The earliest fund document we could fetch covers the six fiscal years to 31 March 1990, and over them Magellan beat the S&P 500 by about 2.5 percentage points a year after its 3% sales charge. Fidelity Magellan's Form 485BPOS filed with the SEC on 19 July 1994 shows a hypothetical $10,000 invested on 31 March 1984, after the sales charge and with distributions reinvested, alongside the S&P 500. Lynch managed the fund throughout that window.
Value of $10,000 invested on 31 March 1984, Magellan and S&P 500: 1985 $11,738 and $11,890; 1986 $18,380 and $16,371; 1987 $22,840 and $20,663; 1988 $20,639 and $18,943; 1989 $25,233 and $22,380; 1990 $30,360 and $26,694. Source: Fidelity Magellan Fund, Form 485BPOS, 19 July 1994, Statement of Additional Information, "Historical Fund Results"; chart data (CSV). The growth rates of 20.3% and 17.8% a year are our calculation; the index carries no costs, and the fund's figures are before tax.
The path was not smooth. In the fund's financial highlights, total return before the sales charge was 56.59% in the year to March 1986 and -9.64% in the year to March 1988, which contained the crash. Calendar-year returns in the same prospectus were 2.03% in 1984, 43.11% in 1985, 23.74% in 1986, 1.00% in 1987, 22.76% in 1988 and 34.58% in 1989.
The headline figures for Lynch's whole tenure come from interviews and press, not from a fund document we could fetch. PBS Frontline introduced its 1996 interview by saying that "Magellan was up over 2700%" from 1977 to 1990 (PBS Frontline, Betting on the Market, January 1997). A 2,700% gain over 13 years equals about 29% a year compounded, our calculation, which matches the figure often quoted.
In the same interview Lynch said: "when I took over in May of 1977, the fund was $20 million", and "When I left Magellan Fund, it was 14 billion." The 1994 filing confirms his retirement on 31 May 1990 and reports net assets of $13,162 million on 31 March 1990, up from $2,364 million on 31 March 1985.
What did Magellan hold, and what did Lynch tell individuals to do?
The fund held well over a thousand stocks; Lynch told individuals to follow a handful of companies they understand. In One Up on Wall Street (Simon & Schuster, 1989; Millennium edition, Fireside, 2000), Lynch wrote in the introduction, "The Advantages of Dumb Money", that "I own 1,400 stocks". In the prologue, "A Note from Ireland", he described deciding during the October 1987 crash "which of the 1,500 stocks in my fund should be sold".
His advice to individuals was narrower. In the same introduction he wrote that "any normal person using the customary three percent of the brain can pick stocks just as well, if not better, than the average Wall Street expert", and that "the amateur investor has numerous built-in advantages".
He also wrote that mutual funds are "a wonderful invention for people who have neither the time nor the inclination to test their wits against the stock market". In the 1996 PBS interview he said: "I think the average person could know three or four or five companies very well."
What does "invest in what you know" really mean?
It means using what you see at work and in the shops to find companies to research, then buying only after checking the numbers. The book's subtitle is "How to Use What You Already Know to Make Money in the Market". Lynch wrote in the introduction that "you can pick the spectacular performers right from your place of business or out of the neighborhood shopping mall, and long before Wall Street discovers them".
The popular version drops the second step, and Lynch corrected it himself. In the introduction to the 2000 Millennium edition he called it a "disclaimer": "Peter Lynch doesn't advise you to buy stock in your favorite store just because you like shopping in the store, nor should you buy stock in a manufacturer because it makes your favorite product or a restaurant because you like the food."
"Liking a store, a product, or a restaurant is a good reason to get interested in a company and put it on your research list, but it's not enough of a reason to own the stock!" He continued that you should never invest "before you've done the homework on the company's earnings prospects, financial condition, competitive position, plans for expansion, and so forth".
What happened to Magellan in the 1987 crash?
Magellan's month-end value fell 33.1% from August to November 1987, more than the US stock market's 29.9%, and took until April 1989 to regain its August 1987 level.
Month-end values, end-August 1987 = 100, Magellan and US stock market: October 1987 72.4 and 75.8; November 1987 66.9 and 70.1; December 1987 73.4 and 75.2; June 1988 87.6 and 86.6; December 1988 90.1 and 88.7; April 1989 103.7 and 99.9. Source: Magellan's monthly values from the chart data in its 1994 prospectus; US market total return from the Kenneth French data library; indexing is our calculation; chart data (CSV); monthly series (CSV).
Month-end data understate the intramonth low; the US market series is the CRSP value-weighted index with dividends, not the S&P 500.
Lynch was on holiday in Ireland when the crash came. In the prologue to One Up on Wall Street he wrote that "the one million shareholders in Magellan Fund had just lost 18 percent of their assets, or $2 billion, in the Monday session" of 19 October.
