Not-So Efficient Market Hypothesis


The Efficient Market Hypothesis (EMH) is an investment theory that states it is impossible to “beat the market” based on skillful selection of stocks.  There are three variants of the hypothesis: “weak”, “semi-strong”, and “strong”.  The weak form claims that all prices already reflect all past publicly available information.  The semi-strong claims that all prices reflect all publicly available and instantly change to reflect new information.  The strong form claims that all prices instantly reflect even hidden ‘insider’ information.  Here at Selective we believe that all three variants, including the weak form, overstate the efficiencies of the market.

This is no small claim.

Belief in the EMH has had an outsized impact on financial markets as the popularity of passive investment strategies, such as ETFs and Index Funds, has exploded over the last three decades.  If the EMH is true, then Selective cannot skillfully identify stocks that are mispriced and add value for clients.  We obviously disagree with this sentiment or we wouldn’t be in business.  In this Selective Insight we’re going to build a case against the EMH.  It is our belief that markets are nearly efficient and, for this reason, the EMH should be restated to read “MOST stock prices reflect all relevant information.”  This is a powerful change that opens the door to superior investment returns based on skillful stock selection over long periods of time.

Changing “ALL” to “MOST” opens the door for

Selective Opportunities.

Our Argument

The stock market is an auction place where thousands of businesses go on sale.  We have observed over the years that this auction place is competitive, and the majority of businesses are reasonably priced at any given time.  However, on occasion we believe that we have observed mis-pricings.  These perceived mis-pricings fall on a spectrum between two categories:

1)      Objective Mis-Pricings. Objective mis-pricings arise when the same exact asset is selling at two substantially different prices.  This type of mis-pricing is far less common due to well known arbitrage techniques that are designed to eliminate and profit from such mis-pricings.

2)      Subjective Mis-Pricings.  Subjective mis-pricings arise when the perceived value of an asset is different from the price.  This generally stems from differences in opinion between various market participants.  We call these mis-pricings subjective because each investor has their own opinion about the value of an asset and these opinions are subjective.

The Spectrum

                 Mis-priced investments don’t necessarily fall neatly into one of the two categories, but rather, appear on a spectrum somewhere between the two.  A strong objective mis-pricing would occur when the same exact security is trading in the same place, at the same time, at different prices.  This very rarely happens today.  Sometimes you may see slight differences when one asset sells in two locations (such as the NASDAQ and Toronto Exchanges), but costs to arbitrage the opportunity are larger than the price differential.  Strong objective mis-pricings are quite rare, since they are nearly a sure fire way to earn a profit.  Unlike objective mis-pricings, it’s almost impossible to definitely state when a subjective mis-pricing has occurred.  At Selective we believe in subjective mis-pricings and believe they occur when two market participants perceive very different values for an asset when analyzing the same data.

Purely Subjective

Gray Area In Between

Purely Objective

Example of an Objective Mis-Pricing

We’d like to begin our case with objective mis-pricings.  This is a logical starting point since they are more clear and if a perfectly objective mis-pricing were discovered it would clearly debunk the EMH and “All” would have to be changed to “Most”.  Unfortunately, over the last 10 years as a professional investor I’ve never discovered a perfectly objective mis-pricing (although I’ve heard stories of them).  The closest instance I found to an objective mis-pricing was when I discovered the same exact company selling for two different prices, at the same location, at the same time, but with nuance.  While this example might not convince strict adherents to EMH, I personally find it a curious case.

In late 2014 I found a small business called National Research Corporation, or NRC, that was publicly traded with two different share classes.  These two share classes, while representing the same exact business, had different economic and voting rights.  In our 2014 research report on the company we detailed the unique opportunity:

Prior to May 2013 NRC had one share class with equal rights in terms of voting power and economic interest in the earnings of the company.  In May 2013 the business announced a recapitalization plan that would create two classes of stock: Class A and Class B.  For each share held prior to the recapitalization a shareholder would receive three shares of Class A stock and ½ of a share of Class B stock.  One Class A share is entitled to 1/6th of the earning power and 1/100th of the voting rights when compared to one Class B share.  There was, however, a unique catch – in the event that the business was acquired, both shares would receive identical proceeds per share.  The economics of each share class are summarized below:

Table: 1 Share Class Summary
Per Share Entire Share Class
Class Earnings Voting
Total Shares Economic
Class A 1/6th 1/100th 1 20,955,678 49.9% 5.6% 85.7%
Class B 1 1 1 3,505,352 50.1% 94.4% 14.3%

Table 1 shows that the Class A shares are entitled to 49.9% of the overall earnings of the business, commanded 5.6% of the vote on all shareholder matters, and would receive 85.7% of the purchase price in the event that the business was acquired.  Class B is entitled to 50.1% of the overall earnings of the business, commands 94.4% of the voting power, and would receive 14.3% of the purchase price in the event the business was acquired.

