Financial Risk Management and Resilience of Small Enterprises Amid the Wartime Crisis

Abstract

This study examines the financial resilience of small enterprises in Ukraine during the wartime crisis, addressing the lack of quantitative evidence on how regional military risks and adaptive strategies jointly shape SME stability. The analysis is based on a sample of 30 small agricultural enterprises from the eastern, central, and western regions of Ukraine using annual data for 2022–2024. To capture multidimensional resilience patterns, the study applies factor analysis, cluster analysis, and taxonomic assessment methods to evaluate financial performance, operational adaptability, and access to external resources. The findings show that resilience variation across the sample is strongly associated with enterprises’ ability to sustain revenue flows, control operating costs, and maintain a balanced capital structure. Three distinct resilience profiles were identified: high resilience in western regions (KT = 0.89), moderate resilience in central regions (KT = 0.81), and low resilience in eastern frontline regions (KT = 0.49). These results indicate substantial regional asymmetry linked to differentiated exposure to military threats. Building on these empirical insights, the study proposes a hybrid risk-management approach that integrates digitalization of financial operations, diversification of funding sources, and enhanced social engagement as mechanisms supporting adaptation under prolonged instability. The novelty of the research lies in combining regional risk exposure with multidimensional financial indicators to develop an evidence-based framework for assessing SME resilience in wartime conditions.

1. Introduction

The full-scale war in Ukraine has fundamentally changed the conditions in which small enterprises operate, creating financial risks that far exceed typical market volatility. Military actions, infrastructure destruction, disrupted logistics, and shrinking access to capital have sharply weakened the stability of small and medium-sized enterprises (SMEs), which are structurally the most vulnerable to external shocks (Hatab & Lagerkvist, 2024; Helwig, 2023). Unlike peacetime crises, wartime instability simultaneously affects liquidity, debt pressure, operational costs, and regional business risks, forming a new configuration of threats that remains insufficiently analyzed in the current literature. Recent studies on SME resilience highlight the importance of financial management flexibility, digitalization, and resource diversification (Hobfoll et al., 2018; Kolodiziev et al., 2024b; Meier et al., 2024). However, most existing works focus on general crises—pandemics, economic recessions, or supply chain disruptions—while the specific mechanisms through which military risks shape SME financial stability remain understudied. Empirical evidence on how regional exposure to military threats interacts with financial indicators to determine business resilience is especially limited. Despite this growing body of research, the central problem remains that current studies do not provide a quantitative and regionally differentiated understanding of SME financial resilience under wartime conditions. In the Ukrainian context, emerging research indicates that enterprises increasingly rely on adaptive strategies such as digital transformation, social engagement, and reorganization of operational flows (Shcherbak et al., 2025; Kolodiziev et al., 2024a; Shenderivska et al., 2022; Singh et al., 2022). Yet the extent to which these strategies compensate for wartime financial pressures—and whether resilience varies systematically across regions with different levels of military threat—has not been comprehensively assessed. A clear research gap exists in empirically determining how specific financial indicators, adaptive behaviors, and regional conflict intensity jointly influence SME resilience. Addressing this gap is necessary to build evidence-based policy and management tools. Therefore, this study focuses on identifying the financial risks that most critically influence SME stability during the wartime crisis, quantifying regional variations in resilience, and developing an adaptive risk-management model grounded in empirical evidence.

1.1. Literature Review

Small and medium-sized enterprises (SMEs) operating in wartime face a set of financial risks that differ fundamentally from those found in economic or pandemic crises. Existing research highlights disruptions in supply chains, loss of assets, liquidity shortages, and rising operational costs as key threats under extreme conditions (OECD, 2023; World Bank, 2023). However, while prior studies examine general crisis-management strategies, evidence on how wartime risks specifically reshape financial resilience—especially across regions with different levels of military threat—remains fragmented. Recent wartime research confirms substantial regional asymmetry and persistent disruptions affecting SME stability (UNDP, 2022; Dligach & Stavytskyy, 2024), yet quantitative evaluations linking financial indicators to conflict exposure are still limited. This gap underscores the need for a comprehensive review of resource-based, adaptive, psychological, and spatial mechanisms that influence SME resilience under war.

1.1.1. Resource-Based Theories and Financial Risk

The Conservation of Resources (COR) theory (Hobfoll et al., 2018) provides an essential lens for understanding financial resilience, suggesting that enterprises attempt to preserve scarce financial and operational resources under external pressure. In wartime, resource depletion becomes faster and more unpredictable, intensifying financial constraints. Building on this logic, Lanivich (2013) argues that entrepreneurs rely on heuristics to maintain business continuity during high uncertainty. These insights align with studies showing that resource-induced coping strategies support decision-making in hostile environments (Adomako, 2020) and that experience in militarized or resource-constrained contexts enhances adaptive capacity (Crecente et al., 2020). Strategic management perspectives also emphasize leveraging firm-specific resources to sustain competitive advantage during instability (Porter, 2000; Butko & Zakharchenko, 2025). Research on business exit (DeTienne, 2008; DeTienne et al., 2014) similarly shows that shocks can force firms to reallocate limited resources, sometimes resulting in temporary closure. Nevertheless, existing studies do not fully explain how resource constraints interact with regional exposure to military risk or how these risks reshape financial indicators such as liquidity, solvency, and access to finance—an essential gap this study addresses.

