I Hope You Don’t Study Stocks

There are countless experts in the world. In Korea, when we talk about “professional occupations,” we typically mean doctors, lawyers, accountants, and the like. What these jobs have in common is that you only receive a license after completing a government‑mandated education process and passing a standardized national exam.

I believe the main reason the state manages professional licenses at the national level is to maintain the quality of medical, legal, and accounting services that are closely tied to our daily lives. Even among doctors who went through medical school and passed the national exam, there are still so‑called “quacks” who misdiagnose patients. But because these people make life‑or‑death decisions, licenses are only granted after they pass through a substantial number of filters. My opposition to increasing medical school admissions by 2,000 students is related to this as well, though I’ll leave a detailed discussion of that for another time.

Finance and investment professionals are a bit different. Finance plays a crucial role in our society by allocating resources through investment, yet most people working at financial institutions do not hold licenses such as the Investment Asset Manager certificate or the CFA. Financial regulators only require a minimum number of licensed professionals—like investment managers and compliance officers—when authorizing the establishment of firms such as investment advisory companies or asset management companies. The people who actually execute investments are not required to hold any license.

At the same time, in a society that aspires to a market economy, it would be quite odd if only government‑certified individuals were allowed to work in finance. Until recently, a “financial expert” or “investment professional” simply meant someone with deep knowledge and experience in the field—for example, an analyst at a securities firm who had gone through years of apprenticeship‑style corporate analysis training. But just as we saw in presidential elections, the rapid rise of social media and the internet has changed things. In the 2020s, the reality is that rather than those who truly deserve the title, it is now enough to be someone whom the public perceives as an expert to be recognized as a financial expert in that field.

Some readers may have noticed that the title of this piece is a parody of the book I Hope You Start Studying Stocks by Lee Hyo‑seok, a former fund manager and head of Lee Hyo‑seok Academy.

When it comes to stock investing, index funds or ETFs that track benchmarks like the S&P 500 are often recommended on the grounds that they outperform most fund managers. Some fund managers might find this unfair, because not all funds are run with the goal of maximizing returns. For example, there is a U.S. hedge fund that borrows the “Noah’s Ark” concept: it is designed to steadily post negative returns in normal times, but to generate very large gains when market volatility spikes or markets crash, thereby helping institutional investors who hold multiple funds to protect their portfolios in times of crisis.

Even so, for most people it is far more effective, in terms of lifetime returns, to forgo funds run by asset‑management professionals—those we might call “financial experts”—and instead allocate between an S&P 500 index‑tracking ETF and U.S. Treasuries, adjusting the weights based on their age, income and asset level, and risk tolerance.

Yet the financial experts we all know—through YouTube channels, Naver Premium Content, and various lecture platforms—insist that investing is an area where studying can improve your win rate and returns, and they sell their content, courses, or financial products on that basis. You might think many of them are simply acting in their own self‑interest, but based on the financial experts I’ve been introduced to and met over the years, most actually carry a strong sense of pride in their own “skills.” One of them is a director at a well‑known U.S. equity research firm, with a résumé that includes stints at major securities houses anyone would recognize. Right before the semiconductor winter hit in 2022, when I was pessimistic about the semiconductor sector, he confidently told me, “Semiconductors are no longer a cyclical industry,” and presented Intel as his top pick in the sector (he did emphasize that he was not a semiconductor specialist).

Another example is Lee Seung‑gyu, CEO of Dumoolmori Investment Advisory, which claims to manage assets using AI algorithms and reports AUM of 207 billion won on its website. In his recent newsletter, “Does the Cigarette Butt Investing Method Really Make Money?”, he argued that if you invest long‑term in the group of stocks with the lowest price‑to‑book ratios (PBR)—stocks that are cheap relative to their asset value—you can earn much higher returns than in the most expensive group. For me, this raised serious questions, because I had understood that over the past 30 years, mechanical “value investing” strategies based on simple metrics like low PBR or low PER had effectively died out.

There are still papers coming out showing that mechanical value‑investing strategies based on low PER or low PBR have not yet died in the U.S. stock market, but it is hard to believe that the low‑PBR group has delivered an average annual return of 12% since 2007.

Since I don’t know exactly how CEO Lee Seung‑gyu collected and processed his stock data, I can’t be definitive. But it appears that this implausible conclusion arose because he failed to remove the most important factor in quant backtesting: survivorship bias.

The image above is a famous illustration that almost always appears when explaining survivorship bias. It shows a model of U.S. fighter planes from World War II that managed to return safely despite being hit by enemy fire. The military marked the bullet holes on returning planes with red dots and reinforced those areas, but surprisingly, the share of planes returning safely from missions did not change much. They had overlooked the fact that the red‑marked areas were places where a plane could take hits without seriously compromising its ability to fly, which is precisely why those planes were able to make it back.

Quant backtesting works the same way. If you run statistical analysis on the universe of stocks that are listed today, any stocks that were delisted in the meantime are excluded from the data. A stock with the lowest PBR is one whose market cap is cheap relative to its book asset value. You can reasonably infer that this is often because (1) the company has failed to grow at all, or (2) its ability to generate cash flow is in doubt, raising questions about its viability. If an extremely cheap low‑PBR stock either (1) returns to a growth path or (2) resolves the uncertainty over its survival, a huge risk disappears, and the stock price is likely to rise sharply. A portfolio made up only of companies that fit (1) or (2) and have survived until now will naturally show very high returns. But if you include the stocks of companies that were delisted from the exchange or went bankrupt and became worthless, the returns would be much lower.

Ironically, the chart above—also from CEO Lee Seung‑gyu—actually explains why you should not buy low‑PBR stocks, even if he did remove survivorship bias: the low‑PBR group has extremely low trading volume, in other words, extremely low liquidity.

Saying that a financial product has low liquidity is effectively the same as saying that the odds your stock will trade at the “market price” are low. As a result, you often end up having to buy at prices above the market and sell at prices below the market in order to execute trades. This leads to higher transaction costs and, naturally, when you actually buy and sell stocks to build a portfolio and rebalance it quarterly, it reduces returns and increases trading risk. Therefore, even if the low-PBR stock group records higher returns than the market after removing “survivorship bias,” you need to calculate precisely what percentage of that excess return is simply “compensation” for low liquidity.

Aside from a handful of super-individual investors with more than 10 billion won in capital, virtually no retail investor trades with a “liquidity discount or premium” in mind. At the scale at which ordinary individuals typically invest, it isn’t all that important anyway, and almost no one is even aware of this concept to begin with. And I don’t think this is a problem that needs fixing.

I just want to ask two questions: