Resilient Leadership and SME Performance in Times of Crisis: The Mediating Roles of Temporal Psychological Capital and Innovative Behavior

Abstract

Small and medium-sized enterprises (SMEs) often face severe resource constraints and operational fragility during crises. However, little is known about how managerial resilience (MR) translates into performance through time-related psychological resources and innovation—two capabilities that are both scarce and critical under such conditions. Drawing on Temporal Motivation Theory (TMT), this study develops and tests a dual-mediation model in which employee temporal psychological capital (TPC) and employee innovative behavior (EIB) transmit the effects of MR on performance. As a core methodological innovation, we adopt a multi-method analytical strategy to provide robust and complementary evidence rather than a hierarchy of results: Partial Least Squares Structural Equation Modeling (PLS-SEM) is used to examine sufficiency-based causal pathways and quantify the mediating mechanisms; Support Vector Machine (SVM) classification offers a non-parametric predictive validation of how MR and its mediators distinguish high- and low-performance cases; and Necessary Condition Analysis (NCA) identifies non-compensatory conditions that must be present for high performance to occur. These three methods address different research questions—sufficiency, classification robustness, and necessity—therefore serving as parallel, equally important components of the analysis. A total of 455 SME managers and employees were surveyed, and results show that MR significantly enhances all three dimensions of TPC (temporal control, temporal fit, time pressure resilience) and EIB (idea generation, idea promotion, idea realization), which in turn improve employee performance. SVM classification confirms that high MR, strong TPC, and active innovation align with high performance, while NCA reveals temporal control, idea generation, and idea realization as necessary bottleneck conditions. By integrating sufficiency–necessity logic with predictive classification, our findings suggest that SMEs should prioritize leadership resilience training to strengthen managers’ adaptive capacity, while simultaneously implementing time management interventions—such as temporal control workshops, workload balancing, and innovation pipeline support—to enhance employees’ ability to align tasks with organizational timelines, execute ideas effectively, and sustain performance during crises.

1. Introduction

In an era characterized by recurrent crises—ranging from economic downturns and global pandemics to technological disruptions—organizations worldwide face heightened uncertainty and operational fragility [1,2]. In China, small and medium-sized enterprises (SMEs) represent over 99% of all registered firms, contribute more than 60% of GDP, and provide approximately 80% of employment [3]. Despite their critical role in economic growth and social stability. Chinese SMEs are particularly vulnerable during crises due to structural disadvantages such as limited financial reserves [4], weaker supply chain bargaining power, and constrained access to emergency financing [5]. The COVID-19 pandemic offered a stark illustration of these vulnerabilities [6]. Prolonged lockdowns, disrupted logistics [7], and abrupt shifts in consumer demand caused severe revenue contractions for many SMEs, with surveys showing that more than half could not sustain operations beyond three months without external support. Beyond public health emergencies, Chinese SMEs have also faced compounding challenges from trade tensions [8], raw material price volatility, and tightening environmental regulations. These pressures expose them to simultaneous demand-side and supply-side shocks, amplifying the risks of business interruption, workforce attrition, and innovation stagnation [9].

While prior studies have recognized the pivotal role of managers in steering innovation and maintaining performance during turbulent periods [10], the existing literature exhibits several notable shortcomings. One major limitation lies in the heavy reliance on indirect proxies—such as demographic traits, tenure, or generalized personality factors—to capture managerial influence [11]. Although convenient, these proxies rest on long causal chains and often exhibit weak construct validity, making it difficult to identify the specific psychological mechanisms through which leaders shape employee behaviors. Such over-reliance on proxy variables risks generating superficial or context-insensitive conclusions and constrains the development of targeted managerial interventions that address the underlying drivers of performance. Moreover, much of the resilience research treats managerial traits as static attributes, neglecting their dynamic interaction with contextual demands and time-sensitive pressures [12]. This neglect is problematic in crisis contexts, where the temporal dimension of managerial decision-making—such as the ability to regulate time pressure, synchronize team efforts, and maintain temporal alignment with strategic goals—is often critical to sustaining performance [13]. Studies that have addressed time-related factors tend to examine them in isolation, for example, focusing solely on time urgency or temporal leadership [13], without integrating these into a broader psychological resource framework such as temporal psychological capital (TPC) [14]. As a result, they fail to capture how multiple temporal capabilities operate together to mediate the effects of resilience on outcomes [15,16].

In parallel, research on employee innovation has primarily emphasized structural or organizational antecedents (e.g., R&D investment, leadership support, resource availability) [17,18], while giving less attention to individual-level psychological enablers that determine whether innovative ideas are generated, promoted, and realized under conditions of uncertainty [19]. The few studies that consider psychological mechanisms often focus on general constructs like self-efficacy or intrinsic motivation [20], overlooking temporal factors that influence the pacing, sequencing, and completion of innovation tasks. This omission limits our understanding of how managerial resilience might activate time-based psychological resources that, in turn, drive innovation behavior in crisis situations. Another gap concerns the treatment of managerial resilience (MR), which has been increasingly acknowledged as a valuable strategic resource in crisis contexts [21]. Much of the existing research has focused on its direct effects on firm performance [10,19], paying insufficient attention to the intermediate processes through which resilience is translated into tangible outcomes. This “black box” treatment is problematic because resilience benefits are rarely automatic; rather, they depend on the activation of employees’ cognitive, motivational, and behavioral resources [8]. Few empirical studies have explored how MR interacts with specific, context-relevant psychological capabilities—particularly those related to time—to facilitate sustained performance [18].

A further shortcoming is the limited theorization of time-related psychological mechanisms in resilience research [22]. Constructs such as time urgency, temporal leadership, and time management have appeared sporadically in the leadership and organizational behavior literature [13], yet are often treated as peripheral moderators or control variables rather than central explanatory factors. This neglect is theoretically limiting because in high-pressure environments, temporal misalignment, resource bottlenecks, and compressed deadlines can derail innovation execution, regardless of other strengths. Temporal psychological capital (TPC)—comprising temporal control, temporal fit, and time pressure resilience—offers a theoretically coherent framework for capturing these capabilities, yet has seldom been integrated into resilience–performance models, despite its direct relevance to SMEs operating under crisis-induced time constraints [23].

Time-related capabilities are central to organizational functioning and become decisive for SMEs under crisis conditions. Because SMEs typically operate with lean staffing, limited resource buffers, and little redundancy, managers and employees must multitask and make rapid decisions with minimal time margins [24,25]. In such settings, temporal control—the ability to plan, allocate, and manage time effectively—proves essential for prioritizing competing tasks and deploying scarce resources efficiently. This capability not only helps maintain operational continuity but also supports the sustained flow of innovation.

Crises frequently bring unpredictable disruptions, including sudden demand shifts, supply chain breakdowns, and regulatory changes [26,27]. Here, temporal fit—the alignment between individual work pacing and evolving organizational and environmental demands—plays a crucial role. Without it, SMEs may fail to synchronize internal processes with external market dynamics, resulting in missed opportunities and eroded performance. Moreover, crises are characterized by compressed decision-making cycles and the need for rapid strategic adjustments [28]. Under these conditions, time pressure resilience—the capacity to maintain cognitive clarity, decisional quality, and well-being amid persistent time demands—proves to be vital to the successful implementation of innovative ideas.

Despite the recognized importance of these capabilities, the three dimensions of temporal psychological capital remain underexplored within resilience-performance frameworks. Existing research often treats time-related constructs—such as time urgency, temporal leadership, or time management—as secondary or supportive factors [29,30], rather than central mechanisms that translate managerial resilience into employee performance. This limited perspective constrains theoretical insight into how resilience functions in crisis-affected SMEs and overlooks practical opportunities for leaders to develop specific temporal capacities that foster adaptability and drive innovation.

