Long-Termism
“We only judge our investment performance over five year plus time horizons. In truth it takes at least a decade to provide adequate evidence of investment skill.”
The investment discipline of holding positions for five, ten, or twenty years — long enough for a company's underlying growth to compound and for short-term price volatility to become irrelevant. Anderson operated Scottish Mortgage on an explicit five-to-ten-year investment horizon.
Anderson's long-termism was not passive patience but active conviction: he held Tesla through near-bankruptcy, Amazon through the dot-com bust, and Illumina through years of underperformance relative to the index. The philosophical basis is that short-term stock price movements are dominated by noise (quarterly earnings beats, macro sentiment, benchmark flows) while long-term prices converge toward underlying business value creation. The structural advantage of a closed-end investment trust like Scottish Mortgage was that it could be genuinely long-term, unlike an open-ended fund that faces redemption pressure whenever performance lags.
Long-Termism
Definition & Origins
Long-termism, in Anderson's practice, is the discipline of holding positions for five, ten, or twenty years — long enough for a company's underlying growth to compound and for short-term price volatility to be revealed as noise. Scottish Mortgage operated on an explicit minimum horizon of five years, judged its own performance only over five-year-plus periods, and in truth accepted that "it takes at least a decade to provide adequate evidence of investment skill."
The idea has roots as old as investing itself, but Anderson's version was forged in a specific institutional and intellectual context. Institutionally, the closed-end investment trust was decisive: Scottish Mortgage could not face redemptions, so it could hold through drawdowns that would force an open-ended fund to sell at the bottom. Intellectually, the commitment deepened as the implications of the power law of returns became clear: if all wealth creation comes from a handful of extreme compounders, then the only way to capture it is to own those companies for the entire journey — through every crash, scare, and derisive headline. Anderson also credits a blunt cultural change at Baillie Gifford: in Aberration or Premonition? he calls the abolition of quarterly performance reviews "the single most important change Baillie Gifford as a whole ever made," recalling the absurdity of mass criticism of a manager "for being underweight the Royal Bank of Scotland in the last three months."
His framing of the concept was never passive patience. It was, as the 2017 Annual Report put it, an active allegiance: "We own companies rather than rent shares."
Core Ideas
Short-term prices are noise; long-term prices converge on value creation. The efficient-market reading of daily prices treats every tick as information. Anderson's counter-position: almost everything that drives returns over five or ten years — learning curves, market creation, founder decisions — is invisible at the one-year horizon, while almost everything that moves prices over one year is irrelevant over ten. Quarterly earnings "beats" and "misses," Federal Reserve speculation, and macro headlines fall into what he called, in a phrase he returned to repeatedly, "the near pornographic allure of news."
The drawdown is the price of admission, not a signal of failure. Citing Bessembinder's companion research, Anderson emphasized that even the greatest compounders suffer savage declines along the way: the top 100 U.S. stocks by decade returns spent an average of ten months underwater, with average losses of 32.5%; for the top 200, the figure approaches 50%. Amazon fell 46% peak-to-trough in 2006, during the very period Scottish Mortgage was building its holding. "The share price drawdowns will be regular and severe. 40% is common. The stock charts that look like remorseless bottom left to top right graphs are never as smooth and easy as they subsequently appear" (Stay on the Road Less Travelled).
Interrupting compounding is the cardinal sin. Because extreme outcomes accrue disproportionately in the later years of an exponential curve, selling early forfeits the most valuable segment. "We mustn't defend investments that have lost their purpose but interrupting compounding is the very worst that we can do" (Aberration or Premonition?). Anderson applied the verdict to his own record: repeatedly trimming Amazon whenever the holding approached 10% of assets was, he wrote in 2021, "misguided" — his clearest public confession of having violated his own principle.
The structure of the vehicle must match the horizon of the strategy. Genuine long-termism is impossible if your capital can flee. The closed-end trust, a century-old Scottish partnership as manager, and shareholders explicitly warned that "Scottish Mortgage is most suited to those who share its patient, long term approach" formed a single aligned system. The same logic later drove the move into private companies, where patient capital could extend the horizon beyond even what public markets allowed.