1.1.2. Adaptation and Financial Resilience Under Crisis Shocks

Studies on SME adaptation during crises emphasize the importance of three key strategies: cost control, digital tools, and operational restructuring (Aidoo et al., 2021; Atiase et al., 2022; Arroyabe et al., 2024; Karmaker et al., 2022). Crisis research highlights that cost-control measures—such as expenditure prioritization, supplier renegotiation, and reductions in non-essential operating expenses—play a critical role in mitigating liquidity pressure during shocks (Aidoo et al., 2021). Likewise, operational restructuring through supply chain reconfiguration, process optimization, and adjustments in production capacity strengthens business continuity and reduces exposure to external disruptions (Karmaker et al., 2022; Dligach & Stavytskyy, 2024). These mechanisms complement digital tools, which enhance cash-flow monitoring and the speed of financial decision-making (Meier et al., 2024). Empirical studies further show that operational flexibility and resource reallocation enhance financial resilience under severe shocks (Xu et al., 2021; Crecente et al., 2020), and Ukrainian wartime evidence confirms that enterprises implementing digital processes tend to maintain higher stability (Kolodiziev et al., 2024b). However, existing literature remains largely descriptive and does not quantify the financial effects of these adaptation strategies, nor does it examine how their effectiveness differs across regions with varying levels of conflict intensity. Importantly, the present study does not treat digitalization as the sole determinant of financial resilience. Instead, cost-control mechanisms, operational restructuring, and digital tools operate as an integrated set of adaptive responses whose effectiveness depends on the severity of wartime disruptions. This integrated perspective maintains conceptual consistency within the section and aligns with the empirical findings discussed later. This gap in understanding the combined and regionally differentiated effects of adaptive strategies constitutes the second key research problem addressed in this study.

1.1.3. Psychological Factors and Their Relevance to Financial Outcomes

Psychological mechanisms—stress, burnout, and perceived uncertainty—significantly influence entrepreneurial decision-making during crises (Sardeshmukh et al., 2018). While not directly financial, these factors help explain why some SMEs maintain financial discipline and adaptive behavior while others face liquidity deterioration or withdrawal from the market. Crisis-induced stress increases the likelihood of business inactivity (Haynie & Shepherd, 2010), whereas supportive networks facilitate recovery (Ucbasaran et al., 2012). In wartime settings, these psychological influences shape leaders’ capacity to implement financial strategies such as cost restructuring, debt management, or investment decisions. In the wartime context, these psychological factors influence leaders’ ability to sustain financial strategies such as cost restructuring, debt management, or investment decisions (Shepherd & Williams, 2014). However, their direct link to wartime financial indicators has not been systematically studied, limiting the understanding of resilience mechanisms under extreme conditions.

1.1.4. Wartime Context and the Need to Study Regional Financial Resilience

War introduces unique institutional and spatial risks. Research shows that the severity of military threats disrupts supply chains, alters demand, and undermines operational continuity (Hatab & Lagerkvist, 2024). Ukrainian studies highlight the role of digitalization and social responsibility in sustaining business activity under such conditions (Kolodiziev et al., 2024a; Shcherbak et al., 2024). However, most works provide qualitative insights without distinguishing how financial resilience differs across frontline, central, and western regions. Although some authors examine military experience as a factor in managerial resilience (Benmelech & Frydman, 2014; Su & Li, 2024), this research does not specifically address war-affected SMEs, nor does it analyze how conflict exposure translates into differences in liquidity, debt burden, or access to finance. Thus, the key knowledge gap concerns measurable regional asymmetry in SME financial resilience during war—critical for evidence-based policy design.

Summary of Gap

Across reviewed literature, three gaps emerge:

-

Lack of quantitative assessments of how wartime financial risks affect SMEs.

-

Insufficient analysis of digitalization’s financial impact, beyond general descriptions.

-

Absence of comparative regional studies that capture how varying conflict intensity shapes financial resilience.

These gaps justify the study’s focus on identifying financial risk determinants, measuring resilience differences across regions, and proposing a model aligned with wartime conditions.

1.2. Justification of the Goal, Objectives, and Hypotheses of the Study

In recent years, small enterprises in Ukraine have found themselves in conditions of unprecedented challenges caused by a combination of economic, humanitarian, and infrastructural consequences of the war. The destruction of production facilities, disruption of logistics chains, loss of markets, and limited access to financial resources have led to a mass reduction in business activity, especially in the eastern and southern regions of the country. According to the World Bank and the Kyiv School of Economics, the total economic losses of Ukraine as a result of the war exceed 400 billion dollars, with a significant part of the damage falling on small and medium-sized businesses. Nevertheless, partial stabilization of entrepreneurial activity and the formation of new centers of resilience are observed in the central and western regions, associated with the relocation of enterprises, the development of digital services, and the growth of domestic demand. This creates a unique opportunity to study the factors of financial stability and mechanisms of risk management in the context of a protracted military crisis.

The goal of the study is to determine the key financial risks affecting the stability of small agricultural enterprises in Ukraine under wartime conditions and to develop an adaptive financial risk-management model that enhances their viability and capacity for recovery. To achieve this purpose, the study integrates financial indicators, regional exposure to military threats, and enterprise-level adaptive strategies into a unified analytical framework.

Objectives of the study:

-

To analyze the current state of small businesses in various regions of Ukraine in conditions of the military crisis;

-

To evaluate and quantify the impact of the main financial risks faced by enterprises (liquidity shortages, credit risk, currency fluctuations, operational losses);

-

To investigate adaptation strategies and resilience mechanisms applied by small businesses in conditions of limited resources;

-

To assess the role of digital transformation, innovation, and social responsibility in increasing the stability of enterprises;

-

To develop recommendations for forming a financial risk management system capable of supporting SMEs in conditions of prolonged instability.

These objectives collectively form a coherent framework that guides the empirical analysis and ensures a comprehensive assessment of financial resilience among small agricultural enterprises during wartime.

Hypotheses of the study:

H1. 

The implementation of adaptive financial management strategies and digital tools significantly increases the resilience of small enterprises to external shocks.

H2. 

Access to financial resources (including grants, loans, and donor support) is a determining factor for the survivability of enterprises in conditions of war.

H3. 

The geographical location of the enterprise significantly affects its ability to adapt to the crisis and maintain financial stability.

H4. 

Entrepreneurs with experience in volunteer and defense initiatives demonstrate a higher level of strategic resilience and readiness for risk management.

This study is aimed at developing practical tools and strategies that will help small enterprises not only minimize losses but also use the crisis as an opportunity for development, innovation, and strengthening financial stability. The analysis is based on data from national and international reports, as well as on examples of adaptation of Ukrainian enterprises in 2022–2024.

