Do You Value Valuation?

by: David Van Knapp


Valuation is a way of estimating the intrinsic or actual value of a stock.

Some investors think valuing stocks is useless.

This article explains why I think it is useful and worth doing.

Various comments over the past couple of years make it clear that some SA readers and contributors don't place much value on stock valuation.

Here are some of the points that have been made:

  • There are specific examples over the years where "good" valuations were attached to stocks that went on to do poorly or even went bankrupt. On the flip side, some stocks that were called "overvalued" went on to smoke the market over the next couple of years.
  • Many investors worried about high current valuations have stopped buying stocks, building up a ton of cash, waiting for a correction or crash. They are missing out on returns they could have been getting. Some of them have been missing out for 2-3 years or more.
  • Valuation is impossible to know, so investors who try to determine valuation are just fooling themselves or chasing a ghost.
  • You can't control the price you pay for any stock's shares, and in 10-20 years you won't care what you paid anyway.
  • You always pay market price, and with large-cap stocks, knowledge is widespread and pricing is professional. They are never mispriced. There is no way that an individual can bring more to the valuation process than the professionals and the market already have.

The purpose of this article is to offer my own perspective on valuation and whether it offers any value to the retail investor.

Quick overview of valuation

What is valuation?

The basic idea behind valuation is to determine whether a stock is priced rationally.

What does that mean? A rationally priced stock is one whose market price approximates its intrinsic value. Investopedia defines intrinsic value like this:

The intrinsic value is the actual value of a company... based on an underlying perception of its true value including all aspects of the business, [including] both tangible and intangible factors. This value may or may not be the same as the current market value.

The key term in that definition is "actual value." The reason it is key is that the true value of a company is essentially unknowable because it involves operations and events that will happen in the future. The future is unknowable, so the current actual value of a company is unknowable too.

However, in my opinion, that is not the same as saying the actual value of a company is random. Clearly, Johnson & Johnson (NYSE:JNJ) is worth more than Joe's Garage down at the corner.

Over the years, countless ways have been devised to ascertain the value of a company like JNJ. Wise investors are fully aware that it cannot be done with perfect accuracy. But that doesn't mean one cannot try to narrow the range of possibilities to inform decisions that investors make: to buy, hold, or sell a stock.

It is not my purpose here to describe or even to catalog the numerous ways that valuation estimates can be made. Many methods have their adherents and detractors.

I simply want to make the point that investors (if they care to) can make reasonable estimates of the actual value of a company, and from those estimates they can decide whether a stock is undervalued, fairly valued, or overvalued at a particular time.

Probabilistic thinking

Over the years, I have often extolled the virtues of probabilistic thinking. I imagine that some readers don't know what I am talking about.

I merely mean that most everything that happens in life is not simply a one-off or a yes-no phenomenon. Life is full of variable factors that interact and produce the actual outcomes that we see, and sometimes those outcomes are surprising.

While some variables are akin to physical laws (the sun will rise tomorrow at 5:45 AM), others are the outcome of processes that are random, inconsistent, incompletely understood, or uncertain. Two simple examples of probabilistic phenomena are the weather and poker.

We are all familiar with probabilities in weather forecasting. I imagine this is the area in life that most people encounter probabilities most often.

Probability and uncertainty can be seen in poker telecasts practically every day. Poker combines mathematically certain probabilities with uncertainty around what cards will be drawn and how players will play their hands. Here is an example.

In No-Limit Hold-em, each player gets two pocket cards face down to start with. The screenshot shows three players' chances of winning after they received Ace-Queen, pocket Queens, and pocket Kings. Two of them bet all of their chips based on seeing their pocket cards (all in). When they made those bets, they could not see what the other players had. But with no more betting possible, the pocket cards are turned up and everyone can see the rest of the hand as it develops.

At this point in the hand, the Kings are a huge favorite. But notice that the probability of the Kings winning is not 100%; it is 71%. The hand is not over, future events have not happened yet, and exact outcomes are unknowable.

In poker, since there a limited number of cards, the probabilities of each being dealt are precisely knowable. But you cannot know which specific cards will come out until they are dealt.

The next event in the game is that three cards are dealt face up, which can be used by all the players. This is called the flop, and here was the flop in that hand.

The middle player got a third Queen. Look at how that flipped the probabilities: the chances of the pocket Kings winning dropped from 71% to 9%, while the now three Queens went from 10% to 87%. The chances of the player at the top dropped almost to zero: 4% to be precise.

