The stock market, which moved mainly sideways from early May to mid-June, has sold off sharply since Ben Bernanke's now infamous press conference on 6/19/13. The Fed Chairman's remarks, which went well beyond the FOMC minutes, stressed that changes in QE policy and timing would be dependent on economic activity. But for the first time, the Federal Reserve suggested timing for the QE wind-down.
While stocks have been hard hit, the bond market has been slammed. The Fed is by far the largest source of demand for Treasury notes. Rates are likely to accelerate higher as the Fed slackens and then ends its purchases of U.S. government debt. When rates rise, how do stocks fare?
They fare reasonably well in the short term, according to our analysis of historical performance pattern dating back to the 1950s. Specifically, when the six-month rise in bond yields tops 20%, stocks have tended to move higher, perhaps as a result of rotation out of bonds. On a longer term basis, for instance when we look at the 12-month change in bond yields, stock performance is more muted. And, particularly when rising rates are associated with inflation, stocks are ultimately hurt.
Daily Price Swings Signal Bull-Bear Tussle
Before looking at the historical record, let's examine the carnage so far in the asset markets. Stocks have given back in June, but they had plenty to give. From the 5/21/13 intraday high of 1,697 (closing high was 1,669), the S&P 500 had retreated just under 7% as of midday on 6/24/13. The S&P 500 advanced 25% from the November 2012 low of 1,353 to the 5/21/13 high. Argus Chief Investment Strategist Peter Canelo believes stocks could give back up to half the rally since November, and is anticipating a decline of 7%-10%; that would put the S&P 500 somewhere between 1,575 (approximately the 6/24/13 level) and 1,535.
Bonds of course are faring much worse. Bond yields are rising even though we are just past phase one (Pundit Taper talk) of a three-stage process that now includes phase two (Fed Taper planning) and phase three (Fed Taper enactment). The 10-year yield, in moving from 1.62% on 5/2/13 to 2.58% as of 6/24/13, has already experienced a trough-to-peak move of more than 58%. The "forceful break" of the October 2011 and May 2012 highs in the 10- and 30-year maturities robs bond prices of support. We now look for the 10-year to trade in the 2.25%-2.75% range in coming months and potentially reach 3.0% by year-end 2013.
While we expect both stocks and bonds to suffer in the early days of QE wind-down, stock bears are predicting cataclysmic implications for equities. You can tell that the conflict between bulls and bears has moved from the usual low-level skirmishing to pitched battle by the ferocity of daily price swings and the rising number of days in which index prices change by 1%-plus or minus.
In the first five months of the year, or roughly the first 100 trading days, daily rate of change on the S&P 500 averaged 0.55%. Since May 21, however the daily rate of change has escalated to 0.73%. And in the first 100 trading days of the year, the S&P 500 recorded 15 days in which prices changed by 1%-plus for a 15% rate of incidence. From 5/21/13 through 6/24/13, the S&P 500 recorded eight 1%-plus change days in 24 sessions, for a 33% rate of incidence. And just since month-end of May, the eight 1%-plus change days in 17 sessions makes for a 47% rate of incidence.
Bond Yields and Stock Index Prices: the Historical Record
In this stand-off, bulls argue that the rising economy justifies further stock gains. Bears posit that the demise of QE will send rate spiraling higher, thus causing the end of the bull market. Each side is fiercely sticking to their hunch about what will happen.
We too have hunches, but whenever possible we ignore them. Instead we go to the numbers. Given the high likelihood of higher yields, we sought to quantify the impact on stocks of rising rates. Beginning back in the early 1950s, we examined every period in which the 10-year Treasury yield rose more than 20% over a six-month span. What we discovered was that stocks, also measured on a six-month rate of change basis, tended to rise - often sharply - in these periods.
We also looked at a 12-month rate of change in bond yields and stocks. When bond yields rise more than 20% over 12 months, stocks perform less well. In periods in which rising yields coincide with rising inflation, such as in 1981, the 12-month rate of change in stocks quickly turns negative.
But we also learned that in most periods, the decline in stocks occurred not amid rising yields. Instead, stock declines were more common when interest rates were falling. This occurred most often in periods of declining economic activity.
For our methodology, we began with month-end pricing on the S&P 500 dating back to 1953. We also used month-end values from the CBOE Interest Rate 10-year T-Note (ticker ^TNX on Yahoo Finance).
Within our survey period, the first time long-bond yields rose more than 20% in six months was in 1958. Yields rose from 2.88% in April 1958 to 4.02% in January 1959 before moderating slightly. Across that period, the six month rate of change in stocks was consistently upward in the mid-teens.
Ten-year yields stayed in that mid-3%-mid-4% range right up to the mid-1960s. Their subsequent rise into the 5% and higher range was so measured that it did not trigger a 20% six-month rate of change in bond yields any time in that span. Amid the Nixon turmoil, the "Great Society" years of Lyndon Johnson, and the single term of Jimmy Carter, 10-year rates climbed patiently and inexorably through the upper single digits.
