Resource Orchestration,Team Faultlines,and Entrepreneurial Resilience

ROT and Entrepreneurial Resilience in Crisis Contexts

Strategic entrepreneurship research has long emphasized that firms create value by exploring entrepreneurial opportunities while exploiting existing advantages (Hitt et al., 2001), and that doing so depends not only on what resources they possess but also on how those resources are deployed and combined (Baert et al., 2016; Sirmon et al., 2011). ROT extends this insight by shifting attention from resources as static stocks to the managerial actions through which they are structured, bundled, and leveraged to create capabilities and value (Sirmon et al., 2007, 2011). Prior research has further refined these processes across diverse settings (Amit & Han, 2017; Baert et al., 2016; Chadwick et al., 2015; Chirico et al., 2025). However, this literature has paid less attention to how new ventures orchestrate distinct internal resource bundles to build capabilities critical for coping with crises.

This gap is important because crisis conditions place unusual demands on new ventures (Anwar et al., 2023). In this study, we define entrepreneurial resilience as an organizational-level capability, namely, a venture’s capacity to maintain stability while remaining flexible in the face of environmental disturbances, rather than as an individual or team-level psychological attribute (Ahmed et al., 2022; Preller et al., 2023). For new ventures, developing such resilience is especially challenging because they are often constrained by the liabilities of smallness and relatively weak margins for error (J. Freeman et al., 1983; Thornhill & Amit, 2003). Although interfirm collaboration can provide access to external resources and knowledge (Katila et al., 2022; Wiklund & Shepherd, 2009), crises often make such ties less reliable, more fragile, or temporarily inaccessible (Batjargal et al., 2023; Williams et al., 2017). Under such conditions, the ability to organize internal resources becomes especially consequential. We therefore draw on ROT to explain how internal resource bundles are configured and deployed under constraints, and how complementarities, compensation, and trade-offs across resource domains shape entrepreneurial resilience under crisis.

The Entrepreneurial Team Perspective and the Faultlines Perspective: The Actor and Team Structure of Resource Orchestration

To understand how internal resource bundles shape entrepreneurial resilience in new ventures, it is necessary to clarify who enacts resource orchestration and how this process is internally structured (Harper, 2008; Ndofor et al., 2015). In new ventures, resource orchestration is rarely enacted by a single founder. Rather, it is typically undertaken by an entrepreneurial team (West, 2007). Resource orchestration therefore needs to be located in the entrepreneurial team context.

The entrepreneurial team perspective emphasizes that entrepreneurial action is inherently collective (Harper, 2008; West, 2007). New ventures are often founded and developed by teams whose members bring heterogeneous knowledge, functional backgrounds, industry experience, and cognitive styles (Jin et al., 2017). Entrepreneurial teams jointly identify opportunities and threats, mobilize stakeholders, allocate scarce resources, and make strategic decisions (West, 2007). They should therefore not be treated merely as collections of individual attributes, but as the immediate actors through which resource orchestration occurs (Ndofor et al., 2015). Team members’ heterogeneous expertise and experience influence how environmental shocks are interpreted, which bottlenecks are prioritized, and which resources are preserved, recombined, or leveraged (Motley et al., 2023). In this sense, the entrepreneurial team perspective specifies who performs resource orchestration.

However, entrepreneurial team literature has paid less attention to how task-relevant differences within the team are structurally organized. The faultlines perspective addresses this issue. Faultlines emerge when multiple member attributes align to create potential subgroup boundaries within the team (Lau & Murnighan, 1998; Meyer & Glenz, 2013). Task-related faultlines are particularly relevant in entrepreneurial settings because they arise from the alignment of work-relevant attributes such as functional background, expertise, or tenure (Cooper et al., 2014; Richard et al., 2019). Unlike relationship-related faultlines, which are more likely to activate identity-based divisions and relational conflict, task-related faultlines primarily shape how knowledge bases, attention patterns, and problem-solving responsibilities are distributed within the team (Richard et al., 2019). In this sense, the faultlines perspective specifies how differences within the entrepreneurial team are organized and become consequential for resource orchestration.

These two perspectives are therefore not substitutes, but complements. The entrepreneurial team perspective identifies the collective actor through which resource orchestration occurs (West, 2007), whereas the faultlines perspective explains how differences inside that collective actor are structured (Chung et al., 2015; Ndofor et al., 2015). Integrating them allows us to conceptualize ETTF as the structural manifestation of task-relevant differences within entrepreneurial teams. ETTF captures not merely the presence of heterogeneity, but the way such heterogeneity is aligned and organized within the team. In this sense, ETTF shapes how entrepreneurial teams allocate attention, integrate specialized knowledge, and coordinate resource-related responses under disruption. This helps explain why entrepreneurial teams face similar resource bundles and crisis conditions may nevertheless differ in how effectively they orchestrate resources.