2. Materials and Methods

2.1. Data for Assessing the Financial Stability of Small Enterprises in Ukraine Under the Conditions of the Military Crisis

For the analysis of financial risks and stability of small enterprises in Ukraine during the period of the military crisis, secondary data from national and international sources were used, including the World Bank, the International Monetary Fund, the OECD, the United Nations Development Programme (UNDP), the European Bank for Reconstruction and Development (EBRD), as well as analytical reports of the Ministry of Economy of Ukraine, the Kyiv School of Economics, and the State Statistics Service of Ukraine for 2022–2024. The empirical analysis is based on a dataset of 30 small agricultural enterprises from the eastern, central, and western regions of Ukraine. The sample includes firms operating under varying levels of military pressure, which ensures sufficient heterogeneity for assessing regional differences in financial resilience. The dataset contains standardized financial indicators, measures of digitalization, access to finance, and social engagement, enabling a multidimensional evaluation of adaptive capacity during wartime. The sample was constructed using explicit inclusion criteria: (1) the enterprise must belong to the category of small businesses according to Ukrainian legislation (annual revenue below EUR 10 million and fewer than 50 employees); (2) it must operate in the agricultural sector, which represents the most affected and regionally diverse SME category during the war; (3) complete financial data for all key indicators must be available for the 2022–2024 period. Data were obtained from official statistical institutions, including the State Statistics Service of Ukraine, the Ministry of Economy of Ukraine, the Kyiv School of Economics wartime enterprise monitoring project, and international databases (World Bank, OECD, UNDP, EBRD). All raw KPI values, sector classification, regional distribution, and cluster membership details are fully presented in [Appendix A]{.mark} to ensure replicability. Each financial indicator reflects an annual value for 2024, while the RRR indicator incorporates a comparison with pre-war benchmarks (2021), and RRE is derived from regional conflict intensity indices for 2022–2024.

The goal of the analysis is to determine which factors contribute to the financial stability of SMEs and which risks critically affect their operations in conditions of war. The analysis specifically targets small agricultural enterprises operating in three regions of Ukraine.

For quantitative assessment, key financial and managerial indicators reflecting various aspects of enterprise functioning were used (

Table 1. Key Performance Indicators (KPI) for Assessing the Financial Resilience of Small Enterprises amid the Wartime Crisis.

№	Indicator	Abbreviation	Description	Thresholds/Interpretation
1	Liquidity Ratio	LR	Liquidity ratio—reflects a company’s ability to meet short-term obligations.	>1.5—good; 1.0–1.5—satisfactory; <1.0—problematic.
2	Debt-to-Equity Ratio	DER	Debt-to-equity ratio, which characterizes the level of financial dependence.	<1.0—low debt; 1.0–2.0—moderate; >2.0—high.
3	Cash Flow Stability Index	CFSI	Cash flow stability indicator—reflects a company’s ability to maintain a positive cash balance.	Scale 0–1. Closer to 1 indicates greater stability.
4	Revenue Recovery Rate	RRR	Revenue recovery compared to the pre-war period.	Percentage (%) of recovery relative to pre-crisis (2021) level.
5	Operational Cost Increase Rate	OCIR	Operating cost growth rate due to military operations and logistical constraints.	Percentage (%) increase relative to pre-crisis period.
6	Digitalization Level of Financial Operations	DLFO	Share of digital financial transactions (online payments, electronic reporting, CRM systems).	Scale 1–5; 1—low, 5—high.
7	Access to External Finance Index	AEFI	Assessment of the availability of loans, grants, and donor programs.	Scale 1–5; 1—very low, 5—very high.
8	Resilience Adaptation Score	RAS	An integrated indicator reflecting a company’s ability to adapt to changing market conditions.	Scale 1–10; higher score indicates greater resilience.
9	Social Responsibility Engagement Level	SREL	Level of company involvement in socially responsible initiatives and volunteer programs.	Scale 1–5; 1—low, 5—high.
10	Regional Risk Exposure	RRE	Regional exposure to military and security risks, based on conflict intensity indices and proximity to frontline zones.	Scale 1–10; higher value indicates greater risk exposure (e.g., 9–10 = frontline, 3–5 = low-risk western regions).

Note: The thresholds and interpretation scales for LR and DER are based on standard corporate financial analysis principles widely adopted in SME assessment frameworks (e.g., OECD Scoreboard, World Bank Enterprise Surveys). The scaling for DLFO, AEFI, RAS, SREL, and RRE follows the methodological approaches used in recent SME resilience studies in conflict-affected economies (UNDP, 2022; Kolodiziev et al., 2024b). The CFSI construction and interpretation align with cash-flow stability metrics recommended by the European Bank for Reconstruction and Development (EBRD) for crisis-period diagnostics.

). The RRE (Regional Risk Exposure) indicator is included alongside financial and operational metrics to account for the spatial dimension of military risk, which directly influences liquidity, costs, and access to finance in wartime settings.

Data on key performance indicators of small businesses, taking into account financial stability in the context of a military crisis, are provided in Appendix B.

2.2. Research Methodology for Assessing Financial Resilience

The methodological framework follows internationally established standards for multidimensional assessment of financial resilience. The selection of analytical tools is based on the approaches recommended by the OECD (2023) for SME vulnerability assessment, the OECD (2023); Dligach and Stavytskyy (2024) methodology for crisis diagnostics, and the UNDP (2022). These sources formally justify the use of factor analysis, cluster segmentation, and taxonomic benchmarking in wartime economic research.

The empirical analysis is structured into five sequential stages:

• Identification of key financial determinants using exploratory factor analysis (EFA).
• Classification of SMEs into resilience profiles using cluster analysis (k-means).
• Benchmarking financial resilience through the taxonomic method to compute an integral resilience indicator (KT).
• Statistical verification of hypotheses using correlation analysis, t-tests, ANOVA, and non-parametric tests.
• Integration and validation of results through cross-validation and sensitivity analysis.