Against this backdrop, the present study develops and empirically tests a dual-mediation model in which MR influences employee performance via TPC and employee innovative behavior. Drawing on Temporal Motivation Theory [31], we position TPC as a context-specific psychological resource that enables employees to manage temporal demands and align innovation processes with organizational objectives under crisis conditions. Methodologically, we employ an integrated sufficiency–necessity framework—combining Partial Least Squares Structural Equation Modeling (PLS-SEM) with Necessary Condition Analysis (NCA)—to capture both the symmetrical (sufficiency-based) and asymmetrical (necessity-based) relationships among variables. This approach advances prior research by (1) replacing proxy-based inferences with direct psychological measurement, (2) unpacking the “black box” linking MR to performance via temporal cognition and innovation, (3) incorporating TPC into resilience theory as a core mechanism rather than a peripheral factor, and (4) identifying non-compensatory bottlenecks that constrain high performance in SMEs during crises.

Furthermore, drawing from Temporal Motivation Theory (TMT), this study systematically investigates how managerial resilience influences SMEs’ performance [31] in crisis contexts through two mediators: employee temporal psychological capital and employee innovative behavior [14]. TMT highlights firms’ capabilities to integrate and reconfigure internal and external resources in response to environmental volatility, offering a robust theoretical foundation for understanding how SMEs can harness managerial resilience to sustain innovation and performance amid crises [32,33]. Specifically, the research addresses the following questions: RQ1: How does managerial resilience influence SMEs’ performance in crisis situations? RQ2: Do employee innovative behavior and employee temporal psychological capital mediate the relationship between managerial resilience and employee performance? RQ3: How does managerial resilience influence employee temporal psychological capital and employee innovative behavior? Addressing these questions will significantly deepen the scholarly understanding of managerial resilience, offering theoretical insights and practical guidance to enhance SME adaptability and innovative capabilities amid uncertainty.

2. Literature Review

2.1. Theoretical Foundation

Temporal Motivation Theory (TMT) was formally proposed by Canadian psychologist Piers Steel in 2007 [34]. Originally developed to integrate multiple leading theories of motivation, TMT primarily aimed at explaining individuals’ decision-making behaviors, particularly when facing tasks involving temporal delays and urgency [35]. Initially, TMT was extensively utilized to interpret procrastination behaviors and time-inconsistent decision-making patterns [36]. Its application has expanded into organizational behavior research, notably in examining how individuals and managers balance short-term pressures against long-term organizational goals.

In crisis contexts, urgency, uncertainty, and compressed timelines become commonplace, intensifying the relevance of TMT. According to TMT, individuals’ motivation is not influenced merely by task value or expectancy but profoundly shaped by their perceptions and management of time under pressure [36]. This study integrates TMT with the concept of “temporal psychological capital” (TPC), defined as positive psychological resources related to employees’ and managers’ temporal cognition, such as temporal control perception, temporal fit, and time pressure resilience. From a TMT perspective, managers with higher levels of temporal psychological capital are better equipped to modulate their motivation levels during crises [37,38], mitigating short-sighted decisions induced by temporal discounting effects. Temporal Motivation Theory has been widely empirically validated and applied in the fields of psychology and organizational behavior abroad [33,34].

2.2. Hypothesis Development

2.2.1. Managerial Resilience

Managerial resilience can be conceptualized as a dynamic capability that enhances an organization’s ability to sense environmental changes, respond rapidly, and reconfigure internal and external resources [39]. Under conditions of high uncertainty, managerial resilience is widely acknowledged as a critical determinant of organizational sustainability and competitive advantage. However, this capability extends beyond strategic intent at the top management level and is ultimately realized through the concrete behaviors of employees [40]. According to Bandura and Wessels (1997), individuals’ self-beliefs are profoundly shaped by their environmental context [41]. Resilient managers establish psychological safety, provide stable temporal structures, and effectively support employees during crises, thereby enhancing employees’ perceived control over time-related tasks. Additionally, Kristof [42] suggests that resilient management promotes shared organizational values and consistent working rhythms, reducing mismatches between individual pacing and organizational expectations. This alignment, referred to as temporal fit, enables employees to synchronize harmoniously with organizational schedules, deadlines, and cycles [43], thereby boosting their confidence and sustained engagement. Base on this, we propose the following hypotheses:

H1a. 

Managerial resilience positively influences employee temporal psychological capital.

H1b. 

Managerial resilience positively influences employee innovative behavior.

2.2.2. Employee Temporal Psychological Capital

Employee temporal psychological capital, conceptualized as a multidimensional construct, refers to employees’ positive psychological resources mobilized to effectively cope with time-related demands in organizational contexts [44]. TPC generally consists of three core components: temporal control perception, temporal fit, and time pressure resilience [45]. Temporal control perception represents employees’ beliefs in their capability to allocate, prioritize, and manage time effectively. Temporal fit refers to the alignment between employees’ individual temporal orientation and the organization’s temporal structures, facilitating better synchronization with organizational rhythms, deadlines, and strategic priorities [46,47]. Time pressure resilience refers to an individual’s ability to maintain focus, adaptability, and emotional stability under time-related stress. This construct is theoretically grounded in the Conservation of Resources (COR) theory, which underscores the importance of preserving psychological resources in high-pressure contexts. Higher temporal control perception facilitates enhanced temporal fit by enabling employees to better align their personal scheduling [48] preferences with organizational demands. Moreover, the development and activation of temporal psychological capital (TPC) are strongly shaped by organizational and managerial contexts. Resilient managers foster psychologically safe environments through transparent communication, adaptive time management practices, and supportive leadership, thereby strengthening employees’ perceptions of temporal control [49]. Moreover, resilient managers contribute to establishing resilient organizations, characterized by consistent planning and participatory decision-making, thereby cultivating an organizational climate that is conducive to collective temporal alignment and reducing time-based stressors [50]. Thus, we propose the following hypotheses:

H2a. 

Managerial resilience positively influences temporal control perception.

H2b. 

Managerial resilience positively influences employee temporal fit.

H2c. 

Managerial resilience positively influences employee time pressure resilience.

In SMEs facing crisis conditions, employee innovative behavior (EIB) is a crucial mechanism for transforming managerial resilience into sustainable performance. Resilience-oriented leadership fosters a climate of psychological safety, where employees are more willing to take interpersonal risks, challenge existing processes, and explore unconventional solutions [51,52]. This environment is particularly valuable for SMEs during crises, as operational disruptions and market instability demand rapid adaptation and creative problem-solving [53,54]. Managerial support and resilience can enhance employees’ capacity to navigate the uncertainty that is inherent in crisis contexts, thereby increasing their likelihood of engaging in behaviors that generate, promote, and implement innovative ideas [55,56]. In SMEs, where decision-making is often informal and resource allocation is highly constrained, managerial resilience enables faster consensus-building and resource mobilization, which are essential for moving innovations from concept to execution [57]. Resilient leaders influence innovation outcomes by maintaining team focus under pressure, reconfiguring limited resources, and sustaining motivation during implementation phases [58]. The ability of managers to remain composed, solution-focused, and adaptive under crisis conditions not only protects ongoing operations but also signals to employees that innovative efforts will be supported even in the face of uncertainty. Managerial resilience in SMEs under crisis conditions acts as both a psychological catalyst—enhancing employees’ confidence to innovate—and a structural enabler—removing organizational barriers that could stall innovation processes. Thus, the following hypotheses are proposed:

H3a. 

Managerial resilience positively influences employees’ idea generation.

H3b. 

Managerial resilience positively influences employees’ idea promotion.

H3c. 

Managerial resilience positively influences employees’ idea realization.

