The strategic importance of corporate social responsibility (CSR) has been widely recognized in recent years as firms’ growth and survival rely on their managers’ ability to satisfy stakeholders’ interests and demands (Aguilera 2007; Freeman et al., 2007; King, 2008). Prior research suggested that corporate strategies may influence firms' commitment to attend to stakeholders’ interests (Kang, 2013; Waddock & Graves, 2006). Corporate divestitures, a prominent form of corporate strategy, are examples of firm actions to remove assets that no longer strategically fit within the firm’s portfolio (Feldman & McGrath, 2016; Scharfstein & Stein, 2000). Separating business units from a firm’s portfolio represents a critical organizational change but also a “moral endeavor” (Parmar et al., 2010, p. 406) because this action may impact various stakeholders of the firm (Phillips, 2003). For example, asset divestitures could affect many employees, suppliers, and community stakeholders who might perceive firms’ divestiture actions as harmful to their interests or even unethical (Harrison & Wicks, 2021). Prior research has mainly focused on the shareholder-centric view of corporate divestitures and their financial consequences (e.g. Miller, 2006; Rau & Vermaelen, 1998). What remains unclear is whether and how divestiture activities may influence executive attention to stakeholders and CSR activities.

Removing strategically unfit units may enhance shareholder wealth (e.g. Lee & Madhavan, 2010) and help firms restore strategic control and long-term investment (Hoskisson et al., 1991), such as increasing human resource training, R&D spending (Hoskisson & Johnson, 1992), or innovation outcomes (Lee & Roh, 2020). Paradoxically, firms’ divestiture transactions may also involve ethical considerations when actions harm the interests of some stakeholders because workforce downsizing or termination of some suppliers often follows organizational break-ups (Harrison & Wicks, 2021). This leads to an important question of whether firms’ corporate divestiture actions lead to value creation or value destruction for stakeholders. If corporate divestitures can help firms enhance their long-term focus, as suggested by several scholars (Hoskisson & Johnson, 1992; Lee and Madhaven 2010), why do firms vary widely in their post-divestiture CSR performance?

To address this question, our study draws from the attention-based view of the firm (Ocasio, 1997, 2011), which argues that firms' strategic actions are the result of what issues, resources, and relationships their decision-makers focus on. The central tenet of the attention-based view lies in the bounded rationality assumption of decision-makers (Simon, 1957) who face numerous stakeholders competing for their attention (Ocasio, 2011). Managers likely experience attentional constraints when their firms’ product portfolio contains strategically unfit assets. Thus, removing such assets from firms’ portfolios would likely shift managerial attention toward a more long-term focus (Hoskisson & Johnson, 1992; Markides, 1992).

While corporate divestitures might help restore firms’ long-term focus, this situation may not automatically increase firms’ commitment to CSR, an important long-term strategy for achieving sustainability. The attention-based view argues that firms’ attentional engagement is often shaped by other managerial, firm, or macro-environmental factors (Ocasio, 1997; Ocasio & Joseph, 2005). Accordingly, we examine three moderators that are likely to play a role in bringing decision makers’ attention into actions, as reflected in post-divestiture CSR performance. We propose that firms’ commitment to CSR is a joint function of their divestiture scale and (1) pre-restructuring financial decline, (2) managerial motives, and (3) task complexity (the type of the business sold). These three factors moderate managers’ attentional engagement in CSR strategies and thus help to explain the heterogeneity in firms’ CSR performance following divestitures.

Our study contributes to the attention-based view by examining factors that help enact firms’ commitment to CSR, from attention to the action, which requires resources and managerial motives during the process. We argue that pre-restructuring financial decline may reduce managers’ attention to CSR activities during divestitures as they face a higher employment risk and more pressure from shareholders who have the urgency, power, and legitimate claims on the firm. In addition, we argue that having strong motives to serve stakeholders helps strengthen managers’ attentional engagement in CSR. We proposed that unlike long-tenured CEOs, new CEOs have greater incentives to build their trust and reputation with various stakeholders by satisfying their interests and demands (Chiu & Walls, 2019). This may be particularly crucial for divesting firms because the asset separation process usually creates high levels of anxiety, uncertainty, and disruptions to firm routines and stakeholders; thus, gaining stakeholders’ support is critically important for achieving the desired outcome. We further analyzed how task complexity (the type of business sold based on the divesting unit’s connection to the firm’s core competency) and divestiture scale may jointly shape managers’ attention to CSR. By examining the interaction between firms’ divestiture scale and type, our study offers a more nuanced insight into how specific task characteristics may influence executive attention to stakeholders, thus differentially affecting their commitment to CSR.

As companies nowadays increasingly depend on a broad range of stakeholders for resources and support, managerial attention to social, financial, and environmental issues becomes a crucial first step in gaining a competitive advantage (McWilliams & Siegel, 2011; Porter & Kramer, 2006). The present study is among the first to systematically examine corporate divestitures’ implications for CSR. Using a longitudinal sample of U.S. firms, we found that firms demonstrated increased commitment to CSR following asset divestment. However, this relationship was weakened among firms experiencing pre-restructuring financial decline or divesting more related businesses, but strengthened among firms with a new CEO. These findings suggest that for top managers to take more strategic actions in CSR following divestitures, their attention to long-term focus may be insufficient and needs to be enacted by firms’ financial condition, managerial motives, or task complexity. Our study, therefore, shed important light on when and how critical corporate change events such as divestitures may enhance or reduce managers’ attention to CSR, thereby leading to divergent CSR performance outcomes.