Our calculation from Yahoo Finance's daily price history for the fund gives a fall of 17.7% that day, against 20.5% for the S&P 500 price index. He spent the afternoon "deciding which of the 1,500 stocks in my fund should be sold to raise cash for the unusual number of fund redemptions", and wrote: "There was enough cash for normal circumstances, but not enough for the circumstances of Monday the 19th."
His own conclusion was about behaviour, not prediction. "If this is any example, less than three percent of the million account-holders in Fidelity Magellan switched out of the fund and into a money-market fund during the desperations of the week", he wrote, adding: "When you sell in desperation, you always sell cheap."
A 33.1% fall needs a 49.5% gain to recover, our calculation, which is why the 20 months from the August 1987 peak to April 1989 mattered for anyone who sold near the bottom. Our article on market crash recovery time explains that arithmetic, and our study of past stock market crashes shows how other asset classes behaved in later falls.
Did Magellan investors earn the fund's return?
We found no Fidelity or Lynch document that measures it, but the fund's own figures suggest the average dollar earned less than the fund because it arrived after the gains and left after the fall. A widely repeated claim that most Magellan investors lost money could not be traced to a document we could fetch, so this article does not use it.
Estimated net flows, fiscal years to 31 March: 1986 +$2.38bn; 1987 +$2.33bn; 1988 -$0.50bn; 1989 -$0.69bn; 1990 +$1.58bn. Source: our estimate from the financial highlights in the 1994 filing (net assets at each 31 March and each year's total return); chart data (CSV).
Each bar is the year-end net assets minus the starting net assets grown by that year's return; it counts distributions taken in cash as outflows and ignores when in the year money moved, so it is approximate. The year to March 1988 combines inflows before the crash with outflows after it.
The largest inflows followed the 56.59% and 24.26% years to March 1986 and 1987; money then left in the two years that contained the crash and its aftermath.
Treating those flows as an investor's cash flows, the average dollar earned between 16.3% and 19.6% a year from April 1985 to March 1990, depending on when within each year the money moved, against 20.9% a year for the fund before its sales charge, our calculation. Lynch described the same pattern from his first years: "the first three years I ran Magellan, I think one-third of the shares were redeemed", he told PBS.
Recent data show the gap persists. Morningstar's Mind the Gap 2026 estimates that the average dollar in US mutual funds and ETFs earned 8.7% a year over the 10 years to 31 December 2025, about 1.2 percentage points less than the funds' 9.9% (p. 1). The gap was 0.4 percentage points a year for the least volatile fifth of funds and 2.1 points for the most volatile fifth (Exhibit 10, p. 14).
The measure has critics: researchers writing in the Financial Analysts Journal estimate that only about 0.10% a year of the 1.2% gap in Morningstar's 2025 edition reflects harmful timing, the rest coming from how the metric treats money that follows past returns (University of Arkansas Walton College, 7 July 2026). The same caution applies to our Magellan estimate.
What did Lynch say about holding through a fall?
Lynch put the investor's capacity to hold first: the fund's return compounded only for money that stayed through the falls. In chapter 4 of One Up on Wall Street, "Passing the Mirror Test", he asks three questions before any stock purchase: "(1) Do I own a house? (2) Do I need the money? and (3) Do I have the personal qualities that will bring me success in stocks?"
His rule for the second: "Only invest what you could afford to lose without that loss having any effect on your daily life in the foreseeable future." In the Millennium edition's introduction he added that it is "important not to buy stocks or stock mutual funds with money you'll need to spend in the next twelve months", because "You don't want to be forced to sell in a losing market to raise cash." Our article on cash and withdrawals during a market crash shows what a reserve changes.
What did the checker return for the ETF illustration?
A 100% S&P 500 ETF example received C overall in copio's checker on 3 November 2026: A for fees, B for long-term assumptions, D for crises and D for diversification, with no historical-efficiency reading. This grades a broad US-equity example, not Magellan or Lynch's skill.
There is a material data limitation: with the same account type and weight, an Amundi ETF tracking the S&P 500 returned diversification F and efficiency B, while this iShares ETF returned D and insufficient history. Until the difference is explained, these recorded outputs cannot support a ranking of equivalent market exposure.
The illustration is not Magellan. Magellan's holdings in Lynch's years are not on EDGAR, and no fund can reproduce a manager's stock picking, so the checker cannot capture Lynch's skill. The S&P 500 appears as a benchmark in the fund's prospectus. Entering that benchmark through an ETF illustrates US large-cap exposure; it does not reconstruct the fund's portfolio.
| ISIN | ETF | Weight |
|---|
| IE00B5BMR087 | iShares Core S&P 500 UCITS ETF USD (Acc) | 100% |
The illustration is held in a French securities account (compte-titres) and was entered in the checker on 3 November 2026.
- Overall: C.
- Fees: A, 0.07% a year, or €7 a year for every €10,000 invested, fund and account included. An index ETF costs little; Magellan's investors paid a 3% sales charge on purchases in the years shown above.