An Integrative Theoretical Framework Linking Resource Bundles, Entrepreneurial Resilience, and ETTF

Building on the foregoing discussion, this study develops a three-theory integration that combines ROT, the entrepreneurial team perspective, and the faultlines perspective to explain entrepreneurial resilience in new ventures under crisis conditions. Each perspective plays a distinct but connected role. ROT explains why resource configurations matter for resilience (Sirmon et al., 2011); the entrepreneurial team perspective specifies who enacts resource orchestration (Harper, 2008; West, 2007); and the faultlines perspective explains how the internal structure of the entrepreneurial team shapes that process (Chung et al., 2015; Richard et al., 2019). Taken together, these perspectives provide the basis for explaining how internal resource conditions are translated into resilience outcomes through entrepreneurial teams.

Within this framework, we focus on two resource domains that are especially consequential for new ventures and that perform distinct yet interdependent functions: social resources and technological resources. Social resources refer to the venture’s relationships with key stakeholders, such as partners, investors, government agencies, suppliers, and customers, and primarily provide external access, legitimacy, and complementary assets (Batjargal, 2003; Florin et al., 2003; Grichnik et al., 2014; J. Tan et al., 2015). Technological resources refer to knowledge stocks, patents, and R&D capabilities, and primarily provide the capability base for continuity, alternative technical pathways, and adaptive change (Bianchi et al., 2014). These two resource domains may each exist under conditions of slack or constraint; more importantly, they do not operate in isolation but form bundles whose internal interplay affects venture responses to shocks (Paeleman & Vanacker, 2015). We therefore conceptualize a two-by-two matrix of resource bundles, which combines slack and constraints in social and technological resources into four representative bundle types (see Figure 1).

OPEN IN VIEWER

Figure 1.

Bundles of social and technological resource slack/constraints.

Their combination yields four representative bundle types: dual slack (Quadrant 1), social resource slack with technological resource constraint (Quadrant 2), technological resource slack with social resource constraint (Quadrant 3), and dual constraint (Quadrant 4).

Bundles of Resource Slack/Constraints and Entrepreneurial Resilience

We now develop hypotheses on how social and technological resource slack/constraints shape entrepreneurial resilience. Given that stability and flexibility correspond to distinct crisis tasks (DesJardine et al., 2019), we propose that stability is primarily driven by a resource compensation logic (Ennen & Richter, 2010; George, 2005; J. Tan, 2003), while flexibility is driven by a crisis-induced resource reconfiguration logic (Attah-Boakye et al., 2023; Colombo et al., 2021).

Resource Misalignment and Entrepreneurial Resilience

The bundle comparisons above show how distinct configurations of social and technological resource slack and constraints shape entrepreneurial resilience. Yet comparing discrete quadrants alone does not fully capture whether the degree of imbalance between the two resource domains also carries systematic implications for resilience (Edwards & Parry, 1993). Two ventures may fall into the same broad bundle category yet differ substantially in how far one resource domain exceeds the other. To address this issue, we extend the quadrant-based comparison with a response surface perspective that focuses on resource misalignment (Edwards et al., 2002). This extension builds directly on the two mechanisms developed above, that is, resource compensation for stability and crisis-induced resource reconfiguration for flexibility. Resource misalignment does not introduce a new logic. Rather, it captures whether increasing divergence between the two domains weakens compensatory buffering while simultaneously stimulating reconfiguration.

We focus specifically on misalignment because entrepreneurial action under disruption depends not only on the absolute level of each resource but also on whether the two resource domains can function together as an actionable system. Entrepreneurial teams rely on social resources to activate, extend, and legitimize technological capacity, while relying on technological resources to give operational content to the support mobilized through social ties (Lee et al., 2001; Yli-Renko et al., 2001). When one domain substantially exceeds the other, this functional interdependence becomes strained.

For stability, greater misalignment should be detrimental because it undermines the compensatory logic required for effective buffering. Stability depends on whether entrepreneurial teams can combine external support with internal continuity in a way that prevents shocks from escalating into severe losses (DesJardine et al., 2019). When social resources substantially exceed technological resources, the venture may obtain patience, coordination, and legitimacy from external actors (Zimmerman & Zeitz, 2002), yet still lack the technical depth needed to maintain reliable operations (Lee et al., 2001). When technological resources substantially exceed social resources, the venture may possess valuable technical reserves, yet lack the relational channels needed to validate, mobilize, and protect those reserves under crisis conditions (Stuart et al., 1999). In either case, the relatively stronger domain cannot fully compensate for the missing contribution of the weaker one. As misalignment increases, buffering becomes more fragmented, protection less coherent, and losses more likely to cascade (Lynn, 2005). Thus, greater misalignment should increase severity of loss.

