The Years He Was Right and Nobody Listened
Reading Notes · The Way of Munger — Part 3
Nothing in this post is financial advice. I'm a software engineer writing reading notes, not an analyst. Everything here reflects Munger's historical remarks — not predictions, not recommendations.
There's a particular kind of frustration that comes from being right too early. You warn people. You explain the mechanism clearly. You name the specific thing that will break. And then — nothing. The system keeps running. People keep making money. You start to look like the cranky old man at the party.
Then the party ends.
Reading the 2003–2010 Wesco Financial chapters of The Way of Munger is a strange experience precisely because you know how the story ends. By the time you reach the 2008 and 2009 meetings, Munger is not triumphant. He's not crowing. He's just... exhausted. And still explaining, in plain English, why the derivatives machine was always going to blow up.
"Mark-to-Myth": The Warning That Predated the Crisis
The mainstream narrative about the 2008 financial crisis tends to credit a handful of short-sellers and a few prescient analysts as the people who "saw it coming." Munger had been saying it, loudly, for years before that — at small annual meetings in Pasadena, to anyone who would listen.
His core argument about derivatives wasn't complicated. He called mark-to-model accounting "mark-to-myth." The logic was simple: when financial instruments are priced using proprietary models rather than actual market transactions, the people running those models have enormous incentive to make the numbers look good. There's no external discipline. "Derivative contracts are opaque," he noted in the transcripts, "and accounting rules allow people to use models to calculate their value — this creates serious loopholes that are easily exploited."
When Berkshire unwound the derivative books it inherited from General Re, the empirical result confirmed the theory exactly. The book value turned out to be overstated by $400 million. Not a rounding error. Not a modeling artifact. Four hundred million dollars of fictional profit, revealed when someone actually tried to settle the positions.
What strikes me, reading these chapters now, is that Munger never positioned this as a clever insight. He presented it as obvious. The incentive structure creates fake profits. The fake profits generate real bonuses. The real bonuses are spent. When the positions eventually unwind — and they must, because reality doesn't care about your model — the losses are also real, but the bonuses are long gone. He compared the entire mechanism to a Ponzi scheme, and I think that comparison is more precise than most financial commentary gave it credit for being.
The crisis, when it came, didn't surprise him. What I sense in the 2008 and 2009 transcripts is not vindication — it's something more like grief.
Stock Options and the Corruption He Couldn't Ignore
The derivatives critique was the loudest alarm, but Munger used these years to extend the same logic to executive stock options — a target that made him far more enemies in corporate America.
His argument was accounting-based, not moral. Stock options are compensation. Compensation is a cost. When options are issued but not expensed, the reported earnings are inflated — often substantially. He cited software companies where the understatement reached 12 to 14 percent of actual operating costs. Then that inflated earnings number gets multiplied by the market's P/E multiple, sometimes 40x or 50x in tech. The distortion at the stock price level is enormous.
"The stock option system," he said, "is very similar to a Ponzi scheme." The first generation extracts wealth through inflated stock prices. The second generation eventually inherits the reckoning.
I find this analysis more useful than most discussions of "pay for performance" I've read, because it ignores the moral dimension entirely and just asks: what does the accounting actually measure? And the answer, Munger argued, was: not quite reality.
The BYD Bet: The Exception That Proves the Rule
In 2008, Berkshire Hathaway acquired a 10 percent stake in BYD for $230 million.
This is worth pausing on. Munger had spent decades warning about the difficulty of predicting technology companies, specifically because technological progress tends to destroy the capital returns of the investors who fund it — even when the technology succeeds. The airplane industry transformed civilization. It also destroyed most of the capital ever invested in it. Why would an electric vehicle company be different?
The research notes describe his analytical framework here as the "dogfish model" — named for a predatory fish that, introduced into a pond of trout, consumes everything in sight. The question wasn't whether EVs would win. The question was whether BYD had the kind of overwhelming cost and manufacturing advantage that would let it eat the pond regardless of what the broader market did.
