Working Capital Management and Firm Profitability in ASEAN: A Cross-Country Panel Analysis

The trade-off between liquidity and profitability constitutes the core theoretical framework for research on WCM. Firms that invest heavily in current assets—by extending trade credit to customers, accumulating inventory, or paying suppliers promptly—maintain operational flexibility but divert capital away from productive investment. Conversely, aggressive policies that compress the CCC free internal resources but may impair supply chain reliability. Shin and Soenen [12] formalised this trade-off by demonstrating that the net trade cycle is inversely related to stockholder value. Deloof [3] extended the framework to accounting-based performance, attributing the negative CCC–profitability relationship to the financing cost imposed on firms that carry excess current assets. Agency theory [13] further predicts that managers may overinvest in working capital to build operational slack, thereby creating a wedge between actual and optimal working capital levels. Dynamic trade-off theory resolves these tensions by positing that firms target an optimal working capital level, with deviations in either direction imposing performance costs [5].

The systematic review by Prasad et al. [5], encompassing 75 highly cited studies across multiple markets, found that approximately 79% document a negative WCM–profitability relationship, establishing this as the empirical consensus. Within developed and upper-middle-income markets, Lyngstadaas and Berg [4] confirm a significant inverse CCC–return on assets (ROA) relationship for Norwegian firms using pooled ordinary least squares (OLS) and two-stage least squares, and recommend flexible, aggressive working capital policies as a value-enhancing strategy. Kumar et al. [14] corroborate this for Indian listed firms, showing that shorter CCC and inventory days increase the probability of achieving superior composite financial performance. In Pakistan, Kafeel et al. [15] employ system generalised method of moments (GMM) alongside static panel estimators and find that reducing the payable deferral period and CCC consistently improves ROA and operating profit, a finding partially supported by Khan et al. [16] who note that cash holdings negatively affect operating profitability, though WCM effects on stock prices vary across manufacturing sectors. Al-Mawsheki et al. [11] extend the evidence to Malaysia, demonstrating that the CCC has a statistically significant negative effect on Economic Value Added, suggesting that WCM inefficiency reduces value creation beyond what accounting-based measures capture.

Within the ASEAN region, evidence specific to Vietnam and adjacent emerging markets reinforces the negative WCM–performance hypothesis. Nguyen and Van Nguyen [10] apply fixed-effects panel regression to Vietnamese non-financial firms and find that the CCC and its components—inventory conversion period, accounts collection period, and accounts payable period—significantly and negatively affect ROA and gross operating income. Their analysis further reveals that trade credit usage in the Vietnamese corporate sector is largely incidental rather than strategically deployed, indicating that firms do not fully exploit trade credit as a financing instrument. Toy et al. [17], using the Driscoll-Kraay estimator for the Fragile Five economies, including Indonesia, show that WCM effects vary with the choice of performance metric and country-specific macroeconomic conditions, highlighting the importance of cross-country heterogeneity. Contrasting evidence comes from Godswill et al. [18] for Nigerian deposit money banks which report positive associations between the payable period and profitability. These divergences are attributable to structural differences: banking firms operate under a distinct intermediation model, and credit-constrained small and medium-sized enterprises (SMEs) in underdeveloped financial environments derive genuine financing benefits from extended payables—conditions that do not apply to the listed non-financial firms examined in the present study.

