Introduction
Do geopolitical considerations influence European citizens’ opposition to incoming Foreign Direct Investment (FDI)? And if so, what specific geopolitical concerns shape these preferences? Traditionally, investment policy has been considered a domain of “quiet politics,” largely insulated from public contestation and dominated by technocratic and business actors.Footnote 1 However, this characterization has become difficult to sustain. Over the past decade, the European Union (EU) and its Member States (MSs) have changed their traditional open-door policy towards inward FDIs and joined other major advanced economies in expanding their capacity for investment screening,Footnote 2 often framing foreign acquisitions not only in economic terms but also as matters of national security, technological sovereignty, and geopolitical competition. This evolution marks an “end of naivety” in the EU’s role in global investment politicsFootnote 3 and is widely deemed a key manifestation of the EU’s broader geopolitical turn.Footnote 4 In this context, understanding whether and how geopolitical considerations influence citizens’ preferences about inward FDIs is central to assessing the political foundations of contemporary economic policymaking in the European context. Thus, this article provides a first step in unpacking the relationship between public opinion and the recent shift toward more restrictive investment policies in Europe, shedding light on the democratic foundations of the EU’s broader geoeconomic turn. While much of the existing literature has focused on elite strategies, institutional dynamics, and state or firm-level actors, we make a key contribution to this Special Issue, which seeks to move beyond state-centric accounts of geopoliticization and to shed light on the role of non-state actors in shaping this process.Footnote 5
Existing research suggests that geopolitical considerations shape public attitudes toward foreign economic policies but also highlights that these considerations are multi-dimensional and operate through different mechanisms: the “geopolitical” label encompasses different types of preoccupations about the implications of FDIs.Footnote 6 Rather than treating these factors as independent, we conceptualize them as part of a broader framework through which individuals evaluate foreign investment under conditions of uncertainty. Because the economic consequences of FDI are complex and often difficult to assess, citizens are unlikely to rely on detailed economic calculations and instead form their preferences using simplified evaluative benchmarks. We focus specifically on how citizens of EU Member States evaluate the security and power implications of foreign investment, distinguishing between two complementary dimensions: system-level considerations related to the expected outcomes of the investment and country-level considerations connected to the properties of the investing country.
More specifically, we test whether citizens’ support for blocking a foreign acquisition of a domestic company is affected by geopolitical considerations concerning (1) the distribution of relative gains between the European country and the country in which the investing company is based, (2) whether the country in which the investing company is based in China, (3) whether the country where the investing company is based is a security ally, (4) whether the country where the investing company is based is an EU Member State. We also control for several economic and political factors that may affect respondents’ attitudes towards incoming FDIs. The former include characteristics of the acquiring company such as size, technological intensity, and economic performance and characteristics of the sector in which the proposed acquisition takes place. The latter include whether the country of the acquiring company is a democracy and provides reciprocal openness to FDIs.
We test these hypotheses through a conjoint experiment embedded in a pre-registered online survey fielded in three key MSs—France, Germany, and Italy—from 15 to 23 April 2024 (N ≈ 1,500 in each survey). Our results show that, out of the potential geopolitical considerations we examined, three are particularly relevant in shaping French, German, and Italian citizens’ attitudes toward incoming FDI: (1) the distribution of relative gains between the host country and the sending country, (2) whether the sending country is a military ally, and (3) whether the investment originates from another EU Member State. This experimental design allows us to assess how different attributes of foreign investments shape respondents’ willingness to support or oppose government intervention.
Our results show that three geopolitical considerations consistently influence public attitudes across all three countries: concerns about relative gains, the military alliance status of the investing country, and whether the investment originates from within the EU. Citizens of the EU Member States are more likely to oppose FDI when it is described as benefiting the investing country more than their own, when it originates from a non-allied country, and when it comes from outside the EU. By contrast, we find no evidence of a consistent “China effect”: Chinese investments are more opposed than those from allied countries, but not systematically more than those from other non-allied countries. Taken together, these findings suggest that citizens evaluate foreign investment through simplified geopolitical evaluative benchmarks emphasizing relative position and alignment rather than detailed economic considerations or other country-specific biases. More broadly, our results indicate that the geopoliticization of investment policy is not solely an elite-driven process but is grounded in structured public preferences.
In addition to unpacking the relationship between public opinion and recent policy shifts, this article helps connect the literature on the politics of FDIs in Europe to the broader International Political Economy (IPE) literature. While this scholarship has come to pay growing attention to the question of how citizens of major world economies form opinions on inward FDIs,Footnote 7 to our knowledge this is the first study that systematically examines the drivers of citizens’ support for (or opposition to) FDIs in EU Member States. Thus, this article contributes to filling a major empirical gap in the IPE literature on mass preferences over foreign economic policies, particularly considering that the EU is the top global destination of FDIsFootnote 8 and it played a pivotal role in shaping the emergence and development of transnational networks of production, distribution, and trade centered on Global Value Chains (GVCs).Footnote 9
European public opinion and incoming FDIs
The EU and its MSs have rapidly changed their open-door approach towards incoming FDIs in recent years by strengthening their institutional capacity for investment screening. By investment screening, we mean the practice of screening, mitigating, or blocking incoming FDI: mechanisms that allow governments to review and condition foreign investment transactions, impose mitigation measures on investments that raise concerns, and, in the most severe cases, deny entry or require divestment. Screening entails the systematic review and monitoring of foreign investment transactions to assess potential risks; mitigation involves imposing conditions or remedial measures on transactions that raise concerns; and blocking—the most consequential and least frequent outcome—entails denying entry to or requiring the divestment of investments deemed unacceptable. In practice, investment screening mechanisms mostly operate through screening and mitigation, with outright blocks representing only a small share of cases. In February 2019, the EU introduced a mechanism to scrutinize incoming FDIs deemed dangerous to European security and critical interests. While falling short of creating a truly binding supranational mechanism for screening incoming FDIs, this reform stimulated the adoption (or strengthening) of investment screening mechanisms across MSs. Since it came into force, the number of MSs with a national screening mechanism has grown from 11 to 23, and in 2023 alone almost 500 FDI transactions were scrutinized by MSs and the European Commission.Footnote 10 Notably, these policies strengthen governments’ ability to screen and eventually block incoming FDIs based on political and security considerations, rather than economic ones. As noted by Bauerle Danzman and MeunierFootnote 11, concepts like “national security,” “essential security,” and “public order” have increasingly shaped the justification for these institutional mechanisms across European governments. This shift is widely seen as one of the most significant expressions of the EU’s broader geoeconomic turn: the growing trend of using economic policy tools for geopolitical purposes.Footnote 12
The emergence of this European investment screening regime has sparked scholarly interest in the domestic societal drivers of this practice at both the EU and MSs’ levels. For instance, Chan and MeunierFootnote 13 use extensive interviews with high-level EU and MSs’ officials to show that support for the EU FDI screening mechanism is higher in countries with greater technological intensity, due to concerns over unreciprocated technology transfer. Vlasiuk Nibe et al.Footnote 14 argue that the EU investment screening regime resulted from a strategy of pre-emptive depoliticization aimed at diffusing public opposition to incoming FDIs. Turning to MSs’ investment screening choices, Doppen et al.Footnote 15 highlight the importance of domestic state-firm relations in shaping the institutional configurations of such choices, while Calcara and Poletti and Vlasiuk NibeFootnote 16 draw attention to the preferences of domestic political parties and organized business groups, respectively.
