The investor library

Geraldine Weiss

1926-2022 · Investment Quality Trends · She signed it "G. Weiss" so it would be read.

Overview

She took a finance degree in 1945 and was offered secretarial work, repeatedly, for twenty years. In 1966 she started her own newsletter from a kitchen table in La Jolla with two thousand dollars, built a valuation method around the dividend yield, and ran it for nearly forty years. She signed it G. Weiss, because subscribers would not take investment advice from a woman, and did not appear publicly as herself until 1977 — when she went on television and let the audience work it out.

“Dividends don't lie.”

Geraldine Weiss

Background

Born in San Francisco in 1926, she studied business and finance at Berkeley and graduated in 1945 into an industry that would not employ her as an analyst. She married, raised children, and read — company reports, dividend records, market history — for two decades while being told the work was not open to her.

Investment Quality Trends launched in 1966. It still runs.

Style, and how it evolved

The dividend yield as the anchor, not the income. This is the part usually misunderstood. She was not a yield-chasing income investor; she used the yield as a measuring instrument. A long-established company's dividend yield oscillates between repeatable historical extremes, and because the dividend is far more stable than the price, the yield tells you where the price sits against its own history. Buy at the historic high-yield boundary, sell at the historic low.

Six tests before a company qualifies at all. Dividends raised five times in twelve years; an A-rating; at least five million shares outstanding; at least eighty institutional holders; twenty-five years of uninterrupted dividends; earnings up in at least seven of the last twelve years. The screen is doing as much work as the yield.

The method barely evolved, and she was explicit that this was the point: a valuation anchor that moves with fashion is not an anchor. What changed was the world around it — the qualifying universe has thinned as buybacks replaced dividends.

Performance

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

Investment Quality Trends was tracked by the Hulbert Financial Digest for decades and ranked repeatedly among the best newsletters on a risk-adjusted basis. ○ The model portfolio's returns are commonly put in the low teens annually — ahead of the market on a risk-adjusted basis rather than dramatically ahead outright, which is what a method built on mature dividend payers should produce.

The important caveat is structural, not sceptical: this is a model portfolio in a newsletter, not managed money. Nobody's capital was audited into those numbers.

✓ documented · ○ tracked by a third party, but a model portfolio rather than managed money

Case studies

The signature. Using an initial was not a flourish; it was the condition of being read at all, and she kept it for eleven years while the letter built a following on the argument alone. When she did appear, the subscribers had already decided the method worked. The mechanism: she removed the only variable her readers were going to judge her on unfairly, and let the record accumulate first. ✓

The yield boundary. Her working device was a chart of a company's own dividend yield over decades, with horizontal lines at the extremes it kept returning to. A price is hard to judge; a yield against its own twenty-year range is not. The mechanism: she replaced an absolute question — what is this worth? — with a relative one the data could actually answer. ✓

Refusing to widen the screen. As dividend payers thinned, the obvious commercial move was to loosen the six tests and keep the universe stocked. She did not, and the letter covered fewer names instead. ○

The other side of the record

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

The method has been quietly disqualifying the market for forty years. Twenty-five years of uninterrupted dividends excludes, by construction, almost every company that has driven returns since 1980. A framework that cannot look at the winners is not wrong so much as increasingly inapplicable, and buybacks have accelerated the problem.

A newsletter model portfolio is the weakest form of record here. No custody, no fees, no slippage, no clients redeeming at the bottom. Hulbert's tracking is real and independent and it still measures a published list rather than money at risk.

Newsletter survivorship is severe. Thousands launched; the ones still discussed are the ones that worked. A single forty-year survivor is not evidence about the population.

The yield anchor assumes the dividend is safe, which is exactly what fails in the situations where the yield looks most attractive. The six tests are there to prevent it, and they are a screen for the past.

Key lessons

• Anchor a valuation to something more stable than the price. The dividend moves slowly and deliberately, which is what makes it usable as a ruler.
• A relative question is answerable when an absolute one is not. Not what is this worth, but where does this sit against its own history.
• The screen is half the method. The yield rule only works on the narrow set of companies the six tests admit, and applying it more widely breaks it.
• Refusing to widen the universe is a discipline with a commercial cost, and she paid it.
• Let the argument arrive before you do. Eleven years of being read as G. Weiss is the most quietly instructive fact on this page.

Reading and links

• Dividends Don't Lie — 1988, with Janet Lowe. The method, in full.
• The Dividend Connection — 1995.
• Investment Quality Trends — still published; the archive is the primary source.

Marked ✓ where documented and ○ where the figure comes from third-party tracking of a model portfolio rather than audited managed money. This is a profile of an investor, not a view on any security.

Primary sources

Go to the thing itself. These are the subject’s own publications or an institution’s own site. No bookseller links, and no referral arrangements — a reading list that earns per click is not a reading list.

• Investment Quality TrendsThe newsletter she founded in 1966. Still published.

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.

← Charlie MungerPeter Lynch →