Macro Wrap – Correlations Change and Shift, $SPY, $TLT

by Enis July 17, 2013 7:52 am • Commentary

As we await Bernanke’s Congressional testimony today, I wanted to highlight the interesting divergence between bonds and stocks in the past 2 months.

First off, let me start with the caveat that correlations in financial markets are constantly changing.  So there is no rule that dictates that stocks and bonds must move together.  Moreover, day-to-day, or even month-to-month correlations often hide the true nature of assets’ relationships with one another.  What do I mean?

The example of SPY and TLT over the last 10 years is illustrative.  Between July 2003 and July 2007, bonds and stocks were often positively correlated on a one month time frame (bonds moved up when stocks moved up, and bonds moved down when stocks moved down).  Since July 2007, bonds and stocks have mostly been negatively correlated on a one month time frame (bonds move down when stocks move up, and vice versa).  But between July 2003 and July 2007, SPY was up 50% while TLT was flat.  Since then, and even more pronounced in the last 4 years, SPY and TLT are both up substantially.  So despite the short-term positive correlation in the last bull market, bonds and stocks did not move together over time.  Whereas in the current bull market, bonds and stocks have both moved higher over time, even though they have had a negative short-term correlation.

Here is the chart:

Top panel shows performance of SPY (orange line) and TLT (black line); bottom panel shows 20 period correlation of SPY and TLT
Top panel shows performance of SPY (orange line) and TLT (black line); bottom panel shows 20 period correlation of SPY and TLT

In the last 2 months though, we have seen a distinct shift in character in the relationship between TLT and SPY.  First, they have become the most positively correlated that they have been in the past 6 years (shown above in the bottom panel).  However, that short-term positive correlation has not meant that TLT and SPY have moved in the same direction.  In fact, here is the 6 month chart:

6 month daily chart of TLT (black) and SPY (orange), Courtesy of Bloomberg
6 month daily chart of TLT (black) and SPY (orange), Courtesy of Bloomberg

Since the start of May, TLT has been in a steady downtrend of lower lows and lower highs.  SPY peaked shortly after TLT, and both fell together on tapering concerns until the June 24th bottom in stocks.  From there, stocks rebounded in essentially a straight line, while Treasury bonds have stabilized, but have not rallied.

One argument that I’ve heard is that both stocks and bonds are anticipating better economic conditions in the coming year (and the Fed’s potential tapering is one signal of that).  As a result, the long-term movement together between these two assets due to QE effects has ended.  But this quarter’s GDP in the U.S. is on track for a 0-1% growth rate, and the Citigroup Economic surprise index in the U.S. has been negative for much of the last 4 months:

Citigroup Economic Surprise Index in the U.S. in 2013, 0 marked in red, Courtesy of Bloomberg
Citigroup Economic Surprise Index in the U.S. in 2013, 0 marked in red, Courtesy of Bloomberg

So I don’t think the bond market is reacting to stronger growth.  Part of the move lower in bonds is certainly just crowded positioning, as the stampede out of bonds in the past month made evident.  But I also view the bond market’s price action as indicating less monetary stimulus going forward.

In that sense, this week’s reaction of both stocks and bonds to Bernanke’s commentary should be telling.  If stocks hold up while bonds fail to rally, no matter his commentary, it’s more evidence that investors in the short run are not worried about tapering’s impact on stocks (and much more worried about its impact on bonds).  Out of mere curiosity, I wonder in what scenario bonds would rally and stocks would sell off.  I don’t see that happening no matter Bernanke’s commentary.  I haven’t heard anyone else voice that view either.  But correlations change and shift.  Interesting times as investors choose between the 2 major asset classes.