Under this scenario it would make sense that the aggregate of the 20,955,678 Class A shares should trade at a slight discount to the aggregate of the 3,505,352 Class B shares.  The logic behind the discount is that the Class B shares receive more income and command almost the entire vote on all shareholder matters.  The takeover provision slightly complicates matters, because the Class A stock would prefer to be acquired due to the 1 for 1 payout the share class would receive, however, with no meaningful ability to vote this desire would likely not materialize.

It is very apparent that the market does not understand the economics of these two share classes and has badly mispriced the two.  At our time of purchase the Class A shares were trading at $15.00 per share for an aggregate value of $314 million.  This share class is entitled to 49.9% of the total earnings of the business or roughly $10 million dollars; therefore trading at a P/E ratio of 31.  The Class B shares were trading for $32.14 per share for an aggregate value of $112 million.  This share class is entitled to 50.1% of the total earnings of the business or roughly $10 million dollars; therefore trading at a P/E ratio of 11.  The Class B shares are roughly 1/3rd of the price of Class A shares and command 19x the voting power.  We bought a sizable portion of Class B shares only.  It should be noted that Michael Hays, the majority shareholder, has the overwhelming majority of his net worth in Class B shares.  Our purchase price of 11x earnings compares very favorably to the average P/E ratio of 20x earnings for an S&P 500 constituent.

This was a very unique case where the same exact business was trading at two very different prices.  It was not purely objective, due to the takeover provision, but was close.  However, it seemed obvious to us that the takeover provision could be ignored and this was nearly an objective mis-pricing (Yes, we realize this statement is a subjective opinion, hence a nearly-objective mis-pricing).  Why could it be ignored?  Michael Hays, the Founder and CEO, had the majority of his net worth in Class B shares and would instantly lose millions of dollars if a takeover occurred – almost irrespective of price.  Due to the voting rights, Mr. Hays held the overwhelming majority of shareholder votes and could block any takeover attempt or sale of the business.  We believed it was unlikely he would vote for an action that lost him millions of dollars and therefore, ignored the takeover risk.  After accumulating a position we wrote a thoughtful letter to the Board proposing to eliminate the two-class structure.  The letter can be found here.

Roughly two and a half years later management adopted a solution that was akin to our original proposal and eliminated the two-class structure.  It should be noted that from the time the arbitrage opportunity was first publicly detailed until final resolution it was nearly 5 years.  The resolution occurred on April 17, 2018 when a recapitalization occurred whereby the Company exchanged one share of its then existing class A common stock, plus $19.59 in cash, for each share of class B stock.

Surely an interesting case stated for an EMH adherent.

Subjective Mis-Pricings

Subjective mis-pricings arise when two market participants look at the same set of data and come to different conclusions on how that data should be interpreted.  We believe this is the most common type of mis-pricing that occurs.  However, we quickly concede the fact that it is nearly impossible to discern whose subjective opinion was correct – even after ample time has passed.  Below we present two examples of how market participants may draw different conclusions from the same set of data.  Both of these instances occurred recently where our opinion here at Selective was different than the general market consensus at the time.

Company Event Market Reaction Selective Outcome
Cognizant Technologies 2016 FCPA Bribery Scandal
September 30, 2016
Fearful of short-term impacts on business operations and potential fines levied against the company.¹

The market capitalization fell 13.3% or $4.4 billion dollars in one day.
Unsure of short-term impacts on business, but assumed fines would be less than the drop-in market value and violation would likely not impact business over a 10-year period The company has posted 6 consecutive quarters of record revenue and profits. Total costs thus far are $63 million or 1.5% of the inital $4.4 billion market decline
Facebook 2018 Cambridge Analytica Scandal
March 16-30, 2018
Fearful of potential shifts in regulation and changes in user behavior. ¹

The market capitalization fell 17.8% or $95 billion dollars in two weeks.
Unsure of short-term impacts on business, but believed the long-term prospects remained unchanged. Net account closures would be immaterial relative to the size of the market capitalization drop. Too soon to tell, but record profits and record user numbers in Q2 2018.

Note: The above situations were opinions expressed by Selective Wealth at the time of event.

Only time will tell which opinion about the future performance of the business is correct.  Even after sufficient time has passed, it will still be unclear whose subjective opinion was ‘correct’.  When it comes to investing, the movement of a stock price can make you appear right when you were wrong and vice versa.  Despite the difficulties, interesting case studies have developed overtime where data seems to suggest that a particular individual, or group of individuals, are more accurate when assessing the subjective value of investments.  We believe one of the most interesting case studies was published by Warren Buffet in a wonderful piece exploring this same topic titled, “The Super Investors of Graham and Doddsville” The piece can be found here.

Some may read the Super Investors article and be unconvinced, because the sample set is too small; or the data is old (investment track records were from decades ago, when the market was less efficient).  These are fair points.  When it comes to ‘proving’ there are subjective mis-pricings it is nearly impossible, but if they do exist we should be able to find instances of skillful investors taking advantage of such opportunities.

