Continuous outperformance: Uncertainty in the 3rd dimension of value creation?

Consistent outperformance?

I have been struggling with reconciling (i) how current expectations of future value creating activities in companies should be relatively well understood by the market, and (ii) the fact that the stocks of high-performing companies can continue to in a persistent manner outperform the market. Let me clarify. The issue I want to dive into is not the fact that the operations of great companies affect share prices. Rather, I want to discuss the fact that they often do so in a persistent manner.

Let me clarify further. I am not talking about the fact that some firms have the capacity to be on the EVA-positive (i.e. economic value added is positive) side of the spectrum for longer periods of time. I am talking about the fact that some firms that are on that side of the spectrum also see share price appreciations (especially augmented with dividend reinvestments) way in excess of their cost of capital, over periods of decades. What is it about companies like Clorox and Colgate Palmolive that historically has controlled expectations to avoid immediate price appreciation followed by modest returns, and instead seen continuous excess returns over decades, excess returns shared by thousands of thousands of shareholders?

Addressing some potential concerns

Before I try to give you my take on this issue, I want to address a few concerns the reader might have at this stage. If you read my last post, you might have issues with my deterministic approach of what drives value, and what the analyst should be concerned about. You might have an entirely other view of value. You might define ROIC differently. You might argue that cost of capital is more of an academic way of viewing risk that what the value investor should be concerned about. You might say that profit and capital metrics must be adjusted to such an extent that EVA often is not even worth working with as a practitioner. You might furthermore claim the “soft” values of companies like management incentives are not considered to a large enough extent. In that case, fair enough, there are some points to consider in all those concerns. Believe me, the last thing I claim is that the investment process can be boiled down to a few simple rules, regardless of the situation. Investing is closer to the realm of art, than it is to the realm of science. Investing has just as much (if not more) to do with psychology, biases, understanding where and when markets fail, evaluating how reasonable expectations are and other qualitative assessments as it has to do with understanding mechanics and maths of a spreadsheet.

Notwithstanding these difficulties, we must stick to the fundamental rules of corporate finance when we assess investment opportunities, and we have to stick to the (same) fundamental rules of valuation when we value companies. Most of the concerns raised above are still means to get to the end: assessing value creation over time. The point of this post is not to assess how each individual factor the investor should consider can affect the value of a company. The point of this post is to provide some insight as to what the mechanism can be that cause extensive, prolonged, time periods of capital gain to the shareholder. Going forward, simply think about EVA as a true measure of the economic value created at any given time, regardless of what might actually be the underlying reasons why the company can create that value, and how we in practice make sure to quantify that value creation perfectly.

Digging deeper

Okay, let’s get back on track. The issue I raised was why it often seems as if expectations of highly profitable companies are, ex ante, chronically depressed, causing persistent outperformance. The logical, top of mind conclusion is simply stating that the intrinsic value simply keeps on changing for the better; a conclusion that obviously makes sense. But I want to dig deeper, as the previous conclusion merely scratch the surface of the relationship between expectations and the marginal investor’s view and practical assessment on fair values on a continuing basis, especially for companies later in the life cycle.

Let me give you an example. Let’s say the year is 1997, and you are valuing Colgate-Palmolive, the 200-year-old consumer products company from the U.S. with trailing revenues of around $8,5bn, approximately a 15% return on capital and a growth rate more or less in line with inflation. This is the type of company you learn to value in business school. With the assumption that the company will continue to grow in line with the economy, you create a spreadsheet where you estimate a likely scenario where profitability start to converge, either to industry means or even to the cost of capital, say after 10 years. Obviously, any company with sustainable advantages can maintain profitability above cost of capital for long periods of time. In this case, let’s assume it starts to converge to a cost of capital of 10% 10 years into the estimation process. Note that setting return on new invested capital (RONIC) equal to WACC does not mean that the entire enterprise suddenly has a profitability of 10%. It is just the new invested capital that does not create value. Plotting the value creation on a graph, the following pattern is implicitly assumed:


