A Good Metric is Hard to Find - Return on Capital

by Scuttlebutt Investor

“Leaving the question of price aside, the best business to own is one that over an extended period can employ large amounts of incremental capital at very high rates of return.” 

– Warren Buffett, 1992 Berkshire Hathaway Shareholder Letter


Return on Capital or Return on Invested Capital (ROIC) is something I think about a lot. The list of metrics I look at when I analyze businesses is long: revenue growth, operating margins, free cash flow, payout ratios, etc. But if I could look at only one metric about a business to judge the quality of that business- ROIC would be the metric. Good businesses generate high returns on capital and they do so with consistency. Return on capital is important because it is a fundamental driver of valuation. Businesses that can generate higher returns on capital can invest less in capital expenditures and thus generate more free cash flow to distribute to shareholders. Since value is the present value of future cash flows, this makes these businesses more valuable all else being equal.    

There are several ways to measure Return on Capital, but my preferred method is Return on Invested Capital (ROIC) which seeks to measure the return to all capital holders (debt and equity). ROIC basically computes the underlying return that a business earns on its cumulative investments in the business, no matter how those investments are financed. ROIC is a ratio of a company's After-tax Operating Income to the Capital Invested in the company. The nuances of calculating ROIC can be hotly debated by finance nerds like myself but my preferred approach is what is often referred to as the Asset Approach. I prefer the asset approach because the denominator consists of the assets that a business has invested in so it is a bit more telegraphic of the core drivers. The formula is:


ROIC = NOPAT / Invested Capital


Let's unpack that a little more though:

Net Operating Profit After Tax (NOPAT) = Operating Income x (1 - normalized tax rate) 

Note: Operating income should be adjusted for special or non-recurring charges as appropriate


Invested Capital = Net Working Capital + Net PP&E + Other Operating Assets - Excess Cash
Net Working Capital = Current Assets - Non-debt Current Liabilities
Excess Cash = Cash on Balance Sheet - Required Cash

Note that the traditional definition of Net Working Capital usually excludes cash from the Current Assets but I include it in this case because we then subtract out the excess cash


Putting It in Practice

I recently calculated ROIC for two very different companies with two very different business models and I thought that the output was worth sharing. Putting both calculations side by side helps exemplify the definition and various drivers a little better. Costco (COST) - a traditional brick and mortar retailer and the Company formerly known as Google -Alphabet (GOOG) - a web search company (among many other things) are two very different companies with very different business models and their respective ROICs drive this home.

ROIC graphic.png

I've made some assumptions here and I can't say with certainty that all of them are perfectly correct. Some of the items lumped in with "Other Operating Assets" may not be pure operating assets. And I've made a high level guesstimate of required cash based on what I know about the businesses. Also, I have not made adjustments to capitalize leases here, which should normally be done for lease heavy business models.  Luckily, Costco owns much of its property so it's not as big an issue here. Nonetheless, these assumptions don't impact the key insights of the comparison very much so I thought it ok for this example.

My observations and insights:

Alphabet and Costco are massive businesses with immense scale. They both generate an incredible amount of revenue - $75bn and $118bn respectively, but their operating margins are vastly different. Costco, as a retailer, operates on razor thin ~3% operating margins while Alphabet, as a software/web services company, churns out robust ~26% operating margins. This results in massive differences in their operating income - Alphabet - the company with much lower revenue ($43bn lower in fact), actually generates vastly more operating income - $13.5bn more.

The higher operating margin alone doesn't make Alphabet a better business than Costco.  After all, it's important how much money is spent to generate this operating income. This is the where the all important denominator comes in - Invested Capital.

I was surprised by Alphabet's total invested capital of ~$52bn, which is much higher than Costco's at ~$15bn. I would have expected the brick and mortar business of Costco with its ~700 gargantuan warehouses to surely be larger than what I thought was Alphabet's asset light business model. I underestimated Alphabet's offices all over the world and their massive investment in data centers (referred to as information technology assets on their BS).  

Despite a significantly larger capital investment, Alphabet still comes out on top when it comes to ROIC - with more than double the ROIC at ~34% versus ~15% for Costco. While the large capital invested denominator for Alphabet is a drag on ROIC, the larger numerator (NOPAT) more than makes up the difference. Costco's ROIC is still respectable on its own and definitely for the retail industry (which has historically averaged ~10% ROIC as an industry), but its low margins are no match for Alphabet.   

