Финансовые факторы деятельности публичных компаний во время и после пандемии COVID: опыт Шри-Ланки

DOI:

The financial performance of publicly listed firms holds critical significance for several stakeholders, including investors, financial institutions, employees, regulatory bodies, the general public, and academics. These organizations play a vital role in the economy, not only through the generation of employment opportunities but also a significant contribution to public revenue through tax payments. In light of recent global events, particularly the adverse impacts of the COVID-19 pandemic, concerns regarding the financial stability of these firms have intensified among shareholders. This has underscored the urgent need for a thorough examination of the factors that influence financial performance [1,2].

In Sri Lanka, the detrimental effects of the pandemic have been compounded by a pre-existing financial crisis plaguing the nation. This compounded economic distress highlights the necessity for focused research aimed at understanding the determinants of financial performance within this specific context. The current study seeks to probe critical factors influencing financial performance, such as liquidity management, capital structure, and firm size, while also exploring the distinct repercussions of COVID-19 on financial outcomes [3,4]. By analyzing firm-year observations from 2020 to 2023, this research endeavors to furnish a comprehensive understanding of how these determinants interacted during the pandemic and in the recovery phase that followed.

Existing literature illustrates that effective liquidity management, an appropriate capital structure, and optimal firm size are fundamental to achieving financial success [5,6]. While numerous studies have examined the interrelationships among these variables and their collective impact on financial performance, research specifically considering the impact of the COVID-19 pandemic on these dynamics remains relatively scarce [7]. Moreover, there exists a significant gap in the research literature that compares the determinants of financial performance during the pandemic with those observed in the post-pandemic period, particularly within the Sri Lankan context. Another notable gap in existing studies is the lack of standardized benchmarks for current and leverage ratios, an omission that this research aims to address [8].

Through this research, this study aims to advance the understanding of the intricate relationships between various determinants of financial performance, particularly in the face of an unprecedented global crisis. The insights gleaned from this study are expected to assist firms in formulating effective recovery strategies and contribute towards fostering a stable economic environment that benefits all stakeholders involved.

The structure of the following sections in this paper will detail the research methodology employed, present the findings derived from the analysis, and engage in an analytical discussion to interpret the results and their implications for listed firms in Sri Lanka as they navigate the post-pandemic economic landscape. The research findings will also aim to identify innovative approaches and strategies that firms can adopt to improve their financial resilience and performance moving forward.

LITERATURE REVIEW AND HYPOTHESIS DEVELOPMENT

The formulation of hypotheses in this study is grounded in a review of both theoretical frameworks and empirical evidence drawn from previous research. Scholars have consistently emphasized the importance of internal and external factors—such as liquidity management, capital structure, and firmspecific characteristics—in shaping a firm’s financial outcomes. Additionally, recent studies have begun to explore the unprecedented influence of the COVID-19 pandemic on firm performance.

Drawing upon the insights from prior literature, including the works of Serrasqueiro et al. [9], Hamidon and Ranjani [10], Chadha and Sharma [11], El-Sayed Ebaid [12], Olufade et al. [13], Abuzayed [14], and Sajiyya et al. [15], this study aims to assess how specific financial and operational determinants affected the financial performance of listed firms in Sri Lanka during the pandemic and in the subsequent recovery period.

The following hypotheses have been developed to empirically test these relationships using panel data from non-financial listed companies:

• •   H1: Liquidity management (LM) is significantly

associated with financial performance (measured by ROA).

• •   H2: Capital structure (CS) is significantly

associated with financial performance (ROA).

• •   H3: Firm size (FSIZE) is significantly associated

with financial performance (ROA).

• •   H4: The COVID-19 pandemic has a statistically

significant negative impact on financial performance (ROA).

These hypotheses are tested using a multivariate regression framework, allowing for the control of additional firm-level and macroeconomic variables that may influence financial outcomes.

METHODOLOGY

Sample and data collection

This study focuses on listed non-financial firms operating on the Colombo Stock Exchange (CSE) over the period from 2020 to 2023. The selection of this timeframe was driven by the need to capture both the effects during the COVID-19 pandemic and the post-pandemic recovery period. Financial institutions were excluded from the analysis due to their unique regulatory environment and capital structure characteristics, which differ significantly from nonfinancial entities.

