Because AGG is such a good proxy for bonds, it also provides us with an opportunity to investigate how movements in the fixed income arena are associated with subsequent movements in the equity markets. I went back to 2004 (N = 154 trading weeks) and took a look at three-week changes in AGG and what happens over the subsequent three weeks in the S&P 500 Index (SPY); the NASDAQ 100 Index (QQQQ); and the Russell 2000 Index (IWM).
When AGG has been up over a three-week time frame--meaning that interest rates are falling over that period (N = 78), the next three weeks in SPY have averaged a very healthy gain of .80% (53 up, 25 down). When AGG has been down over three weeks (N = 76), the next three weeks in SPY have averaged a gain of only .08% (38 up, 38 down). Clearly, falling interest rates (rising bond prices) have been associated with superior large cap returns since 2004 over a several week period; rising rates (falling bond prices) have yielded subnormal returns going forward.
We see identical patterns in QQQQ and IWM. When AGG has been up over a three-week time frame, the next three weeks in QQQQ (IWM) have averaged a gain of .83% (1.32%) with 41 (53) occasions up and 37 (25) down. When AGG has been down over the past three weeks, the next three weeks in QQQQ (IWM) have averaged a loss of -.22% (-.12%) with 38 (32) occasions up and 38 (44) down. Stock performance in those averages has been dramatically better during periods of rising bond prices (lower interest rates) than during periods of falling prices (rising rates).
I'm especially impressed by the magnitude of the differences in IWM performance under different bond conditions. It may well be that rising rates dampen speculative sentiment for equities, particularly affecting small caps. Conversely, an environment of falling rates might be deemed positive for growth, encouraging speculation in the small caps. With the availability of data on a variety of ETFs across asset classes, we are more able than ever to tease apart such intermarket relationships.