Macroeconomic factors, working capital management, and firm performance—A static and dynamic panel analysis

The study investigates how working capital management (WCM)—captured by cash conversion cycle (CCC), average receivable days (ARD), inventory turnover (ITR), and average payable days (APD)—relates to firm performance (gross profit) in the fuel and energy sector of a developing economy, and how macroeconomic conditions (exchange rate and interest rate) moderate these relationships. Motivated by the sector’s strategic importance and unique institutional features, the paper examines whether CCC is negatively associated with profitability and whether exchange rate and interest rate moderate the CCC–profitability link. The authors focus on 21 Pakistani fuel and energy firms (2013–2018) to address a gap on macro-micro interactions in WCM within this sector. Hypotheses: (H1) CCC negatively correlates with gross profit; (H2) exchange rate moderates the CCC–gross profit relationship; (H3) interest rate moderates the CCC–gross profit relationship.

Prior studies generally find that efficient WCM improves profitability and that longer CCC is associated with lower profitability. Research also reports negative impacts of leverage on profitability, and roles of firm age, size, and liquidity. Macroeconomic conditions (e.g., exchange rates and inflation) can influence firms’ operating cycles and profitability, especially in sectors exposed to international inputs and sales. Contradictory findings in the CCC–profitability nexus may reflect unmodeled macro moderators such as the exchange rate. Based on this, the study proposes that exchange rate and interest rate moderate the effect of CCC (and APD) on gross profit, leading to three hypotheses (H1–H3).

Data and sample: Panel of 21 firms in the fuel and energy sector listed on the Karachi Stock Exchange (Pakistan) observed from 2013–2018. Financial data are from company reports and the State Bank of Pakistan; macro variables (interest rate, REER/exchange rate) are from the State Bank.

Variables: Dependent variable: Gross Profit (GP). Key WCM regressors: ARD = (average accounts receivable/net sales)365; ITR days = (average inventory/net sales)365; APD = (average accounts payable/COGS)365; CCC = ARD + ITR − APD. Macroeconomic moderators: interest rate (Irate) and exchange rate (ExR, REER). Controls: firm age, firm size (log total assets), debt-equity ratio (total debt/total assets), net cash flow from operations (NCFO), liquidity (current ratio). Interaction terms: IrateAPD, IrateCCC, ExRAPD, ExR*CCC.

Empirical strategy: Multiple specifications using both static and dynamic panel methods.

• Static models: Pooled OLS, Random Effects (RE), Fixed Effects (FE). Model selection via Breusch–Pagan LM and Hausman tests. Panel-corrected standard errors (PCSE) with AR(1) and heteroskedasticity adjustments are employed to address serial correlation and heteroskedasticity.
• Dynamic models: System GMM (Arellano–Bover/Blundell–Bond) with lagged dependent variable to address endogeneity, unobserved heterogeneity, and dynamic persistence. Instrument validity tested via Sargan test; serial correlation via Arellano–Bond AR(1)/AR(2) tests; finite-sample correction per Windmeijer.

Model forms include: (i) GP on ARD, ITR, APD plus controls and macro variables; (ii) GP on CCC plus controls and macro variables; (iii) inclusion of interaction terms with APD and CCC for both Irate and ExR; (iv) dynamic counterparts with lagged GP.

Diagnostics: VIF for multicollinearity; heteroskedasticity tests; Wooldridge test for serial correlation; Hausman tests for FE vs RE; Sargan and AR tests for GMM validity.

• Descriptive statistics: Mean CCC ≈ −66.87 days; mean APD ≈ 66.88 days; mean GP ≈ 13.72; mean interest rate ≈ 7.95%; mean liquidity ≈ 2.09.
• Correlations: CCC negatively associated with GP; ARD and APD positively associated with GP; liquidity, exchange rate, and interest rate show positive simple correlations with GP, while other controls have mixed associations.
• Static panel results (with PCSE): CCC is negatively related to gross profit; ARD, ITR, and APD are generally positively related to GP. Size, DER, NCFO, liquidity, interest rate, and exchange rate are significant in several specifications, with DER typically negative. Exchange rate and interest rate tend to have adverse direct effects on GP once model corrections are applied.
• Dynamic panel (System GMM): Inclusion of lagged GP addresses serial correlation and endogeneity; size turns positive and significant, ARD and ITR can switch signs relative to static models, indicating dynamic adjustments. Diagnostic tests (Sargan p>0.10, AR(1) present, AR(2) absent) support instrument validity and specification.
• Quantified effects reported: In one specification, a 1-day increase in ARD associates with about a 6.97% increase in GP; a unit increase in ITR associates with ≈0.06% increase in GP; a 1-day increase in APD associates with ≈0.06% increase in GP; a unit increase in CCC associates with ≈0.006% decrease in GP (CCC coefficient ≈ −0.0061).
• Moderation (interaction) effects: • Interest rate × APD: Significant negative interaction; higher interest rates combined with longer APD reduce GP, implying that rising borrowing costs make extended supplier payment periods detrimental. • Interest rate × CCC: Interaction term is negative in several models, indicating that higher interest rates exacerbate the adverse effect of longer CCC on GP. • Exchange rate × CCC: Positive interaction; when the exchange rate rises (domestic currency depreciation) and CCC lengthens, the adverse impact of CCC on GP is mitigated or reversed, improving performance in some models. • Exchange rate × APD: Evidence suggests firms that can pay earlier (lower APD) benefit more when the exchange rate increases; higher exchange rate combined with longer APD can harm GP.
• Overall, H1–H3 are supported: CCC negatively relates to GP, and both exchange rate and interest rate significantly moderate the CCC–GP and APD–GP relationships.

The findings confirm that tighter working capital cycles (shorter CCC) enhance profitability in the fuel and energy sector. However, the profitability impact of WCM decisions depends materially on macro conditions. Rising interest rates increase the cost of financing current assets and payables, amplifying the negative effect of a longer CCC and making extended supplier payments less beneficial. Conversely, exchange rate movements moderate WCM effects: domestic currency depreciation, common in import-dependent energy firms, can alter the profitability consequences of holding inventories and receivables, and in some cases a longer CCC under higher exchange rates aligns with improved gross profit, potentially due to pricing power or pass-through effects. These results underscore the need to integrate macroeconomic expectations into working capital policies—particularly APD and CCC management—to sustain profitability. The dynamic model indicates path dependence in profitability, and that scale (size) and liquidity management play additional roles.

This study contributes evidence from a developing economy’s fuel and energy sector that: (1) efficient WCM—especially a shorter CCC—improves gross profit; (2) macroeconomic variables (interest rate, exchange rate) directly affect profitability and significantly moderate the WCM–performance relationship; and (3) firm size, leverage, liquidity, and operating cash flows materially shape outcomes. Practically, managers should tailor APD and CCC policies to prevailing interest and exchange rate conditions, maintain adequate liquidity, and manage leverage prudently. Future research could extend the time horizon, expand to more firms and sectors, incorporate additional macro factors (e.g., inflation, commodity prices), explore alternative performance measures (e.g., ROA, Tobin’s Q), and employ robustness checks with alternative identification strategies for endogeneity.

• Short panel: six years (2013–2018), limiting long-run inference and exposure to structural breaks.
• Data are secondary and rely on the accuracy of State Bank and firm-reported statistics.
• Small, sector-specific sample (21 KSE-listed fuel and energy firms); results may not generalize beyond this sector or to other economies.
• Potential sensitivity to model specification and instrument proliferation in GMM despite diagnostic checks.