Geopolitics, conflict, and decoupling: evidence of Western divestment from Russia during 2022

How foreign firms doing business in Russia responded to the invasion of Ukraine in 2022 has received extensive media scrutiny. Publication lags in official and private databases on international corporate activity, traditionally used to investigate cross-national commercial ties, preclude short-term assessments of foreign firms’ responses. However, as geopolitical events unfold quickly, such assessments can offer valuable insights. In this paper, we advocate for and implement a replicable, near-time methodology that is executed at arms-length to track foreign corporate responses to geopolitical events which can inform both academic debate and policy deliberation. We apply it to the ongoing Russia–Ukraine conflict, estimating Western corporate divestment rates from Russia during the first 9 months of the conflict. Our findings, confirmed by extensive robustness checks, result in divestment rates in the range of 5–13%. This range is interpreted in light of the prevailing sanctions regime as well as the divestment rate witnessed after Russia’s invasion of Crimea in 2014. We also discuss the competing, often policy-induced, incentives facing international business during a time of growing geopolitical rivalry and draw implications for the design of sanctions regimes. Avenues for future research are also outlined.


INTRODUCTION
Intensifying geopolitical rivalry has led to calls for decoupling of national economies, even though reducing commercial ties has been estimated to be costly (Bekkers & Goes, 2022; Felbermayr,  Mahlkow, & Sandkamp, 2022; International Monetary Fund, 2023;  Javorcik, Kitzmueller, Schweiger, & Yıldırım, 2022).Meaningful decoupling and other thinning of cross-border commercial ties must have a counterpart in the decisions of multinational firms and The online version of this article is available Open Access is therefore of interest to international business scholars.The nature of decoupling must be linked to the manner in which an overseas market is supplied-and abandoning some of those modes of supply may be more expensive, more time-consuming, and more feasible than others.A sophisticated understanding of the corporate dimension of decoupling driven by geopolitical events requires granular information on international commercial presence.
Traditionally, official databases on trade and foreign direct investment-with OECD, UNCTAD, and United States Bureau of Economic Analysis being amongst the most reputed sources-have been deployed to assess the extent and determinants of cross-border commercial ties and, hence, to gauge cross-country corporate connectedness, and, more broadly, globalization patterns.Yet, these databases on firms' international activity publish information with a considerable lag and the same can be said for private sector databases, such as the ones compiled by Compustat.Without near-term approaches to track foreign corporate responses, analysts will be unable to determine the form, extent, drivers, speed, and efficacy of corporate decision-making.At a time of rising geopolitical rivalry, policy initiatives may weigh more heavily on international business decision-making and may indeed be influenced by state, media, and civil society perceptions of corporate response.To address this important matter, in this paper we advocate for and implement a replicable, granular approach to assessing the extent of decoupling that can be executed by analysts far from a conflict zone and that can offer meaningful insights into foreign corporate responses to geopolitical events well before aggregate figures from traditional data sources become available.
The Russian invasion of Ukraine in February of 2022 and the near-term corporate decisions that followed offer insights into the extent to which Western firms are willing and able to quickly sever commercial ties with nations now viewed by their governments as geopolitical rivals.In a way, it is a near-perfect test case.The Russian government's actions in Ukraine have been condemned so vividly by some foreign governments that Western firms that continue commercial operations in Russia could be accused of ''trading with the enemy.'' 1 The hundreds of press releases and statements to shareholders by Western firms about this matter are an indication of commercial sensitivity.Further, Russia's economy is large enough to be a credible test of the willingness to decouple while at the same time not large enough that the scale of potentially stranded corporate assets, or Russia's future growth prospects, are decisive determinants of global strategies for the majority of companies, in particular for the world's largest multinationals.
Much of the data that exist on companies' international commercial footprints is aggregate and sheds some light on the short-term withdrawal of foreign firms from geopolitical hotspots.For example, when Russia invaded Crimea in 2014, the total value of assets owned by majority-owned subsidiaries of American firms in Russia fell 16.3% below the 2013 level.By 2015, the total value of such assets had fallen 23.1% below its pre-invasion level. 2 Notice that two-thirds of this asset drawdown occurred in the same year as the invasion of Crimea.
Russia's second invasion of Ukraine has been condemned by the governments of some 40 nations that have backed up their words with wide-ranging economic, financial, and trade sanctions.These policy steps have been complemented by civil society campaigns urging foreign firms to sever all ties with Russia.How quickly international business responded to these policy-related and other pressures this time around has been of considerable interest, not only for policy analysts and international business scholars, but also for broader civil society.
The first contribution of this paper is to outline a replicable, near-time methodology that can be implemented at arms-length to track certain foreign corporate responses to geopolitical events and apply it to the Russia-Ukraine conflict.Such a methodology enables us to estimate Western corporate divestment rates from Russia during the first 9 months of the conflict.Our methodological approach involves three central design choices that are further detailed later.First, we restrict our analysis to subsidiaries of firms headquartered in the EU and G7 nations.Second, we restrict our research to divestment of equity stakes.Non-equity commercial ties-such as arms'-length exporting, licensing, and franchising-typically involve smaller capital investments than acquiring or building equity stakes.Third, we examine divestments over a relatively short time frame.
We extracted information on equity investments in Russia made by foreign companies headquartered in either an EU member state or a G7 nation at the time of the invasion of Ukraine from the ORBIS database.In April 2022, 2405 subsidiaries owned by 1404 EU-and G7-based companies were active in Russia according to official registries tracked by ORBIS.We then checked how many of these companies had entirely divested at least one of their Russian subsidiaries by late November 2022.Our findings-corroborated by several robustness checks-imply that estimates of the rate of divestment in this sample of firms in the range of 5-13% could be defended.Variation in corporate divestment rates across sanctioning nations was observed.Other differences between the exiting firms and those remaining were found and are discussed.
This paper makes two other contributions to the received literature on this topic.The second contribution is to establish the different incentives to divest from contemporary geopolitical flashpoints arising from what turns out to be at times conflicting policy-related measures implemented by governments of the sanctioning and the sanctioned nations.A heady brew of factors-including innovative counter-sanction measures from sanctioned nations and the legacy of prior attempts to encourage overseas foreign direct investments made by many countries and addressed by intergovernmental organizations such as the United Nations-has generated a potentially underappreciated set of push and pull factors on international business.Establishing the set of disparate incentives faced by Western companies doing business in Russia helps us to better contextualize firms' responses in this geopolitical hotspot and represents an important contribution of our work.
The third contribution is to explore the potential links between the design of economic sanctions regimes and international business responses and how these matters might be taken up in future research.In short, the foreign corporate response to Russia's second invasion of Ukraine is taken as a prism through which several connections between policy and international business can be viewed during an era of intensifying geopolitical rivalry.
The remainder of this paper is organized as follows.The next section ''Literature overview'' provides an overview of the relevant literature on this topic.Section ''Sanctions, other policies, and international norms bearing upon foreign firms' decision-making following the invasion of Ukraine'' discusses the various policy measures and international norms that bear upon foreign firms contemplating the fate of their subsidiaries in Russia.Section ''Methodology'' outlines the methodology and discusses its key facets.Section ''Sample construction and evidence collection'' of this paper describes how we constructed the sample and the way the evidence was collected as well as the sources used.Section ''Empirical findings'' summarizes the empirical findings of this paper as well as the results of the robustness checks performed.The last section ''Discussion and conclusion'' contains a discussion on the assessment of the factors likely responsible for the findings on rates of corporate divestment, identifies future research questions, and concludes by drawing out implications for the design of economic sanctions policies.

