Economic News & Analysis—June 27, 2012
Is there a bubble in bonds?By James Kochan, Chief Fixed-Income Strategist
For three decades, bond prices have trended higher, and yields in several sectors are currently the lowest on record, causing some observers to suggest that the bond market might be the “asset bubble” of this cycle. That is a serious allegation as the word “bubble” conjures up images of huge potential losses, such as the tech-stock bubble burst in 2000 when losses were as great as 90%, and the real estate bubble in 2008 when home values declined by 50% or more in some areas of the country.
Bubble enthusiasts like to speculate what would happen to bond values if there were a repeat of the prolonged bear market of the 1970s and early 1980s. In that cycle, the yield on the 10-year Treasury note peaked at around 15% in 1981. If that level was realized again in the next five years, the price of the current 10-year note would drop from around 101 to around 55, almost a 50% drop in principal value. That would rank with past bubble disasters.
However, a long list of fundamentals suggests that nothing close to that scenario is likely over the foreseeable future. During the 1970s, the rate of inflation, as measured by the consumer price index (CPI), rose from 3% in 1971 to 13% in 1981. Over that period, the price of crude oil increased from $2 to almost $40 per barrel, unit labor costs were increasing at close to double-digit rates, the money supply was growing at rates as high as 15% per year, and nominal gross domestic product (GDP) growth was at a 20% annual rate in 1980.
Today, oil prices, while high, are not accelerating as they were in the 1970s and unit labor costs are no higher than they were in 2008. The CPI is rising at around a 2% pace, money supply growth is around 4%, and the growth rate of nominal GDP is around 3.5%. And, perhaps most importantly, U.S. companies face international competition from low-cost producers that did not exist in the 1970s, preventing them from raising prices and forcing them to control costs.
Another key difference between now and the 1970s is the nature of this business cycle. This recovery fits the historical pattern of very weak recoveries following a severe financial crisis. It typically takes as long as a decade before economic growth returns to a healthy pace after a financial crisis such as we saw in 2008-2009. The failure of the housing sector to recover is one such factor. The chart below illustrates how atypical the current flat-line pattern of housing starts is versus earlier cycles. Because the housing sector is a big user of borrowed money, this ongoing weakness is a key reason why interest rates have stayed low for such an extended period. Even the most optimistic forecasts call for only a slow housing recovery, and that would keep the demand for mortgage credit relatively soft in the years ahead.
Chart 1: U.S. housing starts (1959-2011)
Finally, the crisis in Europe has probably extended the period of extremely low interest rates in the U.S., contributing to slower global growth, reducing U.S. exports, and depressing global commodity prices. This further delays any cyclical increase in inflation and the attendant growth in private sector borrowing and money supply growth. It is probably not unreasonable to suggest, as the Fed has done, that interest rates might stay near current levels for another two years. Moreover, these atypical cyclical developments would suggest that even when rates start to rise, the increases will be moderate. There is, in our view, virtually nothing on the economic horizon that would suggest a return to a 1970s type interest rate environment.
Eventually, however, interest rates and bond yields will start to rise, which would cause bond prices to decline and, at first glance, appear to qualify as a bursting of a bubble. But if the rise in yields is relatively slow and moderate, that process need not produce negative returns. For example, consider a 10-year corporate bond with a coupon of around 6% that is currently priced at a premium to yield 4%. If over the next five years, market yields were to double to around 8%, the total return on such a bond would be 2.25% per year, or 12% for the entire five years. While not a generous return, it is far better than the huge losses usually associated with bubbles bursting. The interest income from bonds cushions the negative effects of falling principal values. Even in periods of rising interest rates, the value of a bond portfolio typically does not plummet as is inferred by the use of the term “bubble”.
ConclusionIn the world of finance, the term bubble serves as a warning that it is time to exit an overvalued asset class and could pose a risk of catastrophic losses. While it could be argued that some sectors of the bond market are overvalued (such as Treasuries and treasury inflation protected securities (TIPS), it does not follow that bond portfolios are likely to suffer huge losses. There are virtually no similarities between the current economic environment and that of the 1970s, when bonds suffered a prolonged bear market. The most likely scenario is another year or more of stable short-term interest rates, followed by a gradual and limited uptrend. Even in that type of rising rate scenario, bond portfolios would probably produce positive returns. We believe the hyperbole associated with the term “bubble” should not frighten fixed-income investors into overly defensive, low return, portfolio strategies.