A chief supply chain officer matters – but only when you are in trouble!

The COVID-19 pandemic, along with shortages of shipping containers, semiconductors, and labor in the logistics industry led to a global supply chain crisis (Gamio and Goodman, 2021), which now further continues with the war in Ukraine, climate change disruptions, trade wars and other major geopolitical disruptions (Bednarski et al., 2025). Even before COVID-19, the strategic importance of supply chain management (SCM) had become evident, given the challenge of managing global and outsourced supply chains (Hendricks et al., 2015). This growing view of SCM as a strategic advantage has been accompanied by an evolution of the function in organizational hierarchies (Villena et al., 2018; Wagner and Kemmerling, 2014). A typical example has been the appointment of a chief supply chain officer (“CSCO”) (Hendricks et al., 2015; Kroes et al., 2021), which recently has been titled by Fortune magazine as rising from obscurity to superhero and as “the toughest job in the C-suite” (Wahba, 2021).

Supply chain managers must deal with supply and demand risks as well as more all-encompassing uncertainties such as those brought on by the pandemic. Existing research into supply chain risk management under exceptional disruptions such as the COVID-19 pandemic is scarce (Munir et al., 2022), and we have a limited understanding of how organizational characteristics—such as CSCO role and hence SCM standing in a firm—impact it (Fan et al., 2017). Yet according to organizational information processing theory (OIPT; Galbraith, 1973), uncertainties like COVID-19 create a heightened need for organizational information processing (Birkel et al., 2023). This can be accomplished through either vertical information systems or lateral relations to ensure high performance under uncertainty (Premkumar et al., 2005). While previous OIPT studies have examined information systems within the context of SCM (see, e.g. Srinivasan and Swink, 2015, 2018) as well as COVID-19 (see, e.g. Birkel et al., 2023), lateral relations are scarcely examined in either setting. Yet from an OIPT perspective, the pandemic for example is likely to create the type of information processing needs for supply chain management that “standard” vertical information systems are ill-equipped to handle. Hence, we argue that a CSCO through their organizational role connecting the function can be a performance differentiator.

Thus, using OIPT, we examine how firms with a CSCO perform in industries with different supply and demand risk characteristics, particularly during COVID-19 induced shocks, compared to firms without a CSCO. Specifically, we seek to answer the following research question: What is the performance impact of a CSCO during heightened uncertainty (COVID-19) and does it differ for firms with high supply or demand risk exposure?

For this purpose, we employ an empirical triple-differences approach to assess the relative performance of North American companies across various operating environments. Specifically, we aim to provide more robust estimates of the CSCO effect by jointly analyzing performance differentials (1) between similar firms with and without a CSCO, (2) across industries with a low and high average exposure to supply and demand risks, and (3) before and after the outbreak of the COVID-19 pandemic. Hence, we compare whether firms with a CSCO perform better than comparable control firms, before and after a global exogenous shock, while accounting for the fact that a CSCO’s presence may not be random. To measure supply and demand risk exposure across industries, as well as the impacts of COVID-19 induced shocks specifically, we employ a text-mining technique newly applied in operations and supply chain management that quantifies how frequently supply and demand risks are discussed during earnings conference calls in each industry (Hassan et al., 2023). Our analysis follows two complementary approaches: (1) to establish baseline supply and demand risk exposure across industries, we analyze pre-pandemic earnings calls from 2017–2019; and (2) to assess COVID-19 induced shocks specifically, we examine 2020 earnings calls for pandemic-related supply and demand discussions.

Our empirical model uses annual financial data from Compustat, executive appointment records from BoardEx, quarterly earnings call transcripts from Refinitiv, and stock market returns from the Center for Research in Security Prices (CRSP). The sample spans 2017–2020 and includes publicly traded U.S. and Canadian companies in manufacturing, wholesale, and retail trade industries.

Our study makes several contributions. First, the negative performance impacts of CSCOs found previously have been attributed to lower-performing firms being more likely to employ CSCOs (Wagner and Kemmerling, 2014) and to the potential impacts of the task environment (Roh et al., 2016). We address both issues by analyzing the CSCO-performance relationship using COVID-19 as an exogenous shock to firm performance and operating environment, in line with calls by van Hoek (2020). Second, understanding whether CSCOs can improve performance in supply chains with high supply or demand risk exposure is strategically important, as firms now operate under persistently turbulent global conditions. Geopolitical instabilities and climate change will continue to disrupt supply chains; understanding the strategic (performance) importance of SCM in industries most prone to such uncertainties can encourage firms to devote more resources to SCM and leverage the role of the CSCO to weather the storms. Third, from an OIPT perspective, both within SCM and regarding COVID-19, studies have focused on understanding the role of vertical information systems in facilitating information processing during uncertainty. Our research examining the role of a CSCO from this perspective thus contributes to both SCM as well as broader upper echelons research around OIPT.

The appointment of a top supply chain executive highlights the role of SCM in a firm, facilitating the allocation of resources, power, support, and strategic focus to it (Villena et al., 2018). A powerful SCM executive has a better span of control and authority, as well as the ability to gain compliance and buy-in from other functions (Kroes et al., 2021). These tasks are essential to successful SCM, given its boundary-spanning nature (Villena et al., 2018).

Previous literature examining the role and impact of CSCOs starts from Wagner and Kemmerling (2014), who showed a gradual increase in their presence, while noting that firms with CSCOs exhibited lower operating profit margins than those without. Others have followed, identifying (1) financial leverage, internationalization, and diversification as antecedents to CSCO appointments as well as moderators for CSCO impact on firm performance (Roh et al., 2016), (2) noting a positive stock market reaction to supply chain (and operations) executive announcements, particularly for newly created positions and outsider appointments (Hendricks et al., 2015), (3) CSCO presence to be associated with shorter cash cycles, lower inventory levels, improved capacity utilization, and more inventory flexibility during high market instability (Kroes et al., 2021), and (4) a significant negative association between the presence of a CSCO and product recalls, particularly for larger firms (Körber and Cotta, 2020). Of the previous research, Kroes et al. (2021) provides the most important grounding for our study. While Kroes et al. examine certain supply chain performance metrics (cash conversion cycle and operational slack) under market instability, we provide a more extensive analysis of the performance impacts of a CSCO, including sales growth and profitability at the higher level, and asset turnover, return on sales, cost of sales and vertical integration as well as operational slack and cash conversion cycle at a more SCM specific performance level. More importantly, while Kroes et al.’s measure of market instability only covers volatility of sales within the industry group over the prior 20 quarters, our examination and empirical operationalization of risk and uncertainty is more extensive, covering both demand and supply side risks of the industry in general as well as demand and supply side uncertainty during COVID-19.