Theory and Hypotheses

Attention is defined as “noticing, encoding, interpreting, and focusing of time and effort by organizational decision-makers on both (a) issues: the available repertoire of categories for making sense of the environment and (b) answers: the available repertoire of action alternatives” (Ocasio, 1997, p. 189). The central point of the attention-based view argues that the issues and answers managers focus on will determine firms' strategic actions (Kaplan and Henderson 2005; Nadkarni & Barr, 2008). However, noticing the issues and solutions may not always lead to desired actions and outcomes because various contingency factors can affect decision-makers' discretions as well as how they set goals and attend to issues facing the firm (Ocasio, 1997; Ocasio & Joseph, 2005).

Strategic leaders’ attention is a scarce cognitive resource (Harrison & Bosse, 2013; Ocasio, 2011). Senior managers often face multiple environmental stimuli, including various stakeholder groups competing for their attention (Bundy et al., 2013). When overloaded by information, decision-makers cannot simultaneously recognize or respond to all the environment stimuli due to bounded rationality (Simon, 1991), which constrains their information processing capacity. To overcome this, managers often allocate their attention to some objects or ideas and away from others. This selective attention influences their decision-making and eventually drives organizational actions and outcomes (Shen & Cannella, 2002; Zhang, 2008).

Successful CSR implementation requires active managerial attention to coordinate firms' value-chain management among the marketing, R&D, procurement, and production processes (Wickert et al., 2016). As managers face competing demands from various stakeholders, it is paramount to this coordination process to integrate heterogeneous stakeholders’ interests (Feldman & McGrath, 2016) while determining what solutions to develop and what strategies to implement to manage CSR (Wong et al., 2011). Substantial research has examined various firm attributes, in particular, financial performance and board governance, as potential antecedents for CSR engagement (e.g. Makni et al., 2009; Walls et al., 2012).

One area that has received increased scholarly attention is how changes in firms’ corporate strategy influence their stakeholder strategy and CSR-related practices. For example, Kang (2013) found that a higher level of product diversification pushed firms to perform more CSR activities due to an increased number of stakeholder demands. In addition, using the mergers and acquisitions (M&A) as a context, Waddock and Graves (2006) found mixed results regarding whether M&A activities provided a strategic advantage to acquiring firms. The authors found that in the post-merger phase, acquiring firms experienced improved performance in some stakeholder-related practices, but worse performance in other stakeholder-related practices. Other research has examined the valuation effects of CSR on M&A activities, showing that investment in CSR creates value in M&A transactions (Bettinazzi & Zollo, 2017; Cho et al., 2021).

So far, existing research has mostly focused on how firms’ growth and expansion strategy would impact their CSR activities and has paid relatively little attention to how firms’ portfolio reduction strategy (such as asset divestiture or refocusing) might influence their stakeholder or CSR strategy. Corporate divestiture activities are not the ‘mirror image’ of acquisition activities as these two corporate strategies are usually driven by different goals and rationales (Brauer & Wiersema, 2012; King et al., 2004; Lee & Madhavan, 2010). This omission is surprising as firms depend on the support of various stakeholder groups for long-term growth, innovative developments, and survival (Hosmer, 1994). Our study, therefore, focuses on the divestiture context (while controlling for firm M&A activity) to better understand the impact of firms’ portfolio reduction strategy on their commitment to CSR. Importantly, we also examine three moderators (pre-restructuring financial decline, CEO motives, and task complexity related to the type of business sold) that are likely to strengthen or weaken managers’ attentional engagement in CSR following divestitures.

Corporate Divestitures and Commitment to CSR

The most popular contention behind firms’ asset divestment decisions is promoting operational efficiency and enhancing market value (Feldman & McGrath, 2016; Scharfstein & Stein, 2000). Some firms perform divestitures to correct previous strategic mistakes while others proactively change their business portfolio to enhance their future market position (Brauer, 2006; Markides, 1992; Shimizu, 2007). Firms’ business subunits can influence executives’ attention selection and engagement (Bouquet & Birkinshaw, 2008). Thus, keeping strategically unfit assets in the portfolio may cause firms to lose their core competencies (Hitt et al., 1996), drain valuable firm resources, and distract and divide managerial attention.

Many firms undertaking restructuring have diversified beyond their optimal level, which limits managers’ attentional capacity in understanding different stakeholder interests and addressing their demands (Wickert et al., 2016). These firms tend to emphasize financial controls (Williamson, 1975) due to executives’ attentional constraints in managing a wide range of businesses concurrently (Hoskisson et al., 1991). A focus on financial controls suggests that managers place a greater emphasis on short-term profitability and returns and are less likely to allocate resources for non-shareholding stakeholders (Stout, 2012). Refocusing the portfolio thus can help firms restore their strategic controls and long-term focus (Bergh, 1995). For example, this strategic approach may free up management time, organizational support functions, and financial capital that can be reinvested to create stakeholder value through building companywide CSR cultures, cultivating community relations, and engaging in more sustainable environmental practices (Jayachandran et al., 2013).

Further, selling off business units that no longer create value helps lower executives’ employment risk since they cannot be held accountable for the unit's performance once it is under different ownership. The reduced employment risk helps enhance executives’ long-term focus, allowing them to pay more attention to social and environmental stakeholders than just shareholders. This circumstance thus enables managers to develop social capital with stakeholders that they may not have time and resources to care for before divestitures, and build a better understanding of how to fulfill their demands (Russo & Perrini, 2010). Accordingly,

Hypothesis 1

Firms’ divestiture scale is positively associated with their commitment to CSR following divestitures.

Moderators for Attention and Commitment to CSR Following Divestitures

As we argue that firms’ divestiture scale will lead to increased commitment to CSR, some factors may either hinder or strengthen this relationship. We propose that firms’ commitment to post-divestiture CSR is a joint function of their divestiture scale and pre-restructuring financial performance, CEOs’ motives, and task complexity associated with the type of the business sold. We argue that having a long-term focus and strategic controls following divestitures may be necessary but insufficient to enact CSR commitment, which also requires managerial discretion to allocate firm resources and strong motives to build trust with various stakeholders to facilitate firms’ engagement in CSR. Also, task complexity may interact with firms’ divestiture scale to influence their ability to attend to stakeholders’ interests and demands following divestitures, which we explain in more detail below.

