US treasury secretary Tim Geithner’s bank rescue has been poorly received by the markets. My proposal to create brand new “good banks” with the limited taxpayer resources available is, instead, the best solution to the crisis.
One reason the Geithner plan has been poorly received is that the money isn’t there to recapitalize US banks as a whole. Geithner has only $350 billion, what’s left of the original $700 billion in Troubled Asset Relief Program. That’s nowhere near enough to get Geithner’s proposed public-private investment fund going on any significant scale.
Given the limited scope US authorities have for increasing the public debt burden without adverse asset market responses, it is best to forget about tax cuts or public spending increases. Instead, the available fiscal resources should be focused on restoring the flow of credit to non-financial enterprises and, to a lesser extent, to households (most of which are already over-indebted).
Rather than wasting the $1.4 trillion of public funds it would take to restore the capitalization of the US banking sector to its fall 2008 level, it would be better to use public money to capitalize new banks that don’t suffer from an overhang of past bad investments and loans—and to guarantee new borrowing or new loans and investment by these banks. This “good bank” model achieves this by identifying the systemically important banks that are kept afloat only by past, present and anticipated future public financial support (“bad banks”) and taking their banking licences away.
The “stress test” proposed by Geithner for major banks (assets in excess of $100 billion) could be used to gather the necessary information to identify the bad banks. New banks, capitalized by the government (possibly with private co-financing) would take the deposits of the bad banks and purchase the good assets from the bad banks. Future government support, through guarantees or other means, would be focused exclusively on new lending and new borrowing by the new good banks and those old banks that passed the stress test.
The legacy bad banks would not be allowed to make new investments or new loans and would simply manage the inherited stocks of assets in the interest of their owners. They sink or swim on their own. If they fail, their unsecured creditors can figure out what to do with the bad assets.
When public resources are scarce, they should be concentrated not on supporting the valuations of existing impaired or toxic assets but on encouraging new flows of lending and borrowing, for which success or failure is still to be determined. To decouple flows of new lending from existing stocks of bad and toxic assets, a legal and institutional separation between the owners of the bad assets and the investors in the new assets is necessary. This objective is achieved by the good bank model.
This model is better than full nationalization, because it does not require the government to trust the valuation of toxic assets implicit in the market capitalization of the banks that own them. It only requires the valuation of good assets. It is better as a recession-fighting policy because it stimulates new lending to the real economy more effectively than would an injection of capital into the existing banks, for which old toxic assets act as a tax on new lending.
The good bank model is also better from the point of view of moral hazard because it does not reward past reckless lending and investment. And it is fairer, because the losses on past failed investments are borne by those who made the bad decisions rather than by taxpayers.
THE WALL STREET JOURNAL
Edited excerpts. Willem H. Buiter is professor of European political economy at the London School of Economics and Political Science. Comments are welcome at firstname.lastname@example.org