The initial dataset comprised all listed firms during the study period. However, firms with incomplete or missing data particularly on control variables such as audit committee characteristics and firm-specific financial indicators were excluded to ensure consistency and reliability in statistical analysis. After applying these criteria, the final sample consisted of 181 non-financial listed firms, resulting in 595 firm-year observations across the four years.

The data required for this study were manually extracted from audited annual reports published on the official websites of respective companies and from the CSE online repository. Macroeconomic variables, such as GDP growth rate, were sourced from the Central Bank of Sri Lanka.

The panel nature of the data allows for a more robust analysis of temporal effects, especially in assessing variations in firm performance before, during, and after the pandemic years.

Research model

To examine the determinants of financial performance among listed non-financial firms in Sri Lanka, this study employs an Ordinary Least Squares (OLS) regression model. The model is designed to evaluate the influence of liquidity management, capital structure, firm size, and the COVID-19 period on firm performance, as measured by return on assets (ROA). Several control variables were also included to account for firm-specific and macroeconomic influences.

The regression model is specified as follows:

ROA = β 0 + β 1 LM + β 2 CS + β 3 FSIZE + β 4 COVID +

β 5 OCF + β 6 FAGE + β 7 ATURN + β 8 ACS + β 9 ACI +

β 10 AFSIZE + β 11 GDPG + Year Fixed Effects + ε i

In this model:

• •   ROA serves  as the dependent  variable,

representing firm performance.

• •   LM  (liquidity management), CS (capital

structure), FSIZE (firm size), and COVID (pandemic period) are the key explanatory variables.

• •    The model includes control variables such as operating cash flow (OCF), firm age (FAGE), asset turnover (ATURN), audit committee size (ACS), audit committee independence (ACI), audit firm size (AFSIZE), and national GDP growth (GDPG) to reduce omitted variable bias.

• •    Year fixed effects are incorporated to account for any unobserved temporal heterogeneity affecting all firms similarly across years.

Each variable is defined in detail in Table 1.

Table 1: definition of variables.

Variables	Definition
Dependent variable: ROA	Return on Assets is calculated as net profit after tax divided by total assets.
Independent variables:
LM	Liquidity Management, measured by the current ratio (current assets / current liabilities).
CS	Capital Structure, measured by the leverage ratio (total non-current liabilities/equity).
FSIZE	Firm Size, measured by the natural logarithm of total assets.
COVID	Dummy variable indicating COVID period; 1 for years 2020 and 2021, 0 otherwise.
Control Variables
OCF	Operating Cash Flow, measured as cash flow from operations divided by total assets.
FAGE	Firm Age, defined as the number of years since the company’s incorporation.

ATURN	Asset Turnover, calculated as total sales divided by total assets.
ACS	Audit Committee Size, measured by the number of audit committee members.
ACI	Audit Committee Independence, measured as the percentage of independent members in the audit committee.
AFSIZE	Audit Firm Size, dummy variable; 1 if the auditor is a Big Four firm, 0 otherwise.
GDPG	Annual GDP growth rate of Sri Lanka.

Data Analysis and robustness check

The analytical process began with the computation of descriptive statistics to summarize the central tendency, dispersion, and distribution of the key variables in the dataset. This step provided an initial understanding of the data and ensured the appropriateness of the variables for further econometric analysis.

Next, a series of independent sample t-tests were conducted to examine mean differences in firm performance across the COVID and post-COVID periods. Correlation analysis was also performed to identify the direction and strength of linear relationships between the independent variables and the dependent variable (ROA), as well as to preliminarily assess the risk of multicollinearity.

To formally test the hypothesized relationships, this study employed Ordinary Least Squares (OLS) regression analysis with year fixed effects. The inclusion of fixed effects controlled for unobserved time-specific influences that could bias the regression estimates.

Prior to model estimation, the Variance Inflation Factor (VIF) was computed for each predictor to detect the presence of multicollinearity. All variables exhibited acceptable VIF values, indicating that multicollinearity was not a significant concern.

To enhance the internal validity of the findings and reduce potential sample imbalances, entropy balancing was applied. This method reweights the sample to achieve covariate balance between treated and control groups particularly for the COVID vs. post-COVID periods, thus reducing bias and improving the precision of the regression estimates.