At Selective, we are optimistic that we can take advantage of these subjective market inefficiencies, but we don’t want to blindly believe we can add value at the expense of our clients.  To guard against unfounded optimism we try to measure outcomes carefully.  After each purchase we make we publish a detailed research report that outlines our expectations for the business operations of any newly acquired Selective Company.  These expectations often are contrarian to market opinion and take advantage of, what we believe, are subjective mis-pricings.  As time passes we provide our clients with actual business results – so they can be compared to our original thesis.  We also, internally, keep track of each investment’s performance relative to the S&P 500.  As indicated in the chart below, from the time the Selective Opportunity Fund was first launched in February 2017 to the end of Q2 2018, we’ve purchased 11 businesses.  The relative performance of each is compared to the S&P 500 below:

Selective Company Symbol Purchase Date Purchase Price Dividends June 30, 2018 Price Gain (Loss) ³ S&P 500 ³ Relative % Allocation ³
MuleSoft Inc. MULE 07/28/2017 $20.90 - $43.93 +110.2% +6.9% +103.3% 0.00%
Syntel SYNT 03/29/2017 $17.19 - $32.09 +86.7% +15.1% +71.5% 10.84%
Cognizant Technologies CTSH 01/31/2017 $52.16 $0.60 $78.99 +51.4% +19.3% +32.2% 2.19%
Ultimate Software Group ULTI 09/27/2017 $183.51 - $257.31 +40.2% +8.4% +31.8% 3.99%
Facebook FB 03/27/2017 $154.00 - $194.32 +26.2% +4.0% +22.1% 4.75%
Baidu BIDU 01/31/2017 $175.88 - $243.00 +38.2% +19.3% +18.9% 9.46%
Shutterstock SSTK 07/28/2017 $38.18 - $47.46 +24.3% +10.0% +14.3% 4.24%
Union Pacific UNP 01/31/2017 $108.55 $2.66 $141.68 +30.5% +19.3% +11.2% 2.02%
Redknee Solutions RKN 06/05/2017 $28.00 - $34.20 +22.1% +11.6% +10.6% 3.22%
Alphabet GOOG 01/31/2017 $869.21 - $1,115.65 +28.4% +19.3% +9.1% 9.36%
Bank of the Ozarks OZRK 05/22/2017 $44.20 $0.37 $45.04 +1.9% +13.5% (11.6%) 6.64%

The above data has been compiled and measured over too short a time interval to draw any meaningful conclusions, but it is the precursor to something that could be statistically interesting.  People often like to say, “Only time will tell.”  Oddly enough, when it comes to investing, time doesn’t tell.  What if we continue with a similar track record for decades into the future?  It would still be very difficult to separate luck from skill.  When it comes to subjective mis-pricings, it is probably impossible to convince a staunch EMH adherent that they exist, but it seems logical to us to consider statically meaningful data.

Note: We are not implying that the short-term results of the Selective Opportunity Fund are statistically meaningful. Past performance does not guarantee future results.



3 As of June 30,2018. Mulesoft was sold during Q1 2018 due to the announced acquisition by Salesforce.

Disclaimers & Disclosures

This document has been prepared and issued by Selective Wealth Management (SWM) on the basis of publicly available information, internally developed data and other sources believed to be reliable. The information contained herein is not guaranteed, does not claim to be comprehensive and is strictly for informational purposes only. SWM does not assume any liability for any direct, indirect or consequential loss that may result from the reliance by any person upon any such information or opinions. Any expressions of opinions are subject to change without notice. The information herein should not be construed as a recommendation to purchase or sell any particular security or an assurance that any particular security held in a portfolio will remain in the portfolio or that a previously held security will not be repurchased. It should not be assumed that any of the security transactions or holdings discussed herein have been or will equal or exceed the investment performance of the securities discussed. The securities discussed herein are for illustrative purposes only and do not constitute recommendations.

As of 6/30/18, National Research Corporation comprised 0.00% of the Selective Opportunity Fund’s portfolio.

Before investing, consider the fund's investment objectives, risks, and charges and expenses.  The prospectus contains this and other information about the fund, and it should be read carefully before investing.  To obtain a prospectus, contact Selective at  The Fund is distributed by Ultimus Fund Distributors, LLC.

Investing involves risk, including loss of principal.  Investing in “value” stocks presents the risk that the stocks may never reach what the Adviser believes are their full market values, either because the market fails to recognize what the Adviser considers to be the companies’ true business values or because the Adviser misjudged those values.  Because the Fund is non-diversified and intends to focus its investments in a relatively small number of companies, an increase or decrease in the value of a single security likely will have a greater impact on the Fund's net asset value and total return than a diversified fund.

The performance quoted is past performance and does not guarantee future results.  The investment return and principal value of an investment in the Fund will fluctuate so that an investor's shares, when redeemed, may be worth more or less than their original cost.  Current performance of the Fund may be lower or higher than the performance quoted.  Performance data current to the most recent month end may be obtained by calling 434-515-1517.  The Fund's gross expense ratio is 1.52% for the Foundation Class and 1.72% for the Service Class.