This study aims to explore how foreign divestment (FD) affects the insolvency risk of parent firms and examines the moderating role of corporate social responsibility (CSR) in this relationship, an underexplored but critical issue in international business. It focuses on how resource losses from divestment destabilize parent firm solvency, offering insights into the financial vulnerabilities of multinational enterprises (MNEs).
Grounded in the resource-based view (RBV) and foreign divestment literature, this study analyzes a longitudinal dataset of 119 publicly listed Nordic MNEs (Finland, Sweden, Norway and Denmark), spanning 667 firm-year observations over 1992–2019. Insolvency risk is measured using Merton’s distance to default (DD) and credit default swap (CDS) spreads, with data from the Credit Research Initiative (CRI), Orbis and Refinitiv Eikon. For the CSR moderation analysis, a sub-sample from 2003 onward is used.
Foreign divestment is positively associated with higher insolvency risk, reflecting the financial strain from strategic asset loss or reallocation. Moreover, parent firms with stronger CSR engagement face greater insolvency risk during divestments.
Integrating RBV and CSR perspectives, this study provides a nuanced view of the financial and strategic consequences of foreign divestment. It offers actionable insights for MNEs managing divestments in volatile markets and underscores CSR’s complex role in mitigating or amplifying financial distress.
1. Introduction
Even in good times there are bankruptcies, product failures, company restructurings, and the like. (Boddewyn, 1979, p. 21).
Foreign divestment, the voluntary or involuntary withdrawal from all or major parts of an initial foreign direct investment (FDI) (Boddewyn, 1979; Sethuram and Gaur, 2024), has received limited attention in international business research compared to foreign direct investment (e.g. Iurkov and Benito, 2018; McDermott, 2010; Schmid and Morschett, 2020; Tan and Sousa, 2019). However, divestment has critical implications for a parent firm’s survival and financial well-being (Powell and Yawson, 2012). Its link to insolvency risk, defined as a firm’s inability to meet its debt obligations (Campbell et al., 2008; Rego et al., 2009), remains underexplored (Rashid Shamim, 2019). Given that foreign operations are often more volatile than domestic ones (Boddewyn, 1979; Iurkov and Benito, 2020), it is important to investigate the relationship between insolvency risk and foreign divestment.
Riskier investment commitments influence firms’ divestment decisions and CSR may increase their reluctance to divest by creating long-term commitments and stakeholder expectations (Belderbos and Zou, 2009; Chen and Wu, 1996; Mata and Portugal, 2000). According to stakeholder theory, CSR generates a competitive advantage and enhances firm value (Freeman, 2010; Makni et al., 2009). As stakeholder demands intensify, CSR becomes increasingly vital for firm growth and survival (Aguilera et al., 2007; Freeman et al., 2007; King, 2008). Conversely, weak CSR performance can damage a firm’s reputation and erode its market position, as investors reward strong CSR and penalize poor performance (Lee and Min, 2015). Therefore, it is important to explore the moderating role of CSR activities in the relationship between foreign divestment and insolvency risk.
To explore these dynamics, we ask the following questions:
How does foreign divestment influence the insolvency risk of parent firms?
How do CSR activities moderate the relationship between foreign divestment and insolvency risk?
We measure insolvency risk using both DD and CDS spreads, as these bondholder-focused indicators reflect market perceptions of financial distress. Our results show that foreign divestment significantly raises insolvency risk in the following year, indicating reduced financial stability. Notably, higher CSR activity amplifies this effect, suggesting that CSR commitments may intensify the negative effect of foreign divestment on parent firms’ financial well-being, thus intensifying insolvency risk.
This study contributes to the foreign divestment literature by applying the RBV (Barney, 1991; Wernerfelt, 1984) to examine the impact of foreign divestment on parent firms’ insolvency risk. We argue that divestment can deplete valuable resources critical to sustaining a competitive advantage. By linking foreign divestment to credit market reactions and exploring CSR’s moderating role, we extend prior research. In practical terms, we highlight the importance of integrating financial risk assessments into divestment decisions. Managers of MNEs should carefully evaluate the long-term financial implications of resource reallocation and assess whether CSR expenditures align with firm capacity during periods of strategic divestment.
2. Theoretical framework
2.1 Resource-based view
Researchers have used various theories to elucidate the motives behind firms’ internationalization and their choices to pursue foreign divestment as a strategic option. We examine the phenomenon of foreign divestment through the lens of international business, using the RBV framework introduced by Wernerfelt (1984). The RBV postulates that firms gain a competitive edge by effectively controlling and leveraging unique resources (Barney and Hesterly, 2006). The origins of the RBV can be traced back to the work of Penrose (1959) and Nelson and Winter (1982), suggesting that enduring qualities, often referred to as the firm’s “winning genes,” are rooted in the development of its most valuable resources and routines. Barney (1991) further elaborated on this perspective, explaining how these attributes are used to achieve above-normal sustainable performance. That is, a firm’s resources are used in a way that leads customers to expect the most valuable products and services. When a firm manages to maintain a competitive advantage and unique conditions over time, it achieves a “sustainable competitive advantage” with above-average returns (Barney and Hesterly, 2006).
These resources include a firm’s assets, processes, capabilities, attributes, knowledge, information, etc. (Barney, 1991). In the extended version of the RBV (Mathews, 2003), external resources, such as law-making entities, customers, suppliers, etc. are also included. However, the process of divesting foreign subsidiaries can disrupt the established routines, stakeholder relationships and operational capabilities that are rooted in the international operations of parent firms. From an RBV perspective, such resource loss and dislocation increase organizational vulnerability and can lead to performance volatility, which can weaken a parent firm’s financial stability and raise its insolvency risk.
Exporting a portion of a company’s sales to international markets is considered a significant measure of competitive success, both on a national and regional scale (O’Farrell et al., 1996), which can help in the growth and survival of smaller and newer firms (D’Souza and McDougall, 1989). While some foreign divestments may be strategic decisions of parent firms aimed at refocusing on core operations (Kolev, 2016), even these moves can temporarily destabilize resource configurations. In addition, forced or reactive divestments can even worsen such effects, as firms are less likely to manage resource loss and redeployment effectively. Therefore, this study applies the RBV lens to argue that foreign divestment is not only a structural adjustment but also a potential cause of financial distress, as it weakens a firm’s resource base that sustains solvency and competitive advantage. Although RBV and survival analysis have different emphases, both are rooted in performance and a connection can be established between these concepts (Li et al., 2010).
While our primary theoretical approach is based on the RBV, we acknowledge that foreign divestments may also act as signals to external stakeholders. Especially in foreign markets with high informational asymmetry, investors and financial institutions may interpret divestments as indicators of operational or financial distress. These signaling effects can also influence market confidence, credit perceptions and a firm’s insolvency risk as a result. Although we focus on internal resource effects, future research could explore this signaling mechanism more directly.
2.2 Foreign divestment, firm performance and firm risk
Scholars have used various terms to describe the phenomenon of exit from foreign business markets, including foreign divestment, de-internationalization, subsidiary failure, business withdrawal, closure, disengagement, sell-off, liquidation, dissolution and total sales (Ameyaw et al., 2023; Burt et al., 2003; Ozkan, 2020). In this study, we encompass these various terms within the broader framework of foreign divestment. Foreign divestment is defined as “the deliberate and voluntary liquidation or sale of all or a major part of an active operation” (Boddewyn, 1979, p. 21). Therefore, foreign divestment takes place when a parent MNE reduces its equity or sells/liquidates its foreign subsidiary (Ameyaw et al., 2023).
The literature on foreign divestment outcomes remains fragmented, with no clear consensus regarding its impact on parent firm performance. There are differing scholarly views on foreign divestment, with some linking it to improved firm performance, others associating it with failure and poor outcomes, and some finding no significant impact at all. The first group of researchers suggest that foreign divestment can enhance firm performance and competitiveness as it allows parent firms to reallocate resources more efficiently, realign their strategic focus, obtain funding, strengthen core operations and so on (e.g. Brauer et al., 2017; Iurkov and Benito, 2020; Kafouros et al., 2022; Kolev, 2016; Lee and Roh, 2020). In addition, it helps in avoiding political conflicts, economic instability and local market volatility (Liu et al., 2022; Owens, 2023).
