One of the easiest things to lose track of during periods of higher than average market valuations is the simplicity of stock returns. It just forms a part of human psyche that we are programmed to spend too much of our time moving from eternal optimists to eternal pessimists; a point which is demonstrated repeatedly during every market cycle. The deceptive point is that each cycle has a thin veneer that tricks us into thinking this time it’s different. In the early 1970s it would have been blue chip growth stocks. In the late 1990s it was the technology sector. In the run up to 2007 it could well have been real estate. And so it goes on.
What I’ve found to be a good way of maintaining focus is to regard stock investing as the accumulation of highly productive assets, from which there are three possible ways that you can see wealth creation. The first is the growth in earnings of your particular slice of the asset’s overall earnings over a given time frame. The second are all dividends paid out to you over that time frame. The third is the change in valuation applied to the asset between the start and end date. That’s pretty much it. If you’re buying a stock or a market tracker fund for long-term wealth creation and you’re not thinking about those three things then something is going wrong.
Example 1: Cisco Stock Returns Since The Dot-Com Bubble
The hard part is that the simplicity of it goes missing when we collectively get swept up in whatever the narrative is of the moment. Take Cisco stock in early 2000 for example. At the peak of the dot-com bubble investors had driven the share price up to the point where folks were paying around 175x the company’s 1999 earnings to own the stock.
Now on its own that doesn’t necessarily tell you much. It could, in theory, be that Cisco’s growth rate is such that a price-to-earnings ratio of 175 is actually cheap. So what do you do as someone who is maybe looking to purchase Cisco stock in early 2000? You ignore the hype and come back to the three points above.
Let’s say at some point in the future Cisco stock will trade at a P/E ratio of 15. This is a very reasonable assumption as there will be a point when the firm is no longer experiencing tremendous earnings growth and so is being priced at around the same level as the historical market average. Let’s say as an investor in 2000 you see that point being in fifteen years time.
You would also like to see at least 7% a year in capital appreciation for the risk you are taking. In addition you figure that you can’t really predict when management will pay a dividend as it depends on the returns they can get elsewhere, so any dividends paid out between now and then will be a kind of “bonus” on the 7% a year you are expecting.
Doing a few calculations with those numbers would tell you that Cisco would needed to have grown its earnings-per-share at a rate 26% per-year over the fifteen year period to deliver 7% annual returns on your investment ignoring any dividends paid out. As it turns out the company “only” managed to grow their per-share profits at around 12% per-year. Not only that but the P/E ratio had compressed to 11-12x earnings by the end of the period because the growth outlook was very poor.
Now ordinarily average earnings-per-share growth of 12% per-year over fifteen years would be great, but because some investors had ignored point (3) above it has actually resulted in huge capital destruction for anyone who invested at the peak in March, 2000 (assuming they didn’t average down).
Example 2: McDonald’s Stock Returns Between 2003 and 2009
Let’s take a look at an example at the opposite end of the returns spectrum. Back in 2003 you happen to notice that McDonald’s stock has fallen heavily from its dot-com era valuation of 30x earnings. The average P/E ratio for the year is now just 14x annual profits.
Let’s say you figure that the company generally earns very high quality income but its growth outlook is modest. You therefore pencil in that the average P/E ratio of 14 is quite likely to stay constant in the future. You also would like to see the same 7% per-year average level of returns that you were looking for with Cisco stock.
You also know that McDonald’s pays out a steady dividend because it is a more mature corporation. For 2003 it will be in the region of 2% on your invested capital. That means you really only need to see 5% earnings-per-share growth to hit your target returns of 7% a year. Not only that but the firm can also use its retained earnings to buy back its own shares because there is little need to reinvest them elsewhere in order to sustain and grow profits. In reality that means the actual organic component of growth is even less than 5% per year.
Looking at all those numbers you conclude that an investment in McDonald’s stock is actually very attractive. Sure there is always the possibility that bad the business could end up badly underperforming but the key point is that the assumptions you have made look not just realistic, but actually quite conservative.
Let’s say you check back just six years later in 2009. The US economy is in a very scary place. Folks are losing their jobs, houses are being repossessed and the markets are in turmoil. Yet when you open up the share price chart of McDonald’s you find that there has barely been a dent. In fact it is actually up by a factor of nearly three-fold since your purchase in 2003. In addition you racked up growing dividend distributions equivalent to a third of your initial stake. In total your annual returns at the absolute nadir of the 2009 bear market would have been equivalent to an average of around 20% compounded.
So how did McDonald’s stock end up doing so well? Well as it turns out the 5% a year earnings growth you had pencilled in was very conservative. The company actually managed to deliver earnings growth of 18.5% per year over the six year period between 2003 – 2009. That made paying 14x annual profits at the beginning of the period an absolute steal. The interesting part is that no component of those returns were dependent on an expanding value multiple: the 14x earnings you were paying in 2003 was roughly the average 2009 valuation of McDonald’s stock.
Okay but isn’t this playing Monday morning quarter back? I mean who can predict that McDonald’s would grow earnings-per-share by 18.5% per-year? Or that Cisco would not continue stellar earnings growth well into the future? The point is not to predict a particular growth rate, but to put yourself on the right side of the balance of probabilities. On balance it was very reasonable to assume that McDonald’s would provide good risk adjusted returns given the conservative nature of the assumptions. Likewise the assumptions for Cisco during the dot-com bubble would have required everything to go right with virtually zero room for downside.