Does this mean that everyone should go out and buy Alphabet stock?  Not quite.  A few things I would say here:

  • Sustainability: We would want to analyze the ROIC for both companies over multiple years to understand how sustainable it is and how susceptible it is to a cyclical downturn
  • Drivers: Need to understand the core drivers of the calculation. What is driving the higher or lower ROIC and are the right inputs being used?
  • Context: We can't look at ROIC in isolation. We need to look at other metrics and also understand the core business and business model.
  • Valuation is the other side of the coin. Alphabet a wonderful business but only if we're able to pay a fair price. A Ferrari is a hell of a car but it's not a great buy if you pay a billion dollars for it. It's also worth mentioning the inherent risk of the business models - one could argue that Costco is less risky than Google because it operates in a more traditional industry where product life cycles aren't changing by the minute. This latter point is debatable (have you heard of Amazon?). But let's pause a moment and talk more about the interplay between Return on Capital and valuation.  


It's the Long Term, Stupid

"In the long term, it's hard for a stock to earn a much better return than the business which underlies it earns.  If the business earns six percent on capital over forty years and you hold it for that forty years, you're not going to make much different than a six percent return-even if you originally buy it at a huge discount.  Conversely, if a business earns eighteen percent on capital over twenty or thirty years, even if you pay an expensive price, you'll end up with one hell of a result."

-Charlie Munger, Poor Charlie's Almanack

I think that this is one of the most powerful yet underestimated principles across all of investing literature and it makes me think that I should have probably started this post with it given its importance. The core idea is exceptional. Investors expend tremendous effort looking for businesses trading at discounts to intrinsic value to reap the difference between price and intrinsic value but Munger is saying don't waste your time trying to exploit these often minute differences. Because over the long term, businesses (and their stocks) will reflect their return on capital and the majority of the return on a given investment will be driven by this factor. Simple and brilliant.  

There are two additional points I would add here: 1) this only works if high returns on capital are sustainable - the company must be able to sustain its return on capital for long periods of time. This requires a high quality business that benefits from a durable competitive advantage and even then, it's not always a given that the company can maintain its returns; 2) Valuation is still important but it becomes less and less important as the holding period increases. If you are holding a stock for 1 or 2 years, then valuation is critical because the majority of return will be driven by the difference between price paid and intrinsic value. However, if you are holding a stock for 30 years, a discount or premium to value becomes insignificant as it gets spread out over many years. Base Hit Investing has a wonderful post explaining this very concept in more detail, which I'll also re-list below.  


On the Shoulders of Giants

I've only scratched the surface when it comes to ROIC. It's a powerful (although not perfect) metric that I could probably devote an entire book to if I had the time. Yes, ROIC isn't perfect and there are some shortcomings including consistency of calculation -you can find about a hundred different ways to calculate it, but there is no perfect metric so it's the best we've got.  There is a litany of writing about ROIC by people far smarter on the topic than myself. For those interested in digging a bit deeper into the nether reaches of ROIC, I thought that some of the best writing on the topic was worth sharing here. It's a combination of both technical posts on the calculation of ROIC and philosophical posts on ways to think about it. One of the posts emanated from a conversation I had with the fine gentlemen at Ensemble Capital.  I'll start there.

Intrinsic Investing Blog by Ensemble Capital

A couple months back - I had the opportunity to meet with Sean and Arif from Ensemble Capital - two very smart guys that manage and invest money with a philosophy that's not very different than my own - buying businesses with durable competitive advantages at a reasonable price, a concentrated portfolio of high conviction positions and a long term holding period. In addition to generating very respectable returns over the last ~15 years, they also pen an astute investing blog- Intrinsic Investing that I've followed for quite some time.  After talking at length about individual names, I was interested to hear more about how they think about ROIC - Sean kindly responded to some of those questions about ROIC that still keep me up at night in a very insightful post.

Base Hit Investing

Base Hit Investing is one of my my favorite investment blogs and John - the author- has penned a series of very insightful posts on ROIC.  