This literature builds on earlier research on the United States, which was an early adopter of investment screening through the Committee on Foreign Investment in the United States (CFIUS). Studies of the U.S. case have long emphasized the role of national security concerns, domestic political pressures, and the strategic importance of key sectors in shaping screening decisions.Footnote 17 These insights have provided an important theoretical and empirical foundation for understanding the domestic politics of the emergence and evolution of investment screening mechanisms in the European context.
However, to our knowledge, no empirical study has yet investigated the drivers of citizens’ support for (or opposition to) FDIs in MSs. While it is common to refer to European citizens’ concerns about the geopolitical and economic implications of incoming FDIs—particularly those originating from China—as a major driver behind the shift towards a less open-door policy on foreign investments in Europe, empirical support for these claims largely takes the form of anecdotal and illustrative evidence.Footnote 18 The few cross-national analyses that do provide systematic evidence on European citizens’ attitudes towards FDIs draw on survey questions that do not allow them to directly address the determinants of attitudes towards FDIs. For instance, Burgoon and Raess’ (Reference Burgoon and Raess2014) study on the preferences of European workers regarding Chinese FDIs relies on a Eurobarometer question concerning perceptions about the alignment between the EU and China in dealing with globalization. Similarly, the cross-national analyses carried out by Daniels et al., Kaya and Walker, and Harms and SchwabFootnote 19 rely on a survey question from the International Social Survey Programme (ISSP) probing opinions on how the activities of multinational companies affect local business. Finally, Lee and XenosFootnote 20 analyze survey data from the Pew Global Attitudes Project, gathering individuals’ opinions on the influence of large companies based in foreign countries. While these studies have contributed several important arguments about the drivers of public opinion on the activities of multinational companies, they rely on data from surveys that cannot test the determinants of citizens’ support for (or opposition to) incoming FDIs.Footnote 21
In contrast, there is a growing body of evidence on the drivers of such attitudes among citizens of countries outside the EU. These studies explore the opinions of citizens in advanced economies, such as the US,Footnote 22 the UK,Footnote 23 Switzerland,Footnote 24 and Japan,Footnote 25 as well as emerging and developing economies, such as China,Footnote 26 and Latin American countries.Footnote 27 These studies have advanced—and empirically tested through both observational and experimental data—several arguments about the drivers of individual-level attitudes towards foreign investments in these countries. In the next section, drawing on both this body of work and the existing literature on the domestic politics of FDI screening in Europe, we derive a set of hypotheses regarding the geopolitical drivers of European citizens’ opposition to incoming FDIs.
Hypotheses: the determinants of European citizens’ attitudes towards incoming FDIs
There is widespread consensus that the EU has undergone a geopolitical turn in its external economic relations in recent years.Footnote 28 For example, in November 2019, President Ursula von der Leyen took office with a pledge to lead a ‘geopolitical’ Commission that would proactively use economic tools for geopolitical purposes.Footnote 29 In setting the guidelines for her second term, von der Leyen further asserted that “in today’s world, geopolitics and geoeconomics go together,” and that “Europe’s foreign and economic policy must do the same”.Footnote 30 Changes in policy regarding how European governments approach incoming FDIs have been regarded as a key pillar of this geopolitical shift.Footnote 31 Indeed, these policy changes represent a shift away from the traditional open-door approach—focused on the economic benefits of incoming FDIs—in favor of a more restrictive stance, in which concepts such as “national security,” “essential security,” and “public order” have gained significance as criteria guiding governments’ decisions to screen and, ultimately, block FDIs.Footnote 32
To address the question of whether this geopolitical shift, which undoubtedly impacts the EU’s approach to incoming FDIs, aligns with the preferences of European public opinion, we must begin by defining what we mean by geopolitical considerations. In line with the definition provided in the introduction of this Special Issue, we broadly define geopolitical considerations as individuals’ evaluations about these economic transactions’ security and strategic implications rather than about their narrowly economic or other political consequences.Footnote 33
Both theoretical and empirical reasons make it plausible that geopolitical considerations play a key role in shaping French, German, and Italian citizens’ attitudes towards incoming FDIs. Theoretically, it is important to note that assessing the distributional consequences of foreign economic policy choices is typically difficult for individuals. This often leads people to rely on cognitive shortcuts and heuristics.Footnote 34 Since the distributional consequences of incoming FDIs are often complex to assess and only visible in the long term, this tendency is likely to be particularly relevant in this case.Footnote 35 Numerous studies, which we discuss in the remainder of this section, lend support to this view, showing that non-economic considerations are indeed key in shaping attitudes towards incoming FDIs. Moreover, several factors suggest that, among various types of non-economic considerations, geopolitical ones should be particularly relevant. Many studies have emphasized the growing politicization of trade and investment policies in MSs in recent years.Footnote 36 Politicization is important because it contributes to increasing the salience of the abstract non-economic components of policies, making it easier to contest them.Footnote 37 Geopolitical considerations can gain importance in a politicized environment when the emergence of military conflicts in close geographical proximity makes individuals better understand the potential security effects of foreign economic policy choices.Footnote 38 Hence, the geopolitical tensions triggered by Russia’s invasion of Ukraine, combined with rising domestic concerns about China’s emergence as a political and economic superpower, can be expected to have strengthened the importance of geopolitical considerations in how European citizens assess the merits of inward FDIs. Unsurprisingly, some authors argue that EU investment politics has become highly geopoliticized in recent years, with geopolitical and security lenses having become increasingly important in how the EU interprets its role in global investment politics.Footnote 39
Empirically, the view that geopolitical considerations influence public opposition to incoming FDIs is well-documented in experimental studies conducted on public opinion outside the EU.Footnote 40 These studies underscore distinct types of geopolitical considerations, suggesting that the relevant question is not only whether geopolitical considerations matter, but also which types of geopolitical dynamics most significantly shape public opposition to incoming FDIs. In the following subsection, we turn to a detailed discussion of existing arguments regarding the geopolitical drivers of such individual-level attitudes and derive testable hypotheses from them.
