By Stocks dey Fall When Lockup Expire?
Stocks dey lose small ground when lockup expire, studies find. But prospectus date dey give market time to price am in, while trading volume fit rise sharply.
Stocks dey fall when lockup expire? On average, dem dey lose small ground, and e much smaller than wetin common belief dey promise. Lockup expiration na the first date wey company insiders and pre-IPO investors fit sell shares wey dem already hold. Published event studies find small negative average abnormal return around that date, together with big and lasting increase in trading volume. Dem print the date for IPO prospectus months before, so market get all that time to price am in.
Wetin lockup expiration really dey change
Lockup agreement na contract between insiders of newly public company and banks wey underwrite the offering. For fixed period after IPO, usually 180 days and sometimes 90, those holders agree say dem no go sell. When the period end, company no issue new shares and nothing about the business itself change. Wetin change na number of shares wey legally free to trade, the free float. Company wey sell 10 percent of itself for the offering fit reach expiration day with several times that number of shares newly eligible to enter market.
Two things for this setup matter. The possible increase in supply big compared with the float, and anybody wey open the prospectus fit know the date. Those two forces dey pull opposite directions. Na there the simple story differ from the measured result. For details of how the agreement itself work, see our IPO lockup expiration explainer.
Stocks dey fall when lockup expire? Wetin research find
Researchers don study this event well for equity markets because contract set the date, no be news. Field and Hanka (2001, Journal of Finance) examine more than thousand lockup expirations. Dem report statistically significant negative abnormal return concentrated around the date, together with permanent increase in average trading volume. Abnormal return na the part of stock move wey remain after dem subtract wetin the broad market do on the same days.
Bradley, Jordan, Roten and Yi (2001) document the same pattern, with price weakness concentrated for venture-backed IPOs. Brav and Gompers (2003) treat lockup as commitment device and connect size of the expiration effect to company characteristics, rather than to the unlock alone. Ofek and Richardson (2000) interpret the price behaviour as evidence of downward-sloping demand curve for shares. This mean say buyers absorb extra supply only at lower price. Cao, Field and Hanka (2004) test whether insider supply damage liquidity around these dates. Dem find say trading conditions hold up instead of getting worse.
Make you carry two points from all this research. The average effect real but small, around the size wey one normal session fit erase. And the spread of results around that average much wider than the average itself.
Volume na the more dependable part of the pattern
The panel below use 6 US IPOs from 2023 and 2024. E compare average daily share volume for the 30 calendar days before the usual 180 day mark with the 30 days after am. E use the 180 day convention instead of each prospectus contractual date. Data notes at the end explain this deliberate simplification.
The exact SQL behind every number
WITH multiIf(
ticker = 'ARM', toDate('2023-09-14'),
ticker = 'CART', toDate('2023-09-19'),
ticker = 'BIRK', toDate('2023-10-11'),
ticker = 'ALAB', toDate('2024-03-20'),
ticker = 'RDDT', toDate('2024-03-21'),
ticker = 'RBRK', toDate('2024-04-25'),
toDate('2024-01-01')) + 180 AS lockup_mark
SELECT
ticker,
round(avgIf(day_volume, session_date < lockup_mark) / 1e6, 2) AS avg_volume_before_m,
round(avgIf(day_volume, session_date >= lockup_mark) / 1e6, 2) AS avg_volume_after_m,
round(100 * (avgIf(day_volume, session_date >= lockup_mark)
/ avgIf(day_volume, session_date < lockup_mark) - 1), 1) AS volume_change_pct
FROM
(
SELECT
ticker,
toDate(toTimeZone(window_start, 'America/New_York')) AS session_date,
toFloat64(sum(volume)) AS day_volume
FROM global_markets.delayed_stocks_minute_aggs
WHERE ticker IN ('ARM', 'CART', 'BIRK', 'ALAB', 'RDDT', 'RBRK')
AND window_start >= toDateTime('2024-01-15 00:00:00')
AND window_start < toDateTime('2024-12-15 00:00:00')
AND (toHour(toTimeZone(window_start, 'America/New_York')) * 60
+ toMinute(toTimeZone(window_start, 'America/New_York'))) >= 570
AND (toHour(toTimeZone(window_start, 'America/New_York')) * 60
+ toMinute(toTimeZone(window_start, 'America/New_York'))) < 960
GROUP BY ticker, session_date
)
WHERE session_date >= lockup_mark - 30
AND session_date <= lockup_mark + 30
GROUP BY ticker
HAVING countIf(session_date < lockup_mark) > 0
AND countIf(session_date >= lockup_mark) > 0
ORDER BY volume_change_pct DESCThe widest gap for the group belong to ALAB at 13.6%. E move from 3.18 million shares per session to 3.62 million. At the other end of the panel dey ARM at -67.6%. Turnover na where the unlock show most clearly: shares wey no fit change hands before now fit trade, and some of dem dey trade.