Damodaran Online

    Aswath Damodaran is an NYU professor and the guru of valuation.  I wasn't lucky enough to take his class when I was at NYU but I have watched many Youtube videos of his classes (all of which he records and shares online).  His book - The Little Book of Valuation- is on my reading list.

    Credit Suisse

    This is a research paper that I randomly came across online.  I thought it was a valuable instructional resource on ROIC.

    McKinsey Quarterly


    Untapped Pricing Power and the EpiPen

    by Scuttlebutt Investor

    "There are actually businesses you will find a few times in a lifetime where any manager could raise the return enormously just by raising prices-and yet they haven't done it. So they have huge untapped pricing power that they're not using. That is the ultimate no-brainer"

    -Charlie Munger (Poor Charlie's Almanack)


    Pricing power is one of the most desirable characteristics of a business. It enables managers to raise prices to not only keep up with inflation of input costs, but it also enables managers to drive revenue and profit increases in excess of inflation. While a price raise in most businesses results in a loss of demand, a price raise in a business with pricing power only minimally impacts demand or not at all. But pricing power isn't a characteristic that exists on its own. Pricing power is typically derived from some or many other characteristics of the business or the industry in which it operates. These other characteristics that drive pricing power are typically competitive advantages that a business possesses like a strong, reputable brand name, owning proprietary patents, or benefitting from a customer base with high switching costs to a competitor product.  

    The topic of pricing power has been top of mind for me for some time as I see more and more press about pharmaceutical companies coming under fire for hiking prices of their drugs with little or no competition. When you're the only game in town, you can price that game at a hefty premium. Sure, companies like Disney, Coca-Cola and See's Candy have some untapped pricing power from the unique experience or strong brand equity as Munger goes on to elaborate, but life saving drugs are the ultimate extreme of pricing power - they mean the difference between life and death for their consumers and as a result command nearly unlimited pricing power.  

    If you're not familiar with the recent controversy surrounding Mylan and the EpiPen, here is an article and the summary version. Basically, Mylan purchased the rights to EpiPen from Merck in 2007 and went on to hike the price of the drug/injector from ~$100 in 2007 to ~$600 in 2016 - a nearly 500% increase in about 9 years. Most of that increase has occurred in the last few years and the most recent increase led to consumer outrage as a larger part of the burden begins to fall on consumers with the increasing prevalence of high deductible health insurance plans.

    Source: CNN Money

    What's most interesting to me is that the EpiPen isn't some fancy new drug with massive investments in R&D that need to be recouped. Rather EpiPen is a simple injecting device that provides a quick way to inject someone experiencing an allergic reaction with epinephrine- a drug that's been around since the early 1900's and is not protected by any patents. While the drug is no longer under patent, Mylan owns patents around the design of the medical device (injector), which apparently is more effective than competitor injector devices. As a result, Mylan commands about 85% share of the market for epinephrine injectors in the US.  

    But we've heard this story before. Turing Pharmaceuticals and its controversial CEO (to put it kindly) similarly came under fire (and a congressional investigation) when it hiked the price of Daraprim- a little known and low volume drug used to treat malaria and HIV patients- nearly 50x overnight.  Another company- Valeant based its entire business model on acquiring drugs with little or no competition and hiking prices to the highest price that the market would bear. And a lot of sophisticated investors including Bill Ackman and the very well respected Sequoia Fund bought into this strategy and thesis eventual leading to major losses.

    This raises a few interesting questions for me. First, how have these pharma companies sustained such sky high price hikes? Sure this makes sense when a company patents a new drug like a treatment for Hep C in the case of Gilead. But it doesn't make as much sense when thinking through some of the examples above. Taking the example of the EpiPen, it seems nearly impossible in today's day and age of free markets and generic competition, that Mylan could raise the price of the EpiPen so high, when it delivers a mainstream drug invented in the early 1900's that is not under any form of patent protection. The answer as with most companies that demonstrate strong pricing power goes back to competitive advantages. Mylan has been able to raise the price of the EpiPen because it benefits from some key competitive advantages (that don't include a patent on the drug). Putting aside the morality of such egregious price hikes (we'll get to that), Mylan benefits from many competitive advantages that enabled it to command pricing power for a commodity drug. Some other industries can stand to learn from how Mylan has been able to command such strong pricing power for a commodity product (the approach on the other hand might need a touch of humanity and morality- as I said, we'll get to that):