Pre-Restructuring Financial Decline as a Moderator

Corporate divestitures can be either reactive or proactive. Some firms engage in divestiture in response to the financial pressure (Hayward and Shimizhu 2006; Kim & Roh, 2014), while others proactively use divestitures as a strategic means to reconfigure their product portfolio, to enhance firms’ innovation capacity (Brunetta & Peruffo, 2014; Lee & Roh, 2020), and to sustain long-term growth and competitive advantage (Darnall & Edwards, 2006). For firms that divest proactively (i.e. under healthier financial conditions), managers have more discretion to allocate resources for social and environmental initiatives. This may not be the case for firms that divest reactively due to consecutive years of financial decline.

When facing prominent corporate issues, managers are likely to narrow their attention to certain stakeholders directly impacted by the situation (Mitchell et al., 1997). The presence of financial trouble will prompt managers to focus their attention on shareholders, who are the most salient stakeholders given the legitimacy of their claims, the urgency of their issues and demands, and the relative power they have compared to other stakeholders in the situation (Agle et al., 1999; Suchman, 1995). Firms face tremendous pressure from financial stakeholders when they encounter “multiple years of declining financial performance subsequent to a period of prosperity” (Pearce & Robbins, 1993, p. 623). Unlike a temporary dip in financial performance, the presence of consecutive years of decline (Barker et al., 2001) poses a significant threat to a firm’s survival, restricting managers’ ability to attend to various stakeholders’ interests and demands (Cyert & March, 1963). When firms’ shareholders take precedence over other stakeholders, managers will likely focus on short-term rather than long-term goals (Gray, 1999).

When firms suffer from consecutive years of financial decline, managers face a strong dismissal threat (e.g. Cannella et al. 2009; Crossland and Chen 2013; Hubbard et al., 2017) and thus, would be reluctant to go against the interests of their shareholders. Also, firms that divest out of financial pressure often incur substantially more costs in the restructuring process (Dranikoff et al., 2002), reducing their ability to allocate resources to other areas that may not generate immediate financial returns. To alleviate shareholders’ concerns, these firms might become more conservative in spending by cutting back or abandoning community services they previously provided, downsizing local employment, and/or reducing their monitoring and funding of social and environmental initiatives. Under this circumstance, divesting firm managers’ priority is to restore financial performance, rectify previous strategic mistakes, and improve cash flows, which may take time to achieve following the divestiture. Accordingly, compared to firms that divest their assets proactively, those that divest reactively due to severe financial decline likely face more challenges in balancing the instrumental (financial) and moral expectations of CSR (Hillman & Keim, 2001) due to managers’ limited attentional capacity (Simon, 1991). Hence,

Hypothesis 2

The level of pre-restructuring financial decline mitigates the positive association between firms’ divestiture scale and their commitment to CSR following divestitures.

New CEOs (Motive) as a Moderator

The attention-based view argues that managers’ motivations will guide their attention and decisions that eventually influence firms’ actions and outcomes (Ocasio, 1997, 2011). Long-tenured CEOs have developed a legacy of relationships with stakeholders in the past and thus, may be disinclined to change the current CSR practices. In contrast, new CEOs are less constrained by firms’ existing practices and routines and more open to new goals and changes (Chiu et al., 2016; Miller & Shamsie, 2001). Being new in the position, CEOs face greater scrutiny from firms' various constituencies (Harjoto & Jo, 2011; Harrison & Fiet, 1999). Research shows that failing to protect the interests of firms’ social and environmental stakeholders may increase the threat of early dismissal for new CEOs (Chiu & Sharfman, 2016). Thus, new CEOs need more stakeholders’ support and partnership than long-tenured CEOs to gain trust and reputation and to ensure their job security in the organization (Shen & Cannella, 2002). Consistent with this logic, recent research showed that new CEOs are more vigilant in attending to stakeholders’ interests and demands (Bernard et al., 2016; Chiu & Walls, 2019).

In the context of corporate divestitures, new CEOs’ attention to various stakeholders may become even more crucial as firms’ divestiture scale increases. This is because CEOs play a substantial role in engaging with internal and external stakeholders, from divestment planning, execution, to post-divestiture integration. Thus, CEOs need to address various stakeholders’ concerns and special requirements and explain how firms’ divestiture actions might influence the existing employees, customers, and suppliers (PricewaterhouseCoopers, 2015). The greater a firm’s divestiture scale, the more stakeholders will likely be involved in the process, which means more support and trust from stakeholders will be needed for new CEOs to facilitate divestiture activities to achieve the desired outcome. This situation should motivate new CEOs to proactively manage their relationships with firm stakeholders and address their interests and demands, thus resulting in stronger post-divestiture CSR performance.

Hypothesis 3

New CEOs strengthen the positive association between firms’ divestiture scale and their commitment to CSR following divestitures.

Task Complexity as a Moderator

This section discusses how task complexity associated with the characteristic (type) of divested units might interact with firms’ divestiture scale to influence CSR following divestitures. At the time of restructuring, some firms might choose to sell off more unrelated than related businesses, while others decide to sell off more related than unrelated businesses (Hoskisson & Johnson, 1992). Selling off unrelated or related businesses is not symmetrical nor exclusive from one another as firms may sell both types of business units concurrently.