Geopolitical determinants
While geopolitical reasons can shape public attitudes towards FDI, it is important to distinguish among different possible pathways through which they can do so. We conceptualize citizens’ evaluations of FDI as structured along two dimensions: system-level considerations and country-level considerations. First, individuals may rely on system-level considerations concerning the systemic implications of the investment—that is, how it affects the distribution of power and the strategic position of their country vis-à-vis others within the international system. Given the centrality of distributional concerns in public opinion on international economic policies, this dimension is likely to serve as a primary evaluative benchmark through which individuals assess whether an investment is broadly beneficial or detrimental for their country.
Second, individuals may rely on country-level considerations, which concern the identity of the investing country—using its characteristics as proxies for the potential security risks and benefits associated with the investment. Whereas system-level considerations ask what an investment implies for the position of one’s own country, country-level considerations ask who the investing country is and how trustworthy or aligned it is likely to be. These considerations are likely to condition or refine initial evaluations by providing additional information about the investing country. These considerations should be understood as complementary rather than competing: they capture different dimensions of how individuals interpret the geopolitical implications of FDI, combining expectations about outcomes with judgments about the properties of the country involved. In short, these two dimensions operate in a structured manner: system-level considerations provide a baseline evaluation of the systemic implications of the investment, which is then refined by country-level considerations.
With regards to system-level considerations, we hypothesize that individuals oppose inward FDIs due to concerns about relative gains, i.e., how the benefits of incoming FDIs are distributed between the host and sending country. The international relations literature has extensively discussed how states’ concerns over relative gains can hinder international economic cooperation:Footnote 41 while cooperation may lead to economic benefits in absolute terms, it can also result (if the rival’s economic benefits are greater) in relative losses, making a country worse off in comparison to its strategic competitors. In this sense, concerns about relative gains are not purely economic in nature: they speak to the strategic implications of economic interdependence, as the distribution of benefits between states bears directly on their relative position and standing vis-à-vis one another. Moreover, concerns about relative gains have been shown to play a key role in shaping citizens’ attitudes towards other foreign economic policies, such as trade policy.Footnote 42 We acknowledge that relative gains in the context of foreign direct investment are more complex and less transparent than in the context of trade. Unlike trade flows, where gains and losses can be more directly linked to production and consumption patterns, the distributional consequences of FDI are often indirect, multi-dimensional, and difficult to assess. Thus, it could be argued that individuals are unlikely to have precise or well-informed beliefs about how the gains from a given investment are distributed between countries. However, several factors suggest that this is a potentially relevant dimension of evaluation for citizens. Precisely because the consequences of FDI are difficult to evaluate, individuals are likely to rely on simplified representations of relative gains, using information about whether an investment primarily benefits their own country or a foreign one as an evaluative benchmark to form opinions on them. Public debates around foreign acquisitions are often framed in simplified terms, where claims about who “benefits more” from foreign investment are typically presented in simplified and categorical terms. Moreover, some scholars have explicitly highlighted the potential significance of relative gains considerations among EU MSs as one of the main drivers of the EU’s more restrictive approach toward inward FDIs. Although not centered on FDI per se, Orenstein and KelemenFootnote 43 demonstrate how economic relationships, including investment ones, can create asymmetric influence and strategic dependence within the EU, suggesting that foreign economic engagement is often interpreted in terms of uneven gains in power and leverage. Moreover, Haroche convincingly argues that the EU’s geoeconomic turn is driven by global power competition, in which economic policies are increasingly evaluated in terms of relative advantage and vulnerability.Footnote 44 More directly on the question of FDIs, DoppenFootnote 45 shows that EU investment screening is increasingly driven by concerns about the distribution of strategic advantages between the EU and external actors, a logic closely aligned with relative gains arguments. Taken together, these arguments suggest that citizens’ attitudes towards incoming FDIs can shift depending on whether the gains resulting from the investment for the host country are smaller, equal, or larger than those for the home country. We therefore test the following hypothesis:
H1. Individuals are less likely to oppose incoming FDIs when they believe the gains for their country are equal to or greater than those for the home country, compared to when they are smaller
Next, we focus on three properties of the investing country that individuals use as additional proxies for the potential risks and benefits associated with the investment. First, we consider whether there is a unique “China effect” that makes citizens particularly wary of investments originating from China. Anecdotal evidence, such as the significant public opposition to Chinese acquisitions of major companies in various EU Member States, supports this notion.Footnote 46 Interview data from EU policymakers also underscore the importance of this factor.Footnote 47 Moreover, as Babić and DixonFootnote 48 show, this reflects an ideational shift among European policymakers, whereby foreign investment is increasingly reinterpreted through a security lens. This growing relevance of security considerations in how European elites view Chinese investment provides a basis for expecting that citizens may evaluate Chinese FDI differently from other sources. Experimental studies on public opinion about FDIs outside the EU provide further support for this view. As discussed in the previous section, experimental evidence from the US,Footnote 49 Switzerland,Footnote 50 Japan,Footnote 51 and the UKFootnote 52 shows that, regardless of the economic implications of foreign investments, citizens in these countries are more likely to oppose FDIs from China than those from other countries. This finding is especially significant when considered alongside research showing that, in the context of prominent geopolitical tensions, voters tend to develop strong policy preferences regarding strategic competitors. For example, Zeng and LiFootnote 53 find that US citizens are less likely to support Chinese FDIs when exposed to information emphasizing the security and economic threats posed by China. In short, we expect individuals to internalize geopolitical concerns about China’s rise as a major political and economic competitor when assessing incoming FDIs. This leads us to propose the following hypothesis:
H2a. Individuals are more likely to oppose incoming FDIs from China than incoming FDIs from any other country
Second, we focus on whether people’s assessments of inward FDIs are influenced by whether the country from which the investment originates is a security ally. Security alliances can foster international economic exchange by easing fears about potential negative side effects.Footnote 54 Existing research supports this view. For example, some studies provide evidence that individuals tend to prefer trading with allies rather than adversaries.Footnote 55 This logic may extend to citizens’ opinions on incoming FDIs. For instance, Chilton et al.Footnote 56 find that US citizens are more likely to oppose acquisitions from companies based in countries perceived as security threats. Similarly, Li and ZengFootnote 57 show that Chinese citizens are more likely to oppose FDIs originating from Asian countries with which China has contentious political relations. These contributions suggest that the nature of the security relationship between the host and sending country may significantly influence individual-level preferences regarding incoming FDIs. Based on these insights, we hypothesize the following:
H2b. Individuals are less likely to oppose incoming FDIs from a military ally than from a country that is not a military ally
Finally, we assess whether citizens tend to oppose FDI from other EU countries less than FDI from outside the EU. This view finds support in various streams of literature. There is a long tradition of studies highlighting how European integration fostered the emergence of a collective European identity, whereby citizens tend to define their identity more in European than in national terms.Footnote 58 Moreover, some studies show that, in the presence of both short- and long-term threats, European identification leads European citizens to favor a more militarily powerful, unified, and autonomous EU.Footnote 59 Empirical work further demonstrates that military tensions in the EU’s proximity have a sizeable and persistent positive effect on European identity. While not directly addressing how citizens view foreign economic policies, these arguments suggest that military tensions near the EU might strengthen European identity, which could, in turn, make individuals more likely to view intra-European economic exchanges positively due to their potential to strengthen the EU’s economic power and political autonomy.
H2c. Individuals are less likely to oppose incoming FDIs from an EU Member State than from countries outside the EU
Control variables
Given the importance that other considerations may have in assessing the impact of incoming FDIs, we also control for several economic and political factors. In the former category, we begin by focusing on three arguments that highlight the potential relevance of the properties of the firm targeted by the FDI.
First, we consider the possibility that European public attitudes are shaped by the size of the targeted company. Existing empirical studies have shown that opposition to foreign acquisitions is higher for large companies with national profiles and lower for smaller ones.Footnote 60 More generally, the public visibility and potential politicization of foreign acquisitions of domestic companies can be expected to increase when such acquisitions concern large domestic companies. We therefore control for the possibility that public opposition is stronger when the FDI targets a large company.
Second, citizens may form their opinions based on economic considerations concerning the targeted company’s economic performance. While companies in economic distress may lobby for protectionist measures, they do not typically seek comparable restrictions on incoming FDIs.Footnote 61 From the public’s perspective, incoming FDIs can “save jobs” and thus be seen as a means to alleviate the economic problems of companies in crisis. The opposite is likely to be true when companies perform well. Not only are these companies less likely to need foreign capital to survive in the market, but they are also more likely to perceive incoming FDIs as foreign attempts to “steal” successful domestic companies.Footnote 62 Hence, we control for whether individuals are less likely to oppose incoming FDIs targeting a company in economic distress.
Third, public opinion could be affected by the level of technological intensity of the targeted company. Chan and Meunier and Bauerle Danzman and MeunierFootnote 63 show that one of the most important factors driving opposition to incoming FDIs is the technological intensity of the targeted company within the host country. While these studies focus on government-level preferences and behavior, we draw on them to control for the possibility that individuals internalize concerns about the potential negative implications stemming from foreign firms obtaining control over high-tech domestic companies. For instance, high-tech companies are increasingly mentioned in public discourse across Europe as essential to advancing long-term economic growth, for instance with reference to Artificial Intelligence.Footnote 64 Public opinion could therefore be concerned that inward FDIs imply the selling out of technological “crown jewels” of the national economy. We therefore control for the possibility that individuals are less likely to oppose incoming FDIs targeting a low-tech company compared to those targeting a high-tech company.
We also consider whether the sector targeted by the FDI influences opposition. Empirical studies show that support for restrictions on FDIs may depend on sectors considered strategically relevant for a country’s economic and industrial growth.Footnote 65 High-profile cases of public uproar against Chinese attempts to acquire companies in the energy, transport, and telecommunications sectors in the USFootnote 66 and the EUFootnote 67 lend support to this view. More generally, the COVID-19 pandemic and rising geopolitical tensions in the EU’s geographical proximity exposed the potential vulnerabilities of a highly integrated international economic system. These circumstances reinforced the trend towards the geopoliticization of EU trade and investment policyFootnote 68 and provided fertile ground for the EU to embark on a systematic re-evaluation of its approach to economic cooperation. To control for the possibility that sector-level characteristics affect public opinion on incoming FDIs, we also examine whether the sector of the targeted company makes a difference, including the following sectors: information and communication, financial and insurance activities, wholesale and retail trade, manufacturing, real estate, transportation and storage, and agriculture.