    • Strong brand name:  EpiPen (as with Band-Aid or Kleenex) is  brand name that defines the category. There are some generic competitors but Mylan has invested considerable marketing dollars to build awareness specifically for the EpiPen brand and ensure that EpiPen is prescribed by name by doctors.
    • High barriers to entry: The FDA has high hurdles for generic competitors and has rejected some competitor products (like one from Teva) from coming to market. It is this lack of competition has provided the window for Mylan to hike prices. Furthermore, Mylan has aggressively lobbied the FDA and other key figures to reject competitor products for their supposed lack of safety.  Whether that is true, we can't opine.
    • Proprietary manufacturing process/technological advantage: To Mylan's credit, they have perfected a manufacturing process that other companies have been challenged with.  Sanofi and Amedra (which made AUVI-Q and Twinject) have been forced to recall their products from market due to device/dosing issues.  The only competitor in the US (Adrenaclick by Impax Labs) has supply issues because they have not yet figured out how to automate their manufacturing process
    • High switching costs- 2 ways:  1) An anaphylactic allergic reaction requires quick and immediate action. In tense, quickly evolving situations like this, it is critical that the person administering the drug knows how to use the device. There are slight differences between the various devices on market and there is a potential learning curve for alternative devices.  Thus, there is value in stocking the device that more people are trained on and are likely to know how to use- the EpiPen. The resources that Mylan has invested in training a broad swath of the population on using the EpiPen are paying dividends in this regard.  2) An EpiPen is classified as both a drug and medical device by the FDA. As a result of this classification, pharmacists in 29 states can't simply switch a generic epi injector for an EpiPen the way they could with medications that do not require a device, such as antibiotics. This ensures that an EpiPen prescription remains a branded EpiPen prescription and not a generic epinephrine injector prescription. 

    Mylan is sitting in a sweet spot as this confluence of competitive advantages has enabled it to garner a significant price premium for its product.  I am convinced though that this pricing power will be short lived. The spotlight that the pricing controversy has created along with the high profits that Mylan is earning is likely to attract new competition and put pressure on the FDA to approve competitor products.


    My second question is a question I ask myself often- WWMD - What would Munger do? What would Munger do if he was running one of these pharma companies? As in the quote above, Munger has on several occasions talked about exploiting untapped pricing power where it exists but he also recently referred to Valeant as "a sewer" at the Berkshire annual meeting.  So what's the disconnect. These pharma companies have in some ways followed the capitalistic playbook that Munger laid out on more than one occasion. These companies have capitalized on their untapped pricing power and nearly inelastic demand for their drugs by consistently raising prices. "It's the ultimate no-brainer," says Munger. If I have to conjecture, I think that Munger would have also increased prices for some of these drugs to tap some of the latent pricing power. However, I am convinced that he would not have been as aggressive and egregious in his approach, but rather tempered and responsible. In the long term, the short term profits captured by these egregious price hikes will inevitably be tempered by the spotlight on these companies and the inevitable entrance of competitors as result.  

    I'm not a public policy expect or an ethics professor so I don't feel qualified to comment on the morality of these massive price hikes on life saving drugs. However, I do believe that there are adverse business consequences for companies that are perceived to be immoral actors, especially in the healthcare industry. See what happened to both the CEO of Turing and Valeant - they got canned! These adverse consequences include scrutiny from Congress (whether it's grandstanding, I'll let you decide) and government bodies like the FDA. Thus there is business rationale to exercising some degree of restraint when it comes to price increases. Also there is a necessity to justify such pricing actions proactively rather then reactively as opposed to the common response of, "because we could".  Restraint, morality and reasonable, justifiable increases make for long term success and keep companies from being demonized as villains.

    I am convinced that these lopsided economics for drug companies cannot last- which is part of the rationale for implementing more reasonable price increases. The spotlight that egregious price hikes attract also attract innovation and competition that is ultimately bad for the long term pricing power.  In the case of Daraprim, CVS found a workaround to create its own generic version that doesn't need FDA approval. Something similar will happen with the EpiPen, although it may take a bit of time.