All key indicators (LR, DER, CFSI, RRR, OCIR, DLFO, AEFI, RAS, SREL, RRE) were calculated using annual data for the period 2022–2024. For each enterprise, indicator values represent the mean of available yearly observations. The taxonomic resilience coefficient (KT) was computed using annual standardized values for the same period.

This structured approach enables a systematic examination of the determinants of SME financial resilience under wartime conditions and provides a transparent workflow for hypothesis testing and model development. All detailed mathematical formulations and computational procedures are reported in the Appendix C, Appendix D and Appendix E to ensure full transparency and replicability. The complete set of code scripts used for data preprocessing and statistical analyses (factor, cluster, and taxonomic methods) is provided in Appendix G.

2.3. Analytical Parameters and Procedures

To ensure transparency and reproducibility, this subsection summarizes the core analytical procedures used in the study. Detailed technical specifications—including software settings, convergence criteria, and parameter values for factor analysis, clustering, and taxonomic benchmarking—are provided in Appendix I for reference.

Figure 1 provides an overview of the sequential methodological framework used to structure the empirical analysis in this study.

This workflow illustrates how each analytical stage builds upon the previous one, ensuring a coherent and replicable methodological pathway.

Figure 2 presents a detailed workflow that outlines the full sequence of analytical operations and their intermediate outputs.

Figure 2 details the analytical operations corresponding to the general workflow shown in Figure 1.

Table 2. Statistical Parameters of Analytical Procedures.

Procedure	Software	Key Parameters	Convergence Criteria	Output Metrics
Factor Analysis	STATISTICA 13	KMO = 0.87; Bartlett’s test: p < 0.001; eigenvalue > 1.0	Eigenvalue > 1.0	90.1% variance explained
Cluster Analysis	STATISTICA 10	k = 3; max iterations = 100	Centroid change < 0.0001	Silhouette score = 0.72
Taxonomic Method	Python 3.10	Min–max normalization; Euclidean distance	Benchmark vector stability check	KT coefficient range: 0.38–0.94

summarizes the key statistical parameters and output metrics used across the analytical procedures.

The presented parameters confirm the methodological robustness of the analyses and demonstrate that each procedure meets the necessary statistical reliability standards.

2.4. Statistical Assumptions, Diagnostic Checks, and Robustness Procedures

The selection between parametric (t-tests, one-way ANOVA) and non-parametric (Mann–Whitney U-test, Kruskal–Wallis test) statistical procedures was based on preliminary distribution diagnostics. The normality of all indicators was assessed using the Shapiro–Wilk test and Q–Q plots. Variables demonstrating significant deviations from normality (p < 0.05) were analyzed using non-parametric methods, whereas indicators with approximately normal distributions and homogeneous variances (Levene’s test, p > 0.05) were examined using parametric techniques. To assess potential multicollinearity among the financial, digitalization, and resilience-related indicators, Variance Inflation Factors (VIFs) were calculated for all explanatory variables. All VIF values remained below the conservative threshold of 5, confirming the absence of statistically critical multicollinearity and ensuring the stability of the estimated parameters. Robustness of the empirical results was additionally evaluated through sensitivity diagnostics of the taxonomic resilience coefficient (KT) and internal consistency checks of the cluster solutions. Due to the absence of direct firm-size indicators in the available dataset, scale effects were indirectly controlled using liquidity and leverage proxies. The methodological limitations associated with this approach are explicitly addressed in Section 4.

3. Results

3.1. Results of the Factor Analysis of Financial Indicators of Small Enterprise Sustainability

To identify the key factors determining the financial sustainability of small enterprises in Ukraine under the conditions of the military crisis, a factor analysis was conducted using the principal component analysis method.

The analysis allowed for grouping the interrelated indicators into a single latent factor, reflecting an integral characteristic of the enterprises’ financial sustainability. Standardized data (Data_nor) were used for the analysis. The threshold value of the factor loading was taken equal to 0.7, which corresponds to a strong correlation of the variable with the factor (

Table 3. Results of Factor Analysis of Financial Stability Indicators of Small Enterprises (Principal Components) (STATISTICA 13 listing).

Variable	Factor Loadings (Unrotated) (Data_nor) Extraction: Principal Components (Marked Loadings Are >0.700000)
	Factor 1
LR	−0.984165
DER	0.982099
CFSI	−0.982463
RRR	−0.993300
OCIR	0.986691
DLFO	−0.885731
AEFI	−0.963329
RAS	−0.989024
SREL	−0.801629
RRE	0.902614
Expl.Var	9.00655
Prp.Totl	0.900654

).

The results of the factorial analysis show that the first principal component explains 90.1% of the total variance of the data, which testifies to the high internal consistency of the selected indicators and their ability to integrally describe the level of sustainability of enterprises. The strongest factorial loads (>|0.95|) are observed for the variables RRR (−0.9933), RAS (−0.9890), OCIR (0.9867), DER (0.9821), and CFSI (−0.9825), which indicates the dominant influence of the indicators of revenue recovery, adaptation sustainability, financial dependence, and cash flow stability on the overall sustainability of enterprises. The variables DLFO (−0.8857) and AEFI (−0.9633) also demonstrate high negative loads, reflecting that the level of digitalization and access to financing have a close inverse connection with financial risks—the higher these indicators, the higher the enterprise’s sustainability. The positive connection between DER and OCIR indicates that an increase in the debt load and growth in costs intensifies financial risks, reducing business sustainability. The high load of the variable RRE (0.9026) confirms that the geographical factor—the degree of regional exposure to military risks—has a critical influence on financial stability. In aggregate, the results confirm that the financial sustainability of SMEs in the military period is determined by the balance between financial risks (DER, OCIR, RRE) and adaptation capabilities (RRR, RAS, DLFO, AEFI, SREL). This indicates the necessity of developing a complex risk-management model, uniting both financial and organizational-managerial instruments of sustainability.

3.2. Classification of Small Enterprises by Zones of Financial Risk

At the second stage of the study, a classification of enterprises by levels of stability in the conditions of the military crisis was carried out by the method of cluster analysis (K-means clustering). Figure 3 illustrates the results of the cluster analysis, reflecting the distribution of small enterprises into three distinct resilience groups.

