Glossary
Recency bias
Giving the most recent events too much weight and assuming that what just happened will keep happening.
Recency bias is the tendency to expect the near future to look like the recent past. After several good years investors expect good years. After a crash they expect more falls. The long record, which contains both, gets less weight than the last few months.
What it leads to
- Performance chasing. Money flows into funds and sectors after they have risen. Studies of fund flows find that the average investor in a fund often earns less than the fund itself, because purchases cluster after gains and sales after losses.
- Shifting risk appetite. Some studies of risk tolerance find that people report somewhat more tolerance after markets have risen, as in a long bull market, and less after a fall, which is the reverse of what would help.
- Abandoning a plan. An allocation chosen calmly is dropped when one part of it lags for a year or two.
- Forgetting rare events. A long quiet period makes a crash feel impossible. A recent crash makes one feel imminent.
Recency and momentum
Recent winners do tend to keep winning for some months, which is the momentum effect. Recency bias differs in that it projects recent results far into the future and ignores mean reversion over longer periods.
What counters it
Fixed rules such as scheduled rebalancing are designed to act against recent performance. Looking at returns over full market cycles, and not the last one to three years, serves the same purpose.