Studies on CSCOs often examine the role as part of the top management team (TMT). Yet Krause et al. (2022) note TMT is often poorly defined, suggesting a consensus definition whereby TMT is noted as “the executives responsible to the CEO who meet regularly to develop organizational strategy and oversee its implementation”. Given complex organizational structures, the exact reporting links between top-level executives often cannot be determined from secondary data. Hence, all criteria from said definition cannot be fulfilled (Krause et al., 2022). We thus do not take a position on whether the CSCOs we examine are on the TMT or not (while it is highly likely many are), and rather examine CSCOs as part of the upper echelon of the firm, i.e. in the top-tiers (Krause et al., 2022). Hence, through their hierarchical influence, reporting, and linkages to others within the upper echelon they likely wield significant impact both within the SCM function and across functions. As Roh et al. (2016), we assume that CSCOs are responsible for enterprise-wide SCM activities. Other than CSCOs, functional TMT members’ performance impacts have been examined and proven for several roles, such as chief marketing officer (Germann et al., 2015), chief risk officer (Li et al., 2022), chief information officer (Taylor and Vithayathil, 2018), and chief sustainability officer (Kanashiro and Rivera, 2019). Research on functional TMT members considering external factors such as industry and market characteristics has been scarce (Menz, 2012), and regarding the moderating effects of uncertainty between TMT composition and performance, previous research has mostly been examining low-tech vs high-tech industries (Aboramadan, 2021). Particularly considering CSCO or any functional TMT member performance impacts during COVID-19, research is very limited, with studies mainly examining chief medical officer roles in the healthcare sector during the pandemic. Calls have been made for theory-grounded research on how different types of TMT compositions interact with dynamic industry environments (Yamak et al., 2014).

The pandemic brought unprecedented impacts and vulnerability to global supply chains and the overall economy (Ivanov, 2021; Seuring et al., 2022), with simultaneous impacts on supply, demand, and the logistics infrastructure (Seuring et al., 2022). These impacts are also unprecedented when compared to prior viral outbreaks, such as SARS and H1N1 (Hassan et al., 2023). Indeed also the supply chain risk management literature had not addressed pandemics prior to COVID-19 (see, e.g. Ho et al., 2015; Louis and Pagell, 2019).

Overall, definitions of risks and uncertainty in the SCM literature vary. There particularly exists inconsistencies in the defitions of (1) risk as source or outcome, and (2) risk versus uncertainty. We briefly discuss both issues in the following, outlining the definitions we use in our study regarding COVID-19 and otherwise.

Risk as source or outcome. Risk is typically noted as, e.g. variation from expected outcomes or a combination of probability and impact (Louis and Pagell, 2019). Sarker (2019) defines supply risk as the input risk that affects the inward flow of resources, noting that a “supply risk simply becomes the disruption”. Jüttner et al. (2003) refer to the same two as risk sources and risk consequences. The term supply chain risk is also often used interchangeably with supply chain uncertainty—when a distinction is made, it typically relates to risk having only negative outcomes while uncertainty leading to either negative or positive outcomes (Simangunsong et al., 2012). Wagner and Bode (2008) suggest that the negative connotation given to risk is a better fit with supply chain reality. As our study is focused on understanding CSCO impact on firm performance based on exposure to supply and demand risks, we focus on risks as negative outcomes/consequences. We examine supply and demand risks primarily at the industry level, where different sectors face varying degrees of exposure to upstream (supply) and downstream (demand) challenges based on their structural characteristics and operating environment. This exposure is measured as a continuous variable based on the proportion of earnings call conversations in 2017–2019 devoted to supply- or demand-related incidents within each industry (Hassan et al., 2023).

Drawing from Knight’s (1921) seminal work, uncertainty is seen as a broader term than risk. Risk is seen as measurable, stemming from repeatable events with an associated probability used in decision-making, while uncertainty is incalculable, linked to unforeseeable and unprepared events that are not manageable a-priori (Müllner, 2016). Referring to the above literature, we see COVID-19 as an uncertainty, which may have caused either negative or positive financial shocks depending on, e.g. industry and firm position in the overall value chain (e.g. Klöckner et al., 2023b). Unlike our baseline industry risk exposure measurement (using pre-pandemic 2017–2019 data), we measure firm exposure to COVID-19-induced supply or demand shocks through separate analysis of 2020 earnings calls. For this pandemic-period analysis, we quantify the proportion of discussions devoted to supply or demand events explicitly mentioning COVID-19 (Hassan et al., 2023) in negative or positive context. This distinct approach captures the direct, directional impacts of the pandemic rather than general industry risk characteristics.

We do not hypothesize about the direct impact of COVID-19 on firm performance, which several studies empirically examine. Instead, we investigate how firms were able to mitigate these impacts, focusing on CSCO presence as the mitigating factor. We account for the possibility that COVID-19 had either positive or negative supply and/or demand effects in the industries of the firms being examined.

Organizational information processing theory (OIPT) argues that organizations are built around information and information flows, with tasks involving greater uncertainty requiring more extensive information processing (Galbraith, 1973). Important concepts within the theory are information processing needs, information processing capabilities, and uncertainty (Fan et al., 2017; Premkumar et al., 2005). Uncertainty stems from the dynamism of the external environment and in the supply chain context, demand and supply uncertainties are major contributors to a firm’s information processing needs (Premkumar et al., 2005), driving the need to develop information processing capabilities (Laari et al., 2023; Yu et al., 2019).

Fit is a central concept of OIPT (Premkumar et al., 2005; Yin et al., 2024); the theory stresses the origin of the uncertainty and its causes and thus how the fit between information needs and processing capacity can be managed (Laari et al., 2023). In this way, OIPT represents a specific outgrowth of contingency theory (Fairbank et al., 2006). The two alternative options for dealing with uncertainty specifically are to reduce information processing needs or increase information processing capacity (Srinivasan and Swink, 2015; Yu et al., 2019). The former can be achieved through, e.g. slack resources, such as additional inventory in the context of SCM (Srinivasan and Swink, 2015, 2018). Of particular relevance to our study is the latter, which refers to investments in vertical information systems or lateral relations (Fairbank et al., 2006). Previous SCM studies have examined the former from an OIPT perspective, focusing on how digitalization, technology-enabled SCM systems, and supply chain analytics improve information processing capacity and consequently performance (see Munir et al., 2022; Srinivasan and Swink, 2015, 2018; Yin et al., 2024). Lateral relations then include organizational relationships, contacts, roles, teams, and integration across functions of a firm (Srinivasan and Swink, 2015). While information systems have been found to better enable the utilization of information for supply chain performance, the concept of lateral relations from an OIPT perspective remains less studied.