However, divestment is also associated with failure (Berry, 2013; Crick, 2004) and poor performance (Park and Russo, 1996), causing negative stock returns due to reduced geographic scope and operational flexibility and increasing market failure risk (Veld and Veld-Merkoulova, 2004). Brahmana et al. (2021) conducted a study in Malaysia on a sample of 319 publicly listed nonfinancial companies during 2012–2016 and found that the divestiture strategy decreased firm performance. Moreover, divestments carried out to address firm-level issues typically result in negative reactions in the stock market (Tsetsekos and Gombola, 1992). Similarly, during periods of financial distress, divestment may not alleviate bankruptcy pressures as intended because asset sales can occur at depressed “fire-sale” prices under crisis conditions (Zhou et al., 2011). Some scholars have found an inconclusive relationship between divestment and parent firm performance and firm value (e.g. Duhaime and Baird, 1987; Colak, 2010; Sharma and Ho, 2002; Shimizu and Hitt, 2005). Overall, the extant literature offers inconclusive results about the impact of divestment on parent firm performance. Thus, we believe it is extremely important to determine how the insolvency risk of parent firms is influenced by foreign divestments.
Although the literature on foreign divestment and parent firm performance remains fragmented, with studies reporting positive, negative or null effects, this very inconsistency highlights the destabilizing nature of divestment decisions. Regardless of its outcome, foreign divestment introduces a high level of operational uncertainty and strategic reconfiguration, which translates into increased volatility in firm performance. Researchers have widely acknowledged performance volatility as a measure of risk (Bloom and Milkovich, 1998; Miller and Bromiley, 1990; Woo, 1987). Furthermore, increased stock performance volatility is associated with higher risk and negative outcomes (Ang and Liu, 2007; Low, 2009). In addition, it has been found that higher performance volatility is linked with a higher risk of firm bankruptcy and credit issues (e.g. Correia et al., 2018; Kim et al., 2001; Merton, 1974). Insolvency risk is often predicted using financial ratios reflecting profitability, leverage and self-financing capacity (Grice and Ingram, 2001; Kaya, 2022; Ohlson, 1980; Pindado et al., 2008). More recent studies have also used market-based indicators, such as a firm’s market value relative to debt (Kaya, 2022; Shi et al., 2018). Building on these insights, we argue that foreign divestment can disrupt firm stability, increase financial volatility and amplify established insolvency risk factors.
However, the extent to which foreign divestments affect a parent firm’s insolvency risk depends on characteristics such as scale, strategic motive and the type of divested operation. Strategically motivated divestments, such as shedding a small or unrelated subsidiary to refocus on core activities, tend to cause only minor resource disruptions and limited performance effects (Berry, 2013; Burt et al., 2019; Iurkov and Benito, 2020; Wan et al., 2015). In contrast, frequent restructurings and divestments of international operations can generate instability and operational uncertainty (Brauer and Wiersema, 2012; Erl et al., 2023), thereby heightening insolvency risks. Moreover, divestment decisions are often challenging, as they can harm corporate reputation in both the host and home countries (Resmini and Vittucci Marzetti, 2020) and create ambiguity about postdivestment value creation, combined with limited transparency in strategic and financial details (Brauer, 2006; Brauer and Wiersema, 2012; Lee and Madhavan, 2010). Accordingly, firms with high divestment activity are likely to face greater operational uncertainty and insolvency risk, particularly those that frequently and repeatedly engage in foreign divestments. As Boddewyn (1983, p. 27) rightly noted, “If investment is a hopeful affair like marriage, divestment is more like divorce.”
More companies went insolvent in Finland than in any other year over the past 25 years, surpassing the peak seen during the 2009 financial crisis (Helsinki Times, 2024; Statistics Finland, 2023), which was driven by the effects of the pandemic, the energy crisis stemming from Russia’s war of aggression against Ukraine, and persistent inflation. These external shocks have increased the financial vulnerability of MNEs, forcing parent firms to divest foreign operations and thereby increasing their exposure to insolvency risk. In addition, divestment of a foreign subsidiary can result in various challenges that can lead to insolvency, including decreased cash flow, loss of market share, lower profitability and a firm’s overall weak financial position. Other challenges can include costs and expenses (Mohr et al., 2020), a reduction in revenue and market share (Chen and Jorgensen, 2018), goodwill and intangible asset reduction (Chakrabarti and Mondal, 2017) and a loss in firm value (Marshall et al., 2021). In addition, in the context of the RBV, scholars posit that the reduction of foreign operations may lead to a decrease in the resources of parent firms (Luo et al., 2022; Marshall et al., 2021). Such increased unpredictability and resource loss challenge a firm’s ability to meet financial obligations and sustain a stable resource deployment. Therefore, even in cases in which divestment might yield performance improvements, the associated volatility and resource dislocation inherently elevate insolvency risk – a research gap within the international business literature that this study aims to address.
2.3 Role of corporate social responsibility
Since the introduction of ISO 26000 in 2010, most large organizations have begun systematically disclosing information about their CSR activities in their annual reports. CSR refers to a firm’s integration of social and environmental concerns into business operations and stakeholder interactions (Commission of the European Communities, 2001). Beyond a disclosure practice, CSR represents an intangible resource configuration that supports legitimacy, reputation and trust among stakeholders, key elements of sustained competitive advantage from the RBV perspective (Barney, 1991; Hart, 1995; McWilliams and Siegel, 2011). This framing positions CSR as a strategic capability rather than a purely normative commitment.
Although numerous studies have examined the relationship between CSR and firm performance (e.g. Lopez-Arceiz et al., 2018; Miras Rodriguez et al., 2014; Petrenko et al., 2016; Stanwick and Stanwick, 1998), the findings remain inconclusive and context-dependent (Margolis and Walsh, 2003; Mishra and Suar, 2010; Oeyono et al., 2011). This ambiguity reflects the dual resource nature of CSR from an RBV standpoint. On the one hand, CSR activities can generate valuable intangible assets, such as legitimacy, trust and stakeholder goodwill, which enhance firm resilience and long-term competitiveness (Abu Bakar and Ameer, 2011; Godfrey et al., 2009; Orlitzky et al., 2003; Van Beurden and Gössling, 2008). On the other hand, CSR may represent a costly and inflexible resource, especially when sustained through financial downturns or strategic restructuring, such as foreign divestments, leading to reduced financial slack and potential liquidity strain (Buchanan et al., 2018; Crisostomo et al., 2011). This theoretical duality provides the foundation for exploring how CSR may moderate the relationship between foreign divestment and insolvency risk.
The literature has long recognized the complexity of managing CSR within MNEs (Arnold and Valentin, 2013; Husted and Allen, 2006). CSR embodies a reputation-based resource that enhances firms’ social legitimacy, stakeholder trust and shareholder value, thereby strengthening their relational capital across diverse markets (Albuquerque et al., 2019; Bhattacharya et al., 2011; Hart, 1995; Porter and Kramer, 2006). Such legitimacy-driven resources usually operate as moral capital that provides stakeholder insurance under stable conditions (Godfrey et al., 2009; Luo and Bhattacharya, 2006). However, the moderating role of CSR becomes especially salient under conditions of organizational stress, such as foreign divestment. Although the proverb that “firms do well by doing good” often holds in stable environments (Lys et al., 2015), periods of strategic exit expose the costly and inflexible side of CSR commitments. From an RBV standpoint, such overcommitment and interrelated community programs, labor agreements and environmental obligations transform CSR from a flexible, value-creating capability into a rigid resource that restricts redeployment during divestment. In addition, managers may also occasionally overinvest in CSR for signaling or personal motives (Benabou and Tirole, 2010; Kruger, 2015), generating agency-related inefficiencies that drain financial slack (Buchanan et al., 2018). This paradox highlights why CSR moderates the relationship between foreign divestment and insolvency risk.
Prior research has largely examined CSR’s direct effects on firm performance and valuation outcomes (e.g. Gregory et al., 2014; Jo and Harjoto, 2012; Servaes and Tamayo, 2013), leaving its role in financial distress contexts relatively underexplored (Nguyen and Nguyen, 2015). Yet, understanding CSR in settings of foreign divestment and insolvency is critical because such events test the true resilience and flexibility of firms’ intangible resource portfolios. Drawing on RBV, we conceptualize CSR as a contingent capability whose value depends on context; it can either safeguard or increase insolvency risk depending on the degree of resource constraint created by divestment. This reasoning advances CSR theory by moving beyond performance correlations toward a dynamic understanding of how socially embedded resources interact with crisis-driven restructuring. To empirically capture this moderating effect, we operationalize CSR through firms’ environmental, social and governance (ESG) scores (e.g. Gillan et al., 2021; Shahbaz et al., 2020; Uyar et al., 2023).