Unlike unrelated business units, related business units typically share common strategic linkages and resources with the firm’s core operations and assets (Bergh et al., 2008). High levels of interdependency, constant mutual adjustments, and strategic stakes between related business units make it more complex and challenging to manage effectively (Thompson, 1967). This complexity stems from frequent interactions and coordination between task teams, the intensity of resource exchanges and information sharing, and reliance on each other’s task inputs/outputs as well as social support (Thompson, 1967; Wood, 1986). Inter-unit coordination becomes more intense and complex as activities performed in one business unit are contingent upon the success or outcomes of activities executed in another business unit (Thompson, 1967). An interruption in the task of one business unit could cause subsequent failure in other business functions that share common resources or part of the same strategic initiatives, which usually occurs among related business units.

Further, due to higher asset specificity (Williamson, 1985), separating assets with close connections with the firm’s primary operations requires more managerial time and energy, from planning, execution, to post-divestiture integration. Selling related businesses also requires more in-depth firm knowledge due to its high strategic stakes from these assets’ tight connections with firms’ core product lines (Chiu et al., 2016). Thus, when the situation warrants divestiture, selling related business units posts greater challenge due to the complexity of task required in order to minimize the interruption to highly coordinated units as well as the embedded relationships among them. By contrast, selling unrelated businesses is more straightforward and creates fewer interruptions to firms’ primary operations. Therefore, from the portfolio management perspective, removing related businesses is more complicated and demands more managerial attention than removing unrelated businesses, thereby limiting managers’ ability to attend to the interests and demands of social and environmental stakeholders.

Practically, restoring firms’ strategic controls and long-term focus is more likely to achieve through ‘refocusing’ firms’ product portfolio (i.e. selling more unrelated than related businesses; Baysinger & Hoskisson, 1989; Chiu et al., 2016). By refocusing the firm’s product portfolio, managers can focus their attention on creating synergies and economies of scope from the joint function of the related product lines, including activities required for CSR implementation (McWilliams & Siegel, 2001). Supporting this view, Lubatkin and Chaterjee (1994) argued that firms with portfolios comprising related businesses have greater incentives to “promote long-term investment horizons” (Lubatkin & Chatterjee, 1994, p. 116). In comparison, selling more related businesses will lead to ‘de-focus’ firms’ product portfolio, which may not help restore a long-term focus for CSR following divestitures. Taken together, we propose that selling more unrelated units will strengthen the positive link between firms’ divestiture scale and their subsequent commitment to CSR, while selling more related units will likely weaken this relationship.

Hypothesis 4

Selling more unrelated business units strengthens the positive association between firms’ divestiture scale and their commitment to CSR following divestitures.

Hypothesis 5

Selling more related business units mitigates the positive association between firms’ divestiture scale and their commitment to CSR following divestitures.

Method

Data and Sample

The study sample comprises a longitudinal dataset of U.S. public firms from 1998 to 2013. We focused only on the U.S. public firms due to the availability and reliability of the data on corporate divestitures and CSR activities. We collected divestiture transactions from the M&A section of Securities Data Company (SDC) Platinum. Following Bergh and Sharp (2015) and Pathak et al. (2014), we used a random sample of 300 firms that had performed voluntary divestitures during 1998 ~ 2013. We used the random sample generator from Excel to perform this task. With these firms, we created a longitudinal panel over the study period, which allowed us to achieve a power above 0.8 to test our hypotheses based on our sample size (Maxwell et al., 2008; Rast & Hofer, 2014). We conducted several t-tests, and the results indicated that our sample firms are not significantly different from the population divesting firms in terms of total assets, total liabilities, cash flow, and accounting performance (ROA and ROE). Our panel includes years with and without divestiture activity to avoid selection bias since the timings of firms’ divestitures and CSR activities are likely non-random. For example, if a firm had some sell-offs in 2005 and 2006, data for that firm in other years (1999–2004, 2007–2013) were also included in the panel dataset.

We focused on sell-off events for two reasons. First, more than 90 percent of the divestiture transactions were classified as sell-offs based on the SDC database. Second, sell-offs represent a cleaner break-up through a 100 percent ownership transfer of the firm’s subsidiary to an external entity, which reduces the bias that may arise due to firms’ restructuring choices (Bergh, 1995). Thus, we removed firm observations that performed spin-offs, leveraged buyout, or equity carve-out, to avoid contamination in the analysis. We also removed firms that received tender offers and deals that were withdrawn after firms’ divestiture announcements.

We collected CSR performance data from MSCI/KLD database (formerly Kinder, Lydenberg, and Domini, or KLD), which has been widely used for stakeholder and CSR research due to its objectivity and comprehensiveness (Chiu & Sharfman, 2016; McWilliams & Siegel, 2000). There might be a sample selection concern about the KLD Social rating database, which has a small bias toward larger organizations with more public information (nonzero scores) in any given CSR subarea (Hart & Sharfman, 2015). According to the KLD manual, firm size is the criterion of inclusion in the database. We thus followed Kang (2013) and controlled for firm size to avoid sample-induced endogeneity. Our firm-related variables (size, financial performance, product diversification, etc.) came from the Compustat North America database; CEO-related variables (e.g. new CEOs and CEO equity holdings) came from Execucomp; governance variables (e.g. board independence) came from BoardEx. After merging the data from various sources and removing missing values due to acquisitions, delisting, or bankruptcies, our final sample includes 1,792 firm years (of which 763 are observations with divestiture events) from 192 unique firms and 40 industries (based on firms’ primary two-digit Standard Industrial Classification (SIC) code).

Dependent Variable

Post-divestiture CSR performance. To effectively capture CSR as a multidimensional construct (e.g. Griffin & Mahon, 1997), we focused on the six major dimensions (community involvement, diversity, employee relations, environment, human rights, and products) that represent a broad range of CSR issues from the MSCI/KLD database. Scholars have shown robust construct validity among MSCI/KLD’s constituent variables and social indicators (e.g. Mattingly & Berman, 2006). Under each KLD dimension, firms are rated based on their CSR Strengths (socially responsible actions) and CSR Concerns (socially irresponsible actions). However, CSR Strengths and Concerns ratings have low convergent validity because their respective indicators are evaluated using different rating schemes (Strike et al., 2006). Given their low convergent validity, using the combined scores (subtracting Concerns ratings from Strengths ratings) would lead to erroneous interpretation (Mattingly & Berman, 2006; Strike et al., 2006). In this study, we focused on the positive aspect of CSR as our study mainly concerns whether firms increase their commitment to CSR following divestitures.

