What Happens to Acquirer and Target Stocks After M&A Filings

We ran an event study on 2,069 M&A announcements from 2000 to 2025. The finding that surprised us most: the acquirer positive signal that shows up in same-day close data vanishes once you enter the next trading day. Here's what the data actually shows, and why the execution model changes everything.

Contents

1. Method
2. What We Found
3. The Data
4. Limitations
5. What This Tells You
6. Screen for Current M&A Activity
7. References

Data: FMP financial data warehouse, 2000–2025. Updated March 2026.

Method

We used the mergers_acquisitions_latest table, which contains SEC-sourced M&A filing data from FMP. Each row is an acquirer reporting a target company to regulators.

A few things to know about the data before we get to results.

It's SEC filing data, not press announcements. The transactionDate field is when the deal was filed with the SEC, which may be days or weeks after the press release. By the time a filing hits, some of the initial price reaction may already be in.

It needs deduplication. The average deal generates 2.9 filings because different share classes of the same acquirer file separately. We collapse to one event per (symbol, transaction date) pair.

Coverage is selective. Not every public M&A deal appears in this dataset. Coverage improved after 2016. About half of target companies have no US price data because they're private or foreign. We only include targets with price data in the target pool.

No deal price or terms. This is the biggest limitation. We can't compute deal spreads, premium percentages, or deal-type breakdowns. This is a post-announcement return study, not a traditional merger arbitrage analysis.

Execution model. We use next-day-close entry (MOC execution): the baseline price is the adjusted close on the first trading day AFTER the filing date. This is the price you'd realistically get if you see a filing at end of day and enter at next day's close. The windows T+1, T+5, T+21, T+63 are measured from that baseline.

The final dataset: 441 target events and 1,628 acquirer events, spanning 2000 through 2025. We measure cumulative abnormal returns (CAR) at T+1, T+5, T+21, and T+63 trading days. Abnormal return means the stock's return minus SPY's return over the same window. We use winsorization at the 1st/99th percentile to limit the impact of extreme outliers.

What We Found

Acquirers: no significant drift after next-day entry.

In earlier versions of this study using same-day-close as baseline, acquirers showed +0.28% at T+1 (significant). Once we shift to next-day-close entry, that signal disappears.

What happened to the old positive signal? The filing-day close already incorporated the market's reaction to the announcement. By the time you can act (entering the next trading day), that reaction is priced. Acquirers show a positive reaction ON the filing day, but no additional drift after.

This doesn't validate the "acquirer curse" either. The long-term numbers (-0.45% at T+21, -0.77% at T+63) are negative but not statistically significant. Acquirers are essentially flat in aggregate from the day after the filing onward.

Targets: consistent negative drift from T+5.

Targets tell a different story. The filing-lag argument explains the muted T+1 (-0.22%, not significant). But T+5 onward is significant.

The T+63 target number requires the same survivorship caveat as before. We had 441 target events at T+1. By T+63, only 348 remain. That's 93 events, or 21% of the sample, that dropped out. Most of those are successfully completed deals where the target was delisted after the acquisition closed.

So the targets still trading at T+63 are disproportionately those where the deal stalled, faced regulatory challenges, or fell apart. The -3.36% is measuring the "troubled deal" population, not M&A targets in general.

But the T+5 and T+21 survivorship bias is smaller. At T+5, we still have 439 of 441 events. The -0.63% there is a genuine post-filing drift, not a survivorship artifact.

Side-by-side comparison

Statistically significant at p<0.05

The Data

* = significant at p<0.05

Note: Target T+63 n=348 (93 events delisted, mostly completed deals). Acquirer n=1,593 at T+63.

The target signal is the cleaner finding here. Persistent negative drift from T+5, spanning 441 events across 25 years, with a t-stat of -2.65 at the 5-day window where survivorship bias is minimal.

Annual event counts

M&A activity follows the credit cycle. The 2009 financial crisis nearly stopped deals entirely (17 total events). Activity rebuilt through the 2010s and peaked in 2021 at 282 events, driven partly by the SPAC-driven deal surge. It's run at 130-170 events per year since.

