Introduction to Quantitative Investing: The Price to Book Factor

Aug. 4, 2023, 12:55 p.m. |Factors |Intermediate

The following is an excerpt from Jin Choi's upcoming book on Quantitative Investing: Building Wealth the Data Driven Way

Value investing is the philosophy of buying securities for less than their ‘intrinsic value’. While there are many ways to describe intrinsic value, I like to define it as the total value you would receive for having purchased a security if you held it until you were compelled to sell it. If the security in question is a bond with a $1,000 principal that pays $50 interest annually until it matures in 10 years, the intrinsic value of the bond is the discounted sum of the $50 interest payments and the $1,000 principal. As another example, if a company is expected to be acquired at $100/share in 2 months, then the intrinsic value of the company’s stock is a little less than $100/share, where the slight discount exists to compensate for the risk that the deal falls through and shareholders don’t receive the promised $100/share in cash. Why value investing works is obvious - if you can correctly identify each security’s intrinsic value, and purchase those securities for cheaper, then you will eventually earn good returns regardless of how prices behave while holding them. Many of the most famous investors, including Warren Buffett and Benjamin Graham, are devotees of the value investing philosophy.

Investor Reading in a Bookstore

Quants have tried to craft a systemic, data-driven method of channeling the power of value investing. The ‘value factor’, which are numerical ratios that hint at whether securities are trading for less than intrinsic value, are the fruit of their efforts. Formulating the value factor, however, is not straightforward. Estimating the intrinsic value of a security is an art that involves many considerations, some of which are hard to assign numbers to (e.g. the strength of a brand). But even with measurable considerations, quants disagree on how to assemble them into ratios that would come closest to approximating intrinsic values. There are many different formulations of the value factor as a result. Fortunately, only a few of them are widely used, so let me explain them in turn.

Perhaps the most widely known value factor is ‘book to price’, which is calculated by taking the book value per share of a company and dividing it by its share price. The book value of a company is found in its financial statements, and it’s equal to the company’s assets minus its liabilities (i.e. what the company owes). You might be tempted to think of book value as the company’s liquidation value - that is, the money shareholders would receive if the company decided to stop operations, sold its assets, and distributed the proceeds. Unfortunately, that wouldn’t be accurate. The value of assets and liabilities are determined using accounting rules, and can be significantly different from the value they would fetch in a liquidation scenario. For example, if a company bought a desk for $1,000, it would record it on the books for $1,000. But if the company tried to sell the desk on Kijiji, it would be lucky to get $800.

The book value of a company is better thought of as the amount it would take to recreate the company. If a company only had a $1,000 desk as its asset and had no liabilities or meaningful operations, it would cost $1,000 to replicate this company. We can therefore conceive a link between book values and intrinsic values, and that stocks trading cheaply relative to their book values are also likely to be trading below their intrinsic values. Historical data also corroborates this view. Though precise numbers differ depending on timeframes and methodologies, stocks with the highest book to price ratios have, over the past several decades, outperformed the market average by a couple of percentage points per year. But there’s a caveat. This factor’s performance dropped off noticeably in the past decade, underperforming the market average by about a percentage point. What gives?

Book value has always been a very loose proxy to a company’s intrinsic value. Take railroads, for example. As of the time of this writing, CN Railway’s book value is about $20 billion, but that doesn’t mean $20 billion would be sufficient to create a company as profitable as CN. Due to accounting rules, CN carries the value of its rail network at the cost to build them, but they built those many years ago. Building the same length of railroads would cost far more today. Furthermore, if someone actually did build that much rail today, they would cause freight rates to plunge and the builder wouldn’t be able to extract the level of profits enjoyed by CN today. Such competitive disincentives, or “moats” as they’re called, allow companies like CN to enjoy great profits and increase their intrinsic values. The book values of such companies therefore greatly underestimate their intrinsic values. Still, despite its flaws, book value had generally held a tight enough relationship with intrinsic value to make it useful. But that began to change with the entrance of technology companies.

To see why, let’s take Meta, the parent company of Facebook and Instagram, as an example. The most valuable assets this company has are its users, its brands, and its software. But users and brands don’t appear on financial statements at all, since their values are hard to measure using rigid rules. As for software, accounting rules have only in recent years allowed software costs to be capitalized - that is, to allow money spent on software to accumulate into assets. Book value therefore greatly underestimated Meta’s intrinsic value, and the same principle held true for other technology companies to varying degrees. Using book to price thus caused investors to miss out on many profitable technology stocks, and underperform the market to such an extent that many started to question whether the factor still works at all. It’s a debate that won’t settle for many years to come.


Leave a comment

Please enter your name!
Please provide a valid email address!
Please enter a comment message!