We further performed the principal component analysis, which generated one factor (eigenvalue = 2.4) with factor loadings above 0.50 cutoff (Hair et al., 1998) on all six KLD dimensions except the ‘products’ category (= 0.47). The Cronbach’s alpha for the CSR index based on our sample is 0.66, which is fairly close to that using the population KLD dataset (alpha = 0.67) over the same period. Since there was no theoretically derived ranking guide to assessing the importance of each category (Ioannou & Serafeim, 2015), we assigned equal weight to all six KLD dimensions, as done in previous studies (e.g. Hillman & Keim, 2001; Ioannou & Serafeim, 2015). We then standardized each dimension's scores before summing them up to create our CSR index variable. As firms may experience internal chaos and turmoil during the first year following divestiture, we estimated post-divestiture CSR based on the firm’s average CSR scores at t + 1 and t + 2 (t was the year of divestiture).

Predictor Variables

Divestiture scale. A firm’s divestiture scale was measured based on the percentage of the transaction value of the divested business divided by the parent firm’s total assets. When a firm had multiple divestiture transactions in a year, we consolidated the transaction values to compute the divestiture scale. For missing transaction values, we searched for available sales or asset data for the divested unit for the previous year (Bergh, 1995; Pathak et al., 2014). If unit sales or asset data was not available for a divested unit, we treated this transaction as missing and summed the values of other transactions to estimate the divestiture scale for that year. Note that our models take into account the number of divestitures each year to reduce the potential bias due to occasional missing transaction values in the SDC database.

Pre-restructuring financial decline. We chose a firm’s return on equity (ROE) as a performance indicator that researchers have frequently used to measure organizational decline and turnaround (e.g. Chen & Hambrick, 2012). Specifically, we took the average value of (ROE t-2—ROE t-1) and (ROE t-3—ROEt-2) to calculate pre-restructuring financial decline. To facilitate interpretation of the results, we used the reverse year calculation so that a positive score indicates a declining financial performance, whereas a negative score indicates an improved financial performance before the focal year.

New CEO. The presence of a new CEO in a given year during the sampling period was coded as 1 and 0 otherwise (Zhang, 2008).

The number of unrelated and related businesses sold. Following prior research (Palepu, 1985; Shimizu, 2007), our study classified a transaction as a sale of the related business if the divested unit shared the same first two-digit SIC code with that of the parent firm’s primary business; otherwise, it was classified as a sale of an unrelated business. Approximately 60% of the sample's divestiture deals were sales of unrelated businesses. We summed the number of unrelated units sold and related units sold, separately, in a given year for each firm.

Control Variables

We controlled for firm size, measured as the natural log of the total number of employees, because larger firms tend to have more resources for CSR investment (Zhao & Murrell, 2016). We also controlled for a firm’s free cash flow (in billions) since financial slack availability can impact CSR (Harrison & Coombs, 2012; Tang et al., 2015). We controlled for a firm’s product diversification, using the entropy measure (Palepu, 1985), as it reflected the scope of stakeholders’ demands (Kang, 2013). We also controlled for a firm’s M&A activities, calculated as the M&A transaction amount divided by total firm assets before the focal divestiture, as these activities would likely influence firms’ internal resource allocations for CSR (Waddock & Graves, 2006). We controlled for prior CSR performance to account for historic patterns within the environment that may be linked to other confounding factors such as the firm’s financial condition and its strategic change decision.

We controlled for governance-related factors including board independence (percentage of independent outside directors on the board) and CEO equity ownership (percentage of shareholdings owned by CEO), as they may influence managers’ attention to the nonshareholder stakeholders (Harrison & Coombs, 2012) and long-term focus for the organization (Johnson & Greening, 1999). We also controlled for CEO tenure (the number of years of serving as a CEO in a firm). Moreover, we controlled for environmental contexts that may influence managerial discretions for implementing CSR strategy (Goll & Rasheed, 2004), including the degrees of industry uncertainty (dynamism), resource availability (munificence), and competition (complexity), using measures from Boyd (1990) and Dess and Beard (1984). Finally, we included the year fixed-effects to reduce unobserved heterogeneity that could affect firms’ CSR over time. All the control variables were lagged by one-year from the dependent variable to rule out the likelihood of reverse causality.

Results

To test our hypotheses, we used firm fixed-effects models for our panel data. Theoretically, using a fixed-effect method is more suitable as our study is mainly concerned about firms' CSR changes following divestiture. The Hausman test result indicates that fixed-effects models are more efficient than the random-effects models (χ2 = 465.16, p < 0.001). We used robust standard errors clustered by firms to reduce heteroscedasticity and autocorrelation issues in the analysis (Hoechle, 2007). Table 1 reports descriptive statistics and correlations for the key variables in this study. Table 2 displays the regressions for post-divestiture CSR, with Model 1 as a control model. Consistent with previous research, the amount of free cash, board independence, and CEO equity, are positively related to post-divestiture CSR (e.g. Tang et al., 2015). We noted that the coefficient for the pre-restructuring financial decline was negative but not significant in predicting CSR performance. This is not surprising as prior research has shown mixed results for the CSR and financial performance relationship, and scholars have argued that various contingency factors may moderate the relationship (e.g. Chiu and Sharfman 2011, Orlitzky et al., 2003).