LITERATURE OVERVIEW
The subject matter of this paper is linked to three distinct academic literatures.The first relates to foreign corporate divestment from nations deemed as pariahs by governments or civil society at home. 3 This matter has been considered at greater length in the case of South Africa under the Apartheid regime (see, for example, Rodman, 1994; Lansing & Kuruvilla, 1988; Feigenbaum & Lowenberg, 1988;  Kaempfer, Lehman, & Lowenberg, 2009). 4A recurring theme in such studies is that the diverse distributional consequences of divestment can rationalize, in part, differential corporate responses to external pressure to cease operations and exit the South African economy.For their part, while some U.S. firms were on record divesting, having reviewed the evidence and the relevant policy debate, Lansing  (1981) suggested that a significant number of U.S. firms decided to remain in South Africa.More recently, when Russia invaded Crimea in 2014, according to the U.S. Bureau of Economic Analysis and noted earlier, by 2015 the total value of assets owned by majority-owned affiliates of American firms in Russia fell 23.1% below its pre-invasion (2013) level. 5Changes witnessed in manufacturing sectors accounted for two-thirds of that reduced footprint; 40% with sales of finance and insurance assets-offset partly by greater American asset holdings in information technology subsidiaries.One empirical question that immediately arises is whether intensifying contemporary geopolitical rivalry results in divestment patterns similar to previous episodes.
Our study should also be seen in light of the longstanding literature on the effectiveness of economic sanctions and the more recent evolution of sanctions tools deployed by governments (Hufbauer  & Jung, 2021; Kirilakha, Felbermayr, Syropoulos,  Yalcin, & Yotov, 2021; Mulder, 2022).Already, econometric analysis of monthly trade flows into Russia suggests some degree of circumvention of the economic sanctions imposed on Russia during 2022 (Chupilkin, Javorcik, & Plekhanov, 2023).Shipping products to Russia via third markets was also found in international trade data when sanctions were imposed in the aftermath of the 2014 Russian invasion of Crimea (Lukaszuk, 2021).This calls into question the utility of bilateral trade-related measures when gauging decoupling between economies.Moreover, international trade data can provide important insights on aggregate flows but shed less light on the actual extent and pattern of economic decoupling pursued by individual firms.In light of this evidence, and other reasons mentioned later, the focus of our work is on how the design of sanctions and their interaction with other national policies and international norms bears upon the near-term equity divestment decisions of individual firms.
International business scholars have been assessing the implications for both the global business environment and corporate strategies of the numerous shifts and shocks witnessed in the world economy during the past 15 years (Kobrin, 2020; Lai, 2021;  Petricevic & Teece, 2019; Witt, 2019).For some, this has led to a revival of interest in the study of Economic Statecraft and its impact on corporate decision-making (Aggarwal & Reddie, 2021; Baldwin, 2020).While the new global commercial landscape is still a work in progress, how firms identify, assess, and manage political risks has greater billing than in some earlier epochs.This is also expected to have an impact on global value chains going forward (Zhan, 2021), and also calls for greater mutual understanding between international business research and the policy practitioner communities (Lundan & Van Assche, 2021).Indeed, to the extent that the cumulative impact of contemporary sanctions, trade, and investment policy measures are fragmenting the world economy into blocks, then our analysis may provide useful insights in the degree to which some firms quickly abandon overseas markets and others are prepared to operate in what some have referred to as a bifurcated world order (Petricevic & Teece, 2019; Vertinsky, Khang, Zhou, &  Cui, 2023; Witt, 2019).

SANCTIONS, OTHER POLICIES, AND INTERNATIONAL NORMS BEARING UPON FOREIGN FIRMS' DECISION-MAKING FOLLOWING THE INVASION OF UKRAINE
Following Russia's invasion of Ukraine on February 24, 2022, governments of the Group of Seven (G7) nations, the Member States of the European Union (EU), Australia, Iceland, Liechtenstein, New Zealand, Norway, the Republic of Korea, Switzerland, and Taiwan imposed economic and financial sanctions on the Russian Federation. 6Countersanctions were enacted soon after by Russia.While these sanctions have garnered a lot of attention, in this section we also consider the incentives created by other, pre-existing policies and international norms that are potentially germane to the divestment decisions of firms operating across borders. 7hose nations sanctioning Russia have put in place several trade-, investment-, and financialrelated measures that curtail commercial ties with Russia, some with the potential to disrupt the commercial operations of foreign firms' affiliates operating in Russia.A review of EU, G7, and U.S. sanctions leads to the following observations.First, to the best of our knowledge, no EU, U.S., or G7 statement on sanctions has included divestment by Western firms as a stated goal.An oftenused formulation is that the sanctions seek to ''counter Russia's capacity to wage its illegal aggression.'' 8 The stated purpose of EU's sanctions is broader: ''the aim of weakening Russia's economic base, depriving it of critical technologies and markets, and significantly curtailing its ability to wage war.'' 9Still, neither the EU nor U.S. sanctions require divestment of Western firms' affiliates in Russia.Having written this, since April 6, 2022, through Executive Order 14071, President Biden banned new investment in Russia by ''US persons, wherever located.'' 10And the EU has imposed a ban on new investment in the Russian energy and mining sectors as part of its fourth sanctions page. 11econd, by their admission, EU, U.S., and G7 sanctions are ''targeted.''That is, they are selective in terms of commercial activities, types of companies, goods, services, and technologies covered.For sure, restrictions on imports and exports of certain goods and services to and from Russia have been imposed. 12Compliance with these sanctions may prevent Western firms' affiliates in Russia from sourcing certain goods and services from sanctioning nations and using their Russian operations as an export platform, both impairing the commercial viability of an affected subsidiary.There are notable exceptions, however.For example, the fifth sanctions package of the EU specifically includes exemptions from a ban on road freight operators for ''pharmaceutical, medical, agricultural, and food products, including wheat and fertilizers whose import, purchase and transport is allowed under this Regulation.'' 13Continuing some commercial activities are fine, it seems.More generally, where specifically permitted, the crossborder supply of a product may be taken by some corporate decision-makers to imply that production of such goods in Russia is not frowned upon either.
Third, sanctioning governments have put in place a sequence of sanctions, whose effects may cumulate or modify earlier sanctions.For example, as of this writing, the EU has put in place ten sanctions packages.Whether, realistically, the full scale of the sanctions imposed on Russia could have been anticipated by corporate executives soon after the invasion of Ukraine is open to question.That a sequence of sanctions unfolded over time may have generated uncertainty as to the potential for disruption commercial operations of a subsidiary operating in Russia.This may be a pertinent consideration inducing delay in corporate divestment decisions.
Fourth, the provision of professional services needed to execute a divestment has been banned by some jurisdictions.The EU banned accounting, auditing, business and management consulting, and legal services in their sixth 14 and eighth 15 sanctions packages.In contrast, on May 8, 2022, the U.S. provided an exemption from its ban for ''any service in connection with the wind down or divestiture of an entity located in the Russian Federation that is not owned or controlled, directly or indirectly, by a Russian person.'' 16According to legal experts we have consulted, no such exemption exists in the European Union or the United Kingdom. 17This legal difference alone may create cross-country differences in foreign firms' divestment rates.
In a series of decrees promulgated during 2022, the Russian government introduced procedures limiting the distribution of dividends and the sales of assets by companies headquartered in ''unfriendly nations.'' 18Approvals for both must be sought from a government agency. 19These decrees apply to all sectors and commercial activities.With respect to dividends, proposals for distribution must not exceed half of the company's net profit in the previous year.A proposal must include a commitment to continue commercial activities in Russia.Dividend distribution is only permitted if a company has met key performance indicators that were set by the Russian authorities.
With respect to sales of shares or other assets, including divestments, agency approval is needed.Such approval is partly contingent on the company in question obtaining a positive opinion from the relevant ministry of the Russian government.An independent valuation of the assets in question must be undertaken and the ultimate sales price must include at least a 50% discount from that valuation.The authorizing agency has the discretion to impose a schedule of payments for the proceeds over a further 1-2 years unless a voluntary donation up to 10% of the sales value is made to the Russian state. 20n addition, in August 2022, a Russian decree created the requirement for certain specified Western financial and energy companies to obtain preclearance from the president of the Russian Federation for transactions in their shares.The named companies are leading American, European, and Japanese manufacturing and financial firms. 21aken together, these measures discourage divestment by foreign firms headquartered in the nations that have sanctioned Russia.
