If Picking Winners Matters So Little, What Decides the Outcome?

Ask most people what separates a good investor from a poor one and the answer involves choosing: the right company, the right fund, the right moment to buy. The search for winners fills the financial pages and a good deal of dinner-table conversation. Yet for someone saving steadily over several decades, the choice of individual holdings is one of the weaker forces acting on the final sum.

The stronger forces are quieter, rarely making news and never appearing as a line on a statement. A handful of pieces of arithmetic, together with a few predictable habits of mind, settle most of what a long-term portfolio ends up worth.

Inflation erodes money that is left still, and compounding rewards time far more than insight while working against the saver through costs as faithfully as it works for them through returns. Gains in the stock market are concentrated in so few companies that owning broadly matters more than choosing well. And the return a fund reports is not the return its owners take home, because their own decisions open a gap between the two.

None of these forces requires a forecast. Each can be measured, and the evidence for each is consistent, if not beyond dispute in every detail. Taken together they give a different picture of what an ordinary investor is actually trying to manage.

Inflation and the cost of standing still

The first force works on money that is not invested at all. A savings balance that never falls feels like the safe option, but real return, roughly the nominal rate minus inflation, is the only measure of whether money has gained or lost ground. At 2% inflation a year, the Bank of England's target, prices double in about 35 years, so idle cash surrenders around half its buying power across a working life. When inflation runs hotter the loss is faster: consumer price inflation in Britain reached 11.1% in October 2022, a forty-year high, while typical savings rates stayed far lower.

The long record suggests this is the normal condition of cash rather than a recent accident. Elroy Dimson, Paul Marsh and Mike Staunton assembled returns for sixteen countries from 1900 onwards; over the twentieth century, short-term government bills, the nearest thing to cash, earned only about 1% a year above inflation in Britain and the United States, and lost real value in several decades, notably the 1940s and 1970s. After 2009, with Bank Rate at 0.5% and inflation near 5% by 2011, British savers holding cash were steadily becoming poorer.

The implication is not that cash is wrong. It has a proper role for emergencies and for bills that are close at hand. The point is that holding still is itself a decision with a cost, and that cost compounds.

Compounding runs in both directions

Compounding is the second force, and it runs in both directions. Returns earn returns, so growth depends on the size of the base, and the base keeps growing. At an illustrative 7% a year, £10,000 roughly doubles in a decade and becomes almost £150,000 after forty years, with most of that sum arriving in the final decades. This is why time left undisturbed does more work than cleverness.

One illustrative saver puts £2,000 a year aside from age 25 to 35 and then stops, contributing £20,000 in total. Another starts at 35 and contributes the same amount every year until 65, a total of £60,000. At 7% a year the early saver ends with about £210,000 and the late saver about £189,000, because the first decade of contributions compounds for three extra decades. The result depends on the assumption: at 6% the two finish roughly level, and real returns never arrive as a smooth annual figure.

Costs compound exactly as returns do

Compounding also turns against the investor through charges, because a fee is taken from a growing balance every year: costs compound exactly as returns do. A 1% annual charge on a 6% return removes a sixth of the return each year, and over forty years cutting an illustrative 6% to 5% leaves a final sum about 32% smaller. When Morningstar's research director Russel Kinnel tested in 2010 which fund characteristics predicted later performance, low expense ratios proved a more reliable signal than the firm's own star ratings. Returns cannot be controlled, whereas costs can be, with certainty.

Returns concentrated in a few winners

If cost matters so much, the obvious objection is that a skilled manager or a sharp stock-picker might earn it back. The third force explains why that is harder than it looks. Hendrik Bessembinder examined the lifetime returns of every US common stock listed between 1926 and 2016. More than half did worse over their listed lives than one-month Treasury bills, while about 4% of companies accounted for all of the net wealth the stock market created.

Returns are concentrated in a few big winners, and those winners are hard to identify in advance. A portfolio of ten familiar names feels diversified but is really a bet that those ten include some of the rare companies that carry the market. Broad ownership is the only dependable way to be sure of holding the few that matter.

Diversification and its limits

The arithmetic of combining holdings adds a further reward. Two assets that each have a volatility of 20%, with unrelated returns, held half and half, produce a mix with a volatility of about 14%, while the expected return is simply the average of the two. In other words, risk falls without giving up expected return, which is why diversification is so often called the only free lunch in investing.

It is not complete protection. Correlations are estimated from the past and can rise under stress: in 2022, as inflation and interest rates climbed, shares and high-quality government bonds fell together after roughly two decades of tending to move in opposite directions when markets were strained. Diversification removes risk that is not rewarded; it cannot remove the risk of the whole market falling.

The same concentration explains the sobering evidence on professional stock-picking. William Sharpe showed in 1991 that active investors as a group must earn the market return before costs, because together they own the market, and that after their higher costs they must, on average, trail it. S&P Dow Jones Indices' SPIVA reports, which correct for funds that close, have typically found around nine in ten US large-company funds behind the S&P 500 over fifteen-year periods. Some managers do outperform; the difficulty is knowing which ones beforehand.

The behaviour gap

The final force is the one investors carry with them. Money tends to flow into funds after strong performance and out after weak performance, so the cash present during the best stretches is often not the cash present during the worst. Morningstar's Mind the Gap studies measure this by comparing the time-weighted return a US fund reports with the dollar-weighted return its investors actually earned, and have generally found investors trailing by around one percentage point a year or more. That shortfall is the behaviour gap, a difference made entirely of the owners' own timing.

Trading more tends to widen it. Brad Barber and Terrance Odean examined the accounts of 66,465 US households at a discount broker from 1991 to 1996. The average household earned 16.4% a year against 17.9% for the market, and the fifth that traded most earned 11.4%, a shortfall explained by trading costs and poor timing.

The size of the gap is disputed. The DALBAR studies report far larger figures, but their method has been widely criticised for overstating the effect, and the popular claim that missing the market's ten best days roughly halves a twenty-year return draws similar objections, since missing the worst days would have helped just as much. The direction, however, is consistent: the average investor earns less than the average fund, and the difference is made of decisions.

What the arithmetic puts first

Put together, these forces rearrange the priorities of an ordinary long-term investor. The questions that matter most are not which company will triumph, but how much is being lost to inflation, how long money can stay invested, what is being paid each year and to whom, how broadly the holdings are spread, and what rule will prevent a panicked sale when prices fall. Each question has a measurable answer, and none depends on predicting the future.

That is the quiet advantage of the arithmetic: forecasts fail often, but the effect of a 1% charge compounded over forty years does not. It carries a limit worth stating plainly, since every figure above describes the past, much of it from one unusually successful market over one particular century, and past returns are no guarantee of future ones. What the arithmetic offers is not a promised result. It offers a way to stop giving away the returns that markets do provide, which for most savers is a larger prize than any winner they might have picked.