Again, however, notice that the probability of three Queens winning did not go to 100%, and even the probability of the player at the top did not go to 0%. The reason is the same as before: the hand is not over, future events have yet to happen, and they are unknowable.

In other words, the winning chances of each player cannot be described with certainty, but only by means of probabilities. Investing is like that, and that is why I recommend thinking probabilistically.

I won't show any more screen shots of the hand. The player at the top, Mr. 4%, won. The final two cards were both spades, which, combined with his pocket spade, gave him 5 spades: a Flush. Even though there was only 4% chance he could win, he did win. The other two players, whose combined probabilities were 96% after the flop, were both eliminated from the tournament.

I hope you can see the relevance to investing. As in poker, in investing:

  • You know certain facts and events that have already happened. We call that data.
  • You don't know anything that is still to happen, although we can make educated guesses - assign probabilities - about what things are more likely than others.
  • The further out you look, the more opaque and uncertain the future becomes.

Here's what I mean by the last point: these were the probabilities that Mr. 4% would win after each round of cards was dealt - 19%... 4%... 20%... 100%. Who could have predicted that sequence?

If you can't be certain about the future, should you decide that it's useless to try to make reasonable estimates about the future? Now we're veering from math into psychology. Some might advocate taking exactly that approach: you can't know, so why waste time even trying? And indeed, we see that attitude among some investors.

That's not my approach. I like what the sportswriter Damon Runyan said:

The race is not always to the swift nor the battle to the strong, but that's the way to bet.

And what Warren Buffett said:

Tens of thousands of [finance] students were therefore sent out into life believing that on every day the price of every stock was "right" (or, more accurately, not demonstrably wrong) and that attempts to evaluate businesses - that is, stocks - were useless... After all, if you are in the shipping business, it's helpful to have all of your potential competitors be taught that the earth is flat.

The goal and value of probabilistic thinking is to reach decisions that are more right more of the time. The fact that it is impossible to be completely right all of the time, in my opinion, is not a sufficient reason to cease trying to improve your odds.

Valuation visualized through F.A.S.T. Graphs

One reason why I think it is worthwhile to try to estimate valuations - even if imperfectly or imprecisely - is the illustration provided by F.A.S.T. Graphs. Chuck Carnevale, F.A.S.T. Graphs' inventor, had a simple but profound idea, which was to plot valuation estimates and actual prices on the same graph.

By now, we all know what they look like.

This is a 20-year chart of JNJ, and I want to linger on it a bit because it illustrates many parts of my perspective on valuation.

First, let's define what we are looking at:

  • The orange line is drawn at a P/E ratio of 15. This is F.A.S.T. Graph's default ratio for most stocks (those with moderate growth rates). What the line shows is JNJ's earnings x 15, divided by the number of shares outstanding. So you get the "fair" price per share at a P/E of 15.
  • The blue line is drawn at a P/E that matches JNJ's average P/E over the time span shown. In this case that is 18.6, so the blue line is higher than the orange line, but otherwise matches it in shape, since both are based on a multiple of earnings.
  • The black line is JNJ's actual price.
  • The extended area past the point where the black line ends are projections based on analysts' estimates about JNJ's earnings over the next 2+ years. (Note the "E" after 2017, 2018, and 2019. They indicate that the orange and blue lines have been drawn based on estimates.)

You can see that we have the elements discussed earlier in the section on probabilistic thinking.

  • Earnings that have already happened and been reported, which are hard data that anyone can look up.
  • Reasonable projections about events that have not yet happened, namely future earnings. These are not knowable, but reasonable predictions about them can be made by experts that cover the company.
  • The cutoff after a couple of years, because projections are likely to become more opaque and less accurate the farther out you go.

The orange and blue lines are estimates of fair value based on one of the countless ways to estimate fair value, which is the ubiquitous P/E multiple method. The orange version is based on a standard reference P/E of 15 and the blue version is based on JNJ's actual average P/E multiple over the period covered by the chart.

It is the relationship between these "fair value" lines and JNJ's actual price that convinces me that it is worthwhile to try to value stocks. Historically, the actual price tends to stick near the fair value lines, wandering back and forth through them. Sometimes it is above the orange or blue line or both, while other times it is below one or the other or both.

That observation leads to this belief: over long periods of time, JNJ's price line will keep returning to the area of its fair value. It will not always be exactly on either fair value line (in the 20 years of the chart, it was right on the orange line only about 5 times and about 9 times on the blue line), but it keeps returning to the area.