- Long-term assumptions: B, 0.82 percentage points a year below the expected annual growth of the checker's modelled growth-optimal portfolio. On those assumptions, a portfolio of US shares alone gives up a modest part of the expected growth on offer, not most of it. This is a model reading, not a forecast.
- Historical efficiency: no reading. The checker reported insufficient history for this ETF on the measurement date.
- Crises: D, -40%, the estimated fall in a 40% shock to US equities. This single-market ETF takes the modelled US-equity shock; it is not a reconstruction of Magellan in 1987.
- Diversification: D, one source of growth. If US large caps stop growing, the checker estimates that the portfolio's expected growth above cash falls from 4.4% to 0.0% a year. The ETF's 503 companies still share broad US-equity exposure; a company count alone does not measure independent risk.
The illustration leaves out Lynch's stock selection, any foreign shares and the cash the fund kept for redemptions, which the documents we fetched do not quantify, and the fund's sales charge. You can enter the same illustration, or your own portfolio, in the free portfolio checker.
What does copio think of this portfolio?
Lynch's record makes the investor's ability to stay invested part of the portfolio decision. The documented period includes a 33.1% fall in three months and a 20-month recovery to the previous peak. The fund's return and the return experienced by incoming and departing money differ, but an aggregate cash-flow gap does not establish why each investor traded. Nor does a 100% S&P 500 illustration reproduce Magellan's stock selection, foreign holdings or redemption cash. Copio's conclusion is to test funding needs and concentration before relying on a manager's headline return.
Three checks for your portfolio
- Calculate what a sharp fall would leave available. Apply a one-third loss to your equity holdings as a scenario and compare the remaining liquid assets with planned withdrawals. Write which expenses would require a sale before a recovery; the historical 20 months is an observation, not a deadline the next crash must meet.
- Look through the number of holdings. Combine overlapping funds and record exposure by company, sector and country. A long list of stocks can still share an economic risk; the S&P 500's US exposure does not prove Magellan had the same portfolio. Use more than one crash period to check which holdings offset losses and which only reduced them.
- Set review rules before a losing stretch. Write what would justify changing a fund: a changed mandate, rising costs, an unsuitable risk level or a broken investment thesis. Measure your personal result with dated contributions and withdrawals, and compare the fund's time-weighted return separately. That makes a decision reviewable without assuming every withdrawal after a loss was a mistake.
Our cash and withdrawals study shows how funding needs affect recovery. The portfolio checker can help explore your own funds, subject to the data limitations described above.
FAQ
How did Peter Lynch's Magellan Fund perform?
Fidelity Magellan's 1994 SEC filing shows that $10,000 invested on 31 March 1984 grew to $30,360 by 31 March 1990, after the fund's 3% sales charge, against $26,694 in the S&P 500. That is about 20.3% a year against 17.8%, our calculation. For Lynch's full tenure from May 1977 to May 1990, PBS Frontline put the gain at over 2,700%; we found no fund document covering the years before 1984.
What happened to Magellan in the 1987 crash?
Magellan's month-end value fell 33.1% from August to November 1987, against 29.9% for the US stock market, by our calculation from the fund's monthly data. Lynch wrote that the fund's shareholders lost 18 percent, or $2 billion, on 19 October 1987 alone, and that he sold stocks to raise cash for redemptions. The fund regained its August 1987 level in April 1989.
What does Peter Lynch's 'invest in what you know' mean?
In One Up on Wall Street (1989), Lynch argued that individuals can spot promising companies through their work and shopping before professionals do. In the introduction to the 2000 Millennium edition he added a disclaimer: liking a store, product or restaurant is a reason to put a company on your research list, but not enough of a reason to own the stock, and you should never invest before doing the homework on earnings, financial condition and competitive position.
Did Magellan investors earn the fund's return?
We could not find a Fidelity or Lynch document that measures it. Our rough estimate from the fund's annual net assets and returns suggests money flowed in after the strong years to March 1987 and out in the two years after the crash, so the average dollar earned between 16.3% and 19.6% a year from April 1985 to March 1990, against 20.9% for the fund. Morningstar's 2026 study finds the same pattern across US funds: investors earned 8.7% a year over the decade to 2025, against 9.9% for the funds.
What grade does copio's portfolio checker give a Lynch-style portfolio?
On 3 November 2026, copio's checker gave a 100% iShares Core S&P 500 UCITS ETF USD (Acc) illustration grade C: A for fees (0.07% a year), B for long-term assumptions (0.82 percentage points a year below its modelled growth-optimal portfolio), D for crises (an estimated 40% fall in a 40% shock to US equities) and D for diversification (one source of growth). It gave no historical-efficiency reading, citing insufficient history. The illustration stands for a fully invested US equity portfolio, not Magellan itself, and cannot capture Lynch's stock picking.