The results of the analysis (Figure 3) showed that the enterprises were distributed into three stable clusters, reflecting various zones of financial risk and sustainability:

The cluster analysis revealed three resilience groups. The detailed characteristics of each cluster are provided after

Table 4. Enterprises of the 1st cluster.

Members of Cluster Number 1 (Data_nor) and Distances from Respective Cluster Center Cluster Contains 10 Cases
Case No.	Distance	Case No.	Distance
C_4	0.4556115	C_20	0.5531445
C_8	0.3167202	C_22	0.4265745
C_9	0.5314636	C_26	0.5478812
C_14	0.4426546	C_28	0.4241928
C_16	0.2676535	C_29	0.6909757

Cluster 1 enterprises: C_4—FE “FERMINSKE”; C_8—FE “POLUYIANIVSKE”; C_9—FE “TORINSKE”; C_14—LLC “MK MedVyn”; C_16—LLC “AGRO OSTERS”; C_20—LLC “AGRO POINT GROUP”; C_22—FE “SKAVA”; C_26—FE “BUKOVIEN”; C_28—FE “KUZHBA”; C_29—FE “V FILVAROK”.

,

Table 5. Enterprises of the 2nd cluster.

Members of Cluster Number 2 (Data_Nor) And Distances From Respective Cluster Center Cluster Contains 13 Cases
Case No.	Distance	Case No.	Distance
C_11	0.266444	C_21	0.5324675
C_12	0.420495	C_23	0.3203908
C_13	0.373412	C_24	0.5405027
C_15	0.3476844	C_25	0.3465283
C_17	0.3587299	C_27	0.2404362
C_18	0.3822872	C_30	0.3606293
C_19	0.3733964

Cluster 2 enterprises: C_11—LLC “PRIMUS KOR”; C_12—LLC “SKYLIGHT”; C_13—LLC “AGROZEMTEKHPROEKT”; C_15—CEC “AMARANT AGRO ZROSHENNIA”; C_17—LLC “NEPTUNE FISH”; C_18—LLC “BT”; C_19—LLC “INKVILIN”; C_21—FE “ZAKHIDNE OPYLLIA”; C_23—LLC “EKO-LISBUD”; C_24—FE MOHORUK K. M.; C_25—LLC “ULTRA FORCE”; C_27—FE “ZAPIDOK”; C_30—LLC “BEST BERRY”.

and

Table 6. Enterprises of the 3rd cluster.

Members of Cluster Number 3 (Data_nor) and Distances from Respective Cluster Center Cluster Contains 7 Cases
Case No.	Distance	Case No.	Distance
C_1	0.3773603	C_6	0.5552688
C_2	0.3879911	C_7	0.3017505
C_3	0.4188683	C_10	0.7467695
C_5	0.3725443

Cluster 3 enterprises: C_1—LLC “AKVA-FISH KH”; C_2—FE “SANTONSKE”; C_3—FE “SEMILONSKE”; C_5—LLC “Semilon-Agro”; C_6—LLC “SANTON AGRO”; C_7—FE “EKOFUD-SLOBODA”; C_10—FE “SOTON AGRO”.

.

The first cluster included 10 enterprises, operating predominantly in the central regions of Ukraine. They demonstrate balanced indicators of liquidity and debt load, a moderate level of digitalization, and a high level of participation in social initiatives.

Enterprises of this cluster have moderate financial stability (LR ≈ 1.2–1.5; DER ≈ 1.0–1.8), which allows them to maintain stability under limited access to financing. Thanks to social engagement (SREL ≥ 0.6) and adaptability (RAS > 0.7).

The second cluster includes enterprises mainly located in the western regions of Ukraine, where military risks are minimal and business activity is gradually recovering. These companies are characterized by high liquidity and active digitalization of financial processes.

Enterprises of the second cluster demonstrate the best indicators of financial health (LR > 1.5; RAS > 0.8; DLFO > 0.6) and have access to external sources of financing (AEFI ≥ 0.7). They are characterized by high adaptive capacity and an innovation-oriented approach, which provides them with leading positions in terms of resilience under crisis conditions.

The third cluster includes enterprises located in the front-line southeastern regions of Ukraine, where active hostilities are taking place and the impact of military risks on economic activity is the highest.

Cluster 3 enterprises exhibit high DER, low LR and high RRE.

3.3. Regional Differentiation of Financial Resilience of Small Enterprises Based on the Taxonomic Method

At the third stage of the study, an assessment of regional differences in the financial resilience of small enterprises was conducted using the taxonomic method. The main goal was to define benchmark standards of resilience and compare deviations of regional indicators from the ideal model.

The benchmark vector z₀ was formed from the maximum values of key financial indicators across the entire sample:

Current liquidity ratio (LR)—2.1;

Cash Flow Stability Index (CFSI)—0.9;

Revenue Recovery Rate (RRR)—105;

Long-term Financial Security (DLFO)—5;

Asset Efficiency Indicator (AEFI)—5;

Resilience Adaptation Score (RAS)—10;

Level of Equity Preservation (SREL)—5.

For parameters that are inversely related in an economic sense (DER, OCIR, RRE), benchmark values were formed based on their minimum levels. The degree of deviation of actual indicator values from the benchmark was determined using the Euclidean distance di, reflecting the enterprise’s distance from the ideal state. Based on the obtained data, the taxonomic resilience coefficient KT was calculated using the formula, where the mean distance đ = 0.55 and the standard deviation s = 0.22.

The analysis results revealed a pronounced regional heterogeneity in financial resilience.

Cluster 2 (KT = 0.87–0.94) included enterprises predominantly from the western regions, characterized by the highest resilience. These enterprises exhibited high liquidity, digitalization of business processes, and access to external financing.

Cluster 1 (KT = 0.72–0.93) grouped enterprises predominantly from the central regions, showing a medium level of resilience, balanced financial position, and moderate risks.

Cluster 3 (KT = 0.38–0.60) included enterprises predominantly from the southeastern front-line zones, most vulnerable to financial shocks and external impacts.

