Firm Profitability and Economic Crises: The Non-Linear Role of the Cash Conversion Cycle

Abstract

This study investigates the non-linear effect of the cash conversion cycle (CCC) on a firm’s profitability for a sample of 6072 firms from five countries (Germany, Spain, France, Great Britain, and Italy) from 2006 to 2015. Additionally, this study explores the sensitivity of economic crises to the non-linear effect of the CCC on a firm’s performance. This study employs fixed-effects unbalanced panel data and weighted least squares (due to heteroscedasticity) to examine a firm’s performance, using return on assets (ROA) to measure profitability. The cash conversion cycle, financial leverage, size, and tangibility are independent variables. The results of this study show that the effect of the cash conversion cycle on firms’ performance is an inverted U-shape (non-linear). It also shows that the economic conditions vis-à-vis crises influence firm performance. This study found the optimal number of the CCC to be 90 days for the entire sample, 85 days for the non-crisis period, and 92 days for the crisis period. It also finds that the marginal effect of the CCC on ROA is 3.9 times higher during economic crises versus non-economic crisis periods. This study contributes to the existing working capital management literature by examining the non-linear effect of the cash conversion cycle on profitability and the sensitivity of these effects during economic crises. Thus, empirical evidence can serve scholars, business policymakers, and corporate finance professionals in managing their working capital strategically.

1. Introduction

This study explores the effect of the cash conversion cycle (CCC) on firm profitability, the non-linearity of the CCC effect, and how this effect varies with a country’s overall economic conditions. Broadly defined, the CCC is the time it takes for a firm to convert cash outflows (from purchases) into cash inflows (from sales). Working capital management, a significant component of the CCC, is a complex process due to the intricate relationship between current assets and liabilities. Creating an optimal structure of these components and managing their interrelationship remains a crucial responsibility for financial managers. The complexity is heightened by the opportunity cost of holding current assets instead of investing in non-current assets. This trade-off between working capital (mainly cash) and other assets varies temporally and cross-sectionally, influenced by market conditions and firm-specific factors.

The relationship between the CCC and firm performance has been widely studied, yet the findings remain mixed and unsettled. Numerous studies (e.g., Nobanee et al., 2011; Wang, 2019; Yazdanfar & Öhman, 2014; Zeidan & Shapir, 2017; Deari & Palomba, 2024) have found a negative relationship between the CCC and firm performance, arguing that shorter CCCs improve working capital management and, consequently, profitability. Alternatively, some studies (e.g., Panigrahi, 2013; Zakari & Saidu, 2016) have found a positive relationship, suggesting that a longer CCC is a buffer during financial hardships and provides insurance against delayed receivables. Additionally, C. H. Chen et al. (2022) demonstrated significant evidence of the CCC’s impact on profitability, further highlighting the variability in findings.

Scholars (e.g., Fernandes et al., 2021; Huynh et al., 2025; Zeidan & Shapir, 2017) have also investigated the non-linear relationship between the CCC and profitability. For example, Fernandes et al. (2021) identified an inverted U-shaped relationship, arguing that an optimal CCC level exists where profitability is maximized. According to their findings, firms with excessively short CCCs may miss opportunities for higher performance, while firms with excessively long CCCs experience diminished profitability. Supporting studies (e.g., Deari & Palomba, 2024; Rey-Ares et al., 2021; Zeidan & Shapir, 2017) emphasize the importance of the CCC as a dynamic metric reflecting the relationship between the debtors’ collection period, creditors’ deferral period, and inventory conversion period. As a versatile tool for analyzing working capital management (Anser & Malik, 2013), the CCC enables firms to adopt aggressive, neutral, or conservative strategies based on their business needs, targets, and operating environments. Additionally, working capital policies influence broader financial decisions, including how firms are financed.

The role of the CCC in economic uncertainty, such as during economic crises (measured by negative real GDP growth), has been explored less. A few exceptions include studies by H. C. Chen et al. (2018), C.-C. Chang and Yang (2022), and Akgün and Memiş Karataş (2021), highlighting how cash reserves can serve as critical buffers during economic crises, mitigating adverse effects on firm performance. While H. C. Chen et al. (2018) and C.-C. Chang and Yang (2022) focus on the effect of cash holding, Akgün and Memiş Karataş (2021) study the effect of the 2008 financial crisis on the relationship between working capital management and business performance. H. C. Chen et al. (2018), using data for the 2000 dot-com crash and 2018 financial crises, find that pre-saved cash mitigates financial shock effects, adding to the precautionary motive of holding money. C.-C. Chang and Yang (2022), using data from 38 countries, find similar results across different crises: currency, banking, external, and domestic debt crises. Akgün and Memiş Karataş (2021) use European countries’ data for the 2003–2012 and find that the 2008 financial crisis significantly negatively affected return on assets. These studies suggest that the relationship between the CCC and profitability may shift during economic downturns, depending on the firm’s ability to manage liquidity effectively and working capital.

Two gaps in the literature emerge from the empirical investigations on the CCC and firm profitability. First, the direction of the relationship between the CCC and firm performance remains unsettled as there are studies supporting a linear and negative relationship and studies that support a non-linear U-shaped relationship. Second, limited research examines the sensitivity of the CCC’s effect on firm performance during periods of economic uncertainty. Despite extensive research, these gaps remain unexplored in detail.

This paper addresses these issues by investigating two research questions: (1) whether the relationship between the CCC and profitability is a non-linear U-shaped relationship and (2) whether the CCC–profitability relationship is sensitive to a country’s economic conditions, particularly during crises. By addressing these questions, this study seeks to contribute to the existing literature by exploring the non-linear effect of the CCC on profitability and the sensitivity of this relationship to economic uncertainty. This study focuses on firms operating in Germany, Spain, France, Great Britain, and Italy from 2006 to 2015, encompassing the 2008 financial crisis and its aftermath. Including these countries provides a diverse economic context to examine the role of the CCC under varying economic conditions. This study uses unbalanced (the number of periods is not the same for all firms) panel data analysis to find strong evidence of the cash conversion cycle’s non-linear effect on return on assets and that these effects are exacerbated during economic crises. On average, the optimal cash conversion cycle is 90 days (across all conditions), 45 days during periods with no economic crises, and 92 days during economic crises.

The rest of the paper is organized as follows: Section 2 presents a literature review of previous studies and hypothesis development; Section 3 outlines the methodology and data inspections; Section 4 presents empirical findings and discussions; and Section 5 offers conclusions and policy implications.

2. Literature Review and Hypothesis Development

In this section, we will review the related studies regarding the relationship between the CCC and firm profitability, the non-linearity of this effect, and the impact of economic crises on this relationship. We will then use these related studies to develop the hypothesis. We will discuss and state the hypothesis along with relevant literature. We discuss the literature on the effect of the CCC on a firm’s performance along four strands. First, we examine studies focusing on the CCC–performance relationship of listed firms. Second, we explore studies that shed light on this relationship for industry-specific reasons. Third, we review the literature on the non-linearity of the CCC’s effect on firms’ profitability. Finally, we will review the literature on the impact of economic and financial crises on enhancing the CCC’s effect on firm performance.

2.1. The CCC and Firm’s Performance: Review of Studies on Listed Firms

We first examine the evidence from listed companies in developed countries. The relationship between the CCC and profitability has been extensively studied in listed firms from developed markets. In the U.S. market, Wang (2019) analyzed data from 1976 to 2015 and found that firms with low CCCs generate higher returns than those with high CCCs, primarily due to the influence of inventory and receivables outstanding. Conversely, X. Chen et al. (2022) found no evidence of the CCC effect in the Chinese market over the sample period 2002–2019, noting that while receivables significantly predict returns, inventory outstanding does not.

The CCC is also a crucial indicator of operating efficiency in fundamental analysis. C. H. Chen et al. (2022) investigated the return predictability of the CCC and its components across global stock markets in 22 developed and 25 emerging countries. They found a negative and robust CCC return spread in their international sample. However, it was more moderate than Wang’s (2019) findings in the U.S. Additionally, they found that each component of the CCC can significantly predict future stock returns, as evidenced by time-series abnormal returns or cross-sectional regressions.

Zeidan and Shapir (2017) support the idea that reducing the CCC can significantly enhance shareholder value. Their analysis of the Brazilian company MRV shows how operational changes reduced the CCC from 508 days in 2012 to 351 days in 2015, significantly lowering working capital requirements without negatively impacting operating margins or sales. This reduction increased profits and free cash flow to equity and led to higher stock prices, reinforcing the value-enhancing potential of effective CCC management. Similarly, Vahid et al. (2012) discovered a negative association between operating profitability and measures of working capital management, such as the average collection period, inventory turnover (in days), average payment period, CCC, and net trade cycle. These relationships are significant, except for the CCC. These results align with U.S.-based findings, suggesting that CCC reduction strategies in developed economies generally enhance financial performance.

Next, we looked at the evidence on the effect of the CCC on a firm’s performance in developing countries. Johan et al. (2024) found that achieving firm performance in the BRICS countries (Brazil, Russia, India, China, and South Africa) requires efficient CCC management. A shorter CCC helps manage cash flows effectively, avoiding over- or underinvestment in inventory, receivables, and debt costs. Using data from manufacturing firms in Egypt, Kenya, Nigeria, and South Africa for 2002–2009, Ukaegbu (2014) found a strong negative relationship between profitability, measured through net operating profit, and the CCC across different industrialization typologies. The study also found a positive relationship between profitability and inventory turnover and firm size, and an inverse relationship between accounts receivable and profitability. This suggests firms should have policies to accelerate accounts receivable collection to enhance profitability.

Some studies take a more comprehensive approach and focus on developing countries. For example, in the manufacturing sector, studies in Pakistan (Raheman et al., 2010; Anser & Malik, 2013) reveal that firms benefit from minimizing receivables and inventory while extending payables. Raheman et al. (2010) explored the impact of working capital management on the corporate performance of manufacturing companies in Pakistan. Their results showed that the CCC inversely impacts the firm’s net operating profit. Anser and Malik (2013) examined the effect of the CCC on profitability in Pakistan’s manufacturing sector. They developed a regression model using return on equity and return on assets as dependent variables, while firm size, debt ratio, and the CCC were independent variables. Their study revealed a negative relationship between the CCC and profitability measures, suggesting that firms should reduce the accounts collection and inventory selling period while extending the payment period to increase profitability. Linh and Mohanlingam (2018) observed similar dynamics in Thailand’s agriculture and food industries, where optimized CCC management enhanced profitability.