    Afternote: After writing this post, I saw that Mylan announced that it would introduce a generic version of the EpiPen (the same exact product without EpiPen written on it) at half the cost - $300 - within a few weeks.  From a pricing strategy standpoint, this seems like a smart move. It quells some of the furor around the price hikes, while also discouraging some potential new competitors from entering the market. While it does cannibalize their existing branded product, Mylan is essentially cannibalizing itself before a new competitor does. Furthermore, it severely debilitates new entrants in the market by lowering to a price point where consumers/doctors might be willing to pay a smaller premium to buy the solution that they know and trust.  Would you rather buy a life saving injector from a new company for $200 or a life saving injector from the company that makes the trusted EpiPen for $300?

    After afternote: After having some technical difficulties with getting this post live, Imprimis Pharmaceuticals (a compounder - different from a traditional pharmaceutical company) announced it was launching a $100 version of the Epipen. Also the US House of Representatives Oversight Committee launched an investigation into Mylan over price gouging.

    Reading Keeping Me Smart - 5/18/16

    by Scuttlebutt Investor

    Some reading piquing my interest this week:

    Mental Models on My Mind

    by Scuttlebutt Investor

    A few months have passed since I received my copy of Poor Charlie's Almanack in the mail so this post has been marinating in my mind for some time.  But I didn't really have the chance to crack open the book and dive into it until recently.  Poor Charlie's Almanack is a compilation of Charlie Munger's (Buffett's right hand man) speeches, lectures and other commentary all rolled up into one absurdly heavy and awkwardly large coffee table book.  The book reads like a stream of consciousness of Munger, which is wonderful if you're like me and eat that stuff up.  The book is such an easy and simple read that the brilliance of Munger is easy to miss if you aren't paying attention.  If the book piques your interest, the best place to get it is here, as it doesn't seem to be readily available through most normal channels.  

    Charlie Munger is the Pippen to Warren Buffett's Jordan or the Thompson to his Curry (as I try to contemporize myself a bit).  I'm sure Warren Buffett would have been very successful without Munger, but I don't know if he would have been as successful.  Buffett often credits Munger with moving him away from the cigarette butt style investing (buying very cheap stocks that have one last puff left in them) he was taught by Benjamin Graham.  

    Munger espouses an approach to investing (and life) that is rooted in multiple mental models.  These mental models consist of theories, tools and general principles from many different disciplines (economics, biology, law, mathematics, physics, philosophy, etc.) that seek to explain more or less how the world works.  The idea is that all of these mental models are interconnected in a meaningful way and to become wise in investing (and life), it is important to make and understand these connections. Further, wisdom is the ability to understand a principle from one domain and apply or extend its learnings to another domain.  Munger refers to the entire group of models as a "latticework of mental models" to highlight their interconnectedness.  

    Munger in his own words from a commencement speech at USC in 1994:

    "You've got to have mental models in your head.  And you've got to array your experience-both vicarious and direct- on this latticework of models.  You may have noticed students who just try to remember and pound back what is remembered.  Well, they fail in school and they fail in life.  You've got to hang experience on a latticework of models in your head.  

    What are the models? Well, the first rule is that you've got to have multiple models- because if you have just one or two that you're using, the nature of human psychology is that you'll torture reality so it fits your models, or at least you'll think it does.  You become the equivalent of a chiropractor, who, of course is the great boob in medicine.  It's like the old saying, 'To the man with only a hammer, every problem looks like a nail.'" 

    In the book, Munger describes many of the mental models that guide his life and investing. Many of these are also discussed eloquently on the Farnam Street blog.  I can't profess to know all of Munger's mental models, but there are several models that I have leveraged many a time in my investing pursuits. Most of these tend to reside in the realm of economics and business.  At first blush, this seems contradictory to Munger's entire theory that a multi disciplinary approach is required, but it's not. I am just covering some models that have benefitted my investing over the years.  Inevitably I have unconsciously mixed these models with others from alternate disciplines.  These models have enabled me to infer what's not being said, ignore the talking heads and maintain my conviction in the face of uncertainty.