Policies in place before the 2022 invasion of Ukraine may be germane.Some governments offer guarantees to firms that invest abroad.For example, the authorities in Berlin offer German companies that invest abroad guarantees against certain political risks, including war.According to the interim report of the agency responsible for issuing these guarantees, in 2022 the total value of guarantees to German firms operating in Russia amounted to 7.4 billion euros, a quarter of the total value of outstanding guarantees worldwide (Federal Ministry for Economic Affairs & Climate Action, 2023). 22On February 24, 2022, the German government stopped issuing guarantees for new investments in Belarus and Russia but stated it would honor old guarantees. 23intershall, a German energy group, was reported in January 2023 to be seeking to recoup some of its multi-billion euro losses from withdrawing from Russia by invoking a guarantee obtained previously from the German government. 24The impact of such guarantees on divestment decisions is unclear.To the extent that they limit the downside from divestment, the latter may be encouraged.However, to the extent that any guarantee is long-lived, then it may encourage a ''wait-and-see'' attitude on the part of a foreign firm.
As to binding obligations under international agreements for corporate behavior in conflict zones, to the best of our knowledge, we could not find any.However, the Working Group on Business and Human Rights, created by the United Nations in 2011, has issued recommended practices for companies operating in ''conflict-affected contexts'' (UN, 2022).This Working Group cautions that a ''hasty exit can be as damaging as one that comes too late.If a business decides to exit, it needs a proper exit strategy.''Moreover, firms contemplating exit are recommended to consider whether ''(a) exiting/suspending could exacerbate tensions; and (b) whether harms to people outweigh the benefits.'' 25In this context, the latter has been taken to imply that firms selling critical or essential goods should not exit conflict zones. 26he OECD's Due Diligence Guidance for Responsible Business Conduct, issued in 2018, recommends that firms ''discontinue operations or business relationships as a last resort, because the risk of an adverse impact is too high or because mitigation efforts have not been successful.'' 27Furthermore, consideration should be given to ''how disengagement might change impacts on the ground, as well as credible information about the potential social and economic adverse impacts related to the decision to disengage.'' 28ased on the above overview of sanctions regimes and other policies and norms applying to corporate conduct in conflict zones, foreign firms active in Russia may nevertheless fail to exit despite official and media campaigns to halt their Russian operations and divest from that country.For example, a multinational firm operating in a sector excluded from official sanctions may decide that it is inappropriate to abandon its Russian customers, who may have played no part in the decision to invade Ukraine or in the prosecution of the armed conflict.In other cases, international firms may not want to abandon long-term relationships with employees or suppliers or to cease operations because of the societal relevance of their products and services (for instance, the supply of life-saving medicines).
In general, in an era of intensifying geopolitical rivalry, one should not be surprised that firms operating across borders are faced with incentives created by policy intervention and extant international norms that push and pull them in different directions.This is almost certainly the case in the current Russia-Ukraine conflict.To the best of our knowledge, no foreign sanctions on Russia contain any official statements requiring corporate divestment; if anything, existing UN and OECD norms provide grounds for companies not to divest their Russian subsidiaries.Yet, those firms minded to divest may not be able to do so at all or on acceptable terms, either because the way foreign sanctions on Russia have been designed or because of hurdles to exit erected by the Russian authorities.Complicating matters further are pre-invasion investment guarantees offered by home governments.While the potential reputational damage from retaining a commercial presence in Russia is a consideration, the complex and conflicting array of policies facing international business in situations like this generate countervailing pressures.

METHODOLOGY
The empirical approach presented here is a replicable, near-time methodology that can be implemented at arms-length to track certain foreign corporate responses to geopolitical events.Here, we apply it to the Russia-Ukraine conflict investigating specifically Western corporate divestment rates from Russia during the first 9 months of the conflict.Ultimately, we estimate the percentage of EU and G7 firms with active equity investments in Russia at the beginning of the conflict that had completed the divestment of at least one of their subsidiaries by the end of November 2022.
Three features characterize our methodology.First, we confine our analysis to subsidiaries of firms headquartered in the EU and G7 nations.We focus on firms' official headquarters because this is factual information verifiable at arms-length and country-level sanctions and policies have been related to where a foreign company formally comes from, i.e., where its parent company is officially headquartered.Moreover, we restricted our attention to firms coming from EU and G7 nations because we are interested in tracking the propensity to divest of those companies that may have been under considerable pressure to leave Russia.This focus reflects the high-profile joint initiatives by governments of those nations to heap opprobrium on Russia for invading Ukraine and any understandable reluctance of firms to be seen to be ''trading with the enemy,'' the notable attempts by certain Western governments to persuade (not mandate) some national firms to exit Russia, as well as the campaigns in those nations' media to encourage exit and condemn those companies that fail to do so.
Second, we confine our research to divestment of equity stakes.Non-equity commercial ties-such as arms-length exporting, licensing, and franchisingtypically involve smaller capital investments than acquiring or building equity stakes.The loss of sales and any capital write-offs are also likely to be lower-and therefore easier to bear-in the case of non-equity ties.Moreover, exit regulations tend to be relatively less stringent for non-equity commercial ties.Conversely, equity stakes often involve higher costs associated with abandoning both profit-making assets and relationships with distributors, employees, and suppliers.For example, Danone, the French consumer goods company, expects to write off 1 billion euros as it disposes of 13 factories that manufacture dairy products in Russia. 29In short, some corporate retreats from the Russian market are harder and more expensive to pull off than others.
Other forms of retreat, such as cessation of shipments to Russia, are extremely difficult to corroborate, especially at arms-length, as discussed previously.Thus, our focus is on a form of retreat where exit costs are higher and where corroboration at a distance is feasible.If we find that large percentages of Western equity investments were divested, then this would reveal much about the willingness and ability of companies based in the EU and G7 nations to abandon commercial operations in a geopolitical hotspot.
Third, we examine divestments over a relatively short time frame.For sure, in the years to come researchers will have more official and commercial data available to gauge the scale of divestment from Russia and its determinants-and surely more insights will follow from that. 30However, following the Russian invasion of Ukraine in February 2022, given the considerable interest in policy circles and in the media in corporate decisions concerning their commercial operations in Russia, we wanted to learn how many companies could rapidly and successfully divest when geopolitical tensions surged unexpectedly.How many foreign corporations are prepared to abruptly drop their subsidiaries in an economy that generates more than a trillion dollars in GDP? Plus, given the way the sanctions and countersanctions regimes unfolded during 2022, perhaps some corporate horses fled before the stable door was bolted?Moreover, given the hurdles to exit erected by the Russian government, was there a ''Hotel California'' effect? 31e were not the first to track the reaction to the 2022 invasion of Ukraine by foreign companies with commercial operations in Russia.Soon after the invasion, a group of researchers at Yale University collected information on the responses of about 1400 individual foreign companies and non-profit organizations conducting business in Russia through a variety of means and classified them into five categories, one of which is withdrawal (Sonnenfeld, Tian, Sokolowski, Wyrebkowski, & Kasprowicz, 2022).Another group of researchers at the Kyiv School of Economics (KSE) assembled a list of foreign company responses from several sources, including that compiled by the group at Yale University. 32Both studies represent valuable research efforts that contribute to our understanding of the responses of international companies following Russia's invasion of Ukraine.
As noted above, the focus of our research is restricted to equity investments made by EU and G7 firms.To corroborate our results, and also to assess the representativeness of our sample, in this paper we provide an extensive comparison of our results with those of KSE's.We chose this comparator for three reasons: (1) KSE researchers tracked the largest number of international companies (2956) doing business in Russia since the beginning of the conflict; 33 (2) KSE has assembled their dataset from multiple sources, including from official registries and the Yale initiative; 34 and (3) they have categorized international companies' responses into four distinct categories, namely ''stay,'' ''wait,'' ''leave,'' and ''exited.''Given that our study focuses on corroborated exits, it is appropriate for us to compare our findings against the list of international companies that KSE deemed have ''exited.'' 35 An important feature of our work is to focus on corroborated exits.We focused on such exits, because, of all the corporate reactions, these stand the best chance of being verified at arms-length.Additionally, given the stakes typically involved, particularly for publicly listed foreign companies, divestment is a costly step that involves public announcements to shareholders, identification of a buying company, a transaction value, and key dates.In current circumstances, divestments are unlikely to be reversed in the near term.This is also why we did not consider promises to withdraw or to suspend operations in Russia as corroborated exits.While promises to suspend operations can be a step towards the completion of a withdrawal, they can still be reversed far more easily, and it remains difficult for analysts to verify the enactment of such promises at a distance.For example, just because a firm promised to stop operating in Russia in July 2022 does not mean it cannot resume operations later, say, in November 2022.Comprehensive tracking of suspensions of operations would require frequent checks on Western subsidiaries operating in Russia and we were unsure how any analyst located outside Russia could effectively do this. 36Thus, we chose to restrict our analysis to the corroboration of a binary event-i.e., corroborated exit or not-as its validation appeared feasible and its reversal less likely.