2.2. The CCC and Firm’s Performance: Review of Studies on Industry-Specific Firms

The second strand of literature concerns studies on the CCC and firms’ performance in the industry-specific. The role of cash as a component of working capital varies by industry. We first discuss such studies and evidence in developed countries. For example, Carnes et al. (2023) found that cash enables firms to obtain higher margins and shorten CCCs, which allows cash-rich firms to enjoy higher performance. However, the benefits of cash are contextual and vary based on the nature of ties with trading partners, as relational ties attenuate them. Fernandes et al. (2021) found a concave relationship between cash holdings and profitability in the banking sector, suggesting diminishing returns from excessive liquidity. Their analysis of banks from developed and emerging markets over 2000–2017 revealed that while higher cash holdings initially correlate with increased profitability, this relationship becomes negative at higher cash holdings. This inverted U-shaped pattern suggests that excess cash may lead to inefficiencies or lower returns beyond a certain point, highlighting the importance of optimizing rather than maximizing cash reserves. These patterns illustrate a trade-off between liquidity and operational efficiency, with non-linear dynamics often observed in cash-intensive industries. During economic downturns, H. C. Chen et al. (2018) noted that pre-saved cash helped financially constrained firms maintain investments, dividends, and acquisitions, emphasizing cash as a critical buffer during crises. However, Panigrahi (2013) reported positive CCC–profitability relationships in Indian cement firms, indicating industry-specific variations in optimal liquidity strategies. Industry-specific studies also provide evidence of the CCC’s role in firm performance, focused on developing countries. Credit access and industry requirements often shape cash management strategies in developing countries. Nwude et al. (2018) found a negative CCC–profitability relationship in Nigerian insurance firms, while Yazdanfar and Öhman (2014) emphasized reducing receivables and inventory in SMEs to enhance performance. These findings support the findings of Zakari and Saidu (2016), who observed positive CCC–performance correlations in Nigerian ICT companies, underlining sector-specific differences.

Studies also evidence the role of the CCC on firms’ profitability for unlisted companies and SMEs. García-Teruel and Martínez-Solano (2007) investigated the effects of working capital management on the profitability of small and medium enterprises (SMEs) in Spain. They found a negative relationship between SMEs’ profitability and the number of days of accounts receivable and inventory. The authors emphasized the importance of working capital management for value creation in SMEs and suggested that they minimize the CCC to a rational extent. Yazdanfar and Öhman (2014) reached similar conclusions for Swedish SMEs, implying that efficient CCC management enables smaller firms to optimize liquidity and maintain competitiveness. In Africa, Mathuva (2010) studied firms listed on the Nairobi Stock Exchange (NSE) and found an inverse relationship between the accounts collection period and profitability. The study also found direct relationships between the inventory conversion period, average payment period, and profitability. These findings are also supported by Kipkemoi et al. (2018), who highlighted inventory management as a critical factor for a firm’s performance. Similarly, Atnafu and Balda (2018) found that firms with better inventory and receivables management outperform their peers in Ethiopia. More recently, Kouaib and Haya (2024) find a negative relationship between a firm’s performance and the CCC in 84 firms (2018–2022) in the Saudi manufacturing sector.

2.3. The CCC and Firm’s Performance: Review of Studies on the Non-Linearity of the CCC Impact

Research into the relationship between the cash conversion cycle (CCC) and profitability has revealed complex, non-linear dynamics that challenge traditional linear assumptions. Earlier studies frequently characterized the CCC–profitability link as linear, highlighting that reductions in the CCC invariably enhance firm performance by improving liquidity and reducing financing costs. However, more recent evidence highlights a non-linear relationship, suggesting the presence of an optimal CCC range that maximizes profitability, beyond which either excessive lengthening or shortening can diminish performance. The theoretical basis for a non-linear relationship comes from the trade-offs inherent in working capital management. For example, while more extended receivable periods or higher inventory levels may support sales growth and operational continuity, they tie up cash and increase holding costs, eroding profitability.

In contrast, reducing receivables and inventory to shorten the CCC may weaken customer satisfaction and disrupt production continuity, leading to lost sales opportunities and a sensitive risk of operational inefficiencies. Empirical findings corroborate this theory. Baños-Caballero et al. (2012) demonstrated that the relationship between working capital and profitability follows an inverted U-shape in the context of Spanish SMEs, emphasizing an optimal CCC range. Their analysis indicates that excessively high and low working capital levels are associated with reduced profitability due to high costs or operational disruptions (Baños-Caballero et al., 2012; Wetzel & Hofmann, 2019). Similarly, Zeidan and Shapir (2017) provided evidence from a case study on MRV, a Brazilian firm, showing that operational improvements reducing the CCC from 508 days to 351 days significantly enhanced shareholder value, free cash flow, and profitability, underscoring the value-destructive potential of overinvestment in working capital (Zeidan & Shapir, 2017). Most recently, Huynh et al. (2025), using data from Vietnam’s Stock Exchange on listed companies’ performance (2012–2022), found strong evidence of a U-shaped relationship between the CCC and ROA.