3. Data and variables
3.1 Sample and data sources
Our main sample consists of Nordic firms that were publicly listed from 1992 to 2019. We chose 1992 as the starting point for our sample period because insolvency risk data from the National University of Singapore (NUS) are available only from that year onwards. We concluded the sample period in 2019, as our study did not consider the impact of the COVID-19 pandemic. In addition, we included only firms that underwent at least one foreign divestment in a given year. For the divestment data, following previous studies (e.g. Wang and Larimo, 2020), we began by manually collecting information on Nordic firms from the Thomson Reuters Eikon and ORBIS databases. Then, we performed a systematic analysis of the investing firms’ annual reports, press releases and data gathered via company surveys to collect information regarding their divestment activities. We focused exclusively on foreign divestments because domestic divestment events could not be coded consistently across sources (e.g. annual reports, databases, announcements, etc.) over firms and time. Therefore, we restricted our focus to foreign divestments and marked the omission of domestic divestments as a study limitation.
We identified 119 publicly listed Nordic MNEs and began with a full panel of 967 firm-year observations (1992–2019). Nordic firms are often characterized by democratic values, equality, informality and consensus-oriented decision-making (Berg et al., 2024). They also exhibit strong corporate governance structures and low levels of corruption (Thomsen, 2016). We merged the foreign divestment data with the insolvency-risk data collected from the CRI database at NUS and combined this with control variable data from the Orbis database (Bureau van Dijk). When the dependent variable was insolvency risk (CDS spread), coverage limitations reduced the estimation sample to 667 firm-years. For the CSR moderation analysis, the requirement of ESG data further narrowed the sample to 278 firm-years (2003–2019). Each regression model used the largest available set of non-missing observations for its variables. Descriptive statistics for the full sample appear in Table 1, and those for the CDS and ESG available subsamples are shown in Appendix 2.
Descriptive statistics
| Variable | Mean | Median | SD | Min. | Max. | P25 | P75 |
|---|---|---|---|---|---|---|---|
| Foreign divestment | |||||||
| FD | 3.36 | 1.00 | 8.88 | 1.00 | 226.00 | 1.00 | 3.00 |
| Divestment dummy | 0.19 | 0.00 | 0.39 | 0.00 | 1.00 | 0.00 | 0.00 |
| FD winsorized | 3.05 | 1.00 | 4.65 | 1.00 | 32.00 | 1.00 | 3.00 |
| Insolvency risk | |||||||
| CDS | 20.85 | 16.07 | 21.51 | 0.00 | 393.69 | 9.75 | 25.51 |
| Log CDS | 2.71 | 2.77 | 0.87 | −7.36 | 5.98 | 2.28 | 3.24 |
| DD | 4.81 | 4.42 | 3.11 | −2.62 | 54.11 | 2.85 | 6.21 |
| Log DD | 1.39 | 1.50 | 0.69 | −3.48 | 3.99 | 1.06 | 1.83 |
| Control variables | |||||||
| ROA | 5.69 | 5.81 | 11.21 | −202.81 | 157.38 | 3.00 | 9.35 |
| Int. experience | 66.94 | 72.54 | 30.28 | 0.00 | 673.15 | 48.26 | 91.86 |
| Size (log of assets) | 13.41 | 13.42 | 1.81 | 8.44 | 18.81 | 12.03 | 14.77 |
| Total assets (millions) | 2,830.17 | 675.70 | 6,168.51 | 4.63 | 147,512.00 | 167.41 | 2,602.90 |
| Leverage | 0.57 | 0.58 | 0.19 | 0.01 | 3.63 | 0.47 | 0.67 |
| Liquidity | 0.49 | 0.49 | 0.18 | 0.03 | 1.00 | 0.36 | 0.62 |
| Tangibility | 0.28 | 0.26 | 0.18 | 0.00 | 0.83 | 0.14 | 0.41 |
| Sales to assets | 1.07 | 1.02 | 0.48 | −0.10 | 3.83 | 0.78 | 1.33 |
| Tobin’s Q | 0.56 | 0.57 | 0.17 | 0.01 | 3.36 | 0.47 | 0.66 |
| CSR | 53.22 | 55.90 | 20.31 | 3.2 | 93.29 | 39.15 | 68.83 |
| Variable | Mean | Median | Min. | Max. | P25 | P75 | |
|---|---|---|---|---|---|---|---|
| Foreign divestment | |||||||
| 3.36 | 1.00 | 8.88 | 1.00 | 226.00 | 1.00 | 3.00 | |
| Divestment dummy | 0.19 | 0.00 | 0.39 | 0.00 | 1.00 | 0.00 | 0.00 |
| 3.05 | 1.00 | 4.65 | 1.00 | 32.00 | 1.00 | 3.00 | |
| Insolvency risk | |||||||
| 20.85 | 16.07 | 21.51 | 0.00 | 393.69 | 9.75 | 25.51 | |
| Log | 2.71 | 2.77 | 0.87 | −7.36 | 5.98 | 2.28 | 3.24 |
| 4.81 | 4.42 | 3.11 | −2.62 | 54.11 | 2.85 | 6.21 | |
| Log | 1.39 | 1.50 | 0.69 | −3.48 | 3.99 | 1.06 | 1.83 |
| Control variables | |||||||
| 5.69 | 5.81 | 11.21 | −202.81 | 157.38 | 3.00 | 9.35 | |
| Int. experience | 66.94 | 72.54 | 30.28 | 0.00 | 673.15 | 48.26 | 91.86 |
| Size (log of assets) | 13.41 | 13.42 | 1.81 | 8.44 | 18.81 | 12.03 | 14.77 |
| Total assets (millions) | 2,830.17 | 675.70 | 6,168.51 | 4.63 | 147,512.00 | 167.41 | 2,602.90 |
| Leverage | 0.57 | 0.58 | 0.19 | 0.01 | 3.63 | 0.47 | 0.67 |
| Liquidity | 0.49 | 0.49 | 0.18 | 0.03 | 1.00 | 0.36 | 0.62 |
| Tangibility | 0.28 | 0.26 | 0.18 | 0.00 | 0.83 | 0.14 | 0.41 |
| Sales to assets | 1.07 | 1.02 | 0.48 | −0.10 | 3.83 | 0.78 | 1.33 |
| Tobin’s Q | 0.56 | 0.57 | 0.17 | 0.01 | 3.36 | 0.47 | 0.66 |
| 53.22 | 55.90 | 20.31 | 3.2 | 93.29 | 39.15 | 68.83 |
This table reports the descriptive statistics for the variables used in this study. The sample comprises 667 firm-year observations from 119 Nordic MNEs over 1992–2019. It uses all firm-year observations with non-missing divestment, insolvency risk (DD and/or CDS), and control variables. Model-specific Ns may be lower due to additional variable requirements or lags; each model reports its own N
Furthermore, we collected ESG data from the Thomson Reuters Eikon database (now Refinitiv LSEG) to analyze the moderating role of CSR in the foreign divestment-insolvency relationship. The panel covers investments from 1990 to 2017 and foreign divestments between 1992 and 2019, ensuring that all investments survived for a minimum of two years and mitigating the honeymoon effect (Gaur and Lu, 2007; Wang and Larimo, 2020). Appendix 1 provides detailed descriptions of all the variables and data sources. Model-specific observation counts may vary slightly due to missing data, lagged specifications, and differences in variable coverage (e.g. ESG or CDS availability).
3.2 Measures of foreign divestment
Following previous studies (e.g. Amiri et al., 2022; Feldman et al., 2016; Humphery-Jenner et al., 2019), we used the number of foreign divestments by a certain firm in a specific year as the metric for foreign divestment. In operational terms, FD represents the number of foreign divestment deals undertaken by each firm in a given year. In our additional analyses, we also used a dummy foreign divestment variable (FD dummy), which equaled 0 if firms experienced no divestment during a particular year and 1 otherwise [1].
3.3 Measures of insolvency risk
In this study, we used market-based insolvency risk measures because they overcome the criticism of accounting-based models due to the forward-looking nature of market data. These measures reflect the expectations of a firm’s future cash flow, making them appropriate for prediction purposes (Beaver et al., 2005). Consistent with prior studies (Ali et al., 2018; Kabir et al., 2020), we used two proxies for insolvency risk. The main proxy for insolvency risk is the CDS spread. CDSs are credit derivatives that allow for the transfer of a firm’s default risk between two agents for a predetermined time period. In a typical CDS contract, the protection seller provides the protection buyer with insurance against the default of an underlying bond issued by a certain firm (the reference entity). If the reference entity defaults, the seller commits to purchasing the bond from the protection buyer for a price equal to its face value. In exchange for this insurance, the buyer pays a quarterly premium, known as the CDS spread, which is quoted as an annualized percentage of the insured notional value. A higher CDS spread indicates a greater insolvency risk [2].
