Visiting Scholar John H. Makin

The policy alternatives in the post-housing-bubble world are painfully unpleasant. In my view, the least bad option is for the Federal Reserve to print money to help stabilize housing prices and financial markets. Yes, use reflation to soften the pain for Main Street and Wall Street. If instead we let housing prices fall another 25%-30%--as predicted by the Case-Shiller Home Price Index--it's almost certain that Washington will end up nationalizing the mortgage business.

So far, the Fed's lending programs have not provided adequate liquidity to financial markets: Reserves supplied to the banking system have grown at a tiny 0.6% annual rate since December. That's because the reserves the Fed is injecting by lending are effectively pulled out or "sterilized" by its sales of Treasury securities. The Fed has been selling these securities to keep the fed funds rate at the level targeted by its Federal Open Market Committee directives.

Congress and the Treasury have proposed voluntary measures to help mortgage borrowers, but the impact on mortgage availability has been nil. As average house prices plummet--declining at a 23% annual rate over the three months ending in January--lenders are sharply curtailing access to mortgage-based, home-equity loans. The 15% of U.S. mortgage holders with negative equity in their homes have no access to credit, and 20% with marginal equity have limited access at best. Overall access to credit is contracting: Ask Americans trying to utilize home-equity lines or arrange student loans.

Meanwhile, the collapse of house prices and the attendant damage to credit markets have become so severe that the Fed has been forced to create new policy measures at a fast clip, including the radical decision to take $30 billion worth of Bear Stearns' risky mortgages onto its own balance sheet, and to open the discount window to investment banks.

The bottom line is this: The Fed could have watched a run on investment banks quickly turn into a run on commercial banks, or protected the creditors of investment banks (like the depositors of commercial banks) at the expense of Bear Stearns' shareholders. The Fed wisely chose the second alternative.

Still, the Fed's intervention has done no more than buy a respite from the crisis in the financial markets. The monetary easing I'm recommending can occur by having the Fed print money to purchase mortgages directly, or purchase Treasury securities directly. The latter is probably more desirable because it adds higher-quality assets to the Fed's balance sheet. The Bank of Japan was also forced to reflate by printing money in 2001, after two years of a zero interest-rate policy failed to lift the economy out of a prolonged recession that had moved Japan to the brink of a deflationary crisis.

Fed reflation--to slow the fall in home prices and alleviate the distress for households and lenders--carries many risks. But the alternative is to struggle with a patchwork of inadequate efforts to shore up mortgage markets, while the Fed sticks to its current tactic of pegging the fed funds rate without increasing the money supply. This, I would submit, is even more risky. It risks a severe recession that will only intensify the drive for reregulation of financial and mortgage markets after the election.

Printing money is a radical step that enables the Fed to stop pegging the federal-funds rate and start increasing market liquidity directly. In any event, there is substantial evidence that the fed funds rate has been well above the equilibrium level. One piece of evidence is the accelerating deterioration in credit markets and the real economy that ensued even while the Fed cut the rate. Even more compelling, consider the sharp widening of the gap between the fed funds rate and the yield on three-month Treasury bills.

That gap, usually close to zero, measures the intensity of demand for riskless assets relative to the Fed's target rate in the interbank market. At the time of the Bear Stearns crisis on March 16, the fed funds rate was an extraordinary 250 basis points above yields on three-month Treasurys. This corresponded to a "10 sigma," or ten-times-the-typical deviation from the mean event. Statistically, 2 or 3 sigma is a very unusual event suggesting, in this case, an unusually strong preference for riskless T-bills. Four or 5 sigma represents a serious risky event, and 10 sigma is an outright panic. Based on this gap criterion, the August 2007 crisis onset was a 5-sigma event, while the October 1998 LTCM crisis and the 1987 stock market crash were each 4-sigma events. This suggests that even at those earlier times of crisis there was less fear as expressed by a run into riskless Treasurys. Ominously, after dipping close to 5 sigma after the Bear Stearns crisis, the gap has crept back above 6 sigma.

The Fed should announce its intention to add to its holding of Treasury securities in order to provide additional liquidity. It should cease pegging the fed funds rate while this policy is in effect. While there is no guarantee, direct injection of money holds some promise of alleviating the worst of the credit crisis. This means that, after the election, Congress will not feel justified in nationalizing mortgage markets.

While there is a substantial risk that inflation may rise for a time--this would be the policy goal--monetization is more easily reversible than nationalization of the mortgage market. Meanwhile, Fed officials concerned about inflation should rethink their view that it is impossible to identify an asset bubble before it bursts.

The postbubble period has yielded some very unattractive policy alternatives. They clearly underscore the rationale for having the Fed target asset prices--in a world where asset markets affect the real economy more than the real economy affects asset markets.

John H. Makin is a visiting scholar at AEI.