A distinct strand of the literature challenges the linear specification, providing evidence of non-linear, concave WCM–profitability dynamics that imply an optimal level of working capital. Korent and Orsag [6] estimated an inverted-U relationship between the net trade cycle and ROA for Croatian software firms using the Arellano-Bond dynamic panel estimator. Anton and Nucu [19] replicated this pattern for Polish listed firms using OLS, fixed-effects, and panel-corrected standard error estimators, and Sensini and Vazquez [20] extended the inverted-U finding to Argentine SMEs for each CCC component individually. In the Norwegian market, Lyngstadaas and Berg [4] include quadratic WCM terms and confirm that performance costs arise from both over- and under-investment in working capital, with over-investment imposing larger penalties. Nguyen and Van Nguyen [10] estimate GLS models with squared WCM terms and identify a U-shaped CCC–ROA relationship, implying that at low working capital levels, further reduction is detrimental. Huang and Dinh [7] adopt Hansen's panel threshold model, with firm leverage as the threshold variable, and find that the negative WCM–ROE relationship intensifies markedly when the debt ratio exceeds 78.45%, thereby establishing leverage as a first-order moderator of WCM effects in Vietnam. A recent study on 88 listed transportation firms in Vietnam (2014–2023) found that the relationship between WCM and profitability is nonlinear and depends on a cash-holding threshold. Below the threshold, longer receivables, payables, and inventory periods significantly reduce ROA. Above the threshold, however, the effects of days sales outstanding (DSO), days payable outstanding (DPO), CCC, and OCC turn positive, suggesting that more flexible working capital policies may enhance profitability when firms maintain sufficient liquidity [21]. Prasad et al. [22] frame non-linearity through the concept of target working capital, demonstrating for Indian firms that deviations from the target net trade cycle—regardless of direction—reduce gross operating income, consistent with the dynamic trade-off prediction. While the present study employs a linear specification, the non-linear evidence implies that ASEAN firms primarily operate on the descending portion of the inverted-U curve, where further CCC reduction yields profitability improvements.

Recent research has identified several firm- and environment-level factors that condition the strength and direction of the WCM–performance relationship. Koroma and Bein [9] show for UK-listed firms that the CCC has a positive baseline association with profitability, but that elevated economic policy uncertainty (EPU) weakens and ultimately reverses this relationship, indicating that macro-level instability introduces time variation into WCM–performance dynamics. This finding is relevant for the ASEAN context, given the region's susceptibility to macroeconomic and policy shocks. Bashir and Ahmad [8] demonstrate for Pakistani firms that market power moderates the CCC–firm value relationship: firms with high market power sustain profitability under aggressive (longer CCC) working capital strategies without incurring the penalties that competitive firms face. El-Ansary and Al-Gazzar [23] show for MENA consumer goods firms that firms' cash levels determine whether the net working capital requirement translates into operating performance, highlighting the role of internal liquidity buffers as a boundary condition. Ahmad et al. [24] further demonstrate, using system GMM for firms in Malaysia, Thailand, and Pakistan, that COVID-19 produced deeper profitability declines than the 2008 financial crisis and temporarily reversed the CCC–ROA relationship as firms extended payables to preserve liquidity—while capital markets continued to penalise working capital inefficiency through lower Tobin's Q. Akgün and Karataş [25] corroborate the crisis argument for EU-28 firms, finding that the 2008 financial crisis substantially suppressed WCM cash-flow dynamics, reducing the translation of working capital efficiency into performance gains.

Despite the breadth of the evidence, several gaps remain. Cross-country panel evidence encompassing the institutional diversity of ASEAN, ranging from the financially developed Singapore to the credit-constrained corporate sectors of Vietnam and Indonesia, is limited. Methodologically, many prior studies rely on OLS or random-effects estimators that do not account for panel heteroskedasticity or serial autocorrelation [5], limiting the reliability of their estimates. Sector-level disaggregation, identified by Prasad et al. [5] as a priority research direction, has been addressed only partially by Khan et al. [16] and Sensini and Vazquez [20] for single-country samples. The present study addresses these gaps through a six-country ASEAN panel estimated by FGLS, with systematic disaggregation by ten Global Industry Classification Standard (GICS) sectors and six countries.

The DSI coefficient for Malaysia is not statistically significant in the joint model (β = −0.007, n.s.), but it is significant in the single-variable specification (β = −0.028***). This outcome indicates that the inventory effect in Model 2 for Malaysia was predominantly influenced by its relationship with DSO; when receivables management is accounted for, inventory holding itself does not independently diminish profitability for Malaysian listed companies. This conclusion is credible given Malaysia's sectoral structure, where large-cap government-linked companies, plantation enterprises, and financial services predominantly influence the exchange and exhibit lower sensitivity to inventory costs than in manufacturing-centric economies like Thailand or Vietnam.