The alternative proxy for insolvency risk is DD, a concept that originates from Merton’s (1974) structural credit risk model. DD is a widely used measure for assessing the proximity of a limited liability firm to insolvency (e.g. Duan and Wang, 2012; Duan et al., 2012). It is a widely used market-based measure of insolvency risk and outperforms many accounting-based measures in estimating insolvency risk (Bharath and Shumway, 2008; Das et al., 2009; Duan et al., 2012; Duan et al., 2018; Miao et al., 2018). A higher DD indicates lower insolvency risk (i.e. greater firm stability). For methodological details on estimating DD, see Duan et al. (2012).
3.4 Measuring control variables
Building on previous studies (e.g. Ali et al., 2018; Chiang et al., 2015; Schultz et al., 2017), we controlled for certain variables that had the potential to influence insolvency risk. First, we controlled for firm size (Size), measured as the natural logarithm of total assets in millions of US dollars. Generally, larger firms are more stable and have a lower default risk. Thus, we anticipated a negative relationship between firm size and default risk. Second, we included firm profitability, measured by the return on assets (ROA), defined as the ratio of net income to total assets. ROA, derived from accounting data, reflects a firm’s ability to generate sufficient returns to support its operations. Firms with higher ROA typically exhibit lower default risk. Third, we controlled for firm leverage (Leverage), defined as the ratio of total debt to total assets. Leverage reflects the capital structure of a firm, indicating how a firm finances its assets and its ability to meet financial obligations. As leverage increases, firms’ financial risks rise, and we expected a positive relationship between leverage and default risk (Chiang et al., 2015; Schultz et al., 2017).
Fourth, we controlled for liquidity (Liquidity), which measures a firm’s capability to meet short-term obligations. Firms with higher liquidity ratios are expected to have a lower default risk compared to those with lower ratios. Liquidity is captured using the current ratio, calculated as current assets divided by current liabilities. Fifth, we included tangibility (Tangibility), measured as the ratio of net property, plant and equipment to total assets. While tangibility is often negatively associated with firm value (Fukui and Ushijima, 2007; Nakano and Nguyen, 2013), it is expected to have a positive relationship with default risk, as firms with higher tangible assets can face reduced flexibility and increased exposure to fixed costs (Chiang et al., 2015).
Finally, we included international experience (Int. experience), measured as the ratio of foreign sales to total sales. In our additional analyses, we also included Tobin’s Q (TobinQ) as a control variable to proxy firm performance, which incorporates market information and is often considered a superior measure of performance and growth opportunities. Tobin’s Q is calculated by dividing the sum of the fair market value and total liabilities by the total assets. A higher Tobin’s Q, indicative of better performance and stability, is expected to have a negative relationship with default risk (Chiang et al., 2015).
To reduce the influence of outliers in our estimates, we winsorized all other control variables at 1%. We kept the independent variable, foreign divestment, at its original values because we observed no issues regarding outliers in the descriptive statistics. Similarly, we retained the dependent variables (DD and CDS) at their original values, as extreme values in DD and CDS indicate poor performance (insolvency).
3.5 Empirical models
We used panel data with insolvency risk as the dependent variable. Our baseline model to examine the association between divestment and insolvency risk follows several alternative panel regressions of equation (1) below:
In equation (1), the dependent variable, Insolvency Riski,t represents one of two alternative measures of insolvency risk: the DD or CDS spread for firm i at time t. FDi,t captures the number of foreign divestments undertaken by firm i in year t. Several firm-level control variables, as discussed above, are included to account for factors that can affect divestment and insolvency risk. These variables are defined in Appendix 1. The model also incorporates firm-specific fixed effects and year-fixed effects to account for unobservable heterogeneity and temporal variations. In addition, country-fixed effects are controlled for using country dummies based on the first two letters of a firm’s ISIN code. To ensure robustness, all independent variables are winsorized at the 1st and 99th percentiles to mitigate potential outlier effects [3]. Standard errors are clustered at the firm level to account for heteroskedasticity and within-firm correlations.
4. Empirical analysis
4.1 Descriptive statistics and correlations
Consistent with Section 3.1, Table 1 reports the descriptive statistics for the full sample, and descriptive statistics for the ESG subsample used in the CSR models are shown in Appendix 2, which indicates notable heterogeneity in our sample firms in relation to the divestment of foreign subsidiaries. Specifically, the variable FD exhibits a range from 1 (minimum) to 226 (maximum), with an average of 3.36. In terms of the number of divestments, the mean of 3.36 indicates that, on average, there were no more than three divestments per year in our sample. In addition, our sample displays significant heterogeneity in terms of insolvency risk. The variable DD has a minimum value of −2.62 and a maximum value of 54.11. Moreover, the CDS ranges from −7.36 to 5.98, with a mean value of 2.71. Finally, Table 2 indicates that our sample exhibits considerable heterogeneity related to the control variables. For instance, there is substantial variation in size, ranging from US$4.63m to 2.26tr.
Correlations
| Variable | (1) | (2) | (3) | (4) | (5) | (6) | (7) | (8) | (9) | (10) | (11) |
|---|---|---|---|---|---|---|---|---|---|---|---|
| (1) CDS5y | 1 | ||||||||||
| (2) DtD | −0.584*** | 1 | |||||||||
| (3) max_divest | 0.0662 | −0.0796* | 1 | ||||||||
| (4) size | −0.0388 | −0.0132 | 0.102* | 1 | |||||||
| (5) ROA | −0.369*** | 0.299*** | 0.0611 | 0.0227 | 1 | ||||||
| (6) leverage | 0.299*** | −0.447*** | −0.0389 | 0.0421 | −0.291*** | 1 | |||||
| (7) liquidity | −0.0683 | 0.0556 | 0.00163 | −0.228*** | 0.196*** | 0.0138 | 1 | ||||
| (8) tangibility | 0.0775 | −0.122** | 0.0992* | 0.123** | −0.100* | −0.0492 | −0.478*** | 1 | |||
| (9) sales_to_assets | −0.0370 | −0.0350 | 0.0268 | −0.451*** | 0.0743 | 0.00326 | 0.470*** | −0.0830* | 1 | ||
| (10) tobinq | 0.298*** | −0.444*** | −0.0393 | 0.0419 | −0.289*** | 1.000*** | 0.0161 | −0.0514 | 0.00425 | 1 | |
| (11) int_experience | −0.0209 | 0.0788 | 0.0508 | 0.277*** | −0.0734 | −0.0148 | 0.0516 | −0.151*** | −0.142*** | −0.0145 | 1 |
| Variable | (1) | (2) | (3) | (4) | (5) | (6) | (7) | (8) | (9) | (10) | (11) |
|---|---|---|---|---|---|---|---|---|---|---|---|
| (1) CDS5y | 1 | ||||||||||
| (2) DtD | −0.584 | 1 | |||||||||
| (3) max_divest | 0.0662 | −0.0796 | 1 | ||||||||
| (4) size | −0.0388 | −0.0132 | 0.102 | 1 | |||||||
| (5) | −0.369 | 0.299 | 0.0611 | 0.0227 | 1 | ||||||
| (6) leverage | 0.299 | −0.447 | −0.0389 | 0.0421 | −0.291 | 1 | |||||
| (7) liquidity | −0.0683 | 0.0556 | 0.00163 | −0.228 | 0.196 | 0.0138 | 1 | ||||
| (8) tangibility | 0.0775 | −0.122 | 0.0992 | 0.123 | −0.100 | −0.0492 | −0.478 | 1 | |||
| (9) sales_to_assets | −0.0370 | −0.0350 | 0.0268 | −0.451 | 0.0743 | 0.00326 | 0.470 | −0.0830 | 1 | ||
| (10) tobinq | 0.298 | −0.444 | −0.0393 | 0.0419 | −0.289 | 1.000 | 0.0161 | −0.0514 | 0.00425 | 1 | |
| (11) int_experience | −0.0209 | 0.0788 | 0.0508 | 0.277 | −0.0734 | −0.0148 | 0.0516 | −0.151 | −0.142 | −0.0145 | 1 |
This table reports the pairwise correlations for the variables used in the empirical analysis. ***, ** and * denote significance at the 0.01, 0.05 and 0.10 levels, respectively. See Appendix 1 for the variable definitions
Table 2 shows the pairwise correlations among the variables used in the analysis. It is evident from the table that the number of foreign divestments has a negative correlation with DD and a positive correlation with CDS, implying that a higher frequency of divestments of foreign subsidiaries corresponds to an increased level of insolvency risk. Moreover, as expected, the two insolvency risk variables, DD and CDS, are negatively correlated by construction (r = 0.93). As the correlation results are not controlled by other factors that may affect the insolvency risk, they should be interpreted with prudence.