Imagine the above picture is the view of the average investor in 1997. Then time passes. What happens to the market value of Colgate if investors simply apply the same logic, 10 years later, again? You might already have the answer to the question, but first, let me clarify again what has happened. Market participants have simply updated their spreadsheets by changing the years in the top row, and increased the size of the company to accommodate for the inflation that has occurred during the years. Looking forward, the assumptions once again looks like this (note the updated years):


Then, 10 years pass again. We just apply the same thought-experience, and we are in 2017. Market participants apply the same logic, and we have the following:


Okay, now let’s think about that question again. What has happened to the value during the years 1997 and 2017? To simplify, let’s think of the value as the Enterprise Value including the dividends received, which are assumed to be reinvested in the business (If I do not make this assumption, any company not growing would have a 100% free cash flows to dividend conversion and we would not see the effect I am after in the wealth of the shareholders). The answer is that the value (in terms of dividends and share price appreciation) has increased, way in excess of cost of capital. For some of you this is probably completely obvious. For some it does not make intuitive sense. The reason why value has inceased is that over time the assumption of zero value creation on new capital has been continuously pushed into the future, causing us to augment the share price continuously with the excess value we previously had not assumed. In the previous pictures we just looked at 3 discrete times, with a very far time between.

The reason why value has inceased is that over time the assumption of zero value creation on new capital has been continuously pushed into the future, causing us to augment the share price continuously with the excess value we previously had not assumed.

In reality, of course, the company and the environment gives us continuous clues as to whether the assumptions we previously held need to be updated. But just to clarify what has changed between the years, ask yourself the following question: What would the value have been in 1997 if we knew that the company had the capacity to maintain the EVA spread for 30 years before it started to decline?


Obviously, the value would then have been the present value of all the EVA created in the future – which we now know starts to decline first in 2027) – together with the Invested capital today. As a reminder, when we are using an EVA model to value a business, we sum up the invested capital today with the present value of all the future economic profits or losses (either defined as NOPLAT – WACC*Invested capital, or (ROIC-WACC)*Invested Capital, where the second one is easier to grasp with the current graphical representations). All I have been doing here is describing that those excess profits differ due to different assumptions, and that when time passes, that effect is substantial on value, but often comes in incremental steps. It still might not be clear to you how it is possible that the same future assumptions in any given time can yield an increase in value. Let me then give you another angle of understanding. Moving from one year to another, the value will be greater due to the fact that the invested capital will be bigger, both today, and in all the future EVA calculations discounted back to today. Either, the reinvestment has caused the business to be a couple of percentages bigger, and with everything else equal, the intrinsic value has obviously increased (since all EVA assumptions remain unchanged), or we have simply reinvested into the company, causing our stake to be bigger than before. This last point might also be a source of misconception. It does not aim to make the argument that dividends, nor growth, created value. It does not. But in the EVA framework, we can see value through those tools, even though the true value being created is by the company’s actions in terms of pushing the profitable period longer into the future (or, the aggregate market realizing that it has the capacity to do so as we move forward in time).

If you know the history of Colgate-Palmolive, you know that the reason behind their outperformance is not pushing the EVA conversion into the future, it is actually improving the spread. Sometimes, there are implicit assumptions in an already mature company that profitability won’t become much improved. In this case, there has been a lot of shareholder value created since the 90’s due to this false preconception.

But why should you care?

At this point, you are probably have one of the following thoughts:

  1. I did not learn anything from this dumb post
  2. I disagree with Oscar’s points
  3. This was interesting, and I learnt something new

In response to the first point, I am glad you have though of this yourself, and I am sorry this was to trivial for you. I think you should care anyways. Why? You can still augment this thinking in your own way. You can try to reconcile it with different investment philosophies. I furthermore encourage you to then think about what can cause disagreements about value, what the clues are to future economic profits using this framework, and what combinations of multiples and key ratios that can be used to assess whether the market is wrong in its assessment of value, or what combinations of EVA and growth might be implicitly built in to the current enterprise value.

If you disagree with me, let me know in what way in the comment section. I want every response I can get! A couple of arguments I might expect you to bring up though, are how growth fits into this picture more in detail, and also what the role of expectations are, and whether or not expectations equally likely can’t work in the opposite direction.

Regarding growth, I will post another text on how I look at value “in 3D”, which I hope will clear up any confusion that might exist. This post is called the 3rd dimension of value creation simply since time (on the x-axis in my graphs) and all uncertainty it brings to valuation is what underlies the potential for continuous outperformance. But as you can see, my graphs only have two dimensions. The first dimension as I see it is Invested capital. This dimension is not seen in the graphs I have shown you in this post, but hopefully the role of invested capital, and thus growth, will be clearer when I found a good way to present it. The second dimension, if it is not clear yet, is the ROIC-WACC spread.

When it comes to the role of expectations, which might be another point of disagreement, I think about the issue the following way. There is always an expectations treadmill, as so famously stated by McKinsey. Logically, the share price reflects the speed the company is running at (in terms of fundamentals like Revenue growth and ROIC), and therefore the performance of a manager or company can’t reasonably only be evaluated from share price performance (especially in the short term), as it might be extremely difficult to uphold current performance, and any decrease in performance relative previous performance can, and will, be punished by the market.  With this as a backdrop, it is tempting to draw the conclusion that I am basing my arguments on something that does not make sense given how the markets react, and that I therefore am in direct conflict with the expectations treadmill hypothesis. You might argue that my view that unchanged performance – both in terms of spreads and growth – will lead to above market returns is complete garbage. The problem with that counterargument is that no one should with a straight face be able to make the case that the market is so inefficient that it assumes that a company will run at sprint speed forever, just because it does so today. My argument is therefore that the market, in its expectations treadmill thinking, still assumed that the treadmill will come from an incline sprint-mode to a flat strolling pace in due time. When someone starts sprinting the first 500 meters up Mt. Everest in 20 minutes, I would be naive enough to extrapolate on that pace and say the person would reach the top under 6 hours. Similarly, I therefore argue that the market builds in assumptions of decline in fundamentals, and try to value a company based on a range of potential outcomes, all from extremely positive to extremely negative, with that decline as a key insight going into the valuation process.

Just like the average mountain climber will face changes in weather, different physical obstacles, lower oxygen levels near the top of the mountain and risk of complete failure (read death), the average company won’t go through the later stages of its life cycle without facing competition and market environments doing its best to push the maturing company's profitability to very low levels. Reflecting back on the main issue of this post, I would then argue that great companies have long-term qualities often overlooked by the market, simply because the average outcome of companies have the capacity to provide us with such a convincing picture of what should be the case, which is: the treadmill must slow down eventually. Companies like Colgate Palmolive and Apple, though, are like a native Nepalis with an unprecedented cardio, navigation skills and experience starting to sprint up Mt. Everest dressed like tourists. We continuously have to revise the point in time when we assume the speed will have to decrease, thus continuously underestimating their performance. The speed, in this case, primarily refers to the EVA spread if it is not clear.

Companies like Colgate Palmolive and Apple are like a native Nepali with an unprecedented cardio, navigation skills and experience starting to sprint up Mt. Everest dressed like tourists.

Lastly, if you are one of the people that experienced that post changed or augmented your view on how to think about value creation and continuous outperformance, I am very pleased. Hang in there for the next one!

I want to end this post by really pushing the fact that a great analysis has the capacity to forsee value creation in the long-term, and the payoff to such an analysis can be huge. But that payoff does not have to come tomorrow, this month, or the coming year for that matter. As Joel Greenblatt famously said:

I always tell my students,” if you do a good job valuing a stock, I guarantee that the market will agree with you”, I just don’t tell them when.
— Joel Greenblatt

// Oscar