2.2.3. Employee Temporal Psychological Capital and Employee Performance

In highly uncertain and time-sensitive organizational contexts, particularly in SMEs during crises, employee performance depends not only on skills and motivation but also on the ability to perceive, manage, and adapt to temporal demands [33]. Temporal psychological capital (TPC), encompassing temporal control, temporal fit, and time pressure resilience, provides employees with the cognitive and emotional resources necessary to sustain performance under volatile conditions. From a person–environment fit perspective, temporal fit—the alignment between employees’ personal work rhythms and organizational temporal structures—facilitates smoother workflow integration and more effective resource coordination [29,59]. Empirical studies have shown that such alignment improves efficiency, enhances collaborative performance, and reduces temporal conflicts within teams [34]. This is particularly valuable for SMEs in crisis environments, where misaligned timing between individual and organizational pacing can exacerbate operational bottlenecks and delay critical responses [53]. From a Conservation of Resources (COR) standpoint, time pressure resilience enables employees to maintain composure, resist burnout, and sustain cognitive functioning under compressed deadlines [60,61]. Employees with higher resilience to time pressure exhibit better adaptability and responsiveness, particularly in high-intensity work settings [35]. In SMEs operating under continuous uncertainty, such resilience directly supports consistent performance, as it allows employees to adjust priorities rapidly and execute tasks without sacrificing quality.

Moreover, temporal control—the ability to plan, prioritize, and allocate time effectively—has been positively linked to both task performance and discretionary behaviors that create organizational value [46,62]. In SMEs, effective temporal control helps optimize scarce resources, minimize task delays, and create time buffers for innovation activities, all of which are critical for a sustaining competitive advantage in turbulent markets. TPC operates as a core psychological capability that enables employees to align with organizational pacing, sustain effectiveness under temporal strain, and contribute to adaptive and innovative performance outcomes. In SME crisis contexts, where operational disruptions, resource constraints, and time compression are common, TPC serves as both a protective and enabling factor—reducing performance losses from temporal misalignment and facilitating rapid, coordinated responses to emerging challenges. Thus, the following hypotheses are proposed:

H4a. 

Temporal control perception positively influences employee performance.

H4b. 

Temporal fit positively influences employee performance.

H4c. 

Time pressure resilience positively influences employee performance.

2.2.4. Employee Innovative Behavior and Employee Performance

Employee innovative behavior (EIB), encompassing the multidimensional process of idea generation, promotion, and realization, has been widely recognized as a critical antecedent of individual and organizational performance [63,64]. Idea generation involves employees proposing original, useful, and creative solutions related to products, processes, or work methods [65]. Employees frequently engaged in idea generation are more likely to proactively identify inefficiencies, propose improvements, and continually contribute to organizational innovation, thereby enhancing task efficiency and problem-solving effectiveness [66]. By proactively promoting innovative ideas [67], employees not only increase their own organizational visibility and influence but also improve both in-role and extra-role performance [68]. Idea realization, the final stage in innovative behavior, involves transforming generated and promoted ideas into practical outcomes. This stage has the most direct connection to performance because idea realization effectively translates innovative concepts into visible achievements [69]. Employees proficient in idea realization typically exhibit high levels of initiative, autonomy, and task execution capability, all of which are essential to achieving superior performance. Hammond et al. [70] further confirm that innovative behaviors affect performance not merely in terms of quantity, but also by significantly enhancing quality, timeliness, and adaptability. Thus, the following hypotheses are proposed:

H5a. 

Idea generation positively influences employee performance.

H5b. 

Idea promotion positively influences employee performance.

H5c. 

Idea realization positively influences employee performance.

2.2.5. Employee Performance

Employee performance (EP) has long been a central research theme within organizational behavior and human resource management [71], typically defined as employees’ task efficiency, output quality, and behaviors exhibited to achieve organizational objectives [72]. As organizational contexts become increasingly complex and dynamic, researchers recognize that employee performance is influenced not only by individual knowledge, skills, and motivation but also by psychological resources, innovative behaviors, leadership characteristics, and organizational contexts [73,74]. Piening, Baluch and Salge (2013) argues that higher temporal control perception strengthens employees’ confidence in task execution [48]. Kristof (1996) suggests that temporal fit reduces cognitive conflict and improves work efficiency [42], while Hobfoll (2011) highlights that time pressure resilience enables employees to maintain psychological stability and performance under high-pressure conditions [75]. Empirical research by Aeon and Aguinis (2017) further confirms that effective time management, temporal alignment [76], and adaptive stress-coping strategies positively correlate with employee performance outcomes.

Moreover, employee innovative behavior (EIB) is widely regarded as a crucial behavioral mechanism enhancing employee performance. Idea generation enables the identification of problems and the proposal of improvements. Idea promotion fosters organizational support and resources, strengthening cross-departmental collaboration and knowledge sharing [55]. Idea realization directly translates innovative ideas into tangible outcomes, thereby enhancing task performance, adaptive performance, and overall output quality [66]. Thus, this study proposes the following conceptual model(See Figure 1).

3. Methodology

3.1. Research Design

This study employs a quantitative research design, utilizing structured questionnaires as the primary instrument for data collection. All measurement items were adapted from previously validated scales and carefully refined to align with the conceptual definitions of the core constructs, thereby ensuring content validity and contextual appropriateness within the Chinese business context. The Idea Implementation items were adapted from the Innovative Work Behavior (IWB) scale developed by De Jong and Den Hartog [69], with supplementary references to Scott and Bruce (1994) [63]. Items measuring managerial resilience were primarily drawn from the validated frameworks of Lengnick-Hall et al. (2011) [77] and Kantur and Iseri-Say (2012) [78], and were further refined to reflect the resilience capacity models proposed by Lee and Zhao [79,80]. To capture temporal control perception, items were adapted from the Mastery Scale by Pearlin and Schooler (1978) [81] and the Sense of Control Scale developed by Lachman and Weaver (1998) [82], both of which are widely recognized in the literature on perceived control. The temporal fit dimension was derived from the Future Time Perspective subscale of the Zimbardo Time Perspective Inventory [83], along with items adapted from the Consideration of Future Consequences Scale by Strathman et al. (1994) [84]. Measurement of time pressure resilience was based on the Connor–Davidson Resilience Scale by Connor & Davidson, (2003) [85], a widely used instrument for assessing psychological resilience in high-pressure environments.

All items were measured on a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree), designed to comprehensively capture participants’ attitudes, perceptions, and behavioral tendencies (see Appendix A).

3.2. Data Collection

Data were collected through an online survey platform (www.wjx.com), targeting employees and their immediate supervisors from 11 representative small and medium-sized enterprises (SMEs) across China. These SMEs encompassed a range of industries—including manufacturing, services, and technology-driven innovation—ensuring sectoral diversity and enhancing the generalizability of the findings. Prior to the formal data collection, a pilot study was conducted to evaluate the clarity of the questionnaire [86], the relevance of individual items, and the accuracy of translations. Based on participants’ feedback, the wording and structure of the questionnaire were carefully revised to improve comprehensibility and contextual fit.

Data analysis was performed using Smart-PLS 4.0 and R software (version 4.4.2) [87]. PLS-SEM was chosen due to its suitability for exploratory research, particularly when sample sizes are relatively small, models are complex, and path analyses involve mediation and moderation effects. Additionally, R software was employed for Support Vector Machine (SVM) boundary analyses [88]. PLS-SEM examined the direct effects of managerial resilience on intermediary variables, including the three dimensions of temporal psychological capital (temporal control perception, temporal fit, time pressure resilience) and employee innovative behavior (idea generation, idea promotion, idea realization), as well as the subsequent impact of these mediating variables on employee performance.