Limitations

The filing-vs-announcement gap. The most important limitation. transactionDate is systematically days or weeks after the press announcement. The T+1 through T+5 results are measuring post-filing drift, not announcement reaction. The acquirer non-result at short windows likely reflects the fact that any positive reaction on press day has already resolved.

No deal terms. We can't split by cash vs stock deals, premium size, deal size relative to acquirer, or hostile vs friendly. All of these affect how markets react. We're averaging across a heterogeneous population.

No deal outcome tracking. Failed deals get pooled with completed deals in the T+1 through T+21 windows. For targets, failed deals produce extreme negative returns. The T+5 negative signal (-0.63%) is a mix of deal-uncertainty drag and actual deal failure.

Coverage selectivity. Not every M&A deal is in this dataset. Coverage expanded sharply after 2016 (2021 has 6x more events than 2010). We don't know what systematic differences exist between covered and uncovered deals.

SPY as benchmark. SPY captures broad market movement but doesn't control for sector or size. An acquirer in a bull tech market will look artificially good vs SPY even if its M&A reaction was flat.

What This Tells You

For anyone holding an acquirer stock when a deal is announced: based on 25 years of data, if you can enter at the filing-day close, acquirers show a slight positive reaction on that day. If you enter the day after, the signal is gone. The "acquirer drops on deal news" narrative doesn't hold in aggregate for large-cap US acquirers, but neither does a persistent positive drift.

For anyone holding a target: stocks drift negative starting around 5 trading days after the filing. This is partly deal uncertainty (companies in limbo tend to drift lower) and partly the survivorship artifact at longer horizons.

The more actionable takeaway is negative: the T+5 acquirer signal that research has historically cited as exploitable doesn't survive when you use tradeable execution timing. What's left is a neutral acquirer signal and a mildly negative target signal explained partly by deal mechanics.

Screen for Current M&A Activity

To see deals filed in the last 90 days, run this query on Ceta Research:

WITH recent AS (
    SELECT
        symbol AS acquirer,
        targetedSymbol AS target,
        companyName AS acquirer_name,
        targetedCompanyName AS target_name,
        CAST(transactionDate AS DATE) AS deal_date,
        ROW_NUMBER() OVER (
            PARTITION BY targetedSymbol, CAST(transactionDate AS DATE)
            ORDER BY acceptedDate DESC
        ) AS rn
    FROM mergers_acquisitions_latest
    WHERE CAST(transactionDate AS DATE) >= CURRENT_DATE - INTERVAL '90' DAY
      AND targetedSymbol IS NOT NULL AND TRIM(targetedSymbol) != ''
      AND NOT (symbol LIKE '%-WT' OR symbol LIKE '%-WS'
               OR (symbol LIKE '%W' AND LENGTH(symbol) > 5))
)
SELECT acquirer, target, acquirer_name, target_name, deal_date
FROM recent
WHERE rn = 1
ORDER BY deal_date DESC
LIMIT 30

Run this query

References

• Mitchell, M. & Pulvino, T. (2001). "Characteristics of Risk and Return in Risk Arbitrage." Journal of Finance, 56(6), 2135-2175.
• Baker, M. & Savasoglu, S. (2002). "Limited Arbitrage in Mergers and Acquisitions." Journal of Financial Economics, 64(1), 91-115.
• Roll, R. (1986). "The Hubris Hypothesis of Corporate Takeovers." Journal of Business, 59(2), 197-216.

Data: Ceta Research data warehouse (FMP/SEC-sourced M&A filing data). 2,069 events, 2000-2025. US stocks only (NYSE/NASDAQ/AMEX), market cap above $1B. CAR = cumulative abnormal return vs SPY. Winsorized at 1st/99th percentile. transactionDate = SEC filing date, not necessarily press announcement date. Entry: next-day close after filing (MOC execution). This is educational content, not investment advice.