Incorrect Inferences When Using Residuals as Dependent Variables

We analyze a procedure common in empirical accounting and finance research where researchers use ordinary least squares to decompose a dependent variable into its predicted and residual components and use the residuals as the dependent variable in a second regression. This two‐step procedure is used to examine determinants of constructs such as discretionary accruals, real activities management, discretionary book‐tax differences, and abnormal investment. We show that the typical implementation of this procedure generates biased coefficients and standard errors that can lead to incorrect inferences, with both Type I and Type II errors. We further show that the magnitude of the bias in coefficients and standard errors is a function of the correlations between model regressors. We illustrate the potential magnitude of the bias in accounting research in four commonly used settings. Our results indicate significant bias in many of these settings. We offer three solutions to avoid the bias.     

The Economics of Disclosure and Financial Reporting Regulation: Evidence and Suggestions for Future Research

This paper discusses the empirical literature on the economic consequences of disclosure and financial reporting regulation, drawing on U.S. and international evidence. Given the policy relevance of research on regulation, we highlight the challenges with (1) quantifying regulatory costs and benefits, (2) measuring disclosure and reporting outcomes, and (3) drawing causal inferences from regulatory studies. Next, we discuss empirical studies that link disclosure and reporting activities to firm‐specific and market‐wide economic outcomes. Understanding these links is important when evaluating regulation. We then synthesize the empirical evidence on the economic effects of disclosure regulation and reporting standards, including the evidence on International Financial Reporting Standards (IFRS) adoption. Several important conclusions emerge. We generally lack evidence on market‐wide effects and externalities from regulation, yet such evidence is central to the economic justification of regulation. Moreover, evidence on causal effects of disclosure and reporting regulation is still relatively rare. We also lack evidence on the real effects of such regulation. These limitations provide many research opportunities. We conclude with several specific suggestions for future research.     

Fifty‐year overview of judgment and decision‐making research in accounting

As part of the celebration of the fiftieth anniversary of AFAANZ, we consider the breadth of judgment and decision‐making (JDM) experimental research in accounting over that 50‐year period. Our review is divided by decade and between auditing, financial accounting and management accounting. In four major journals, we found 5745 papers between 1970 and 2009, which we consider impressive and strong support for the opportunity to publish in this field. Our aim is to encourage more JDM research from Australians and New Zealanders, and to allow researchers in particular specialisations to get a better understanding of the JDM research in other specialisations.     

Regression-Based Tests of the Market Pricing of Accounting Numbers: The Mishkin Test and Ordinary Least Squares

The test developed in Mishkin [1983] (hereafter, MT) is widely used to test the rational pricing of accounting numbers. However, contrary to the perception in the accounting literature, the exclusion of variables from the MT's forecasting and pricing equations leads to an omitted variables problem that affects inferences about the rational pricing of accounting variables. Only if the omitted variables are rationally priced is their exclusion irrelevant. Failure to recognize this issue leads accounting researchers to employ the MT without appreciating how omitted variables affect the inferences they draw. We demonstrate that when additional explanatory variables are included in the MT, the rational pricing of accruals is not rejected. That is, the accrual anomaly documented in Sloan [1996] vanishes when additional explanatory variables are incorporated into the MT. We also show that in accounting research settings, where samples are large, ordinary least squares (OLS) is equivalent to the MT. As a result, accounting researchers should consider using OLS or be more explicit about the exact advantages of the MT over OLS in their research setting.     

Does incorporating non-linearity into discretionary accrual models improve their performance?

Using a large sample of firms that restated earnings, this study investigates whether incorporating non-linearity (conditional conservatism) into discretionary accrual models improves their performance in detecting earnings management. The findings of this study are important because discretionary accrual models play a prominent role in several streams of accounting research and the models' ability to isolate the discretionary (managed) component from the non-discretionary (unmanaged) component of total accruals is critical. If the conventional linear discretionary accrual models are mis-specified, it is likely to result in misleading inferences about earnings management behavior. The findings indicate that the non-linear specification improves the performance of most linear models. The findings also indicate that a more sophisticated linear model that incorporates a performance measure and a future growth measure outperforms other simple models.     

Investment Dynamics and Earnings‐Return Properties: A Structural Approach

We propose the standard neoclassical model of investment under uncertainty with short‐run adjustment frictions as a benchmark for earnings‐return patterns absent accounting influences. We show that our proposed benchmark generates a wide range of earnings‐return patterns documented in accounting research. Notably, our model generates a concave earnings‐return relation, similar to that of Basu [1997], and predicts that the earnings‐return concavity increases with the volatility of firms’ underlying shock processes and decreases with the level of firms’ investments. We find strong empirical support for these predictions. Overall, our evidence suggests that our proposed benchmark is useful for understanding the joint dynamics of variables of interest to accounting research (e.g., earnings, returns, investment, market‐to‐book) absent accounting influences, a necessary precondition for inferring the effects of accounting from these dynamics.     