4.2 Baseline regression results
Table 3 presents the results for the six alternative specifications of equation (1). Models 1–3 use DD as the dependent variable, while Models 4–6 use CDS as the dependent variable. Models 1 and 4 adopt a parsimonious specification, including only Size, ROA and Leverage as control variables. Models 2 and 5 expand the analysis by incorporating the full set of control variables, along with year fixed effects. Finally, Models 3 and 6 provide the most comprehensive specifications, including both year and firm fixed effects, in addition to the full set of control variables. In addition, all models account for country-fixed effects.
Foreign divestment and insolvency risk
| Distance to default | Credit default swap spread | |||||
|---|---|---|---|---|---|---|
| DD | DD | DD | CDS | CDS | CDS | |
| Variable | model (1) | model (2) | model (3) | model (4) | model (5) | model (6) |
| FD | −0.044** (−2.148) | −0.049** (−2.174) | −0.043* (−1.863) | 0.016*** (2.757) | 0.017*** (2.772) | 0.017*** (2.844) |
| Size | 0.054 (1.184) | 0.011 (0.213) | −0.385* (−1.736) | 0.025 (1.619) | 0.028 (1.609) | 0.131* (1.761) |
| ROA | 0.069*** (4.161) | 0.062*** (3.747) | 0.028* (1.854) | −0.025*** (−4.017) | −0.023*** (−3.668) | −0.011* (−1.754) |
| Leverage | −6.153*** (−8.012) | −6.825*** (−8.802) | −4.594*** (−5.523) | 1.821*** (8.234) | 1.904*** (7.924) | 1.036*** (3.301) |
| Liquidity | 0.142 (0.237) | 0.804 (0.631) | 0.168 (0.973) | −0.455 (−1.121) | ||
| Tangibility | −0.760 (−1.164) | 0.006 (0.005) | 0.361* (1.806) | 0.330 (0.748) | ||
| Int_experience | 0.006* (1.877) | 0.003 (0.705) | −0.001 (−1.264) | 0.000 (0.156) | ||
| Constant | 4.060*** (4.947) | 5.036*** (5.309) | 6.293* (1.875) | 2.238*** (8.265) | 2.023*** (6.732) | 1.027 (0.962) |
| Observations | 656 | 625 | 625 | 667 | 635 | 635 |
| R-squared | 0.508 | 0.522 | 0.704 | 0.387 | 0.394 | 0.601 |
| Compno | 119 | 108 | 108 | 123 | 111 | 111 |
| Year FE | YES | YES | YES | YES | YES | YES |
| Firm FE | NO | NO | YES | NO | NO | YES |
| Country FE | YES | YES | YES | YES | YES | YES |
| Distance to default | Credit default swap spread | |||||
|---|---|---|---|---|---|---|
| Variable | model (1) | model (2) | model (3) | model (4) | model (5) | model (6) |
| −0.044 | −0.049 | −0.043 | 0.016 | 0.017 | 0.017 | |
| Size | 0.054 (1.184) | 0.011 (0.213) | −0.385 | 0.025 (1.619) | 0.028 (1.609) | 0.131 |
| 0.069 | 0.062 | 0.028 | −0.025 | −0.023 | −0.011 | |
| Leverage | −6.153 | −6.825 | −4.594 | 1.821 | 1.904 | 1.036 |
| Liquidity | 0.142 (0.237) | 0.804 (0.631) | 0.168 (0.973) | −0.455 (−1.121) | ||
| Tangibility | −0.760 (−1.164) | 0.006 (0.005) | 0.361 | 0.330 (0.748) | ||
| Int_experience | 0.006 | 0.003 (0.705) | −0.001 (−1.264) | 0.000 (0.156) | ||
| Constant | 4.060 | 5.036 | 6.293 | 2.238 | 2.023 | 1.027 (0.962) |
| Observations | 656 | 625 | 625 | 667 | 635 | 635 |
| R-squared | 0.508 | 0.522 | 0.704 | 0.387 | 0.394 | 0.601 |
| Compno | 119 | 108 | 108 | 123 | 111 | 111 |
| Year | ||||||
| Firm | ||||||
| Country | ||||||
The table reports the estimates of Six alternative versions of the following panel regression specification:
where the dependent variable, Insolvency Riski,t represents One of Two alternative measures of insolvency risk: the DD or CDS spread for firm i at time t. FDi,t captures the number of foreign divestments undertaken by firm i in year t. Several firm-level control variables, as discussed above, are included to account for factors that may affect divestment and insolvency risk. These variables are defined in Appendix 1. is a dummy variable for firm i, and is a dummy variable for fiscal years. In addition, country-fixed effects are controlled for using country dummies based on the first Two letters of a firm’s ISIN code. The reported adjusted R2s are the overall R2s that account for the explanatory power of the firm and year fixed effects. The t-statistics (reported in parentheses) are based on robust standard errors clustered at the firm level. ***, ** and * denote significance at the 0.01, 0.05 and 0.10 levels, respectively
Table 3 shows that the number of foreign divestments has a statistically significant and negative coefficient in Models 1, 2 and 3 when DD is the dependent variable. These results indicate that an increased divestment of foreign subsidiaries amplifies the insolvency risk of parent firms. In addition, Table 3 shows coefficient estimates for the number of foreign divestments that are positively associated with the CDS spread (in Models 4, 5 and 6). This indicates that a greater frequency of divestment of foreign subsidiaries corresponds to an elevated level of insolvency risk. The coefficients related to firm-specific variables also provide important insights. For instance, firm Size is positively connected to the CDS spread (in Model 6) and negatively linked to DD (in Model 3), suggesting that larger firms carry higher insolvency risk. This finding is consistent with the view that larger firms tend to take more risks. As expected, the return on assets has a negative relationship with the CDS spread and a positive relationship with DD, suggesting that firm stability increases with greater profitability. Finally, our findings indicate that firms with higher leverage are exposed to increased insolvency risk.
In summary, Table 3 indicates that firms engaging in a higher frequency of divestment activities are associated with heightened insolvency risk. These findings provide an answer to RQ1, demonstrating that firms divesting their foreign subsidiaries are prone to greater insolvency risk. Overall, the findings reported in Table 3 broadly align with the existing literature on firm risk (see, e.g. Mohr et al., 2020).
4.3 Lagged variables
Regressions based on contemporaneous variables are susceptible to endogeneity bias due to reverse causality. In contrast, a regression based on lagged values of independent variables helps control for reverse causality and thus tends to be less vulnerable to endogeneity effects. To address this concern, we followed Li et al. (2021) and re-estimated equation (1) by using the number of foreign divestments by firms and control variables in the prior period (t – 1), along with the CDS (DD) from the current period (t). The results reported in Table 4 are similar to those presented in Table 3. The relation between divestment and insolvency risk remains significantly negative with DD and positive with CDS in the one period lag specification. These results support the interpretation that firms’ divestment actions influence the current insolvency risk, rather than the alternate interpretation of past insolvency risk influencing the current divestment activity. These findings provide additional support for the baseline results and suggest that the foreign divestment of subsidiaries can predict insolvency risk. That is, a high level of divestment activity in the current period prompts a higher level of insolvency risk in the subsequent period.
