The Disciplines of M&A
What a deal acquires when it crosses a border: extra regulators, extra currencies, extra employee consultation, and a documentary record that has to work in two legal traditions at once.
Cross-border M&A is the acquisition of a company by a buyer domiciled in a different country. The defining feature is not distance but the number of legal, tax and regulatory systems that must simultaneously say yes. A domestic acquisition clears one merger control regime, one company law, one employment code and one tax authority. A cross-border acquisition may clear a dozen of each, with different standards of proof and no coordinating body.
The two most-cited sources measure different things. UNCTAD's World Investment Report and Global Investment Trends Monitor report net cross-border M&A sales — foreign purchases less foreign-owner divestitures, on a completed basis, as an FDI component. Commercial databases such as LSEG (formerly Refinitiv) and Dealogic report gross announced deal value. The two series can differ by an order of magnitude in the same year; the difference is methodological, not a discrepancy to be reconciled.
On UNCTAD's basis, cross-border M&A totalled US$403 billion in 2025 across 6,823 deals — down 10 per cent in value and 8 per cent in count on 2024. Developed economies accounted for US$389 billion (down 8 per cent), developing economies for US$14 billion (down 47 per cent). Global FDI flows that year were US$1.6 trillion, up 14 per cent (UNCTAD, January 2026).
On the commercial-database basis the numbers are far larger. LSEG recorded US$1.05 trillion of cross-border M&A in January–July 2026, roughly 33 per cent of the US$3.19 trillion global total and the highest January-to-July cross-border figure since 2007 (LSEG). BCG puts the cross-border share of global deal value at close to 50 per cent at the 2007 peak and around 30 per cent in 2024, with 2024 activity at 0.7 per cent of global GDP against a long-run average of 1.3 per cent (BCG, 2025).
The single-market programme, the run-up to monetary union and the privatisation of state telecom and utility monopolies produced the first genuinely large European cross-border wave. Its emblem is Vodafone's acquisition of Mannesmann, agreed on 4 February 2000 after a hostile campaign — an unsolicited bid for a German company being, in Goldman Sachs's description, "unprecedented at the time." Goldman puts the value at more than US$190 billion (Goldman Sachs); other accounts cite US$180–183 billion, the variation reflecting share-price movement between announcement and completion. It remains the largest acquisition on record.
Cross-border M&A reached US$1,637 billion in 2007 on UNCTAD's then-current basis, 21 per cent above the previous record set in 2000, alongside record FDI inflows of US$1,833 billion (UNCTAD). This was the high-water mark for cross-border activity as a share of global M&A; the credit crisis ended it abruptly.
Beijing loosened outward FDI restrictions in 2014 and Chinese acquirers moved rapidly into European and American assets. Rhodium Group records US$140 billion of completed Chinese outbound M&A in 2016, almost double the previous record, against an annual range of US$30–60 billion in 2010–15 and under US$10 billion in 2005. China's share of global cross-border M&A value rose from 1 per cent in 2007 to 14 per cent in 2016, second only to the United States at 19 per cent (Rhodium Group). The landmark transaction was ChemChina's US$43 billion acquisition of Syngenta, cleared by CFIUS in August 2016 and completed in 2017.
The reversal was equally sharp. Chinese FDI into the United States fell from US$46 billion in 2016 to under US$10 billion by 2019 and less than US$5 billion in 2022, averaging US$667 million a year over 2019–22 (Rhodium Group). Two forces acted together: Beijing's 2017 tightening of capital controls and crackdown on leveraged private acquirers, and Washington's tightening of screening through FIRRMA in 2018, followed by export controls and the investment restrictions in the CHIPS Act and Inflation Reduction Act.
Cross-border activity in 2025–26 is recovering in absolute terms while remaining below its historic share. Composition has shifted: activity concentrates in transatlantic and intra-European flows, with Chinese outbound largely absent from developed-market targets and Asia-Pacific volumes down 8 per cent in the first seven months of 2026 (LSEG).
More than 130 countries and transnational organisations operate merger control regimes, with further adoptions expected in Africa, the Middle East and Asia (American Bar Association). Thresholds are usually turnover- or asset-based and calculated differently in each regime, so establishing where a filing is required is itself a diligence exercise. Most regimes are suspensory: closing before clearance is gun-jumping and attracts fines. A single unresolved filing in a jurisdiction contributing trivial revenue can hold the entire transaction.
Substantive risk concentrates in a few large regimes. Adobe's US$20 billion acquisition of Figma was abandoned on 18 December 2023 after the UK Competition and Markets Authority and the European Commission indicated no clear path to clearance; the CMA's contemplated remedies "would effectively amount to a prohibition" (Gibson Dunn). Adobe paid Figma a US$1 billion termination fee (CNBC).