Banks’ Financial Reporting and Financial System Stability

The use of accounting measures and disclosures in banks’ contracts and regulation suggests that the quality of banks’ financial reporting is central to the efficacy of market discipline and nonmarket mechanisms in limiting banks’ development of debt and risk overhangs in economic good times and in mitigating the adverse consequences of those overhangs for the stability of the financial system in downturns. This essay examines how research on banks’ financial reporting, informed by the financial economics literature on banking, can generate insights about how to enhance the stability of the financial system. We begin with a foundational discussion of how aspects of banks’ accounting and disclosures may affect stability. We then evaluate representative papers in the empirical literature on banks’ financial reporting and stability, pointing out the research design issues that empirical accounting researchers need to confront to develop well‐specified tests able to generate reliably interpretable findings. To this end, we provide examples of settings amenable to addressing these issues. We conclude with considerations for accounting standard setters and financial system policy makers.     

Strengthening causal inferences in positivist field studies

This essay discusses how incorporating qualitative analyses and insight in positivist field studies can strengthen researchers’ ability to draw causal inferences. Specifically, I review how the rich institutional knowledge available in field settings can be used to increase internal validity by improving the specification of empirical models and tests and by providing greater insight into statistical results, particularly through the investigation of the causal processes linking the accounting practices and outcomes.     

Robust anomalies? A close look at accrual‐based trading strategy returns

The last 40 years have seen an extensive literature documenting so‐called anomalies in major capital markets. Evidence of ‘abnormal’ returns associated with trading strategies based on readily observable phenomena such as accounting‐based data involves experimental design choices that can be expected to influence the results. We show how evidence of an accrual anomaly in Australia is sensitive to research design specifications such as the choice of proxy for total accruals; the definition of abnormal returns (i.e. the return generating model); the impact of data trimming as a response to exceptionally large returns; and the choice between value and equal weighting of returns. We show that research design choices do matter and help reconcile conflicting prior evidence of any accrual anomaly in Australia. More broadly, our results suggest the need for caution in drawing inferences from trading strategy tests which claim to identify anomalies.     

A Real Effects Perspective to Accounting Measurement and Disclosure: Implications and Insights for Future Research

Accounting measurement and disclosure rules have a significant impact on the real decisions that firms make. In this essay, we provide an analytical framework to illustrate how such real effects arise. Using this framework, we examine three specific measurement issues that remain controversial: (1) How does the measurement of investments affect a firm's investment efficiency? (2) How does the measurement and disclosure of a firm's derivative transactions affect a firm's choice of intrinsic risk exposures, risk management strategy, and the incentive to speculate? (3) How could marking‐to‐market the asset portfolios of financial institutions generate procyclical real effects? We draw upon these real effects studies to generate sharper and novel insights that we believe are useful not only for the development of accounting standards, but also for guiding future empirical research.     

A manipulationist view of causality in cross-sectional survey research

This essay discusses issues related to establishing causal relationships in empirical survey research. I adopt a manipulationist view of causality because it matches the context of (management) accounting research where we are commonly interested in studying the effects of changes. Strictly speaking, causal relationships cannot be unequivocally proven when the researcher employs cross-sectional surveys—that is, correlation is not causation. Notwithstanding, survey research can be fruitfully engaged to inform pertinent management accounting topics. I discuss four “markers” of causality—theoretical coherence, empirical covariation, temporal/physical separation, and internal validity—and how the researcher can lever these to suggest compelling survey-based inferences. Of these four markers, I particularly emphasize the first as I believe that one piece of any reasonable observer’s considerations will be whether the proffered causal relationships are theoretically plausible. Moreover, a stronger theoretical foundation also helps causal inference by suggesting a reasonably complete set of control variables that are useful to eliminate alternative explanations. Overall, I focus rather pragmatically on the limitations of causal inference when using the survey method and what may be done to try and alleviate, although not eliminate, them.     

Addressing unobserved endogeneity bias in accounting studies: control and sensitivity methods by variable type

Together with their associated statistical routines, this paper describes the control and sensitivity methods that can be employed by accounting researchers to address the important issue of unobserved (omitted) variable bias in regression and matching models according to the types of variables employed. As with other social science disciplines, an important and pervasive issue in observational (non-experimental) accounting research is omitted variable bias (endogeneity). Causal inferences for endogenous explanatory variables are biased. This occurs in regression models where an unobserved (confounding) variable is correlated with both the dependent (outcome) variable in a regression model and the causal explanatory (often a selection) variable of interest. The Heckman treatment effect model has been widely employed to control for hidden bias for continuous outcomes and endogenous binary selection variables. However, in accounting studies, limited (categorical) dependent variables are a common feature and endogenous explanatory variables may be other than binary in nature. The purpose of this paper is to provide an overview of contemporary control methods, together with the statistical routines to implement them, which extend the Heckman approach to binary, multinomial, ordinal, count and percentile outcomes and to where endogenous variables take various forms. These contemporary methods aim to improve causal estimates by controlling for hidden bias, though at the price of increased complexity. A simpler approach is to conduct sensitivity analysis. This paper also presents a synopsis of a number of sensitivity techniques and their associated statistical routines which accounting researchers can employ routinely to appraise the vulnerability of causal effects to potential (simulated) unobserved bias when estimated with conventional regression and propensity score matching estimators.     

