At the start of September, Morgan Stanley's global investment committee released their assessment of possible scenarios for the S&P 500 (NYSEARCA:SPY) along with their projections for future S&P 500 price levels during the remainder of 2014 and 2015. You can download a version of it right here. Having put together a regression model on the historical relationship between earnings and price for the S&P 500, I thought I would also run the Morgan Stanley future estimates through the model and see if we reached the same conclusions with respect to the S&P 500 price level.
But first, here are the various scenarios for the S&P 500 per the Morgan Stanley analysis:
On the left-hand side, the Morgan Stanley projected earnings denoted in light blue, fall somewhat lower versus the analyst consensus. The S&P 500 is projected to fall in the Bear Case by a -22.1% if earnings remain stagnant. Currently the P/E of the S&P 500 is at about 19 - 20. Naturally if earnings held constant, the assumption is that the price would fall on missed expectations.
However there are some good reasons for believing that the S&P 500 will grow along with earnings for the next year and a base case scenario is probably quite likely. In Morgan Stanley's view even a 5% - 6% increase in earnings would still result in nearly no increase over the next year with respect to the S&P 500. This would be a type of "correction" in the market as the P/E would adjust to more modest levels without a decline in price (at least on average for the course of the year).
The speculation that the market will continue to do well, with earnings increasing, plays on several factors within the MS analysis including for the most part CAPEX and Manufacturing.
To quote from the Morgan Stanley analysis:
US industrial production has been resurgent since mid-2009, outpacing GDP by the widest margin in more than 50 years. Many factors have contributed to this renaissance. For instance, relative labor costs have continued to decline, the result of a decade of double-digit wage gains in the emerging markets. The Federal Reserve's aggressive Quantitative Easing suppressed interest rates and put downward pressure on the dollar, helping make exports more competitive. And the US's march toward energy self-sufficiency has improved prospects for reindustrialization and onshoring by lowering costs in energy-intensive manufacturing.
Also included in the article was the following chart for manufacturing showing the growth of that sector:
All good things - and since the writing of this analysis we saw the ISM index rise much higher than anticipated (57.1 to 59.0) on September 2nd. Strong growth in capital spending has occurred with 8.4% increase quarter over quarter. Consumer confidence is up overall which is another important factor contributing to earnings and price increases.
Now let's assume that the bull case occurs and that earnings grow for a total in 2014 of 122.6, 134.9 in 2015, and 141.6 in 2016. Let's see how these numbers, if right, would play out in their contribution to price within the historical context and relationship between earnings and price. A week ago I posted an article comparing the accuracy of utilizing 10 years of earnings versus 1 year of earnings in projecting the future pricing of the S&P 500. In order to do this I put together a regression model which you can download in excel (link is at the bottom of the page) if you would like. In any case I'm going to take the future expected values per this Morgan Stanley analysis and swap out the historical values to see what the price comes to given both the 10 year averaged earnings paradigm and the single year of earnings model.
First - here are some of the numbers being used for the regression line in both models:
Some things to quickly note - the accuracy of the historical relationship between earnings as a predictor or explanation of the current price of the S&P 500 is denoted by R^2 which here shows roughly 76% accuracy for ten years of earnings and 78% for 1 year of earnings. Also the method to find what the price would be in the future is to take the expected earnings as the independent variable and price as the dependent variable. The item denoted as X-Variable 1 is our slope and the Intercept is self-explanatory. So we use the basic equation of a line here [y=mx+b (yes it's only linear - let's not get too crazy!)] in order to find the predicted future value of the S&P 500.
And here is a quick chart of this for just a basic one year of earnings approach utilizing the Morgan Stanley estimates for a bull case:
Our average here is 2569.4 which is only about 30 points higher than the targeted S&P 500 in the Morgan Stanley diagram posted above. So we can be fairly certain that the Morgan Stanley analysis utilizes some level of historical relationship between earnings and price in their model (as we would expect them to) and now the only question is with respect to whether or not the earnings estimate is sufficiently justified (which is the subjective part of the analysis).
However if we really want to give ourselves a hard time, we might consider digging deeper. Most readers here will remember Dr. Shiller the recent Nobel Laureate and the man who in part developed the CAPE Ratio to determine whether or not the market is overpriced overall. At the moment, the CAPE Ratio is over 26 as opposed to the standard P/E of 19 which shows the S&P 500 as overvalued.
Within this paradigm I have created a regression model that uses the previous 10 years of earnings as an explanation for movement in price. We will take the previous years and combine them with the Morgan Stanley estimate running forward. We have earnings denoted in "real" dollars (assuming that the Morgan Stanley analysis was stated in "real" dollars) so that our analysis is stated as a snapshot in time. I do this so as to make life easier as I don't wish to go down the road of calculating present value, etc. ... additionally the linearity assumption of the regression model requires a "steady point in time" approach.
First - here is a chart showing the year ending real earnings for the S&P 500:
And now here are the subsequent average 10 years of earnings for 2014, 2015, 2016 based on the Morgan Stanley estimate:
As we can see, if Shiller's theories with respect to the implication of 10 years of earnings and their impact on prices tend to hold, then we are probably still looking at trouble on the horizon even in a bull case scenario from the Morgan Stanley analysts.
A bull case scenario with respect to 10 years of earnings would appear to leave the S&P 500 at a relatively stagnant level for several years. Whether or not this is an accurate perspective seems to be mostly a matter of personal preference at this point for investors. Hopefully you will download the Morgan Stanley analysis for a better look at all the nice charts and estimates. After reading this analysis, I really tend to believe that we may be looking at another year of positive increases for the S&P 500.
However the Cyclically Adjusted P/E ratio has only been over 26 three other times in history and each time was followed by severe "corrections" in the market. You can read more about this in Shiller's NY Times piece, or also in my article here which looks at worldwide equity markets and their CAPE Ratios. In any event, while looking one year forward might show the S&P 500 as increasing, I tend to think that eventually we will see some sort of correction. And if Shiller's perspective holds true, then even a bull case with respect to earnings such as that outlined by Morgan Stanley, might still result in a poor equity market.
Disclosure: The author has no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.
The author wrote this article themselves, and it expresses their own opinions. The author is not receiving compensation for it (other than from Seeking Alpha). The author has no business relationship with any company whose stock is mentioned in this article.