The DSO remains negative and highly significant across all six nations in the joint model, with coefficient magnitudes essentially retained compared to the single-variable specification, demonstrating that the receivables effect is not an artefact of collinearity with inventory or payables cycles. The robustness of DSO establishes it as the paramount and most dependable WCM determinant of financial performance within the ASEAN environment, applicable to both individual-variable and joint specifications.

Significantly, in three nations — Philippines (β = −0.077***), Thailand (β = −0.134***), and Vietnam (β = −0.106***) — the DSO coefficient in the combination model is markedly greater than in the single-variable model. In the Philippines, the single-variable DSO coefficient was −0.033***, increasing to −0.077*** in the combined model. This amplification arises because the single-variable model conflates DSO with DPO: Philippine enterprises that provide credit to customers also tend to extend payables to suppliers, resulting in a positive partial correlation that diminishes the DSO coefficient when DPO is omitted. Upon controlling for the independently favourable effect of DPOs, the whole cost of extending receivables becomes apparent. This discovery underscores the notion that receivables management is a vital — and hitherto undervalued — performance determinant, especially in economies dominated by conglomerates.

Thailand demonstrates the highest DSO coefficient in the combined model (β = −0.134***), surpassing its substantial single-variable estimate (β = −0.120***). The amplification demonstrates a similar confounding mechanism: Thai companies with extended receivables cycles typically exhibit prolonged payables cycles (positive DPO), which, in the univariate model, somewhat mitigate the DSO effect. Accounting for DPO's beneficial impact in Model 5 elucidates the full extent of Thailand's receivables inefficiency costs — a notably significant finding for Thai export-oriented manufacturing enterprises, where delays in international buyers' payments constitute a continual operational obstacle.

The joint-model DSO coefficient for Vietnam (β = −0.106***), although marginally lower than the single-variable estimate (β = −0.119***), remains among the highest in the sample. The slight decrease confirms that Vietnam's DSO and DSI effects are predominantly autonomous: Vietnamese enterprises concurrently experience both inventory accumulation expenses and collection inefficiencies, with these two WCM deficiencies operating through separate mechanisms rather than being driven by a shared underlying factor. This outcome aligns with the structural assessment of Vietnam's working capital landscape: an inadequate trade finance infrastructure incurs expenses at both stages of the working capital cycle — in acquiring and maintaining inputs and in receiving payments from clients.

The combination model yields the most theoretically informative DPO results of the comprehensive analysis. Three separate groups of countries arise, each embodying a unique institutional equilibrium for trade credit within ASEAN:

The initial group — Indonesia (β = −0.043***), Malaysia (β = −0.029***), Singapore (β = −0.020*), and Vietnam (β = −0.006, not significant) — exhibits negative or neutral DPO effects in the combined model, affirming that the relational cost of delayed payment endures despite adjustments for inventory and receivables management. The DPO coefficient for Vietnam, which was significant in the univariate model (β = −0.156***), becomes statistically insignificant in the multivariate model (β = −0.006, not significant). This significant reduction is attributed to the high collinearity between DPO and DSO in Vietnam: companies with extended receivables cycles tend also to have prolonged payables cycles, indicative of a pervasive 'slow payment' culture. Once DSO is accounted for, DPO loses its independent explanatory capacity. This does not diminish the economic significance of DPO management in Vietnam; rather, it suggests that, in the Vietnamese context, delayed payments to suppliers and delayed collections from customers are two expressions of the same systemic WCM inefficiency, more comprehensively captured by DSO in the integrated model.