While the KSE study and the possibility to compare our findings with theirs were valuable, the methodology outlined here does not require a comparator for its implementation and has unique features.First, we provide a stepwise approach to replicate our near-real-time methodology in other geopolitical contexts, paying particular attention to the sampling process and the representativeness of our sample.Second, our key design choices discussed above-in particular, the focus on the corroboration of a binary event-represent another distinctive feature of our approach that can yield credible, short-term insights into firms' corporate behavior in other geopolitical hotspots which can be established at arms-length.
To summarize, our research methodology requires the creation of a representative sample of subsidiaries operating in Russia and owned by EU and G7 firms, and then, using additional publicly available information, checks how many of these companies have completed the divestment of at least one of their Russian subsidiaries by the end of November 2022.With a list of corroborated exits, estimates of divestment rates readily follow.We also perform additional robustness checks and provide an extensive comparison with the results obtained by KSE.Lastly, we perform additional exploratory analyses to compare the commercial footprint of corroborated exiting firms with others and we examine whether firms from certain sectors or from certain Western nations have higher divestment rates than others.

SAMPLE CONSTRUCTION AND EVIDENCE COLLECTION
To construct a sample of Western equity investments in Russia, we use the internationally recognized ORBIS database, which contains detailed information on over 400 million (public and private) companies across the globe and is widely regarded as one of the most comprehensive and reliable record of equity investments by foreign firms.The Baker Library of Harvard Business School describes this database in these terms: ''ORBIS, a global company database, produced by Bureau van Dijk, is unique in breadth of geographies and extent of companies covered as well as the availability of private company financial information.'' 37 As a result of the widespread usage of ORBIS in research, the coverage and representativeness of this database has been investigated.In their recent assessment, Bajgar, Berlingieri, Calligaris, Criscuolo, and Timmis (2020) benchmark ORBIS database entries against several OECD datasets and some official samples of firm data.They find that ''firms in Orbis are disproportionately larger, older and more productive, even within each size class.''ORBIS was also found to be more representative of populations of firms with at least ten employees.These authors state that ORBIS data is better suited for analysis of ''top performers and multinationals rather than underperforming firms.''Given the focus of this paper on foreign firms, this finding offers some reassurance. 38Having written that, other reasons why the ORBIS database may omit data on some firms operating in the Russian economy should also be noted.ORBIS' data provider in Russia is Ruslana.According to Kalemli-Ozcan, Sørensen, Villegas-Sanchez, Volosovych, and Yesiltas (2022) ''approximately 40% of all active companies file their accounts.So, if there are 1,500,000 registered active companies in Russia, the accounts are available for approximately 800,000 companies.Most of these are included in Ruslana.''These authors too contend that ORBIS covers relatively fewer small and medium-sized firms.Thus, our approach is likely to yield a meaningful sample of Western firms operating in Russia with relatively larger representation of larger, older, and more productive firms.
The ORBIS database identifies the geographic location of the global ultimate owner (GUO), defined as the entity or person holding a controlling stake of at least 50.01%.On the face of it, this sounds ideal for excluding owners located in Russia and outside of the G7 and EU.However, the method used by ORBIS to assign ultimate ownership based on direct shareholdings has come under scrutiny.Rungi, Morrison, and Pammolli (2018)  argue that control of a company may also occur through the aggregation of multiple shareholdings that are related to some upstream owner found in the ORBIS database.Having said that, they acknowledge that a GUO identified in ORBIS can indeed be the ultimate owner, suggesting there is some information value in GUO tags.In light of these considerations, we proceeded as follows: we used ORBIS' GUO tags to identify candidate ultimate owners and their nationality.Then, we checked the resulting lists of firms headquartered in the G7 and EU nations to see if there were any anomalies.In addition, we perform a series of robustness checks to ascertain whether the inclusion in our sample of firms listed in ORBIS as headquartered in the G7 and EU but whose ultimate owners may be a Russian meaningfully changes estimated divestment rates.
Data on approximately 36,000 firms with active operations in the Russian Federation was downloaded from the ORBIS database on April 22, 2022.To be included in our analysis, a firm must have had at least 1 million USD of operating revenue in Russia in at least one of the years 2017 to 2021, years that precede any economic disruption to the Russian economy that followed the second invasion of Ukraine.Once firms whose headquarters were registered in Russia were removed, this left 3444 subsidiaries of foreign companies for which information was available in the ORBIS database.Of those 3444, a total of 2405 were subsidiaries of companies located in members of the G7 group of nations or in Member States of the EU.Some foreign companies have more than one subsidiary that are commercially active in Russia.For example, Renault has 22 subsidiaries listed in the ORBIS database that met the operating revenue filter mentioned above.For this reason, we distinguish between the total number of EU and G7 ''subsidiaries'' and the total number of EU and G7 ''foreign companies'' that own them.A total of 1404 foreign companies headquartered in the EU and G7 countries with commercial activities in Russia were found in the ORBIS database.To replicate the construction of this database of EU-and G7-based firms with commercially active equity stakes in the Russian Federation around the time of the 2022 invasion of Ukraine, please consult ''Appendix 1''.
We then reviewed the list of 1404 foreign companies.The first finding is that there were a small number of foreign state-owned companies and non-profit foundations in the list.We also noticed that Cyprus and Luxembourg were probably overrepresented in the list, and these are well-known investment locations for Russians.Next, we spotted that there were 366 entities owned by individuals, some of whose names might be Russian.In the robustness checks, we ran additional analyses to ascertain how sensitive our results were to filtering out these individuals.
Finally, we noticed that there were a few highprofile firms in our base sample that some might contend are Russian.Yandex is a good example.The company is owned by a holding company, Yandex N.V., based in Amsterdam in the Netherlands.Of course, a large share of Yandex customers and staff are Russian or based in Russia.However, the company has offices in seven countries, including Switzerland, Israel, the US, China, and others.What criteria should we use to decide if it is Russian or not?
Upon reflection, we maintained our stated criteria-i.e., based on whether the GUO is formally based in the G7 or EU.Other researchers might use different criteria to construct their datasets, say, the nationality of the founder, the location of the bulk of employees, or where most revenues are generated.Each of these alternative metrics has pros and cons.For many alternative approaches, data availability will be an impediment to implementation.For instance, relying on revenue generation to determine a company's ''true'' origin requires access to sales data by host country that many companies do not publish.For all of these reasons, while we proceeded with the sample of 1404 EU and G7 firms, we created robustness checks that eliminated one-by-one state-owned companies and non-profit foundations, firms located in Cyprus and Luxembourg, individual owners, as well as all individual owners and firms with Russian-sounding names.The results of these robustness checks can be found in Section ''Additional analyses and robustness checks'' of this paper.
For each and every one of the 2405 EU and G7 subsidiaries that were commercially active in Russia, we checked if any post-invasion exit from Russia could be corroborated.We separately searched relevant company websites for announcements and, where available, their financial statements.Particular weight was given to statements concerning completed sales of subsidiaries where the buyer was named and other pertinent transaction details mentioned, such as reporting the value of the transaction and dates of the transaction.We acknowledge that firms included in our sample could in principle have completed divestments without making any public announcements and, hence, would be missed during our evidence collection.Still, we consider this unlikely.Given the pressure on Western firms to exit Russia since its invasion of Ukraine, companies completing divestments from Russia have an incentive to bring this to the attention of stakeholders.Yet, as will become evident, we checked all the entries in the KSE sample, especially the Western firms recorded in their ''exited'' category, to see whether they had been missed in our evidence collection.All these additional analyses are reported and discussed in the following section.