Finally, we control for the possibility that citizens care about reciprocity. Building on a long tradition in international relations,Footnote 69 several studies have demonstrated that reciprocity plays a crucial role in shaping public foreign policy preferences.Footnote 70 This also concerns attitudes towards FDIs. For instance, evidence suggests that US politicians are more likely to oppose incoming Chinese FDIs in sectors where US companies face similar investment restrictions in China.Footnote 71 Additionally, Chilton et al.Footnote 72 underscore the importance of reciprocity, showing that respondents in both the United States and China are consistently more likely to oppose foreign acquisitions when the acquiring country does not provide reciprocal market access.
Research design and data
We test our hypotheses with three survey experiments fielded from 15 to 23 April 2024. The surveys sampled French, German, and Italian adults (N ≈ 1,500 per country) and were conducted by Bilendi, which uses opt-in panels. All surveys are representative of the population by age, gender, education, and region of residence. For the empirical analysis, we included only respondents who successfully passed a “screener” question that we employed to measure their attention in completing the survey.Footnote 73
Following standard comparative logic, we select cases that are similar enough to be meaningfully compared but sufficiently diverse to allow for analytical leverage.Footnote 74 These three countries can be meaningfully compared because they represent the largest economies in the EU, had investment screening mechanisms in place well before the introduction of the EU-level investment screening coordination framework, and have been the key political supporters of the emergence of the EU investment screening mechanism.Footnote 75 This means that citizens in all three countries are accustomed to state intervention in the context of FDI and have been exposed to comparable national debates about EU-level investment screening, ensuring a comparable baseline across cases.
At the same time, France, Germany, and Italy differ along several dimensions allowing us to assess whether the patterns we identify are robust across different national contexts. First, they have exhibited distinct economic trends over the past three decades, with Germany experiencing consistent growth since reunification, while Italy and, to a lesser extent, France have seen slower economic growth. Second, the political systems in these countries vary considerably, along with significant socio-economic differences. Third, and more importantly for us, they have designed national investment screening mechanisms that display significant institutional variation. Their screening mechanisms differ in key dimensions, including the breadth of sectoral coverage—ranging from more narrowly defined strategic sectors to broader conceptions of critical technologies and infrastructure—as well as the thresholds and criteria used to trigger review, with some systems relying on formal ownership thresholds and others allowing more discretionary intervention based on control or influence.Footnote 76
This combination of similarities and differences has important implications for the scope of our findings. While they should be interpreted as applying most directly to large, advanced EU Member States with established screening regimes, the consistency of the results across cases that differ in economic performance, political systems, and institutional design would help us make a strong case that the mechanisms we identify are not purely context-specific. In short, this design allows us to identify individual-level mechanisms while assessing their consistency across different national contexts, thereby strengthening the external validity of the findings within the scope of comparable EU Member States.
Experimental design
In our conjoint experiment, respondents randomly see attributes that operationalize our causal factors, forming a description of a hypothetical acquisition. The description, in which the order of the attributes was itself randomized, informed respondents about the country in which the acquiring company is based, the degree of reciprocity with regard to FDI market openness, the distribution of relative gains from the investment for home and host country, size and other features of the targeted company and the characteristics of the sector in which the targeted company operates (see Table 1 for a complete list of all attributes and formulations). We chose to use the names of real countries (with the exception of “an EU Member State”) rather than describing their abstract features (e.g., “a security rival”, “a non-democratic country”, etc.). We ensured that the countries’ features did not overlap (e.g., China is a non-democratic country and it is not a security ally, India is not a security ally but it is a democracy, Saudi Arabia is an ally but it is a non-democratic country) in order to better appreciate the mechanisms at play. One potential drawback of our choice of using real-world country labels (e.g., China, United States, EU Member State) is that they may bundle together multiple dimensions, such as regime type, alliance status, and economic development, potentially complicating causal interpretation. This trade-off is inherent to experimental design choices regarding the level of abstraction. As recent work has shown, varying actor identity involves balancing experimental control against realism, and neither more abstract nor more concrete designs are unambiguously superior.Footnote 77 We decided to prioritize a relatively realistic design in order to capture how individuals evaluate foreign investments in settings that resemble real-world decision contexts. While this implies that country labels may activate multiple associations simultaneously, it also enhances construct validity by ensuring that respondents interpret the scenarios in a meaningful and policy-relevant way. At the same time, our conjoint design allows us to isolate the average effect of country attributes across randomized profiles, mitigating concerns that results are driven by any single bundled dimension.
Attributes and formulations of the conjoint experiment
Table 1. Long description
A table with two columns: Attribute and Formulations. The table has seven rows under the Attribute column and multiple rows under the Formulations column. The Attributes are: 1. Relative gains, 2. Country in which the acquiring company is based, 3. Size of the targeted company, 4. Economic performance of the targeted company, 5. Technological intensity of the targeted company, 6. Sector in which the targeted company operates, and 7. Reciprocity in openness to foreign investment. The Formulations column lists specific options for each attribute. For Relative gains, the options are: the investment will bring more economic benefit to the country where the acquiring company is based than to [RESPONDENT’S COUNTRY], the investment will bring equal economic benefit to the country where the acquiring company is based and [RESPONDENT’S COUNTRY], and the investment will bring fewer economic benefits to the country where the acquiring company is based than to [RESPONDENT’S COUNTRY]. For Country in which the acquiring company is based, the options are: an EU Member State, China, United States, United Kingdom, Saudi Arabia, and India. For Size of the targeted company, the options are: a small company (less than 50 employees), a medium company (between 50 and 249 employees), and a large company (250 or more employees). For Economic performance of the targeted company, the options are: the company is in economic distress and the company is doing well. For Technological intensity of the targeted company, the options are: high-tech firm and low-tech firm. For Sector in which the targeted company operates, the options are: information and communication, financial and insurance activities, wholesale and retail trade, manufacturing, real estate, transportation and storage, and agriculture. For Reciprocity in openness to foreign investment, the options are: the country where the acquiring company is based is less open to foreign investment than [RESPONDENT’S COUNTRY], the country where the acquiring company is based is equally open to foreign investment as [RESPONDENT’S COUNTRY], and the country where the acquiring company is based is more open to foreign investment than [RESPONDENT’S COUNTRY].
Note: Attributes 1 and 2 are used to test the hypotheses presented in the “Geopolitical determinants” section. Attributes from 3 to 7 are used as control variables (see “Control variables” section). Formulations with an asterisk are used as reference category.
