One central tenet of value investing has always been mean reversion. If you buy something for less than its worth, eventually the market will notice and push the price up. The key question of course is how do you measure ‘worth’. In 1934 when Benjamin Graham wrote Security Analysis, his definition of worth was based on the physical assets held by the business. If the company was selling below its liquidation value, you had a margin of safety. Graham measured liquidation value using the book value or net current asset value (current assets – total liabilities). His thinking was likely influenced by the Great Depression which caused him to be overly cautious.
Buying a basket of companies selling below their net current asset values which have been profitable for the last 5 years is likely to work out well. The key here is a long history of profitability because a company losing money gradually erodes any margin of safety you may have from the assets.
Searching for companies selling below their book value can also be a good strategy especially for banks, utilities or companies that have a lot of land. A company with lots of land on its books carried at cost will understate the book value further increasing the margin of safety.
However, the economy has changed a lot in over the years. Back then, a majority of a stock’s value was tied to its tangible book value. Intangible assets now account for over 80% of the average company’s market value.
Companies had fewer competitive advantages in an asset heavy economy because a competitor could come along and steal market share by simply acquiring a similar asset base. If you were in the business of making plastics and a salesman convinced you to buy a new piece of machinery that would increase production, you would have been very excited. This excitement would be short lived after you realised the salesman managed to convince your competitors to buy the same machine. In a competitive market where the products were undifferentiated, you had no choice but to pass savings back to the customer.
The increasing importance of intangible assets like brands changed everything. Now you had an asset that could not be easily replicated by your competitors. In an asset heavy economy, buying stocks at a discount to book value was a good strategy because mean reversion was likely to return prices back to book value. In an asset light economy, expecting price to book values to revert back to historic averages doesn’t make sense because a company can increase its real economic value without increasing its tangible assets.
The increased importance of intangible assets such as brands, patents, R&D and how they are treated by accountants has led to distortions of traditional value metrics such as low P/E or P/B ratios.
To summarise, two things have changed over the last 60 years:
- Companies now rely more on intangible assets and the link between tangible book value and economic book value has been severed for many companies. In the 1950’s, the S&P 500 consisted of 85% industrials, 10% utilities and 5% rail companies. The three asset heavy sectors which previously accounted for 100% of the index now account for less than 15%. Information technology and healthcare are now the fastest growing and largest sectors in the index accounting for 36% of its value.
- The accounting treatment of intangible assets can distort earnings and book value therefore metrics like the P/E and P/B ratios can often be meaningless in determining the real economic value of a company.
These two points have accelerated the demise of traditional value factors for many sectors of the economy. The first point is simply a fact and can’t be adjusted. The second point about accounting changes allows for adjustments to be made to translate the realities of modern accounting into the real economics of the business
One important accounting principle that has distorted the income statement and balance sheet is the treatment of intangible assets. We will focus on R&D and marketing spend in particular.
Prior to 1975, companies in the USA could capitalise R&D. This meant that investing $100 million into developing a new product or process was not immediately expensed on the income statement. Instead it was capitalised on the balance sheet and amortised over a fixed amount of time. Today, GAAP and IFRS accounting require R&D to be expensed so a $100 million investment impacts the income statement straight away. R&D is treated slightly differently under IFRS, which allows companies to capitalise part of the R&D expense. More about that here.
A brand is also an intangible asset that may not be properly valued on the balance sheet. Coca Cola has spent almost $100 billion on advertising throughout its history. None of this is shown on the balance sheet. However, if Coca Cola was to be acquired at a premium to book value, the acquirer will record this excess as goodwill. This makes no sense because a brand developed internally and the same brand acquired by another company are economically identical. However, accounting principles do not show the value of a brand on the balance sheet if it was developed internally.
How can you make sensible adjustments to intangible assets such as R&D and brands? Firstly, these adjustments can’t be automatically applied to every company. Not all R&D and brands are created equal. Examples of the good, bad and ugly of R&D and brands and how to make adjustments to the financial statements can be found here.
For R&D to be classed as ‘good’ the company needs to have a history of improving its competitive position because that is the ultimate goal of R&D.
For a brand to be classed as ‘good’ it needs to change consumer behaviour. Having a well-known brand is not good enough. Sony is a well-known brand but since they sell products that have been commoditised like TVs, their ability to alter consumer behaviour is limited. You wouldn’t pay much more for a Sony TV compared to an LG, Panasonic or Samsung. The same can’t be said about Apple’s iPhone. Ultimately, a good brand is able to keep old customers and attract new ones without competing on price. The iPhone does not compete on price and its retention rate is over 80%. This is the power of a strong brand.
What is the value of Apple’s brand on the balance sheet? About $6 billion. Interbrand values Apple’s brand at over 180 billion. Valuing a brand is subjective however; we don’t need a precise value to know that the $6 billion stated on the balance sheet significantly understates the value of the Apple brand.
When a company invests in new machinery, this is capitalised on the balance sheet and depreciated over the expected life of the machine. R&D and marketing spend by companies with a history of good R&D investments and good brands should also be capitalised and amortised on the balance sheet. This will remove distortions in the income statement and valuation by increasing the EPS and reducing the P/E ratio.
Mean reversion can sometimes sound like a religion for some investors. There is no law of physics that says one variable must mean revert to its historical trend. In 1958, the dividend yield on stocks in the U.S. market fell below the long-term bond yield for the first time in history. Anyone expecting mean reversion would still be waiting.
Mean reversion needs to be understood in the context of changes in the economy like the larger importance of intangible assets and how current accounting principles can distort true economic value.
Value investing is still very much alive but simply using a low multiple as an indication of value can’t be applied to companies with lots of intangibles. If it was that easy, everyone would be a value investor and excess returns would be arbitraged away.