It is possible to believe this while still recognizing that it will not always happen. Just because it has happened repeatedly over the past 20 years does not mean it will ever happen again. Nobody knows the future.

If you look at enough F.A.S.T. Graphs, you see over and over again that actual stock prices have tended to oscillate back and forth around the fair value lines. This makes sense theoretically, and for most stocks it happens in real life. The theory to explain the observations would be that the market, over time, tends to pull prices towards fair valuation.

In the spirit of "that's the way to bet," I am willing to bet that sort of price magnetism towards fair value is likely to keep happening repeatedly in the future. The stock is like a horse with a track record. I can look up the track record and make reasonable predictions about how it will do in the future.

That's my case for why valuing stocks is worthwhile. Now let's consider the viewpoints expressed at the beginning of the article and evaluate them based on the belief that valuation is worthwhile.

1. There are specific examples from the past where "good" valuations were attached to stocks that went on to do poorly or even went bankrupt.

Thinking probabilistically, that is bound to happen. Companies sometimes perform unexpectedly badly, and even well-established companies can plunge into bankruptcy.

Knowing this, it is a good idea to do more as an investor than simply value stocks. Other elements of fundamental research are important, such as evaluating business quality, financial records, balance sheets, and the like. For myself, valuation is the last thing I do. Some investors start with valuation. It doesn't matter what order you do things in so long as you do them. Be like a pilot: use a checklist, and never skip a step.

2. Some stocks that were called "overvalued" went on to smoke the market over the next couple of years.

Thinking probabilistically, this is bound to happen too.

One defense against this is to use both the orange and blue lines in your valuation, as well as other methods too. I use four methods and average them out.

  • F.A.S.T. Graphs orange line.
  • F.A.S.T. Graphs blue line. (I go back 10 years to be sure I am including the Great Recession and the crash of 2007-2009.)
  • Morningstar star rating and fair value estimate. Morningstar uses a different method of estimating fair value; it does not use the P/E ratio.
  • A comparison of the stock's yield versus its historical yield.

Here is an article detailing how I value stocks: "DGI Lesson 11: Valuation."

3. Many investors worried about high current valuations have stopped buying stocks, building up a ton of cash, waiting for a correction or crash. They are missing out on returns they could have been getting. Some of them have been missing out for 2-3 years or more.

I believe it is true that many investors are holding onto more cash than a few years ago. Some of them have sold highly appreciated stocks to lock in gains and raise cash, awaiting better buying opportunities somewhere down the road. Others simply stopped investing.

I don't see these as reasons not to value stocks, because the decision not to invest or to raise cash is a separate one. It's the action you decide to take based on the information you have.

I don't hold investable cash myself. I am fully invested, and have been throughout most of the bull market that began in 2009. (My wife and I do have a 1-2 year non-investable cash bucket.)

But thinking probabilistically, I can see why some investors are shying away from stocks or selling them to actualize cash profits. They are in effect placing a bet that over the long term, and they will be better off holding onto cash now rather than losing it in a correction or crash. Their bet is that over the long haul, the profits they will make by investing when stocks go on sale at some time in the future will overcome the opportunity costs they have experienced by not investing since 2012 or 2014 (whenever they stopped) and/or by selling out now to lock in profits.

Guess what? Like the poker hand shown earlier, we won't know who does better until the hand plays out. In 2020 or 2030, maybe we will know who did better. A lawyer could make compelling arguments for either continual investing or holding cash. What's your standard of proof? Beyond a reasonable doubt? Clear and convincing? More likely than not?

Everyone is different, we all have different time frames, and nobody knows the future.

4. Valuation is impossible to know, so investors who try to determine valuation are just fooling themselves or chasing a ghost.

If by "know" you mean 100% certain, then, of course, no one can know what the true or fair value of a stock is. We went over that earlier. Valuation has a lot of uncertainty built in.

But that is different from concluding that it is not even worth trying to estimate valuation. Most people would not approach buying a car or a house with that attitude. In my opinion, it is not the most helpful attitude when considering whether to buy, hold, or sell a stock either.

If the answer to the question "Is the stock priced too high or too low?" were random, then we could expect to be right 50% of the time and wrong 50% of the time. But if we can move that needle even a little bit, and it doesn't cost much money or take much time, then estimating valuations is worthwhile.

Does that mean every undervalued stock you choose to buy will be a good decision? Of course not. Does it mean every overvalued stock you choose to sell or avoid buying will be a good decision? Same answer: of course not.