Table 1 Summary Statistics and Correlation Matrix
Full size table
Table 2 Regressions Results Predicting Post-divestiture CSR Performance (H1-H3)
Full size table

Hypothesis 1 predicts that a larger divestiture scale is positively associated with firms’ commitment to CSR. The coefficient for the divestiture scale was positive and significant (Model 2 of Table 2: B = 0.679, SE = 0.255, p = 0.008). Specifically, a one percent increase in firms’ divestiture scale resulted in an average increase of 0.679 units in their CSR following the divestiture. Thus, Hypothesis 1 was supported. For the four moderating hypotheses, we examined both the coefficients for the interaction terms and the marginal plots of the relationships. Hypothesis 2 states that pre-restructuring financial decline weakens the positive relationship between firms’ divestiture scale and subsequent CSR. The interactive effect of firms’ divestiture scale and pre-restructuring financial decline was negative and significant (Model 3 of Table 2: B = − 0.971, SE = 0.412, p = 0.019). In practical terms, among firms with an average divestiture scale, one unit increase in pre-restructuring financial decline resulted in a 0.97 unit decrease in post-divestiture CSR performance. Figure 1 displays the plot for the interactive effect with simple slope test results, showing that firms exhibited a lower commitment to CSR when their pre-restructuring financial decline increased from low levels to high levels. Thus, Hypothesis 2 was supported.

Fig. 1
figure 1

Divestiture Scale and Financial Decline on CSR Performance

Full size image

According to Hypothesis 3, new CEOs accentuate the positive effect of firms’ divestiture scale on their subsequent commitment to CSR. The interactive effect of divestiture scale and CEO succession was positive (Model 3 of Table 2: B = 1.784, SE = 0.455, p < 0.001), indicating that the relationship between firms’ divestiture scale and CSR became stronger in firms led by a new CEO. Firms with a new CEO had 1.78 units higher post-divestiture CSR performance than those without a new CEO. Figure 2 displays the interactive effect, offering further evidence for the relationship. Simple slope tests showed that firms’ divestiture scale resulted in stronger post-divestiture CSR in the presence of a new CEO (dy/dx = 2.029, SE = 0.437, p < 0.001). Therefore, Hypothesis 3 was supported.

Fig. 2
figure 2

Divestiture Scale and New CEO on CSR Performance

Full size image

Hypotheses 4 and 5 stated that selling more unrelated (related) businesses strengthens (mitigates) the positive relationship between firms’ divestiture scale and their subsequent CSR, respectively. We focused on firm-years with divestiture activities because the type of business sold was relevant only for firm observations with divestiture activity. In a supplemental analysis, we tested the relationships using the full sample and found similar results. The interactive effect of the divestiture scale and the number of unrelated businesses sold was positive but not significant (Model 5 of Table 3, B = 0.003, SE = 0.196, p = 0.990). Thus, Hypothesis 4 was not supported. In comparison, the interactive effect of divestiture scale and the number of related business sold was negative and significant (Model 5 of Table 3, B = − 0.381, SE = 0.134, p = 0.005). Specifically, each additional related business sold resulted in a decrease of 0.381 units in post-divestiture CSR. Figure 3 displays how the effect of firms’ divestiture scale on their CSR is shaped by four different values (0, 1, 2, and 3) of the number of related businesses sold. The simple slope test results showed that the positive relationship between firms’ divestiture scale and CSR was weakened with each additional related unit sold. Taken together, these results supported Hypothesis 5.

Table 3 Regressions Results Predicting Post-divestiture CSR Performance (H4-H5)
Full size table
Fig. 3
figure 3

Divestiture Scale and Related Business Sold on CSR Performance

Full size image

Supplemental and Robustness Analyses

Assessing potential omitted variable bias associated with divestiture scale. To examine whether a potential omitted variable bias related to the divestiture scale might influence our results, we followed recent research (e.g. Gamache & McNamara, 2019; Harrison et al., 2018) and employed the Impact Threshold of a Confounding Variable (ITCV; Frank, 2000) test. The ITCV test calculates the impact of a confounding variable needed to alter a statistical result, which helps to evaluate the robustness of a statistical finding when a potentially confounding variable is included (Frank, 2000). The results showed an ITCV of − 0.024, which means that for an omitted variable to overturn our results, it would need to be correlated |r|> 0.155 with both divestiture scale and post-divestiture CSR (α = 0.05). None of the variables in our model exceeded the impact threshold with both divestiture scale and post-divestiture CSR performance, suggesting that our findings are unlikely to be driven by an omitted variable.

For additional robustness tests, we conducted a two-stage residual inclusion model (2SRI; Hausman, 1978) to assess potential endogeneity related to a firm’s divestiture scale. In the first stage equation, we used two instrumental variables, the median industry divestiture scale and the sum of state divestiture scale (the U.S. state where the firm headquarter resides), and all the control variables to predict a firm’s divestiture scale. The first stage results showed that both instruments were significant predictors of divestiture scale (Model 1 of Appendix Table 4: p < 0.01). The results of the weak identification test (F- statistic = 25.77; Stock et al. 2002) and the Sargan-Hansen test (p = 0.773) suggested that the two instruments met both relevance and exclusivity conditions. We then performed the Durbin-Wu-Hausman test, and the result was not significant (F- statistic = 0.40, p = 0.526). The ITCV and 2SRI results indicated that our models did not suffer from omitted variable bias related to a firm’s divestiture scale.

Assessing potential endogeneity associated with new CEO. We further examined the potential endogeneity of new CEOs because the occurrence of CEO change was unlikely to be random and could be influenced by confounding factors within the firm and the environment. We again employed a 2SRI model with two instrumental variables, average board tenure and the average number of years of board directors to retirement (excluding the CEO), which are strong correlates of CEO succession but not highly correlated with post-divestiture CSR. The first stage regression showed that both instruments were significant predictors of divestiture scale (Model 2 of Appendix Table 4: p < 0.05). The week identification test (F-statistic = 15.34) was greater than 11.59 and the Sargan-Hansen statistic was not significant (p = 0.715), indicating that these instruments met the relevance and exclusivity conditions. Further, the Durbin-Wu-Hausman test result was not significant (F- statistic = 0.73, p = 0.393), suggesting that our models were robust against potential endogeneity associated with CEO change.