Practical Application
Holding Amazon since 2004–05. The defining case. Scottish Mortgage became a major Amazon owner around 2004–2005, held through the 46% drawdown of 2006, through years of "it doesn't have any profits," through peers at client conferences "declaring Amazon their favourite short," and through the skepticism directed at Prime and "Amazon Elastic Compute" — the latter becoming AWS. Anderson's own verdict was not triumph but self-critique: "what took us so long?"
Holding Tesla through near-death and derision. From the initial 2013 purchase, Tesla was attacked by short sellers as a capital-incinerating fraud. Anderson's response, set out in Stay on the Road Less Travelled, was that the battery and manufacturing learning curves made electric dominance "as close to inevitable as investing allows," so the correct action was to "listen to experts and wait." The trust held through the 2018–19 production hell and was rewarded with one of the great public-market outcomes of the era.
Institutional design. Long-termism was engineered, not merely espoused: no performance fees ("it increases pressure and narrows perspective"), no quarterly reviews, low turnover ("we still own 29 of the top 30 shares from the previous year" — 2016 report), and a willingness to look eccentric. As Anderson told Ritholtz, if you compete on the short-term game you face 150 rivals predicting the Fed; the long-horizon game "is a field where it is a little bit easier."
Stewardship through crises. The principle extended to behavior as a shareholder: quiet support for Rolls-Royce's multi-year renewal, public regret over ARM's sale to SoftBank, and the general rule that at critical moments the trust would be "willing accomplices" of founders rather than another source of quarterly pressure — patient capital as an explicit competitive advantage in winning allocations.
Common Misconceptions
Misconception 1: Long-termism means buy-and-forget passivity. Anderson's version is intensely active: continuous re-examination of whether a company retains "the possibility to be in the top five percent of outcomes," fresh reading of learning curves, and direct engagement with management. The 2016 review showed him adding Tesla shares after the Model 3 unveiling; the 2021 interview showed him trimming Tesla when the upside mathematics changed. The horizon is long; the scrutiny never stops.
Misconception 2: It is a counsel of stubbornness. The loyalty/stubbornness distinction is explicit in Aberration or Premonition?: "The advantages of loyalty usually trump the disadvantages of stubbornness" — but the essay immediately insists that investments which have "lost their purpose" must not be defended. Long-termism is not an excuse to hold the failed; it is a demand to hold the still-compounding through temporary disgrace. Amazon's later years show the exit side too: once it was "seen as good value, safe and acceptable" and no longer led by its founder, "our enthusiasm waned."
Misconception 3: Anyone can do it — the returns make it easy. The entire psychological machinery of markets works against the holding of volatile outliers. "It's important to learn how to suffer," Anderson wrote — recalling colleagues "kindly pointing out that 'it doesn't have any profits,'" Barron's declaring Alibaba "the reincarnation of Enron," and "the daily barrage of hedge funds attacking Tesla." Even an omniscient investor would suffer: the "God's portfolio" study he cites showed that perfect foreknowledge of the best stocks still produced severe relative and absolute drawdowns along the way.
Anderson's Own Words
"We are long term in our investment decisions. It is only over periods of at least five years that the competitive advantages and managerial excellence of companies become apparent... We own companies rather than rent shares. We do not regard ourselves as experts in forecasting the oscillations of economies or the mood swings of markets."
"It's important to learn how to suffer. We're now trained to know that Amazon 'misses' earnings a couple of times a year. We cope with the numerous downward lurches in share price that result from this and from colleagues and clients kindly pointing out that 'it doesn't have any profits.'"
"In retrospect enduring the siege looks easy. In each great investment what looks like a straight and exponential line of bottom-left to top-right compounding is in reality jagged and painful."
"We should apologise for our willingness to trim Amazon back repeatedly when our holding size approached 10% of assets. That was misguided."
"We only judge our investment performance over five year plus time horizons. In truth it takes at least a decade to provide adequate evidence of investment skill."
Thought Evolution
Key Letters / Related Concepts
Key letters: Scottish Mortgage Annual Report 2016 · Scottish Mortgage Annual Report 2017 · Aberration or Premonition? (2018) · Stay on the Road Less Travelled (2021) · Masters in Business Interview (2022)
Related concepts: Patient Capital · Power Law of Returns · Refusal to Forecast · Benchmark Irrelevance · Epistemic Humility