For flexibility, however, greater misalignment can be advantageous because it sharpens the focal problem around which crisis-induced reconfiguration occurs. Rapid recovery requires entrepreneurial teams to identify the most serious bottleneck and redirect available resources toward that bottleneck without delay (Williams et al., 2017). When social and technological resources are relatively balanced, teams may be more likely to remain within inherited bundling patterns and pursue incremental adjustment (Gilbert, 2005). By contrast, pronounced misalignment makes the weak point in the venture’s resource structure more visible and creates stronger pressure to act (Greve, 2003). This pressure activates a reconfiguration process in which the relatively stronger domain is mobilized to offset deficiencies in the weaker one (Karim & Capron, 2016). Social strength, for example, can be used to obtain external knowledge, collaboration, legitimacy, or market access that compensates for technical shortfalls (Gil et al., 2022; Li et al., 2023); technological strength can be used to redesign offerings, processes, or problem-solving routines when external support is thin (Haase & Eberl, 2019; Parida & Örtqvist, 2015). In this sense, misalignment does not merely represent uneven resource endowments. It creates a clearly identifiable adaptation challenge that organizes search, recombination, and restructuring. As misalignment increases, entrepreneurial teams are more strongly pushed away from passive preservation and toward active recovery-oriented reconfiguration (Acar et al., 2019; Baker & Nelson, 2005). Because such reconfiguration is guided by a more visible bottleneck, teams can reduce dispersed search, prioritize resource redeployment move quickly, and more faster from recognizing the problem to implementing adaptive responses. Thus, greater misalignment should shorten time to recovery. Accordingly, misalignment between social and technological resources should have opposite implications for the two dimensions of entrepreneurial resilience.

The Moderating Effect of ETTF

As argued above, the effects of social and technological resource bundles on entrepreneurial resilience depend on how effectively resources are orchestrated under disruption (Sirmon et al., 2011). In new ventures, such orchestration is enacted by entrepreneurial teams (West, 2007), and we argue that ETTF constitutes an important boundary condition shaping this process. ETTF matters because it structures how task-relevant expertise is distributed within the team, which in turn affects how members allocate attention, process resource problems, and coordinate responses under crisis (Chung et al., 2015; Richard et al., 2019; Rico et al., 2012).

With respect to stability, ETTF should strengthen the resource compensation logic developed above. Stability depends on whether social and technological resources can be combined into a coherent buffering system that preserves external support and internal continuity (DesJardine et al., 2019). This requires entrepreneurial teams not only to recognize where the venture is most vulnerable but also to mobilize complementary resources quickly enough to prevent temporary shortfalls from escalating into severe loss (Klotz et al., 2014; Williams et al., 2017). Stronger task-related faultlines can facilitate this process in two ways. First, they promote selective attention across resource domains (Georgakakis et al., 2017). Some subgroups may focus more on external coordination such as stakeholder support, legitimacy, and access to complementary assets, whereas others may focus more on process continuity, technical reserves, and fallback options (Chung et al., 2015; Richard et al., 2019). Such differentiation enables the team to monitor multiple sources of vulnerability simultaneously rather than relying on a single undifferentiated response (Thatcher et al., 2024). Second, task-related faultlines can improve compensatory coordination across domains (Gibson & Vermeulen, 2003). Because subgroup differentiation is organized around task demands rather than social identity, specialized subgroups can contribute domain-specific knowledge while still recognizing the need for cross-domain integration (Chung et al., 2015; Espinosa et al., 2007). As a result, entrepreneurial teams with higher ETTF should be better able to identify where one resource domain is falling short and mobilize another domain to compensate for it. In this sense, ETTF increases the likelihood that resource bundles are translated into coordinated buffering responses, thereby attenuating differences in severity of loss across resource bundles when ETTF is high.