The CCC’s impact on firm value also exhibits non-linear patterns. Effective CCC management enhances cash flow and operating efficiency, directly contributing to increased shareholder value. In contrast, inefficiencies stemming from aggressive CCC reductions or excessive expansion can compromise profitability and erode firm value (Zeidan & Shapir, 2017). Such findings are echoed by Wetzel and Hofmann (2019), who demonstrated that firms with supply chain finance (SCF)-oriented working capital strategies tend to achieve higher profitability and optimized CCC levels compared to firms adopting purely self-centered approaches (Wetzel & Hofmann, 2019). The evolving literature suggests that reconciling linear and non-linear perspectives on the CCC–profitability relationship requires a better understanding of factors, including industry dynamics, firm size, and financial constraints. Yazdanfar and Öhman (2014) observed that CCC efficiency yields more profitability benefits for smaller firms and those operating under constrained credit environments, as efficient cash management reduces reliance on costly external financing (Yazdanfar & Öhman, 2014; Baños-Caballero et al., 2012).

Hypothesis 1.

There is a non-linear U-shaped association between a firm’s profitability and the cash conversion cycle.

2.4. The CCC and Firm’s Performance: The Impact of Economic Crises

In the fourth strand of related studies, we review those concerned with the impact of economic and financial crises on enhancing the cash conversion cycle’s (CCC) effect on firm performance. Economic and financial crises influence the cash conversion cycle (CCC) dynamics and impact firm performance. During economic uncertainty, liquidity constraints, and diminished demand, the CCC becomes critical for firms to maintain operational efficiency and financial stability. Crises create a context where the interaction between working capital components and firm performance intensifies, amplifying the significance of efficient CCC management. Economic contractions, financial market disruptions, or widespread liquidity shortages typically mark crises. For example, the 2008–2009 global financial crisis was characterized by reduced credit availability and demand shocks, severely impacting firms’ operational cycles. These crises are particularly significant for firms reliant on external financing and those sensitive to demand changes (Castellares & Salas, 2019; Nguyen & Qian, 2014). The importance of liquidity increases during crises.

Firms with shorter CCCs often demonstrate resilience by efficiently converting inventory and receivables into cash, thereby mitigating reliance on external financing. H. C. Chen et al. (2018), Hofmann et al. (2022), and Carnes et al. (2023) highlight that firms with optimized CCC management experience better financial performance during downturns by maintaining liquidity buffers. Demand shocks, a hallmark of financial crises, disproportionately impact firms with inefficient CCCs. Nguyen and Qian (2014) emphasize that the primary adverse effect during the 2008–2009 crisis was a collapse in demand, underscoring the need for firms to maintain lean and efficient operational cycles. Firms with shorter CCCs adapted better to the reduced-demand environment, maintaining profitability despite external pressures. Similarly, H. C. Chen et al. (2018) extended the discussion to the role of pre-saved cash in financially constrained firms during economic downturns. Their study, spanning the 2000 dot-com crash and the 2008 financial crisis, provided evidence that firms with more pre-saved cash maintained capital investments and dividends, engaged in acquisitions, and were less likely to default during these periods. This suggests that cash reserves can serve as a critical buffer, enabling firms to navigate financial turbulence more effectively.

The sparse literature studying the role of economic crises on CCC’s impact on firm profitability reveals that economic and financial crises increase CCC’s impact on firm performance, making liquidity management critical. Efficient CCC practices mitigate the adverse effects of crises and enable firms to sustain their economic activity and performance during periods of economic uncertainty. This further emphasizes the need to explore CCC dynamics across different industries and crisis contexts.

Hypothesis 2.

The maximum (i.e., the turning point) in the association between profitability and the cash conversion cycle occurs at a higher cash conversion cycle during economic crises.

In Figure 1 below, we graphically show the hypothesized relationship between a firm’s profitability, measured by ROA, and its cash conversion cycle. This figure illustrates the U-shaped relationship between ROA and the CCC for the entire sample (curve 1), for the non-crises period (curve 2), and for the crises period (curve 3).

3. Data and Methodology

3.1. Data

The data used in this study were selected and extracted from the Amadeus database, provided by Bureau van Dijk, covering the period 2006–2015. This database has financial information on private firms across European countries. Firms in this study operate in different business sectors in countries such as Germany, Spain, France, Great Britain, and Italy. The focus on these selected countries is justified based on the affinity of these countries with business practices. In addition, these countries account for 87% of the GDP and 73% of the trade of the entire European Union (authors’ calculation from World Development Indicators).

The data are organized as unbalanced panel data. There are 6072 firms observed over ten years (2006–2015). There are firms for which data has been missing for some years. Instead of 60,720 (if data were balanced), we have only 41,185 observations. Data have been fairly distributed over the past ten years. The sample composition is as follows: Great Britain—15,666 or 38%; Germany—9571 or 23%; Italy—6128 or 15%; France—5904 or 14%; Spain—3916 or 10%.

Accordingly, Figure 2 shows that, on average, firms in Germany and Great Britain have operated with positive working capital. Firms in Spain have negative working capital, and only in 2015 did they operate with positive working capital. Firms in France from 2010 to 2014 have positive working capital and negative values for the rest of the years. Finally, firms in Italy for 2012–2014 had negative working capital, and for the rest of the years, they operated with positive values. Overall, these preliminary results show differences between and across countries, and consequently, no unique strategy or policy can be expected to manage the CCC.

Undoubtedly, firms would prefer to tighten the collection period and expand the payment period, consequently increasing short-term liquidity. They prefer to use more non-interest-bearing funding sources (by suppliers, i.e., obtained trade credit) rather than financial debt to finance business activities.

This study measures a firm’s profitability using return on assets (ROA hereafter). ROA is defined as a ratio of net income to total assets. ROA captures the effects of income and balance sheet statements, making it a useful measure of profitability.