While information systems can be used to process structured information, there can be situations of multiple and conflicting interpretations of information when face-to-face communications, meetings, discussions, and collaboration work better (Ye et al., 2022), that is, lateral relations. Supply chain risks relate to abnormal information about inventories, logistics, finances, market, politics, and other factors, and this information can be random, complex, ambiguous, and uncertain (Fan et al., 2017; Munir et al., 2022). To effectively tackle such uncertainties, an appropriate configuration to collect, process, and distribute information must be in place (Fan et al., 2017), but information systems built around “routine” information flows may not suffice. When it comes to COVID-19 and OIPT, again information processing capacities (in SCM context but also more broadly) are mainly examined from a vertical information systems perspective (see, e.g. Birkel et al., 2023). We thus see a clear gap in examining how internal information processing capacities related to SCM and uncertainties can manifest beyond information systems and technical solutions.

In response to the above-noted gap and in line with OIPT, we argue that a CSCO within the top management team can increase an organization’s information processing capabilities, thereby enhancing performance. Specifically, we argue that in highly uncertain contexts, it is specifically these more flexible CSCO’s relationships, contacts and boundary spanning-enabled information processing capacities that provide the best fit for driving performance. Consequently, CSCO performance impacts are particularly prevalent in industries with high risk exposure and during COVID-19.

Corporate appreciation of SCM will drive supply chain managers to consider the long-term implications of their decisions (Villena et al., 2018), which can often mean not only considering short-term costs and efficiencies but also resilience. A supply chain executive positioned high in the organizational hierarchy has an extensive span of control and the ability to coordinate SCM across business units, as organizational structure defines the resource configurations, power dynamics, communication channels, and decision-making rights of individuals and groups within an organization (Roh et al., 2022). This CSCO position from an OIPT perspective means structures and routines, timely participation in decision-making, information distribution, and joint action and commitment that enable information processing between functions (Fairbank et al., 2006; Laari et al., 2023; Srinivasan and Swink, 2015). This provides higher supply chain visibility, agility, and responsiveness (Swink et al., 2012).

The above-described impacts are likely to be especially important for firms with riskier supply chains—the negative effects of supply chain vulnerability can be mitigated by placing priority on SCM. Directions from the top provide guidance for supply chain decision-making at lower levels and help address the risk-benefit trade-off (Villena et al., 2018). A CSCO can influence a firm’s supply chain uncertainty, variability, and lead time through strategic internal and external supply chain integration (Körber and Cotta, 2020; Roh et al., 2016). This improved ability for information processing capacities through enabled by the CSCO is a key factor why firms exposed to supply and demand risks will particularly benefit from the CSCO presence, that is, from an OIPT perspective these conditions provide the best fit for the increased information processing capacities a CSCO provides. Consequently, firms with increased information processing capacities enabled by CSCO presence may reduce their downside exposure to unexpected disruptions of production schedules. Additionally, a CSCO can facilitate the management of demand fluctuations through closer collaboration with the marketing function.

There are several mechanisms through which a CSCO can influence firm performance. Contracts and collaborative information sharing are important integrative mechanisms for managing supply and demand risks, and they require strategic decisions on technology investments, sales, and operations planning – all improved when a top-level executive has responsibility for them (Hendricks et al., 2015). Supply chain collaboration with senior executives better enables obtaining signals from the environment—an example of information processing—that provides operational advantages for risk management (Yu et al., 2019). The quality of sales and operations planning is also dependent on the extent to which inputs to the process and decision-making are internally co-managed (Laari et al., 2023). Kroes et al. (2021) particularly show that CSCO presence is linked to shorter cash conversion cycles, likely due to the integration with suppliers, customers, and internal units that the executive has enabled. Shorter cycles, in turn, enable firms to adjust to supply and demand volatility more efficiently and quickly (Klöckner et al., 2023a). Roh et al. (2016) indicate that lacking a CSCO is likely to impair supply chain performance for firms with high product diversity and global sales, both conditions tied to elevated demand risk exposure. Kroes et al. (2021) support this with findings of enhanced inventory flexibility for firms with a CSCO during high market instability.

The mechanisms discussed above indicate that a CSCO can drive increased firm responsiveness to risks. Yet they may be overly expensive in firms with low vulnerability, reducing profitability (Sun et al., 2009). An approach that maximizes efficiency without specific preparation works well for firms operating under relatively stable supply and demand conditions (Ketchen and Craighead, 2021). It can require significantly less cross-functional coordination and integration than the strategies required under high demand or supply risk exposure. The coordination structures brought on by CSCOs under certain settings may only add unneeded complexity (Roh et al., 2016). Hence, we believe the benefits a CSCO brings matter more in operating environments with high-risk exposure, be it to supply or demand side risks. Specifically, we hypothesize:

A firm’s operating environment is a typical contingency factor highlighted in operations and supply chain management studies (Simangunsong et al., 2012), with decision-makers being recommended to align their organization with the risks in their supply chains (Wagner and Bode, 2008). The COVID-19 pandemic drastically changed the contingencies under which supply chains operated and highlighted the need for firms to approach SCM in a coordinated way (Alexander et al., 2022). Its scale and duration created huge uncertainties and placed great demands on SCM practices and capabilities (Kähkönen et al., 2023; Xiong et al., 2021). Under such “black swan” or catastrophic events as COVID-19, the information-processing needs of companies rise significantly (Birkel et al., 2023) and organizational information processing capabilities become particularly highlighted (Phillips et al., 2023; Wong et al., 2020). In such times, the role of the CSCO becomes more prominent (Douglass and Viniak, 2019), helping to stabilize the firm (Swink et al., 2012) and facilitate the processing of information across the firm through TMT relations. As a result of the pandemic, boards have heavily interacted with their CSCOs on supply chain issues (Gartner, 2020; Oglesbee, 2022). Having a CSCO in the TMT during the pandemic was crucial, as exemplified by the following earnings call quote:

Klöckner et al. (2023a) have particularly demonstrated that shorter trade cycles and higher vertical integration – which both fall under impact areas of a CSCO as discussed above – attenuate the negative impact of COVID-19. Kroes et al.’s (2021) findings on CSCO presence to be associated with shorter cash conversion cycles, improved capacity utilization, and larger inventory buffers during sales instability all support the argument that CSCO presence will have better prepared firms for the unprecedented uncertainty brought on by COVID-19. Further support is found from post-COVID-19 surveys, which show that CSCO-level respondents rank resilience as their first priority while mid-level management is still focused on more traditional measures, including costs (GEP, 2022).