Following entropy balancing, the weighted dataset was used to re-estimate the regression model. This entropy-balanced regression analysis allowed for a more accurate assessment of the impact of the key determinants on firm performance, mitigating confounding effects and enhancing the robustness of the empirical results.

RESULTS AND DISCUSSION

Descriptive statistics

The descriptive statistics offer an overview of the firm-level characteristics in the sample and reveal considerable variation across several financial and governance-related indicators. As shown in Table 2, the average Return on Assets (ROA) is 4.79%, with a relatively high standard deviation of 13.71%, suggesting a wide dispersion in profitability among listed non-financial firms in Sri Lanka. The median ROA value of 3.69% further indicates that while some firms experienced negative or low profitability, a substantial number performed reasonably well during the study period.

Liquidity Management (LM), as measured by the current ratio, has a mean value of 3.15, pointing to a generally healthy liquidity position across the sample. However, the large standard deviation (5.29) signals significant heterogeneity in firms’ ability to meet short-term obligations. Capital Structure (CS), represented by the leverage ratio, has a mean of 0.46, which implies that on average, firms finance around 46% of their operations through long-term debt. The relatively large standard deviation of 2.07 suggests diverse financial strategies and risk exposures among firms.

Firm Size (FSIZE), measured as the natural logarithm of total assets, averages 21.94, indicating a prevalence of relatively large firms in the sample. The COVID-19 variable is constructed as a binary indicator, with a mean value of 0.50, confirming an even split between observations during (2020–2021) and after (2022–2023) the pandemic, thereby facilitating comparative analysis.

Operating Cash Flow (OCF) averages 0.26, but the standard deviation of 4.11 reflects major differences in cash flow efficiency and liquidity management across firms. The average Firm Age (FAGE) is approximately 45 years, suggesting that the dataset largely comprises well-established entities, although age ranges significantly (standard deviation = 27.43).

Asset Turnover (ATURN), with a mean of 0.59, shows broad variability in operational efficiency among firms, pointing to differing capabilities in utilizing assets to generate revenue. Audit Committee Size (ACS) and Audit Committee Independence (ACI) are fairly consistent, with mean values of 3.20 members and 83.13% independence, respectively, reflecting adherence to good governance practices. Audit Firm Size (AFSIZE), with a mean of 0.84, reveals that a majority of firms are audited by one of the Big Four accounting firms. Finally, the average GDP growth rate (GDPG) during the study period is –2.48%, underscoring the economic contraction experienced in Sri Lanka due to the pandemic and related macroeconomic disruptions

Table 2: Descriptive statistics

Variables	Observations	Q1	Mean	Std.Dev	Median	Q3
ROA	595	-1.62	4.79	13.71	3.69	9.51
LM	595	0.70	3.15	5.29	1.27	2.86
CS	595	0.06	0.46	2.07	0.17	0.42
FSIZE	595	21.29	21.94	1.80	22.10	22.97
COVID	595	0.00	0.50	0.50	1.00	1.00
OCF	595	-0.02	0.26	4.11	0.02	0.08
FAGE	595	29.00	45.37	27.43	40.00	55.00
ATURNOVER	595	0.07	0.59	1.61	0.36	0.78
ACS	595	3.00	3.20	0.74	3.00	3.00
ACI	595	66.67	83.13	17.38	80.00	100.00
AFSIZE	595	1.00	0.84	0.36	1.00	1.00
GDPG	595	-4.62	-2.48	4.29	-2.30	4.21

T-test analysis

To explore the effect of liquidity on financial performance, a two-sample t-test with equal variances was conducted. The sample was divided into two groups based on the median current ratio (1.27). Group 1 consists of firms with liquidity ratios above the median (high liquidity), while Group 0 includes those below the median (low liquidity).

As shown in Table 3, firms in Group 1 reported a significantly higher mean ROA of 9.60%, compared to a mean ROA of –0.18% for firms in Group 0. The mean difference in ROA between the two groups is 9.78 percentage points, and the t-statistic is –9.31, with a p-value of 0.000, indicating that the difference is statistically significant at the 1% level.

These findings suggest that firms with stronger liquidity positions were better able to sustain or enhance their financial performance during the studied period. This supports Hypothesis H1, which posits a significant relationship between liquidity management and firm performance.

Table 3: Two-sample t-test with equal variances

Group	Obs	Mean	Std. err.	Std. dev.	[95% conf. interval]
0	293	-0.18	0.59	10.03	-1.33	0.97
1	302	9.60	0.86	15.03	7.90	11.31
Combined	595	4.79	0.56	13.71	3.68	5.89
Diff		18.33	1.05		-11.85	-7.72
T						-9.31
Pr(|T| > |t|)						0.00

Correlation Analysis

The correlation matrix presented in Table 4 provides preliminary insights into the associations between the key variables in this study. The results indicate a significant positive correlation between Return on Assets (ROA) and Liquidity Management (LM) (r = 0.36), suggesting that firms with stronger liquidity positions tend to exhibit better financial performance. This finding reinforces the theoretical assertion that effective short-term asset and liability management enhances profitability.

A significant negative correlation is observed between ROA and the COVID-19 dummy variable (r = –0.13), implying that the pandemic had a dampening effect on firm profitability, albeit with a modest correlation strength. Firm Size (FSIZE) also shows a positive and significant correlation with

ROA (r = 0.10), indicating that larger firms were better positioned to sustain performance during the period under review.

Other explanatory and control variables, including capital structure (CS), operating cash flow (OCF), and governance indicators, do not show statistically significant correlations with ROA. Importantly, the Variance Inflation Factor (VIF) values for all variables are below 2, suggesting that multicollinearity is not a concern in this dataset. The low VIF scores further confirm the suitability and reliability of the regression model used in subsequent analysis.

Table 4: Correlation matrix.

Variables	1	2	3	4	5	6	7	8	9	10	11	12	VIF
ROA (1)	1.00
LM (2)	0.36	1.00											1.04
CS (3)	-0.06	-0.08	1.00										1.02
COVID (4)	-0.13	-0.04	-0.03	1.00									1.43
OCF (5)	0.05	0.03	-0.01	0.03	1.00								1.02
FSIZE (6)	0.10	-0.04	0.11	-0.08	0.01	1.00							1.10
FAGE (7)	-0.03	0.02	-0.08	-0.03	-0.03	-0.12	1.00						1.03
ACS (8)	0.01	-0.09	0.01	0.00	-0.09	0.17	-0.03	1.00					1.13
ACI (9)	0.00	-0.06	0.00	0.00	0.03	0.01	-0.03	-0.23	1.00				1.08
AFSIZE (10)	0.05	-0.06	0.04	0.01	0.03	0.20	0.01	0.14	-0.08	1.00			1.09
ATURN (11)	0.03	-0.09	0.00	0.02	0.01	0.01	-0.07	-0.06	-0.02	-0.14	1.00		1.04
GDPG (12)	-0.04	-0.03	-0.03	0.54	0.07	-0.03	0.00	0.02	0.02	0.00	-0.01	1.00	1.43

Note(s): Figures in bold represent significance at 1%, and figures in bold and italic represent significance at 5%.

Regression analysis

To evaluate the relationships between financial and governance variables and firm performance, we employed a multiple regression model with Return on Assets (ROA) as the dependent variable. The model is specified as follows:

ROA = β 0 + β 1 LM + β 2 CS + β 3 FSIZE + β 4 COVID +

β 5 OCF + β 6 FAGE + β 7 ATURN + β 8 ACS + β 9 ACI +

β 10 AFSIZE + β 11 GDPG + Year Fixed Effects + ε i where:

• •    ROA represents the dependent variable, a measure of firm performance.

• •    LM (Liquidity Management), CS (Capital Structure), FSIZE (Firm Size), and COVID (Pandemic Period) are key explanatory variables.

• •    OCF (Operating Cash Flow), FAGE (Firm Age), ATURN (Asset Turnover), ACS (Audit Committee Size), ACI (Audit Committee Independence), AFSIZE (Audit Firm Size), and GDPG (National GDP Growth) are included as control variables to mitigate omitted variable bias.

• •    Year Fixed Effects account for unobserved temporal heterogeneity that may affect all firms similarly across years.

Model interpretation

The regression coefficients (β) reflect the impact of each independent variable on the dependent variable,

ROA, while holding all other factors constant. For instance, a coefficient of β1 for LM means that a one-unit change in liquidity management would result in a β1-unit change in ROA, assuming other variables are unchanged.

The error term (ε i ) captures the unexplained variation in ROA, accounting for all factors not directly included in the model that might influence firm performance.