3.3. Sample Description

To enhance the reliability and validity of the research instruments, a multi-wave survey design was employed. In the first phase, a pilot study was conducted in March 2025, during which a small sample of 37 valid responses was collected. The primary aim of the pilot study was to assess the clarity, reliability, and contextual relevance of the questionnaire items. Based on the feedback and statistical diagnostics obtained from the pilot data, several items were revised or refined. Following this preliminary phase, the formal data collection commenced in May 2025 using the revised questionnaire, targeting a broader sample population for the main study. A total of 462 questionnaires were collected. After removing invalid responses, 455 valid questionnaires remained, resulting in an effective response rate of 98% (see

Table 1. Demographic profiles of respondents.

Variable	Category	Frequency	Percentage (%)	Cumulative Percentage (%)
Gender	Male	169	37.1	37.1
	Female	286	62.9	100
Age	18–25	34	7.5	7.5
	26–40	270	59.3	66.8
	41–55	139	30.5	97.4
	56+	12	2.6	100
Education Level	High School and below	99	21.8	21.8
	College	106	23.3	45.1
	Bachelor’s degree	232	51	96
	Graduate and above	18	4	100
Job Level	Entry-level employee	302	66.4	66.4
	Middle management	106	23.3	89.7
	Senior management	25	5.5	95.2
	Owner/Executive level	22	4.8	100
Years of Experience	Less than 1 year	55	12.1	12.1
	1–3 years	99	21.8	33.8
	4–6 years	113	24.8	58.7
	7–10 years	63	13.8	72.5
	10 years or above	125	27.5	100
Industry Type	Manufacturing	31	6.8	6.8
	Service industry	222	48.8	55.6
	High-tech industry	5	1.1	56.7
	Education	7	1.5	58.2
	Government sector	190	41.8	100

).

3.4. Common Method Bias Analysis

To assess the potential threat of common method bias (CMB), we conducted Harman’s single-factor test using principal component analysis (PCA) without rotation. As shown in

Table 2. Total variance explained.

Factor	Initial Eigenvalues			Extracting the Sum of Squared Loadings
	Total	Percentage of Variance	Cumulative%	Total	Percentage of Variance	Cumulative%
MR	2.09	26.129	26.129	1.827	22.838	22.838
TC	1.75	21.88	48.009	1.297	16.217	39.055
TF	1.13	14.131	62.14	0.51	6.375	45.431
TP	1.024	12.797	74.936	0.407	5.092	50.523
IG	0.714	8.923	83.86
IP	0.681	8.508	92.368
IR	0.456	5.705	98.073
EP	0.154	1.927	100

, eight factors exhibited eigenvalues greater than 1, jointly explaining 100% of the total variance. The first factor accounted for 26.13% of the variance, which is well below the threshold of 40% commonly recommended in the literature (Podsakoff et al., 2003 [89]). Furthermore, the cumulative variance explained by the first two factors (48.01%) and the first three factors (62.14%) indicates that no single factor dominated the total variance. The extraction sums of squared loadings also show that the variance explained by the first factor (22.83%) after factor extraction remained below the critical value.

4. Result

4.1. Descriptive Statistical Results and Correlation Analysis

Table 3. Descriptive statistics of main variables.

Variable	Mean	SD	Min	Max
MR	4.12	0.58	2.3	5
TC	3.87	0.63	2.1	5
TF	4.03	0.55	2.7	5
TP	4.01	0.61	2.5	5
IG	3.94	0.6	2	5
IP	3.81	0.66	1.9	5
IR	3.89	0.64	2.1	5
EP	4.07	0.57	2.8	5

Managerial resilience (MR), temporal psychological capital (TPC), employee innovative behavior (EIB), employee performance (EP), temporal control (TC), temporal fit (TF), idea generation (IG), idea realization (IR), idea promotion (IP).

summarizes the descriptive statistics of the core constructs examined in this study, namely managerial resilience (MR), temporal psychological capital (TPC: TC, TF, TP), employee innovative behavior (EIB: IG, IP, IR), and employee performance (EP).

All variables have mean values above the mid-point of the 5-point scale, indicating generally high levels of MR, TPC, EIB, and EP among respondents. Within TP, TF (M = 4.03) is highest, suggesting good alignment between employees’ work pacing and organizational demands, while TC (M = 3.87) is relatively lower, pointing to room for improvement in planning and time allocation. In EIB, IG (M = 3.94) and IR (M = 3.89) are higher than IP (M = 3.81), implying that securing stakeholder support may be a bottleneck in SMEs during crises. EP (M = 4.07) is relatively high, though variation exists, reflecting differences in resilience, temporal management, and innovation execution. Practically, enhancing temporal control and strengthening idea promotion could be key for SMEs to sustain high performance under volatile conditions.

To examine the interrelationships among the core constructs in this study—managerial resilience (MR), temporal psychological capital (TPC: TC, TF, TP), employee innovative behavior (EIB: IG, IP, IR), and employee performance (EP)—a Pearson correlation analysis and hierarchical clustering analysis were conducted. Figure 2 presents the correlation heatmap and the dendrogram-based clustered heatmap of these variables.

The results reveal several noteworthy patterns. First, strong positive correlations were observed among the three dimensions of employee innovative behavior, with coefficients exceeding 0.70, indicating that idea generation (IG), idea promotion (IP), and idea realization (IR) are highly interrelated and often co-occur within organizational settings. Similarly, the three dimensions of temporal psychological capital are moderately to strongly correlated (r = 0.58–0.63), suggesting that employees who perceive higher control over time tend to experience better temporal alignment with organizational rhythms and greater resilience under time pressure. Managerial resilience demonstrated robust positive correlations with all mediating variables (r = 0.60–0.82), highlighting its pivotal role in fostering both time-based psychological resources and innovative behaviors among employees. These results are consistent with the theoretical assumption that resilient managers create supportive and adaptive environments, thereby enhancing employees’ cognitive and behavioral capabilities under crisis conditions.

The hierarchical clustering analysis further confirmed these findings. Two primary clusters emerged: (1) TPC dimensions (TC, TF, TP) grouped closely with MR, forming a psychological resource cluster; and (2) EIB dimensions (IG, IP, IR) clustered together with EP, reflecting the behavioral pathway leading to performance outcomes. This clustering structure provides empirical support for the dual mediation model proposed in this study, in which managerial resilience first influences employees’ temporal psychological states, subsequently enabling innovative actions that directly enhance performance. The correlation and clustering analyses validate the hypothesized relationships among the constructs, demonstrating clear linkages between managerial resilience, employee-level psychological resources, innovation behaviors, and ultimate performance outcomes.

4.2. PLS-SEM Result

4.2.1. Assessing the Outer Measurement Model

The reliability and validity of the measurement model were evaluated through key psychometric indicators [90], assessing the internal consistency and convergent validity of each construct [91]. As shown in

Table 4. Reliability and convergent validity.

