Some are out of line. But don’t take that as a trading strategy without further investigation; it’s future ROE and growth in ROE…and the risk to ROE… that determines P/B. Current ROE is only a cue to look further; it can be due to one-time effects on earnings. That said, current ROE is typically a good indicator of future ROE.

“Most people in research still like to study PE ratios and dividend yields. However, I am different from them. I like to study PB (Price-to-Book) ratios and ROE (Return on Equity). This makes it easier for me to grasp the essence of things.

For some companies that rely on their business models and have light assets, using the PB and ROE models doesn’t hold much meaning.  Take a high-quality company with a 20% ROE. After four years, this company’s net assets will have doubled.

If it is a leading internet or liquor company, for example, which is a light-asset company with a high PB ratio, the success of its business is not dependent on its net assets. For instance, a company with a market value of 1 trillion RMB and net assets increasing from 100 billion to 200 billion RMB — what would its market value be four years later? It’s hard to predict. Its stock price may remain 1 trillion RMB, maybe less than 1 trillion RMB, or it may exceed 1 trillion RMB.

But for another type of company, one with heavy assets that continuously takes on projects, its business expansion depends on the growth of net assets and quality projects. With a 20% ROE and an initial PB ratio of 1.25, a market value of 1 trillion RMB, and 800 billion RMB in net assets — after four years, the net assets could grow to 1.6 trillion RMB. Of course, 50% of the increase in net assets, or 800 billion RMB, could be distributed as dividends, and 50% could be used for capital expenditures, with part of it replacing capacity and part of it going toward expansion.

In fact, this example of the business model for real estate companies can also apply to upstream resource stocks. After four years, if the PB ratio remains stable, the market value will grow in proportion to the net assets. If no dividends are paid, the net assets will become 1.6 trillion RMB, and the market value will be 2 trillion RMB. If, during this period, 400 billion RMB is distributed as dividends, the net assets will be 1.2 trillion RMB, and the market value will be 1.5 trillion RMB. Over four years, the company would have distributed 400 billion RMB in pre-tax dividends.

After explaining this, everyone can understand why I use the PB and ROE models as my theoretical foundation for analyzing such companies. As long as the business has a monopoly, and the product is universal, the company’s price range can allow for sensitivity analysis on profitability or ROE. When the PB ratio is at a reasonably low level, anywhere below 1 and up to 1.5, I consider it a low PB range, and you can hold the stock, capturing the 20% ROE annually and essentially recreating the company in four years. This way, you won’t be scared by fluctuations. The company’s ability to make money, combined with its valuation, can lead to the company doubling in value every four years. So why not hold onto it and let it double every four years, and in eight years, double again? This business model is suited for this kind of analysis.

In contrast, the previous examples of internet and liquor giants do not apply to this method. Therefore, I don’t care how much the stock price has risen in the past or where it is on the chart. I only focus on the PB level and the ROE’s ability to represent profitability, allowing me to hold the stock perpetually. This investment model allows me to make money in a secure and steady way, without worrying during market adjustments, because I know exactly how I am making money.

I’ve seen those who analyze economic cycles, recent market trends, how to rotate sectors, and how to analyze growth stocks using PE and PEG ratios. They engage in all sorts of stock price speculation, but I find it overwhelming. It’s nowhere near as stable and reliable as my system.

In 2017, I achieved my greatest success with the real estate sector. Now, once again, I’m relying on resource stocks with high ROE and low PB. These companies acquire new projects, replacing their old, depleted ones (like exhausted mines). The difference is that in real estate, projects typically have lower profit margins and rely on leverage to increase ROE. However, in the resource sector, where profit margins are high, even with low leverage, ROE can still be high. Therefore, the risks in the latter are much lower, though the income growth rate is slower.    But if the essence of these companies’ profitability is ROE, and their valuation method is PB, does it really matter if their income growth is slow? Does the world only have one type of performance growth that can increase valuations? Does performance decline always lead to a lower valuation? This only applies to certain companies that follow the PE and PEG model. But the essence of investing is to make money. If my net assets are 800 billion RMB this year and 1 trillion RMB next year, even if my growth rate slows down, the company’s value will increase. Why should the market value fall? It should rise.

So, I understand the theory behind valuing growth stocks, but I don’t believe it can be applied universally. Similarly, the idea that cyclical stocks’ prices should fall when commodities peak is nonsensical to me. When my PB ratio is below 1, and my profits over 2-3 years exceed the market value, why should the stock price drop? Some people say cyclical stocks should be sold when they become cheap, but they don’t realize that theory is based on buying at a high PE and selling at a low PE. They don’t understand the significance of an extremely undervalued PB ratio. Various strange theories about cyclical stocks and growth stock valuations have trapped most people. They fail to understand the most fundamental value theory: year after year, the company’s net assets are increasing, its cash flow is growing, and so its value should rise.

