Deep Value Investing: Tobias Carlisle

Gyan on Treadmill dated 04-Apr-2018

Tobias Carlisle has written the book Acquirer's multiple, another approach to value investing. I have this book in Kindle and will review it sometime later.

The Google Talk is about 'Deep Value Investing', how to find value stocks that can give you surprising outperformance in the long run.

This is a matter of fact presentation, full of surprising insights. The ideas made in this speech upends many of your conventional investing ideas. !!!

This is a number driven presentation, I will try to cull out the key points in this post.

You can watch the presentation here

The key idea in this presentation is an unusual one. It is that investing in a portfolio losing stocks with pathetic financials and high losses can sometimes lead to phenomenal portfolio returns. At the outset, Carlisle gives a disclaimer that this approach is very risky. There is 6% chance that the stocks will become zero. However, a portfolio will ensure that what remains gives huge returns.

Should you pick and choose from the basket of pathetic stocks? The answer is no. The reco is to invest as per the model. 

What explains the portfolio out-performance. Simple answer. Mean Reversion. In the long run, the performance of all the stocks will tend to revert to mean. Since these pathetic stocks are beaten down so much, they are so far away from the mean that any mean reversion will ensure tremendous out-performance.

The next question is when will mean reversion happen and what causes it. 

Regarding first question, no one knows when it will happen. However, if there is a significant difference between price and intrinsic value of the company, the price will gradually move towards the intrinsic value. Some of the causes of mean reversion, according to Graham, are:

1. Creation of an earnings power commensurate with company's assets, in other words improving Return on Assets. This can happen due to general improvement in Industry or due to favorable changes in company's operating policies, with or without a change in management. 

2. A sale or merger (aka. special situations)

3. Complete or partial liquidation

Read the next two paragraphs carefully. This will blow your mind.

Carlisle explains that he and his team divided the universe of stocks into three groups. Large Cap, Midcap and Small Cap. They further divided each of the above to three subgroups, high growth, moderate growth and low growth. Then they did a historical trend analysis (so called 'Back Test') of portfolio returns. Over 5 year and 10 year, the stocks in the low market cap, low growth portfolio, consistently outperformed all the rest.

Wait there is further. Among the low cap, low growth stocks, portfolio of stocks that was in loss gave better returns that the portfolio of stocks that made profit. Among the latter (Portfolio of Low Market Cap, Low Growth Profit making stocks), the stocks that gave dividends did poorer than those that did not give dividend !!

Amazing !!

Does it mean that the growth rate of low growth undervalued portfolio spurted and overtook that of the high growth undervalued portfolio? No. What happened is that mean reversion lifted the market value of the undervalued portfolio to normal valuation. I saw this phenomenon play out in Indian markets about 4-5 years ago when HPCL was available at a price of 190 and a PE of 1 !!. Stupid me, did not buy it then. From then it has become a 10 bagger.

I was reminded of parallels in India market. In 2007, if you had invested Rs.10 Lakhs (a million) equally in a portfolio of 25 such stocks, one of them would have been Avanti Feeds (you would have got about 6000 shares). It has given stratospheric returns since. Indo count in 2009, Mayur Uniquoters in 2011 all are testaments of this approach.

So that is lesson 1: Portfolio of Low Growth, Low Market Cap stocks will outperform the high growth,  high market value companies.

The author also talks about the 'Net net approach' popularized by Ben Graham in his book Intelligent Investor.  I have mentioned about my experience with this approach in this blogpost.

Net-net companies are those where the weighted net current assets (different weights for different current assets based on their liquidity risks) is greater than the market capitalization of the company. Even deeper value is if the weighted net current assets minus Cash and Cash Equivalents is greater than the market cap of the company. Effectively you are getting a company with fixed assets and cash free of cost.

One key point made is about the need for a model. Studies after studies in different areas including medicine and psychology show the need for and importance of a model. Surprising finding is that a model, even a simple model, will outperform the arbitrary decisions taken by experts even!!! Your model could be just the PE screen. But even that will tend to outperform the complex models chosen after deep analysis by experts. The reason that simple model outperforms is that people tend to make mistake while trying to outperforming a model.

Golden rule? Simple models outperform experts, even when experts have access to the results of the simple model.

If that is not a call to arms by the retail investors, I don't know what is.

Lesson 2: Models are required. And simple models outperform complex models and expert discretions.

As mentioned above, Carlisle has written a book called Acquirer's Multiple. What is that? 

This is one of the simple valuation models propounded by the presenter. He looked at various models including 'Net current assets', 'franchise model (See's Candy)', 'great companies going thru temporary stress (American Express)', 'Magic Formula of Greenblat' etc.

Before we move to Acquirer's multiple, let us look at the Magic Formula propounded by Joel Greenblat. In his book 'Little book that beats the market' and its cousin 'Little book that still beats the market', he uses a combination of ROIC and Earnings Yield to identify companies that are worth investing. (Earnings Yield is calculated on EBIT basis, not on net profit basis to remove the impact of the capital structure on returns). This formula comprehensively beat the S&P 500 when back tested over 20 years.

In their research, Carlisle and team goes one step further. They break up the magic formula into its components, ROIC and Earnings Yield and then check the performance. Surprisingly, they find that while earnings yield alone outperforms even the magic formula, the ROIC pulls down the magic formula returns. 

The reason is the mean reversion of ROIC. 

Carlisle do not explain if he tested a portfolio of high earnings yield and low ROIC. May be that IS the magic formula...

In summary, since earnings yield is the inverse of PE, all we need to do is to buy a company with low PE. May be also low ROIC. Mean reversion will do the rest for us.

Acquirer's multiple, also known as 'Enterprise Multiple' is calculated by the formula:

(Market Capitalization + Debt + Preferred Stock - Cash) / (EBITDA or EBIT)

This is the matric used by acquirer's, PE funds, hedge fund etc to value the business. The numerator is the cost of acquisition and the denominator is the Earnings from that investment. Note that this is a more complex calculation of Earnings Yield than the inverse of PE. PE just look at the market capitalization, while this formula also considers Debt, Preferred Stock and Cash In Hand, which could be a more accurate value of the business to an investor.

Lesson 3: Acquirer's multiple is a simple model that outperforms all the other models.

That is it. That is a lot of wisdom for a day....