Driven by national security concerns, a growing number of countries screen foreign investment in an expanding range of sectors, yet evidence on the economic effects of such screening is scarce. This column uses new sector-level data for 43 OECD and EU countries to show that screening reduces the number of cross-border acquisitions by 11–18% but leaves the total value of inward investment unchanged. There is evidence for re-structuring of deals: minority stakes fall while controlling acquisitions rise. The effects are not concentrated on investors from China or other politically non-aligned countries.
In June 2026, the EU adopted a revised regulation on the screening of foreign investment. From January 2028, all member states are required to review foreign acquisitions in at least a common set of sensitive sectors, from defence and critical infrastructure to semiconductors, quantum technologies, artificial intelligence, and finance. Moreover, screening will extend to EU-based investors ultimately controlled by individuals or entities from outside the Union (European Union 2026). This completes a long phase of policy development and practice. Almost two decades ago, Véron and Röller (2008) argued here on VoxEU for a common European framework to assess foreign takeovers on security grounds. Today, screening has become a core instrument of the EU’s economic security agenda.
The EU is not alone. Screening foreign investment has been increasingly implemented across advanced economies (OECD 2025). Existing studies show that the introduction of screening and its restrictiveness are associated with exposure to Chinese investment and geopolitical alignment with the US (Bauerle Danzman and Meunier 2023, Bencivelli et al. 2023). Screening is part of a broader shift towards geoeconomic fragmentation, which likely imposes costs on the world economy (Aiyar et al. 2023). Foreign direct investment is increasingly flowing along geopolitical lines (Boeckelmann et al. 2024, Grover and Vezina 2025). Yet, while the political drivers of screening are well studied (Eichenauer et al. 2021, Chan and Meunier 2022, Bauerle Danzman and Meunier 2023, Bencivelli et al. 2023), evidence on its economic effects is scarce and mostly limited to single countries (Connell and Huang 2014, Frattaroli 2020, McCalman and Walter 2026). As screening has become the norm across advanced economies, a central question is whether it affects only the security-threatening investments it targets, or the welcome type as well.
In a recent paper (Eichenauer and Wang 2026), we provide the first cross-country causal evidence on this question.
We hand-code the introduction and sectoral coverage of investment screening in 35 critical sectors for 43 OECD and EU countries from 2007 to 2022. Our sector definitions follow the PRISM dataset of Bauerle Danzman and Meunier (2023). In 2007, 11 of these countries screened foreign acquisitions in at least one of those sectors. By 2022, 29 did, and the number of screened country-sector pairs rose from 47 to 389 (Figure 1).
Figure 1 Investment screening is spreading across countries and sectors, 2007–2022
Screening has also moved well beyond its traditional domains (Figure 2). In 2007, it was concentrated in traditional security sectors such as defence and essential services such as energy, telecommunications, transport, and water infrastructure. By the end of 2022, more than 20 countries screened all those sectors. What is more, in 2022, more than ten countries screened finance, food security, media, biotechnology, robotics, artificial intelligence, microprocessors, and quantum technologies, although almost none screened them in 2007. Both the traditional and the new sectors are by and large also sectors that the new EU regulation makes mandatory.
Figure 2 Number of countries screening foreign investment, by sector, 2007 and 2022
Formal prohibitions remain rare. Across the EU, only 18 to 39 transactions per year were blocked or withdrawn in 2020-2022 (European Commission 2021, 2022, 2023). But screening may affect many more deals than it blocks. Screening criteria such as “public order” are vague, reviews take weeks or months, and legal advice is costly. In 2020, 80% of the cases notified to EU authorities turned out not to require formal screening, suggesting that many investors file to be safe on the legal side.
Screening can affect cross-border investment through several channels (Eichenauer and Wang 2026). Some are intended: authorities may prohibit a deal or impose conditions that lead investors to abandon it, and screening may deter investment that could potentially threaten national security. Others are not. Screening creates uncertainty about whether a deal is covered and, if so, whether it will be approved, which is known to depress M&A activity (Bonaime et al. 2018). It also entails compliance costs for legal advice, filings and delays. In France, outside counsel alone was estimated to cost 3.5–4% of transaction value (Ministère de l’économie et des finances 2018).
