Ok, full disclosure, we have never created a price target on anything. Not for the S&P 500, nor our favorite commodity, the C$ nor an individual stock. The reason we don’t put targets on things is we know there are simply too many moving parts. That being said, we do have internal detailed discounted cash flow models for companies we own or have researched. And these models do have intrinsic value outputs, but they are just numbers or loose guides at best. We do implement trading strategies around certain share prices, bond yields, currency levels, but we are not hubris enough to peg a target. Take the S&P 500 Index, some days the market will pay 15x for earnings, other days 10x and others 20x. So even if you can get the earnings forecast correct (a tall feat in itself), good luck guessing on the correct multiple the emotional market will pay for those earnings in a year’s time.
Valuations
Now that we have dispelled any desire for a target, let’s see what we can provide because it is more about direction than any specific target. When it comes to investing, just getting the direction right is winning. Here goes.
Absolute Valuation Levels – are slightly elevated
Few would call this market cheap. The top chart is the price-to-earnings ratio for the S&P 500 and TSX Composite over the past decade. During this abbreviated time period 12-14x has been cheap and 16-18x elevated. This is a bit of a short time period, so the 2nd chart is the S&P 500 PE ratio going way back to 1935 (earnings model from my Merrill days, they had great data). Again, mildly elevated but not crazy by any means. Valuations do matter though. The 3rd chart breaks down valuations into ranges with the blue bars as the number of occurrences dating back to 1935 and the orange triangles showing the forward 12-month return for the S&P 500. Not a hard and fast rule, but notice how the triangles tend to be lower in the valuation ranges that have a higher PE. Valuations matter.
So at the aggregate index valuation level, we will say things are mildly elevated which would have folks expecting low single digit returns going forward. However it is not that simple and there are more moving parts.
Dividends + Earnings + Multiple
There are three components to index returns including the dividend, earnings growth and the market multiple (top chart for the past few years). As you can see the multiple change is most volatile followed by earnings growth. Dividends are the most stable, contributing between 1-3% each year. And this variability is mainly driven by changes in the index price level. Most years see dividends rising and while dividends fall in recessions, lower index levels means they still contribute a good deal. So let’s say we have a solid 2% coming from dividends going forward.
Earnings growth can be very volatile, especially during economic slowdowns. In the 2008 recession, quarterly S&P 500 earnings dropped from $24 to -$0.10. However in most environments earnings grow. This had been the norm from 2009 to 2014 when the S&P 500 fell into an earnings recession. Earnings recession is negative earnings growth while the broader economy does not suffer a recession. This stumble in earnings was mainly due to lower capex related to the collapse in resources and the negative impact of the strong US dollar on earnings. Still, it didn’t help the market as negative earnings growth weighed on returns.
Since this earnings recession started in 2014, the S&P 500 has been stuck in a range and unable to move higher (2nd chart). However now we are seeing earnings growth return (3rd chart). This has helped returns in 2016 and appears set to continue to be a positive contribution in 2017. Forecasts are fairly optimistic, ranging from 10-20% earnings growth. If we could just add this to the dividends, 2017 would be a party, unless the earnings multiple spoils the dividend/earnings party.
Earnings Multiple – this one gets tricky
Ok, maybe it isn’t tricky, it will likely decline in 2017. A big component of the multiple is market emotion which tends to be rather volatile. Given how optimistic the market is as we near the exit of 2016, a simple revisionist argument would be it is set to decline. However, let’s throw some math at this as well. The multiple is impacted by bond yields and higher bond yields are generally not good for the multiple (bottom chart previous page). If we continue to see yields rise, this does increase the likelihood of multiple compression in 2017.
Of course this does raise the question why we have not seen the earnings multiple compress over the past few months as the 10-year bond yield has risen from 1.6% to 2.5%. Well there is another factor, credit spreads. And spreads have been compressing, which is viewed as positive for the multiple (top chart). It denotes the market pricing risk at a lower level.
So if the market multiple has been able to weather high bond yields thanks to compressing credit spreads, what happens next year. High yield credit spreads have fallen from over 8% to 3.6%. We have seen lower spreads, but not very often or for very long. Same can be said with investment grade spreads that are down to 69bps. If the credit spreads are running out of steam, another 50bps higher in 10-year yields may land more squarely on compressing the market multiple.
Putting it all together
You guessed it, no price target for the market. But if pressed with the question as to where do we see things going next year – our answer is higher. Earnings growth is a huge positive and that should be enough to keep the S&P 500 continuing its bull market into 2017. However, the likelihood of a sharp multiple contraction (read correction) due to higher bond yields has become an increasingly big risk. And it will likely be swift. One day we will wake up and the market will all of a sudden care about the impact of higher yields. While we don’t know what day that will be, we would view as a buying opportunity given earnings growth, the economy and a host of factors that continue to favour a continuation of this bull.
Article by Craig Basinger, CFA – Richardson GMP
See the full PDF below.
