On December 27, 2005, the difference in yield between the ten-year and the two-year US Treasury Note slipped into negative territory. From that day to this, the spread has averaged -0.2% and has not been greater than 0% since August 11. On July 17, 2006, the yield spread between the 10-year Treasury note and the three-month Treasury bill also turned negative. Since then it has averaged -0.3%. These below-zero spreads are like below-zero temperatures. They make for cold, hard days in the banking world, where most of the profit is still made the old-fashioned way: borrowing money short-term from depositors and lending it long-term to borrowers.
In this month’s commentary, we will have a look at how we got to such an unpleasant pass and whether there is any way out. Our story begins early in 2000 when the air was just starting to leak from the dotcom bubble. The Fed, preoccupied as usual with inflation, missed the initial hissing noise. Even as the NASDAQ index dropped from its peak of 5,048 on March 1, the Fed kept raising the fed funds rate: to 6.00% on March 21 and 6.50% on May 16. The Treasury yield curve—that is, an imaginary line connecting the yields of Treasury securities of increasing maturity—began the year with a nice upward slope and a spread of 1.2% between ten-year and three-month maturities. But faced with the Fed’s inability to hear, or failure to comprehend, the stock market’s message, the curve quickly turned negative.
The Treasury curve was saying, “By keeping short-term interest rates high, the Fed is going to drive the economy into recession. It’s safe to own long-term bonds because soon the Fed will be forced to change course.” And that’s just what happened. An investor who bought the ten-year Treasury note at its then-current yield of 5.50% in December 2000 was making a very smart decision, even though fed funds were at 6.50%. Soon enough, the economy faltered and the Fed began to cut short term interest rates. By the end of 2001, the fed funds rate was down to 1.75%. Our ten-year T-Note investor had a very tidy profit that would only grow larger in the years ahead. The Fed held the funds rate at 1.75% for most of 2002, before easing to 1.25% near year-end. Finally, on June 25, 2003, the Fed dropped the cost of overnight loans between banks to 1.00% where it remained for a year.
Several Fed officials have admitted recently that keeping the fed funds target so low for so long was a mistake, but the Treasury market did not wait to draw its own conclusion, which was that the Fed had lost its collective mind. Treasuries began selling off the day after the last Fed ease. Between June 25 and Labor Day 2003, the ten-year Treasury note yield rose from 3.4% to 4.6%. Because bond yields and prices move in opposite directions, the change represented a 10% decline in value. The Treasury yield curve had now become historically steep. As a natural consequence, the carry trade flourished as never before. Borrowing at 1.00% and investing in instruments without credit risk at 4.00% or even 5.00% was too profitable an opportunity to pass up. Extending maturities to five or even 10 years seemed well worth the risk. It had taken a long time for the fed funds rate to fall, and a quick rise seemed ever more unlikely the longer the Fed stood pat.
While the effect of the super-low fed funds rate on the Treasury market might have been predicted, the impact on housing was something new under the sun. Low short-term interest rates were the spark for a revolution in so-called “sub-prime” home-ownership. I use the word “revolution” very deliberately and in distinction to the simultaneous housing boom. The boom represented the confluence of low long-term, fixed rate mortgage rates and increasingly affluent Baby-Boomers widening the market for upscale housing, whether as primary residences, vacation homes or investment property. The revolution required teaser ARM rates of 1.00% to 2.00% and similar innovations made possible by cheap short-term borrowing costs. According to a report by the Center for Responsible Lending, sub-prime mortgage loan originations grew 20-fold from $35 billion in 1994 to $665 billion in 2005 and 2006. Adjustable-rate mortgages predominate in this market.
Meanwhile, starting in June 2004, the Fed began raising the fed funds rate at “a pace that was likely to be measured.” In fact, it was measured in 17 consecutive 25-basis point increases stretching over two years. The last increase, on June 29, 2006, brought the fed funds rate to its present level of 5.25%, but Treasuries maintained a contrarian stance. The difference between the fed funds target at Labor Day 2003 and today is 4.25%. The difference between the ten-year T-Note yields on those two dates is a mere 0.2%. Virtually all of the increase in interest rates has been at the short end of the curve, progressively diminishing with lengthening maturities. The carry trade is dead, and with it, a significant source of bank profitability. By the third quarter of 2006, net interest margins across the banking industry were at their lowest since 1989, not coincidentally a period when the yield curve was also inverted.
We saw how in 2000, the falling stock market preceded an inverted Treasury yield curve, which in turn preceded, and some would say, predicted, a weakening economy and a period of Fed easing. The prime candidate for derailing the expansion this time has been the housing slowdown. Many economists have predicted that weakness in housing will spill over into the broader economy lowering employment in affiliated industries, curbing inflation fears and ultimately convincing the Fed to reduce short-term interest rates. But the evidence is growing that the real pain will be limited to those sub-prime borrowers, particularly those unable to refinance into conventional fixed-rate mortgages.* In the wider housing market, sales of both new and existing homes have stabilized in the last few months. Housing starts surprisingly increased in November and the Housing Affordability Index has been trending up since August. Construction employment, which had dipped sharply in October and November, leveled off in December. In contrast, the most recent mortgage delinquency statistics show a sharp increase in problem sub-prime loans. More than 12.5% of sub-prime mortgages were 60 or more days delinquent in the third quarter, up from a low of 10.3% in the second quarter of 2005. The delinquency rate for sub-prime adjustable rate mortgages was even worse at 13.2%.
If serious problems in housing do not spread beyond the sub-prime sector, then the economy as a whole won’t slow down. The Fed, whose focus remains steadfastly on the threat of inflation, may not ease at all in 2007, which means the yield curve may stay inverted. The prospect of a static, directionless bond market hovering indefinitely at today’s levels is a considerably less than delightful prospect for banks. In order to maintain spreads, banks have been forced to keep the lid on consumer deposit rates, a strategy not without risk. As The Wall Street Journal recently reported, total bank deposits have dipped to their lowest level since the FDIC was created in 1933. To support loan demand, banks of all sizes have turned to the wholesale money markets, once the preserve of only the largest institutions. Since wholesale money is much more expensive than retail money, this trend has only contributed to the net interest margin squeeze.
No exit to this dilemma is apparent, but if history tells us anything, it’s that markets always change and usually in some unexpected manner. Here’s hoping the change comes soon.
*And, of course, the usual last-in speculators who always end up holding the bag.