Correlation coefficients and related analyses reveal patterns of co-movement between indicators, but they do not establish cause–effect mechanisms. Therefore, the results should be interpreted as evidence of statistical association only. The average value of the taxonomic resilience coefficient across the sample was KT = 0.74, confirming significant regional differences. Enterprises in the western regions were closer to the benchmark model (KT > 0.9), whereas southeastern enterprises significantly deviated from it, indicating the need for targeted government support, expansion of credit guarantees, and promotion of digitalization to improve their resilience.

Appendix H presents standardized KPI values, calculated multidimensional distances to the benchmark, and the final integral resilience coefficients KT. The higher the KT value, the closer the enterprise is to the benchmark level of financial resilience. Appendix I presents the extended statistical outputs, including the full PCA eigenvalue table, factor loadings matrix, cluster membership distances, taxonomic distance distributions, and KT sensitivity analysis.

Figure 4 shows the distribution of average taxonomic resilience coefficient (KT) values for small enterprises, grouped into clusters according to their level of financial resilience.

The heat map visually highlights the geographic distribution of financial resilience levels of small enterprises across the regions of Ukraine (Figure 5).

Regions are colored according to the values of the taxonomic coefficient KT, with darker shades representing higher resilience levels and lighter shades corresponding to lower resilience scores. The western regions display the most intense coloring, confirming the taxonomic analysis results showing the highest enterprise resilience in these areas (KT ≈ 0.89). Central regions exhibit moderate KT values (≈0.81), while eastern and southeastern frontline zones show the lightest coloring, indicating critically low financial resilience (KT ≈ 0.49). This visual representation facilitates the identification of regions requiring special attention and targeted support to enhance small business resilience.

3.4. Hypothesis Testing Using Statistical Methods

At this stage, the proposed hypotheses aimed at identifying factors influencing the financial resilience of small and medium-sized enterprises (SMEs) under conditions of military crisis were tested. Non-parametric and parametric statistical methods were used to test the hypotheses, including Spearman’s correlation analysis, independent samples t-test, one-way analysis of variance (ANOVA), and the Mann–Whitney U test. The choice of methods was determined by the type of data, the distribution of indicators, and the research objectives. Figure 6 presents a scatter plot illustrating the relationship between the digitalization level of financial operations (DLFO) and the resilience adaptation score (RAS) of small enterprises. The observed pattern indicates a strong positive association, consistent with the hypothesis H1.

To test hypothesis H1 (“The implementation of adaptive financial management strategies and digital tools increases SME resilience”), Spearman’s correlation analysis was conducted. The resulting coefficient ρ = 0.898 (p < 0.001) indicates a strong positive relationship between DLFO and RAS, confirming the hypothesis that effective financial management and digitalization enhance enterprise resilience (

Table 7. Correlation Analysis between DLFO and RAS (Hypothesis H1).

Indicators	Spearman’s Coefficient (ρ)	p-Value	Interpretation
DLFO vs. RAS	0.898	0.0000	Strong positive correlation

).

Figure 7 presents a comparison of average return on assets (RAS) of small enterprises across three levels of access to external financing (AEFI), reflecting the impact of financial support on business resilience.

In Figure 7, the X-axis represents three categories of access to external financing (AEFI): low (1–2) with an average RAS = 4.0, medium (3) with RAS ≈ 5.5, and high (4–5) with RAS = 8.7. The Y-axis shows the corresponding average values of the return on assets (RAS) indicator. The diagram demonstrates a clear positive relationship between AEFI level and enterprise financial resilience: as access to external financing increases, RAS values rise, indicating an enhanced ability of enterprises to adapt to financial challenges. Thus, the visualization results support hypothesis H2 regarding the key role of external financing in ensuring the survival and resilience of small businesses.

Figure 8 presents the distribution of the return on assets (RAS) of small enterprises across the regions of Ukraine, visualized as a box plot for testing hypothesis H3.

The diagram reflects differences in financial resilience across three regions: eastern, central, and western. For the eastern region, the median RAS is 4, with an interquartile range of 3–5 and values ranging from 2 to 5. In the central region, the median is 6, the interquartile range is 6–7, and values vary from 5 to 7. The western region demonstrates the highest resilience—the median RAS is 9, the interquartile range is 8–9, and values range from 8 to 10. Thus, there is a clear trend of increasing financial resilience of enterprises from east to west, confirming hypothesis H3 on the statistically significant influence of geographic location on the resilience level of small businesses.

Figure 9 presents box plots illustrating differences in the distribution of return on assets (RAS) between enterprises with high and low levels of social engagement (SREL), used to test hypothesis H4.

The diagram shows that the median RAS for enterprises with a high level of social engagement (SREL ≥ 4) is approximately 9, whereas for enterprises with a low level (SREL ≤ 3) it is around 6. The range of values in the first group is significantly narrower, indicating a consistently high level of efficiency among such enterprises. These differences support hypothesis H4 regarding the positive impact of participation in social and volunteer initiatives on the resilience of small and medium-sized businesses (

Table 8. Testing Hypothesis H4 Using the Mann–Whitney U Test.

Indicator	Group 1: High Engagement (SREL ≥ 4)	Group 2: Low Engagement (SREL ≤ 3)	Justification
Number of observations (n1, n2)	n1 = 9	n2 = 17	Division by level of social activity
Mean RAS	8.56	6.00	Entrepreneurs with high engagement demonstrate higher resilience
Median RAS	9	6	Confirms the shift in distribution
Standard Deviation	1.17	2.01	Greater variation in the low engagement group
Rank Sum (R1, R2)	R1 = 256	R2 = 149	Higher rank sum in the high SREL group
U Statistic (calculated by formula 14)	U1 = 28	U2 = 125	Minimum U value used for comparison
Critical U Value (α = 0.05)	Ucrit = 37		For n1 = 9 and n2 = 17
Decision	U1 < Ucrit → H0 rejected		Difference is statistically significant
Interpretation	Entrepreneurs with high SREL demonstrate significantly higher strategic resilience		Hypothesis H4 is confirmed

).