    The Power of Secular Shifts

    I don't know if this is a mental model that Munger would officially endorse, but secular shifts are an important phenomenon that are sometimes under-appreciated and other times over-appreciated by market participants.  Secular shifts are a change in volume, demand or relevance driven by secular or permanent forces.  This is as compared to a cyclical shift or change that is driven by temporary forces like supply and demand.  Secular shifts have the power to put entire companies out of business if they are no longer relevant (think Kodak with the advent of digital photography).  Secular shifts pose two interesting challenges for long term investing. First, it is difficult to determine if a shift is driven by secular or cyclical forces in the early days.  Second, it is difficult to determine when a secular shift is slowing down and reaching some point of equilibrium.  The reality is that secular shifts can go on for long periods of time without slowing down all the while market prognosticators predict their imminent doom. Often times, companies benefitting from a secular shift can defy gravity and logic because they are able to thrive despite a declining or volatile operating environment.   

    The examples of secular shifts impacting companies and industries are numerous, but let's discuss two- Google (GOOG) and coal - not in any way related.  

    Google has managed to grow its revenue by double digits nearly every year for the last decade. Mind you, this is amidst one of largest recessions that the US has ever seen.  Even while total advertising revenues were declining by double digits in 2009, Google's revenues increased by some 9%. How could this be?  Well, online advertising and specifically search advertising was gaining share from traditional advertising mediums (newspapers, yellow pages, radio, etc.).  In other words, Google's slice of the pie was growing even while the total advertising pie was shrinking or staying flat.  This shift is still in play today as digital advertising continues to take share from traditional advertising, hence the success of Facebook in growing its revenue and earnings stream over the last several years. I expect this phenomenon to continue to play out over several more years as marketers seek the superior targeting and higher ROIs that these digital mediums can provide, especially as consumers shift their media consumption habits.  

    Coal is a very different story. The demand for coal as an energy source is also experiencing a secular shift, although this shift has been in a direction opposite that of Google.  Coal fired power generation is declining in the US primarily because it is environmentally unfriendly and recent regulatory policies make the economics less compelling.  Net net this is probably good for society in the long term although it’s not so good for coal mining jobs in the short term.  Demand for coal is in a secular decline and will continue to decline for many years.  This is a secular shift to the downside and will continue to impact the coal industry and other industries that benefit from coal - like the railroads. Although developing countries will continue to drive some demand for coal in the short term, these countries too will eventually learn that the long term detriment to society and the environment isn't worth a few jobs and cheaper energy.  

    Secular shifts are a powerful phenomenon but they sometimes make it difficult to differentiate skill from luck (even though management will still often take the credit). Sometimes a company may grow because it is lucky enough to be part of a category or segment that is experiencing a secular shift.  After all, a rising tide lifts all boats.


    The Inevitability of Reversion to the Mean

    The basic premise of reversion to the mean is that nothing can go on endlessly - whether in a positive or negative direction.  Eventually there are factors that act to stop or decelerate growth or decline to some normal or equilibrium level. To put it concisely: Extreme outcomes tend to be followed by more moderate ones. For example, a company that is growing at 20% per year cannot grow at that rate forever.  If it did, it would eventually become bigger than the entire economy.  Eventually, the growth of a given company slows down and begins to resemble the industry average.  The more interesting question though is how or why this happens and how companies can work to delay (stopping it is more challenging) the inevitable reversion.  

    On the former statement, growth or margin expansion slows down or speeds up due to a few key reasons:

    1. High growth or high margins (success in general) invite competition and imitation until those higher margins are competed away to equilibrium.  Other companies realize that high margins and juicy profits are being had and decide to compete because they want a piece of the action.  Similarly, declines are often stemmed because competitors exit a given product line or industry given low profits or margins.  As competition and supply is rationalized, pricing power becomes more normalized and declines slow down or turn into growth.  