We also believe that firms operating in industries with high pre-COVID supply and demand risk exposure, that is, those with the “closest possible practice”, will have benefited even more from having a CSCO during the pandemic. Experiential learning from disruptions is noted as an antecedent to supply chain resilience (Scholten et al., 2019). CSCOs that have operated in a risky supply-chain environment will have prioritized resources and means to respond to disruptions and developed uncertainty-relevant information processing capabilities more so than those who have faced a more stable environment. Thus, we further hypothesize:

Our initial dataset includes 1,112 public firms from the United States and Canada operating within manufacturing (Standard Industrial Classification codes 2000–3999), wholesale trade (SIC 5000–5199), or retail trade (SIC 5200–5999) industries. We focus on firms with at least 500 employees and available sales data from the Compustat/CRSP Merged Database. Our dataset spans four years, including a three-year pre-pandemic period (2017–2019) and the pandemic year (2020). We augment this financial data with executive appointment records from BoardEx, merging the two sources via the Wharton Research Data Services linking table. After this merge, our interim dataset comprises 786 firms and 2,757 firm-year observations, including 206 observations from 85 firms with an active CSCO.

We then employ coarsened exact matching (Iacus et al., 2012) within this interim sample to match the 206 firm-year observations with CSCOs to similar-sized industry peers without a CSCO (Arora et al., 2020; Roh et al., 2016). Matching criteria include exact two-digit SIC industry groups and coarsened lagged total sales. Industry definitions consistently follow two-digit SIC codes throughout our analysis. Utilizing the Stata “cem” procedure with five size bins, we obtain a final sample comprising 477 firms and 1,663 firm-year observations, including 206 CSCO observations matched to 1,457 observations from non-CSCO peer firms. Major industries in the final sample include chemicals (SIC 28), industrial machinery (SIC 35), and instruments (SIC 38). Industries with the fewest observations include printing and publishing (SIC 27), building and gardening (SIC 52), and furniture stores (SIC 57), though our results are robust to merging small industries with their nearest neighbors. Table 1 (columns 1–4) provides a full industry distribution.

Coarsened exact matching mitigates model misspecification and omitted variable bias in observational studies by matching similar treated and control firms based on pretreatment covariates, thus minimizing disparities and potential confounding effects (Iacus et al., 2012). We use matching to compare similar firms with and without a CSCO before and after the COVID-19 pandemic, an external shock impacting supply and demand conditions in the firms’ industry. We hypothesize that accounting for an industry’s inherent susceptibility to supply chain disruptions is essential, as a CSCO may be less effective or even redundant in industries with low exposure to supply or demand risks (Wagner and Kemmerling, 2014). To address this, we introduce an additional layer of industry differentiation, creating a triple-difference approach. This method compares firms with and without a CSCO across industries with low- and high-risk exposure before and after COVID-19. By using a three-way interaction model, we leverage risk exposure variations to capture the heterogeneous effects of CSCO presence, allowing for a more nuanced understanding of its impact across industries and market conditions.

Consistent with prior research, we assess firm financial performance through sales growth and profitability. Sales growth, our primary metric, reflects the year-over-year change in sales (sg1) and may be influenced by CSCO-driven operational improvements. We also examine profitability via operating return on assets (roa), calculated as operating income after depreciation divided by average total assets, to evaluate the impact of CSCO presence on financial performance. Detailed variable definitions and data sources are provided in the  Appendix Table A1.

CSCO presence. To identify firms with a Chief Supply Chain Officer (CSCO), we follow a systematic approach outlined in earlier research (Kroes et al., 2021; Körber and Cotta, 2020; Roh et al., 2016; Wagner and Kemmerling, 2014). Using the BoardEx database, we first conducted a case-insensitive regular expression search for the terms “chief,” “supply,” “chain,” and “officer” (Kroes et al., 2021), as well as a broader search with the term “CSCO” or the combination of “supply chain” with “corporate,” “group,” or “executive vp” (Roh et al., 2016). After rigorous manual verification of role descriptions, we code CSCO presence as a binary variable (csco_i,t = 1) when firm i maintains an active CSCO in year t with at least nine months tenure. We tracked these positions annually throughout our entire sample period (2017–2020), applying consistent coding rules. We excluded regional and divisional CSCOs to focus on corporate-level supply chain leadership, as well as executives with CSCO responsibilities but less than nine months tenure, who were coded as absent (csco_i,t = 0) to ensure adequate executive experience (Körber and Cotta, 2020). For instance, if a CSCO entered service in June 2018 and departed in October 2020, we would code absence in 2017, absence in 2018 (insufficient tenure), and presence in 2019 and 2020. For consistency, we assumed CSCO presence during years with missing disclosure when the same executive held the position in both preceding and following years. The average CSCO tenure in our sample is 3.11 years, remaining stable across the sample period with no significant year-to-year variations. We acknowledge that our identification method relies on job titles, which may miss executives functioning as CSCOs but not labeled as such. However, Securities and Exchange Commission regulations requiring companies to fully disclose all positions held by executives and their business experience help mitigate this limitation (Wagner and Kemmerling, 2014). Additional analyses using alternative CSCO definitions are detailed in the Supplementary Materials of this study.

Supply and demand risk exposure. Our study employs a text-based approach to quantify industry-level supply and demand risk exposure. Through computational linguistics, we analyze the extent of supply and demand risk incidents discussed across different industries in quarterly earnings conference calls. This approach builds on recent research validating textual analysis for measuring exposure to political risk (Hassan et al., 2019), country risk (Hassan et al., 2024b), and external shocks including Brexit (Hassan et al., 2024a), the Fukushima nuclear disaster (Hassan et al., 2024b), and epidemics (Hassan et al., 2023). Their validation tests demonstrate strong economic significance—for example, a one-standard-deviation shock to demand risk exposure correlates with revenue changes of −2.0% to +1.3% (Hassan et al., 2023).

