The following op-ed, “How to give banks confidence to lend to businesses,” was co-written by Harvard Law School Professor Lucian Bebchuk LL.M. ’80, S.J.D. ’84 and Italy Goldstein, a professor at the Wharton School of Business. It appeared in the December 19, 2008, edition of the Financial Times.

An important aspect of the economic crisis has been the drying up of credit that US banks normally extend to Main Street companies. Borrowing by businesses remains costly and difficult, with spreads between yields on corporate bonds and treasuries at extremely high levels.

Why does credit fail to flow despite the infusion of so much additional capital into the financial sector? The Treasury has been arguing that banks still lack confidence and we just need to give them time to adjust. The chair of the congressional oversight panel has suggested that banks’ reluctance to lend reflects their rational assessment of borrowers’ bleak prospects. But there is a third explanation: banks may not be lending because of their self-fulfilling expectations that other banks will not lend.

In a modern economy, the prospects of businesses are likely to be interdependent, with each company’s success (and ability to repay) depending on whether other companies obtain financing. Companies commonly use components and services from other businesses and often sell their output to other companies or their employees.

Consider a bank choosing whether to lend to companies or park its capital in treasuries. Suppose that lending to any given company will generate an expected return of 10 per cent if other businesses obtain financing but an expected loss of 5 per cent if they do not. In such circumstances, the economy may get stuck in an inefficient credit freeze in which banks expect other banks to avoid lending and, given these expectations, rationally choose to hoard their capital to avoid the expected loss from lending when other banks do not.

Unfortunately, we cannot count on interest rate cuts and capital infusions into banks to get the economy out of such a credit freeze. Even if banks have ample capital and the yield on treasuries is barely positive, not lending and avoiding the 5 per cent expected loss will remain each bank’s rational choice as long as other banks are not lending.

Is there anything more the government could do? Yes, it can go beyond intervening at the level of the financial sector and intervene in lending to companies. It can take on itself some of the credit risks involved in extending substantial new lending to businesses.

Suppose that the government wishes to get at least $200bn (€140bn, £133bn) of additional lending to companies. Under one possible mechanism, the government would facilitate banks’ putting together a diversified portfolio of newly originated loans by agreeing to bear part of any losses to the portfolio in return for a share of the upside. In the example considered above, to induce banks to put together a portfolio of new loans it would be sufficient for the government to agree to bear any losses to the portfolio of up to 10 per cent of the value of the extended loans.

The share of the upside received by the government could be determined through a competitive process. Banks would submit bids indicating the share they would be willing to offer and the government would accept the highest offers that would collectively produce additional lending of $200bn.

Under an alternative mechanism, the government would place $200bn in a number of funds. Each would be run by a private manager charged with putting together a portfolio of loans and compensated with a share of the profits generated by the fund.

Under each of these mechanisms, the party putting together the portfolio of new loans (the bank or the fund’s manager) would have incentives to lend only to companies with good projects. And the government’s taking upon itself credit risks would directly lead to $200bn flowing to companies.

But the programme’s contribution to producing a credit thaw would go much beyond this direct effect. With many companies expected to receive financing, banks’ willingness to lend their own capital, which they might otherwise elect to hoard, would increase. Furthermore, to the extent that the programme would produce a credit thaw, the programme’s costs to the government would be limited, because the credit risks the government took upon itself would not materialise.

When capital infusions and interest rate cuts do not work, the mechanisms we propose might provide effective tools for unfreezing credit markets.

Lucian Bebchuk and Itay Goldstein, professors at Harvard and Wharton respectively, are co-authors of ‘Self-fulfilling Credit Market Freezes’, just issued by the Harvard Olin Center for Law, Economics and Business.