In addition, one of the ways to measure working capital management is the cash conversion cycle (see, e.g., among others, Attari & Raza, 2012; Anser & Malik, 2013; Yazdanfar & Öhman, 2014; Aminu & Zainudin, 2015). Thus, the cash conversion cycle is “the net time interval between actual cash expenditures on a firm’s purchase of productive resources and the ultimate recovery of cash receipts from product sales” (Richards & Laughlin, 1980). Accordingly, the cash conversion cycle (CCC) could be more beneficial to decision-makers in the business operation period compared to the working capital conventional and static formula, calculated as the difference between current assets and current liabilities.

Our study calculates the CCC as the inventory conversion period + debt collection period—creditors deferral period. In addition, the CCC components are calculated as follows: (1) Inventory conversion period = Stock/(Operating revenue/360); (2) Debtors collection period = Debtors/(Operating revenue/360); and (3) Creditors deferral period = Creditors/(Operating revenue/360).

The rest of the independent variables, shown in

Table 1. Variables, definition, abbreviation, and calculation.

Description/Variables	Abbreviation	Calculation
Return on assets	ROA	Net income/Total assets
Independent variables:
Cash conversion cycle	CCC	Inventory conversion period + Debtors collection period − Creditors deferral period
Financial leverage	LEV	Total debt/Total assets
Firm’s size	SIZE	Logarithm of total assets
Tangibility ratio	TAN	Fixed assets/Total assets

Source: Authors’ calculations.

, are defined as follows: financial leverage (LEV) is defined as total debt to total assets; the firm’s size (SIZE) is measured as the logarithm of total assets; and tangibility ratio (TAN) is measured by the ratio of fixed assets to total assets.

Table 2. Descriptive statistics and correlation coefficients.

Variables/Descript. Stats	ROA	CCC	LEV	SIZE	TAN
Mean	0.0487	42.1981	0.1971	19.8092	0.1062
Max.	10.3755	488.3469	0.9785	27.1698	0.9861
Min.	−15.9733	−345.303	0	12.08765	0
Std. Dev.	0.1366	54.9941	0.1992	1.3804	0.1819
Obs’s	40,833	40,833	40,833	40,833	40,833
ROA	1
CCC	−0.0008	1
LEV	−0.090	0.0578	1
SIZE	−0.0376	0.0175	0.1719	1
TAN	−0.0301	−0.0484	0.1137	0.1529	1

Source: Authors’ calculations.

shows the descriptive statistics and correlation coefficients for all variables used in this study. The data show that the average return on assets is 4.8%, and the average CCC is 42 days. The firm’s leverage ratio is relatively low, 0.19, and the tangibility ratio is 0.1.

The correlation coefficients show that all dependent variables (CCC, LEV, SIZE, and TAN) are weakly negatively correlated with return on assets. The correlation coefficients among independent variables are small (weak correlation), which is the first sign of no multicollinearity.

3.2. Methodology

Given the longitudinal nature of the observed data, the many firms, and the economic affinity of countries in the sample, we employ an unbalanced panel data methodology. Data are unbalanced due to missing observations for some firms for some years. The measures of the firm’s performance, i.e., ROA, are regressed on CCC, squared CCC, and control variables. This study employs the unbalanced panel data regression for ROA for the entire sample and ROA during economic crises (periods during which a country where the firm operates has had negative real GDP growth, hence crises, and for periods during which a country where the firm operates has had positive real GDP growth).

This study considers that firm profitability, measured with ROA, is a function of cash conversion cycle (CCC), squared CCC, financial leverage (LEV), firm size (SIZE), and tangibility ratio (TAN). This equation is also estimated for the entire sample, for the subsample when economic crises are present, and for the subsample when there are no economic crises. Notice that variables measuring financial leverage (LEV), firm size (SIZE), and tangibility ratio (TAN) are used as control variables.

Controlling for factors influencing financial performance is crucial in examining the relationship between the cash conversion cycle (CCC) and firm profitability, measured by return on assets (ROA). Leverage (LEV) is a key determinant of a firm’s capital structure and has long been debated in financial literature since the seminal work of Modigliani and Miller (1958). The role of leverage in firm profitability is complex, as firms strategically adjust their financing decisions based on market conditions and investment opportunities (C. C. Chen et al., 2022; Li & Mauer, 2016). Empirical studies provide mixed evidence on the relationship between leverage and firm profitability, with some findings suggesting a negative association due to increased financial risk and debt servicing costs (C. Chang, 2018; Afrifa & Padachi, 2016), while others indicate that leverage can enhance firm value under specific conditions (Altaf & Shah, 2018; Deari et al., 2024).

Firm size (SIZE) is another essential control variable, as larger firms may benefit from economies of scale, greater access to financial resources, and more substantial bargaining power, which can impact both CCC management and profitability (Altaf & Shah, 2018; Deari et al., 2024). However, the relationship between firm size and profitability remains inconclusive, with some studies reporting a negative association, while others find that larger firms tend to outperform smaller firms (C. Chang, 2018).

Lastly, tangibility (TAN), the proportion of tangible assets to total assets, affects a firm’s ability to secure financing and influences capital structure decisions (Afrifa & Padachi, 2016; Altaf & Shah, 2018). Firms with higher asset tangibility may have greater access to debt financing, potentially impacting profitability through capital investment strategies (C. Chang, 2018). Given the varying effects of leverage, firm size, and tangibility on financial performance, incorporating these control variables ensures a more robust analysis of the CCC–profitability relationship by accounting for key firm-specific characteristics that influence profitability.