Yet, they insist on applying ridiculous cyclical stock or growth stock valuation theories. These theories aren’t true investment theories, they are stock speculation theories, mere stock price gambling theories, not real value investment theories. Today, I’ve shared all my most valuable knowledge. I believe very few people discuss these things, because everyone’s understanding and experience with investing are different. I feel like I’ve already grasped the essence of things. Many people may not even “get” what I’ve shared today.”

(Translated by Zhi (Oliver) Yan)

E/P = E/B x B/P = ROE x B/P

So a given E/P (P/E) can be any mixture of ROE and B/P, a high ROE with low B/P or low ROE with high B/P, for example.

For the historical average P/E since 1960 of about 16 (E/P = 6.7%) with a P/B of 2.99 (a B/P of 0.334) and an ROE of 20%,

E/P = 6.7% = 20% x 0.334

But an E/P of 6.7% can also be accounted for by

E/P = 6.7% = 5% x 1.34

That is, an ROE of 5% and a P/B of 0.75 (a B/P of 1.34).

Now, as chapter 2 instructs, a high P/B is justified by a high ROE (as in the first calculation here) and a low P/B is justified by a low P/B (the second calculation). The interesting case for a value investor is that with a high ROE and a low P/B. It suggests the market is underpricing the ROE: Why is the P/B so low while the book value is generating a high return on book value? See also the testimonial on a P/B-ROE strategy on this webpage.

There is a Vignette in chapter 2 pointing to some interesting cases. Here are the subsequent returns over one year and two years for the firms exhibited there. Note that for Newell Brands and China Eastern, the second year ahead was the Covid pandemic year.

This is not a completely informed strategy for three reasons.

First, it’s not current ROE that the P/B prices but expected future ROE. So look at the forward ROE but, even then, ROE in later years also matter.

Second P/B is not only based on expected ROE but also the amount of investment that the firm puts in place to earn the ROE; more investment to earn at a high ROE adds value. The two generate growth.

Third, expected ROE needs to be discounted for risk to ROE. That affects the price in P/B. If expected ROE is high but at risk of falling a lot, price should be lower.

These are the additional features to the P/B-ROE strategy that the book brings in as it proceeds.

(This will only work for positive ROE.) P/B is determined fundamentally by what book value will earn in the future, that is, expected ROE, so a low PBROE (low P/B relative to expected ROE) suggests a mispriced stock. But the measure comes with the same warning about a trap as with PEG: P/B can be low because the risk to future ROE is high. Further, P/B is determined by the future ROE stream over many years, not just forward ROE, so growth in ROE must also be incorporated. Tread softly.

At December 31, 2024, the PRROE for the S&P 500 stood at 0.235. Here is measure on the same day in previous years, along with the subsequent year’s return for the index:

The PBROE can also be calculated with trailing ROE. Here are PBROE ratios for selected stocks in January 2025 on that basis:

To give some perspective, a firm with a required return of 10% and an expected ROE of 10% should trade at a P/B of 1.0 (no value added to book value). LGI Homes above with a P/B of 1.09 and an ROE of 10.4% is close to this benchmark.

However, screening can help by filtering out stocks for that further analysis. There are many firms to which we can bring value investing analysis, and sorting out those where there might be payoff to the work involved would be helpful. You can be stimulated to investigate a firm by a number of catalysts: Recognizing the prospects of a business, recognizing good management, recognizing a promising business model, an insightful newspaper report, an analyst’s informed report, an unexplained drop in the stock price, and much more. But the chapter suggests other filters which can be combined with these filters to identify potential investing targets.

One is a low PEG with low indicated risk. The PEG ratio compares the forward P/E to a forecast of earnings growth in the near term. In effect, it challenges P/E with an (informed) growth estimate: If the P/E is low but the growth estimate is high (that is, PEG is low), that could be mispricing. That is a cue to investigate further. As the chapter informs, P/E could be low for a given growth because the expected growth is risky, discounting the price in the P/E ratio. So, a low PEG must be evaluated against the risk involved. Investigate further with a complete value investing analysis.

So, here’s the filter to help identify a promising stock to buy: A low PEG with indications that risk is low. And avoid stocks with high PEG and high perceived risk. That, of course, requires a risk indicator, a topic for later in the book. Chapter 2 refers to beta, not to be ignored but not a fundamental measure. Keep in mind that the integrity of the PEG ratio requires the earnings in the P/E to be sustainable earnings, not affected by one-time items that cannot grow. And a good estimate of future earnings growth is also important. The five-year estimate from analysts’ forecasts are not very accurate.