Construct	Item	Loading	Cronbach’s Alpha	rho_a	CR	AVE
EP	EP1	0.93	0.964	0.964	0.972	0.874
	EP2	0.93
	EP3	0.939
	EP4	0.924
	EP5	0.952
IG	IG1	0.922	0.962	0.963	0.971	0.869
	IG2	0.941
	IG3	0.945
	IG4	0.905
	IG5	0.948
IP	IP1	0.924	0.965	0.966	0.973	0.879
	IP2	0.936
	IP3	0.941
	IP4	0.954
	IP5	0.931
IR	IR1	0.95	0.976	0.976	0.981	0.913
	IR2	0.959
	IR3	0.953
	IR4	0.96
	IR5	0.956
MR	MR1	0.906	0.953	0.954	0.964	0.843
	MR2	0.922
	MR3	0.917
	MR4	0.938
	MR5	0.907
TC	TC1	0.903	0.938	0.94	0.953	0.802
	TC2	0.928
	TC3	0.885
	TC4	0.85
	TC5	0.908
TF	TF1	0.905	0.956	0.957	0.966	0.852
	TF2	0.917
	TF3	0.927
	TF4	0.943
	TF5	0.922
TP	TP1	0.921	0.965	0.965	0.973	0.877
	TP2	0.934
	TP3	0.945
	TP4	0.936
	TP5	0.946

, the results of the reliability and convergent validity analyses for each latent variable (EP, IG, IP, IR, MR, TC, TF, and TP) are presented. First, regarding reliability, Cronbach’s α coefficients ranged from 0.93 (TC) to 0.97 (EP), all surpassing the commonly accepted threshold of 0.7 [92,93], indicating good internal consistency for each construct. Similarly, CR values also exceeding the recommended threshold of 0.7, further confirming the reliability of the measurement scales. In terms of convergent validity, average variance extracted (AVE) values were all above the benchmark of 0.5, indicating that the latent constructs adequately explained the variance of their respective measurement items.

The path coefficient in Figure 3 illustrates how managerial resilience (MR) influences employee performance (EP) through the dual mediating mechanisms of employee temporal psychological capital (TPC) and employee innovative behavior. MR significantly and positively predicts all three dimensions of TPC (TC: β = 0.872; TF: β = 0.770; TP: β = 0.810) as well as all three dimensions of EIB (IG: β = 0.826; IP: β = 0.796; IR: β = 0.815), indicating that resilient managers can enhance employees’ time management capacity and innovation execution. Further analysis shows that TC (β = 0.278), TF (β = 0.005), TP (β = 0.112), IG (β = 0.161), IP (β = 0.300), and IR (β = 0.259) all significantly and positively impact EP. The model demonstrates good explanatory power (EP R² = 0.591, Q² = 0.678), providing empirical support for understanding performance enhancement mechanisms under crisis conditions.

Discriminant validity was further assessed using the Fornell–Larcker criterion. As shown in

Table 5. Forel–Larcker criterion.

	EP	IG	IP	IR	MR	TC	TF	TP
EP	0.935
IG	0.832	0.932
IP	0.853	0.891	0.937
IR	0.853	0.894	0.916	0.956
MR	0.753	0.826	0.796	0.815	0.918
TC	0.811	0.817	0.795	0.804	0.812	0.895
TF	0.797	0.805	0.827	0.855	0.77	0.858	0.923
TP	0.77	0.872	0.835	0.843	0.81	0.819	0.841	0.936

, the model demonstrates satisfactory overall discriminant validity. According to the criterion, the square root of the average variance extracted (AVE) for each construct should be greater than its correlations with other constructs. Similarly, the square root of the AVE for MR is 0.811, which is also higher than its correlations with any other construct. All constructs have AVE square roots greater than their inter-construct correlations [94], confirming that the model possesses good discriminant validity.

This study employed the Heterotrait–Monotrait Ratio of Correlations (HTMT) method and constructed a heatmap matrix.

Table 6. HTMT matrix heatmap.

	EP	IG	IP	IR	MR	TC	TF	TP
EP
IG	0.864
IP	0.883	0.824
IR	0.879	0.822	0.843
MR	0.785	0.862	0.83	0.844
TC	0.852	0.859	0.835	0.84	0.858
TF	0.83	0.838	0.861	0.885	0.806	0.815
TP	0.798	0.805	0.865	0.868	0.844	0.86	0.875

presents the HTMT values among employee performance (EP), the three dimensions of employee innovative behavior (IG, IP, IR), managerial resilience (MR), and the three dimensions of temporal psychological capital (TC, TF, TP). Most HTMT coefficients range between 0.80 and 0.88. Although some values slightly exceed the conventional threshold of 0.85, they remain below the more lenient cutoff of 0.90 (Henseler et al., 2015 [95]), indicating that the model constructs exhibit acceptable discriminant validity.

4.2.2. Structural Path Analysis

The results of the structural model reveal several key relationships among the variables in

Table 7. Confidence intervals—one-tailed/two-tailed.

	Original Sample (O)	Sample Mean (M)	Confidence Intervals		Confidence Intervals
			One-Tailed		Two-Tailed
			2.50%	97.50%	2.50%	97.50%
IG- > EP	0.161	0.168	−0.016	0.349	−0.03	0.336
IP- > EP	0.3	0.303	0.101	0.491	0.091	0.483
IR- > EP	0.259	0.246	0.029	0.458	0.043	0.475
MR- > IG	0.826	0.827	0.788	0.862	0.785	0.86
MR- > IP	0.796	0.797	0.751	0.838	0.747	0.835
MR- > IR	0.815	0.816	0.773	0.852	0.77	0.849
MR- > TC	0.812	0.813	0.763	0.859	0.759	0.855
MR- > TF	0.77	0.771	0.712	0.823	0.706	0.82
MR- > TP	0.81	0.811	0.762	0.856	0.757	0.853
TC- > EP	0.278	0.277	0.154	0.4	0.158	0.401
TF- > EP	0.055	0.061	−0.089	0.219	−0.097	0.205
TP- > EP	−0.112	−0.114	−0.252	0.014	−0.256	0.012

. Managerial resilience (MR) demonstrates strong, statistically significant positive effects on all three dimensions of employee innovative behavior (EIB)—idea generation (IG: β = 0.826, 95% CI [0.785, 0.860]), idea promotion (IP: β = 0.796, CI [0.747, 0.835]), and idea realization (IR: β = 0.815, CI [0.770, 0.849]). MR also exerts significant positive influences on all three facets of temporal psychological capital (TPC), including temporal control (TC: β = 0.812, CI [0.759, 0.855]), temporal fit (TF: β = 0.770, CI [0.706, 0.820]), and time pressure resilience (TP: β = 0.810, CI [0.757, 0.853]). These findings empirically support the hypothesized role of MR in enhancing employees’ psychological and behavioral capacities under crisis conditions.

With regard to performance outcomes, idea promotion (IP) (β = 0.300, CI [0.091, 0.483]) and idea realization (IR) (β = 0.259, CI [0.043, 0.475]) are significant predictors of employee performance (EP), while idea generation (IG) shows a weaker, marginally significant effect (β = 0.161, CI [−0.030, 0.336]). These results suggest that the later stages of innovation (i.e., promotion and realization) may play a more instrumental role in translating creativity into measurable performance gains.

Among the temporal psychological resources, only temporal control (TC) is positively and significantly associated with EP (β = 0.278, CI [0.158, 0.401]), underscoring the importance of perceived control over time-related tasks. In contrast, temporal fit (TF) shows a weak and non-significant relationship (β = 0.055, CI [−0.097, 0.205]), and surprisingly, time pressure resilience (TP) has a negative association with EP (β = −0.112, CI [−0.256, 0.012]). The inverse relationship between TP and EP may reflect the psychological cost of sustained pressure coping, which, while beneficial in the short term, might lead to fatigue or disengagement over time.

Collectively, these results support the proposed dual-mediation model, wherein MR indirectly enhances employee performance through its positive influence on both TPC and EIB. The model highlights the nuanced roles of different temporal and innovative capacities in the performance process and suggests that not all dimensions contribute equally to outcome effectiveness.

4.2.3. Inner Structural Model

The variance inflation factors (VIFs) were calculated for all predictors in the structural model. As indicated in

Table 8. Assessment of structural model.