    2. Law of large numbers.  As companies get bigger, it becomes harder for them to grow at the same rate because each percentage point of growth becomes worth more on an absolute basis and thus more difficult to attain.  On the decline, the inverse is true - each percentage point of decline is worth less on an absolute basis.  But, at the same time, each absolute dollar becomes more difficult for competitors to steal away as a company acts to defend itself

    An example of reversion to the mean might help articulate the point even better:

    I am a big NBA basketball fan.  If you haven't been living under a rock for the past year, you know that Steph Curry and the Golden State Warriors have redefined the sport of basketball in the most recent regular season. They have pretty much destroyed every NBA record that matters by playing a different style of basketball and could go down as the greatest team in the history of the NBA.  Some of the records they set this season: 

    • Best regular season record of all time with 73 wins and 9 losses surpassing the 1995-1996 Chicago Bulls at 72 and 10
    • Best start to a season in the NBA at 24-0 surpassing 15-0 set by the Houston Rockets and Washington Capitals and the best start in professional sports surpassing the 1884 St. Louis Maroons in baseball
    • Best regular season home game win streak of 54 games surpassing the 1995-1996 Chicago Bulls
    • Most three pointers in a season at 402 by Steph Curry, surpassing his own previous record at 286 (side note: that is an absurdly huge margin by which to beat a record)
    • Most three pointers by a team in a season at 1,077 surpassing the 2014-2015 Houston Rockets at 933

    As much as I admire the Warriors and the three point master that is Steph Curry, I do not expect next season to be as good as this season. The Warriors are an extremely skilled team and they actually improved from their incredible 2014-2015 championship run to the 2015-2016 season that just ended. I am sure they will be very good next season, but they are unlikely to be as good.  

    In his book Fooled by Randomness, Nassim Taleb says, "A result is that in real life, the larger the deviation from the norm, the larger the probability of it coming from luck rather than skills."  

    In the case of the Warriors, their record setting regular season record was a large deviation from the norm and next season is more than likely to revert to the norm (note that for the Warriors, the norm is still likely to be a great season).  First, no matter how skilled the Warriors are, part of the stupendous result this season was inevitably driven by luck and luck doesn't come with consistency. Second, there will be several countervailing forces that will act to normalize their success next season. The latest example is that of Luke Walton recruited to become head coach for the Lakers next season. We could debate how much of the team's success can be attributed to Walton, but inevitably he added something that contributed to that record (especially during the first 43 games where he was on the sidelines as coach in place of Steve Kerr).  Other talented players (bench or otherwise) that also contributed to that record in some way will undoubtedly be recruited by other teams. And finally, other basketball teams will learn to imitate the successful playing style of the Warriors - a faster pace and more three-point shooting. All of these factors will work in concert to normalize the Warriors success next season.  Don't get me wrong; I think they will be very successful, but just not as successful as this season.  

    The interesting thing is that most observers are accustomed to relying on historical results as a way to predict future results. But when historical results are outliers, it is more likely that future results will revert to the mean. This applies to basketball team performance as much as returns on capital for companies.  

    Taleb goes on to say, "Consider that two average-sized parents (dogs) produce a large litter.  The largest dogs, if they diverge too much from the average, will tend to produce offspring of smaller size than themselves, and vice versa. This "reversion" for the large outliers is what has been observed in history and explained as regression to the mean."   

    Reversion to the mean in performance applies to pretty much any sport and in many different ways.  With the Warriors I chose to highlight the track record of the entire team over an entire season, where the sample size is quite large – 82 games. It could be argued that the impact of luck is lessened as the sample size is increased and this is true.  The impact of luck and the lesson of reversion to the mean is much more obvious when analyzing one players’ performance in a specific game. For example, in February of 2016, Steph Curry made twelve three-pointers in one game to tie the NBA record for the most three pointers in a game.  He also played a few other games during the season where he put up ten or eleven three pointers in a single game. Extrapolating Curry’s performance from these few games would be dangerous as the data is not likely to be representative of Curry’s performance over the entire season.  In fact, Curry’s average number of 3 pointers per game over the entire 2015-2016 season was much lower than 12 at 5.1 (mind you – still incredibly high). Over time, extreme outcomes are dampened by more moderate outcomes.  While Curry’s high three point performance in one game was an extreme outcome driven partly by incredible skill and partly by luck, his three-point average over an entire season is much more representative of his expected performance in any given game.

