Improving assessment capability for accounts receivable pool financiers An empirical investigation

Xiao Song (First Author), Xiande Zhao (Participant Author), Lujie Chen (Participant Author), Hao Ying (Participant Author)

Research output: Contribution to journalJournal

2 Citations (Web of Science)


Purpose The accounts receivable pool (ARP) is an e-invoice management system that provides suppliers with easy access to financial service. The purpose of this paper is to focus on the strength and weakness of ARP mechanism and suggest efficient methods to identify creditworthy borrowers. Design/methodology/approach By decomposing the sales records of 348 ARP borrowers and predicting the occurrence of overdue incidences, this study first portrays the creditworthy borrowers by sales features. Then, content analysis was applied to measure the loadings of soft and hard information, and examined the effectiveness of different information structures in creditworthiness assessment. Findings For ARP borrowers, upward trend and low volatility reveal their creditworthiness. In order to identify creditworthy borrowers beforehand, ARP financiers who have elaborated more soft information and less hard information can perform better. Originality/value This study first discussed ARP finance from a critical perspective and underlines borrower assessment to eliminate the defect of loose recourse. The empirical evidence presents the sales features of creditworthy borrowers. Moreover, the results suggest an efficient approach for ARP financiers to conduct better assessment.
Original languageEnglish
Pages (from-to)547-566
JournalIndustrial Management & Data Systems
Issue number3
Publication statusPublished - 2020

Corresponding author email


  • Accounts receivable pool finance
  • Information structure
  • Sales information

Indexed by

  • ABDC-A
  • SCIE
  • Scopus


Dive into the research topics of 'Improving assessment capability for accounts receivable pool financiers An empirical investigation'. Together they form a unique fingerprint.

Cite this