Lagged foreign divestment and insolvency risk
| Distance to default | Credit default swap spread | |||||
|---|---|---|---|---|---|---|
| DD | DD | DD | CDS | CDS | CDS | |
| Variable | model (1) | model (2) | model (3) | model (4) | model (5) | model (6) |
| FD | −0.053*** (−3.273) | −0.056*** (−3.434) | −0.040** (−2.485) | 0.019*** (2.975) | 0.020*** (3.131) | 0.016** (2.151) |
| Size | 0.063 (1.292) | 0.015 (0.278) | −0.445* (−1.890) | 0.022 (1.362) | 0.030* (1.747) | 0.174** (2.198) |
| ROA | 0.048*** (2.661) | 0.046*** (2.594) | 0.014 (1.064) | −0.013** (−2.501) | −0.013** (−2.438) | −0.001 (−0.191) |
| Leverage | −6.335*** (−7.554) | −6.958*** (−8.413) | −4.317*** (−5.044) | 1.893*** (6.952) | 2.068*** (8.013) | 1.123*** (3.563) |
| Liquidity | −1.809** (−2.466) | −1.926 (−1.419) | 0.745*** (3.398) | 0.408 (0.956) | ||
| Langibility | −2.295*** (−3.418) | −2.182* (−1.674) | 0.782*** (3.906) | 0.822* (1.697) | ||
| Int_experience | 0.004 (1.228) | 0.003 (0.860) | −0.001 (−0.733) | −0.001 (−0.446) | ||
| Constant | 4.276*** (4.259) | 6.570*** (5.926) | 7.755** (2.249) | 2.075*** (6.560) | 1.349*** (3.992) | −0.722 (−0.647) |
| Observations | 656 | 625 | 625 | 666 | 634 | 634 |
| R-squared | 0.468 | 0.501 | 0.730 | 0.313 | 0.353 | 0.608 |
| Compno | 114 | 103 | 103 | 118 | 106 | 106 |
| Year FE | YES | YES | YES | YES | YES | YES |
| Firm FE | NO | NO | YES | NO | NO | YES |
| Country FE | YES | YES | YES | YES | YES | YES |
| Distance to default | Credit default swap spread | |||||
|---|---|---|---|---|---|---|
| Variable | model (1) | model (2) | model (3) | model (4) | model (5) | model (6) |
| −0.053 | −0.056 | −0.040 | 0.019 | 0.020 | 0.016 | |
| Size | 0.063 (1.292) | 0.015 (0.278) | −0.445 | 0.022 (1.362) | 0.030 | 0.174 |
| 0.048 | 0.046 | 0.014 (1.064) | −0.013 | −0.013 | −0.001 (−0.191) | |
| Leverage | −6.335 | −6.958 | −4.317 | 1.893 | 2.068 | 1.123 |
| Liquidity | −1.809 | −1.926 (−1.419) | 0.745 | 0.408 (0.956) | ||
| Langibility | −2.295 | −2.182 | 0.782 | 0.822 | ||
| Int_experience | 0.004 (1.228) | 0.003 (0.860) | −0.001 (−0.733) | −0.001 (−0.446) | ||
| Constant | 4.276 | 6.570 | 7.755 | 2.075 | 1.349 | −0.722 (−0.647) |
| Observations | 656 | 625 | 625 | 666 | 634 | 634 |
| R-squared | 0.468 | 0.501 | 0.730 | 0.313 | 0.353 | 0.608 |
| Compno | 114 | 103 | 103 | 118 | 106 | 106 |
| Year | ||||||
| Firm | ||||||
| Country | ||||||
This table reports the estimates of Six alternative versions of the following panel regression specification:
where the dependent variable, Insolvency Riski,t represents One of Two alternative measures of insolvency risk: the DD or CDS spread for firm i at time t. FDi,t captures the number of foreign divestments undertaken by firm i in year t. Several firm-level control variables, as discussed above, are included to account for factors that may affect divestment and insolvency risk. These variables are defined in Appendix 1. is a dummy variable for firm i, and is a dummy variable for fiscal years. In addition, country-fixed effects are controlled for using country dummies based on the first Two letters of a firm’s ISIN code. The reported adjusted R2s are the overall R2s that account for the explanatory power of the firm and year fixed effects. The t-statistics (reported in parentheses) are based on robust standard errors clustered at the firm level. ***, ** and * denote significance at the 0.01, 0.05 and 0.10 levels, respectively
4.4 Moderating role of corporate social responsibility
To examine the moderating role of CSR in the divestment-insolvency relationship, we reestimated equation (1) by including an interaction variable, Foreign Divestment × high CSR. In this context, high CSR denotes firms that fall in the top quartile based on their CSR activities. The results, as reported in Table 5, feature DD as the dependent variable in Models 1–3 and CDS as the dependent variable in Models 4–6. Consistent with the findings addressing RQ2, the interaction terms exhibit negative and statistically significant coefficients with DD (in Models 1–3). This suggests that parent firms with higher levels of CSR activities face increased insolvency risk when involved in foreign divestments, as the DD reduces. Similarly, the interaction terms demonstrate positive and statistically significant coefficients with CDS (in Models 4–6), reinforcing the notion that parent firms with higher CSR activities face increased insolvency risk in the case of foreign divestments as the CDS increases. An alternative explanation of these findings suggests that the higher level of CSR activities is considered costly, time consuming and can harm trust and resources when divestment is used as a strategy to reduce the financial stability of parent firms.
CSR Moderation of the divestment-insolvency relationship
| Distance to default | Credit default swap spread | |||||
|---|---|---|---|---|---|---|
| DD | DD | DD | CDS | CDS | CDS | |
| Variable | model (1) | model (2) | model (3) | model (4) | model (5) | model (6) |
| FD | 0.021 (0.843) | 0.025 (0.953) | 0.042 (1.522) | −0.004 (−0.534) | −0.003 (−0.413) | −0.005 (−0.511) |
| FD × high CSR | −0.107** (−2.284) | −0.121** (−2.320) | −0.146*** (−3.339) | 0.033** (2.347) | 0.034** (2.167) | 0.037*** (2.643) |
| High CSR | 0.451** (2.248) | 0.535*** (2.593) | 0.528** (2.480) | −0.133** (−2.129) | −0.140** (−2.207) | −0.099 (−1.298) |
| Size | 0.064 (1.470) | 0.022 (0.445) | −0.458** (−2.022) | 0.022 (1.494) | 0.025 (1.487) | 0.153** (2.033) |
| ROA | 0.071*** (4.227) | 0.064*** (3.811) | 0.030** (1.983) | −0.025*** (−4.092) | −0.024*** (−3.735) | −0.011* (−1.841) |
| Leverage | −6.191*** (−8.096) | −6.852*** (−8.951) | −4.785*** (−5.820) | 1.833*** (8.317) | 1.913*** (8.039) | 1.071*** (3.408) |
| Liquidity | 0.238 (0.393) | 0.793 (0.616) | 0.142 (0.824) | −0.455 (−1.113) | ||
| Tangibility | −0.824 (−1.264) | −0.024 (−0.020) | 0.375* (1.883) | 0.310 (0.691) | ||
| Int. experience | 0.006* (1.936) | 0.003 (0.780) | −0.001 (−1.331) | 0.000 (0.100) | ||
| Constant | 3.588*** (4.511) | 4.426*** (4.820) | 7.233** (2.119) | 2.373*** (8.852) | 2.178*** (7.415) | 0.809 (0.763) |
| Observations | 278 | 277 | 277 | 278 | 277 | 635 |
| R-squared | 0.515 | 0.530 | 0.713 | 0.394 | 0.402 | 0.608 |
| Compno | 53 | 53 | 53 | 53 | 53 | 53 |
| Year FE | YES | YES | YES | YES | YES | YES |
| Firm FE | NO | NO | YES | NO | NO | YES |
| Country FE | YES | YES | YES | YES | YES | YES |
| Distance to default | Credit default swap spread | |||||
|---|---|---|---|---|---|---|
| Variable | model (1) | model (2) | model (3) | model (4) | model (5) | model (6) |
| 0.021 (0.843) | 0.025 (0.953) | 0.042 (1.522) | −0.004 (−0.534) | −0.003 (−0.413) | −0.005 (−0.511) | |
| −0.107 | −0.121 | −0.146 | 0.033 | 0.034 | 0.037 | |
| High | 0.451 | 0.535 | 0.528 | −0.133 | −0.140 | −0.099 (−1.298) |
| Size | 0.064 (1.470) | 0.022 (0.445) | −0.458 | 0.022 (1.494) | 0.025 (1.487) | 0.153 |
| 0.071 | 0.064 | 0.030 | −0.025 | −0.024 | −0.011 | |
| Leverage | −6.191 | −6.852 | −4.785 | 1.833 | 1.913 | 1.071 |
| Liquidity | 0.238 (0.393) | 0.793 (0.616) | 0.142 (0.824) | −0.455 (−1.113) | ||
| Tangibility | −0.824 (−1.264) | −0.024 (−0.020) | 0.375 | 0.310 (0.691) | ||
| Int. experience | 0.006 | 0.003 (0.780) | −0.001 (−1.331) | 0.000 (0.100) | ||
| Constant | 3.588 | 4.426 | 7.233 | 2.373 | 2.178 | 0.809 (0.763) |
| Observations | 278 | 277 | 277 | 278 | 277 | 635 |
| R-squared | 0.515 | 0.530 | 0.713 | 0.394 | 0.402 | 0.608 |
| Compno | 53 | 53 | 53 | 53 | 53 | 53 |
| Year | ||||||
| Firm | ||||||
| Country | ||||||
This table reports the regression results for the moderating role of CSR on the relationship between foreign divestment and insolvency risk. The dependent variable, Insolvency Riski,t represents One of Two alternative measures of insolvency risk: the DD or CDS spread for firm i at time t. FDi,t captures the number of foreign divestments undertaken by firm i in year t. high CSR denotes firms that fall in the top quartile based on their CSR activities. The reported adjusted R2s are the overall R2s that account for the explanatory power of the firm and year fixed effects. The t-statistics (reported in parentheses) are based on robust standard errors clustered at the firm level. ***, ** and * denote significance at the 0.01, 0.05 and 0.10 levels, respectively. The sample is restricted to firm-years with ESG data (2003–2019). Model-specific observations may differ due to additional variable inclusion and non-missing requirements (e.g. lags, controls)
4.5 Additional analysis
To check the robustness of our empirical findings, we performed several additional tests. First, we examined the relationship between foreign divestment and insolvency risk using an alternative proxy for insolvency risk: the probability of default. We used a probability of default for one month, three months, six months, one year, two years, three years and five years in our analyses. The results (untabulated) show that higher divestment of foreign subsidiaries is positively associated with the probability of default. These results suggest that higher divestment activity by parent firms leads to a greater insolvency risk.
