Earnings benchmarks, earnings management and future stock performance of Chinese listed companies reporting under ASBE-IFRS

Abstract

Purpose Prior literature has revealed three key earnings benchmarks: earnings level; earnings change; and analysts’ expectations. The purpose of this paper is twofold. First, the authors seek to establish which earnings benchmark induces the largest extent of earnings management. Second, the authors explore the implications of earnings management on firm future performance. Both of these purposes are investigated for Chinese listed companies during China’s IFRS/ISA reporting era. Design/methodology/approach The authors rely upon the unique regulations and incentives for Chinese listed companies in order to develop four testable hypotheses. Next, the authors employ both logistic and ordinary least squares regressions to test the hypotheses. Findings The results suggest that Chinese listed firms have the highest level of income increasing discretionary accruals around the earnings level benchmark, followed by the earnings change benchmark. The authors do not find any evidence of earnings management to beat analysts’ expectation. In addition, the authors find evidence that Chinese listed firms with relatively high level of earnings management and low earnings exhibit relatively weak future stock performance. Originality/value The findings are the first to document an earnings management benchmark hierarchy with respect to the extent of income increasing discretionary accruals, while simultaneously establishing a link between earnings management and firm future stock performance, for Chinese listed companies. The findings are valuable for regulators and investors by suggesting that management intervention in the reporting process during China’s IFRS/ISA reporting era may act to circumvent delisting regulations and cloud earnings signal for firms that beat certain earnings benchmarks.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 21 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.