	β	Std	T Values	F-Square	VIF	p Values	Result
IG- > EP	0.161	0.093	1.726	0.016	1.984	**	Support
IP- > EP	0.3	0.099	3.018	0.057	3.223	**	Support
IR- > EP	0.259	0.112	2.305	0.037	3.154	**	Support
MR- > IG	0.826	0.019	43.928	2.155	1	***	Support
MR- > IP	0.796	0.022	35.874	1.734	1	***	Support
MR- > IR	0.815	0.02	40.651	1.973	1	***	Support
MR- > TC	0.812	0.024	33.833	1.932	1	***	Support
MR- > TF	0.77	0.028	27.278	1.455	1	***	Support
MR- > TP	0.81	0.024	33.459	1.911	1	***	Support
TC- > EP	0.278	0.061	4.523	0.08	1.798	***	Support
TF- > EP	0.055	0.079	0.694	0.02	2.168	**	Support
TP- > EP	−0.112	0.068	1.661	0.01	2.48	0.097	Not Support

Note: p < 0.001 ***, p < 0.01 **.

, the VIF values ranged from 1.798 to 3.223, remaining below the commonly accepted threshold of 5, thereby confirming that collinearity does not pose a problem. Additionally, a bootstrapping procedure involving 5000 samples was carried out to assess the significance of the proposed hypotheses [96].

The structural model assessment results in Table 5 indicate the strength, significance, and effect sizes of the hypothesized relationships among constructs. Managerial resilience (MR) significantly and positively predicts all dimensions of employee innovative behavior (EIB)—idea generation (IG: β = 0.826, p < 0.001), idea promotion (IP: β = 0.796, p < 0.001), and idea realization (IR: β = 0.815, p < 0.001)—as well as the three dimensions of temporal psychological capital (TPC)—temporal control (TC: β = 0.812, p < 0.001), temporal fit (TF: β = 0.770, p < 0.001), and time pressure resilience (TP: β = 0.810, p < 0.001). All paths show strong effect sizes (f² ranging from 1.455 to 2.155) and low multicollinearity (VIF = 1), supporting H1a–H1f and H2a–H2c.

Regarding the influence on employee performance (EP), IG (β = 0.161, p < 0.01), IP (β = 0.300, p < 0.01), IR (β = 0.259, p < 0.01), and TC (β = 0.278, p < 0.001) significantly enhance performance, with moderate to small effect sizes (f² = 0.016–0.080). TF also shows a positive, albeit weak, relationship (β = 0.055, p < 0.01), indicating marginal support for its role. However, TP unexpectedly exhibits a negative but non-significant effect (β = −0.112, p = 0.097), suggesting that under high time pressure, resilience alone may not translate into enhanced performance outcomes. These findings collectively affirm the multi-pathway influence of MR through both psychological and behavioral mechanisms, while highlighting TP as a potential bottleneck or boundary condition in crisis contexts.

4.3. SVM Decision Boundary Analysis

To further explore the relationship between managerial resilience (MR) and the mediating variables, this study employed the Support Vector Machine (SVM) method to analyze the classification boundaries between MR and the three dimensions of temporal psychological capital (temporal control perception, temporal fit, and time pressure resilience), as well as the three dimensions of employee innovative behavior (idea generation, idea promotion, and idea realization). The Support Vector Machine (SVM) is a supervised machine learning algorithm that is widely used for classification tasks [97]. Its primary goal is to identify the optimal decision boundary—also called the hyperplane—that best separates observations into distinct categories based on their features. The algorithm determines this boundary by maximizing the margin between the closest data points of each class, known as support vectors [98]. This maximization enhances the model’s generalization ability, ensuring robust performance on unseen data. In the context of this study, the SVM was employed to complement the Partial Least Squares Structural Equation Modeling (PLS-SEM) results by providing a visual and non-parametric classification perspective. Whereas PLS-SEM tests hypothesized relationships and estimates average effects, the SVM enables us to observe how well the predictor variable—managerial resilience (MR)—can discriminate between high and low levels of its outcome variables (e.g., temporal psychological capital and employee innovative behavior dimensions). This classification approach is particularly valuable in high-dimensional settings and when relationships may be nonlinear.

We were constructed six Support Vector Machine (SVM) classification models with MR as the predictor and each of the following variables as classification targets in Figure 4: employee innovation—IG, IP, IR; and temporal psychological capital dimensions—TC, TF, TP. All models utilized a linear kernel with a regularization parameter C = 1. The resulting decision boundary visualizations indicate that MR possesses strong classification capabilities across all six outcome variables. Notably, MR exhibits the clearest margin in distinguishing high and low levels of IG and IP, suggesting its substantial influence on employees’ innovative thinking and proactive advancement. Similarly, MR shows moderate separability for IR and TC, implying its potential role in shaping realization behaviors and confidence under time pressure. The boundaries for TF and TP, while slightly less distinct, still demonstrate meaningful classification patterns. These findings visually reinforce the structural model results, affirming MR as a pivotal antecedent of psychological and innovation-related variables. Overall, SVM boundary analysis provides a robust complementary validation of the model’s theoretical assumptions through nonlinear pattern recognition.

4.4. Necessary Condition Analysis

Necessary Condition Analysis (NCA) provides a unique approach to examining complex causal relationships by pinpointing the indispensable conditions that impact outcome variables [99]. Diverging from traditional methods, NCA not only identifies these critical conditions but also measures their scope and limitations. This method is particularly valuable for uncovering “indispensable yet insufficient” links between dependent and independent variables [100,101]. Complementing standard sufficiency-based evaluations, NCA quantitatively assesses necessary conditions needed to attain particular outcome levels [102], providing insights into their influence and revealing potential constraints. The initial step in Necessary Condition Analysis (NCA) consists of plotting a ceiling line that intersects the highest data points on an x-y graph, effectively defining the limits of necessary conditions. This methodology is visually represented in Figure 5, showcasing scatter plots for all pertinent relationships.

The CR-FDH method is applied to estimate ceiling lines and assess the necessity effects of the mediating constructs involved in achieving high levels of employee performance (EP).

Table 9. Necessary Condition Analysis ceiling line (Method: CR-FDH).

	Effect Size	Obs. Above Ceiling	Slope	Condition Inefficiency	Outcome Inefficiency	Rel. Inefficiency	Abs. Inefficiency	p-Value
EP
IG	0.065	3	0.632	52.91	72.406	87.006	15.524	**
IP	0.22	7	1.866	50.039	11.844	55.957	10.175	***
IR	0.196	7	1.286	42.593	31.543	60.701	10.825	***
TC	0.277	2	1.022	35.514	14.155	44.642	11.177	***
TF	0.055	1	0.588	58.959	73.227	89.012	18.973	0.05
TP	0.102	4	0.406	33.715	69.123	79.533	17.556	0.09
IG
MR	0.218	4	0.913	37.906	29.818	56.421	11.562	***
IP
MR	0.116	1	1.234	60.702	41.123	76.863	16.052	***
IR
MR	0.165	3	1.285	54.432	27.492	66.959	13.714	***
TC
MR	0.073	2	0.636	48.839	71.299	85.316	24.533	***
TF
MR	0.06	1	0.693	58.995	70.541	87.92	21.523	***
TP
MR	0.054	1	0.643	58.995	73.639	89.191	22.611	***

Note: p < 0.001 ***, p < 0.01 **.

provides a comprehensive summary of each variable’s effect size, slope coefficient, condition inefficiency, outcome inefficiency, relative inefficiency, absolute inefficiency, and associated p-values.

The necessity effects of key mediating variables in achieving high levels of employee performance (EP) were evaluated using the Ceiling-Regression Free Disposal Hull (CR-FDH) method. As presented in Table 8, results demonstrate that several constructs play a critical necessary role in determining EP, with varying degrees of explanatory strength and inefficiency. Among the mediators predicting EP, IP exhibits the highest effect size (0.22, p < 0.001), followed closely by IR (0.196, p < 0.001) and TC (0.277, p < 0.001). These variables show low condition inefficiencies and high explanatory power, suggesting that the absence of high IP, IR, or TC levels may severely constrain the achievement of superior EP. IG, although statistically significant (p < 0.01), shows a relatively small effect size (0.065) and higher inefficiency metrics, indicating that it is less critical as a bottleneck condition. Conversely, TF and TP exhibit weaker necessity effects with marginal significance (p = 0.05 and p = 0.09, respectively), reflecting their secondary or complementary roles in shaping EP.

