Here as a follow-up of Gurufocus.com’s interview with him, Dr. Price continues explain the finer points of conscious value calculations.
Q: Warren Buffett once summarized Ben Graham's teachings into the three cornerstones of sound investing? What would you say if you were to summarize conscious investing into a few fundamental ideas?
A: My basic framework for conscious investing is that we need to do three things.
1. We need to have a measure of value. (I mean value per share, but I won’t keep saying “per share.”) It could be assets, equity (book value), earnings, dividends, free cash flow, enterprise value, etc. Or it could be a combination of these.
2. We need to be able to make statements such as: In N years (say, 5 years) I am confident that the level of this value will be at least $x.
3. We need to be able to make statements such as: “In five years I am confident that the market will pay me at least $y for every dollar of value.”
With this done, I can then conclude that I am confident that a single share will be worth $xy in 5 years. I can also figure out my expected return if I buy a share today. On top of this, I might get extra profit from dividends.
As investors we do not like negative surprises in levels of value, but we love positive surprises. So we are trying to specify a level $x so that we are confident that future value at the specified time will be at or above that level.
I am not saying this is easy. This is where all the information about the company comes in. Debt levels, stability of growth, economic moat, etc. The more doubt you have, make your estimate of $x lower. It is like Buffett’s quote when he says he want companies whose “earnings are virtually certain to be materially higher, five, ten, and twenty years from now.”
In addition, as investors, we do not like negative surprises in any estimates of what the market will pay for this value. So this time we choose a level $y so that we are confident that the market will pay at least this much for every dollar of value.
If you base your investment on the projection that $y will be materially higher in the future than it is now (without too much growth assumptions of $x), you are usually called a value investor.
As an example, suppose that our measurement of value is in terms of book value. A desirable investment would be a company that the market is currently paying a low price for its book value. Consider BRK. Over the past 25 years the price to book ratio has ranged from a lows of 1.17 and 1.19 to a highs of 2.02 and 2.23. Currently it is around 1.20.
Now we have to make projections for the value the market will pay for the book value of BRK. Let us be conservative and suppose that we are confident that book value will be at the same level or higher in one year. In other words, there is no growth in book value. Also suppose we are confident that the market will pay 1.35 per dollar of book value in one year. This means that we are saying that the price of BRK will rise by at least 12.5% over the next year. (This is the growth from 1.20 to 1.35.)
Alternatively, if you base your investment on the projection that $x will be materially higher in the future (without too much consideration of $y), you are usually called a growth investor.
Of course, as most people will be aware, Buffett and others said that restricting yourself to value or growth investing was unwise since all investing involves finding value. Buffett said: “We think the term ‘value investing’ is redundant. What is ‘investing’ if it is not the act of seeking value at least sufficient to justify the amount paid?”
Q: The expected return method seems to rely on future earnings projections and the terminal value at the end of the projection period. Many believe those projections are less reliable than asset value and replacement value. What do you think?
A: In the framework I described, it is up to you to choose which measurement of value you use. I agree that asset value is more reliable at each point in time. But you still have to factor in the market’s reaction to asset value. In other words, how much it is willing to pay for asset value? If you think that the market is currently undervaluing a stock in terms of assets, you have to make a projection that in the future the market will pay a higher value for the assets in the future. You also have to make a projection about the growth of the assets.
I use earnings. There are many reasons. The main one is that this is the core aim of (most) businesses, namely to grow their earnings. Even levels of management bonuses are often based on growth of earnings. For good reason earnings or net profit are referred to as the bottom line.
However, it is possible to use other measurements. For example, book value is used in the BRK case above.
Q: Dr. Price, your method places a very high priority on predictability of earnings and revenue. Do you think the market will ever go through extended periods of time (4+ years) where you could lose money because the stable equities are priced too high? Thanks!
A: Absolutely. There are always overall market cycles with times when “the market” could lose money. This is why it is better to focus on individual companies, rather than “the market.” The market was extraordinarily high around 2000. So since then market indices have done very poorly. But there have still been excellent individual companies.
There are always companies selling at sensible, even bargain, prices. Sometime there are more, sometime there are fewer. But they are there.
In “Beating the Street”, Peter Lynch said: “The key to making money in stocks is not to get scared out of them.” In May 2007 at the Annual Meeting, Buffett said “Something bad will happen, but you could go back at any time in the last 100 years and say the same thing … you can freeze yourself out indefinitely.”
Q: How do your models predict the future by using historical data IF despite consistent historical data the future will not replicate the past? My point being, no matter how great a company's historical data is, the future will only prove to replicate the past et ceteris paribus. Even the largest moats fall over a long enough time span. Oh, and please don't say something like “If company X traded at 15 times earnings historically, I assume it will only trade for 12x earnings to incorporate a margin of safety”. That is not an acceptable answer.
A: I take the position that all investing involves making forecasts or projections. Sometimes these forecasts are more like wishes such as: “The price will be higher in the future than it is now.” Not much help for serious investing.
Other times the forecasts are very mild such as: “The company will be able to sell its assets in the near future for the amounts described in the balance sheet.”
Still other times the forecasts are highly problematical such as in intrinsic value DCF methods: “The growth rate of free cash flow out to infinity will be 3%.” or “The discount rate out to infinity will be 12%.”
In fact, the whole idea of investing is to do something with the anticipation of a particular outcome or range of outcomes in the future. (Of course, for investing I don’t mean forecasts in the usual sense. Rather I mean in the sense of forecasting that a particular level will be reached or better. Also, sometimes the forecasts are implicit and hidden, rather than being explicit. But this is whole other topic.)
If you accept that investing involves making forecasts, then we have to decide what we are going to forecast and for how long. Also, to make these forecasts, of course all we have available is current and historical information. So we have to use this information in the best way that we can. I mostly base my calculations as ultimately heading towards projections about earnings (more particularly, earnings per share).
But, expanding on what I said for an earlier question, for most companies I do not know how to forecast earnings with any reliability. Fortunately, for a few companies because of modest debt levels, stability of growth of sales and earnings, economic moat, quality management, and so on, I think that it is possible to set out steps to be able to make statements with reasonable confidence. These are statements such as it is likely that the earnings will grow by at least a certain rate over some specified reasonable time span.
Also, at certain times the market may only be willing to pay a lower price for these earnings. Again, at times I believe that it is possible to make statements such as it is likely that the market will pay a certain level for these earnings at some specified period in the future
Of course, sometimes neither of these outcomes will come about. As you say: “Even the largest moats fall over a long enough time span.”
This is why I have crunched through very large databases over extended periods to check the methods. Also, I try not to rely on time spans that are too long. And having done all this, I introduce margins of safety calculated using computer algorithms.
Of course there are no guarantees. But all the assumptions are testable. And for me, at least, they provide a rational framework which has proved to be successful.