other words, it will probably need to suffer a bond market shock first.