Writing accounting research for publication and impact

This paper suggests guidelines for writing accounting research papers that may be useful to accounting researchers who want to improve the chances that their work will be accepted by top journals and, if accepted, will have a significant impact on their field. The advice is illustrated mainly in the context of experimental research in audit judgment, but it is also applicable to other types of accounting research. Several guidelines for communicating the results of accounting research are offered, and these are broken down by the typical sections and subsections that are (or should be) included in well-written accounting research papers. While guidelines such as these cannot guarantee success (either in publication or impact), they nevertheless may provide a valuable framework for communicating and evaluating accounting research.     

Discussion of “CEO discretionary power,unconstrained stock ownership,and stock trading: Theory and evidence”

Hong (2023) provides both an analytical model and empirical archival evidence to explain why CEOs hold vested own-firm shares when doing so comes at the cost of reduced CEO portfolio wealth diversification. I discuss Hong (2023) in terms of the intuition provided by its analytical model and the inferences one can draw from its empirical results. Moreover, I briefly discuss (the lack of) multi-methods research in accounting and consider how accounting scholars can add insight to the cross-disciplinary literature on executive power and contracting.     

Accounting based valuation models: what have we learned?

The present survey article formed the basis of a presentation by G. Richardson to the 8 July 2003 plenary session of the Accounting and Finance Association of Australia and New Zealand Conference in Brisbane, Australia. The present article reconciles the historical and forecasting branches in the published accounting literature. Prior survey articles have primarily focused either on the historical branch or the forecasting branch. While these approaches have yielded useful insights, they do not attempt to synthesize the link between the two branches of the published literature. An obvious link between the two branches is that the Ohlson model begins with the Residual Income Model as an initial assumption. We believe that there are other links that need further emphasis. In the process, we also review the empirical issues and the evidence within these two branches. We know of no paper to date that has surveyed the empirical evidence on both the historical and forecasting branches of the published literature. In particular, we draw inferences on the following question: on balance, what have we learned from nearly a decade of research on accounting based valuation models and its applications?     

Co‐authorship Network Analysis in the Accounting Discipline

The purpose of this study is to explore co‐authorship structures in the accounting discipline through social network analysis. For this purpose, we hand‐collected the authorship data of 10 863 papers published in 22 accounting journals listed on the Web of Science for the period 2000–2016. The findings indicate that the proportion of multi‐author papers has increased over time while the trend away from sole authorship has continued. The network indicators and visualisations reflect that accounting research manifests a small‐world property. Despite the large network size, a high number of ties among nodes, the size of giant components and a high clustering coefficient all support this inference. We hope that emphasis on collaboration metrics motivates and encourages researchers to engage with other scholars both locally and abroad. Furthermore, the development of research networks in the analysis period indicates that the trend is moving towards more collaboration; thus, being part of a team or team‐making is key to integration with other scholars.     

Detecting Accounting Fraud in Publicly Traded U.S. Firms Using a Machine Learning Approach

We develop a state-of-the-art fraud prediction model using a machine learning approach. We demonstrate the value of combining domain knowledge and machine learning methods in model building. We select our model input based on existing accounting theories, but we differ from prior accounting research by using raw accounting numbers rather than financial ratios. We employ one of the most powerful machine learning methods, ensemble learning, rather than the commonly used method of logistic regression. To assess the performance of fraud prediction models, we introduce a new performance evaluation metric commonly used in ranking problems that is more appropriate for the fraud prediction task. Starting with an identical set of theory-motivated raw accounting numbers, we show that our new fraud prediction model outperforms two benchmark models by a large margin: the Dechow et al. logistic regression model based on financial ratios, and the Cecchini et al. support-vector-machine model with a financial kernel that maps raw accounting numbers into a broader set of ratios.     

Accounting and Accountability in Fiji: A Review and Synthesis

This paper reviews accounting and accountability research in Fiji. The review is based on 47 papers published since 1978 in accounting refereed journals, professional journals, edited book chapters and theses and other refereed journals outside accounting. In addition to categorisation of the reviewed papers according to accounting topics, theories and methods of data collection, some themes to which the papers could be related are discussed. Financial reporting/accountability research is the most popular research in Fiji followed by new public management research, with sustainability accounting research the third most popular. The paper's findings suggest some directions for future accounting research in Fiji and where the data can possibly be sourced for such research. We conclude that more research is needed in the Fijian context that focuses on accounting and the state, performance auditing, indigenous accounting, SMEs and accountability.