When a payout comes off a share, does its price fall by less than the payout?
Study 71 · Changed a rule or plan · recorded 26 October 2026 · answer: Not at the size tested · full paper with figures
The question
A share's price should drop by about the payout on the first day it trades without it, and an earlier look at 2013 to 2017 found it dropped by less, about a tenth of a percentage point. We compare each paying share's return on that day with the same day's average for similar paying shares with no payout nearby.
Rules fixed before the result
Fixed before any outcome: each payout-day return was compared with the same day's average for paying shares of similar price and trading value with no payout nearby, the answer read on the gap's 95% range against a tenth of a point. A range wholly above zero reaching a tenth meant a shortfall; one including zero and inside minus a tenth to a tenth meant none at that size. It required 2,000 payout days, 300 paying shares and 40 months.
Data and method
Shares priced at five or more as traded the day before were followed on their payout days in 2018 to 2022; the ranges allow for days sharing a session and for a share's repeated payouts. Before any return in those years was formed, the earlier years set the expected precision: the smallest detectable gap, 0.061 points, was within the tenth, so the plain average decided. The same shares' other sessions, two to ten either side, served as a check.
The answer
Not at the size tested. Over 8,579 payout days of 622 paying shares in 2018 to 2022, the return net of similar paying shares averaged -0.002 percentage points, with a 95% range from -0.052 to 0.048, below the tenth of a point set in advance; smaller shortfalls remain compatible with this range. The typical price fall was 1.007 times the payout (0.949 to 1.062). From the previous close to the opening price the share gained 0.088 points on similar shares (0.062 to 0.114), and the rest of the day's return was -0.087 (-0.133 to -0.041), so the gain was traded away within the day. Against the market fund alone the day's return was -0.028 (-0.093 to 0.036).
Main estimate
-0.002 percentage points (95% range -0.052 to 0.048), from 8579 observations
What remains uncertain
Earlier studies used these years, so this is a check within our development data, not a test on fresh data. For 2,635 payout days the price data's own adjustment did not match the payout within 2%, more than the one in ten the plan allowed, so the dates or amounts carry some doubt; where they matched, the gap was -0.008 (-0.057 to 0.041). Taxes, quotes and holders are not observed, so the causes of any shortfall cannot be told apart.
How it was made
Run took 63 minutes; independent code review: approve.
Exploratory work on development data; not a performance claim.