SPY: What's Driving The S&P 500?

| About: SPDR S&P (SPY)

Summary

We introduce a model of the S&P 500 to gauge key drivers and market sentiment.

Over the last few months, economic activity and surprises have replaced earnings and oil as the key drivers.

Our model suggests the S&P 500 remains rich to fair value.

The S&P 500 Index and ETF (NYSEARCA:SPY) has traded in a relatively narrow range over the last few months as the market has been digesting contradictory information, preparing for a number of upcoming events and analyzing the running commentary from the Fed. As the market has taken a pause, it has allowed investors to take a breath and re-evaluate fundamentals, analyze the key market drivers of the last year and gauge the market's potential response to various scenarios that may play out into year-end and beyond.

In this article we introduce a framework for thinking about these issues. Our conceptual starting point is the assumption that the S&P 500 Index returns are driven by changes in macro-economic and market "drivers". We regress the Index returns on a number of these drivers to get estimates of their market beta to the Index. We then calculate an out-of-sample expected market performance as the product of the changes in these drivers and their betas.

How can this approach be useful? Firstly, it can disaggregate the performance of the Index into more familiar factors to give us a more intuitive sense of the market e.g. "the market rallied because of a jump in the oil price and a drop in yields". Secondly, it can give us a sense of the shifting nature of market drivers e.g. "oil was driving the market over the first half of the year but has since taken a back seat to China and yields". Thirdly, it can allow us to run scenarios of market performance based on our view of the potential future path of market drivers e.g. "assuming a view of $40 Oil by the end-of-the-year, we would expect the market to fall X points, all else equal". Finally, the unexplained portion of the Index performance can be viewed as market risk sentiment e.g. "the market may be getting ahead of itself if the actual index is trading much higher than the model".

Before going further into the details, here is a chart to start us off:

The chart above shows us three things:

  1. Blue Line is the familiar actual S&P 500 Index from July-2015 to Today
  2. Gray Line is the model of the S&P 500 based on a 2-year regression from July-2014 to July-2016
  3. Orange Line is the out-of-sample projected performance of the S&P 500 since July-2016 based on the regression model and contemporaneous changes in market drivers

The market drivers we use in the model are the following:

  • USD Index
  • Chinese Economic Activity
  • Earnings Revisions
  • Actual Earnings
  • Yields [10-year Treasuries]
  • Oil [WTI]
  • US Economic Activity
  • US Economic Data Surprises
  • Financial Conditions [ex-Equities]

The R2 of the 2-year regression is fairly robust at 68% - this can also be seen by the good relationship between the Actual [Blue] and Model [Gray] S&P 500.

This is what the out-of-sample results [Orange] are telling us:

  • In July, the model predicted nearly a 100-point sell-off in the S&P 500 due to the 8-point drop in Oil. This was offset somewhat by economic activity and earnings however the market mostly chose to ignore the move in Oil.
  • In August, the model expected a strong rebound also largely due to the recovery in Oil which traded back up to the low 50s.
  • Over the last few weeks the model predicted a sell-off based on worse-than-expected Chinese and US data.
  • Overall, the model level of the S&P 500 is below the actual level, suggesting that the S&P 500 may be trading somewhat rich.

Another way to visualize the performance of the S&P 500 since July is to disaggregate its actual performance by Driver with the remaining unexplained portion being represented as a "Residual". We do this in the chart below where we can see the following:

  • Positive impulse to the S&P 500 has come from 10y Treasury Yields (20bp increase with a positive beta contributing +22 points to the Index), stronger than expected performance in China since July (contributing 34 points) and a stronger USD (having a positive beta over the regression period).
  • Negative impulse has come from negative economic surprises and earnings.
  • The unexplained portion of the performance is 25 points (actual S&P 500 has outperformed the model by 25 points) which is shown below under a "Residual".

It is important to keep in mind that the explanatory power of the factors tends to shift over time. We already mentioned above how the market has mostly ignored the volatility in Oil over the last few months suggesting that the beta to Oil has gone down. In the chart below we show the R2 of each factor with 6-month running window.

In this chart we can see that the explanatory power of Oil, Earnings and Financial Conditions has decreased substantially while that of Economic Activity and Data Surprises has gone up.

So, all in all, what can we do with this framework and results for those investors following Equities and in particular?

First, it can focus our attention on "what matters" right now in the market. For instance, we can see that the market is discounting volatility in the Oil price so, all else equal, we shouldn't expect a big move if we see the kind of Oil price volatility (+/- 10-15 points) that we have seen over the last few months. One can argue that the fundamental argument for this is that the industry has had enough time to readjust to the world of lower Oil prices and the worst case scenarios of defaults have not been realized. Therefore, the market may not be as sensitive to Oil as it had been on the first leg lower in prices.

Secondly, it can give us a sense of whether the market is trading rich or cheap or, in other words, whether the risk sentiment is bullish or bearish. The cumulative Residual or unexplained gap between the Actual and Model S&P 500 since July is 25 points suggesting the market has "outperformed" its drivers. Of course, it is possible that we have missed a driver that would perfectly explain this gap so we must take this analysis with some grain of salt, however the combination of the mean-reverting nature of the Residual as well as the narrative of the market slightly getting ahead of ourselves which rings true to us does give us some comfort.

We should keep in mind however that all of the analysis above does not tell us whether the market is going to rally or sell-off. Hopefully, however, it can provide a framework around which investors can hang their own discretionary views, scenarios and analyses. Our own view is that the market does feel a touch rich and oddly impervious to the continuing weakness in the Oil & Gas industries. Adding that to a basket of Risk events over the coming months: US election in November, potential ECB Tapering over the coming months and Fed action in December suggests to us a cautious stance into year-end.

If you would like to be notified of regular updates of this and other analyses please Follow us.

Thanks and Good Luck!

Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.

I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.