Extrapolation bias
Also known as: Representativeness in expectations
Investors expect recent trends to continue, projecting the past straight into the future.
What it means
The tendency to form expectations by over-weighting recent outcomes and extending them forward, so a run of high returns breeds forecasts of more high returns. Rooted in the representativeness heuristic, it leads investors to chase past performance and to misperceive temporary streaks as durable trends. Extrapolation inflates bubbles, depresses neglected assets, and helps explain the value premium and return-chasing in mutual funds. It matters because expectations data confirm that real investors extrapolate even when rational models say they should expect reversion.
How it works
Extrapolation is typically modeled as a weighted average of past returns in which recent observations dominate and older data fade geometrically. A single parameter, the degree of extrapolation, sets how fast a fresh streak takes over the forecast, and it is not constant: Cassella and Gulen estimate that it drifts over time, so markets swing between hard trend-chasing and near-indifference. Barberis and colleagues add that investors often hold two conflicting signals at once, the trend and a nagging sense of overvaluation, and waver over which to trust. That wavering is what lets a bubble keep inflating past the point where value-minded traders expect it to break.
What the evidence shows
The signature finding comes from Greenwood and Shleifer, who assembled six independent series of investor return expectations spanning 1963 to 2011. All six climb after the market has risen and after past returns were high, yet they move opposite to what valuation models predict, since expected returns are low precisely when prices are high. Reported expectations are thus not a noisy copy of rational forecasts; they point the wrong way. Bordalo, Gennaioli, La Porta and Shleifer sharpen the point with analysts: stocks carrying the most optimistic long-term growth forecasts subsequently disappoint and underperform the most pessimistic ones, and the errors are systematic, the fingerprint of beliefs that overshoot.
Where it shows up
Extrapolation is not confined to equities. In housing, mid-2000s buyer surveys recorded expectations that recent gains would keep coming for years, financing purchases on that belief just before the bust. Mutual-fund flows chase past performance, arriving after strong runs and leaving after weak ones, mechanically buying high and selling low. Credit markets follow the same rhythm: tight spreads and easy issuance trail good news and precede rising defaults. IPO and takeover waves cluster after price run-ups, when extrapolated optimism makes financing look cheap. The common thread is a recent trend treated as information about the future when it is mostly noise.
Diagnostic expectations: a sharper account
Newer work reframes extrapolation as diagnostic expectations: forecasters do not mechanically project a line but overreact to news, inflating the probability of whatever the latest data made more likely, a kernel of truth exaggerated. The distinction predicts more than trend-chasing. Because good news is overweighted, beliefs eventually overshoot and then correct, producing predictable reversals and, in credit, boom-bust cycles. It also fits a subtlety in the data: Bordalo and colleagues find individual forecasters overreact even when the consensus looks sluggish. Extrapolation is best read not as naive momentum but as over-sensitivity to recent information whose errors are, in principle, forecastable.
Limits and caveats
Two cautions temper the picture. Extrapolation is not universal: the same surveys contain contrarians, institutions often lean against retail flows, and even trend-followers expect reversal at long horizons, which is why demand can flip. And most evidence rests on survey expectations, so skeptics ask whether reported forecasts drive the trades that move prices or merely track mood. The strongest reply is that these expectations predict actual portfolio choices and later returns, not just talk. Extrapolation is therefore better treated as a strong average tendency with wide individual variation than as a law every investor obeys, a tilt in the crowd, not a switch in every head.
Examples
After several booming years, households pour money into stocks expecting the boom to last — just before returns mean-revert.
After a decade of rising house prices, buyers stretch to the top of what the bank will lend, reasoning that prices only go up and next year's valuation will rescue the mortgage.
A start-up posts three record quarters and hires as though the curve runs on forever, then cuts a third of the staff when demand simply returns to trend.
A worker whose pay has climbed sharply for three years running treats the streak as permanent and stretches into a car loan and rent sized to that trajectory; when the raises stall, the fixed payments no longer fit the paycheck.
An oil producer reads two straight years of high prices as the new normal and sanctions expensive new wells; the added supply lands just as prices sink back toward cost.
First described in Greenwood & Shleifer (2014); Lakonishok, Shleifer & Vishny.
Key references
- Bordalo, P., Gennaioli, N., Ma, Y., & Shleifer, A. (2020). Overreaction in macroeconomic expectations. American Economic Review, 110(9), 2748-2782. doi.org/10.1257/aer.20181219
- Bordalo, P., Gennaioli, N., La Porta, R., & Shleifer, A. (2019). Diagnostic expectations and stock returns. The Journal of Finance, 74(6), 2839-2874. doi.org/10.1111/jofi.12833
- Cassella, S., & Gulen, H. (2018). Extrapolation bias and the predictability of stock returns by price-scaled variables. The Review of Financial Studies, 31(11), 4345-4397. doi.org/10.1093/rfs/hhx139
- Barberis, N., Greenwood, R., Jin, L., & Shleifer, A. (2018). Extrapolation and bubbles. Journal of Financial Economics, 129(2), 203-227. doi.org/10.1016/j.jfineco.2018.04.007
- Greenwood, R., & Shleifer, A. (2014). Expectations of returns and expected returns. The Review of Financial Studies, 27(3), 714-746. doi.org/10.1093/rfs/hht082
- Lakonishok, J., Shleifer, A., & Vishny, R. W. (1994). Contrarian investment, extrapolation, and risk. The Journal of Finance, 49(5), 1541-1578. doi.org/10.1111/j.1540-6261.1994.tb04772.x