Initial estimate of Western corporate divestment rate
Of the total of 1404 EU and G7 companies with commercially active equity investments in Russia before the invasion of Ukraine, we could confirm that, by the end of November 2022, a total of 120 had actually left.Therefore, our initial estimate of the Western corporate divestment rate was 8.5%.The final column of ''Appendix 2'' reports the evidence used to corroborate the divestments of each of these 120 firms.Figure 1 summarizes our initial findings.By comparison, of the 2956 foreign companies tracked by KSE by December 2022, the KSE team found that 143 of them had ''exited''.Put differently, KSE found that 4.8% of all foreign companies they tracked had left Russia.That the percentage we find is higher is unsurprising given we confine our analysis to EU and G7-headquartered companies which have likely faced more pressure at home to leave Russia. 39e further compared our list of 120 corroborated exits with KSE's list of exiting firms, again see ''Appendix 2'' for such comparison.A total of 69 exits on our list also appear on the KSE's list as ''Exited''.In the robustness checks, reported in ''Appendix 3'' and described in the following subsection, we discuss further checks that we performed on the 74 companies that are included in the ''exited'' category on the KSE list but did not qualify for inclusion on our list of 120 corroborated exits.
In principle, the 8.5% of exiting Western firms with equity investments could constitute the lion's share of Western investment in Russia.This is not the case, as the statistics reported in Figure 2 reveal.Corroborated exits by EU and G7 firms that had equity stakes in Russia account for 6.5% of total profit before tax of all the EU and G7 firms with active commercial operations in Russia for which information is available in the ORBIS database, 8.6% of tangible fixed assets, 8.6% of total assets, 10.4% of operating revenue, and 15.3% of total employees. 40These findings imply that, on average, the exiting Western firms tended to underperform in terms of profitability and had larger workforces (which in turn may have contributed to their higher public profile). 41he left-hand panel of Figure 3 shows the countries of origin of those foreign companies that have divested at least one of their Russian subsidiaries to date.The top nine such economies are reported along with a catch-all category for the others.The right-hand panel reveals the countries of origin of those foreign companies that are still active in Russia-i.e., firms where divestments of their Russian subsidiaries cannot be corroborated.Comparing across the panels is revealing.American and Finnish firms constitute considerably higher percentages of exiting firms than those that remain.To a lesser degree that is true for British, Danish, and French firms.
Italian-headquartered firms are more heavily represented among remaining than exited firms, and to a lesser degree so are Japanese firms.A sixth of subsidiaries remaining in Russia have parent companies that are formally headquartered in Cyprus.Some of the latter may be Russian-owned, confirming our earlier decision to conduct a robustness check dropping all Cypriot-headquartered firms from the sample.
It should be evident, then, that the propensity to exit and to remain in Russia differ across G7 and EU members.As multiple factors can, in principle, determine the propensity to sever commercial ties with economies now deemed geopolitical rivals, it would be inappropriate to automatically conclude from these findings that American, British, Danish Finnish, and French multinational firms are especially susceptible to pressure from home governments or from other sources.

Additional analyses and robustness checks
Since the ORBIS database reports the type of foreign investor, the nation their headquarters is located in, and the sector of economic activity, it is possible to conduct a barrage of robustness checks as well as to report additional information on the commercial footprint of exiting firms.The checks and associated evidence are presented in ''Appendix 3''.
The key findings are as follows: • Excluding from the analysis equity investments by individuals, state-owned enterprises, and foundations does not alter the findings much.imply that fewer than 18% of U.S. subsidiaries have actually divested.In other words, while U.S. companies have divested more often than their EU and G7 counterparts, towards the end of 2022 fewer than one in five have completed exits.• EU and Japanese firms that have exited to date tend to have low levels of profitability. 42• There are fewer corroborated exits by EU and G7 firms in the agricultural and resource extraction sectors than in manufacturing and services sectors.Those firms from the former two sectors that did leave had above-average levels of profitability.
The commercial footprint of exiting firms in the latter two sectors is broadly in line with the findings for the entire sample.The profitability of the exiting manufacturing firms is very low. 43 also performed additional checks relating to the companies included in the KSE list.First, starting from the information on corroborated exits that is summarized in ''Appendix 2'', we reviewed the 51 companies that are included in our list of 120 corroborated exits but do not appear in the KSE ''exited'' list.A total of 46 of them actually appear in the KSE inventory but in a different category, primarily in the ''Leave'' category. 44he remaining five are not included in the KSE listour data collection efforts identified corroborated exits not detected by others.We also checked the 74 companies that are included in the KSE list of 143 exits, which do not appear in our list of 120 corroborated exits.Eleven of them are companies that are not headquartered Fig. 2 Commercial footprint of corroborated exits from Russia.ORBIS database does not report the financial metrics (shown in figure) for all subsidiaries included in our analysis.Still, for each reported statistics it was possible to rely on information from over 96% of all 2405 subsidiaries considered in our analysis.The exact number of subsidiaries for which information was available for each statistic shown here is reported in ''Appendix 3''.
in the EU or in a G7 nation.A further 56 of them are companies that are not associated with any of the 1404 EU and G7 companies and, moreover, for most of them we were not able to find any information on an equity divestment.Finally, seven of them could be linked to companies included in our sample of 1404 EU and G7 companies but, based on our assessment of the information publicly available, could not be included in our list of corroborated exits.
The latter check led to the conclusion that in 53 cases 45 our assessment of ''corroborated exit'' differed from the one made by the KSE study-which represents just 4% of the 1404 foreign companies included in our sample.Hence, differences in interpretation of available information do not markedly change the initial percentage (8.5%)reported here.Indeed, a sensitivity analysis of how much the overall percentage of corroborated exits would vary if we were to use KSE list shows that the percentage could range from approximately 5-13%, with both extremes being unlikely. 46he KSE team has performed additional analyses on a restricted list of 1405 unique foreign groups of companies for which they were able to identify financials for their commercial operations in Russia (Sonnenfeld, Mylovanov, Shapoval, Tian, Onoprienko,., Hirsty, & Wyrebkowski, 2023).The fact that, independently of our research effort, KSE was able to identify a comparable number of foreign companies with an equity presence in Russia-i.e., 1405 foreign companies identified by KSE versus our list of 1404 EU and G7 companies-provides us some comfort with respect to the overall representativeness and relevance of the sample used for our analysis.
Moreover, this KSE analysis shows that, as of January 29, 2023, 179 out of the 1405 unique foreign groups of companies identified were classified as having exited Russia.Thus, as they also explicitly state in their paper (Sonnenfeld et al.,  2023, pp.25-26), the percentage of completed exits is 12.7%.This finding is similar to our results as 12.7% is at the upper end of the range that we reported above.Moreover, our initial finding of a divestment rate of 8.5% corresponds to completed exits finalized by late November 2022, while KSE's finding refers to completed exits up to late January 2023.This lends further confidence to our overall finding, namely, that few Western firms had divested from Russia during the first 9 months of the conflict. 47o summarize, our analysis shows that those EU and G7 firms that exited by the end of November 2022 accounted for small shares of the Western corporate footprint in Russia.When weighted by sensible metrics of commercial activity, the overwhelming majority of EU and G7 firms operating in Russia have stayed put or have not completed planned divestments.
A preliminary analysis of differences between the sub-samples of divesting and remaining firms reveals tendencies that may be the subject of further research.Comparisons of the median values of both sub-samples reveals that divesting firms were on average twice as profitable as remaining subsidiaries.The former had a quarter more total assets than the latter but the same average value of fixed assets.Subsidiaries where exits were corroborated had on average a third more employees and a third more operating revenues than remaining firms.However, differences between exiting and remaining firms are not confined to medians.''Appendix 4'' reproduces decile-by-decile comparisons of the distribution of key characteristics of the sub-sample of corroborated divestments and the sub-sample of EU-and G7-owned subsidiaries that remain in Russia. 48Note the implied differences in the extremes of some of the distributions-in the first and tenth deciles.In short, there are likely to be notable counterexamples to any conclusions drawn from comparing average tendencies.

DISCUSSION AND CONCLUSION
The sharp adverse reaction by Western governments to Russia's invasion of Ukraine presents an opportunity to examine how willing, able, and quickly Western companies sever overseas commercial ties in the face of acute geopolitical risks.Many Western companies have spent decades and billions of U.S. dollars and euros building up operations in economies now deemed geopolitical rivals.How ready and able are those firms to abandon those investments?