The second group — Thailand (β = +0.057***) — demonstrates a strong positive DPO effect in the combined model, but its unimpressive outcome in the single-variable analysis. This reversal is analytically significant: it demonstrates that, after eliminating the confounding effect of Thailand's substantial negative DSO coefficient, extending payables yields a net liquidity advantage for Thai enterprises. The contingent nature of this outcome — favourable solely when DSO is accounted for — indicates that Thai companies who adeptly handle receivables can convert prolonged payables into authentic enhancements in profitability. This aligns with the formalised contractual trade credit standards set by Japanese multinational supply chains in Thailand, where the extension of DPO is a negotiated and generally recognised practice rather than an indication of financial difficulty [28]. The favourable joint-model DPO effect in Thailand indicates a developed trade credit relationship that functions independently of, and is obscured by, Thailand's significant receivables management issue.

The third group — the Philippines (β = +0.045***) — has a sustained positive DPO effect in the joint model, the most substantial in the sample. The consistency of this coefficient between Models 4 and 5 suggests that the buyer power mechanism influencing the Philippines' positive DPO effect is robust and not due to collinearity. Philippine conglomerates utilise their supply chain supremacy to delay payables without incurring relational costs, irrespective of their receivables management strategies—demonstrating that the trade credit dynamics in the Philippines operate under a fundamentally distinct institutional logic from those in Vietnam, Indonesia, or Thailand.

The control-variable outcomes in the joint model strengthen the institutional narrative. Firm size (SIZE) exerts a substantial positive influence in Indonesia (β = +0.140***) and Singapore (β = +0.212***), a moderate influence in Malaysia (β = +0.067***) and the Philippines (β = +0.008, not significant), and a negative influence in Thailand (β = −0.023***) and Vietnam (β = −0.004, not significant). The positive SIZE effect in Singapore aligns with the prevalence of large, well-managed firms on the Singapore Exchange, which benefit from economies of scale, enhanced capital market access, and lower financing costs. The adverse SIZE impact in Thailand is paradoxical and may indicate the existence of large, capital-intensive corporations characterised by substantial fixed costs and narrow operating margins, in which scale fails to yield short-term profitability.

The leverage penalty (DE) is the most consistently observed phenomenon across all six nations and all model configurations. The Philippines has the most pronounced effect (β = −0.663***), succeeded by Thailand (β = −0.648***) and Singapore (β = −0.526***). The substantial leverage penalties in the Philippines and Thailand indicate elevated financial distress costs associated with concentrated debt structures in economies characterised by protracted bankruptcy resolution and inadequate enforcement of creditor rights. The relatively minor leverage penalty in Malaysia (β = −0.041***) aligns with the financial robustness of prominent GLCs and blue-chip corporations on Bursa Malaysia, which maintain investment-grade credit ratings and secure diverse, low-cost funding sources. The intermediate leverage effect in Vietnam (β = −0.241***) indicates that leverage among publicly listed companies is increasing, many of which are former state-owned enterprises with legacy debt frameworks, and that the banking system continues to pose a systemic issue due to non-performing loans.

Liquidity (LIQ) is both positive and significant across all six countries, with the largest effects in Indonesia (β = +0.350***) and Vietnam (β = +0.287***). This discovery offers the most robust empirical validation for Pecking Order Theory within the ASEAN framework: in nations where external borrowing is the most expensive and least attainable, domestic liquidity emerges as the most important financial asset for enterprises. The substantial liquidity premium in Indonesia and Vietnam—approximately quadruple the Malaysian coefficient (β = +0.082***)—correlates directly with the relative underdevelopment of their capital markets and a heightened dependence on bank lending, which is both rationed and costly for non-blue-chip borrowers.

Sales growth (GRO) is consistently positive in all six countries, with Indonesia (β = +0.194***) and Singapore (β = +0.209***) exhibiting the most significant benefits. The strength of the growth-profitability relationship in Singapore is significant: in a mature, low-growth economy, companies that attain above-average revenue growth do so through authentic competitive advantage rather than market expansion, and this competitive advantage directly results in a profitability premium. Capital expenditure (CAPEX) is generally negative, except in the Philippines (β = +0.001, not significant), indicating that CAPEX generally leads to short-term earnings dilution across most ASEAN economies. However, the positive, albeit insignificant, result for the Philippines suggests a more efficient capital allocation within conglomerate structures featuring integrated project pipelines.