    Many industry forecasts and projections often don't factor in reversion to the mean but it is very real. Many analysts are guilty of projecting out the good times or bad times into perpetuity. But this is almost never the case.  Extreme performance tends to moderate over time as external and internal forces act to normalize it.  Having said that, there are ways to extend the good times (high growth, high returns on capital and/or margin expansion) or delay reversion to the mean. The first way is if said company is benefitting from the first mental model - a secular shift.  When a secular shift is at play, the good times can often go on for a long time and even seem to defy gravity when not viewed in context of the losing party in the industry.  Most of the time, that growth is not coming out of thin air, but is a result of share gains from another category competitor (direct or indirect).  The second way to delay reversion to the mean is by leveraging a durable competitive advantage, which leads me to the next mental model.  


    Building Durable Competitive Advantage

    Buffett talks about competitive advantage often in his writing and also refers to it as a moat similar to the body of water around a castle that affords it protection from raiders.

    In the 2007 Berkshire Hathaway annual report, Buffett wrote:

    "A truly great business must have an enduring “moat” that protects excellent returns on invested capital. The dynamics of capitalism guarantee that competitors will repeatedly assault any business “castle” that is earning high returns."

    Competitive advantages are characteristics of a business or business model that help keep competitors at bay or drive pricing power. These characteristics can exist in the form of: 

    • High barriers to entry for new competitors: Think of the US freight railroads like Union Pacific (UNP) or Norfolk Southern (NSC). It would be nearly impossible to build a new railroad competitor in the US given the massive amount of land and permits required.
    • Patents that grant a legal monopoly: Think of pharmaceutical drugs like Pfizer's (PFE) Viagara (now off patent) where a patent prevents competitors from selling the same drug for a set period of time.
    • High switching costs for customers: Think of a new software competitor to Microsoft Excel (MSFT) where new training would be required.  Businesses and customers would be reluctant to switch lest they have to retrain themselves and suffer a loss in productivity
    •  Strong network effects (winner take all being the holy grail): Think of switching to a new social network from Facebook (FB). The new social network probably wouldn't provide much value unless all your friends and family also switched.
    • Strong brand name: Think of a competitor to Mondelez's (MDLZ) Oreo cookies, the iconic creme filled chocolate cookie.  Sure, a generic cookie can probably grab some market share but likely only at a discounted price.
    • Strong technological advantage: Think of a new electric car competing with Tesla (TSLA). Tesla has a big head start on state of the art battery technology and electric car design that even the large auto companies with significantly more resources have struggled to produce a worthy competitor

    All of these competitive advantages help prevent competitors from entering a given business or help an existing business defend itself when competitors do enter. Merely possessing a competitive advantage, however, is not enough. The goal is to find businesses that demonstrate a durable or sustainable competitive advantage. Can the competitive advantages driving excess returns today continue to drive excess returns well into the future? Ultimately, most competitive advantages deteriorate or wear down over time so the key is finding companies that are investing to continuously widen that moat whether it is investing in R&D to develop better technology, investing in marketing and brand building to drive consumer willingness to pay for a brand or adding features to a product to further embed it into peoples' lives.  The Ensemble Capital team explores moats and competitive advantage a bit deeper on their wonderful Intrinsic Investing blog.  

    The canon of work on mental models is broad and deep and I have barely even splintered the surface of it, which is why I would still direct you to some of the other wonderful blogs I have peppered throughout this post. I should add that the three mental models covered here are merely those that have impacted my investing most meaningfully but the models certainly don’t end there. Some of the other mental models that I would give a nod to are Confirmation Bias, Circle of Competence, Operating Leverage, Diminishing Marginal Returns, Second Level Thinking, Tragedy of the Commons and Margin of Safety.

    Thinking through the models discussed here, there is an interesting takeaway – these models are inherently conflicting. I’ve said that reversion to the mean is inevitable for every business...that is unless a business builds durable competitive advantages or benefits from a secular shift. So there are forces that push a business in one direction and conflicting forces that push it in the opposite direction. To truly understand the fates and fortunes of a business, one must understand how these models interact and compete.  Here I go back to the expert:

    "More commonly, the forces coming out of these one hundred models are conflicting to some extent. And you get huge, miserable trade-offs. But if you can’t think in terms of tradeoffs and recognize tradeoffs in what you’re dealing with, you’re a horse’s patoot. You clearly are a danger to the rest of the people when serious thinking is being done. You have to recognize how these things combine."

    Charlie Munger, Outstanding Investor Digest, December 29, 1997