Our choice to rely on a conjoint experiment to examine these hypotheses is motivated by the fact that this method is particularly well-suited to assess the multi-dimensional preferences of respondents, allowing for the precise estimation of causal relationships and helping avoid potential problems of reverse causality that may characterize empirical analyses based on observational data.
The experiment was administered at the end of the survey, after a number of questions concerning socio-demographic factors, political attitudes, attitudes towards specific policies, and views about the economy. It consisted of three stages. First, respondents were presented with an introductory text (see Section G.1 in the Appendix). After the introduction, each respondent saw a table in which the attributes in Table 1 were randomly ordered and the formulations randomly assigned (see Section G.2 in the Appendix for an example). Finally, respondents were asked whether the government should block the acquisition. The presentation of attributes and the question were repeated five times for each respondent.
Our conceptualization of the dependent variable—i.e., whether the government should block or not the acquisition—may reflect multiple underlying motivations, including security concerns, economic protectionism, or broader attitudes toward globalization. While investment screening mechanisms in practice operate through screening, mitigation, and blocking, we focus on the most consequential outcome—blocking—as a demanding indicator of opposition to an acquisition. However, this feature is a defining characteristic of conjoint experiments, not a limitation. Conjoint analysis is specifically designed to recover the marginal contribution of different attributes to an overall evaluative judgment. By examining how variation in these attributes affects the probability of opposing an investment, we are able to identify the relative importance of these different motivations, even though they are not directly observed. In this sense, our design allows us to infer which considerations matter most for respondents’ decisions, rather than requiring respondents to explicitly articulate their motivations. In addition, from a measurement perspective, this choice also reflects a trade-off between construct validity and conceptual isolation. While more disaggregated outcome measures could target specific motivations, they would do so at the cost of departing from the real-world decision context we seek to capture. Thus, our outcome aligns with the argument that more contextually grounded experimental designs can enhance construct validity by capturing how individuals evaluate complex policy choices in realistic settings.Footnote 78
Empirical analysis
Main analysis
We analyze the results of the experiment by estimating average marginal component effects (AMCEs)Footnote 79 using a generalized linear regression model with standard errors clustered by respondent, following the method outlined by Hainmueller et al.Footnote 80. The regression models (we run a separate model for each country sample) can be written as follows:
$\eqalign{ & {Y_i} = {\beta _0} + {\beta _1}\left( {RelativeGain{s_i}} \right) + {\beta _2}\left( {CountryAcquiringCompan{y_i}} \right) + {\beta _3}\left( {Siz{e_i}} \right) \cr & \ \ \ + {\beta _4}\left( {Performanc{e_i}} \right) + {\beta _5}\left( {Technolog{y_i}} \right) + {\beta _6}\left( {Secto{r_i}} \right) + {\beta _7}\left( {Reciprocit{y_i}} \right) + {\varepsilon _i} \cr}$
where Yi is a binary variable that equals 1 if the respondent wants to block the acquisition and 0 otherwise, β 0…β 7 are the coefficients of the experimental treatments listed in Table 1 and ϵi is a random error term. The results of the regression models are illustrated in Figure 1.Footnote 81
Average marginal component effects calculated from the conjoint experiment.
Notes: The plot displays coefficients’ estimates and 95% confidence intervals. Standard errors are clustered by respondents. Each respondent evaluated five proposed acquisitions. Nominal sample sizes are 6,115 (France), 6,635 (Germany) and 5,915 (Italy). Points and lines with higher transparency indicate coefficients that are not statistically significant. Full results in Tables B.1, B.2 and B.3 in the Appendix.
While H1 can be tested by simply comparing the responses of RelativeGains when the expected benefit is “equal” or “greater for home country” versus the baseline “greater for foreign country”, for the other hypotheses we compute linear contrasts of AMCE meansFootnote 82 for each country sample. More specifically, for H2a, we compare China to each of the other five countries (5 pairwise contrasts); for H2b, we contrast military allies (EU Member States, United States, United Kingdom) versus non-allies (China, India, Saudi Arabia) (1 aggregate contrast); for H2c, which posits that EU Member States are preferred over non-EU countries, we compute a linear contrast comparing EU Member States to non-EU countries (1 aggregate contrast). We report the results of these linear contrasts in Table 2.
Average marginal component effects (AMCEs) and contrasts for all hypotheses
Table 2. Long description
A table comparing average marginal component effects (AMCEs) and contrasts for various hypotheses across France, Germany, and Italy. The table has 15 rows and 6 columns. Columns are labeled Contrast, Estimate, SE, P-value, and 95% CI. Each country has multiple hypotheses (H1, H2a, H2b, H2c) with specific contrasts. For France, hypotheses include equal benefit vs more benefit to foreign country, more benefit to home country vs more benefit to foreign country, and comparisons of China with other countries and groups. Similar hypotheses and contrasts are listed for Germany and Italy. Each row provides the estimate, standard error, P-value, and 95% confidence interval for the respective contrast.
Notes: Estimates represent percentage point differences in probability of opposing FDI. H1 contrasts are regression coefficients comparing each treatment level to the baseline “More benefit to foreign country”. H2a contrasts compare China to each other country. H2b contrasts compare military allies (EU Member States, United States, United Kingdom) versus non-allies (China, India, Saudi Arabia). H2c contrasts compare EU Member States versus non-EU countries (China, United States, United Kingdom, Saudi Arabia, India). P-values for H2a, H2b, and H2c contrasts are adjusted using the Holm (Reference Holm1979) procedure for 7 comparisons per country; confidence intervals are computed using stepwise critical values from the Holm procedure. Significance levels: *** p < 0.001, ** p < 0.01, * p < 0.05. Under the more conservative Bonferroni correction, all previously significant contrasts remain significant, except “China vs United Kingdom” in Italy (p = 0.064). Full results in Table C.1 in the Appendix.