Further, testing whether a non-linear relationship exists (U-shaped or inverted) between firm profitability and the CCC requires squared CCC to be included in the regression model. Several prior studies confirmed a non-linear relationship between profitability and the CCC (e.g., Deari et al., 2024; Deari et al., 2024; Fernandes et al., 2021).

Therefore, the general panel regression models, which have to be estimated, can be written as follows:

$${ROA}_{it}=\propto +{ \beta }_1{CCC}_{it}+{\beta }_2{{CCC}^2}_{it}+{\beta }_3{LEV}_{it}+{\beta }_4{SIZE}_{it}+{\beta }_5{TAN}_{it}+{\omega }_i+{ϵ}_{it}$$

(1)

Using Equation (1), regressions are estimated for the entire sample, the subsample for the firms operating during the years when the country experienced economic crises (when the country experiences a negative real GDP growth), and the subsample for the firms operating during the years when the country did not experience economic crises (when the country experiences positive real GDP growth). This study ends with three estimated equations. It estimates the entire sample, the sub-sample for the period with crises, and the sub-sample for the period without crises. If Hypothesis 2 is to be confirmed, we expect that ${\beta }_1$ to be positive and ${\beta }_2$ negative.

We will start with estimating unbalanced panel data with fixed effects and then with random effects. The Hausman test determines which of the two estimating methods (fixed-effect or random-effect) gives the most efficient estimates.

Next, using Baños-Caballero et al. (2014), we calculate the optimal (maximum for the inverted U-shape and the minimum for the U-shape) level of the CCC. The maximum performance will be achieved at

$$CCC=-{\displaystyle \frac{{\beta }_1}{{2\times \beta }_2}},$$

where ${\beta }_1$ > 0 and ${\beta }_2$ < 0 (for more see, Baños-Caballero et al., 2014).

4. Empirical Analysis

In this section, we conduct an econometric estimation of Equations (1) and (2). These equations measure the effect of cash conversion period, financial leverage, size, and tangibility on a firm’s profitability. Equation (1) establishes the ground for testing the non-linear effect of the cash conversion cycle on profitability and, when run for subsamples (crises versus non-crises periods), establishes the ground for the differing effects of the cash conversion cycle during economic crises.

First, we test for stationarity of the series used in the regression analysis. The results are presented in

Table 3. Unit root test results.

Variables	ADF-Fisher Chi-sq	PP-Fisher Chi-sq
ROA	5669.46 (0.0000) ***	8963.79 (0.0000) ***
CCC	5435.56 (0.0000) ***	7898.66 (0.0000) ***
LEV	5784.52 (0.0000) ***	7739.13 (0.0000) ***
SIZE	4263.27 (0.7556)	6248.36 (0.0000) ***
TAN	5668.13 (0.0000) ***	29802.2 (0.0000) ***

Note(s): Probabilities, in parentheses, are computed using an asymptotic Chi-square distribution. Stars show the significance of the calculated statistic at a 1% significance level. Source: Authors’ calculations.

below.

The study uses two stationarity tests: augmented Dickey–Fuller (ADF) and Phillips–Perron (PP). The augmented Dickey–Fuller (ADF) test uses τ (tau) estimated statistics. The null hypothesis in both tests (ADF and PP) is that there is a unit root (unit root process). Rejecting the null hypothesis means that the series is stationary at the level of measurement. The results suggest we can reject the null hypothesis and conclude that there is insufficient evidence on unit root processes. The only variable that does not satisfy this condition is SIZE when considering the ADF test, but it is stationary under the PP test. We confidently say that the series’ stationarity is satisfactory, thus avoiding spurious regressions.

Next, we estimate Equation (1), including only the CCC and squared CCC. The results of such an estimation are presented in

Table 4. Fixed-effects panel regression for ROA [CCC and CCC² only].

Variable	ROA (All Sample)	ROA with Crises	ROA Without Crises
C	0.0509 *** (0.0012)	−0.049 *** (0.0387)	0.0486 *** (0.0008)
CCC	−0.00002 (0.00003)	−0.000037 (0.000036)	0.000039 (0.000000003)
CCC2	−0.0000003 ** (0.00000014)	0.00000000026 (0.00000002)	−0.0000003 ** (0.00000014)
Adj. R-sq	0.2601	0.2620	0.2620
F-statistic	3.3841	3.4044	3.4051
P(F-stat.)	0.0000	0.0000	0.0000

**, and *** show the estimated coefficient’s statistical significance at 5%, and 1%, respectively. The numbers in brackets below the estimated coefficients are the respective standard errors. Source: Authors’ calculations.

below. The regression results in Table 4 show the effect of the CCC and squared CCC alone on the firm’s profitability. These preliminary results suggest further investigation. The coefficients of the effect of the CCC on ROA are not statistically significant, and the coefficient of squared CCC is statistically significantly different from zero. Due to misspecifications, this inconclusiveness leads us to the necessity of estimating Equation (1), including all dependent variables.

Next, we estimate the regressions using all control (dependent) variables and the CCC and squared CCC. Three estimated results are summarized in

Table 5. Summary of regression results [fixed effects].