With respect to flexibility, ETTF should strengthen the crisis-induced resource reconfiguration logic developed above. Flexibility depends on whether entrepreneurial teams can identify bottlenecks, abandon inherited deployment patterns, and redirect available resources toward adaptive recombination (DesJardine et al., 2019). This requires both rapid search and integrative recombination: teams must generate alternative solutions while also assembling them into a new workable action path (Attah-Boakye et al., 2023; Colombo et al., 2021). Stronger task-related faultlines can facilitate this process by increasing both the variety and the organization of crisis response. First, differentiated subgroups make it easier for entrepreneurial teams to conduct parallel search across distinct knowledge domains, thereby increasing the speed with which potential alternatives are surfaced (Georgakakis et al., 2017; Gibson & Vermeulen, 2003). Second, because the outputs of these subgroup efforts are organized around specialized task domains, they are more recombinable and therefore more useful for constructing a new viable action path (Chung et al., 2015; Richard et al., 2019). In this way, ETTF does not merely increase variety; it organizes parallel search and makes the results of that search more actionable for crisis-driven adaptation. ETTF therefore helps entrepreneurial teams convert resource pressure into parallel search, focused experimentation, and faster integration of alternative action paths. As a result, entrepreneurial teams with higher ETTF should be better able to translate differences in resource bundle conditions into more pronounced differences in reconfiguration effectiveness and recovery timing.

Data Sources

Our study utilizes financial and stock data from new ventures listed on the National Equities Exchange and Quotations (NEEQ). Following prior studies (Buyl et al., 2019; DesJardine et al., 2019; Sajko et al., 2021), we investigate the impact of external shock events on entrepreneurial resilience at the organizational level. In this context, crises are defined as systemic shocks that are low-probability, unanticipated, and high-impact, thereby creating widespread market disruption with enduring implications for firm survival and performance (Williams et al., 2017). Specifically, we focus on three systemic crises that generated economy-wide disruption: the 2015 Chinese economic downturn, the 2018 U.S.–China trade frictions, and the COVID-19 pandemic. We classify these events as “economic crises” because they produced systemic economic losses across a broad population of firms. The observation period spans from 2013 to 2020, allowing us to capture these shocks and observe sufficient post-shock recovery time (Buyl et al., 2019). We constructed an unbalanced short-panel dataset based on startups affected by these three shocks.

Data were drawn from multiple databases. Financial and stock data were obtained from the WIND database, covering information on all companies listed on China’s NEEQ. Missing data were supplemented using the iFinD database. Patent data were sourced from the PatSnap database. Additionally, executive data for the new ventures were manually extracted from company prospectuses and annual financial reports.

The final dataset was constructed following these criteria: we included firms established within the last 10 years (Kiss & Barr, 2017) and listed on the NEEQ before 2019, ensuring that these firms had encountered at least one external shock. We excluded companies that were specially treated (e.g. ST and *ST), those with missing data, and firms that experienced significant events during the sample period, such as litigation, mergers, or major management changes. This was done to minimize the potential impact of other significant events on firm stock price volatility (DesJardine et al., 2019). The final database contains 345 new ventures and 528 observations.

Measures

Method of Analysis

Given the unique characteristics of our variables, we employ two distinct estimation methods to test our hypotheses. First, to assess the relationship between resource bundles and severity of loss, we use a two-way fixed effects model with robust standard errors, accounting for firm and year effects. This approach partially addresses endogeneity concerns arising from omitted variables (Hill et al., 2021). Second, for the relationship between resource bundles and time to recovery, we follow Buyl et al. (2019) and DesJardine et al. (2019) by applying an event history model. This method accommodates right-censored data, as not all ventures recovered from the shocks within the observation window. Excluding such observations could introduce bias (DesJardine et al., 2019). Accordingly, we use parametric survival-time models with an exponential distribution, a common method in event history analysis (Buyl et al., 2019), and apply robust variance estimation to compute standard errors. To further examine our hypotheses concerning resource misalignment, we complement our main analyses with response surface analysis (RSA). RSA allows us to move beyond discrete bundle comparisons and capture the continuous effects of divergence between social and technological resources. Following prior congruence research (Croonen et al., 2025), we estimate polynomial models including the linear, squared, and interaction terms of social and technological resources, and use these coefficients to derive the response surface parameters.

Results for Severity of Loss

Table 3 presents results from the fixed effects model examining the relationship between resource bundles and severity of loss (stability).

OPEN IN VIEWER

Table 3.

Fixed Effects Model Results for Severity of Loss.

