Home / Learn / The lost decade
Learn

The lost decade

Between 2000 and 2009 the S&P 500 went backwards even with dividends reinvested. A lump sum invested at the start finished down 9%. The same money paid in at £100 a month finished up — no timing, no skill, no decisions after the first one.

By Nathan Wickham-Hurd · Founder, Oak Growth · First-class Economics & Finance, MBA · Last reviewed August 2026

The short answer

Someone who put a lump sum into the S&P 500 in January 2000 and left it there had less money ten years later — down about 9% even with every dividend reinvested. Someone who paid in exactly the same amount at £100 a month finished ahead. Same index, same total, same decade. The only difference was when the money went in.

What actually happened, month by month

The decade opened at the dot-com peak and contained two crashes. This is the real path — 120 monthly readings with dividends reinvested, not a smoothed line.

9001,1001,3001,5001,700Mar 2009 low · 873Jan 2000 · 1,426Dec 2009 · 1,296-5.3%-12.6%-20.2%-39.7%2000200120022003200420052006200720082009S&P 500 total return, monthly, Jan 2000 – Dec 2009 (dividends reinvested)

This is an index level, not a sum of money. Red marks each negative calendar year. With dividends reinvested the index still finished the decade below where it started — down 9.1% — and at the March 2009 low it was some 39% under its starting point.

YearTotal returnIndex at year end
2000-5.3%1,350
2001-12.6%1,180
2002-20.2%941
2003+22.1%1,149
2004+12.7%1,295
2005+6.9%1,384
2006+14.0%1,578
2007+6.1%1,673
2008-39.7%1,008
2009+28.5%1,296

The £100-a-month investor

Now the same decade from the other side. The green line is the portfolio value of someone paying in £100 every month; the grey dashed line is what they had paid in so far. Watch where it finishes.

£4k£8k£12k£12,492paid in £12,0002000200120022003200420052006200720082009£100 a month — portfolio value against money paid in (dividends reinvested)

The portfolio spent 53 of the 120 months below what had been paid in — three unbroken years from late 2000 to late 2003, then again from October 2008 to October 2009. That is what it feels like at the time, and it is why most people stop. The 2009 recovery lifts it clear because so many units had been bought cheaply on the way down.

Side by side

Two people, £12,000 each, the same ten years, every dividend reinvested in both cases.

Paid inEnded withResult
All £12,000 in Jan 2000, then left alone£12,000£10,908−9.1%
£100 a month × 120 months£12,000£12,492+4.1%

A gap of £1,584 on identical money in an identical index — and the difference between finishing behind and finishing ahead. The monthly investor made no decisions after setting it up, never picked a bottom, and never needed a view on anything.

Why the monthly investor did better

A much lower average price

The lump-sum investor bought everything at 1,426. The monthly investor's average cost across 120 purchases was about 1,245 — roughly 13% lower, because the money arrived across the whole decade rather than all at its most expensive moment.

Most of the money arrived after the falls

Barely anything was invested during the 2000–2002 decline. Contributions were still going in through 2008 and into the March 2009 low, when the index was around 39% below where the decade started.

No decision was required

The monthly investor didn't identify a bottom or wait for confirmation. The advantage came from consistency, which is a behaviour rather than a skill.

Worth stating the mechanism precisely, because it's often described wrongly. Contributing during a fall does not “recover” a loss — nothing you add makes an earlier purchase worth more. It lowers your average cost per unit, so the recovery lifts a larger holding. That distinction matters, because “averaging down to get back to breakeven” is exactly the reasoning that keeps people buying a failing individual company.

The honest other side

Monthly investing is not universally better. Over most historical ten-year periods a lump sum has finished ahead, for the simple reason that money invested earlier is invested for longer and markets rise more often than they fall.

The lost decade is the case where lump-sum timing was worst — a peak entry followed by two crashes. What the comparison really shows is not that one method beats the other, but that the sequence of returns matters enormously, and nobody chooses the sequence they get. The practical argument for monthly investing is that it works acceptably in every sequence rather than brilliantly in some and catastrophically in others — and it's what most people can actually do from income.

The version of this decade that really was a disaster

Reverse the direction of the money and the story changes completely. Someone drawing income through 2000–2009 was selling units into two crashes, realising losses to fund living costs and permanently shrinking the capital that had to recover. Analysis of that decade has found a retiree taking an inflation-adjusted 5% a year from an S&P 500 portfolio starting in 2000 could have exhausted it entirely.

Same index, same decade, opposite outcome — decided by which way the money was flowing.

What to take from it

Ten years is not long enough to be sure of anything

The idea that shares always win over a decade is not supported by this one. Plan on longer horizons than you think you need.

Keep contributing when it feels worst

The months that felt like the biggest mistakes — late 2002, early 2009 — bought the units that closed the gap.

Money you need within five years shouldn't be in shares

Because you might be the person forced to sell into the fall rather than through it.

The index isn't the only option

A decade where the index returned nothing was not a decade in which every company returned nothing — see how intrinsic value is calculated and what a value trap is.

Method: monthly average closing prices for the S&P 500 from the Shiller dataset, with dividends accrued evenly at the rate that reproduces the published January 2000 to December 2009 total return of −9.1%. That works out at about 1.6% a year. Figures are in dollars, before fees, taxes and inflation; a sterling investor's experience would also reflect the currency move. Past performance is not a reliable indicator of future results and capital is at risk.

A decade of nothing, company by company

Oak Growth estimates what roughly 1,000 companies are worth across eight markets and shows the gap against today's price — because an index going nowhere never meant every business inside it did.

Explore Oak Growth

Common questions

What was the lost decade in the stock market?

The ten years from January 2000 to December 2009, when the S&P 500 produced a negative total return of about -9% even with every dividend reinvested. It opened at the dot-com peak and contained two crashes, and at the March 2009 low the index was roughly 39% below where the decade began.

Did anyone make money in the lost decade?

Yes — anyone contributing regularly rather than investing a lump sum at the start. £12,000 invested in January 2000 finished at about £10,908, down 9%, while the same £12,000 paid in at £100 a month finished at about £12,492, ahead of everything contributed. Both figures include dividends reinvested.

Why does monthly investing help in a falling market?

Because a fixed monthly amount buys more units when prices are low. Across the decade the monthly investor's average cost was around 1,245 against the 1,426 the lump-sum investor paid — roughly 13% lower — so the eventual recovery lifted a larger holding.

Is monthly investing always better than a lump sum?

No. Over most historical ten-year periods a lump sum finishes ahead, because money invested earlier is invested for longer and markets rise more often than they fall. The lost decade is the case where lump-sum timing was at its worst, which shows that the sequence of returns matters rather than that one method is superior.

Also see: What if you invest £1,000 a month? → · What is compound interest? → · Bull market vs bear market → · Why is the stock market falling? →