The direct return to the auditor of fictitious confirmations is an important assumption embodied in the confirmation procedure. The effectiveness of accounts receivable confirmations depends on auditors gaining assurance that both the account balance is correct and that the account is valid. To determine that the account balance is correct, auditors rely on recipients to provide the correct information. However, the effectiveness also depends on the validity of the accounts. If the account is fictitious because of a fictitious name and/or address, the auditor depends on the National Postal Service to return the letter. Some fictitious confirmations could still be delivered, so then the auditor relies on the recipients to return the confirmations. The purpose of this paper is to report the results of a study that tested the performance of the U.S. Postal Service (or the recipient if the confirmation was delivered) in returning confirmation requests containing fictitious names and/or addresses. In this study, seven hundred twenty confirmations were mailed to twelve U.S. cities. The independent variables tested were type of recipient (individual or business), size of city (large or small), type of error (fictitious name and fictitious address, fictitious name and valid address, and valid name and fictitious address), and type of zip code (9-digit or 5-digit). The results indicated that, in total, 83.9 percent of the confirmations were returned, and the mean time to return the confirmations was 14.1 days. The most important conclusions from this study relate to the type of error variable. When both the name and address are fictitious, or when a valid name and fictitious address are used, the Postal Service does an excellent job of returning these confirmations, and the auditor should rely on the Postal Service. However, when the confirmations contain a fictitious name and valid address, only slightly less than half of the confirmations are returned by the Postal Service, and the auditor needs to be concerned by this potential risk. Also, for these confirmations that are delivered, the recipients cannot be relied upon to return the confirmations. This indicates that auditors should be skeptical of all nonreturned confirmations, and conduct extensive alternative procedures

În prezentul articol autorii prezintă rezultatele unui studiu care a testat eficienţa serviciului poştal al SUA în revenirea cererilor de confirmare care conţin nume fictive şi/sau adrese. Cercetarea în cauză se bazează pe şapte sute douăzeci de confirmări care au fost trimise în douăsprezece oraşe din SUA. Cele mai importante concluzii din acest studiu se referă la ponderea confirmărilor care au fost returnate, timpul mediu de returnarea confirmărilor şi tipul aparent de eroare..