The Lost Decade
The idea of the “lost decade” has become one of the most widely repeated narratives in investing. The claim is simple and striking: from 2000 to 2009, the S&P 500 delivered essentially zero returns. At face value, it suggests that an entire decade of equity investing was wasted. But, like most clean, compelling narratives in markets, this one falls apart under closer inspection.
The first issue is the framing itself. The so-called lost decade begins at one of the most extreme valuation peaks in modern market history, the dot-com bubble of early 2000, and ends near the depths of the global financial crisis in 2009. In other words, the measurement period is anchored at a market top and concludes at a market bottom. This introduces a severe timing bias.
Why does this matter? Because market returns are highly sensitive to starting valuations. Beginning at a peak compresses forward returns, while ending at a trough understates the recovery that typically follows. The “lost decade” is therefore less a reflection of the underlying return-generating ability of equities and more a function of where the clock starts and stops.
When we adjust for this timing bias, the narrative changes meaningfully. Shifting the start date even slightly, away from the absolute peak, or extending the end date to include the early recovery period, produces dramatically different results. If an investor had bought near the bottom of the dot-com crash in 2002, when the S&P 500 was roughly in the 800–900 range, and sold near the pre-crisis peak in 2007, around 1,500, they would have seen a gain of roughly 70–90%. Annualized, that’s around 11–13% per year. Returns begin to normalize when timing bias is removed, and the idea of a “lost” decade is refuted.
More broadly, the fixation on this period highlights a common cognitive trap in investing: the tendency to overemphasize discrete time windows. Markets do not operate in clean ten-year blocks, and performance rarely conforms to calendar boundaries. By isolating a particularly unfavourable slice of history, we risk drawing conclusions that do not generalize. The S&P 500 did not suddenly stop working for ten years; it simply went through normative cyclical movement.
In that sense, the “lost decade” reveals the importance of context. It reminds us that returns are more dependent on timing, time invested, and amount invested rather than depending on arbitrarily drawn windows of market activity. Once we strip away the timing bias, what remains is far less dramatic: a period of volatility, two major drawdowns, and ultimately, a continuation of the long-term behaviour of equities.