The usefulness of accounting measures in predicting future cash flow

Table of contents

1. Introduction

2. Literature Overview

3. Research design

3.1 Models and methodology

3.2 Data and sample

3.3 Descriptive statistics

4. Results

4.1 In-sample regressions

4.1.1 Aggregate earnings and cash flow

4.1.2 Disaggregated earnings

4.1.3 Cash flow decomposition

4.2 Rolling window regressions

4.2.1 Parameter estimates

4.2.2 Summary statistics of the model-based and analysts’ forecasts

4.2.3 Error measures and performance of the model-based forecasts

4.2.4 Relative performance of the model-based and analysts’ forecasts

5. Additional tests

5.1 Undeflated data

5.2 Alternative earnings variable

6. Conclusion

Research Objectives and Themes

This thesis investigates the relative predictive ability of accounting earnings and cash flow measures in forecasting future cash flows from operations. It aims to evaluate the "accrual vs. cash flow" debate in accounting research and determine whether disaggregating these components provides superior forecasting accuracy compared to aggregate measures.

• Comparison of the predictive power of aggregate accounting earnings versus cash flow from operations.
• Examination of the incremental information content provided by accrual components.
• Analysis of time-series trends in the explanatory power of these variables.
• Evaluation of mechanical model-based forecasts against analysts' cash flow forecasts.
• Impact of research design choices, such as data frequency and estimation techniques, on forecasting outcomes.

Excerpt from the book

Summary of Chapters

1. Introduction: This chapter defines the core research problem regarding the predictive superiority of earnings over cash flows and outlines the primary aim of the study.

2. Literature Overview: This section reviews existing academic research, highlighting the mixed empirical findings and the ongoing debate surrounding accrual-based versus cash-based performance measures.

3. Research design: This chapter details the regression models and methodology used to test predictive ability, including the definition of earnings, accrual, and cash flow variables.

4. Results: This chapter presents the empirical findings from in-sample and rolling window regressions, comparing various model specifications and performance metrics.

5. Additional tests: This chapter conducts robustness checks using undeflated data and alternative earnings definitions to confirm the stability of the empirical conclusions.

6. Conclusion: The final chapter summarizes the findings, asserting the predominance of cash flow as a predictor and discussing the implications for market participants and standard setters.

Keywords

Cash flow from operations, Accrual accounting, Financial reporting, Earnings prediction, Forecasting, Predictive ability, Accounting standard setters, Cash flow decomposition, Accrual components, Financial performance, Rolling window regressions, Analysts' forecasts, Operating cycle, Accounting conservatism, Forecast accuracy.

Frequently Asked Questions

What is the core focus of this research?

The study examines whether accrual-based accounting earnings are superior to cash flow from operations in predicting a company's future operating cash flows.

What are the primary themes discussed in the work?

Key themes include the evaluation of accounting standard setter assertions, the incremental predictive power of accruals, the decomposition of cash flows into core and non-core components, and the comparison of mechanical models with analysts' forecasts.

What is the central research question?

The study asks which accounting variable—earnings or cash flow—serves as a more reliable indicator for forecasting a firm's future cash flows from operations.

Which methodology is employed in the empirical analysis?

The author employs pooled cross-sectional regressions on annual data, supplemented by 10-year rolling window regressions to test out-of-sample forecasting performance.

What is covered in the main section of the paper?

The main section investigates the predictive ability of aggregate and disaggregated earnings, decomposes cash flow into various operating and non-operating components, and evaluates models using different error metrics.

Which keywords best characterize this study?

Important keywords include Cash flow from operations, Accrual accounting, Financial forecasting, Predictive ability, and Earnings decomposition.

How does the author account for the difference between core and non-core cash flows?

The author follows the approach of Cheng and Hollie (2007) by categorizing components related to primary business activities as core, while grouping items like tax and interest as non-core to test their respective persistence.

Why are analysts' forecasts included in the study?

Analysts' forecasts serve as a real-world benchmark to evaluate the validity and performance of the author's mechanical model-based forecasts.

How does this study address the influence of firm size?

The author ranks observations by total assets to test how predictive accuracy changes with firm size, finding that relative absolute error impacts often diminish for larger companies.