摘要 : Using a generalized aggregate -level difference -in -differences analysis across 32 countries over the 1991-2017 period, we find that the ability of aggregate earnings to predict one-year ahead GDP growth is greater for countries ... 展开 Using a generalized aggregate -level difference -in -differences analysis across 32 countries over the 1991-2017 period, we find that the ability of aggregate earnings to predict one-year ahead GDP growth is greater for countries that adopted International Financial Reporting Standards (IFRS) than those that did not. IFRS adoption also enables aggregate earnings to better predict growth in GDP components and related factors. We show that aggregate accruals drive this effect, not aggregate cash flows. The mechanism for the enhanced predictive ability of IFRS-based aggregate earnings for future GDP growth is due to fair value -based accruals, which we proxy with IFRS-driven special items. In additional analyses, we find that our main results are stronger for adopting countries with greater differences between local accounting standards and IFRS and robust to controls for enforcement. Our findings suggest that IFRS adoption improves aggregate earnings' ability to reflect fundamental economic news in a timely manner. 收起