The pressures on companies to decouple from geopolitical rivals are growing.National security experts have criticized Western firms for being naı ¨ve in their dealings with autocratic regimes (Inkster, 2021). 49Proponents of reshoring production, friend-shoring, and the like have had the wind in their sails since the onset of the COVID-19 pandemic (Deputy Prime Minister of Canada [Chrystia Freeland], 2022).Ultimately, the question is whether such pressures are translating into changes in the international footprint of companies.
In this paper, the fallout from Russia's invasion of Ukraine in February 2022 was used as a prism to answer this question.We designed and executed an empirical approach that provides meaningful information about divestment rates in the near term, so that analysts, policymakers, and commentators need not wait years for official data and commercial databases to reveal how much the Western corporate footprint in Russia changed during 2022.Divestment rates of EU and G7 companies with equity investments in Russia in the range of 5 to 13% were found to be defensible.Arguably, such exits take time and are typically more complex to unwind than non-equity ties.So, this percentage could increase as those companies that have announced their intention to leave 50 follow through assuming, critically, they can navigate the hurdles put in place by the Russian authorities.
Having written this, it is worth noting that twothirds of the reduction in U.S. multinational assets in Russia witnessed after the 2014 invasion of Crimea occurred by the end of the first calendar year. 51he rate of divestment by U.S. firms was found to be 17.6%, accounting for 14% of the total value of assets invested in U.S. subsidiaries in Russia before the 2022 invasion.The latter percentage is close to the 16% reduction in the total value of assets owned by American firms in their Russian subsidiaries witnessed in 2014, the year that Russia invaded Crimea.Given the recent invasion was followed by more far-reaching Western sanctions, combined with sustained and high-profile media and other campaigns to encourage U.S. firms to leave Russia, one might have expected a larger percentage draw down in U.S. corporate foreign assets based in Russia during 2022.
We recognize that Western companies doing business in Russia may have altered their commercial operations in ways short of exit.However, as discussed earlier, each of those ways-such as ceasing exports and deliveries to Russia or suspending operations-remain difficult to corroborate in the near term by analysts located outside the Russian Federation.Corporate statements about these intermediate steps can be reversed.Over time-with the publication of additional information by Western government agencies and in corporate databases-a richer picture may emerge.In this paper, we focused on a binary, non-recurring event that can be established by arms-length observers to provide a near-term evidence-based estimate of the Western divestment rate following Russia's second invasion of Ukraine.
Other caveats are in order.In principle, we could have missed reports confirming the completed divestment of other G7 and EU firms that were commercially active in Russia.Additionally, while ORBIS remains a high-quality and reliable source of data, it could be the case that some equity investments made by EU and G7 firms have not been captured in that database.However, given the extensive checks that we have done and the overall comparability of our findings with those obtained by researchers at the KSE, the likelihood that our estimated range of corporate divestment rates represents a substantial underestimate of the actual rate is slight.
To the contrary, our divestment rate estimate of 8.5% may actually overstate the degree of completed divestment.That is also because, for the purposes of calculating divestment rates, we counted a foreign company as completely exiting if one or more-but not necessarily all-of its subsidiaries in Russia have been divested.In essence, we treated partial divestments as if they were full divestments.
Furthermore, in the process of divesting, some Western firms have inserted ''buy-back'' clauses in contracts with buyers of their Russian subsidiaries.For example, Nissan reportedly sold its Russian subsidiary to state-owned NAMI with a 6-year buyback provision. 52It appears that McDonalds can reacquire its Russian operations within 15 years, according to a report of a statement by Russia's Federal Anti-Monopoly Service. 53It would seem that even a completed divestment need not mean leaving Russia forever.
Our findings raise questions that could be tackled in future research and provide relevant insights for the design of economic sanctions policies going forward.First, as far as international business decision-making is concerned, what is the best approach to managing a ''trapped subsidiary,'' that is a subsidiary in a foreign nation a group headquarters wants to dispose of but is unable to do so?Second, what obligations does senior management have to a trapped subsidiary's customers, staff, and suppliers and to its own shareholders?Third, should the OECD and UN guidelines on Responsible Business Conduct be revised now that geopolitical rivalries are intensifying?If so, what logic should govern this non-binding guidance?
Fourth, the decision to invest in a particular foreign jurisdiction in the first place may be influenced by assessments of the likelihood that host governments will take measures that make subsequent exit more costly or impossible.Similar considerations apply to decisions to reinvest profits from existing subsidiaries.A ''Hotel California'' effect may chill cross-border investments in the economies whose host governments are deemed current or potential future geopolitical rivals to other states.
The extension of US sanctions to prohibit investments in Russia by Americans located anywhere, coupled with measures taken to prevent supplying Russian buyers through third nations, call into question the ability of multinational firms to execute certain bifurcated global strategies.Such policy steps probably increase the weight attached to future scenarios where the world economy breaks up into fuzzy, if not explicitly defined, blocs.In light of the fact that the governments of over 100 nations have refused to take sides following the Russian invasion of Ukraine, this point about emerging blocs should not be pushed too far as multinational operations in the large number of non-aligned nations may not be much affected by trade frictions and sanctions regimes associated with intensifying geopolitical rivalry. 54ur research also contributes to the literature by establishing the different incentives to divest from contemporary geopolitical flashpoints arising from what turns out to be at times conflicting policyrelated measures implemented by governments of the sanctioning and the sanctioned nations.This enabled us to identify some incoherence in Western government policies towards multinational companies operating in Russia.On the one hand, a stated objective of Western commercial sanctions policy is to deny the government in Moscow resources to prosecute its military campaign in Ukraine.Yet, given Western firms must now pay 10% of the value of any proceeds from divestment to the Russian government, this objective is undermined by Western government and other pressure on multinational companies to exit Russia.Moreover, EU and UK's bans on the provision of accounting, auditing, and management consulting services necessary to execute divestments cuts against campaigns and pressure to encourage Western corporate exit from Russia.Additional research could further explore the degree to which the multiple objectives of Western sanctions regimes cut against one another.
In addition to concerns about the coherence of the Western sanctions' packages, it is unclear that Western government demarches towards their companies operating in Russia have been designed with Russian countersanctions in mind.Given the steps Russia has put in place during 2022, there is a serious risk that pressure on Western multinationals to divest from Russia will, if successful in attaining that objective, result in the transfer of valuable technology and other assets to Russian owners at fire sale prices.In the future, it would be interesting to investigate how many divestments resulted in the transfer of assets to Russian owners without seriously compromising subsequent commercial operations in Russia of the acquired assets.Such analyses would shed light on what Western sanctions regimes have actually achieved and would contribute another chapter to the literature on commercial sanctions regimes.
Another consideration is the design of a sequence of sanctions' packages.At this time of writing, the European Union has imposed ten sanctions packages and arguably Western sanctions regimes on Russia have become more far-reaching over time.While the thinking may be that ratcheting up Western sanctions over time signals to the Russian government that the economic price it will pay for sustaining its invasion in Ukraine will keep rising, this begs the question of what is the optimal sequence of sanctions?Surely, the capacity of the Russian state to retaliate directly against sanctioning nations or against the latter's commercial interests are considerations here.
Moreover, to the extent that more severe sanctions pose a large threat to ongoing business operations in Russia, given any uncertainty over the imposition of future sanctions, some multinational corporations may be tempted to adopt a ''wait and see'' attitude towards divesting subsidiaries in geopolitical hotspots-in this case, exiting when the sanctions regime become too much to bear.The impact of sequences of sanctions packages on business conduct, the uncertainty about the future evolution of such packages, and the political economy of sanction package determination are matters deserving of more attention from researchers in international business and policymakers going forward.
