Two Crucial Tweaks to the Treasury Plan

Two tweaks could make a big difference in the prospects for success of Treasury Secretary Timothy Geithner's plan to revive U.S. credit markets.

They are critical, but even so, in the bigger scale of things, they're tweaks — the underlying plan makes a lot more sense than most commentators admit.

It's not fair to say Geithner is throwing the kitchen sink at the U.S. financial crisis because that makes the new bailout sound like a hodgepodge of measures pulled together in desperation. In fact, the new measures are a rational plan to fix a set of Bush Administration initiatives that suffered from a lot of uncertainty about everything — except for the certainty that the principal architects were about to leave office.

For example, Geithner's plan — announced February 10, 2009 — reflects three key lessons we have learned as the financial crisis has unfolded. First, it reflects the fact that banks need both capital and other funding to make new loans — capital to make up for losses on bad investments and funding from investors and wholesale depositors that won't materialize until they're absolutely sure they know the extent of the banks' losses.

Former Secretary Henry Paulson moved forward with new bank capital because it was easy and popular with senior bank managers. But it was a mistake on the part of most analysts not to insist on putting a floor under banks' losses at the same time so banks could attract funding on top of their new capital for new loans.

Second, it reflects the fact that we should use all three major sources of finance in the U.S. credit market at the same time — money appropriated by the U.S. Congress such as the Targeted Asset Relief Program, money created by the Federal Reserve and money we can mobilize from the private sector once we've committed to both of the other two. This is why Geithner wants to put the Federal Reserve to work buying troubled assets using TARP funds as credit support.

And third, we've learned to treat assets that banks trade differently from assets they hold until maturity. The reason is that regardless of how much the value of the latter dips during a crisis, those market values will eventually return to their original value so long as the obligations they represent are repaid. Trading assets are different because you're never sure when, and at what price, a bank will dispose of them.

So it makes sense, as Geithner proposes, to give banks a guarantee for the stuff they intend to hold until maturity, while buying up troubled trading assets outright. Overall, the basic structure of the new plan reflects these three lessons.

Now, on to what needs improvement. The first tweak would help answer the question investors keep asking about how to value troubled assets of both kinds. The trick is to get banks to disclose at least as much as they know about the values of those assets — and the best way to do that is through auctions.

Treasury could assign assets in each major class of mortgages, mortgage-backed securities, security-backed securities, card debt and commercial and industrial loans to pre-defined credit categories — and ask banks to bid on them. The bids would propose prices for each individual asset that Treasury could compare with bids on other assets in the same credit category.

While banks would normally have an incentive to overbid — and ask for too much money — the auction would restrain them. After all, the government would not be guaranteeing that it would purchase every asset. In fact, it would purchase only those with the most favorable bids in each asset class. Auctions give people a reason to show what they're thinking about value.

The second crucial tweak has to do with how monetary policy encourages spending. The Geithner plan is open about the maturity of the assets the Federal Reserve would buy. It might start buying 30-year mortgages, for example.

This is a huge departure from traditional practice, where the Fed has focused mostly on Treasury bills due in 90 days. Most economists assume that is because the Fed prefers the safety of debt that is due soon. But at a time like this, the Fed needs to worry a lot more about the safety of the credit system than the safety of its own balance sheet.

In reality, the Fed's traditional focus on short-term securities is just as important today as it has ever been. By buying short-term securities, the Fed lowers short-term interest rates compared to long-term interest rates.

Why not lower both? Because the real solution to the credit collapse lies in giving banks an incentive to provide the kind of credit that borrowers want — long-term credit, whether you're a business or a house-buyer.

The trouble is that banks have to fund those longer-term loans with short-term deposits, which exposes them to the risk of losses if short-term rates rise compared to long-term rates. If you want banks to take that kind of risk — a risk that's critical for re-starting the credit markets — you have to give them a reward for doing so.

And that is why it is such a bad idea for the Fed to buy long-term securities. Doing so will depress long-term interest rates compared to short-term interest rates — cutting into the premium banks need to earn to start solving borrowers’ problems themselves.

In short, when it comes to troubled assets, rely on auctions — ask the banks to bid on what they want to pay and buy from the lowest bidder. And when it comes to asset purchases by the Federal Reserve, focus on short-term securities so the banks can make a profit from new lending.

David Apgar is the author of “Risk Intelligence: Learning to Manage What We Don’t Know,” Harvard School Business Press 2006.