Variable	ROA (All Sample)	ROA Without Crises	ROA with Crises
C	−0.1682 ***	0.2398 ***	−3.0277 ***
	(0.0379)	(0.0328)	(0.2272)
CCC	0.0000175	−0.00008 ***	0.0006 ***
	(0.000035)	(0.00003)	(0.00017)
CCC2	−0.0000003 **	−0.00000014	−0.00000011 *
	(0.00000014)	(0.00000013)	(0.0000006)
LEV	−0.0999 ***	−0.1021 ***	−0.0378
	(0.0066)	(0.0057)	(0.0398)
SIZE	0.0118 ***	−0.0083 ***	0.1516 ***
	(0.0019)	(0.0017)	(0.0115)
TAN	0.0217 ***	0.0005	0.6338 ***
	(0.0047)	(0.0037)	(0.0649)
Adj. R-sq	0.3739	0.5298	0.4774

*, **, and *** show the estimated coefficient’s statistical significance at 10%, 5%, and 1%, respectively. The numbers in brackets below the estimated coefficients are the respective standard errors. Source: Authors’ calculations.

. The estimated method is that of unbalanced panel data fixed effects. This method is justified (Green, 2018) because this study uses short panel data (i.e., ten years of data from over 6072 firms). Because of the short period and, relative to that, an extensive set of panel units, the panel least squares technique is one of the estimation techniques that best estimates the unbiased and efficient estimates. Table 5 presents a summary of the regression results. We follow Kukeli et al.’s (2023) battery of tests and justification of techniques to produce the estimates reported.

In addition to using a fixed-effect panel, we run a random-effect panel and use the Hausman test to determine whether the random effect would give more efficient estimates. The results of this test (Hausman test—null hypothesis: GLS (random effect) estimates) are consistent. The asymptotic test statistic (Chi-square (5) = 84.2114 with p-value = 0.000001) favors the fixed-effects model.

The results presented in Table 5 offer a better picture of the impact of the CCC and squared CCC on ROA, with and without crises. We proceed with testing for heteroscedasticity as a primary source of efficient estimates. The FE panel data models do not pass the Breusch–Pagan/Cook–Weisberg heteroscedasticity test. As Green (2018) suggests, we run a weighted least square (WLS) panel regression to correct for heteroscedasticity, where the weights are based on per-unit error variances. The new regressions have a better overall fit, shown by a higher adjusted R-square, and all coefficients are statistically significantly important and different from zero at all three conventional levels of significance (1%, 5%, and 10% significance level). The results of WLS unbalanced panel data estimations are presented in

Table 6. Weighted least square (WLS) regressions.

Variables	ROA (All Samples)	ROA (No Crises)	ROA (Crises)
C	0.0821 *** (0.0017)	0.0851 *** (0.0018)	0.0493 *** (0.0006)
CCC	0.0001 *** (0.000003)	0.000068 *** (0.000004)	0.00014 *** (0.000001)
CCC2	−0.0000005 *** (0.00000001)	−0.0000004 *** (0.00000002)	−0.00000076 *** (0.000000005)
LEV	−0.0532 *** (0.0005)	−0.0543 *** (0.0005)	−0.0578 *** (0.0002)
SIZE	−0.0014 *** (0.00009)	−0.0013 *** (0.00009)	−0.00004 *** (0.00003)
TAN	−0.0102 *** (0.0006)	−0.0064 *** (0.0005)	−0.0308 *** (0.0006)
Adj. R-sq	0.5187	0.2697	0.9343
Obs’s	41185	32254	8931
Units	6072	6038	4553

*** show the estimated coefficient’s statistical significance at 1%. The numbers in brackets below the estimated coefficients are the respective standard errors. Source: Authors’ calculations.

below.

The estimated equations are as follows:

$$\begin{array}{cc}\hfill \widehat{{ROA}_{it}}=0.0821& +\left(0.0001\right){CCC}_{it}-\left(0.0000005\right){{CCC}^2}_{it}-\left(0.0532\right){LEV}_{it}\hfill \\ & -\left(0.0014\right){SIZE}_{it}-\left(0.0102\right){TAN}_{it}\hfill \end{array}$$

(2)

$$\begin{array}{cc}\hfill \widehat{{ROA}_{it}}=0.0851& +\left(0.000068\right){CCC}_{it}-\left(0.0000004\right){{CCC}^2}_{it}\hfill \\ & -\left(0.0543\right){LEV}_{it}-\left(0.0013\right){SIZE}_{it}-\left(0.0064\right){TAN}_{it}\hfill \end{array}$$

(3)

$$\begin{array}{cc}\hfill \widehat{{ROA}_{it}}=0.0493& +\left(0.00014\right){CCC}_{it}-\left(0.00000076\right){{CCC}^2}_{it}\hfill \\ & -\left(0.0578\right){LEV}_{it}-\left(0.00004\right){SIZE}_{it}-\left(0.0308\right){TAN}_{it}\hfill \end{array}$$

(4)

Overall, Equations (2)–(4), using the unbalanced panel data weighted least squares (WLS) method, estimate satisfactory yield results and confirm the stated hypothesis. The Wald test for omitting variables suggests that none should be removed from the estimation. The results for omitting variables fail to reject Wald’s test null hypothesis: the regression coefficients are zero.