	(1)	(2)	(3)
Variables	All firms	Low FLSTask	High FLSTask
Resource constraints in technological resources (Quadrant 2)	0.189**(0.079)	0.255**(0.098)	0.079 (0.116)
Resource constraints in social resources (Quadrant 3)	0.245***(0.078)	0.328***(0.114)	0.066 (0.080)
Dual resource constraints (Quadrant 4)	0.271***(0.079)	0.357***(0.095)	0.095 (0.154)
TMT size	0.032 (0.028)	−0.010 (0.064)	0.053 (0.038)
Firm size	−0.009 (0.077)	0.111 (0.128)	−0.081 (0.102)
Firm age	1.062**(0.510)	1.236**(0.622)	0.615 (0.851)
Share price	0.009*(0.005)	0.007 (0.008)	0.008 (0.006)
Profitability	−0.263 (0.397)	−0.712**(0.330)	0.332 (0.645)
Operational efficiency	0.227 (0.379)	0.310 (0.408)	−0.156 (0.605)
Constant	−1.654 (1.025)	−2.599**(1.225)	−0.236 (1.821)
Firm dummy	Included	Included	Included
Year dummy	Included	Included	Included
N	528	268	260
Adjusted R2	0.152	0.211	0.119

Note. SE = Standard error; TMT = Top management team; FLS = Faultline strength.

Robust SE in parentheses, *p < .1. **p < .05. ***p < .01.

Following Paeleman and Vanacker (2015), we use dual resource slack (Quadrant 1) as the reference category, with the remaining three quadrants interpreted relative to it. Model 1 reports results for the full sample, testing H1a and H1b. Models 2 and 3 focus on subsamples based on ETTF, testing H4a.

First, Model 1 results support H1a and H1b. Specifically, ventures with social resource slack but technological constraint (Quadrant 2) incur greater losses compared to those with dual resource slack (β = .189, p < .05). Similarly, ventures with technological resource slack but social constraint (Quadrant 3) show greater losses (β = .245, p < .01), and those with dual resource constraints (Quadrant 4) suffer the highest losses (β = .271, p < .01). Second, coefficient magnitudes in Model 1 show a clear ordering (Q2 < Q3 < Q4). The coefficient ordering is directionally consistent with H1b. Third, the subgroup pattern in Models 2 and 3 is consistent with H4a. Model 2 reports outcomes for ventures with low faultlines, while Model 3 focuses on those with high faultlines. In Model 2, ventures with resource constraints (Quadrants 2, 3, and 4) in the low faultline subsample exhibit significant differences in severity of loss, with all coefficients significant (β₁ = .255, p < .05; β₂ = .328, p < .01; β₃ = .357, p < .01). However, in Model 3, where ventures have high faultlines, none of the coefficients are significant, indicating that high faultlines diminish the differences between ventures with constraints and those with dual resource slack. This shift from significant differences under low faultlines to nonsignificant differences under high faultlines suggests that high ETTF attenuates differences in severity of loss across resource bundles, which is directionally consistent with H4a.

Results for Time to Recovery

Table 4 presents results from the parametric survival-time model, assessing the relationship between resource bundles and time to recovery. Following Paeleman and Vanacker (2015), dual resource slack (Quadrant 1) serves as the reference category. In the event history analysis, exponentiated coefficients are interpreted as proportional changes in the recovery hazard relative to the reference category, calculated as [exp(coefficient)−1] × 100% (DesJardine et al., 2019). Model 1 examines the full sample, testing H2a and H2b, while Models 2 and 3 analyze subsamples based on ETTF, examining H4b.

OPEN IN VIEWER

Table 4.

Parametric Survival-Time Model Results for Time to Recovery.

	(1)	(2)	(3)
Variables	All firms	Low FLSTask	High FLSTask
Resource constraints in technological resources (Quadrant 2)	0.687**(0.276)	−0.206 (0.328)	1.449***(0.351)
Resource constraints in social resources (Quadrant 3)	0.677***(0.261)	−0.050 (0.299)	1.147***(0.366)
Dual resource constraints(Quadrant 4)	0.571**(0.240)	0.007 (0.248)	0.891***(0.346)
TMT size	−0.067 (0.048)	−0.056 (0.063)	−0.107 (0.068)
Firm size	0.021 (0.059)	−0.003 (0.072)	−0.007 (0.091)
Firm age	−0.281 (0.309)	−0.278 (0.401)	−0.271 (0.435)
Share price	−0.001 (0.009)	−0.001 (0.010)	−0.014 (0.013)
Profitability	−0.480*(0.272)	−0.198 (0.315)	−0.557 (0.479)
Operational efficiency	0.497*(0.276)	0.376 (0.347)	0.427 (0.466)
Constant	−2.776***(1.063)	−2.160*(1.251)	−2.125 (1.426)
Industry dummy	Included	Included	Included
Year dummy	Included	Included	Included
Observations	528	268	260
Recovery events	496	255	241
Time at risk	1,863	856	1,007
Log likelihood	−817.247	−375.500	−397.582

Note. Recovery events indicate the number of firms that completed recovery within the observation window, while the remaining cases are right-censored (i.e. recovery not observed by the end of the study period). Time at risk represents the total duration of exposure across all firm–crisis spells until either recovery occurred or the observation was censored. SE = Standard error; TMT = Top management team; FLS = Faultline strength.