In addition, MR (managerial resilience) shows consistent necessity across all intermediate outcomes. It is particularly essential in driving IP (effect size = 0.116, p < 0.001), IR (0.165, p < 0.001), and IG (0.218, p < 0.001), confirming its foundational role in supporting innovation and adaptive behaviors. Its influence on temporal psychological capital dimensions (TC, TF, and TP) also reached statistical significance, though the effect sizes were modest, suggesting a baseline requirement for managerial resilience to enable effective time-related psychological capabilities. Overall, the CR-FDH analysis provides robust evidence that IP, IR, and TC serve as necessary conditions for high employee performance, while MR functions as a cross-cutting, enabling factor for both innovative behaviors and time-related psychological resources. These findings underscore the asymmetrical, non-compensatory nature of certain organizational factors, which should be prioritized in practice-oriented interventions aiming at performance optimization.

To identify critical bottlenecks in achieving employee performance (EP), this study employed the CR-FDH (

Table 10. Bottleneck table (Method: CR-FDH).

EP	IG	IP	IR	TC	TF	TP	MR
0.00%	NN	NN	NN	NN	NN	NN	NN
10.00%	NN	NN	NN	NN	NN	NN	NN
20.00%	NN	−2.604	NN	−4.143	NN	NN	NN
30.00%	NN	−2.369	NN	−3.714	NN	NN	NN
40.00%	NN	−2.134	−2.449	−3.285	NN	NN	−2.78
50.00%	NN	−1.899	−2.108	−2.856	NN	NN	−1.781
60.00%	NN	−1.664	−1.766	−2.428	NN	NN	−1.725
70.00%	NN	−1.429	−1.425	−1.999	NN	−3.652	−1.725
80.00%	−2.211	−1.194	−1.084	−1.57	−3.036	−2.572	−1.725
90.00%	−1.516	−0.959	−0.743	−1.141	−2.291	−1.491	−1.725
100.00%	−0.822	−0.724	−0.402	−0.712	−1.546	−0.41	−1.725

Managerial resilience (MR), temporal psychological capital (TPC), employee innovative behavior (EIB), employee performance (EP), temporal control (TC), temporal fit (TF), idea generation (IG), idea realization (IR), idea promotion (IP).

) method to generate bottleneck tables. At low performance levels (≤20%), no variables acted as bottlenecks, indicating that baseline performance does not rely on high predictor values. However, from the 30% level onward, IP and Innovative Role IR consistently emerged as limiting factors. MR became essential from the 40% level and remained stable above 60%, suggesting its foundational role. At high performance thresholds (≥80%), additional constraints appeared—IG, TC, and TF became necessary, indicating that temporal psychological resources and innovation are critical for peak performance. At the 100% level, nearly all conditions must exceed specific thresholds, confirming their non-compensatory role.

Overall, the bottleneck analysis reveals that IP, IR, and MR are central constraints across most performance levels, while TC and TF are crucial at the upper end. These findings reinforce the necessity results and emphasize that organizations aiming for exceptional employee performance must address these critical conditions, particularly by fostering intellectual competencies, resilience, and temporal psychological resources.

5. Implications and Discussion

5.1. Theoretical Implications

This study makes several theoretical and practical contributions to understanding employee performance (EP) and organizational resilience in small and medium-sized enterprises (SMEs) under crisis conditions. By integrating Temporal Motivation Theory (TMT), Conservation of Resources (COR) theory, and the Job Demands–Resources (JD-R) model, we offer a richer explanation of how managerial and employee-level resources interact to sustain performance in high-uncertainty environments.

From a TMT perspective, our results show that managerial resilience (MR) enhances performance both directly and indirectly through temporal psychological capital (TPC) and employee innovative behavior (EIB). Managers with high resilience set temporal norms, clarify priorities, and maintain momentum, aligning individual pacing (temporal fit) with organizational urgency and fostering an environment where innovation can occur under compressed timelines. COR theory helps explain why temporal control (TC) and temporal fit (TF) are positive drivers: they allow employees to allocate limited resources efficiently and avoid depletion, thus preserving their capacity for adaptive and innovative work. In contrast, time pressure resilience (TP) exhibited a negative association with performance, a finding that challenges the assumption—embedded in psychological capital frameworks—that all resilience dimensions are inherently beneficial. The JD-R model provides insight here: sustained high time pressure represents a chronic job demand that can exhaust cognitive and emotional resources, leading to reduced task quality and lower long-term performance. Employees who continually absorb excessive time pressure may cross a tipping point where coping strategies become maladaptive, suggesting that resilience to time pressure in crisis contexts must be actively managed to avoid resource depletion.

Methodologically, our integration of PLS-SEM and NCA advances resilience research by combining sufficiency and necessity perspectives. While PLS-SEM identifies average causal pathways (e.g., MR→TC→EP; MR→IR→EP), NCA pinpoints non-compensatory conditions that must be present for high performance: idea promotion (IP), idea realization (IR), and MR across most performance thresholds, with TC and TF emerging as essential at top performance levels. This dual approach increases theoretical precision by distinguishing between factors that can drive performance and those that must exist to achieve it.

Practically, the findings have several implications for SMEs navigating crisis environments:

(1)

Strengthen temporal alignment—Implement training and systems to improve temporal control and fit, such as structured priority-setting sessions, shared digital timelines, and synchronized workflow planning.

(2)

Manage time pressure proactively—Monitor workload patterns to prevent chronic overexposure to high-pressure demands, introduce rotating relief roles, and encourage realistic deadline setting to preserve employee resources.

(3)

Support innovation execution—Provide targeted support for idea promotion and realization, including stakeholder-mapping exercises, cross-departmental collaboration platforms, and micro-budgets for pilot testing ideas.

(4)

Embed resilience in leadership development—Equip SME leaders with skills to maintain composure under uncertainty, mobilize resources quickly, and signal consistent support for innovation even during disruption.

Theoretically, this study extends multilevel resilience theory by demonstrating that managerial psychological traits cascade into employee-level temporal and innovation capacities, and that the benefits of certain resilience dimensions (e.g., TP) are context-dependent rather than universally positive. We re-conceptualize TPC as a multidimensional construct that operates differently across crisis conditions, integrating temporal cognition into resilience and performance models. By situating these mechanisms in the SME crisis context, we respond to calls for situated theorizing in organizational resilience research, highlighting how lean structures, resource scarcity, and innovation dependence shape resilience–performance dynamics in ways that differ from large organizations.

5.2. Practical Implications

This study provides several practical recommendations for SME leaders, HR professionals, and policymakers aiming to foster organizational resilience and employee performance under volatile and uncertain conditions [77]. The findings confirm the following:

Build resilience at the leadership level: Given MR’s foundational role, SMEs should implement leadership development programs focused on adaptive cognition, stress regulation, and strategic foresight. For example, scenario-based crisis simulations can strengthen leaders’ ability to maintain stability while making rapid decisions under uncertainty.

Target temporal psychological capital, especially TC and TF: Since TC exerts a strong positive effect on EP, interventions such as time management workshops, workload redistribution, and digital scheduling tools can help employees maintain a sense of control. TF can be improved through aligning work cycles with employees’ natural pacing, e.g., flexible shifts or synchronized deadlines.