5. Discussion of Results

The regression analysis results in Table 6 and Equations (2)–(4) provide a comprehensive understanding of the relationships between the CCC, squared CCC, and the control variables in the variation of corporate performance, measured in terms of return on assets (ROA). The analysis spans different conditions, including the entire sample, periods with crises, and periods without crises. It is important to notice that both the CCC and squared CCC are statistically significantly different than zero at a 1, 5, and 10% significance level for the entire sample and crises and non-crises. The adjusted R-squared values suggest that the models explain a moderate to substantial portion of the variability in ROA.

These results strongly suggest that the effect of the CCC in ROA (i.e., cash version cycle on profitability) is non-linear, suggesting strong evidence in favor of Hypothesis 1. Because of the squared CCC, its marginal effect can be calculated using the partial derivative of ROA. We find that the marginal effect of the CCC on ROA during periods with no economic crises is 1.65 times higher than the marginal effect for the entire sample. The marginal effect of the CCC on ROA for periods with economic crises is 2.36 times higher than the same marginal effect for the entire period. In addition, there are striking differences in the marginal effect of the cash conversion cycle on return on assets during economic crises. This effect is 3.9 times higher during economic crises than during periods without economic crises. These results strongly suggest that economic crises play a role in the effect of the CCC on ROA, lending evidence in favor of Hypothesis 2. Results show that the CCC has a non-linear effect on ROA, and the effect is magnified during crises. Additionally, there is a strong inverted U-shape for the effect of the CCC on firms’ return of assets. The non-linear relationship indicated by CCC² in ROA suggests diminishing returns, corroborating C. Chang’s (2018) finding that the benefits of reducing CCC taper off or reverse at lower levels.

The mean value of the CCC (Table 2) is 42. We also calculate separately the mean CCC value for the non-crisis period (41 days) and for the crisis period (46 days). Using the estimated coefficients of Table 6 and the mean values of the CCC, we calculate the optimum (maximums) of 90 days for ROA for the entire sample, 92 days for ROA with economic crises, and 85 days for ROA without economic crises. This evidence suggests that firms find it challenging to convert cash paid into cash received during economic crises.

Next, we discuss the effects of control variables. Leverage (LEV) consistently shows a negative and highly significant impact on ROA across all conditions. The estimated coefficients are −0.0532 (entire sample), −0.0543 (no crises), and −0.0578 (during crises). These estimates indicate a stable negative effect of leverage on asset returns. The consistent negative impact of leverage on ROA supports the findings of Bontempi et al. (2020). Their work highlights how leverage choices significantly impact firm performance, with higher leverage often leading to lower profitability due to increased financial risk and costs.

The firm size (SIZE) variable negatively impacts ROA across all conditions. The estimated coefficients are −0.0014 (entire sample), −0.0013 (no crises), and −0.00004 (during crises). This indicates that larger firms tend to have lower asset returns.

Asset tangibility (TAN) exhibits a negative and significant impact on ROA, with estimated coefficients of −0.0102 (entire sample), −0.0064 no crises), and −0.0308 (during crises), highlighting that firms with less tangible assets tend to perform better regarding asset returns. This suggests that tangible assets might burden performance, especially during crises, due to higher maintenance or depreciation costs. The negative impact on a firm’s performance, especially during non-crisis periods, might be explained by the maintenance and depreciation costs associated with tangible assets, as discussed in Fernandes et al. (2021), who noted the cost implications of asset management in banks and other sectors.

6. Conclusions

This study investigates whether the effects of the CCC on firm performance are non-linear and whether economic crises matter for profitability. The study uses weighted least squares (WLS) unbalanced panel data from 2006 to 2025 for selected European countries. The results reveal strong evidence of the non-linearity of the effect of the CCC on firms’ performance, with different lengths of the CCC within and across firms, providing evidence in favor of Hypotheses 1 and 2. Similarly, this study provides evidence that the non-linear effect of a CCC firm’s performance differs during crises and non-crisis periods, supporting Hypothesis 2. The relationship between firm performance (measured by ROA) and CCC is examined by adding firm characteristics such as financial leverage (LEV), firm size (SIZE), and tangibility ratio (TAN). In addition to the non-linear effect of the CCC, all three control variables significantly influence performance (return on assets). Furthermore, there are striking differences in these effects during economic crises and periods with no economic crisis, with economic crises exacerbating these effects on ROA.

Therefore, the regression results highlight distinct relationships between financial metrics and corporate performance. A firm’s size, leverage, and tangibility consistently negatively impact ROA. The role of the CCC is decisive and non-linear, affecting asset returns and emphasizing the complexity of working capital management in financial performance.

Based on the regression results and the supporting literature, several policy implications can be drawn for firms seeking to optimize their performance. Firms should optimize their CCC to enhance performance. Given the negative impact of leverage on ROA, firms should carefully manage their debt levels to avoid excessive financial risk. The negative effect of tangible assets on ROA highlights the importance of managing maintenance costs and reducing access stocks.

This study’s limitations are due to a specific region, country, and period chosen to capture the crisis’s effect on the relationship between the CCC and firm performance. This study limits the econometric analysis to the unbalanced panel data weighted least square method. In the case of balanced panel data, other econometric techniques could be explored in future studies. In addition, future research on the relationship between the cash conversion cycle (CCC) and corporate performance could focus on various pertinent areas to extend the current study. Such an extension could involve investigating the non-linearity and the crisis effect for different industries and small- and medium-sized enterprises.