Li Lu

Li Lu's most consequential portfolio category is not buy or sell. It is "too hard" - a decision that preserves cash, attention, and capacity for the rare business whose economics can be understood well enough to act.

The Right to Reject Almost Everything

Li Lu founded Himalaya Capital in 1997 and remains its founder and chairman. The firm's current description says it primarily owns public companies in Asia and North America for endowments, foundations, family offices, individuals, and pension plans. The partnership is generally closed to new capital, and some holdings date back roughly twenty years.

Those facts help explain the method. A manager paid to gather assets or stay fully invested may need a continuous supply of acceptable ideas. A mostly closed partnership that tolerates cash can wait. Li's 2013 Columbia Business School interview divided propositions into yes, no, and too hard, with most in the last category. Concentration follows from saying no repeatedly; it is not a reason to lower the standard for knowing.

Li traced his starting point to a Warren Buffett lecture in Bruce Greenwald's class: buy good securities at bargain prices. His process then changed. He began with statistically cheap securities because he lacked experience, but company study shifted his attention toward strong businesses whose development and competitive advantages were more predictable. The price requirement remained. Quality did not excuse overpayment.

From Information to a Portfolio Decision

Ideas, Li said, came from extensive reading, studying many companies, and speaking with knowledgeable operators. That is sourcing, not an edge by itself. Further research must identify what is knowable, test the strongest counterargument, and expose what remains outside the analyst's competence.

Management assessment illustrates the limit. A capable and trustworthy team can change the business outcome. But a polished meeting can be staged. Li's answer was not that personal access always reveals character; it was that if management cannot be assessed, the investor should admit that it cannot be assessed.

Cash and current holdings provide the next control. In the 2013 interview, Li described cash as the starting opportunity cost. A candidate had to improve the risk-adjusted combination already owned. He also described three reasons to sell: the original analysis proves mistaken, price reaches an extreme of euphoria, or a clearly superior opportunity requires the capital. A falling quote without thesis failure was not one of them.

BYD: Predicting the Learner, Not Every Product

BYD is the strongest public test because Li explained the reasoning himself. The company made batteries, electronics, cars, and other engineered products in fast-changing markets. Li did not claim that every technology could be forecast. He focused on founder Wang Chuanfu's record, an engineering culture that repeatedly solved difficult manufacturing problems, the company's learning speed, a large field of potential applications, and a purchase price that he said supplied a margin of safety.

The thesis included conditions outside the company. Li noted the favorable industry environment and government support. That matters because a result later credited to corporate culture may also depend on industrial policy, access to capital, labor and supply chains, regulation, and the growth of China's vehicle market.

Li's interview says Charlie Munger became impressed by BYD and that this led to Berkshire's investment. An exchange-filed BYD announcement independently defines the transaction: Berkshire subsidiary MidAmerican agreed in September 2008 to buy 225 million new H shares at HK$8 each, about 9.89% of the company after issuance.

The filing does not name Li or disclose Himalaya's holdings. The causal link comes from Li's retrospective interview; the share count and price come from the exchange record. Himalaya's purchase dates, position size, currencies, sales, and fund return remain private. BYD's later success therefore tests the stated analysis but cannot supply a Li Lu performance number.

Patient Capital Is Part of the Method

In 2013 Li said his partnership had no management fee, gave investors the first 6% annual return, and then took a 25% incentive allocation. He also said all of his investment capital was in the funds. Those historical terms reduced the incentive to accept new money merely to increase asset-based fees and made unused cash less costly to clients. The current website does not confirm that the same terms remain in force.

Closing a fund protects capacity. A large inflow can dilute a concentrated portfolio or force purchases outside the manager's best ideas. A patient partner base can also reduce the chance that redemptions force sales during a temporary decline. These conditions do not remove agency risk or guarantee that the manager is right; they make waiting and rejection materially feasible.

Cross-border familiarity adds useful questions, not immunity. Investing in China and the United States involves different accounting enforcement, ownership structures, capital controls, currency, political authority, and access. A policy change can alter cash flows or foreign shareholder rights without appearing in a product analysis. No amount of company research grants control over that decision.

What the Public Portfolio Does Not Show

Himalaya files quarterly US holdings reports. Its March 31, 2026 Form 13F reported $3.201 billion of qualifying securities. That is not firm assets under management or a complete portfolio. Form 13F generally omits non-US securities, cash, private investments, many derivatives and shorts, and positions granted confidential treatment. It records quarter-end market value, not purchase cost or return.

No complete audited Himalaya return series appears in the official sources used here. Claims that Li compounded at a specific rate, manages a specified portion of the Munger family fortune, or always holds fewer than ten securities are therefore excluded. The firm's public archive contains speeches, essays, and interviews, not investor letters supplying those missing figures.

Concentration Is the Last Step

Li's method is sometimes summarized as deep knowledge followed by conviction. That skips the institutional and evidentiary work. Interest narrows the circle of competence. Rejection protects attention. Cash sets opportunity cost. Patient partner terms permit waiting. Research identifies both predictable elements and facts outside the analyst's control. Only then can a rare proposition justify substantial capital.

Even then, concentration magnifies an undiscovered error and can make exit difficult. The point of knowing what one does not know is not to become certain. It is to act only when the remaining uncertainty, price, position, and partnership can coexist.

Inside CompanyGraph

The deep-value configuration is observable: companies priced well below book value while current assets exceed current liabilities by a wide margin and the equity ratio sits high for the industry.