Saturday, April 12, 2008

How Markets and Intermarket Relationships Have Changed Since Mid-2007

One of the great dividing points between experienced, successful market participants and amateurs is that the former realize that markets are continuously changing. They make active attempts to identify and adapt to these changes. Amateurs look for fixed patterns across all markets and all periods of time. For them, financial markets are relatively static entities.

This post illustrates how markets do, indeed, change over time. The chart above tracks volatility--the median price movement--for equities (S&P 500 Index, SPY) and interest rates (10-year Treasury rates, $TNX) both during the overnight period (close to open) and during the day period (open to close). Note that we're looking at the absolute size of movements--pure volatility--not the directionality of those moves.

What you can see is that the volatility of those movements has expanded significantly since mid-2007. In each case, the red bar is quite a bit higher than the blue bar, meaning that volatility since July, 2007 has increased for both stock price movements and the movement of interest rates. Indeed, for the most part, the movements lately have been twice as large since July, 2007 as they were from 2006 through June, 2007.

But it's not only the *size* of these movements that has changed dramatically. The relationship between stock price movements and the movements of interest rates has also shifted meaningfully. Below, here are the correlations between equity and rate movements:

Overnight, 2006 - Mid 2007: .16
Overnight, Mid 2007 - Present: .45

Day Session, 2006 - Mid 2007: .09
Day Session, Mid 2007 - Present: .40

In other words, the size of movements in equities are rates were relatively uncorrelated from 2006 through mid-2007. Since mid-2007, however, the size of those movements has been significantly more correlated: when we see big moves in one, we tend to see large moves in the other.

And, of course, when we look at the directionality of those movements, we see greater correlation during the recent period as well. In past periods, falling interest rates (rising Treasury prices) have been associated with bullish movements in stocks, as reduced rates have spurred lending and economic activity. Since mid-2007, however, Treasuries have acted as a safe haven when there has been uncertainty about stocks and the economy. As a result, we've seen a tight correlation between falling yields and falling stock prices.

Specifically, the correlation between end-of-day changes in SPY and $TNX was -.03 from 2006 through mid-2007. Since mid-2007, that correlation has soared to .56.

Volatility has changed, intermarket relationships have changed. We've also seen dramatic changes in trending over these time periods, across bonds, commodities, currencies, and stocks. Identifying and understanding these shifts is essential to successful investment and portfolio management. I would also argue that, on a day-to-day basis, it's also invaluable for intraday and swing traders.


Intermarket Relations


Ziad said...

Hi Dr. Brett,

From your experience, have things also changed in terms of the relationships between sectors since 2006? i.e. Back then were you also tracking the strength of the various sectors in the same way you do now, with XLY, XLF, and XLK being risk-seeking sectors while XLP and XLV being risk-averse sectors?

I would assume that some basic relationships like this that reflect the business cycle don't really change much, but they may vary in the degree of their importance as time goes by depending on what is taking center stage in the economy. For instance XLF is of paramount importance now, and oil taking a greater role than it may have before, but technology was the place to watch for investor sentiment in the late 90's. So while general principles may remain the same for certain things, precedence and importance varies. Does that make sense?

Brett Steenbarger, Ph.D. said...

Hi Ziad,

Very good point: the themes--and hence the relative importance of individual sectors--change from one market cycle to the next. Tech was *the* theme in the late 90s and then during the subsequent decline; now we're seeing financials as a key theme. These themes seem to be drivers of speculative sentiment, bullish *and* bearish--