Investment screening is now the layer that most often changes outcomes. In the EU, 24 Member States had screening mechanisms in place at the time of the Commission's fifth annual report, covering 2024. National authorities reviewed 3,136 transactions; 41 per cent (1,286) went to formal screening, of which 86 per cent were cleared unconditionally, 9 per cent with conditions, 4 per cent withdrawn and 1 per cent prohibited. US acquirers accounted for roughly 30 per cent of screened acquisitions (White & Case).
The UK's National Security and Investment Act regime received 1,324 notifications in its 2025-26 reporting year, called in 60 acquisitions, made 9 final orders and blocked one. Of called-in deals, 30 per cent involved Chinese acquirers and 23 per cent US acquirers; the median from call-in to final order was 69 statutory days (UK Government).
CFIUS received 116 declarations and 209 written notices in calendar 2024. Twenty-three per cent of notices were withdrawn or abandoned, up from 19 per cent in 2023; 55 per cent proceeded to a second-stage investigation. China led acquirer countries with 26 notices, followed by France and Japan at 23 each, and CFIUS made 76 formal inquiries into non-notified transactions (White & Case).
Where the target holds controlled technology, transferring it to a foreign parent — including granting access to foreign-national employees — can itself be a licensable export. Sanctions diligence extends beyond the target's compliance to its counterparties and supply chain, and post-closing the acquirer inherits the exposure.
Between signing and closing, a buyer paying in a foreign currency carries unhedged exposure that can run several months and, where regulatory approvals are involved, for an unknown duration. A conventional forward locks a rate but must be unwound at market cost if the deal fails. The market answer is the deal-contingent hedge: a forward or option that voids automatically if the specified conditions — antitrust clearance, completion — are not met, so the buyer pays nothing if the transaction dies. Cost is embedded in the rate rather than paid as an upfront premium, optimising "costs compared to 'full' options" (Natixis). The structure has extended from FX into interest rate, commodity and inflation risk.
Debt raised in one currency to fund an acquisition in another creates a structural mismatch, managed through currency swaps or by borrowing in the target's currency. Guarantee and security packages run into local law limits — financial assistance rules, corporate benefit requirements, thin capitalisation — determining which subsidiaries can guarantee the debt and when security can be granted. Extracting cash from an acquired foreign subsidiary triggers withholding tax on dividends, interest and royalties unless a treaty or directive reduces it; this is the primary driver of holding-company jurisdiction choice.
In the EU, the Acquired Rights Directive (2001/23/EC, implemented in the UK as TUPE) provides that on a transfer of an undertaking, employment contracts transfer automatically on existing terms, dismissal by reason of the transfer alone is prohibited, and both transferor and transferee must inform and consult employee representatives. The regime bites on asset and business transfers rather than share sales, making carve-outs materially more complex than share acquisitions in Europe. Works council obligations in Germany, France and the Netherlands add fixed periods to the timetable and, in some cases, a duty to consult before signing rather than after.
Populating a data room with personal data and transferring it to an acquirer outside the EEA is a restricted transfer under Chapter V of the GDPR. Following the CJEU's Schrems II judgment, reliance on Standard Contractual Clauses requires a transfer impact assessment of the destination country's legal regime, with supplementary measures where protection falls short (Pinsent Masons). Anonymisation and staged disclosure are the usual responses.
Acquiring a company in a higher-corruption jurisdiction imports its conduct history, and under the FCPA and the UK Bribery Act an acquirer can inherit liability for pre-closing conduct. The US Department of Justice's M&A Safe Harbor Policy, announced 4 October 2023, offers a presumption of declination where an acquirer discloses criminal misconduct within six months of closing and fully remediates within one year, extendable for transaction complexity. Companies that fail to conduct effective diligence face "full successor liability for that misconduct under the law," and individuals at the acquired entity remain prosecutable regardless (Steptoe).
The available structures are set by the target's jurisdiction of incorporation, not the buyer's. The scheme of arrangement, used in the UK, Ireland, Hong Kong, Singapore and Australia, is proposed by the target, voted on by shareholders and sanctioned by a court; approval binds all shareholders, delivering 100 per cent ownership in one step, but it is target-cooperative by construction and unsuitable for hostile approaches. The contractual tender offer is the hostile route in the UK and the default in Japan and much of continental Europe: the bidder offers directly to shareholders, with squeeze-out requiring a statutory threshold — 90 per cent in the UK. The statutory merger is the US structure, approved by target board and shareholders, commonly as a reverse triangular merger; Delaware's Section 251(h) allows a tender offer followed immediately by a back-end merger without a second vote.
