Commentary on the Dark Side of CEO Social Capital: Evidence from Real Earnings Management and Future Operating Performance

Paul A. Griffin^*
*Correspondence: Paul A. Griffin, Department of Management, University of California, Davis, USA, Tel: +1 530 752 7372, Email:

Keywords

CEO social capital • Real earnings management • Future operating performance

Description

Firm managers have a natural tendency to produce upbeat earnings, most often in the form of beating analysts’ expectations. The stock market also rewards better-than-expected earnings, which increases firm value and executive compensation tied to stock price performance. However, investors and the media have become smarter in recent years, and can generally easily see when a manager pumps up earnings by fiddling with the accounting accruals, such as under-provisioning for bad debts or by accelerating yet-to-beearned revenue into the income statement. Company shareholders can also lose big-time and senior executives can be terminated when the Securities and Exchange Commission investigates or shareholders sue the firm for financial fraud in a class action. The upshot of this is that today’s managers are more likely to resort to subtler ways of cooking the books [1-5].

One of these is called “real earnings management”, whereby a manager purposely alters the firm's cash flow to report earnings based on departures from the timing or structuring of normal or optimal operations. For example, a manager could cut this year’s research or advertising budget. A firm could purposely overproduce to lower the cost of product. These and other techniques produce higher earnings in the short term. Despite this, they are routinely portrayed by managers as smart moves that are part of a longerterm strategy. Jack Welch, the legendary former CEO of General Electric was famous for this. He used the discretion allowable in accounting for merger transactions to produce a steady stream of positive earnings surprises. Following his retirement, however, his successor Jeffrey Immelt wasn’t so lucky or creative. His moves to bolster earnings were soon discovered, GE was sued for accounting fraud, and the company has still yet to fully recover. Even Volkswagen cut back substantially on R&D spending soon after the emission cheating scandal broke supposedly to shelter the high cost of the litigation that it had to book as an accrued expense [6-8].

The activities of Jack Welch and Jeffrey Immelt illustrate well what we report in our recent paper in the Journal of Corporate Finance. Jack Welch, a well-connected and highly-respected CEO, was highly successful in managing earnings, whereas Jeffrey Immelt with fewer connections was not. Our paper indicates that this is a widespread phenomenon. Based on a large sample of CEOs at US firms, we find that well-connected CEOs successfully engage in significantly more real earnings management than less well-connected CEOs. We argue that this occurs because of social capital. Well-connected CEOs (with large social capital) are considered more trustworthy and influential. Their social network connections also allow them to make smarter decisions in their own private interests because they can gather valuable information from a greater number of fellow well-connected CEOs.

But wouldn’t such well-connected CEOs eventually be found out in the long-term? Probably not, because well-connected CEOs have “connections” that help insure that they will have a new CEO position at another firm in the event that they agree to relinquish their current position. In today’s world, CEOs are seldomly fired for cause and so, can leave the firm without a tainted reputation [9,10].

We test this idea in our study. While a CEO with a large network can generate higher earnings in the short-term using low-detection forms of real earnings management (our first result), as our second result, we find that in the longer-term (three years out or more) firm cash flows and operating profits decline more for firms run by well-connected CEOs compared to CEOs with limited social capital. In other words, a CEO's preference for real earnings management would appear to be a symptom of misaligned interests leading to poor operating performance in the long-term. This is the dark side of real earnings management. Consistent with this interpretation, we find that the relations between larger CEO network size and higher amounts of real earnings management and between larger CEO network size and lower future operating performance concentrate in firms with low CEO share ownership. This is where the misalignment of shareholder and CEO private interests is most severe.

These benefits of being well-connected CEO may help explain the pervasive and successful use of real earnings management in practice.

References

1. Brown, Lawrence D and Marcus L. Caylor. "A Temporal Analysis of Quarterly Earnings Thresholds: Propensities and Valuation Consequences." Acct Rev 80(2005): 423-440.
2. Griffin, Paul A and David H. Lont. "Evidence of an Increasing Trend in Earnings Surprises Over the Past Two Decades: The Role of Positive Manager-Initiated Non-GAAP Adjustments.", (2021).
3. Bhattacharya, Nilabhra, Ervin L. Black Theodore E. Christensen and Chad R. "Assessing the Relative Informativeness and Permanence of Pro Forma Earnings and GAAP Operating Earnings.” J Acct Econ 36(2003): 285-319.
4. Cohen, Daniel A, Aiyesha Dey and Thomas Z Lys. "Real and Accrual-Based Earnings Management in the Pre-and Post-Sarbanes-Oxley Periods." Acct Rev 83(2008): 757-787.
5. Graham, John, Campbell Harvey and Shivaram Rajgopal."The Economic Implications of Corporate Financial Reporting." J Acct Econ 40(2005): 3-73.
6. Reinhardt, Uwe E. "Jack Welch and the Lone Ranger Theory." (2009).
7. Goldman, David. "GE pays $50M to settle SEC fraud suit." (2009).
8. Wilmot, Stephen. "The Long-Term Cost of Volkswagen’s Emissions Scandal.", (2019).
9. Liu, Yun. "Outside Options and CEO Turnover: The Network Effect." J Corporate Financ 28(2014): 201-217.
10. Taylor, Lucian. “Why are CEOS Rarely Fired? Evidence from Structural Estimation.” J Financ 65(2010): 2051-2087.