Financial statement change and equity risk

• Author: David L. Stowe,John D. Stowe,Michael H. Senteney
• Published date: 01 January 2020
• Date: 01 January 2020
• DOI: http://doi.org/10.1002/rfe.1069

Rev Financ Econ. 2020;38:63–75. wileyonlinelibrary.com/journal/rfe

|

63

© 2019 University of New Orleans

1

|

INTRODUCTION

Accounting information is essential for the work of investors, managers, and researchers. Given its widespread applications,

it is surprising that there is no holistic measure of the amount of change in financial statements in general use. Statistical de-

composition measures have been used to determine whether a substantial change has occurred in a firm's financial statements.

Unfortunately, a fundamental problem with decomposition measures is that they are undefined if any of the accounts on which

they are based are zero or negative. Researchers have finessed this problem by deleting observations and using highly aggre-

gated financial statements in which negative numbers do not frequently appear. They often used only two asset categories and

two liability categories, such as short‐ and long‐term assets and short‐ and long‐term liabilities. They also do not typically apply

decomposition analysis to income statements.

This paper proposes three alternative “distance” measures to provide indices of the amount of change in financial statements as a

solution to the negative number problem in the computation of statistical decomposition measures. Distance measures can also be called

measures of similarity or measures of proximity. These alternative measures are adapted from cluster analysis, and they enable the use

of detailed balance sheet categories and the ready use of income statements. The properties of these alternative distance measures are

compared, and the measures are related to several empirical measures of corporate equity riskiness and to corporate bond ratings.

The empirical results confirm the logical advantage of the distance measures over decomposition measures. The proposed

distance measures allow detailed financial statements, avoid the necessity of deleting observations, and allow the use of income

statements. The distance measures provide an index of the overall amount of change in the structure of the balance sheet or

income statement. The distance measures are correlated positively to the monthly equity return volatility for a set of Compustat

firms. In addition, the distance measures are correlated with (a) alpha, market beta, and R‐squared in the CAPM, (b) alpha,

market beta, size beta, value beta, and R‐squared in the Fama–French three‐factor model, (c) bond ratings, and (d) corporate

distress (Altman Z‐scores).

Received: 25 January 2019

|

Revised: 6 June 2019

|

Accepted: 7 June 2019

DOI: 10.1002/rfe.1069

ORIGINAL ARTICLE

Financial statement change and equity risk

Michael H.Senteney

|

David L.Stowe

|

John D.Stowe

Ohio University, Athens, Ohio

Correspondence

John D. Stowe, Finance Department, Ohio

University, 222 Copeland Hall, Athens, OH

45701, USA.

Email: stowej@ohio.edu

Abstract

While financial statement analysis is a rich tool, there is no widely used holistic

measure of the amount of change in corporate financial statements. Statistical de-

composition analysis has been employed as an index of the amount of change, but

has fallen into disuse because it does not allow negative accounting numbers. As a

remedy, this paper suggests three distance measures adapted from cluster analysis

that avoid this critical data limitation. We successfully apply these proposed distance

measures to explain the total and systematic risk of stock returns (in the CAPM and

Fama–French model), corporate bond ratings, and corporate distress.

KEYWORDS

accounting statistical decomposition measures, Altman Z‐score, CAPM, corporate bond ratings, distance

measures, Fama–French model, financial statement change

JEL CLASSIFICATION

G11; G12; M41