The Cross-Border Mergers Directive (2005/56/EC), later codified into Directive (EU) 2017/1132, created the first workable procedure for merging limited liability companies incorporated in different Member States, resting on a pre-merger certificate issued by each company's home authority. Directive (EU) 2019/2121, the Mobility Directive, extended the framework to cross-border conversions (redomiciliation) and divisions, adding an anti-abuse review, shareholder exit rights, creditor safeguards and strengthened employee information and participation provisions. Implementation has been uneven; EY maintained a country-by-country tracker into 2025 precisely because transposition ran late in a number of jurisdictions (EY).
Where two companies of similar size in different countries wished to combine, the dual-listed company (DLC) structure let them equalise economics by contract while each shareholder base retained shares in its own company. The Reserve Bank of Australia's 2002 survey identified six: Royal Dutch/Shell (from 1903), Unilever (1930), Reed Elsevier (1993), Rio Tinto (1995), BHP Billiton (2001) and Brambles (2001). The motivations were avoiding capital gains tax that a conventional merger would crystallise, preserving national identity where the transaction was politically sensitive, and avoiding change-of-control triggers in debt and joint venture documents. The drawbacks proved decisive: price divergence between the twin listings "can be very large and highly variable," with premiums and discounts exceeding 30 per cent, alongside dual regulatory compliance and split liquidity (RBA). Most have since unified — Unilever, Shell and BHP among them — and Rio Tinto has faced sustained shareholder pressure to do the same (Rio Tinto).
Luxembourg, the Netherlands, Ireland, Singapore and Delaware recur as holding-company jurisdictions for consistent reasons: a participation exemption removing tax on dividends and capital gains from qualifying subsidiaries, low or eliminable withholding tax on distributions upward, a wide treaty network, and predictable corporate law. Luxembourg's SOPARFI exempts dividends and capital gains where the holding is at least 10 per cent of share capital (or EUR 1.2 million acquisition cost) held for twelve months, applies no withholding tax to interest or liquidation proceeds, and sits within roughly 75 covered tax treaties. The OECD Multilateral Instrument's principal purposes test and domestic anti-abuse rules now deny treaty access to arrangements lacking genuine commercial purpose, and financing companies must show real local presence through qualified personnel, resident directors and local decision-making (Elvinger Hoss Prussen). Structuring has shifted from treaty positioning towards substance.
Under the UK Takeover Code, a bidder announcing a firm offer under Rule 2.7 must have its financial adviser confirm that resources are available to satisfy full acceptance in cash. UK public offers therefore carry no financing condition: funding must be committed on a "certain funds" basis before announcement, with the lender's conditionality to draw reduced to a short list of matters within the bidder's control. US merger agreements routinely proceed on debt commitment letters with wider conditionality, and failure to fund is addressed through a reverse termination fee and specific performance provisions rather than a pre-announcement funding certificate.
The 2011 UK Takeover Code reforms, effective 19 September 2011 and prompted directly by the Kraft–Cadbury outcome, prohibited offer-related arrangements including break fees, non-solicitation undertakings, matching rights and exclusivity, subject to narrow exceptions of normally no more than 1 per cent of offer value in formal sale processes or for a competing white knight (CMS). The same reforms cut the "put up or shut up" period to 28 days from public identification of a bidder, required detailed financing disclosure even for all-cash offers, and made statements about employees and post-offer intentions binding for twelve months.
In the United States, where deal protections remain permissible, reverse termination fees are the standard instrument for allocating regulatory risk, sized against the probability and consequence of failure. The US$1 billion Adobe paid Figma, on a US$20 billion transaction, is a benchmark for a deal with identified competition exposure in multiple jurisdictions.
Material adverse change clauses are read very differently on each side of the Atlantic. Delaware's standard, developed through IBP v Tyson, Hexion v Huntsman and finally Akorn v Fresenius in 2018 — the first case in which the Court of Chancery found an MAE permitting termination — requires a decline in earnings power that is durationally significant, measured over a commercially reasonable period of years rather than quarters. English courts have had far fewer opportunities; in Travelport v WEX [2020] EWHC 2670 (Comm) the Commercial Court construed pandemic-related carve-outs on ordinary contractual principles, without importing the Delaware gloss. The same words therefore carry different content depending on governing law, and drafting cannot be lifted between systems.
Cross-border private acquisition agreements are commonly governed by English or New York law regardless of where the parties sit. Where one party sits in a jurisdiction whose courts the other will not litigate in, or where judgment enforcement would be difficult, arbitration under ICC, LCIA or SIAC rules is preferred, relying on the New York Convention for enforcement of awards.