Our analysis leverages two distinct but complementary earnings call datasets: (1) pre-COVID data (2017–2019) to establish baseline industry risk exposure, and (2) COVID-period data (2020) to measure pandemic-induced shocks. For baseline risk exposures, we analyzed 48,503 quarterly earnings call transcripts from Q1 2017 to Q4 2019 (pre-COVID-19) from 4,318 North American companies across various industries using the Refinitiv Eikon database. Following Hassan et al. (2023, Table 3) for each firm we identified keyword patterns related to supply and demand risks and applied a textual search algorithm to measure both the frequency and intensity of these patterns in each firm’s earnings calls, yielding firm-level risk exposures. We then aggregated these firm-level measures by averaging firm-level exposures across all firms within each two-digit SIC industry group over the 2017–2019 period. Aggregation at the industry level enhances statistical power, mitigates endogeneity issues, and captures systematic risks that better align with our theoretical framework. We rescale industry-level risk exposures to vary from zero (no exposure) to one (maximum exposure), thereby simplifying empirical interpretation. Table 1 (columns 5 and 6) displays these industry-level risk exposures, reporting average supply and demand exposures of 0.59 (±0.22) and 0.59 (±0.19), respectively. These estimates leverage our full dataset, ensuring precise measurements even in smaller industries. For example, the supply (0.17) and demand (0.96) exposures for SIC-27 were computed from 359 earnings call transcripts from 26 firms, not only from the four firms listed in Table 1. The Supplementary Materials provides further details of the text-based approach.

COVID-19 pandemic. To evaluate COVID-19’s impact on financial performance, we use a dummy variable, covid, which is set to 1 for observations from 2020 and 0 otherwise. This provides a simple and effective method to account for the pandemic’s external shock, isolating its effects on the response variable within a difference-in-difference framework. This approach also minimizes confounding effects that may arise from using text-based COVID-19 metrics, which could correlate with supply and demand risk exposure variables. In Section 5.2, we explore the negative and positive effects of COVID-19 across various industries in more detail.

To account for residual size differences after coarsened matching, we include the natural logarithm of company sales (“firm size”) as an initial control. We also control for a firm’s capital structure using financial leverage (“leverage”), defined as the book value of long-term debt divided by total assets (Kroes et al., 2021; Roh et al., 2016). Both firm size and leverage are lagged by one year to reduce simultaneity bias. Additionally, we control for lagged performance variables, including sales growth as “sales trend” and return on assets as “profitability,” since current performance is likely influenced by past results (Hendricks et al., 2009; Kroes et al., 2021). This approach mitigates potential biases arising from performance-related omitted variables and unobserved heterogeneity. Table 2 provides summary statistics for all dependent, independent, and control variables, along with a correlation matrix. None of the correlation coefficients exceed 0.7, indicating collinearity is not a significant concern. To minimize the influence of outliers, we winsorize all variables at the 1% level in both tails, though this treatment does not materially affect our coefficient estimates or their statistical significance.

Using a matched sample from the years 2017–2020, we empirically test our hypotheses by estimating the following triple interaction model:

where y is a measure of firm performance; i and t are firm and year subscripts; k is firm i’s industry; covid is a binary variable for the COVID-19 pandemic; csco is an indicator variable for CSCO presence; exposure_k is the firm’s industry exposure to supply or demand risks measured from earnings calls 2017–2019; controls include one-year lagged control variables, including lagged firm performance.

We estimate our triple-differences model using OLS regression with firm-clustered standard errors. To facilitate the interpretation of the triple interaction effects, Table 3 presents the estimated marginal effects of COVID-19 and CSCO presence on firm performance for different levels of industry exposure (the full results of the triple interaction model are available in the Supplementary Materials). Marginal effect analysis offers an intuitive way to assess the total effect of CSCO presence by combining the direct effect with relevant interaction terms, while holding industry risk exposure constant. Low, Mid, and High refer to estimated marginal effects at the 1st, 50th, and 99th percentiles of industry risk exposure, respectively.

Columns 2–7 in Table 3 show that the effect of CSCO (csco) is generally small, inconsistent, and not statistically significant during normal times, except in industries with low-risk exposure where the effect on sales growth is negative (0.017 or 1.7%, p < 0.10), consistent with Wagner and Kemmerling (2014). Regarding Hypothesis H1a, the CSCO effect on sales growth (−0.016) and return on assets (−0.002) is not statistically significant in high supply risk industries. Similarly, in high demand risk industries, the CSCO effect on sales growth (−0.027) and return on assets (0.001) remains insignificant. Thus, hypotheses H1a and H1b, which postulate a positive CSCO effect for highly exposed industries under normal conditions, are supported neither statistically nor economically by our results.

Table 3, Column (1) shows that COVID-19 had a notable negative effect on sales growth and return on assets across all industries, with declines of −0.096 (p < 0.01) and −0.012 (p < 0.01), respectively. Economically, this nearly 10% decline in sales and more than one percentage point decrease in profitability represent substantial losses, reflecting meaningful negative shocks to firm performance and competitive positioning. The first rows in Panels A and B also show that COVID-19 has a monotonically increasing negative effect on firm performance as one moves from low to high industry exposure. This pattern holds for both supply and demand risk exposure measures and for both firm performance measures, supporting the construct validity of our industry risk exposure measures. Regarding Hypothesis H2a, the interaction effect between COVID-19 and CSCO presence (0.8% sales growth; 0.9% return on assets) is economically small and statistically insignificant, suggesting a limited practical benefit from CSCO presence across all industries during the pandemic. Thus, these results do not support Hypothesis H2a.

Column 4 shows that during COVID-19, the CSCO’s influence on firm performance becomes significantly positive and economically meaningful in industries with high supply risk exposure. Specifically, the marginal effect analysis indicates that CSCO presence in high supply risk industries during COVID-19 correlates with a 9.8% (p < 0.05) increase in sales growth and a 3.0% (p < 0.10) rise in return on assets compared to non-CSCO firms. These findings offer support for Hypothesis H2b. Similarly, Column 7 shows that in high demand risk industries, CSCO presence is associated with a 13.4% (p < 0.01) increase in sales growth and a 4.0% (p < 0.10) increase in return on assets during COVID-19, providing statistical and economic support for Hypothesis H2c.

In summary, our results indicate that CSCO presence improves firm performance in high-risk industries during the COVID-19 pandemic, supporting Hypotheses H2b and H2c. However, our results do not support Hypotheses H1a and H1b, which predicted a positive impact of CSCO presence in highly exposed industries under normal conditions, nor do they support Hypothesis H2a, which anticipated a universally positive CSCO effect during COVID-19 irrespective of industry-specific risk exposure. This nuance may explain prior mixed findings and emphasizes the importance of contextual factors in evaluating the effectiveness of CSCOs.