On relative gains (H1), we find that citizens care strongly about how FDI benefits are distributed between their country and the investing country. Citizens in the three countries are consistently more likely to accept FDIs when the expected benefit for their country is equal to, or greater than, that for the acquiring company’s country. At first glance, this pattern may appear consistent with recent work emphasizing the role of fairness in shaping public attitudes toward international economic policies.Footnote 83 However, our findings point to a distinct mechanism. Fairness-based accounts predict a preference for equal or equitable outcomes, even when these do not maximize national advantage. By contrast, our results do not indicate a systematic preference for equal outcomes over favorable ones. Instead, respondents react most strongly to scenarios in which their country is relatively worse off, while showing comparatively weaker responses when it is equally or more advantaged. This pattern is more consistent with a concern about relative position—particularly an aversion to falling behind—than with a normative evaluation of distributive fairness. In this sense, the relative gains effects we identify appear to reflect a geopolitical consideration focused on positional considerations, rather than a fairness-based judgment grounded in principles of equality or equity. It is noteworthy that, while citizens’ concerns about relative gains have been documented with respect to trade policy,Footnote 84 to our knowledge this is the first study that shows that such concerns also influence citizens’ opinions on inward FDIs.
For hypotheses about the investing country (H2), results are mixed. First of all, we find no consistent “China effect” (H2a): while China is associated with the highest estimated probability of opposition and it differs significantly from EU Member States, the United States, and the United Kingdom in all three countries, it does not differ significantly from India and Saudi Arabia in France and Italy, nor from Saudi Arabia in Germany (China vs. India is, however, statistically significant in Germany). In short, while Chinese investments are consistently more opposed than those originating from EU Member States and other allied countries, they do not differ systematically from investments originating from other non-allies, such as India or Saudi Arabia. By contrast, we find consistent support for H2b: opposition to foreign investment is significantly lower when the investing country is a military ally. This effect is present across all three countries, although its magnitude varies, being strongest in Germany and weakest in Italy. Taken together, these results suggest that respondents rely on a more general “ally versus non-ally” heuristic when evaluating foreign investments, rather than singling out China as a uniquely threatening actor. Given that the survey was fielded in the post-Ukraine war period—i.e., in a context marked by multiple sources of geopolitical tension—we interpret these findings not as evidence of the absence of China-specific geopolitical concerns, but rather as indicative of a broader reclassification whereby China appears to be perceived less as an exceptional case and more as part of a wider category of non-allied strategic competitors. Lastly, we find strong support for H2c, according to which individuals are less likely to oppose incoming FDIs from an EU Member State than from countries outside the EU. Overall, German voters appear to be the most “geopolitical” in our sample, and Italian voters the least, with French voters falling in between.Footnote 85
What do the results reveal about the control variables? The size of the targeted company does not generally appear to matter. German respondents show a modest preference for blocking acquisitions of medium-sized firms, contrary to expectations, which pointed to large firms, and the effect is weak. In contrast, the performance of the targeted company does matter for French and Italian respondents: when the company is “doing well,” they are more likely to oppose the foreign acquisition. In Italy (coefficient = 0.052, p < 0.001) and France (coefficient = 0.035, p < 0.05), higher performance increases opposition, consistent with the “stealing crown jewels” interpretation of economic nationalism. Germany shows a different pattern (coefficient = 0.019, n.s.), where performance does not significantly predict opposition. We find no consistent evidence of a significant effect for company size across all three countries, and no evidence of a significant effect for technological intensity. Overall, the economic-related controls influence respondents far less than the geopolitical variables underpinning our main hypotheses.
Finally, we find little and uneven support across the three countries for the impact of reciprocity. German citizens appear to value it, but the treatment effect is not statistically significant in France and remains ambiguous in the Italian sample.
Robustness checks and heterogeneous effects
To probe the robustness of our findings, we estimated Average Marginal Interaction Effects (AMIEs) between the relative gains variable and the three contextual factors for which we formulated hypotheses: military alliance status, whether the partner country is China, and if it is an EU Member State.Footnote 86 While a few interaction terms reach conventional levels of statistical significance (e.g., the negative effect of “More benefit to foreign country × Non-ally” in Germany, and the positive effects of EU membership under “Equal benefit” and “More benefit to home country” in Italy), these effects are not consistent across countries. Given this heterogeneity and lack of cross-national replication in the two-level analysis, we caution against strong substantive conclusions. Overall, the two-level AMIE analysis shows no robust, consistent interaction effects.
To explore treatment-effect heterogeneity across subgroups, we follow Leeper et al. and compute differences in marginal means.Footnote 87 We compute differences in marginal means by dividing the sample into two groups according to six factors: education, attitude towards the government, attitude towards trade, employment status, ideology, exposure to the “China shock.” The analyses are reported in Section E of the Appendix (Figures E.1–E.6).
Education’s effect aligns with the well-known link between education and support for free trade.Footnote 88 As more educated citizens are consistently more supportive of free trade, Figure E.1 in the Appendix shows that college graduates tend to oppose incoming FDIs less than the rest of the sample. Figure E.2 in the Appendix confirms that attitude towards trade is a strong correlate of attitude towards incoming FDIs, since trade skeptics are significantly more likely to oppose incoming FDIs in all three countries.
Another test that we carried out examined the reaction to the experiment of citizens with different degrees of support for the government. While the reference to the current government was intended to enhance the vividness and realism of the scenario, it raises the possibility that some respondents were influenced in their answers by how much they trusted (or simply liked) their government. The results of this test (see Figure E.3 in the Appendix) show that government supporters and opponents do not have significantly different preferences on the issue.
We also examined whether employment status conditions preferences, expecting that employed voters might be more skeptical toward incoming FDI due to fears of increased offshoring risks. As shown in Figure E.4 in the Appendix, employment status does not yield significant differences. If anything, the point estimates suggest that employed and self-employed respondents are slightly less opposed to incoming FDI.
We then tested if political ideology affected the respondents’ reaction to the experimental treatment. Figure E.5 in the Appendix clearly illustrates that this is not the case. This result diverges from other experimental surveys that show how political preferences affect respondents’ choices, with nationalist preferences, for instance, pushing stronger relative-gains preferences.Footnote 89
Finally, we tested if exposure to the “China shock” (measured as living in a region most severely exposed to an import shock before the Global Financial Crisis) made a difference in how people responded to the experiment. Also in this case (see Figure E.6 in the Appendix), there is no evidence of heterogeneous effects in any of the three countries.