Second, we assessed robustness by excluding observations from 2007 to 2009 to mitigate potential global financial crisis (GFC) effects, which may have temporarily increased foreign divestments and influenced our results. The (untabulated) results show that our study findings did not change due to abnormal observations of GFC. Our use of year fixed effects also controlled for macroeconomic shocks, including major global crises across the sample period.
Third, while firm size was controlled for in our main analysis to account for its baseline effects, we conducted additional analyses to explore whether firm size moderates the relationship between foreign divestment and insolvency risk. The (untabulated) results show that as the size of parent firms increases, their insolvency risk decreases in the event of foreign divestment and vice versa. These results are in line with earlier studies (e.g. Berry, 2013; Daley et al., 1997; Schlingemann et al., 2002), which state that larger firms try to divest unrelated and poorly performing subsidiaries to focus on the core business. Fourth, we found that international experience had no impact on the relationship between foreign divestment and the insolvency risk of parent firms. The untabulated results show that foreign divestment has the same impact on insolvency risk for firms with higher or lower international experience.
Finally, to further validate our main findings, we re-estimated equation (1) using an alternative measure of foreign divestment: a binary divestment dummy variable (FD dummy), which equals 1 if a firm undertook any foreign divestment in a given year, and 0 otherwise. The results (untabulated) show a consistent pattern with our main results, i.e. the divestment dummy is positively associated with the CDS spread (higher insolvency risk) and negatively associated with DD (lower stability). These supplementary analyses further reinforced our main findings, suggesting that foreign divestment significantly increases the insolvency risk of parent firms.
5. Conclusion
Although prior research has mainly focused on the antecedents of divestment, the outcomes of divestment have received comparatively limited attention (Mohr et al., 2020). This study investigated the relationship between foreign divestment and the insolvency risk of parent firms, focusing on Nordic MNEs and using both traditional (Merton’s DD spread) and innovative (CDS spread) market-based measures of insolvency risk. Our findings reveal a positive association between foreign divestment and increased insolvency risk for parent firms, offering new insights into the financial vulnerabilities introduced by divestment decisions. This contribution extends the international business and firm risk literature, emphasizing the effects of resource loss associated with foreign subsidiary divestments (e.g., Batsakis et al., 2024; Brauer and Wiersema, 2012; Schmid and Morschett, 2023; Teschner and Paul, 2021).
In addition to this core relationship, we examined the moderating role of CSR activities. While CSR is often associated with positive organizational outcomes (e.g. Orlitzky et al., 2003), some studies have highlighted its potential financial downsides (e.g. Crisostomo et al., 2011). Consistent with this perspective, our findings show that higher levels of CSR activities exacerbate the insolvency risk associated with foreign divestments. This suggests that despite fostering goodwill and enhancing reputation, CSR initiatives can impose significant resource demands, intensifying financial strain during crises. These findings contribute to the CSR literature by shifting the focus from success metrics, such as profitability, to the less-explored area of financial risk and failure.
Our research also addresses gaps in international business literature by exploring the implications of foreign divestments on credit market reactions and providing evidence of how bondholders perceive the risks faced by parent firms in the context of divestment. By linking foreign divestments to CDS spreads, this study underscores the critical need for MNEs to consider the financial repercussions of their strategic divestment decisions on firm solvency. These insights complement prior studies that have largely focused on operational and performance outcomes (e.g. Denning, 1988; Hillier et al., 2009).
For managers, these findings highlight the importance of integrating financial risk assessments into divestment planning in addition to immediate operational considerations. Before foreign divestment commences, firms need to carefully evaluate the potential impacts it has on resource configurations, performance stability and insolvency risk. Parent firms with high CSR commitments should particularly assess whether such investments remain sustainable and beneficial during restructuring periods, as excessive CSR spending may amplify firms’ financial distress.
Theoretically, our research makes important contributions by being among the first to link foreign divestment to parent firms’ insolvency risk, expanding the research beyond performance outcomes to financial stability and risk management among multinational firms. By using the RBV in this context, we show how divestments can destabilize parent firms, leading to their increased insolvency risk. In addition, we extend the CSR literature by explaining that, while CSR is often seen as beneficial, it may increase the financial vulnerability of firms in cases of foreign divestment. Together, these contributions offer a new perspective on the financial outcomes of foreign divestments and open new avenues for future research on risk management in the field of international business.
However, this study is not without limitations. Our analysis covers foreign divestments only, as consistent data on domestic divestments are hard to collect. While the RBV logic applies equally to domestic exits, cutting proximate, long-standing home-market linkages could create even greater resource disruption and signals of distress than foreign exits. Therefore, our findings may represent a moderate lower bound of the broader divestment and insolvency relationship. Future research could test this by incorporating domestic divestments. Still, the fact that even geographically distant foreign exits raise insolvency risk highlights the general importance of divestment-related solvency management.
The scope of this study is restricted to Nordic firms and relies on specific data sources, which may limit its generalizability to other regions or contexts. Future research could extend this work by examining larger, more diverse samples and exploring additional factors, such as industry-specific characteristics or variations in market conditions, that may influence the relationship between foreign divestments, CSR and insolvency risk. Moreover, subsidiaries divested in developing countries may have a significantly different impact on this relationship, particularly when considering the role of CSR activities, compared to those divested in developed countries. Future studies could also explore how motivations, divestment scale and voluntary vs involuntary divestments (Makino et al., 2007; McDermott, 2010) shape insolvency risk.
This study includes only firms that have undertaken foreign divestments, which may introduce selection bias, as these firms could systematically differ from nondivesting firms. Due to data limitations, we were unable to apply correction techniques, such as the Heckman two-stage model. In addition, our divestment measure does not differentiate by scale or type, potentially oversimplifying the divestment impact. This measurement may weaken the ability to capture theoretical nuances and limit the internal validity. Future research could incorporate more divestment characteristics to improve construct validity. Finally, future research could also examine whether divestment is not only a cause of financial vulnerability but also a symptom of poor firm alignment, as suggested by transaction cost and internalization theories (Hennart, 1982).
In summary, this study advances the understanding of the interplay between foreign divestments and insolvency risk by incorporating the role of CSR activities. It provides valuable theoretical and practical insights for MNEs navigating the complexities of divestment and resource management in volatile global markets. Future research could further unravel these intricate dynamics, offering a more comprehensive understanding of firm survival in the context of foreign divestments.
Acknowledgements
The authors would like to thank Robert Czernkowski, Pervez Ghauri, Carlos Sousa, Min Zhu and participants at the BAFA Joint Conference of the Corporate Finance & Asset Pricing SIG and the Northern Area Group, CIMaR 29th Annual Conference, the 2023 AFAANZ conference, 17th Vaasa International Business Conference, the Brownbag Seminar at the University of Queensland Business School and the Seminar at University of Vaasa for the valuable comments and suggestions. Arshed Iqbal gratefully acknowledges financial support from the Liikesivistysrahasto. Jamshed Iqbal gratefully acknowledges financial support from Suomen Arvopaperimarkkinoiden Edistämissäätiö, Suomen Kulttuurirahasto (grant number: 00230480) and Liikesivistysrahasto (grant number: 200116 and 220140). Any errors are their own.