Our main analysis shows that CSCO firms have higher sales growth and profitability during COVID-19 in high-risk industries. To better understand the sources of these advantages, we next analyze various channels through which the positive effect of the CSCO in highly exposed industries can be understood. Focusing first on the profitability channels, we break down the return on assets ratio into its profit and asset components (Swink and Jacobs, 2012). Specifically, we analyze asset turnover (ato; defined as the ratio of sales to total assets) on the asset side as a measure of productivity. On the income side, we examine the return on sales (ros; operating income/sales) and the cost of sales (cos; cost of goods sold/sales) as two determinants of financial efficiency. We also investigate operational efficiency channels of the CSCO effect, including operational slack via the cash conversion cycle (ccc; Hendricks et al., 2009), and capacity slack (slack; Kroes et al., 2021), measuring how effectively a firm uses its physical assets to generate sales. Finally, the level of vertical integration (intg; Klöckner et al., 2023a) assesses production adaptability by examining the firm’s dependence on external resource providers. Detailed variable definitions are in  Appendix Table A1.

To evaluate the relative importance of each channel, we re-estimate the triple interaction model in Equation (1), using each channel as the outcome variable and including its lagged value as a control. Table 4 presents the estimated marginal effects of the main variables, focusing on high supply risk (Panel A) and high demand risk (Panel B) industries. The coefficient values for covid imply that the COVID-19 pandemic has led to decreased asset turnover, longer cash conversion cycles, decreased capacity utilization (positive slack coefficient in high supply risk exposure industries), and decreased vertical integration (in high demand risk exposure industries). Overall, these results demonstrate the substantial impact of COVID-19 on various aspects of firm performance in industries with high supply and demand risk exposure. Apart from reducing the cost of sales, the isolated impact of a CSCO (csco) on profitability and efficiency channels is not statistically significant during normal market conditions.

Table 4 also provides further insights into the mechanisms through which the presence of a CSCO influences financial and operational efficiency during COVID-19. Focusing on the profitability channels, we find that the interaction term (covid × csco) has a negative and significant effect on the cost of sales in both high supply risk exposure (−0.118, p < 0.01) and high demand risk exposure (−0.137, p < 0.01), indicating that CSCO presence leads to economically meaningful cost savings by approximately 11.8% and 13.7%, respectively, during the pandemic. In terms of operational efficiency channels, the interaction term (covid × csco) shows a negative, statistically and economically significant effect on the cash conversion cycle (ccc) and capacity slack (slack) in both high supply and demand risk exposure industries. For example, the parameter value for the cash cycle (−0.068, p < 0.01) implies roughly 25 days (−0.068 × 365) improvement in working capital efficiency. Additionally, the positive interaction term for asset turnover (0.086, p < 0.10) in high supply-risk industries indicates an 8.6% improvement in productivity (sales per asset), while the positive effect on return on sales (0.011, p < 0.10) in high demand-risk industries translates into approximately a 1.1% point increase in profit margins during the pandemic.

In summary, our analysis shows that a CSCO positively impacts the drivers of firm performance primarily during COVID-19 in high supply and demand risk industries. The main mechanisms include reduced cost of sales, faster cash conversion cycles, and improved asset utilization, collectively clarifying how CSCO presence enhances firm resilience under challenging market conditions.

In this section, we examine how CSCO presence interacts with COVID-19′ diverse impacts across industries. Our Section 5 analysis assumed COVID-19 had uniform negative effects, but likely the pandemic created both winners and losers. To capture this variation, we first categorize industry-specific pandemic effects using Hassan et al. (2023)’s framework of three distinct shocks: negative supply shock, negative demand shock, and positive demand shock (we exclude positive supply shock due to its rarity). We identify these shocks by analyzing 13,473 quarterly earnings call transcripts from 2020 (covering Q1-Q4), representing 3,642 North American firms. Using computational linguistics, we scanned these transcripts for COVID-related terminology alongside supply and demand risk language. We assigned sentiment scores to each relevant text segment, then averaged and normalized these scores (0–1 scale) to quantify each industry’s exposure to each shock type. See Supplementary Materials for complete methodological details.

We apply these industry shock measures in a two-way interaction model for the 2020 cross-section of the sample by using the measured intensity of the specific shock type (“shock”) as a moderator for the CSCO effect. Table 5 presents the results of this estimation with the variable of interest being csco × shock. Panel A reports the regression coefficients, and since interactions are not directly interpretable, Panel B shows the marginal effects of the CSCO presence in high-shock industries.

The analysis reveals that CSCO presence significantly influences firm performance during the COVID-19 pandemic, with economic effects varying by the type of COVID-19 shock. In industries facing severe negative supply shocks (columns 1–2), CSCO presence substantially enhances sales growth by approximately 13.0% (p < 0.05) and return on assets by about 3.7% (p < 0.05), economically implying strong resilience against supply disruptions. For industries experiencing negative demand shocks (columns 3–4), CSCO presence economically improves return on assets by 3.2% (p < 0.10), though its impact on sales growth (8.0%) remains economically relevant but statistically insignificant. Conversely, in industries encountering positive demand shocks, CSCO presence is associated with economically significant sales growth of about 14.9% (p < 0.01) and an improvement in return on assets of around 4.8% (p < 0.10), highlighting substantial economic gains in capturing market opportunities. Overall, the presence of CSCOs impacts firm performance differently depending on the type of COVID-19 shock, generally yielding positive effects, most prominently by mitigating negative supply shocks and amplifying positive demand shocks.

To address the complexity of COVID-19 disruptions affecting multiple supply chain dimensions simultaneously, we extended our analysis to combined shock scenarios: concurrent negative supply and negative demand shocks (columns 7–8) and the particularly challenging scenario of negative supply coupled with positive demand shocks (columns 9–10). In scenarios involving simultaneous negative supply and demand shocks, CSCO presence economically enhances sales growth by 9.3% (p < 0.10) and return on assets by 3.1% (p < 0.10), though the economic magnitude is somewhat smaller compared to the isolated negative supply shock scenario. Notably, in the challenging scenario combining negative supply shocks with positive demand opportunities, CSCO presence yields exceptionally strong economic benefits, significantly improving sales growth by about 20.3% (p < 0.001) and return on assets by approximately 5.7% (p < 0.001). These economically substantial effects underscore the strategic value of CSCOs in simultaneously mitigating supply disruptions and capitalizing on emergent market opportunities during periods of unprecedented supply chain disruptions caused by the COVID-19 pandemic.