The results of the Mann–Whitney U test showed that the difference in financial resilience (RAS) between entrepreneurs with high and low levels of participation in social and defense initiatives is statistically significant (U = 28 < 37 at α = 0.05). This confirms hypothesis H4, which states that social activity and volunteer experience positively influence the strategic resilience of small enterprises.

3.5. Interpretation and Visualization of Results

The results of the statistical and cluster analyses were visualized using three key analytical tools: dendrograms, correlation matrices, and regional risk maps (RRE). These tools allow for a comprehensive assessment of the interrelationships between financial, social, and territorial factors of small enterprise resilience.

The dendrogram (Figure 10) was constructed based on DLFO, AEFI, RAS, and SREL indicators using hierarchical agglomerative clustering with the Euclidean distance metric.

The dendrogram (Figure 10) visually shows that Cluster 2 separates at an early stage of merging, demonstrating a high degree of homogeneity, whereas Cluster 3 is characterized by significant dispersion, reflecting the heterogeneity and vulnerability of enterprises in the eastern region.

The results showed a clear division of enterprises into three resilient groups, as presented in

Table 9. Cluster Characteristics and Regional Distribution of Enterprises.

Cluster	Main Characteristics	Predominant Region	Average KT	Interpretation
Cluster 1	Medium liquidity and digitalization, moderate risks	Central	0.81	Balanced strategies focused on maintaining stability
Cluster 2	High DLFO, AEFI, and RAS values	Western	0.89	Most resilient enterprises with access to external financing
Cluster 3	Low DLFO and RAS, high debt load	Eastern	0.49	Vulnerable enterprises requiring targeted support and credit guarantees

.

To identify relationships between financial resilience indicators, a Pearson correlation coefficient was calculated (

Table 10. Correlation Matrix between Key Enterprise Resilience Indicators.

Indicator	LR	DLFO	AEFI	RAS	SREL
LR	1.00	0.68	0.72	0.65	0.59
DLFO	0.68	1.00	0.74	0.83	0.61
AEFI	0.72	0.74	1.00	0.79	0.64
RAS	0.65	0.83	0.79	1.00	0.58
SREL	0.59	0.61	0.64	0.58	1.00

Interpretation: The strongest positive correlation is observed between DLFO and RAS (r = 0.83), confirming the interdependence of long-term financial security and profitability. The significant correlation between AEFI and RAS (r = 0.79) indicates the important role of access to external financing. The moderate correlation of SREL with other indicators shows that social engagement enhances resilience but is not a determining factor.

).

To analyze the spatial heterogeneity of small business financial resilience under a military crisis,

Table 11. Summary of Ukrainian Regions by Financial Resilience and Risk Exposure.

Region	Average KT	Average RRE	Risk Level	Characteristics
Western	0.89	0.15	Low	High resilience, active digitalization, EU and grant program support
Central	0.81	0.37	Medium	Balanced indicators but dependent on domestic demand
Eastern	0.49	0.68	High	Low resilience, significant impact of military risks and resource shortages

was created, summarizing average values of the taxonomic resilience coefficient (KT) and the regional risk exposure index (RRE) for the three macroregions of Ukraine. This approach allowed for comparing the level of enterprise resilience with the intensity of external threats and identifying territorial differences in the ability of businesses to adapt to crisis conditions. A complete dataset of all small enterprises with full financial indicators and cluster classification is provided in Appendix A.

Table 11 demonstrates a pronounced gradient of financial resilience: from the western regions, characterized by high adaptability and institutional support, to the eastern regions, where the impact of military risks and resource shortages significantly undermines the ability of enterprises to recover. The results confirm the territorial asymmetry of economic resilience of small businesses in Ukraine and are consistent with the findings of cluster and taxonomic analyses, indicating the need for regionally differentiated entrepreneurship support strategies in the post-war period.

Based on the results of cluster and taxonomic analyses (KT, RRE), targeted recommendations were developed for each type of enterprise to establish a financial risk management system aimed at enhancing SME resilience to external shocks, optimizing liquidity flows, and adapting to prolonged instability (

Table 12. Recommendations for Financial Risk Management System to Support SMEs under Prolonged Instability.

Cluster	Main Enterprise Characteristics	Key Risks	Strategic Priorities	Risk Management System Recommendations
Cluster 1—Medium Resilience (Central Region, KT ≈ 0.81)	Balanced liquidity and debt indicators; moderate digitalization; high social engagement	Dependence on domestic demand; limited access to external financing	Increase financial flexibility and operational efficiency	Develop stress-testing systems to assess the impact of reduced domestic demand; implement ERP systems for financial planning; create reserve funds to cover cash gaps; strengthen ties with regional funds and business incubators
Cluster 2—High Resilience (Western Region, KT ≈ 0.89)	High liquidity, active digitalization, high RAS, access to grants and credit	Currency fluctuations, rising energy costs	Strengthen innovation and export resilience	Create a multi-level monitoring system for currency risks; implement contract hedging and supplier diversification; develop digital platforms to monitor cash flows; participate in European grant and acceleration programs (Horizon Europe, EBRD, USAID)
Cluster 3—Low Resilience (Eastern Region, KT ≈ 0.49)	Low access to financing, high debt load, low liquidity, high regional risk exposure	Loss of markets, logistical disruptions, infrastructure damage	Financial recovery and cooperative integration	Introduce government guarantees for SME loans and targeted microfinance programs; develop cooperative networks for joint resource use; apply digital tools for supply chain and inventory management; create regional crisis consulting centers at OVAs and Chambers of Commerce

).

The recommendations reflect the need for a differentiated risk management approach: enterprises in the western regions should focus on innovation and digital financial control, central regions on stabilizing domestic demand and flexibility, and eastern regions on financial rehabilitation and institutional support.