Assessing divestiture scale as a mediator for pre-restructuring financial decline and CSR. Another concern might exist about the situation where a firm’s pre-restructuring financial decline indirectly influences post-divestiture CSR performance through its divestiture scale. To examine this possibility, we tested the mediation model (conditional indirect effects) of firms’ pre-restructuring financial decline on their post-divestiture CSR performance via the divestiture scale. We used the Monte Carlo Method (Selig & Preacher, 2008) with 50,000 bootstrapping samples. The conditional indirect effect was not significant (B = − 0.004, SE = 0.003, 95% CI = [− 0.010, 0.001]), suggesting that a firm’s divestiture scale did not mediate the relationship between pre-restructuring financial decline and post-divestiture CSR performance.

Testing hypotheses using generalized method-of-moments (GMM) models. While we have examined omitted variable bias and endogeneity related to divestiture scale and CEO change, as described above, there might still be unobserved fixed effects correlated with the lagged values of our dependent variable. To account for this possibility that might influence the consistency of our findings, we tested our hypotheses using the GMM models with the Arellano-Bond dynamic panel estimator (Arellano & Bond, 1991). We treated all the predictors as endogenous and used their lagged values for previous years as instruments in the models. The results using the GMM models generated similar regression patterns as those based on the firm fixed effects in the primary analyses (see Appendix Table 5).

Discussion

Corporate divestitures are often used to restore firms’ strategic controls and long-term focus by removing strategically unfit assets from their product portfolio. Drawing from the attention-based view (Ocasio, 1997, 2001), the present study examines whether and to what extent corporate divestitures may lead to increased firm commitment to CSR. We argue that a higher level of corporate divestiture scale will result in more CSR activities; however, this relationship is likely shaped by pre-restructuring financial performance, CEO motives, and task complexity associated with the type of business sold. A better understanding of managerial attention, as well as the factors that may change their attentional engagement in CSR, helps to enrich our knowledge of why firms vary widely in their post-divestiture CSR performance.

Theoretical Implications

Our study contributes to stakeholder and CSR research by connecting firms’ divestiture actions to their commitment to CSR. We showed that disposing strategically unfit businesses from firms’ product portfolios positively influences firms’ post-divestiture CSR. This finding is an important contribution as prior research has mainly examined corporate divestitures from the perspective of neoclassical economics to predict financial outcomes or impacts on investors. Our result echoes Lee and Roh (2020), who argued that corporate sustainability may be achieved through “vigorous structuring and maintaining the firm’s operation efficiently by nurturing core values simultaneously” (p. 1). With more knowledge about the impact of divestiture actions on CSR and its boundary conditions, we offer fresh insights into the strategic determinants of CSR through asset divestitures. In doing so, our study provides meaningful implications for firms’ long-term prosperity and sustainability which has increasingly relied on the support of a wide range of stakeholders, not just shareholders.

We found support that firms facing pre-restructuring financial decline would exhibit a lower commitment to CSR as the divestiture scale increases. Managers in these firms face greater attentional constraints in serving the different interests and demands of stakeholders competing for their attention. This finding adds to the debate in the CSR literature regarding the relationship between corporate financial performance (CFP) and CSR performance (e.g. Orlitzky et al., 2003; Margolis and Wash 2003; Waddock and Graves 1997). In their replication of Waddock and Graves (1997), Zhao and Murrell (2016) found that the relationship between CFP and CSR performance (and vice versa) was not straightforward, and the effect sizes were smaller when using a larger, longitudinal dataset with a broader range of firms and industries involved. While several researchers have cast doubt on the virtuous circle argument (Waddock and Graves 1997), which suggests that improved CSR (CFP) leads to better CFP (CSR), our findings highlight the importance of recognizing decision makers’ limited attentional capacity to fulfill the interests and demands of various stakeholder groups simultaneously. Notably, our findings suggested that firms’ commitment to CSR is likely driven by the joint function of specific corporate strategies and prior financial performance. In other words, financial performance alone may not be sufficient to guide managerial attention to allocate more/fewer resources for CSR-related initiatives.

On the other hand, we showed that firms led by new CEOs exhibited stronger post-divestiture CSR performance (while controlling for pre-restructuring financial performance), which is consistent with the attention-based view that decision-makers’ attention is often influenced by their motivations. Given their lack of in-depth knowledge about and relationships with the existing stakeholders, new CEOs are more motivated to commit to CSR activities as they need to gain support, trust, and partnership with various stakeholders during and after the divestitures. This situation becomes more prominent as the firm’s divestiture scale increases. Thus, although both pre-restructuring financial decline and new CEOs could trigger organizational change events, as found in previous studies (e.g. Boeker, 1997; Chiu et al., 2016), they have opposing implications for CSR performance following divestitures based on our findings.

Our results regarding task complexity associated with the type of business sold suggested that considering both the scale and the type of the firm’s divestiture activity is critical in understanding the effect of corporate divestitures on firms’ commitment to CSR. Prior research has rarely examined both simultaneously or the interaction between them in predicting organizational outcomes. Our findings suggested that divesting more related businesses may constrain managers’ attentional capacity due to higher task complexities in separating assets with close operational and strategic connections, thereby adding more challenge to firms to attend to the needs of social and environmental stakeholders.