We conduct several additional tests to validate the performance effect of a CSCO, extend our core analysis, and strengthen methodological robustness. These tests include firm fixed effects estimation to control for time-invariant characteristics; alternative risk exposure specifications such as U.S.-only estimates, year-specific measures, and firm-level risk exposures (Wu, 2024); matching on prior performance trends; additional control variables (TMT characteristics, Chief Operating Officer/Chief Risk Officer presence, industry fixed effects); an expanded CSCO definition; and placebo tests to verify statistical validity. Our core finding remains robust across these analyses: CSCO presence has a statistically significant positive impact on firm performance in high-risk industries during COVID-19. One notable exception occurs when using firm-specific risk measures (purged of industry-level variation), where CSCO effects disappear. This exception provides additional support for our methodological approach, suggesting that industry-level measures may more appropriately capture the relevant CSCO performance aspects. Collectively, these analyses strengthen our confidence in the CSCO’s role in improving financial performance under uncertainty, particularly when firms face systematic industry-level risks. These tests are further discussed in the Supplementary Materials.

Our results indicate that the presence of a CSCO enhances financial performance under extreme conditions. Expanding on this, we performed a post-hoc analysis to examine the influence of CSCOs on stock market performance from a shareholder perspective, focusing on the period before and during the COVID-19 pandemic. The pandemic’s disruption of global supply chains intensified market volatility and exposed firms to heightened risks, thus providing a valuable context for evaluating the role of CSCOs in managing these market risks.

For this purpose, we analyze years 2017–2020 using three stock return metrics from CRSP: total 12-month return, maximum monthly return, and minimum monthly return, to assess overall shareholder wealth and price volatility. Using coarsened exact matching, we pair each CSCO firm with a non-CSCO industry peer of similar size and market characteristics, yielding 2,046 matched observations. We then apply a triple interaction model, controlling for firm size, valuation, momentum, and market beta, to capture the pandemic’s stock impact on CSCO and non-CSCO firms in high supply and demand risk industries.

Our post-hoc analysis, detailed in the Supplementary Materials, shows that firms in high-risk industries experienced substantial but volatile returns during the pandemic. Firms in industries with high supply and demand risk saw significant total returns (29.77% for supply-risk and 37.98% for demand-risk sectors) but faced considerable volatility, with monthly returns ranging between 21–25% for highs and -15-17% for lows. This created almost a 40-percentage point range in monthly returns. While CSCO presence does not significantly impact total stock returns, economically, it significantly stabilizes monthly stock price volatility, reducing price fluctuations by approximately 30% points in highly exposed industries. Thus, our results suggest that CSCO presence offers investors meaningful economic benefits by decreasing risk exposure during extreme market conditions.

Surprisingly, our hypotheses regarding CSCO performance impacts in high supply or demand exposure industries (H1a&b) or during COVID-19 in general (H2a) were not supported. Our results do, however, demonstrate that CSCO presence enhanced firm performance during COVID-19 for firms in industries generally associated with high supply (H2b) and demand (H2c) risk exposure. Hence, we would argue that exposures to demand and supply risks in these industries were not all bad: they came with the silver lining of having prepared the CSCOs and their firms for the uncertainty of the pandemic. Economically, CSCO impact translated into meaningful financial performance improvements through substantial cost savings in sales (∼12–14%), accelerated cash conversion cycles, and improved capacity utilization, demonstrating practical and economically valuable benefits during periods of heightened uncertainty. More generally, the presence of CSCO was found to alleviate negative supply shocks and reinforce positive demand shocks. Particularly for firms experiencing concurrent negative supply and positive demand shocks, CSCO presence exhibits the strongest positive impact on both sales growth and return on assets. This effect was the largest of any shock scenario, showing CSCO performance impacts are present particularly under the most challenging of supply chain operating environments. Our post-hoc analysis of stock market returns also implies that CSCO presence in a firm significantly dampened extreme stock price movements during the pandemic.

In this section, we outline our contributions (1) to CSCO and TMT research more broadly, (2) to OIPT theory, and (3) in terms of datasets and methods.

Wagner and Kemmerling (2014) found a negative association between CSCO presence and firm performance, prompting the need to understand under which circumstances this link could be positive (Roh et al., 2016). In contrast to previous studies with associative analysis, we use triple differencing to provide a more robust identification of a link between CSCO presence and firm performance under COVID-19 and for firms in industries with prior supply chain risk exposure. Our findings show that improved cost of sales, faster cash conversion cycles, and improved capacity utilization were all important mechanisms through which a CSCOs impact firm performance under COVID-19 for firms in industries previously associated with high supply (H2b) and demand (H2c) risk exposure. We infer that firms with a “risk-experienced” CSCO were better able to match supply and demand under the COVID-19 induced uncertainty and hence make better use of their physical assets during a time when material shortages, delays, unexpected demand spikes, and large order cancellations collectively plagued global supply chains. The significant effect of these particular mechanisms highlights the importance of improved internal and external integration (Körber and Cotta, 2020; Roh et al., 2016), visibility in the chain (Swink et al., 2012), and flexibility (Kroes et al., 2021) that CSCO can enable, ultimately making a difference.

Roh et al. (2016) note how the performance effects of a CSCO occur for early adopters, while Kroes et al. (2021) suggest the benefits increase as the position becomes more established. Our results do not speak of the early adoption per se but emphasize the role of learning when it comes to managing uncertainty. The positive performance impact of a CSCO during COVID-19 was conditional on the firm having been exposed to high supply and/or demand risks prior to the pandemic; giving the CSCO necessity and experience in building a resilient supply chain. These findings provide interesting contributions not only to SCM but to upper echelons research in general, complementing previous findings of the importance of top executive experience (Hutzschenreuter and Horstkotte, 2013). We show the overall impact of a CSCO on sales growth (profitability) during COVID-19 to be 9.8% (3.0%) in industries prone to high supply risks and 13.4% (4.0%) in industries prone to high demand risks, respectively. These results demonstrate the significant performance impact that CSCOs can bring to firms where they have likely driven previous efforts to develop a more risk resilient chain based on industry risk background. This is in line with Hora and Klassen (2013), who examine the impact of experiential learning through own as well as operationally similar firms’ risks.