Price results dey scatter
Na the same six names and the same windows, but this time measurement na closing price instead of volume.
The exact SQL behind every number
WITH multiIf(
ticker = 'ARM', toDate('2023-09-14'),
ticker = 'CART', toDate('2023-09-19'),
ticker = 'BIRK', toDate('2023-10-11'),
ticker = 'ALAB', toDate('2024-03-20'),
ticker = 'RDDT', toDate('2024-03-21'),
ticker = 'RBRK', toDate('2024-04-25'),
toDate('2024-01-01')) + 180 AS lockup_mark
SELECT
ticker,
round(100 * (argMaxIf(day_close, session_date, session_date < lockup_mark)
/ argMinIf(day_close, session_date, session_date < lockup_mark) - 1), 1) AS return_before_pct,
round(100 * (argMaxIf(day_close, session_date, session_date >= lockup_mark)
/ argMaxIf(day_close, session_date, session_date < lockup_mark) - 1), 1) AS return_after_pct
FROM
(
SELECT
ticker,
toDate(toTimeZone(window_start, 'America/New_York')) AS session_date,
toFloat64(argMax(close, window_start)) AS day_close
FROM global_markets.delayed_stocks_minute_aggs
WHERE ticker IN ('ARM', 'CART', 'BIRK', 'ALAB', 'RDDT', 'RBRK')
AND window_start >= toDateTime('2024-01-15 00:00:00')
AND window_start < toDateTime('2024-12-15 00:00:00')
AND (toHour(toTimeZone(window_start, 'America/New_York')) * 60
+ toMinute(toTimeZone(window_start, 'America/New_York'))) >= 570
AND (toHour(toTimeZone(window_start, 'America/New_York')) * 60
+ toMinute(toTimeZone(window_start, 'America/New_York'))) < 960
GROUP BY ticker, session_date
)
WHERE session_date >= lockup_mark - 30
AND session_date <= lockup_mark + 30
GROUP BY ticker
HAVING countIf(session_date < lockup_mark) > 0
AND countIf(session_date >= lockup_mark) > 0
ORDER BY return_after_pct DESCThe 30 days after the mark carry 51% for ALAB at one end of the panel, and 1.9% for BIRK at the other. Six names na illustration, no be full study. Na the reason wey we show am: variation across small group of ordinary unlocks much bigger than the small average wey research measure across thousand of dem. Anybody wey quote one expected percentage for an unlock dey quote the middle of very wide distribution.
How positioning before the date dey look
Prospectus make the expiration date public, so professional traders fit position before am. One visible sign na short sale volume, meaning the part of each day reported share volume wey marked as short sale. E no be short interest, which na outstanding borrowed-share balance wey dem publish twice every month. Short volume na daily flow measure, and e high for most liquid stocks under normal conditions.
ALAB get the highest pre-mark reading for the group: 58.4% of reported volume marked short during the 30 days before, compared with 48.7% during the 30 days after. Focus on the levels, no be only the change. Short-marked prints already form large part of daily turnover well before the unlock arrive. Na so anticipation dey show for the tape. Borrow market often tighten as these dates approach too. Stock wey costly to borrow na stock wey other traders don already crowd into.
One name, session by session
Averages fit hide the real shape. This one na Reddit across the weeks around its own 180 day mark during autumn 2024. Closing price dey on top, with daily volume underneath.