Notes
The data used in this study was manually collected and provides little information on the deal size.
The CRI refers to the CDS spread as the “actuarial spread.” The actuarial spread is built using the traditional CDS design but without an upfront fee, under the assumption that market participants are risk neutral. Thus, the actuarial spread has the same features as the standard CDS spread. This adjustment allows for calculation of the CDS spread for a large number of firms.
We adopt the approach of Ellul and Yerramilli (2013) and apply winsorization to the independent variables. It is worth noting that the results remain robust even when not applying winsorization.
References
Appendix 1
Variables
| Variable name | Notation | Definition | Source |
|---|---|---|---|
| Foreign divestment | |||
| Foreign divestment | FD | Number of foreign divestment deals a firm completed in a given year. Each transaction represents a distinct divestment of a foreign subsidiary or equity stake, identified from different sources. Conceptually, it reflects any voluntary or forced reduction of a firm’s engagement in cross-border activities (Benito and Welch, 1997) | Primary data from annual reports, press releases, personal contacts, and databases |
| Foreign divestment dummy | FD dummy | Equals 1 if the firm undertook at least One foreign divestment in a given year and 0 otherwise | Primary data |
| Insolvency risk | |||
| Credit default swap spread | CDS | Credit derivative spread that transfers a firm’s default risk between Two agents for a fixed time period; higher values indicate greater insolvency risk | RMI-NUS |
| Distance to default | DD | Annual average of a firm’s distance to default based on stock price volatility; higher values indicate greater stability | RMI-NUS |
| CSR activities | |||
| Corporate social responsibility score | CSR score | Composite ESG score (0–100) represents the level of corporate social responsibility activity performed by a firm | LSEG (refinitiv eikon) |
| High CSR | Dummy variable equal to 1 for firms in the top quartile of CSR scores, 0 otherwise | LSEG | |
| Controls | |||
| Firm size | Size | Natural logarithm of total assets | LSEG |
| Leverage | Leverage | Ratio of total debt to total assets (TLTA), reflecting a firm’s capital structure | LSEG |
| Liquidity | Liquidity | Current assets divided by current liabilities, measuring short-term solvency | LSEG |
| Profitability | ROA | Net income divided by total assets | LSEG |
| Firm performance | Tobin’s Q | Market value of a firm divided by the replacement cost of its assets | LSEG |
| Tangibility | Tangibility | Ratio of net property, plant, and equipment to total assets | LSEG |
| International experience | Int. Experience | Ratio of foreign sales to total sales | LSEG |
| Variable name | Notation | Definition | Source |
|---|---|---|---|
| Foreign divestment | |||
| Foreign divestment | Number of foreign divestment deals a firm completed in a given year. Each transaction represents a distinct divestment of a foreign subsidiary or equity stake, identified from different sources. Conceptually, it reflects any voluntary or forced reduction of a firm’s engagement in cross-border activities ( | Primary data from annual reports, press releases, personal contacts, and databases | |
| Foreign divestment dummy | Equals 1 if the firm undertook at least One foreign divestment in a given year and 0 otherwise | Primary data | |
| Insolvency risk | |||
| Credit default swap spread | Credit derivative spread that transfers a firm’s default risk between Two agents for a fixed time period; higher values indicate greater insolvency risk | RMI-NUS | |
| Distance to default | Annual average of a firm’s distance to default based on stock price volatility; higher values indicate greater stability | RMI-NUS | |
| Corporate social responsibility score | Composite | ||
| High | Dummy variable equal to 1 for firms in the top quartile of | ||
| Controls | |||
| Firm size | Size | Natural logarithm of total assets | |
| Leverage | Leverage | Ratio of total debt to total assets ( | |
| Liquidity | Liquidity | Current assets divided by current liabilities, measuring short-term solvency | |
| Profitability | Net income divided by total assets | ||
| Firm performance | Tobin’s Q | Market value of a firm divided by the replacement cost of its assets | |
| Tangibility | Tangibility | Ratio of net property, plant, and equipment to total assets | |
| International experience | Int. Experience | Ratio of foreign sales to total sales |
Appendix 2
Descriptive statistics (Sub-sample)
| Variable | N | Mean | SD | P25 | Median | P75 | Min. | Max. |
|---|---|---|---|---|---|---|---|---|
| CDS spread (bps) | 667 | 20.42634 | 19.89703 | 10.64611 | 16.81137 | 25.38325 | 1.344162 | 393.6864 |
| Log (CDS spread) | 667 | 2.77919 | 0.688815 | 2.365193 | 2.822055 | 3.23409 | 0.295771 | 5.975554 |
| Divestment (max_divest) | 667 | 2.827586 | 4.265085 | 1 | 2 | 3 | 1 | 65 |
| Size | 667 | 14.44354 | 1.559303 | 13.41783 | 14.60837 | 15.50866 | 9.493412 | 17.69237 |
| ROA | 667 | 5.803388 | 7.398989 | 3.055 | 5.66 | 8.67 | −45.64 | 60.55 |
| Leverage | 667 | 0.573269 | 0.13226 | 0.501597 | 0.590334 | 0.657433 | 0.08085 | 1.646398 |
| Liquidity | 661 | 0.462994 | 0.158952 | 0.350731 | 0.447818 | 0.567071 | 0.077318 | 0.998794 |
| Tangibility | 667 | 0.290807 | 0.160906 | 0.163503 | 0.262052 | 0.41378 | 0 | 0.687675 |
| Sales-to-Assets | 667 | 1.054401 | 0.452586 | 0.795486 | 0.995409 | 1.202612 | 0 | 3.829487 |
| Tobin’s Q | 648 | 0.572374 | 0.132489 | 0.501234 | 0.589097 | 0.657474 | 0.081354 | 1.646528 |
| International experience | 640 | 72.4659 | 25.4371 | 52.73 | 82.82 | 94.21 | 0 | 111.95 |
| ESG score | 278 | 56.63514 | 19.46941 | 43.4625 | 60.12 | 70.3225 | 3.55 | 93.29 |
| Distance to default | 656 | 4.681002 | 2.349801 | 3.020371 | 4.476373 | 5.988119 | −0.82999 | 14.94562 |
| Variable | N | Mean | P25 | Median | P75 | Min. | Max. | |
|---|---|---|---|---|---|---|---|---|
| 667 | 20.42634 | 19.89703 | 10.64611 | 16.81137 | 25.38325 | 1.344162 | 393.6864 | |
| Log ( | 667 | 2.77919 | 0.688815 | 2.365193 | 2.822055 | 3.23409 | 0.295771 | 5.975554 |
| Divestment (max_divest) | 667 | 2.827586 | 4.265085 | 1 | 2 | 3 | 1 | 65 |
| Size | 667 | 14.44354 | 1.559303 | 13.41783 | 14.60837 | 15.50866 | 9.493412 | 17.69237 |
| 667 | 5.803388 | 7.398989 | 3.055 | 5.66 | 8.67 | −45.64 | 60.55 | |
| Leverage | 667 | 0.573269 | 0.13226 | 0.501597 | 0.590334 | 0.657433 | 0.08085 | 1.646398 |
| Liquidity | 661 | 0.462994 | 0.158952 | 0.350731 | 0.447818 | 0.567071 | 0.077318 | 0.998794 |
| Tangibility | 667 | 0.290807 | 0.160906 | 0.163503 | 0.262052 | 0.41378 | 0 | 0.687675 |
| Sales-to-Assets | 667 | 1.054401 | 0.452586 | 0.795486 | 0.995409 | 1.202612 | 0 | 3.829487 |
| Tobin’s Q | 648 | 0.572374 | 0.132489 | 0.501234 | 0.589097 | 0.657474 | 0.081354 | 1.646528 |
| International experience | 640 | 72.4659 | 25.4371 | 52.73 | 82.82 | 94.21 | 0 | 111.95 |
| 278 | 56.63514 | 19.46941 | 43.4625 | 60.12 | 70.3225 | 3.55 | 93.29 | |
| Distance to default | 656 | 4.681002 | 2.349801 | 3.020371 | 4.476373 | 5.988119 | −0.82999 | 14.94562 |