Robust SE in parentheses, *p < .1. **p < .05. ***p < .01.

First, our results support H2a, indicating that new ventures with dual slack take longer to recover than those facing resource constraints. Ventures with constraints in technological resources (Quadrant 2) exhibit a higher recovery hazard than those with dual resource slack (β = .687, p < .05). Similarly, ventures with constraints in social resources (Quadrant 3) and dual resource constraints (Quadrant 4) also exhibit higher recovery hazards than those with dual resource slack (β = .677, p < .01; β = .571, p < .05, respectively). Second, the coefficient pattern is directionally consistent with H2b, with Quadrants 2, 3, and 4 ordered as predicted in terms of recovery hazard, showing that ventures with constraints in technological resources outperform those with constraints in social resources, and both perform better than those with dual constraint.

Third, Models 2 and 3 are consistent with H4b, showing that differences in time to recovery across resource bundles are more pronounced when ETTF is high. Model 2, which includes the subsample with low faultlines, reveals no significant differences in recovery times among new ventures with resource constraints (i.e. Quadrants 2, 3, and 4) compared to those with dual resource slack (i.e. Quadrant 1). However, in the high faultlines subsample, Model 3 demonstrates significant coefficients (β₁ = 1.449, p < .01; β₂ = 1.147, p < .01; β₃ = .891, p < .01), indicating that ventures with resource constraints recover more effectively than those with dual slack. This pattern is directionally consistent with H4b, suggesting that high ETTF makes differences in time to recovery across resource bundles more pronounced.

RSA Results on Resource Misalignment

We test Hypotheses 3a and 3b by examining the effects of resource misalignment on severity of loss and time to recovery using RSA (Table 5 and Figure 2). Figure 2 presents polynomial response surface plots, with Panel A depicting predicted loss severity and Panel B depicting predicted recovery hazard. In both panels, the surfaces display the predicted outcomes when social and technological resources take values ranging from two standard deviations below the mean to two standard deviations above the mean, including all intermediate values.

OPEN IN VIEWER

Table 5.

Polynomial Regression and Response Surface Analysis Results.

	(1)	(2)
Variables	Severity of loss	Time to recovery
Social resource slack	−0.031**(0.013)	−0.053***(0.018)
Technological resource slack	0.009 (0.008)	−0.010 (0.013)
Social resource slack2	0.003***(0.001)	0.002**(0.001)
Social resource slack × Technological resource slack	−0.006**(0.003)	−0.013***(0.003)
Technological resource slack2	−0.001 (0.000)	−0.001 (0.001)
TMT size	0.028 (0.025)	−0.054 (0.046)
Firm size	−0.031 (0.075)	−0.009 (0.058)
Firm age	1.233**(0.528)	−0.289 (0.307)
Share price	0.008 (0.005)	−0.001 (0.008)
Profitability	−0.169 (0.403)	−0.473*(0.271)
Operational efficiency	0.112 (0.378)	0.476*(0.271)
Constant	−1.444 (1.043)	−2.483**(1.202)
Model statistics
Adjusted R2	0.143
Log likelihood		−806.503
Joint significance of second-order terms	F-statistic = 3.53, p < .05	Wald chi-square = 36.90, p < .01
Slope of the line of congruence	−0.022 (0.016)	−0.064***(0.023)
Curvature of the line of congruence	−0.004 (0.003)	−0.012***(0.004)
Slope of the line of incongruence	−0.040***(0.015)	−0.043**(0.021)
Curvature of the line of incongruence	0.008**(0.004)	0.014***(0.002)

Note. n = 528. Industry and year fixed effects included in our models.

Robust SE in parentheses, *p < .1. **p < .05. ***p < .01. SE = Standard error; TMT = Top management team.

OPEN IN VIEWER

Figure 2.

Polynomial response surface graphics for severity of loss and recovery hazard. Panel A: Predicted severity of loss; Panel B: Predicted recovery hazard.

Consistent with Hypothesis 3a, the curvature along the line of incongruence (LOIC) for severity of loss is positive and significant (a₄ = 0.008, p < .05), indicating that as the discrepancy between social and technological resources increases, predicted loss severity increases. As shown in Figure 2A, the response surface bends upward along the LOIC, suggesting that greater resource misalignment exacerbates the immediate impact of shocks on entrepreneurial ventures.