The exact SQL behind every number
SELECT
toString(toDate(toTimeZone(window_start, 'America/New_York'))) AS session_date,
round(toFloat64(argMax(close, window_start)), 2) AS close,
round(toFloat64(sum(volume)) / 1e6, 2) AS volume_m
FROM global_markets.delayed_stocks_minute_aggs
WHERE ticker = 'RDDT'
AND window_start >= toDateTime('2024-08-12 00:00:00')
AND window_start < toDateTime('2024-10-25 00:00:00')
AND (toHour(toTimeZone(window_start, 'America/New_York')) * 60
+ toMinute(toTimeZone(window_start, 'America/New_York'))) >= 570
AND (toHour(toTimeZone(window_start, 'America/New_York')) * 60
+ toMinute(toTimeZone(window_start, 'America/New_York'))) < 960
GROUP BY session_date
ORDER BY session_dateAcross the 53 sessions wey dey shown, from 2024-08-12 to 2024-10-24, the close move from $52.51 to $78.62. No cliff dey for the middle of the chart. Na the normal case, and normal case still be evidence.
When the move big
The conditions wey researchers and trading desks connect with bigger reaction get one common feature: the unlock big compared with wetin market fit absorb, and holders get clear way to exit.
- Small free float compared with large unlocking block, so the shares released several times bigger than wetin normally trade in one week.
- Ownership concentrated inside few venture or private equity funds near the end of their fund life.
- Stock trading far from IPO price in either direction, which change the gain or loss wey holders dey sit on.
- Thin daily liquidity, where even modest sale program become meaningful part of the tape.
When nothing happen at all
Just as often, the date pass without any visible mark on the chart. The reasons dey structural.
- Insiders don already sell through follow-on or secondary offering, so supply reach market weeks earlier inside one negotiated block.
- Underwriters release the lockup early, fully or partly, as dem get permission to do.
- Agreement release shares in tranches, or contain early-release conditions tied to earnings report or price threshold. So the 180 day date na only one of several release points.
- Biggest holders na strategic holders, no be financial investors, and dem no plan to sell on any particular date.
Checklist for one specific upcoming expiry
- Get the contractual date, no be the 180 day estimate. Dem write am inside underwriting section of prospectus. Our guide on how to find a lockup expiration date show where to look.
- Compare the unlock size with the float: shares becoming eligible divided by shares wey currently trade freely.
- Check whether secondary offering don already move part of that block.
- Read the release conditions for tranches, price thresholds or earnings-linked early releases.
- Check wetin don already position into the date, including short volume share and cost to borrow.
- Note how concentrated the holder list be, and whether those holders don publicly state any exit plan.
For context on how many of these dates dey pipeline this year, see our review of the 2026 IPO market so far.
FAQ
Stocks always dey fall when lockup expire?
No. Published event studies find small negative average abnormal return around lockup expirations across large samples. Individual results spread widely on both sides of that average. Plenty individual unlocks pass without any visible price effect.
How long typical IPO lockup period dey last?
Most US IPO lockups run for 180 days from the offering. 90 day agreements and staggered multi-tranche structures common too. The binding period na wetin the prospectus state for that particular deal, no be the usual convention.
Wetin happen to trading volume when lockup expire?
Volume na the most consistent part of the pattern. Field and Hanka (2001) find permanent increase in average trading volume after expiration. The panels above show turnover stepping up across most names for the sample window.
Lockup fit lift before expiration date?
Yes. Underwriters fit waive lockup early, fully or partly. Many agreements also get early-release conditions tied to earnings report or price threshold. Na one major reason why date calculated as "IPO plus 180 days" fit miss completely.
High short volume before expiration date unusual?
No be by itself. Short-marked prints form large part of daily reported volume for most liquid stocks under normal conditions. So compare the level before unlock with that stock own baseline, no be with zero.
Data notes and method
The three cross-sectional panels use the 180 day convention measured from each company IPO date, wey visible inside the SQL. Na simplification: contractual expiration inside prospectus fit differ, fit come in tranches, or underwriters fit waive am early. Treat the panels as illustration of how the window behave, no be record of six contractual dates.
Sessions come from regular-hours minute bars converted to New York time, so extended-hours prints no dey included. Short volume figures come from daily reported short sale volume, deduplicated per ticker and date. Short volume measure marked sell orders inside one day reported flow. E no be short interest, and short-marked print fit close out minutes later.
The academic findings we mention come from Field and Hanka (2001), Bradley, Jordan, Roten and Yi (2001), Ofek and Richardson (2000), Brav and Gompers (2003), and Cao, Field and Hanka (2004). We describe magnitudes by direction only here because the samples and windows for those papers differ from the window we use for these panels.
Every panel for this page come with the SQL wey produce am, so you fit check the windows and ticker list line by line. To run the same before-and-after comparison on an unlock wey you dey watch, ask am in plain English for the Strasmore terminal.