4. Discussion

The findings of this study indicate that the financial resilience of small enterprises operating under the conditions of military conflict is a multidimensional phenomenon shaped by the interaction of internal resource capacities and the broader institutional environment. The interpretation of the identified patterns aligns with the principles of the Conservation of Resources (COR) Theory by Hobfoll et al. (2018), which emphasizes that an entity’s ability to preserve, accumulate, and efficiently allocate resources under threat is associated with higher adaptive resilience. In the context of Ukrainian SMEs, these resources comprise financial, organizational, human, and social capital, all of which contribute to maintaining operational continuity during prolonged instability.

Although robustness checks were performed, the results remain correlational and may be affected by unobserved confounders such as firm size or infrastructure access. These limitations should be considered when interpreting the findings.

According to the concept of entrepreneurial resilience by Lanivich (2013), resilience in a crisis environment is not a static property but a dynamic capacity for adaptation and strategic rethinking under external shocks. This is illustrated in the present study through the identification of statistically observable differences between clusters of enterprises, which reflect distinct response patterns associated with varying levels of exposure to military risks. For enterprises in western regions, digitalization and access to external financial resources appear to be associated with higher resilience levels, whereas for eastern enterprises, liquidity constraints and rising operational costs were critical. These patterns are broadly aligned with conceptual arguments by Williams and Shepherd (2016), though the present study identifies associations rather than causal mechanisms. The interpretation of these differences aligns with resilience theory, which identifies liquidity buffers, operational flexibility, and balanced leverage as factors typically associated with an enterprise’s capacity to endure external shocks. Prior empirical studies on SME crisis adaptation also demonstrate that firms with diversified financial structures and stable cash-flow positions exhibit higher functional stability during disruptions. This theoretical grounding provides a conceptual frame for interpreting the identified cluster patterns without implying causal relationships. Sensitivity and robustness checks, including KT stability analysis under alternative weighting schemes, further confirm that the results remain consistent across varying data conditions (Appendix H).

The psychological aspect of entrepreneurial behavior under threat, described by Sardeshmukh et al. (2018), confirms that stress caused by uncertainty and loss of control significantly influences managerial decisions of small business owners. The observed regional differences in strategies may reflect differing degrees of resource depletion and recovery effectiveness. Similar results are reported by Haynie and Shepherd (2010), emphasizing that cognitive flexibility and rapid business model adaptation are key survival factors in highly uncertain environments.

The identified differences in adaptation levels between enterprises in various regions of Ukraine can also be explained through the lens of research by Benmelech and Frydman (2014) and Kolodiziev et al. (2024a) on the influence of military experience and crisis leadership on managerial behavior. Recent studies further indicate that managerial adaptability, including proactive strategic decision-making and resource reallocation, significantly enhances enterprise resilience under crisis conditions (Odoch et al., 2025). Entrepreneurs with experience in resource-constrained or high-uncertainty conditions tend to take more risks and adopt proactive strategies, which is particularly relevant for Ukrainian managers adapting their businesses to new geopolitical realities.

The comparison of resilience clusters can be interpreted in light of the concepts of business exit and recovery developed by DeTienne et al. (2014) and Wennberg et al. (2009). Enterprises with low resilience tend to adopt temporary activity conservation or partial market exit strategies, whereas resilient companies focus on diversification and the search for new forms of value creation through cooperation and social initiatives. Empirical evidence supports that firms engaging in proactive resource reallocation and diversification strategies are more likely to maintain financial stability under crisis conditions (Cefis et al., 2022). This suggests that even under wartime conditions, entrepreneurs aim not only to minimize losses but also to generate social and community benefits, strengthening ties with local communities and volunteer organizations.

Thus, the study demonstrates that the resilience of small enterprises in Ukraine under military crisis conditions is formed at the intersection of three interrelated dimensions:

(1)

Resource potential (financial, human, and organizational resources);

(2)

Managerial adaptability (digitalization, business model reorientation, access to external financing);

(3)

Social responsibility and engagement (integration into community and volunteer initiatives as a means of strengthening reputational and social capital).

The study’s findings are consistent with existing theories of resilience and entrepreneurial behavior but also extend them by demonstrating the influence of military factors as a system-forming element in shaping a new paradigm of SME financial risk management. This conclusion has practical significance for the development of government support programs for small businesses, anti-crisis financing mechanisms, and regional development policies aimed at reducing spatial inequality and strengthening financial resilience in the long term. It must be noted that the analysis identifies statistical associations rather than causal relationships; therefore, the interpretations reflect patterns consistent with theoretical frameworks but do not establish causality.

5. Conclusions

This study offers an evidence-based examination of the associations between financial indicators, regional risk exposure, and the observed resilience patterns of small enterprises in wartime Ukraine. The empirical results consistently show that the ability to sustain revenue flows, manage operational costs, and maintain a balanced capital structure is central to financial stability. These findings demonstrate that resilience appears to be closely associated with measurable financial behaviors rather than by contextual or descriptive factors.

A clear regional pattern also emerges from the analysis: western regions exhibit higher resilience, central regions hold a moderate position, and eastern frontline regions remain the most vulnerable. These differences may reflect variations in access to financing, digitalization intensity, and exposure to military risks. While these trends are statistically robust, they should be interpreted within the limits of secondary data availability and the specific timeframe (2022–2024).

The study’s main contribution lies in developing an integrated, evidence-driven perspective on SME financial resilience under wartime conditions. By combining factor, cluster, and taxonomic methods, the research demonstrates how multidimensional financial indicators can be systematically used to identify risk profiles and inform tailored support strategies. This approach provides a structured analytical basis that can inform the development of regionally differentiated financial risk-management strategies.

The implications of the findings suggest three areas that warrant policy attention for strengthening resilience:

(1)

Expanding access to financial instruments for enterprises in high-risk regions;

(2)

Encouraging digitalization of financial operations to stabilize cash-flow management; and

(3)

Supporting adaptive practices that help SMEs sustain operations under prolonged instability.

These conclusions reflect the empirical evidence obtained in the study without extending beyond the scope of the available data. Further research should incorporate micro-level and primary data to deepen understanding of the mechanisms behind SME adaptation in extreme environments. Given that the study relies on correlational and classification techniques, the patterns identified should be interpreted as associations rather than causal effects.