A similar cue can be given by the P/B ratio: A low P/B with a high forward ROE indicates possible mispricing. But, again, recognizing risk is crucial: P/B should be low if the risk to future ROE is high.

In January 2025, China Sunshine Paper Holdings Company Limited (2002.HK) traded at a P/E of 4.73, a P/B of 0.4, and an ROE of 9.2%. Worth looking into?

The returns to B/P are clear in the period from the early 1960s to 1999. But the strategy has not worked since 2007.

Ignore Information at Your Peril. A real value investor understands choosing an investment based on just one piece of information is dangerous. B/P, despite being called “value,” does not challenge the price because it leaves out information about value that this book directs you to. Return on book value, ROE introduces more information as chapter 2 and the P/B-ROE discussion on this web page show. More so a full residual earnings model introduced in chapter 3 and its application in subsequent chapters.

See Arnott, R., C. Harvey, V. Kalesnik, and J. Linnainmaa. 2021. Reports of value’s death may be greatly exaggerated. Financial Analysts Journal 77 (1), 44-67.

and thus

But, while helpful in demonstrating the point, the model is a bit too simple. It only works for positive earnings, and it assumes full payout of earnings with dividends. However, the P/E can be expressed with the favored valuation model of chapter 3, the residual earning model where growth is valued only when earnings cover the required return.

To see this, think of the residual earnings model with no growth:

and the forward P/E ratio = This is the P/E with no growth in the simple valuation model.

So, a P/E with no growth is just given by r, the required return. If r = 10%, P/E = 10 and E/P = 10%.

A P/E higher than this is one with growth and that is given by

Note, again, that g here is not growth in earnings but rather growth in residual earnings; it is residual earnings growth that is valued, not earnings growth. Note also a theme in chapter 2 and throughout the book: Buying growth is risky, so buying more g can mean higher risk and thus a higher r, as with the analysis of the P/E in chapter 2.

This discussion points to two mistakes an investor can make in evaluating P/E ratios.

First, the investor might evaluate a P/E ratio based on expected earnings growth rather than value-added earnings growth, residual earnings. Setting a high P/E based on earnings growth from M&A is a good example: The acquisition may add considerable earnings growth but not necessarily value-added earnings that covers the cost of capital.

Second, the investor might not appreciate that growth does not necessarily add to price. If residual earnings growth is at risk of not being realized, it adds to r and thus should be discounted in the price. A risky acquisition might add to residual earnings growth but might not add to price because the expected growth is fully discounted for the risk.

But, beware. The calculation imbeds a trap: A seemingly attractive low PEG ratio can be due price bring discounted because buying growth is risky. Chapter 2 explains.

On March 1, 2024, the P/E for the S&P 600 small stock index was14.7 while that for the S&P 500 (weighted towards large stocks) was 21.2. Small stocks typically have higher growth prospects so, based on growth expectations, the P/E of 14.7 looks cheap. But buying growth in small firms could be quite risky so, with a discount for that risk, the lower P/E could be appropriate. As chapter 2 emphasizes, P/E is determined by expected earnings growth but also the risk to growth.

Buying small firms has typically yielded higher returns that large firms. This so-called small-firm return premium is sometimes seen as gravy, the yield to a trading strategy to earn abnormal returns (alpha). But higher returns go with higher risk, so the return premium could just be reward for taking on higher risk. Chapter 12 examines the small-firm return premium from this perspective.

“The p/e ratio of any company that’s fairly priced will equal its growth rate.”

… Peter Lynch, One Up on Wall Street.

Peter Lynch, manager of the Magellan Fund at Fidelity Investments from 1977-1990, is an heralded value investor with significant returns. Here is a video where he lays out his investing principles:

But this quote, P/E = g, is not quite spot on.

Chapter 2 explained with a simple valuation model:

P/E increases with g as the Lynch quote says, but it is reduced by risk that determines r. That is one of the themes of the book: Growth is risky. So, while price anticipates expected earnings growth it is also discounted to the extent that growth is at risk.

Look at Walgreens in early January 2025, traditionally viewed as a franchise company (with a brand and wide distribution) and thus a potential growth company. Look up its business profile.

Walgreens Boots Alliance (WBA): Trailing P/E = 5.98; Forward P/E = 6.18

Why do you think the P/E is so low? Could it be underpriced or is there other reasons? The PEG stood at 3.55. That might give you a clue. Your answer will determine whether you would include Walgreens in a low P/E screen.