NOTES 1 An overview of the competing pressures facing Western companies with operations in Russia can be found in this March 18, 2022 Financial Times article on Western pharmaceutical firms.The case of pharma firms is particularly interesting as medicines and associated medical supplies were exempted from sanction regimes imposed on Russia.Thus, pharma firms faced pressure to ''scale back its presence in Russia'' even though their products were not covered by trade sanctions.Source: https://www.ft.com/content/43378dea-3a7d-40e9-8be7-bafea1afaa1f 2 The source for these statistics on U.S. multinational investment were downloaded from the following interactive site maintained by the U.S. Bureau of Economic Analysis https://www.bea.gov/international/di1usdop. 3 This literature is to be distinguished from the assessment of the effect of economic sanctions on inward flows of foreign direct investment (see, for example, Biglaiser, & Lektzian, 2011). 4A timeline of foreign sanctions and civil society pressure on foreign corporations operating in South Africa can be found using the link at the end of the note.Source: https://www.piie.com/commentary/speeches-papers/case-62-2-and-85-1 5 The source for these statistics on U.S. multinational investment were downloaded from the following interactive site maintained by the U.S. Bureau of Economic Analysis https://www.bea.gov/international/di1usdop. 6There is no suggestion here that Western nations have adopted the same sanctions against Russia, nor that enforcement is uniform. 7We do not claim to provide here an extensive inventory of the many sanctions packages imposed on Russia and the countersanctions enacted by the authorities in Russia.A detailed sanctions tracker for Australia, EU, Japan, and the United Kingdom can be found here: https://www.ashurst.com/en/news-and-insights/hubs/sanctions-tracker/.SPGlobal has assembled a useful ''timeline'' of sanctions imposed by many nations on Russia that is available here: https://www.spglobal.com/marketintelligence/en/news-insights/latest-newsheadlines/sanctions-against-russia-8211-a-timeline-69602559.
16 https://ofac.treasury.gov/faqs/1034. 17In its ninth sanctions package the EU did allow European firms seeking to exit Russia to apply for an authorization for the sale of sanctioned goods and technologies as part of a divestment from the Russian market.However, there is no certainty that an authorization will be provided.Nor does this measure address the ban on the provision of business services necessary to proceed with divestment.
18 For an overview of the relevant decrees and their implications see: https://schneider-group. com/en/news/countries/new-restrictions-forcompanies-controlled-by-non-friendly-countries/.
19 Formally, the Sub-commission of the Government Commission on Monitoring Foreign Investment. 20Press and consultancy company reports in March and April 2023 consistently report that the Russian government has issued guidance that this ''donation'' will now become mandatory.See, for example, the following report by Ernst & Young: https://globaltaxnews.ey.com/news/2023-0738russian-authorities-require-exiting-foreign-companies-to-pay-contributions-at-5-percent-or-10-percent-of-market-value-implications-for-usmultinationals.
21 For a partial list, see: https://www.clearygottlieb.com/news-and-insights/publicationlisting/sanctions-developments-resulting-from-thegeopolitical-conflict-in-ukraine-russia. 22See the statistics presented at the top of page 20 of this report.The portfolio of the German government included 103 such guarantees to firms operating in Russia.
23 https://www.exportkreditgarantien.de/en/ news/messages/news-20220228.html. 24As reported by the Financial Times on January 24, 2023.Source: https://www.ft.com/content/ 3f98d24b-8ed5-4c26-94e3-36c75354cf89. 25An entire section of this report is devoted to guidance on ''exit'', see section IV.4.C. 26 We thank Mr. Michael Meier, a member of the Swiss delegation to the United Nations Human Rights Council, for bringing to our attention the contribution of this UN Working Group.
28 OECD (2018, p. 31). 29See the articles in the Financial Times published on October 14, 2022.Source: https://www.ft.com/ content/ab44a126-47b5-4f81-af5c-6bfed23fdec9. 30For example, given the normal publication lag by the U.S. Bureau of Economic Analysis in releasing data on the footprint of American majorityowned subsidiaries abroad, analysts interested in the how American firms responded during 2022 and 2023 to the Russian invasion of Ukraine would have to wait until the middle of 2026 at the earliest.At this time of writing (May 2023) the data available of U.S. direct investments abroad is available in excel form through to 2020, see https:// www.bea.gov/international/di1usdop. 31In the sense meant by the Eagles: ''you can check out, but you can never leave.'' 32A detailed description of the KSE project can be found on the following website: https://kse.ua/about-the-school/news/.Extensive weekly updates of their data gathering are provided.At the time of our data collection and analysis, the latest available update by the KSE was given on December 5, 2022: https://kse.ua/about-the-school/news/30th-issueof-the-weekly-digest-on-impact-of-foreigncompanies-exit-on-rf-economy/.We note that, in January 2023, the KSE team incorporated some of our evidence into their database and acknowledged doing so.study can be found here and is updated regularly: https://leave-russia.org/companies-that-exited?flt% 5B147%5D%5Beq%5D%5B%5D=9061.
On December 8, 2022, a total of 2956 international companies were tracked and their decisions to the armed conflict reported on the website.As stated on their website, ''the KSE database is more complete and comprehensive and contains * 40% more information than most other similar databases.''Four and a half months later, on April 23, 2023, the KSE database has information on 3163 international companies.This implies the average number of new entries recorded has fallen from approximately 320 per month in the first 9 months of the invasion to 50 per month during the past 4-5 months. 34The list of sources used by KSE is reported in their website: ''''KSE database is partly based on the Yale's School of Management database, epravda.com.ua,squeezingputin.com,leave-russia.orgwebsites and other open sources.Data is verified and KSE status is assigned.''Source: URL https://kse.ua/about-the-school/news/32nd-issue-of-the-weeklydigest-on-impact-of-foreign-companies-exit-on-rfeconomy/. 35The KSE team issue weekly or biweekly statistics on the status of foreign firms operating in Russia and maintain a website (https://leave-russia.org/)that contains many sources. 36Additionally, suspension of operations-while a potentially meaningful corporate response-implies that affected Western firms still maintain a series of obligations with local stakeholders-e.g., they continue to pay taxes, have employment contracts in place, keep relationships with suppliers-and are thus not examples of corporate decoupling from the Russian economy. 37The Baker Library of Harvard Business School provides further information related to ORBIS here: https://www.library.hbs.edu/find/databases/orbis.
ORBIS database regularly appears as reliable data source in academic articles published in top-ranked academic journals in international business, strategy, and management.Examples of recently published peer-reviewed papers using the ORBIS database include: Yan, Li, and Zhang (2022), Dwenger and Treber (2022), and Rios (2021). 38Given the emphasis in this paper on divestment, it is noteworthy that these authors caution against using the ORBIS dataset to study exit dynamics.We do not use ORBIS for the latter purpose-we use it to construct a sample of representative foreign-owned firms operating in the Russian economy at the time of the invasion of Ukraine in 2022. 39The likely inclusion in the KSE study of firms supplying the Russian market before the invasion through non-equity means is another potential reason for any divergence in our results. 40Recall a foreign company may have more than one subsidiary in Russia.Some of those subsidiaries may have been divested, some not.As the ORBIS database provides financial information at the subsidiary level, then this was used in the calculations underlying Fig. 2 and ''Appendix 3''. 41Our overall findings are comparable with those obtained by the KSE researchers.According to their research published on December 5, 2022, the companies that exited Russia ''employed almost 22.0% of the personnel employed in foreign companies,'' ''owned about 11.2% of the assets,'' and ''generated revenue of $35.4 billion or 12.0% of total revenue''.Source: https://kse.ua/about-theschool/news/30th-issue-of-the-weekly-digest-onimpact-of-foreign-companies-exit-on-rf-economy/.Note: KSE provides these statistics using 142 companies that exited up to December 5, 2022.While it is important to keep in mind the differences in the samples employed-the sample for the KSE study comprised all international companies doing business in Russia-the similarity in our reported summary statistics should further increase the confidence in our findings. 42One Japanese company made significant losses and this has pulled down the average for that country.