These channels differ in which deals they affect most. Prohibition and deterrence should hit controlling acquisitions hardest, as these attract most scrutiny. Uncertainty and compliance costs should weigh more strongly on small, non-controlling deals. Fixed costs associated with screening represent a larger share of the expected returns for smaller investment, and investors can more easily switch to alternative small targets. In contrast, large strategic acquisitions are more costly to abandon, especially when no other target holds the assets of interest. Effects through uncertainty are expected to fade as investors learn how the rules are applied.
We link our screening data to all cross-border mergers and acquisitions (M&As) targeting firms in 43 OECD and EU countries, drawn from Bureau van Dijk’s Orbis M&A (formerly Zephyr) database. We then compare deals in a sector before and after they become subject to screening with comparable deals in unaffected sectors of the same country and in the same sector in countries that do not (yet) screen it. There are no differences in trends before screening is introduced, and the results are robust to alternative estimation methods and to excluding the years of the COVID-19 pandemic. Three findings stand out, in line with the channels discussed above.
First, screening reduces the number of cross-border deals (Figure 3, panel a). After a sector becomes subject to screening, inward M&A deals fall by 11–18%, or roughly 24–40 deals per year in the average country. This is far more than the number of deals formally blocked, pointing to deterrence, uncertainty and compliance costs rather than prohibitions. The decline is immediate but fades after about two years, consistent with investors learning how the new rules work in practice. There is no evidence for postponed deals.
Second, the decline is entirely driven by minority deals (Figure 3, panel a). The number of acquisitions of stakes below 50% falls. By contrast, majority acquisitions rise (significantly so in the European sample) and full takeovers seem unaffected.
Third, total investment value does not fall (Figure 3, panel b). Because the deals that experience a reduction are small and those that increase are large, the average value of a cross-border deal rises significantly, while the effect on the total value of inward M&A investment is statistically indistinguishable from zero. We thus conclude that screening does not reduce the amount of foreign capital inflows. But it shifts its composition towards larger, controlling investments.
Figure 3 Screening shifts cross-border M&As from minority stakes to controlling acquisitions
Existing studies document that countries with more Chinese FDI are more likely to introduce investment screening and their screening tends to be stricter (Bauerle Danzman and Meunier 2023, Bencivelli et al. 2023). A natural concern is therefore that screening may primarily deter investors from particular countries such as China. We find little support for this (Figure 4). Screening reduces deals by all groups of investors. The estimated declines are somewhat larger for Chinese, US and state-owned investors, but the differences relative to other investors are generally not statistically significant. The few marginally significant differences appear in one sample but not the other. We find no difference between investors from autocracies and democracies, or between investors routed through tax havens and others.
Figure 4 Screening reduces deals across investor types, not only those from particular countries
The one clear contrast is between deals within the European Single Market and deals involving investors from outside the EU/EFTA. Deals among EU and EFTA investors, which most European regimes exempt in most sectors, are not significantly affected, whereas deals involving investors from outside the EU/EFTA significantly fall. Screening thus pulls up a wall around the Single Market rather than around particular, politically non-aligned countries.
Our results offer both reassurance and new considerations. The reassurance is that screening has not caused a collapse of foreign investment into sensitive sectors most closely scrutinized by regulators: our evidence points to a mild, rather than severe, deglobalisation. Policy makers should, however, take into account that screening reshapes who invests and how this could consequently affect the economy.
Our findings matter as EU member states implement the new regulation. Its extension to EU-based investors controlled from outside the EU will widen the footprint of screening, since our results show that exempt intra-EU deals have so far been unaffected. Three design choices could limit unintended costs. First, clear criteria for notification and more transparency about the risk vectors would reduce precautionary filings and uncertainty. Second, fast-track procedures or safe harbours for small, non-controlling stakes in certain sectors would lower the fixed costs that fall disproportionately on minority investors. Third, more transparency about screening outcomes would help investors learn faster what is, and is not, a concern. Screening can protect security without quietly filtering out the investments that bring capital and know-how to the European economy.
Source : VOXeu
Spending on public pensions through Social Security and on elderly healthcare through Medicare accounted for…
Nominal interest rates increased significantly in the US and euro area following the post-pandemic inflation…
Official development assistance fell by 23% in 2025, the largest drop on record. Aid disbursements…
EU debt has become a noticeable feature of Europe’s fiscal architecture. However, EU fiscal rules…
The European banking sector is known for high regulatory complexity, market fragmentation, and uneven implementation…
Emerging market capital flows remain highly sensitive to global shocks. This column analyses foreign investors…