European private M&A favours the locked box: price fixed by reference to a historical balance sheet date, with seller covenants against value leakage between that date and completion. US practice favours completion accounts — a post-closing true-up against actual cash, debt and working capital. The CMS European M&A Study, covering 582 European private deals in 2024, records rising use of purchase price adjustments and earn-outs as buyers gained leverage, and notes locked box popularity rose between 2010 and 2019 in seller-friendly conditions; a headline locked box percentage was not published in the material reviewed here (CMS).
W&I (in the US, representation and warranty) insurance originated in Europe and is now standard on both sides. The ABA's 2025 Private Target Deal Points Study, covering 139 agreements from 2024 and Q1 2025, found RWI referenced in 63 per cent of US deals, up from 55 per cent (ABA). CMS records an 8 per cent increase in W&I use across Europe in 2024, and 14 per cent in German-speaking markets, attributing it to falling premiums and mid-market adoption. Market-by-market penetration rates on a comparable basis could not be verified from the sources consulted.
The evidence does not support a simple claim that cross-border deals underperform domestic ones. BCG's 2025 analysis of two-year relative total shareholder return found domestic deals at approximately –0.9 per cent, intra-regional cross-border deals at +1.2 per cent, and inter-regional deals at approximately +0.6 per cent — cross-border deals outperformed domestic ones on average, with regional proximity helping. European acquirers were the strongest cross-border performers; Asia-Pacific and North American acquirers found it harder (BCG). Earlier academic work, notably Moeller and Schlingemann's 2005 study in the Journal of Banking & Finance, found lower announcement returns for US acquirers in cross-border deals; the abstract was not accessible through the sources consulted here, so the magnitude is not stated.
Failure, where it occurs, concentrates in governance and integration rather than strategy. Daimler-Benz acquired Chrysler for US$36 billion in 1998, presented as a merger of equals; in 2007 DaimlerChrysler sold 80.1 per cent of Chrysler to Cerberus for US$7.4 billion, retaining 19.9 per cent, and took a charge of up to US$5.4 billion against 2007 profit (NBC News). The gap illustrates cross-border integration risk: incompatible governance traditions, a supervisory-board structure unfamiliar to the American side, and a framing that obscured which party was in control.
Carve-outs across borders are the hardest category. BCG's review of more than 50 divestitures found one-time separation costs of 1 to 5 per cent of the divested business's revenues, reaching up to 13 per cent in demanding cases, proportionally higher for smaller units because separation effort is broadly fixed (BCG). Across borders the drivers multiply: new legal entities in each operating country, employees transferred under local acquired-rights regimes, licences and permits reissued rather than transferred, and shared IT estates separated jurisdiction by jurisdiction. Transitional service agreements bridge the gap, the seller providing HR, IT and accounting services post-closing at cost, cost-plus or fixed per-unit rates. Duration is contested because "the buyer often does not know exactly how long it will need the services" (Mayer Brown); TSAs that run long produce stranded costs at the seller and dependency at the buyer.
Kraft-Cadbury (2009–10). Kraft's offer went unconditional on 2 February 2010. On 9 February Kraft reversed its stated position on keeping the Somerdale facility open, having repeated that intention across four announcements from September 2009. The Takeover Panel publicly criticised Kraft for breaching Rule 19.1, which requires offer documents to be prepared to the highest standards of care and accuracy (Takeover Panel). The episode produced the 2011 Code reforms described above.
Pfizer-AstraZeneca (2014). Pfizer's final proposal, valuing AstraZeneca at approximately £69 billion (about US$118 billion), was rejected by the AstraZeneca board on 19 May 2014; Pfizer walked away on 26 May under the Code's deadline (NBC News). Tax domicile was a stated motivation, and UK political concern about research commitments featured heavily in the debate.
Nippon Steel-US Steel (2023–25). Announced December 2023. President Biden prohibited the transaction by executive order on 3 January 2025; President Trump permitted it on 13 June 2025 subject to a National Security Agreement, and it closed on 18 June 2025. The agreement requires US incorporation and Pittsburgh headquarters, a majority-US board, US-citizen CEO and key management, and no transfer of jobs abroad; the US government holds a golden share giving it one director appointment and presidential consent rights over capital reductions, redomiciliation, headquarters relocation and facility closures. Nippon Steel committed approximately US$11 billion of new US investment by 2028 (Hunton).
Broadcom-Qualcomm (2017–18). Broadcom's unsolicited approach, valued at over US$100 billion, was prohibited by presidential executive order on 12 March 2018 on CFIUS recommendation — only the fifth such presidential block in thirty years. Broadcom's Singapore incorporation made it a "foreign person" within CFIUS jurisdiction; when it sought to redomicile to the United States to defeat that jurisdiction, CFIUS prohibited the corporate actions required to do so. The stated concern was that the combination would curb US 5G investment to Huawei's advantage (Goodwin).
Daimler-Chrysler and Vodafone-Mannesmann are covered above.