Discussion of results and conclusions
We used conjoint experiments embedded in CAWI surveys conducted in France, Italy, and Germany to test a number of hypotheses about the geopolitical determinants of European citizens’ support for (or opposition to) incoming FDIs. These hypotheses were derived from both the broader IPE literature and the nascent literature on the domestic politics of incoming FDIs in Europe. In doing so, we sought to clarify not only whether geopolitical considerations affect voters’ attitudes on FDI, but, more importantly, which of these considerations matter most—and what these micro-level findings reveal about the broader process of geopoliticization of EU investment politics and its implications for the geopoliticization-GVCs nexus.
Our results identify three factors shaping citizens’ attitudes toward incoming FDI. First, citizens strongly oppose investments that benefit the investing country more than their own, indicating that concerns about relative gains play a central role in European views on inward FDI. Second, preferences vary by the investor’s geopolitical alignment: investments from military allies face less opposition, confirming that security considerations spill over into economic evaluations. Third, origin within the EU matters; across all three country samples, respondents are significantly less likely to oppose investment from fellow EU Member States. By contrast, we find no evidence of a uniform “China effect.” While Chinese investments face higher opposition than those from military allies and EU partners, this pattern does not extend consistently across all non-ally countries. These findings have clear implications for the geopoliticization-GVCs nexus. As long as FDI is a key mechanism through which GVCs are organized, citizens’ tendency to oppose investments from geopolitical competitors while favoring those from allies may exert bottom-up political pressure on policymakers. This, in turn, could shape the macro-level conditions under which GVC restructuring takes place.
Our article makes three contributions. First, we provide micro-level evidence of geopoliticization in EU investment politics. We show that geopolitical considerations—relative gains concerns, military alliance status, and EU membership—consistently matter in shaping citizens’ attitudes. This demonstrates that geopoliticization operates not only among elites but also in the general public. The findings are also theoretically relevant: while relative gains concerns and security alliance effects have been extensively studied at the state and elite level, this is—to our knowledge—the first study to show that these mechanisms shape individual citizens’ preferences over FDI.
Second, these findings are particularly interesting when considered in combination. Taken together, these findings suggest that citizens evaluate foreign investment through a relational geopolitical lens. Rather than focusing solely on the economic characteristics of investments, individuals appear to rely on simplified evaluative benchmarks that combine concerns about relative position with assessments of the properties of the investing country. The consistent importance of relative gains indicates sensitivity to positional disadvantage, while preferences for military allies and intra-European investments point to the role of perceived alignment and group membership. In this sense, attitudes toward FDI are shaped not only by what an investment does, but by who is involved and how it affects the country’s position relative to others.
Third, the article expands the scope of existing empirical analyses of individual-level attitudes towards incoming FDIs. While a number of studies in the IPE literature have shed light on this important question with regards to a number of developed and developing countries, this is the first study that investigates citizens’ attitudes in three major EU MSs. Considering the systematic move towards less open-door policies across MSs and the relevance of the EU as a global investment power, our study helps fill a major empirical gap in the extant literature on both the politics of FDIs and the societal roots of the globalization backlash. While there is growing consensus on the idea that the EU has experienced a geo-economic turn—whereby trade and investment decisions are increasingly shaped by security concerns and geopolitical goals—it remains unclear whether this policy shift is elite-driven or rooted in the demands of societal actors. Even though the results presented here do not allow us to make any claims regarding the causal effect of public opinion on such a policy shift, they show that some geopolitical considerations are indeed present in how the public assesses FDIs. Therefore, some alignment between public opinion and policymakers is present. This alignment is consistent with the bottom-up pathway in the micro-meso-macro framework articulated in this Special Issue’s Introduction, whereby micro-level geopoliticization—the growing salience of geopolitical considerations in European citizens’ policy preferences—can generate the societal pressures that inform and legitimate geopolitically motivated policy shifts at the macro level. In this respect, our analysis complements two other contributions to the Special Issue that investigate the micro-level dimension of geopoliticization, examining, respectively, how geopolitical considerations shape public support for green industrial policy in the US,Footnote 90 and how they affect European firms’ decisions to disengage from China.Footnote 91
Turning to the limitations, our analysis does not (and cannot) address two important issues. First, we do not examine how citizens respond to situations in which other EU Member States may accept investments that their own government rejects. This dimension is increasingly relevant in light of ongoing debates about the coordination and harmonization of investment screening mechanisms within the EU, as well as concerns about potential “forum shopping” by foreign investors. While our analysis shows that citizens are generally more supportive of intra-European investment, it does not address how they evaluate the distribution of gains and risks across Member States. Future research could examine how individuals respond to cross-national dynamics within the EU, thereby capturing an additional dimension of the geopoliticization of investment policy. Second, our analysis does not investigate the causal link between public opinion and policy shifts. Unpacking this relationship would be a valuable extension of our work, and it would offer insights into the democratic foundations of Europe’s evolving geo-economic strategy. This analysis could also be extended to citizens of other EU countries, as citizens of countries with lower levels of economic development might show different preferences. Third, our primary treatment concerns the distribution of relative gains, whose actual consequences are difficult to assess ex ante—particularly for ordinary citizens. Although this difficulty is common to most, if not all, economic policies, we must be cautious about interpreting our findings as evidence of citizens’ evaluation of complex geopolitical considerations. Moreover, our study focuses on public attitudes towards FDI-related policies, but not on the actual impacts of these policies. How recent shifts in regulatory frameworks and investment screening practices influence citizens’ perceptions—and whether these effects reinforce or reshape public sentiment—remains an open and pressing question. Future research may examine how public opinion on FDI relates to views on global supply chains, assessing whether voter preferences meaningfully influence changes in production networks or if such complex decisions remain largely insulated from public scrutiny.
Supplementary material
The supplementary material for this article is available at https://doi.org/10.1017/bap.2026.10027.
Open access