For time to recovery, we again observe a positive and significant curvature along the LOIC (a₄ = 0.014, p < .001). Given that the event history specification models the recovery hazard, this result indicates that greater misalignment between social and technological resources is associated with a higher hazard of recovery—that is, faster recovery and a shorter time to recovery—supporting Hypothesis 3b. As illustrated in Figure 2B, the response surface also bends upward along the LOIC, indicating that extreme misalignment is associated with faster post-shock recovery.

Taken together, these findings reveal that resource misalignment has distinct implications for the two dimensions of entrepreneurial resilience. While greater misalignment intensifies loss severity, it simultaneously accelerates post-shock recovery. This dual effect suggests that misalignment weakens ventures’ immediate shock-absorbing capacity but may also trigger more rapid post-shock adjustment.

Robustness Checks

To ensure the robustness of our analysis, we specified several alternative models and variables. Detailed results of these robustness checks are provided in the Supplemental Appendix.

Theoretical Implications

This study makes several contributions to theory and empirical research. First, we contribute to ROT by explicitly integrating entrepreneurial team and faultline perspectives into ROT, thereby advancing a three-theory framework that explains how internal resource bundles shape entrepreneurial resilience under crisis conditions. Although ROT has long emphasized that capabilities arise from structuring, bundling, and leveraging heterogeneous resources (Amit & Han, 2017; Sirmon et al., 2011), empirical research has more often focused on isolated resources or on the alignment between resource investments and leveraging strategies (Sirmon & Hitt, 2009; Symeonidou & Nicolaou, 2018). Our study shifts attention to within-firm resource configurations by showing that the resilience implications of social and technological resources depend not only on whether each domain is characterized by slack or constraint but also on how the two domains are bundled together and the degree to which they are misaligned. In this way, we extend ROT’s bundling logic by demonstrating that different combinations of social and technological resource slack and constraints generate systematically different resilience outcomes under crisis conditions. Our findings further show that the degree of misalignment itself also matters: greater misalignment undermines stability while facilitating faster recovery. We further clarify why these effects differ across resilience dimensions by linking stability to a resource compensation logic and flexibility to a crisis-induced resource reconfiguration logic. Taken together, these findings move ROT beyond single-resource reasoning and static bundle typologies by showing that both bundle composition and resource misalignment shape the development of distinct resilience capabilities in new ventures under crisis.

Second, we contribute to research on entrepreneurial teams (Harper, 2008; West, 2007) and faultlines (Lau & Murnighan, 1998; Thatcher et al., 2024) by extending the explanatory role of ETTF beyond their direct implications for organizational outcomes. Prior entrepreneurial team research has primarily examined the direct effects of team characteristics—such as team size, heterogeneity, or aggregated human capital—on venture outcomes (Jin et al., 2017; Klotz et al., 2014). Faultline research, in turn, has mainly focused on how aligned subgroup structures shape team processes such as conflict, coordination, and information processing (Lau & Murnighan, 1998; Thatcher & Patel, 2012). By integrating these two perspectives, we conceptualize ETTF as a structural condition of the entrepreneurial team that shapes how effectively internal resource bundles are translated into resilience capabilities under crisis. More specifically, our theorizing and findings suggest that ETTF matters because it structures the distribution of task-relevant expertise within the team (Chung et al., 2015; Richard et al., 2019), thereby influencing how members allocate attention across resource domains, recognize bottlenecks, and coordinate compensatory or reconfiguration responses under disruption. In this way, ETTF does not merely characterize team composition; it conditions the process through which entrepreneurial teams convert resource conditions into resilience outcomes. Our findings further suggest a more nuanced moderating pattern: for stability, higher ETTF attenuates differences in severity of loss across resource bundles by strengthening cross-domain compensatory coordination, whereas for flexibility, higher ETTF makes differences in time to recovery across resource bundles more pronounced by enabling more effective reconfiguration. This helps clarify why ventures with similar internal resource conditions may nevertheless exhibit different resilience outcomes under crisis.

Third, our study offers an empirical contribution to resilience research in entrepreneurship by uncovering how different resource configurations generate contrasting resilience outcomes under crisis conditions. Prior studies have largely emphasized the beneficial role of resource availability in supporting resilience (Anwar et al., 2023; Williams et al., 2017), often treating slack resources as uniformly advantageous buffers. Moving beyond this view, we show a clear double-edged effect: new ventures with dual slack exhibit stronger stability but lower flexibility, whereas, among constrained bundles, those combining social slack with technological constraints achieve relatively lower severity of loss and faster recovery. We further show that resilience should not be treated as a single outcome, because the same resource condition may support one dimension while constraining another. By highlighting the trade-offs between stability and flexibility, our findings offer a more nuanced understanding of how new ventures withstand and recover from disruption.

