Management Forecast Accuracy and Its Role in Financial Reporting Credibility

Abstract

This research introduces a novel, multi-dimensional framework for assessing
management forecast accuracy that moves beyond traditional point-estimate comparisons to incorporate temporal consistency, uncertainty calibration, and narrative
alignment. While existing literature predominantly treats forecast accuracy as a
univariate metric influencing market reactions, this study reconceptualizes accuracy as a dynamic credibility signal embedded within the broader financial reporting ecosystem. We develop and validate the Integrated Forecast Credibility Score
(IFCS), a composite measure derived from machine learning analysis of linguistic
features in management discussion, historical forecast error patterns, and the structural properties of uncertainty disclosures. Applying this framework to a unique
longitudinal dataset of 850 publicly traded firms from 2015 to 2023, we demonstrate
that forecast accuracy, when measured multidimensionally, exhibits a non-linear
and context-dependent relationship with reporting credibility as perceived by sophisticated investors and analysts. Crucially, we find that moderate, well-calibrated
inaccuracy in volatile periods can enhance credibility more than robotic precision,
challenging the prevailing ’more accurate is always better’ paradigm. The results
reveal that credibility is built not through infallibility, but through a transparent,
consistent, and explainable forecasting process, where deviations are rationally justified and linguistically coherent with the firm’s strategic narrative. This research
contributes a new theoretical lens for understanding financial communication, offering practical tools for managers to build trust and for regulators to assess the
quality of forward-looking disclosures