Dow Boost and a (Nearly) Private Sector Solution to the Crisis

Ricardo Caballero 22 February 2009



Hope is in short supply during these trying economic times. Nowhere is this clearer than in the financial system. Since Secretary Geithner’s announcements last week, shares of the main financial institutions have yet again imploded. To make matters worse, politics has decidedly entered into the process of economic-policymaking, which makes it all the more likely that we will end up with the wrong policy response; one which is too late anyway. Talk of nationalization has become widespread, as if it were a panacea, further reinforcing the deadly spiral of fear and panic.

By now, this illness has spread to the entire world economy and has ravaged the wealth of citizens around the globe by approximately 40 trillion dollars by some estimates. This daily wealth destruction, which has no clearer and quicker indicator than in the stock market indices, has frozen consumers and companies alike, so now the real economy is in a free-fall as well. How do we stop and reverse this process?

A simple proposal

Here is a “simple” proposal: The government pledges to buy up to twice the number of shares currently available, at twice some recent average price, five years from now. (Obviously the specific numbers are only an example.)

While the policy is about future (and unlikely) interventions, the immediate impact would be enormous. In particular, it would turn around the negative stock markets dynamics, and it would allow banks to raise private capital.

The most direct effect would be an increase in the price of banks’ shares by much more than 100 percent, as the pledge puts a floor on the price five years from now, but the upside potential is huge once we get over the crisis hurdle. That is, buying equity from these banks would become like buying Treasury bonds plus a call option on the upside. By the strong forces of contagion, this rise would immediately spread to non-financial shares. Consumers, especially retirees, would see some of their wealth replenished; insurance companies would see their balance sheets improve; destabilizing short-sellers and predators would be wiped out (a la Hong Kong in 1997); and we would have the foundations for a virtuous cycle.

Amplifying effects and taxpayer costs

The second and reinforcing effect would be the stabilization of the financial sector, as banks would now possess the conditions necessary to raise private capital. Until now, banks have not wanted to raise capital because this would be highly dilutive at the current prices. Potential investors have no interest in injecting capital either because there is an enormous fear of further dilutions, especially through public interventions and, worse yet, outright nationalizations. A pledge to support the shares in the future, instead of the threat of nationalization in the short run, would reverse these bad dynamics and quickly recapitalize the banking sector.

How much will this cost the taxpayers? Most likely nothing. It is highly unlikely that the crisis will last five years, especially in the presence of an aggressive policy response, and most banks’ shares are likely to trade for many times current prices. In fact, I wrote “twice the recent prices” in my proposal to reduce the shock effect, but it may well be better to go for four times (and perhaps lengthen the period to ten years).

Separating the proposal’s two goals

Note also that the proposal’s two goals can be separated. For example, if for political or other reasons the idea of boosting the stock market is unattractive, but the private capitalization is, then the mechanism can be changed to insure only the new shares. This would allow the banks to raise capital at reasonable rather than the current fire-sale panic-driven prices, but would not offer the insurance to the existing shareholders and hence it would not be too effective in creating a stock market boom.

For example, suppose that after studying a bank’s books the government concludes that valuing the assets fairly, rather than at current market (or no-market) prices, the shares are worth five times the current market value of $1, and that in order to be resilient to an aggregate catastrophe the bank needs $5billion in capital. Then the government gives new capital enough put options so that they are willing to buy 1 billion shares at $5 each. This dilutes current shareholders, but only in that they are forced to hold a buffer of capital to ensure systemic resilience against a catastrophe, rather than being diluted through forced equity issuance at panic-drive prices.

The converse is also possible, to simply support the stock market as Hong Kong did so successfully in 1997, but without facilitating recapitalization directly. Even in this case, there is an argument to focus on the top financial institutions rather than on the broad markets. There are (at least) three reasons for this:

  • First, the largest fire-sales are probably concentrated in these institutions since nationalization talk began and the illiquidity of their assets have ravaged their equity value.
  • Second, it is not possible to assess the balance sheets of a very large number of firms with the necessary speed required, so it is better to focus on those that have the largest systemic effect and do it quickly.
  • Third, the notional government liability has to be kept under control. Focusing on the top financial institutions would likely keep the 5-years-out notional liability below half a trillion dollars.

The specific amount in each case depends upon how much capital a bank needs to survive the worst catastrophe and on what is the expected value of its shares once the bank is well capitalized.

An example

To be concrete, suppose that after the stress test and the evaluation of its books, the conclusion is that Citigroup needs $25b of new capital (which is about twice its current cap value) and that its non-panic share price is $10, about half of what it was in the third quarter of 2008 (the last “sensible” quarter). Then new shareholders would get 2.5 billion shares at the price of $10 each, which would probably require that the government guarantees between $5 and $8 a share, for a total of between $12.5 billion and $20 billion in guarantees. And if the guarantee was to be extended to existing shareholders, then it would add another $20 billion or so. These are small numbers by current standards, especially considering that the scenario where the government needs to honour these guarantees is a very small probability event. The real (expected) liability from an extreme case such as Citigroup probably does not exceed $5 billion and the mode and medians are surely zero.

Implementation issues

As always, there are many implementation issues to deal with. For example, it may be useful to customize the different multiples to each bank’s needs, and determine these on the basis of stress tests and of the extent of the liquidity discount of the different portfolios. Also, this probably needs to be done through a non-TARP facility, since the accounting of the latter is ill-designed for guarantees (these count one-for-one rather than on the basis of some expected value against the TARP funds). But these are not insurmountable problems and the market needs good news sooner rather than later. Given the extreme current pessimism, this time it may be wise to go public with the general principles of the policy before all of the specific details are ironed out.

First steps

To be concrete, and in the spirit of changing expectation dynamics, I would immediately announce a minimum package making very clear that more, but not less, will be done once more information is gathered about banks’ and other financial institutions’ needs. This package could be something along the lines of: As of today, the government guarantees for the top five banks measured by assets, share prices equal to twice their average prices for the first week of February (i.e., before the previous announcements) five years from today, for as much as twice the outstanding shares. The government reserves its right to increase these multiples and to extend the program to other financial institutions.


Editors’ Note: A greatly abbreviated version of this appeared first in the Washington Post print edition, 22 February 2009. The author wrote this on 19 February 2009.


Editors' note: This column is a Lead Commentary on Vox's Global Crisis Debate where you can find further discussion, and where professional economists are welcome to contribute their own Commentaries on this and other crisis-linked topics.




Topics:  Financial markets

Tags:  global crisis debate, bank rescue

Professor of Economics, MIT


CEPR Policy Research