After buying another company, does a buyer trail similar companies more when its shares had risen most beforehand?
Study 39 · Inconclusive · recorded 5 October 2026 · answer: Not settled · full paper with figures
The question
In the year after a company completes a purchase of another business, do its shares fall behind similar companies, and if so, is the shortfall largest for buyers whose shares had risen most in the year before, or does it come with the deal whatever came before?
Rules fixed before the result
Fixed before any outcome: the run-up explanation holds if the run-up slope's 95% range lies wholly below zero, the deal explanation if a buyer with no run-up has its range wholly below zero while the slope's range holds zero, both if both lie below zero, and otherwise neither. The study needed at least 100 deals by 75 companies, 30 in each period, and would stop unless a check made without outcomes showed it could detect a 12-point shortfall. It passed.
Data and method
We kept 150 of 205 deals completed late 2013 to 2019 in development data to 2022; most others lacked a full prior year of prices or fell outside the study's categories. Against five same-industry peers of similar trading size, a regression gave a buyer's year-after gap with no prior-year run-up and how the gap moves with that run-up. It was rerun on these buyers' quarterly dates at least a year from their deals; companies were redrawn 2,000 times for ranges.
The answer
The result is inconclusive: neither the buyer's run-up nor the deal itself is established, as we had expected at low confidence. A buyer with no run-up trailed its comparison companies by about 3 percentage points over the year after (95% range: 9 behind to 2.5 ahead); each standard deviation of run-up, about 39 points, went with about 2 points better, not worse (5 worse to 10 better), the opposite direction to the run-up explanation, and both ranges hold zero. On the same buyers' ordinary dates, both measures were near zero. The study was built to detect a 12-point shortfall, which neither range reaches, and could detect about 9 points either way; smaller effects remain compatible, so this is not a statement that nothing is there. The 150 deals by 117 companies met every minimum set in advance.
Main estimate
0.020 fraction (95% range -0.051 to 0.098), from 150 observations
What it changes
No plan changed. Neither the buyer's run-up nor the deal itself becomes a condition in any trading rule, and work on these deals rests where study twenty-three left it, with the deals counted in our records.
What remains uncertain
Study twenty-three counted these deals before; deals on nearby dates share market moves the redrawing ignores. Completion dating blends the run-up with the deal's first market reaction, but a check excluding the half-year before, when most deals are announced, was also inconclusive. Run-up only approximates dear shares; deal size, payment and later years are unstudied; that buyers who had lagged did worse after, the run-up explanation's opposite, is not established, though one of nine side checks pointed that way.
How it was made
Run took 41 minutes; independent code review: approve; the recorded calculation reruns to the same result.
Fingerprints
plan 04ff8344e933a921653c351ee4850fcc73979da1a2149e73c274eb99dbab2ac0 result ff119ed64d90383b107ecc1ee5fccff7ffd4c4e64dad8214fb156878b7af297a
Exploratory work on development data; not a performance claim.