As per organizational information processing theory, information processing capacity can be increased through vertical information systems or lateral relations. While the role of vertical information systems in the context of SCM and COVID-19 has been shown to enhance performance, lateral relations from an OIPT perspective have been scarcely examined in either context. Srinivasan and Swink (2018) suggest that regardless of the maturity of information systems, OIPT would suggest that competitiveness will still also rely on organizations ability to shift priorities and resources. They further argued for a need to examine such other assets from an OIPT perspective. We argue that a CSCO can provide such high-level lateral connections through the top management team which enable enhanced information processing capabilities that are particularly relevant in a highly uncertain environment. While the “standard” industry risk exposure associated results did not prove this to be the case, under COVID-19 this was demonstrated, particularly if previous risk exposure existed – perhaps having helped develop the use of those lateral information processing capabilities to then work in extreme situations. This aligns with Klöckner et al. (2023b) who emphasize the need for firms to build dynamic capabilities that can activate the necessary crisis responses. While the pandemic has had serious impacts on firms’ supply chains (Xiong et al., 2021) and hence performance (Hu and Zhang, 2021), not all companies have been equally impacted and companies with more coordinated supply chain strategies have emerged as winners (Choi et al., 2020). Our results suggest that having a CSCO is a key organizational mechanism for building those capabilities to respond to an uncertain operating environment. In essence, we suggest that the CSCO’s high-level mandate to develop SCM within the firm contributes to building the necessary internal information processing capabilities, which then emerge as performance-enhancing during periods of extreme uncertainty. Indeed, as our results suggest it is particularly during the most difficult conditions (combining negative supply shocks with positive supply shocks) that CSCO performance impact becomes most prominent, it appears extremely uncertain environments are where CSCOs are the best “fit” in terms of OIPT.

Earlier studies have not explicitly accounted for CSCOs’ self-selection to best-performing firms or to industries where supply chain risks are more acute. This study largely avoids these selection issues by subsampling, controlling for confounding factors, and exploiting the unpredictable, exogenous shock in firm performance caused by the COVID-19 pandemic. In this study, the triple-differences approach is employed to enable more robust identification regarding the impact of the CSCO on firm performance under supply and demand risk exposure, as well as the uncertainty introduced by COVID-19 in terms of negative and positive shocks. We have further strengthened the validity of our results by using alternative model specifications and risk exposure measurement, and alternative variable definitions (e.g. Supply Chain and Operations Management Executive, SCOME, instead of CSCO).

We have used a novel dataset in the operations and supply chain management field to estimate firm exposure to supply and demand risks and COVID-19 shocks, namely, earnings call transcripts. Public corporate disclosures have been a key source for textual sentiment analysis in finance, given their role as official releases from insiders with better knowledge of the firm (or in our case, its supply chain). The style and tone of these disclosures can provide useful information beyond financial statements (Loughran and McDonald, 2016). Drawing from recent studies that use textual analysis to quantify firm risk and uncertainty (Hassan et al., 2019, 2023, 2024a, b), we have demonstrated how the dataset can be utilized to differentiate industries and firms based on their supply and demand risk exposure levels as well as positive and negative shocks to supply and demand. Furthermore, also computational linguistics represents a methodological advancement as textual and sentiment analysis has only scarcely been used in supply chain studies (Wood et al., 2017).

The European Central Bank suggests that supply chain shocks are the cause of a third of the strains on global production networks (Attinasi et al., 2021). With potential future pandemics, extreme weather events from climate change, and ongoing and increasing geopolitical uncertainty, firms need to rethink their SCM strategies. The learnings in dealing with COVID-19 will likely help CSCOs deal better with the next “black swan” event. This is also demonstrated by the risk-mitigating effect of CSCOs documented in our post-hoc analysis of the stock market returns. Markets increasingly appreciate the strategic competitiveness that SCM can bring and reward firms that can demonstrate it through structure and actions. Since the pandemic, there has indeed been a 72% increase in the hires of supply chain directors (Oglesbee, 2022). However, our results did not prove a universal performance improvement from the presence of a CSCO, it was demonstrated only during COVID-19 for firms with previous supply or demand risk exposure. Our findings emphasize the learning effect for CSCOs, which may suggest the need to give CSCOs sufficient time to build supply chain resilience through experience that only comes from having had to deal with risk outcomes. Thus, firms should not be too quick to judge the worth of a new hire. Yet we do suggest being quick to hire CSCOs as there is “no time like the present to learn from risks and uncertainties” in the current global operating environment and create the information processing capacities needed to deal with the next black swan.

Our results regarding the CSCO performance impact during COVID-19 being contingent on the firm belonging to an industry with a history of high risk exposure are also interesting in light of the findings on stock markets reacting more positively to outsider SCOME appointments (Hendricks et al., 2015). Our results suggest that CSCO impacts may not be immediate and may indeed depend on the demand and supply risk and uncertainty exposure characteristics of the CSCO’s operating environment.

There has been limited research on CSCOs, but the current turbulent environment for supply chains, along with our positive performance results derived from CSCO presence during COVID-19 for previously risk-exposed industries, strongly argue that the role is likely to become more common – and more useful. We encourage more research on the performance impacts of CSCOs under different external environments.

While we examined the industry-specific impacts of COVID-19, we did not control for geographical location factors such as port delays and pandemic-related labor restrictions for the firms we examined. We acknowledge this as a limitation of our study.

Furthermore, and importantly from our theoretical (OIPT) perspective, our methodology identifies the presence of executive positions but cannot assess how effectively these roles are performed. A CSCO’s impact likely stems not only from the existence of the position but also from the individual’s leadership qualities and coordination abilities and the specific amount and strength of the lateral relations, which are not directly observable. Thus, our data permit us to examine only whether dedicated supply chain leadership exists structurally and whether it brings with it information processing capabilities, but we cannot examine the quality or effectiveness of these capabilities. We suggest further research should engage in more qualitative inquiry on how and in which ways CSCOs create and use such capabilities.

It is also important to note from an OIPT perspective that investments in vertical information systems and creating lateral relations are not mutually exclusive (Fairbank et al., 2006). There is research examining the former, and we have focused on CSCOs as a manifestation of the latter. Yet from a complementarity theory perspective, it is argued that the value of one resource increases when combined with other related resources (Al-Sheyadi et al., 2019). Hence it would be interesting to examine the performance impacts of a CSCO when combined with different maturity levels of SCM related information systems.

The novel textual dataset and the application of computational linguistics offer interesting research avenues. Specifically, the data and method combined provide opportunities to understand how SCM function visibility has evolved in corporate disclosures and in what contexts SCM is discussed. These are particularly relevant given the availability of increasingly sophisticated textual analysis tools and field-specific dictionaries (Bochkay et al., 2022). Supply chains interest a broad set of stakeholders, and hence what corporate disclosures communicate about SCM needs to be examined.

The authors wish to acknowledge Andreas Wieland, Bjørn Jørgensen, Emma-Riikka Myllymäki, Sami Vähämaa, Seongtae Kim, Simone Traini, and Ulf Mohrmann for their insightful comments and suggestions. They also thank the seminar participants at Aalto University and the Norwegian School of Economics, and attendees of the “Organizing Risk and Managing Supply Chains Workshop” at Copenhagen Business School.

The supplementary material for this article can be found online.