Accounting for Asset Securitizations: Fair Values and Earnings Management

This study investigates whether the potential for unreliable fair value estimates leads to earnings management. I investigate this question within the context of accounting for an asset securitization, a technique that involves pooling of loans and selling a portion of these pools to third party investors via a Special Purpose Entity (SPE). I hypothesize that firms manage the gains from securitization to meet either the prior year earnings benchmark and/or the consensus analyst earnings forecast. Gains from securitizations are determined, in part, by (a) the volume of loans securitized and (b) the fair value of the portion of pool retained by the securitizer. In particular, for firms close to the earnings benchmark before recognizing the gain from securitization, I hypothesize that managers will manage upward the volume to meet these benchmarks. Once the volume has been determined, managers may manage the assumptions underlying the fair value of the retained interests to increase the gain if the benchmark is not met. However, if the benchmark has been exceeded, managers may manage the assumptions to lower the gain to avoid potential future write-offs of the retained interests. Therefore, I hypothesize that managers manage the fair value assumptions but make no prediction about the direction. I find limited evidence consistent with managers managing the volume securitized to meet the consensus analyst earnings. This limited result is consistent with the events of the period. In August 1998, the Russian government defaulted on its debt, resulting in many investors switching their investments to U.S. Treasury Bills, which made it difficult for many firms to securitize loans. The results also are consistent with firms managing gains upwards to meet prior year earnings targets and managing gains downward to ensure that the analyst benchmark is just met after controlling for the volume securitized. In addition, I find preliminary evidence consistent with new disclosures required by SFAS No 140 being useful at identifying earnings management.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Accountancy

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2015

Author dc:creator

• Shakespeare, Catherine

Contributors dc:contributor

• Ziebart, David A.

• Business Administration, Accounting

Language dc:language

eng

Identifier

(MiAaPQ)AAI3044219

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/87151