Interestingly, our result showed that selling more unrelated business units did not positively moderate the divestiture scale and CSR link. We conducted additional tests and found that the marginal effect of the divestiture scale on CSR was positive and significant when the firm sold a modest number (1 ~ 4) of unrelated business units, but turned non-significant when the firm sold five or more unrelated business units a year. It is likely that firms that sold a large number of unrelated business units (> = 5) required substantially more efforts, costs, and time to integrate and recover from internal chaos and turbulence following divestitures. Thus, there might be a potential ceiling effect for selling more unrelated business units in shaping the corporate divestiture and CSR relationship. Taken together, these findings suggested that the mechanisms for selling unrelated businesses are not the mirror opposite of those for selling related businesses. Our study thus provides a more nuanced theoretical contribution for connecting corporate strategic change to CSR, showing that it is important to account for both the scale (the proportion of the assets sold) and the type (related or unrelated units sold) to better understand the multifaceted impact of corporate divestiture activities on CSR.

Practical Implications

Our findings offer practical insights for top managers and boards of directors who strive to achieve both strategic change and better CSR in the future. Like other corporate strategies, restructuring through asset divestitures requires managers to assess how their actions might benefit or harm firm stakeholders during the strategic planning process (Hosmer, 1994). While it is challenging to fulfill the interests and demands of all the stakeholders who compete for managers’ attention, firms may benefit from knowing the conditions under which corporate divestiture actions might create values for their stakeholders. For example, our result suggested that CEOs’ attention and commitment to post-divestiture CSR are guided, at least in part, by their motivations to build trust and relations with firms’ stakeholders, as in the case of new CEOs. Directors of the board, given their fiduciary duty to hire/fire a CEO and protect stakeholders’ interests (Rao & Tilt, 2016), should pay close attention to new CEOs' role in influencing the scale and scope of firms’ divestiture and CSR strategies and outcomes.

Second, as our findings suggested that divesting more related business units may hinder firms’ commitment to CSR due to higher task complexity associated with separating related assets from the product portfolio, managers that sell more related assets may require closer board supervision and guidance to ensure their divestiture strategy not deviate significantly from the firm’s long-term focus in CSR. If retaining firms’ commitment to CSR is critical for divesting firms, refocusing the product portfolio by selling more unrelated assets is more likely to help achieve this goal. Moreover, as stakeholders of divesting firms with financial decline are likely to suffer more from poor CSR performance, managers and boards should engage with stakeholders in more thoughtful divestiture planning and execution. In doing so, these key corporate leaders will be able to make more optimal divestiture decisions and strike a better balance in achieving instrumental and moral expectations of CSR performance (Hillman & Keim, 2001).

Limitations and Future Research

Our study has some limitations that should be noted. First, we focused on sell-offs, the most widely used divestiture strategy, in predicting firms’ post-divestiture CSR. Our findings may not be applicable for firms that perform other forms of divestiture strategies, such as spin-offs, equity carve-outs, or leveraged buyouts. Second, our empirical analyses focused on the positive aspect of CSR outcomes following divestitures. It is likely that firms’ asset divestiture may generate actions that are deemed socially irresponsible (CSiR; Harrison & Wicks, 2021), such as mass employee lay-offs or service termination for the divested product line. Prior research showed that firms might engage in CSR and CSiR activities concurrently (e.g. Strike et al., 2006). Since high levels of CSiR could undermine the firm’s competitive advantage (Orlitzky & Benjamin, 2001), understanding corporate divestitures’ implications for CSiR will allow managers to better prepare for effective corporate changes while avoiding harm to their stakeholders. Such research will help shift the dialog from whether a specific corporate strategy will benefit or harm stakeholders to under what conditions a corporate strategy may create the ‘variability’ in the magnitude of responses between different stakeholder groups and individuals.

More future research should consider the role of business ethics (Freeman & Gilbert, 1988) in stakeholder management during corporate strategic change. For example, it could be fruitful to examine how the impact of corporate divestiture on social and financial performance is moderated or mediated by the stakeholder engagement process. As Hosmer (1994) stated, “the strategic planning must be both analytical and ethical” (p. 32), suggesting that it is important to incorporate ethical and moral considerations to protect stakeholders’ interests and rights in the process. Managers that show more ethical and moral considerations for their employees, communities, customers, and suppliers during strategic change planning and implementation will likely gain more trust and support from their stakeholders, thereby enhancing their potential to achieve stronger firm performance and competitive advantage in the long run.

In addition, future research might explore alternative explanations for low post-divestiture CSR performance behind firms that restructure their assets reactively due to financial trouble. Prior research showed that underperforming firms are more likely to engage in illegal acts such as bribery (Xu et al., 2019), which might partially explain the lack of CSR commitment in these firms. Such research will also shed light on the intricacy of the relationship between CFP and CSR performance with a contingency perspective. Further, as activist stakeholders become more involved in firms’ strategic actions including CSR-related practices, it is crucial to pay attention to activist investors (Chen & Feldman, 2018), which can be critical carriers of attention (King, 2008). As stakeholder activism continues to spread in the U.S. and other countries, firms should closely watch the growing trend of embracing social, environmental, and governance issues from activists and institutional investors (JP Morgan, 2019) while undertaking critical corporate change initiatives such as asset divestitures.

Conclusion

According to a recent survey by Financial Times Group (Global Corporate Divestment Study, 2018), 53 percent of the firms had a stakeholder communications plan for their divestiture activities, up from 19 percent in 2016. This indicates that firms are increasingly aware of the importance of treating stakeholders, not just shareholders, as crucial partners during corporate divestitures. Our study is among the first to examine the implications of firms’ divestiture actions in relation to CSR. We contribute to the attention-based view of CSR strategy based on firms undertaking corporate divestitures. Our findings suggest that firms’ post-divestiture CSR performance is a function of their divestiture scale and financial performance, (new) CEOs’ motives, and task complexity associated with the type of business sold. Overall, this study supports the notion that firms’ divestiture actions can have a meaningful effect on their CSR engagement.