The investor library

Bill Miller

1950- · Legg Mason Value Trust, Miller Value Partners · Fifteen years, then a collapse, then a comeback.

Overview

He beat the S&P 500 in fifteen consecutive calendar years, from 1991 to 2005 — the longest streak any fund manager has recorded. Then he lost roughly 55% in 2008, about 72% peak to trough, and watched a $16.5bn fund shrink to under a billion as investors left. Then, from a much smaller base, he produced one of the best runs of the following decade.

“Lowest average cost wins.”

Bill Miller

Background

Born in 1950 in Laurinburg, North Carolina; his father managed a truck terminal. He studied economics at Washington and Lee, served as an army intelligence officer overseas, and then spent several years in a philosophy doctorate at Johns Hopkins that he never finished — which shows in how he argues.

He joined Legg Mason in 1981 as a securities analyst, co-managed the Value Trust from 1982, and took it over outright in 1990.

Style, and how it evolved

Value defined by future cash flows, not by a low multiple. This is the substantive contribution and it was heretical at the time. If a business's worth is the cash it will produce, then a company on fifty times earnings can be cheap and one on eight times can be expensive, and the price-to-book screen that defined value investing is measuring the wrong thing.

Which is why he owned technology when value managers did not. Amazon, AOL, Dell and later Google sat in a fund with "Value" in its name, and he defended the position on the arithmetic rather than the label.

Lowest average cost wins. Buy more as it falls, because a lower average cost raises the return if the analysis holds. The whole method is in that sentence, and so is the disaster.

Concentration, and a high tolerance for looking wrong.

The evolution came after the collapse: the second act at Miller Value Partners was more concentrated still, less constrained by a mandate, and included positions — Bitcoin from around 2014 — that no institutional fund would have let him hold.

Performance

paid offa decisiona losswalked awaythe lifeshaded columns are the crashes — hover any mark

Fifteen consecutive calendar years ahead of the S&P 500, 1991-2005 ✓ — a public, daily-priced, audited mutual fund, so unusually checkable.

Then about −55% in 2008 and roughly −72% peak to trough ✓, with assets falling from around $16.5bn to under $1bn as investors redeemed. Measured across his whole tenure on the Value Trust, the long-run record is far closer to the index than the streak implies — a great deal of the compounding was given back.

Then the comeback ✓ — the funds he ran from 2012 produced several of the strongest years of the decade, before another sharp fall in 2022.

✓ documented — public mutual funds with audited, daily-priced records

Case studies

1999 — owning technology in a value fund. He held Amazon and AOL while the value establishment said the label forbade it, and argued that a discounted stream of future cash does not care which style box a consultant files it under. He was right about the principle and it made much of the streak. The mechanism: the value/growth dichotomy is a classification convention, not a fact about businesses. ✓

2008 — the method, applied to the wrong companies. Bear Stearns, Countrywide, AIG, Freddie Mac, Washington Mutual. He averaged down into all of them, exactly as the method says, because a lower average cost raises the return if the analysis holds. The analysis did not hold: these were not cheap businesses, they were insolvent ones. The mechanism: averaging down converts an analytical error into a fatal one, and there is nothing inside the rule that tells you which case you are in. ✓

2014 onwards — Bitcoin. He bought early and held through drawdowns that would have ended most professional careers, and it became a very large share of his personal wealth. Note what this says about the second act: freed of an institutional mandate, the same temperament that destroyed the fund in 2008 produced an extraordinary result. ✓

The other side of the record

A record this good is where scepticism is most worth spending, so:

The streak is a calendar artefact and he says so. Fifteen Decembers in a row is a much weaker claim than it sounds, and it was marketed as a much stronger one for a decade and a half. Anybody quoting it should also quote his own view of it.

The long-run number is far worse than the famous one. Investors who arrived because of the streak — and most of the $16.5bn did — experienced the collapse, not the run. It is the same gap between a fund's return and its investors' return that Lynch's and Wood's pages turn on, and here it is at its most severe.

"Lowest average cost wins" is a rule with no stopping condition. It is excellent discipline against a falling price and catastrophic against a failing business, and the rule itself cannot distinguish them.

He was buying the financial system's worst credits while arguing it was a liquidity crisis, not a solvency one. That was a specific analytical call, made publicly, and it was wrong.

And the comeback ran on a much smaller base, with a much freer mandate, which makes it a different exercise from the one that made his name — impressive, and not the same test.

Key lessons

• Value is about future cash, not about a low multiple. The most useful thing he argued for, and it cost him credibility to argue it.
• A record's shape depends on where you put the boundaries. Fifteen calendar years and zero rolling twelve-month streaks are the same data.
• Averaging down needs a solvency test in front of it, or it turns a mistake into a ruin.
• The money mostly arrives after the run, so the average investor's experience can be the opposite of the manager's.
• A career can have a second act, and his is the clearest evidence in this library that a collapse is not necessarily the end of one.

Reading and links

• Richer, Wiser, Happier — William Green, 2021. Contains the fullest account of the collapse and what he made of it afterwards.
• His shareholder letters and market commentaries — public, and unusually willing to restate the argument after it failed.

Profiles of investors, not views on any security, and not personal advice. Figures are marked ✓ where they are documented and ○ where they rest on secondary or private accounts — investing biography is heavily mythologised and the well-known numbers drift with each retelling. Capital is at risk.

Corrections and right of reply. Where a criticism is made of a named person it is stated as a specific measure over a stated period, attributed to its source, and separated from opinion. Assessments are opinion, honestly held, on facts believed accurate at the date shown. If anyone profiled here — or anyone acting for them — believes a fact is wrong, it will be corrected promptly and visibly, and a reply will be published alongside it on request.

← Jim SimonsSeth Klarman →