By early 1980, our expanded post-Vietnam global presence and the cost of "Great Society" social programs were causing a surge in inflation. The 10-year yield rose from 9.0% in August 1979 to 12.8% by March 1980, or by 380 bps in seven months. Yet stocks held firm in this turbulent period, recording mid-single-digit gains on a six-month rolling basis across most of this period. Yields briefly backed down to high-single-digits in mid-1980, only to rise again. Even though rolling six-month appreciation in bond yields again topped 20% during the fall 1980 election months, the six-month change in stocks was 23% in November 1980 as "morning in America" and the Reagan presidency approached.
Most of the Reagan presidency was characterized by 10-year yields in upper-single digits, even though this period was marked by deficit spending and vast expansion in the size of the Treasury market. By May 1987, the 10-year yield had jumped to 8.6%, up 15 bps or 21% from 7.1% in December 1986. Yields continued rising, peaking at 9.5% immediately before the "Black Monday" market sell-off on 10/19/87. All Across this period, stocks were rising in mid-teens on a six-month rate of change basis.
The October 1987 crash has been variously blamed on program trading, currency manipulation and a stare-down between the Federal Reserve and Germany's Bundesbank. Whatever the cause, as stocks were crashing so too were bond yields. The 10-year yield, which flirted with double-digits immediately before Black Monday, downshifted into the 8% range amid the 20%-plus tumble in stocks. Despite a few feints, 10-year yields have not been above 9.3% at any time since the 1987 market crash.
As the 1990s approached, the 10-year yield had again stabilized in the low 7% to high 8% range. Following the early 1990s recession, which ushered out the first Bush presidency and ushered in Bill Clinton, yields drifted down below 6% amid slack economic activity. But a new phenomenon - the "internet economy" - coincided with and perhaps contributed to rising economic and stock-market activity.
From 5.3% in October 1993, the 10-year yield reached 6.5% by March 1994. Yields continued to rise, causing the six-month rate of change to remain above 20% for most of the first half of 1994. In this period, stocks did not perform well; but they did not sell off sharply either. Between the low of 5.3% in the 10-year yield in October 1993 and the high of 8.0% reached in November 1994, the S&P 500 declined by 14.14 points. That is a decline less than 3% over a 13-month span in which long yields increased by 263 basis points, or by more than 49%.
The 10-year yield has never again been a high as it was in November 1994. In June 1996, when yields were in the high 6% range, the six-month rate of change in bond yields briefly topped 20%. Stocks were beginning their massive late-1990s rally. In an unusual set of circumstances, even as the S&P 500 was averaging 26% annual appreciation across 1995 to 1999, the 10-year yield was working lower: from 7.8% in January 1995 to 5.0% in February 1999. At the market top in March 2000, the 10-year yield was 6.3%.
As the market fought back from 9/11 and the internet implosion in 2003, bond yields recorded another brief period of 20%-plus appreciation on a six-month rolling basis. This was occurring, of course, at structurally lower rates, reflecting the globalization of money and interest rates. Ten-year yields moved from 3.3% in June 2003 to 4.3% by October. Of course this was in the early phase of the mid-2000s bull market, and stocks on a rolling six month basis were up in mid-teens across this period.
The recession brought the 10-year yield to what (at the time) seemed an impossibly low 2.4% in December 2008. By June 2009, the yield was up by more than 130 basis points, or by 47%, to 3.7%. But, as in early 2003, stocks in 2009 were in the first year of a major bull market. Rather than fold amid rising yields, stocks across 2009 rose in the 25%-30% range on a six-month rolling basis.
We have demonstrated that stocks tend to do well (and sometimes very well) in periods in which rates rise across a half-year period. How do stocks fare within longer-term trends of rising rates, such as across one year? Around the time of the transition to the Reagan presidency, stocks maintained their strong performance. But as yields (on a 12-month rolling basis) shot above 15% in summer and fall 1981, the one-year trend in stocks turned negative - and stayed that way until autumn 1982. The 12-month trend in stocks was weaker than the six-month trend in stocks during other rising-rate periods, including 1993-94, 2003 and 2007.
Most of the major stock-market crashes coincide not with rising rates but with falling rates. That was not true in 1981, a year in which the S&P 500 declined 10% while the 10-year yield rose 120 basis points. But in the significant stock-market decline years of 1987, 2000 and 2001 and 2008, bond yields ended the year lower than where they began the year.
History should be taken with a grain of salt and like salt on the table, history is best used sparingly in overall market analysis. Every situation is different. That said, over the past 60-plus years, stocks have done well on an interim basis in periods of rising Treasury yields.
(Jim Kelleher, CFA, Director of Research)