Munger believed it did. He also believed in Wang Chuanfu specifically — he compared Wang's combination of engineering depth and entrepreneurial execution to a blend of Edison and Welch. Coming from someone who spent most of his career being skeptical of founder-worship, that's a notable endorsement.
The stock went nowhere for five years after the purchase. Then it exploded — eventually trading at nearly 200x earnings. Munger held through all of it.
The BYD investment is interesting to me not as a template to follow — it absolutely is not — but as a calibration point for understanding where his "circle of competence" actually ended. He talked constantly about the "too hard" pile. AI, Bitcoin, most technology: too hard. But a manufacturing company with identifiable cost advantages and a founder he could evaluate? That he felt he could understand. The circle was smaller than it looked from outside, but more precisely drawn.
The Efficient Market Hypothesis, Again
Throughout this period, Munger kept returning to academic finance with a kind of weary exasperation. He called the strong-form efficient market hypothesis "pure nonsense." He dismissed the use of beta and standard deviation as risk measures, calling them "useless."
His critique wasn't just that these models are wrong. It was that they're wrong in a specific way that produces specific harms: they make students think they've understood risk when they've only learned to measure volatility. Risk is the possibility of permanent loss. Volatility is not risk. A great business whose stock temporarily drops 50 percent is less risky than a mediocre business that never moves.
He also noted, somewhat acidly, that a major investment firm once hired brilliant graduates from Harvard and Wharton and asked each one to contribute their single best investment idea. The resulting portfolio — a collection of high-IQ recommendations from highly credentialed people — failed to beat the index three times in a row. His diagnosis: pooling independent smart ideas without a unified analytical framework is the modern equivalent of alchemy.
I find this observation genuinely useful as a software engineer who occasionally encounters similar dynamics. Aggregating expert opinions doesn't compound intelligence. It averages it.
The End of Wesco
Wesco Financial was folded into Berkshire Hathaway in June 2011 — technically after this period, but it's impossible to read the final Wesco-era transcripts without sensing the conclusion approaching. Munger had been at the helm for nearly 35 years.
What I expected, going into these chapters, was some sentimentality. A valediction. Instead, what I found was characteristic bluntness. Wesco had become, in his telling, a kind of anachronism — a separately-listed company that was almost entirely owned by Berkshire anyway, creating complexity without corresponding benefit. The structure made less and less sense. So they ended it.
There's a lesson here that I keep coming back to: the willingness to close things that don't need to remain open. Munger didn't have an emotional attachment to Wesco as an institution that overrode his judgment about whether it should continue as a separate entity. He built it, ran it, and when the right time came, dissolved it cleanly. That kind of clarity about endings is harder than it sounds.
What Matured in These Years
Reading the 2003–2010 remarks as a sequence, something does change across the decade. The earlier Wesco meetings feel like a man still building his framework, still discovering what he thinks. By this period, the framework is fully formed and the project has become application — bringing the same models to bear on new situations, consistently, without flourish.
The core remains what it always was: concentrate rather than diversify, because diversification is what you do when you don't know what you're doing. Use opportunity cost as your filter — not "is this good?" but "is this better than my best current option?" Stay inside the circle of competence and don't pretend it's larger than it is. Read everything. Lower your expectations for happiness but not for rigor.
The 2008 crisis chapters add one more thing to this list, implicitly: be prepared to be right too early. Munger spent years explaining derivatives, years explaining the accounting games, years warning about the Ponzi-scheme dynamics of modern finance. He was correct. The market ignored him for long enough that being correct felt, for a while, like being wrong.
I don't think there's a comfortable resolution to that. You can't know in advance how long the wrongness will last. All you can do is be honest about the mechanism and be patient.
That's not a satisfying lesson. But I think it's the accurate one.
This is Part 3 of a reading series on 《芒格之道》 (The Way of Munger). Parts 1 and 2 covered the earlier Wesco years. Part 4 will move to the Daily Journal Corporation era.
Adrian is a performance engineer. Nothing in this series constitutes investment advice.
Leave a comment ✎