Financial Prominence and Financial Conditions: Risk Factors for 21st Century Corporate Financial Securities Fraud in the United States

Abstract

Following the global financial crisis wrought by fraudulent activities of prominent US companies and considering the field’s lack of understanding about precursors to such vast corporate illegalities, this study measured firm prominence and financial conditions to identify corporate-organizational factors related to twenty-first century corporate financial fraud. We compiled data on 250+ US public companies involved in corporate securities frauds identified in 1,000+ Securities and Exchange Commission filings over 2005–2013; we randomly selected a comparable control group of 500+ US public companies from Compustat. Based on logistic multivariate regression analyses, marginal profitability, a strong growth imperative, and firm prominence were significant fraud risk factors. Prominent Fortune 500 firms were more susceptible to marginal profitability and/or strong growth-opportunities as risk factors. Findings were robust to various empirical measures and additional controls for undetected fraud. Anomic growth pressures and profit constraints, especially among America’s most prestigious firms, were linked to corporate securities fraud.

Keywords:

• White-collar crime
• corporate crime
• reputation
• status
• strain

Notes

1 For example, WorldCom, while misrepresenting its financials, charged consumers low prices and increased its market share. Competitors Sprint and AT&T responded by cutting their prices, spurring substantial, unsustainable declines in operating margins. In response, AT&T laid off 20,000 employees, cut $7.5 billion in costs, and ultimately split into three units, a decision that undermined AT&T’s value. WorldCom’s actions distorted the competitive market in telecommunications during its fraud, but also long after, by impacting industry-wide opportunities for external financing, expectations about profitability, rivals’ stock prices, and so forth. Moreover, federal and state governments used the false disclosures in setting regulatory fees and implementing other regulatory decisions (Velikonja 2013).

2 That is, these firms were not subject to an Accounting and Auditing Enforcement Release by SEC. Recognizing that not all financial frauds were detected, we took several steps in our supplemental analyses to address such concerns and test robustness of findings.

3 The US Securities and Exchange Commission (SEC) requires all publicly traded firms to file an audited financial statement.

4 Using another approach to measure securities fraud/financial wrongdoing, some researchers employed financial restatements (Prechel and Morris 2010). However, only 20-40% of restatements result in SEC enforcement actions (Karpoff, Lee, & Martin 2008), yielding a high rate of false positives. Beyond our scope, future research might consider whether our study findings hold using financial restatements as the fraud outcome.

5 Moreover, an anomic environment shaped by relative deprivation generates “milieu” effects that sustain and spread the attenuation of norms and regulatory guidelines within and across firms. The fraternal sense of relative deprivation contributes to a shared definition of the problem and to a collective breakdown in trust of the system, giving rise to redefined “rules” and other cultural adaptations of norms and values. Individual moral orientations shift as vocabularies of motive and techniques of neutralization become more widespread through a firm or industry.

6 This projection might vary depending on organizational characteristics and type of white-collar corporate crime under scrutiny, an issue we return to in the discussion.

7 If undetected fraud firms were wrongly categorized among the control group, group differences in financial or other factors would be harder to detect and significance levels might be underestimated. However, few unofficial options exist to measure securities fraud. We consider later the matter of enforcement bias and steps we took to address it.

8 Comparisons of sample characteristics showed no differences among the included/excluded frauds that would skew results; regression results were robust to various substitution techniques for missing data.

9 A randomly selected control group was preferred over a matched case-control design for several reasons. A case-control design would match fraud-firms with equivalent non-fraud firms to eliminate confounding factors and gain efficiency. However, overmatching can occur if the matching criteria are associated with independent rather than dependent variables, or if they are variables along the causal pathway, leading to loss of information or even spurious results. Matching variables are chosen a priori and therefore eliminated as objects of analysis – but this seemed premature for a first study using newly available fraud-data. Second, matched-control groups are no longer representative of the population of public firms as a whole, but our aim was to identify organizational and financial factors that distinguished fraud from non-fraud firms. Finally, unmatched designs are more efficient and produce less biased results than matched-case controls for most studies where controls outnumber cases by a factor of two or more, and the n is not small (Rose and van der Laan 2009).

10 We elected to present results for the sample including all types of schemes, but because corporate securities fraud can take different forms, we also investigated multivariate results for only the most common act of asset misstatements (results upon request). Conclusions were unchanged when we excluded bribery and self-dealing. Case counts for bribery and self-dealing were low, however exploratory multivariate regression results suggested broad similarities for each with asset misstatements. Insider trading was driven somewhat more by growth opportunity factors (and less so by profitability concerns); bribery, too, was responsive to growth pressures and especially reputation (Fortune 500 status). Future research should examine these nuances further.

11 Linear effects of profitability on fraud were positive and significant; however, theoretically driven inspection of the data revealed profitability exerted a non-linear effect, so we elected to model its effect via dummy variables.

12 Because replacement cost of assets is difficult to measure, the Tobin’s Quotient formula generally substitutes a firm’s book value of equity. Book value is the company’s historical cost of assets minus accumulated depreciation, or its value according to general accounting practices.

13 Market value of equity, or market capitalization, equals stock price per share times total number of shares.

14 Prominence is of course multi-dimensional (Sorenson 2014) and conceptualized as comprised of status (relative social standing, or position in the hierarchy) and reputation (perceived positive distinction among peers) (King and Whetten 2008).

Additional information

Funding

This work was supported by the National Institute of Justice [2013-R2-CX-0055 to D.S. & J.S.]