43 This raises the possibility that pressure on EU and G7 firms to exit Russia may have provided the pretext to sell underperforming manufacturing operations. 44The different categorization of these 46 companies is related to the way in which we deemed a subsidiary's exit from Russia to be confirmed.If we were to remove these 46 companies from our list of completed exits, our overall percentage of corroborated exits would fall to 5.2% (almost entirely eliminating the gap between KSE's headline percentage of 4.8%).Having said that, we considered that the information available to us regarding these companies' exit was sufficient to deem them corroborated exits. 45The 53 cases correspond to the sum of 46 cases discussed beforehand and the seven cases mentioned in the previous paragraph.The 46 cases correspond to companies that we categorized as corroborated exit but in the KSE list they are reported in a category other than ''Exit Completed''.The seven cases correspond to companies that we could associate with one of the companies included in our sample of 1404 EU and G7 companies but, based on our assessment of the information retrieved, were not deemed to be included in our list of corroborated exits. 46The 5% figure is obtained removing the 46 companies that according to the KSE list have not ''exited'' from our list of 120 corroborated exits (hence (120-46)/1404 = 0.053).The 13% figure is obtained by adding to our list of 120 corroborated exits (a) the 56 companies that according to the KSE list have exited but that we were not able to link to any of the 1404 subsidiaries included in our sample (thus assuming that were missed out from our list of 120 exits because of the incompleteness of the ORBIS sample); and (b) the seven companies that could be linked to companies included in our sample of 1404 EU and G7 companies but, based on our assessment of the information retrieved, were not deemed to be included in our list of corroborated exits ((120 + 56 + 7)/(1404 + 56) = 0.125).47 Sonnenfeld et al. (2023), on page 25, state that ''the most correct way to calculate the real % of different statuses vs totals would be to take only those companies where we have financials collected in our database.For the rest of the companies, we were not able yet to identify companies represented in Russia with revenue earned physically in the country.''Their analysis shows that, as of January 29, 2023, 179 foreign companies out of the 1405 unique foreign groups of companies for which KSE has financials collected in their database had exited Russia.As clearly shown in their chart on page 25 and explicitly stated in the text on page 26, the percentage of completed exits is therefore 12.7%. 48A chart comparing the distribution across deciles of both sub-samples in respect of the total value of fixed assets was omitted as no differences could be discerned. 49In July 2022, the heads of the FBI and MI5 issued a joint warning concerning the growing threat from China.FBI Director Wray is quoted as arguing that the Chinese government ''poses an even more serious threat to Western businesses than even many sophisticated businesspeople realize,'' and is ''set on stealing your technology.''Source: https://www.reuters.com/world/heads-mi5-fbi-give-joint-warning-growing-threat-china-2022-07-07/ 50Recall that KSE researchers have found a significant number of firms that fall into this category.
Having written this, charts produced by the KSE team reveal that since the fourth quarter of 2022 the total number and shares of firms indicating their intention to leave and the total number and shares of firms that ''exited'' have grown very slowly; see the chart in the middle panel of https://kse.ua/about-the-school/news/44th-issueof-the-regular-digest-on-impact-of-foreigncompanies-exit-on-rf-economy/.In fact, the division of companies between those waiting and staying in the KSE sample and those leaving and exiting is remarkably stable has barely changed since September 2022, with a 55-45% split in the favor of the former.Before that the split was in favor of leaving and exiting firms. 51The 2014 conflict was shorter-it lasted 1 month and 6 days.The present conflict has lasted longer than one year and pressure on Western companies with commercial operations in Russia has not abated.
53 See Reuters article published on June 2, 2022.Source: https://www.reuters.com/business/retailconsumer/mcdonalds-will-have-15-year-optionbuy-its-restaurants-russia-back-2022-06-02/. 54As noted in the last section another fruitful line of research inquiry is to ascertain why some foreign firms were able to divest from Russia and others were not.Since some multinational firms have multiple subsidiaries in Russia, divestment may be partial.Of particular interest is whether the design of sanctions regimes has influenced directly or indirect divestment decisions. 55Given our interest in creating a replicable methodology, it is pertinent to note here that ORBIS is a live dataset that does not permit downloads of data that was available at a given point in time in the past.Over time, some companies are added to ORBIS and some are taken out.Furthermore, financial data are added regularly implying that, for example, a company that had no 2021 financial data available in April 2022 may have these data available at a later date.This means that replicating the retrieval of companies using the criteria outlined in this Appendix after April 2022 leads to the identification of a slightly different set of companies.To illustrate, independent researchers that contacted us in June 2023 and that closely followed the criteria outlined in this Appendix were able to replicate our results, with the corresponding numbers of the sub-samples deviating only marginally (by 5-10%) from the ones we obtained in each step of the sampling process and reported below.These differences are entirely explainable by the way ORBIS works as discussed earlier in this footnote. 56According to ORBIS, foreign companies are identified as GUOs, i.e., ''global ultimate owner'' (GUO): the individual or entity at the top of a corporate ownership structure, or DUOs, i.e., ''domestic ultimate owners'': the highest owning entity in a country.Definition of the UO: min.path of 50.01%, known shareholder.Hence, several companies can be associated with the same GUO (or DUO).As information on firms' performance is provided in ORBIS at the subsidiary-level, it is necessary to focus on subsidiaries to calculate the operating revenue, profits, total assets, and number of employees in the dataset. 57We should emphasize the condition that we restricted our analysis to companies that had recent financial data, in particular subsidiary-level revenues.The latter information is needed to gauge the commercial footprint of the companies where exit was corroborated and compare it with the ones of those for which corroboration was not possible.We did so by focusing on those subsidiaries that reported subsidiary-level revenue data from 2021 and/or 2020 and/or 2019 to maximize the likelihood of including subsidiaries in our sample that (a) were still active in Russia in February 2022 (at the beginning of the conflict) and (b) had financial data that makes possible the analysis that we reported in Fig. 2 and ''Appendix 3''.When imposing the ''recent financial data'' constraint to the Russia-based companies obtained from ORBIS, it turned out that, as the latest available subsidiaries' turnover was becoming smaller, the number of subsidiaries for which financials were not available was increasing, thus jeopardizing our ability to address points (a) and (b) adequately.In addition, as discussed in the paper, studies addressing the reliability of ORBIS in capturing representative data across countries confirm that the smaller a company's size, the less likely is ORBIS to provide a representative sample.These considerations led to the focus on the 36,000 firms reported in the paper.

APPENDIX 1: CONSTRUCTION OF THE SAMPLE OF WESTERN EQUITY INVESTMENTS IN RUSSIA
• The original data for this analysis was downloaded from the ORBIS database on April 22, 2022. 55It consists of 36,000 companies operating in Russia.
The criteria to retrieve this sample from ORBIS was: •

Fig. 1
Fig.1How the estimated Western divestment rate was determined.

Fig. 3
Fig. 3 Location of headquarters of foreign companies where exit from Russia was corroborated and where it was not.
1. Status: Active companies • 2. World region/Country/Region in country: Russian Federation • 3. Consolidation code: LF (Limited financial), U1 (unconsolidated accounts with no consolidated companion), U2 (unconsolidated accounts with a consolidated companion).• 4. Subsidiaries with Ultimate Owners (UO) by profile: UO is one of the following types: Banks and Financial companies, Insurance companies, Corporate companies, Private Equity firms, Hedge funds, Venture capital, Mutual & Pension Funds/Nominees/Trusts/ Trustees, Foundations/Research Institutes, Public authorities, States, Governments, One or more named individuals or families, Employees/Managers/Directors; GUO and DUO. 56• 5. Operating revenue (Turnover) in Russia: minimum of one million USD for at least one of the following years 2017-2021; exclusion of companies with no recent financial data and Public authorities/States/ Governments. 57• Next, only information on foreign companies operating in Russia were retained.That is, entries whose GUO-Country ISO code != ''RU''.This yields a dataset with a total of 3444 companies.• Lastly, only EU and G7 companies are included in the sample.This reduces the sample to 2405 foreign subsidiaries.Here we filter for companies whose GUO-Country ISO code belong to the EU or G7 countries: • EU countries: AT, BE, BG, HR, CY, CZ, DK, EE, FI, FR, DE, GR, HU, IE, IT, LV, LT, LU, MT, NL, PL, PT, RO, SK, SI, ES, SE. • G7 countries: CA, DE, FR, GB, IT, JP, US. • Out of the 2405 companies, there are 1404 unique EU and G7 GUOs.• The total number of corroborated exits found in this analysis is 120 i.e., 120 unique GUOs are confirmed to have exited Russia./www.reuters.com/markets/deals/ball-corp-sells-russianbusiness-arnest-group-530-mln-2022-09-21/BITTEN OG MADS CLAUSENS FOND https://www.reuters.com/article/ukraine-crisis-danfoss-russia-idUKL1N2YW0YHBP PLC https://www.reuters.com/business/energy/britains-bp-says-exitstake-russian-oil-giant-rosneft-2022-02-27/