Regression results

The regression analysis was conducted for three sub-samples: the Full Sample, COVID-19 Period, and Post-COVID Period. The results are summarized in Table 5, and the key findings are as follows:

• •    Full Sample:

Liquidity Management (LM) shows a significant and positive relationship with ROA (Coef = 0.96, p < 0.01). This result supports Hypothesis H1 and highlights the importance of maintaining strong liquidity, especially during times of financial uncertainty. Firms with better liquidity management are more likely to perform well in terms of ROA.

Firm Size (FSIZE) also positively impacts ROA (Coef = 0.68, p = 0.03), confirming Hypothesis H3. This result is consistent with previous literature that suggests larger firms enjoy economies of scale and greater financial resilience, which contribute to higher profitability.

COVID has a significant negative effect on ROA (Coef = -3.63, p = 0.01), supporting Hypothesis H4.

The pandemic led to a decline in firm performance, as expected, due to disruptions in business operations across industries.

• •    COVID-19 Period:

During the COVID-19 period, Liquidity Management (LM) continued to have a positive effect on ROA, though the relationship weakened (Coef = 0.60, p < 0.01). This suggests that while liquidity was important during the pandemic, its role in determining firm performance was less pronounced compared to the full sample period.

Audit Committee Size (ACS) showed a significant and positive impact on ROA (Coef = 1.71, p = 0.04), indicating that governance structures, particularly audit committees, played an important role in maintaining firm performance during the crisis.

• •    Post-COVID Period:

Liquidity Management (LM) exhibited an even stronger positive relationship with ROA (Coef = 1.36, p < 0.01) in the post-COVID period. This suggests that liquidity management became even more critical during the recovery phase, as firms focused on regaining stability.

Audit Firm Size (AFSIZE) and Asset Turnover (ATURN) emerged as significant predictors of ROA during the post-COVID period. Larger audit firms and higher asset turnover rates were associated with better firm performance.

DISCUSSION

The regression results suggest that liquidity management and firm size are crucial determinants of firm performance. Liquidity management, in particular, emerged as a strong predictor of ROA across all periods, emphasizing the strategic importance of liquidity both during and after the COVID-19 crisis. Additionally, governance factors such as Audit Committee Size (ACS) and Audit Firm Size (AFSIZE) were found to play significant roles in maintaining firm performance, especially in crisis situations like the pandemic.

These findings have important implications for corporate governance and financial management, particularly during periods of economic uncertainty. Firms with robust liquidity management strategies and strong governance structures tend to perform better in terms of profitability, as reflected in the positive relationships with ROA.

Table 5: Regression results.

Variables	Full Sample			During COVID			Post COVID
	Coef	T	P>|t|	Coef	T	P>|t|	Coef	t	P>|t|
LM	0.96	9.63	0.00	0.60	4.95	0.00	1.36	9.20	0.00
CS	-0.33	-1.30	0.20	0.09	0.26	0.79	-0.53	-1.50	0.14
COVID	-3.63	-2.85	0.01
OCF	0.12	0.95	0.35	0.11	0.98	0.33	0.06	0.20	0.84
FSIZE	0.68	2.25	0.03	0.24	0.74	0.46	0.93	1.91	0.06
FAGE	-0.01	-0.71	0.48	-0.01	-0.65	0.51	-0.01	-0.20	0.84
ACS	0.69	0.92	0.36	1.71	2.09	0.04	0.08	0.07	0.95
ACI	0.02	0.73	0.47	0.03	0.81	0.42	0.00	0.07	0.94
AFSIZE	2.58	1.73	0.08	0.70	0.42	0.68	4.73	2.03	0.04
ATURN	0.66	2.02	0.04	0.04	0.14	0.89	7.89	5.82	0.00
GDPG	0.16	0.55	0.59	0.11	0.84	0.40	-0.02	-0.05	0.96
Constant	-16.79	-2.22	0.03	-11.86	-1.47	0.14	-27.09	-2.30	0.02
Observations	595.00			300.00			295
Prob > F	0.00			0.00			0.00
Adj R-squared	0.15			0.06			0.15
Year fixed effects	Yes			Yes			Yes

Robustness check

To validate the robustness of the regression findings, an entropy balancing technique was employed. This approach reweights the dataset to ensure that the control variables are balanced between the treatment group (high-liquidity firms) and the control group (low-liquidity firms).