Manage time pressure resilience carefully: The negative TP–EP relationship suggests that more resilience is not always better if it is achieved by enduring excessive time pressure. Leaders should monitor workload intensity and recovery periods to prevent burnout, for instance, by enforcing micro-breaks or rotating high-pressure tasks.

Close the gap between idea advocacy and execution: The non-significant IP–EP link indicates that advocacy without follow-through is insufficient. SMEs should streamline innovation pipelines—allocating resources, decision rights, and cross-functional support to move viable ideas swiftly from conception to implementation.

The necessary condition analysis further emphasizes that certain variables—such as IP, IR, and MR—serve as non-compensatory constraints at multiple performance levels. Particularly at higher performance thresholds, temporal cognition (TC, TF) and innovation drivers (IG) become critical bottlenecks [76]. As such, leaders should adopt early-warning diagnostic tools to monitor deficiencies in these areas and direct resources strategically to resolve them before expanding broader performance initiatives.

Collectively, these insights provide actionable direction for SMEs navigating crisis contexts, highlighting the centrality of resilience, innovation execution, and time-based psychological resources in building sustainable organizational agility.

5.3. Discussion

The present study provides empirical evidence that managerial resilience (MR) significantly enhances employee performance (EP) in SMEs under crisis conditions, primarily through two key mediating pathways: temporal psychological capital (TPC) and employee innovative behavior (EIB). This finding extends the resilience literature [77,103] by demonstrating that resilience at the managerial level not only provides emotional stability but also activates employees’ time-based cognitive resources and innovative actions, thereby contributing to sustainable organizational performance.

First, invest in resilience-building at the managerial level. Training programs should focus on adaptive cognition, strategic foresight, and stress regulation to ensure that leaders can maintain psychological stability while supporting teams under high uncertainty. Given SMEs’ flatter structures and limited redundancy, leadership resilience has an immediate, cascading impact on employee temporal and innovation capacities.

Second, strengthen employees’ temporal psychological capital (TPC) with a targeted approach. Our results show that temporal control (TC) exerts a strong positive influence on performance, while temporal fit (TF) plays a moderate role and time pressure resilience (TP) may even suppress performance when overactivated. To maximize benefits, SMEs should provide time management training and personalized scheduling tools to enhance TC; align task pacing and project timelines to employees’ working rhythms, improving TF; and monitor for signs of TP overuse, such as fatigue and burnout, by implementing workload redistribution and flexible scheduling to prevent resource depletion.

Third, close the gap between idea advocacy and implementation. Although idea generation (IG) and realization (IR) significantly enhance performance, idea promotion (IP) alone does not yield measurable benefits. This suggests a bottleneck in the innovation pipeline. SMEs should establish cross-functional innovation taskforces, grant decision-making authority to project champions, and allocate resources explicitly tied to implementation milestones to ensure ideas progress from conception to delivery.

Finally, use diagnostic tools to detect and address bottlenecks early. The Necessary Condition Analysis reveals that certain variables (e.g., MR, IP, IR, TC) are non-compensatory for achieving high performance. Leaders should adopt early-warning metrics to identify deficiencies in these areas and intervene before broader performance initiatives are rolled out. This proactive approach is particularly critical in SMEs, where resource misallocation can have immediate and lasting negative effects. Collectively, these measures can help SMEs navigate crisis environments with greater agility, ensuring that resilience, temporal alignment, and innovation execution work in tandem to sustain competitive advantage.

6. Conclusions

This study examines how MR in SMEs under crisis conditions translates into sustained EP. Drawing on JD-R, COR, and TMT, the findings show that MR enhances EP through two main pathways: building TPC and enabling EIB. Within TPC, TC and TF consistently support performance, whereas TP—when overextended—has a negative effect, indicating that not all resilience dimensions are universally beneficial. For EIB, IG and IR emerge as strong predictors of EP, while IP alone does not yield significant gains, emphasizing that execution capacity matters more than advocacy in resource-constrained SMEs.

Theoretically, this work advances resilience and innovation research by positioning TPC as a context-specific psychological resource and demonstrating its mediating role between MR and EP. Methodologically, the integration of PLS-SEM and NCA distinguishes between drivers of performance and necessary conditions (IR, IP, and TC), offering a more diagnostic and nuanced understanding of performance mechanisms under high uncertainty.

Practically, the results suggest that SME leaders should align individual and organizational pacing, strengthen TC, protect employees from chronic TP, and invest in the full innovation cycle from IG to IR. Resilience-driven performance is ultimately a matter of balance—mobilizing the right temporal, cognitive, and behavioral resources in ways that fit the crisis context, while avoiding the overuse of coping strategies that may erode capacity over time. This study closes a key gap in understanding how resilience is operationalized in SMEs and offers a roadmap for sustaining competitiveness amid volatility and change (see Figure 6 and

Table 11. Summary of key contributions.

Contribution Type	Description
Theoretical advances	➀ Clarifies the nuanced role of TP in crisis contexts. ➁ Integrates TPC into resilience–performance models. ➂ Emphasizes execution-focused innovation (IG and IR) as critical for SME performance.
Methodological contributions	➀ Applies a combined sufficiency–necessity framework (PLS-SEM + NCA) in SME resilience research. ➁ Distinguishes between performance drivers and non-compensatory prerequisites (IR, IP, TC).
Practical relevance	➀ Guides SME leaders to align individual–organizational pacing and strengthen TC. ➁ Recommends managing TP to prevent resource depletion. ➂ The need to support the full innovation cycle from IG to IR for sustained competitiveness.

).

7. Limitations and Future Research

Although this study offers robust theoretical and practical contributions, several methodological, contextual, and cultural limitations should be acknowledged. First, the cross-sectional design precludes causal inference between MR, TPC, EIB, and EP; as such, the observed associations may be subject to reverse causality or omitted variable bias. Longitudinal or panel data would allow for the examination of how these relationships evolve across different crisis stages, clarifying whether resilience investments generate compounding benefits or diminishing returns over time. Second, reliance on self-reported measures may introduce perceptual or social desirability bias, and while procedural and statistical remedies (e.g., Harman’s test) were employed to mitigate common method variance, residual bias cannot be entirely ruled out. Third, the study is situated within the Chinese SME context, where high collectivism, moderate-to-high uncertainty avoidance, and institutional factors such as government–business relations may shape how resilience and innovation mechanisms operate. These cultural and institutional characteristics may limit the generalizability of findings to SMEs in more individualistic, low power-distance, or less state-influenced environments. Fourth, although the sample covers multiple industries, sector heterogeneity may mask industry-specific dynamics—such as stronger TPC–EIB–EP linkages in technologically turbulent sectors compared to more stable service industries. Finally, while recommendations for enhancing MR, strengthening TPC, and improving innovation pipelines are provided, practical implementation in SMEs may be constrained by limited resources during crises, requiring external support such as targeted policies, financial incentives, or public–private partnerships.

Building on the current findings, future studies can extend and refine this research in several concrete ways. First, conduct longitudinal studies that track MR, TPC, EIB, and EP across pre-crisis, crisis, and post-crisis phases. Such designs would capture the dynamic evolution of resilience and performance mechanisms, clarifying whether certain pathways strengthen or weaken over time. Second, pursue multi-country comparative research to test the model in different cultural and institutional contexts, thereby identifying how variations in collectivism, power distance, or uncertainty avoidance influence the MR–TPC–EIB–EP relationships. Third, adopt mixed-methods designs by combining large-scale surveys with in-depth interviews or case studies to uncover qualitative nuances—such as leadership narratives, employee sensemaking, and organizational routines—that underlie the quantitative patterns. Fourth, examine industry-specific resilience strategies by comparing high-tech and service-sector SMEs, where innovation demands, resource constraints, and temporal pressures differ markedly. Such comparisons could reveal sector-tailored approaches to leveraging TPC and EIB for sustained performance.