GDP growth (GDPG) is both positive and substantial in Indonesia (β = +0.048***), the Philippines (β = +0.052***), Thailand (β = +0.031***), and Vietnam (β = +0.034***), which are the four economies exhibiting the highest cyclical sensitivity of firm performance to domestic production. Conversely, GDPG is negligible in Singapore (β = +0.016, not significant) and Malaysia (β = +0.006, not significant), indicating that publicly traded companies in these more open, trade-dependent economies are predominantly influenced by global rather than domestic economic factors. The inflation coefficient (INF) persists in demonstrating the cross-country heterogeneity observed in earlier model specifications: positive in Malaysia (β = +0.056***) and Indonesia (β = +0.021, not significant), and negative in the Philippines (β = −0.060***), Thailand (β = −0.031***), Vietnam (β = −0.064***), and Singapore (β = −0.019**). This disparity supports the notion that large-cap firms in Malaysia and Indonesia have the pricing power to capitalise on inflationary conditions, but firms in the Philippines, Thailand, Vietnam, and Singapore experience margin compression during inflationary surges.

The studies whose findings differ from ours [8, 9, 18] share specific contextual features that elucidate the reasons for their difference. These characteristics include the following: the examination of firms with exceptional market power that enables them to mitigate the adverse effects of long working capital cycles, the focus on banks and financial institutions with working capital structures that differ from those of nonfinancial firms, and the use of developed financial markets with lower financing constraints.

Interestingly, our conclusions are not actually contradicted by the nonlinear findings from some studies [6, 7, 19]. However, they contribute to a deeper understanding of the relationship between profitability and WCM. In particular, our discovery of a negative linear impact of the currency conversion cycle on profitability is entirely consistent with the notion that firms in the Southeast Asian region are operating on the downward sloping portion of an inverted U-shaped profitability curve. Performance improves until an optimal point is reached on this portion of the curve, as additional reductions in the CCC are implemented. Consequently, the nonlinear studies do not refute our findings; rather, they provide an additional layer of significance: the adverse effects of extending the working capital cycle may not be strictly linear. The effect may become greater or weaker as it exceeds a certain threshold. This implies that future research on Southeast Asian markets should formally test for nonlinearity using methods such as threshold regression or smooth transition regression.

The country-level results indicate that financial market development, corporate ownership structure, and the dominant commercial relationship model in each economy are systematically correlated with WCM sensitivity across ASEAN. The Pecking Order Theory predicts that internal financing capacity is most valuable where external capital is most costly, and frontier and bank-dominated markets (Vietnam, Indonesia) exhibit the largest WCM sensitivity and the strongest liquidity premia. The availability of trade finance instruments that enable decoupling of operational and financial working capital cycles is consistent with the smaller, more balanced WCM effects observed in more institutionally advanced economies (e.g., Singapore and Malaysia). By contrast, the Philippines' comparatively low WCM sensitivity arises through a different channel — concentrated conglomerate ownership and buyer power — rather than financial market depth, illustrating that muted WCM sensitivity in ASEAN can arise from either institutional advancement or structural market concentration.

The structural distinction between SME-dominated, network-governed supply chains (where relational costs of delayed payment are high) and conglomerate-dominated, power-asymmetric supply chains (where delayed payment confers a genuine liquidity benefit) is reflected in the DPO heterogeneity, which is negative in Vietnam and Indonesia, positive in the Philippines, and insignificant or conditionally positive in Thailand. This discovery emphasises the importance of institutional conditioning in WCM research for emergent market contexts.

The macroeconomic findings further substantiate institutional heterogeneity. In Vietnam and the Philippines, inflation is negative due to margin compression, while it is positive in Malaysia and Indonesia (pricing power and inflation tolerance), and negative in Singapore and Thailand (low-inflation, cost-sensitive environments). In frontier economies (Vietnam, Indonesia, Philippines), GDP growth is more significant than in more globalised economies (Singapore, Malaysia), where firm performance is decoupled from domestic output cycles.