As shown in Panels A and B of Table 6, the distribution of means, variances, and skewnesses across the groups becomes highly comparable after balancing, thus eliminating observable confounding biases. The entropy-balanced regression model (Panel C) largely confirms the baseline regression results.

Liquidity management remains a significant and positive determinant of ROA (Coef = 9.21, p < 0.01). Interestingly, capital structure becomes statistically significant (Coef = 0.86, p = 0.01) after balancing, suggesting that its effect may have been masked by covariate imbalance in the unadjusted sample.

Table 6: Entropy balancing results.

Panel A: Before entropy balancing

	Treatment group: Early Childhood Matured Students			Control group: Early Childhood I mmature Students
	Mean	Variance	Skewness	Mean	Variance	Skewness
CS	0.23	0.89	-3.38	0.70	7.67	6.53
COVID	0.46	0.25	0.15	0.55	0.25	-0.19
OCF	0.49	31.46	14.03	0.03	1.80	-7.86
FSIZE	21.95	2.65	-1.03	21.92	3.87	-1.26
FAGE	44.83	761.20	1.63	45.93	745.60	1.44
ACS	3.16	0.57	0.74	3.25	0.51	0.83
ACI	82.64	295.10	-0.19	83.64	310.00	-0.40
AFSIZE	0.86	0.12	-2.09	0.83	0.14	-1.72
ATURN	0.55	0.36	2.31	0.64	4.91	14.09
GDPG	-2.52	18.16	0.56	-2.45	18.80	0.58

Panel B: After entropy balancing

	Treatment group: Early Childhood Matured Students			Control group: Early Childhood Immature Students
	Mean	Variance	Skewness	Mean	Variance	Skewness
CS	0.23	0.89	-3.38	0.23	2.96	-1.88
COVID	0.46	0.25	0.15	0.46	0.25	0.15
OCF	0.49	31.46	14.03	0.49	3.45	3.61
FSIZE	21.95	2.65	-1.03	21.95	3.54	-1.11
FAGE	44.83	761.20	1.63	44.83	695.80	1.59
ACS	3.16	0.57	0.74	3.16	0.46	0.88
ACI	82.64	295.10	-0.19	82.64	307.50	-0.26
AFSIZE	0.86	0.12	-2.09	0.86	0.12	-2.09
ATURN	0.55	0.36	2.31	0.55	1.74	20.03
GDPG	-2.52	18.16	0.56	-2.52	18.79	0.55

Panel C: Entropy balancing regression results

	Coefficient	T	P>|t|
LM	9.21	7.93	0.00
CS	0.86	2.55	0.01
COVID	-2.43	-1.71	0.09
OCF	0.26	1.23	0.22
FSIZE	0.56	1.63	0.10
FAGE	-0.01	-0.54	0.59
ACS	0.53	0.52	0.60
ACI	-0.02	-0.71	0.48
AFSIZE	1.24	0.74	0.46
ATURN	1.25	1.27	0.21
GDPG	-0.05	-0.14	0.89
Constant	-12.70	-1.59	0.11
Observations	595
Prob > F	0.00
Adj R-squared	0.15
Year fixed effects	Yes

CONCLUSION

This study examined the key determinants of financial performance among listed non-financial firms in Sri Lanka from 2020 to 2023, focusing on liquidity management, capital structure, firm size, and the impact of the COVID-19 pandemic. Analyzing 595 firm-year observations, the findings show that liquidity management was the most significant predictor of profitability, especially during the post-COVID recovery phase.

Additionally, larger firms demonstrated better financial performance post-pandemic, while capital structure did not significantly affect profitability.

The study highlights the importance of effective liquidity management for corporate decision-makers, suggesting it should be prioritized for long-term financial stability. Policymakers are encouraged to support sound liquidity practices in emerging markets, which are more susceptible to economic fluctuations.

Limitations include the focus on listed firms in Sri Lanka, restricting generalizability to smaller private firms or other regions. The use of Return on Assets (ROA) as the sole performance measure may overlook other factors like shareholder value. The relatively short post-COVID analysis period and the lack of sector-specific disaggregation could also affect the results.

Future research could benefit from broader performance indicators, a longer post-pandemic analysis, and cross-country comparisons to gain deeper insights into how firms adapt their financial strategies in response to systemic shocks.