Practical Implications

This study offers practical guidance for new ventures seeking to enhance entrepreneurial resilience under external shocks. More broadly, our findings suggest that entrepreneurial resilience should not be treated as a single, undifferentiated objective. Rather, managers need to recognize that stability and flexibility involve distinct managerial challenges: the former concerns limiting loss severity through buffering and compensation, whereas the latter concerns accelerating recovery through timely reconfiguration. Resource arrangements that help ventures withstand shocks are therefore not always the same as those that help them recover quickly, and entrepreneurs should assess resilience preparedness along both dimensions rather than assuming that more slack will uniformly improve crisis responses.

First, new ventures should optimize existing resource structures rather than indiscriminately pursue additional resources. Managers should periodically assess the balance between social and technological resources and identify whether current bundle configurations create buffering capacity or expose vulnerabilities under crisis. In particular, ventures with dual slack should avoid complacency and maintain reallocation capacity, whereas ventures with social or technological constraints should invest selectively in the weaker domain to improve resilience. In this way, regular resource review and targeted adjustment can strengthen both stability and flexibility.

Second, managers should not treat resource misalignment as uniformly undesirable. Our RSA findings suggest that pronounced misalignment is harmful when the immediate priority is loss containment, but it may also be leveraged when the priority shifts to rapid recovery. As a result, when the managerial objective is to reduce loss severity, firms should avoid substantial imbalances between social and technological resources and maintain a more coherent buffering structure. By contrast, when the objective is to accelerate recovery, managers may mobilize the stronger domain to compensate for the weaker one. The practical challenge, therefore, is not simply to eliminate misalignment, but to judge when it should be reduced and when it can support reconfiguration.

Third, our findings highlight the importance of strengthening resource orchestration through deliberate team design and management. In particular, ETTF can reshape how resource bundle differences translate into loss severity and recovery timing when managed effectively. Ventures should therefore consider how members’ functional backgrounds and tenure combine within the team, rather than evaluating individuals in isolation. By building teams that support differentiated expertise and coordinated integration, new ventures can improve their ability to mobilize, recombine, and deploy resources under crisis conditions.

Limitations and Future Directions

This study has several limitations that offer directions for future research. First, our empirical analysis is based on Chinese new ventures. Although most sample firms are private enterprises operating in competitive, market-oriented environments, the generalizability of our findings may still be limited. Prior research suggests that cultural orientations (e.g. collectivism versus individualism) can shape how entrepreneurs mobilize and redeploy resources (Pinillos & Reyes, 2011), while institutional and regulatory arrangements (e.g. bankruptcy regimes, labor protections, and government intervention) may influence how resources are recombined during crises (Boudreaux et al., 2022). Future research could therefore extend our framework to Western economies and other emerging markets to examine how such contextual differences condition the resilience effects of resource orchestration.

Second, our study focuses primarily on social and technological resources. Yet entrepreneurial resilience may also depend on other underexplored resource types, such as cognitive resources (e.g. managerial attention and learning capacity) or cultural resources (e.g. organizational values). Future research could examine how these alternative forms of slack and constraints influence resilience and interact with social and technological resources.

Third, although stock prices provide a useful market-based lens to capture firm-related information, relying solely on stock market signals may constrain the generalizability of our findings. Future studies could incorporate richer data sources—such as financial accounting indicators, survey data on managerial perceptions, or text-based measures from corporate disclosures—to triangulate resilience more robustly at the organizational level. Importantly, our operationalization of severity of loss means that the current study’s conclusions apply only to ventures that experienced negative shocks. We do not capture cases where firms benefited from crises, for example by seizing market opportunities. Future research could extend our framework by explicitly modeling outcome directionality (loss vs. gain) and examining whether resource bundles differentially predict the trajectories of “losers” and “gainers.”

Finally, our study examines how different bundles of social and technological resources influence entrepreneurial resilience, but it does not provide direct evidence for the mechanisms linking these resource bundles to stability and recovery. Drawing on resource orchestration and slack- and constraint-based perspectives, we propose that ventures with distinct bundles may differ in their ability to redeploy resources quickly, their willingness to take risks, and their creativity in making do under constraints. Future research could therefore use finer-grained or survey-based evidence to directly test how these behavioral and cognitive mechanisms mediate the relationship between resource bundles and resilience outcomes, thereby shedding further light on the micro-foundations through which ventures translate resource endowments into resilience.