But thinking probabilistically, we are not expecting certainty. We are just trying to improve our odds. I believe that you can improve your odds with intelligent valuation techniques, so like Damon Runyan, my conclusion is that it's the way to bet.

5. You can't control the price you pay for any stock's shares, and in 10-20 years you won't care what you paid anyway.

I think this is generally false, except maybe for the part about not caring.

First off, you can control the price you pay. You can refuse to purchase above a certain price, or stated better, above a certain valuation. Some investors do this with limit orders, some do it mentally. But what you pay is in your control.

If refusing to pay above a certain valuation means you go a long time, or forever, without purchasing the shares of an excellent company that you want to own, that's a consequence that you will need to consider. That puts us into the realm of psychology again.

Maybe you will regret never having owned, say, 3M (NYSE:MMM) because it always seemed overvalued. Perhaps the anticipation of regret overrides your refusal to pay above a certain value. That's fine, we're all human. Who hasn't overpaid for some things in their life? I sure have.

The second part about not caring in 10-20 years what you paid may be psychologically common, but mathematically it does make a difference. I used to think that over time price differences faded in importance, until I did the research for this article: "Does Valuation Matter for Dividend Growth Investors?"

In the article, I compared two scenarios: one in which the investor paid 25% above fair value for a stock, and another in which the investor got the stock for fair value.

Here was my key finding:

It is hard to avoid the conclusion that valuation matters. On every metric [price, dividends, etc.], the stock bought at fair value delivered more than the stock purchased at an overvalued price. Not only that, the performance gap widens with each passing year.

The final sentence was the key. The hypothesis that I was testing in the article was that in 10 years I wouldn't care what I had paid for the stock. The result, which surprised me, was that as more time passed, the favorable results from not overpaying actually kept growing each year. They didn't fade, they strengthened.

Of course, "not caring" is a psychological state, not a mathematical one. It could well be that someone won't care much in 10 years what they paid for a stock, especially if they got good results from it. When I stated at the beginning of this section that I thought the comment was generally false, I was only referring to the mathematics of it.

6. You always pay market price, and with large-cap stocks, knowledge is widespread and the pricing is professional. They are never mispriced. There is no way that an individual can bring more to the valuation process than the professionals and the market already have.

This is demonstrably false.

First, while you always pay market price for anything you buy, you don't have to buy. As Warren Buffett says, it's a no-called-strikes game. You don't have to swing at every pitch, or any pitch for that matter.

Second, the idea that large-cap stocks are never mispriced ignores the data. It is clear that even the largest companies go through periods of mispricing. The following table shows the differences between the 52-week highs and lows for the biggest members of the S&P 500 right after the Brexit vote last June.


The average gap is nearly 50%. Is it logical that the true value of a company like Microsoft (NASDAQ:MSFT) or JNJ varied by more than 40% in 52 weeks? Not to me. Far more likely is that the market, which is focused so intently on daily, weekly, or quarterly results, routinely puts an incorrect price tag on even the most widely followed stocks.

As the author of the article from which the chart is derived said:

Many of these businesses have very stable earning power with mature, slow-changing business models. It is remarkable to me that a company like Johnson and Johnson can see a $100 billion fluctuation in value between its 52 week high and its 52 week low. Of course, these fluctuations are often correspond very closely to the fluctuations in the overall market, but the fact is that there is almost no chance that a company like JNJ is worth $225 billion in one month, and a few months later is worth $325 billion. There isn't that much that changes in the course of a few months in most companies of this size.

He concluded that there are bound to be some stock market bargains from time to time, even in the largest companies in the world. He also noted that it is the ability to adopt a long-term mindset that enables one to capitalize on this category of mispricing.


These are my personal conclusions:

  1. Valuation is worthwhile.
  2. Valuation is inherently approximate - like a house assessment - but often directionally correct.
  3. The goal of valuation is not to avoid all mistakes but to improve your odds of success.
  4. Buying stocks at better valuations pays off not only immediately (as in getting more shares that yield more in dividends right away) but also over the long term.
  5. Performance gaps based on original valuation actually widen over time.
  6. Even large-cap, widely followed stocks can be mispriced. The market price is not always "correct."
  7. You may have reasons to ignore what valuation tells you. It's not a crime to "pay too much" for a stock. Over a long time, the results from a high-quality but overvalued company may be quite worthwhile, and psychologically you really may not care what you paid many years ago, even if someone demonstrates to you that you could have done better.

Disclosure: I am/we are long JNJ.

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.