In April 2010, the Securities and Exchange Commission found itself in headlines no regulator wants. The agency’s Office of Inspector General revealed that 33 SEC employees and contractors had been regularly viewing pornography on government computers during work hours. More than half were relatively senior staff, and the details were vivid: One regional office accountant received more than 16,000 access denials from the agency’s internet filter in a single month. The press was merciless (The Atlantic ran a piece titled “Did Porn Cause the Financial Crisis?”), and Congress piled on, with Representative Darrell Issa charging that “high-ranking officials within the SEC were spending more time looking at porn than taking action to help stave off the events that put our nation’s economy on the brink of collapse.”

One might have expected the episode to remain an embarrassing but inconsequential personnel matter. The misconduct involved fewer than 1 percent of the agency’s workforce, was personal and idiosyncratic, and had no obvious connection to the offices’ enforcement competence. The SEC’s own disciplinary response was muted: No implicated employee was fired. In a new study, however, we show that the scandal left measurable fingerprints on the SEC’s core output (investigations and enforcement actions), offering an example of how one financial regulator behaves when its reputation is damaged.

Political scientists’ bureaucratic reputation theory (Carpenter 2010; Carpenter and Krause 2012) is well suited to scandals like this. Reputation is an asset that earns an administrative agency autonomy, legitimacy, and credibility with the audiences that matter, and it has several dimensions: moral (does the agency behave ethically?), performative (can it do the job?), procedural, and technical. The scandal was first and foremost a blow to the SEC’s moral reputation, but the congressional reaction converted it into an attack on the agency’s performative reputation as well, casting the SEC as failing at its core mission in the shadow of the financial crisis.

A key feature of our research setting is that the most natural place to look for a response, investigative activity, was not publicly observable at the office level. During our sample period, data on SEC investigations were confidential, reaching researchers only later through Freedom of Information Act requests (Blackburne et al. 2021). But each office’s investigation activity was visible to SEC leadership in real time through internal case-tracking systems, and Congress, which learned of the scandal through the inspector general, could pressure that leadership. So, if implicated offices increased enforcement, this was unlikely to be an attempt to send a signal to markets or the public: The only audiences that could observe a response were the agency’s own hierarchy and its political principals, a sharp contrast with the corporate setting, where reputation repair is aimed at the market.

Our tests exploit two event dates. When the scandal broke in April 2010, the identities of the implicated offices were known only within the government. After a lawsuit, a judge in March 2011 ordered the SEC to disclose the offenders’ offices (though not their names), and the agency revealed that they worked at its Washington, D.C., headquarters and six of its 11 regional offices: Atlanta, Boston, Chicago, Denver, Fort Worth, and Los Angeles. Implicated offices were thus exposed first to internal and political scrutiny and only later to public naming.

Using the FOIA investigation data, we compare the six implicated regional offices with the five non-implicated offices from 2008 through 2013 in an event study design. Because we compare offices of the same agency in the same quarter, anything that hit the SEC as a whole (the Dodd-Frank Act, the Madoff fallout, agency-wide budget changes) is absorbed; what remains is the comparison between implicated and non-implicated offices in the same quarter. Our analyses present four notable findings.

First, implicated offices more than doubled the rate at which they opened new investigations of public companies in their regions (an increase of roughly 145 percent) in the quarter the scandal became public (the second quarter of 2010) and again in the quarter the offices were publicly identified (the second quarter of 2011). At the mean, the typical implicated office went from starting about two new investigations per quarter to about five.

Second, SEC headquarters responded, too, but selectively. The number of headquarters-led investigations more than doubled following both events, but only of firms in the jurisdictions of implicated offices (the first jump is statistically significant; the second falls short). The pattern is consistent with SEC leadership losing some confidence in the implicated offices and subjecting their regions to closer oversight.

Third, the response went beyond opening case files. Both investigative surges were short-lived, essentially confined to a single quarter, raising the concern that they were window dressing. To probe this, we examine Accounting and Auditing Enforcement Releases (AAERs), which are issued only in serious cases. AAERs in implicated offices’ jurisdictions rise in the quarter the scandal broke and, at a statistically significant level, in the quarter after the offices were publicly named. Because AAERs typically conclude investigations that run for years, these releases cannot stem from the newly opened investigations. Instead, implicated offices appear to have accelerated cases already in their pipelines, producing real, costly enforcement output their overseers could see, though we cannot definitively distinguish substantive effort from strategically timed output.

Fourth, the scandal had labor-market consequences. Using employment data built largely from LinkedIn profiles, we find that staffing at implicated offices fell by about 5.5 percent, but only after the offices were publicly named, through both fewer new hires and more departures. The timing makes sense: Prospective employees could react to an office’s implication only once it became public.

Taken together, the pattern helps identify whom regulators believe they answer to. Implicated offices increased enforcement both when only SEC leadership and Congress knew which offices were involved and when the public learned the identity of those offices, even though office-level investigations remained invisible to the public throughout. We read the second response not as signaling to the public but as public naming of the intensifying scrutiny implicated offices faced from their overseers. Reputation repair inside a regulator, in other words, can run through internal accountability channels, a mechanism absent from the firm-focused reputation literature, where the relevant audience is necessarily external.

Our findings speak to the SEC’s critics and defenders alike. To those who describe the commission as a sluggish bureaucracy indifferent to its standing, our results suggest otherwise: Reputational concerns moved the agency’s observable output almost immediately. But that responsiveness cuts both ways. SEC investigations are costly for target firms; prior research finds median abnormal returns of roughly negative 10 percent when an investigation comes to light (Karpoff et al. 2008). At the same time, consistent, predictable enforcement supports well-functioning markets, and enforcement intensity that swings with a professionalism scandal is hard to square with that ideal.

The cleanest solution, of course, is for regulators to avoid such scandals. Our results suggest that ethics codes, training, and other professionalism guardrails at agencies like the SEC are more than superficial box-checking: They protect the reputational capital that keeps enforcement steady. And with similar scandals surfacing at other regulators (the FDIC’s recent harassment scandal comes to mind), the lesson is unlikely to be SEC-specific.

REFERENCES

Blackburne, T., J. D. Kepler, P. J. Quinn, and D. Taylor. 2021. Undisclosed SEC Investigations. Management Science 67 (6): 3403–3418.

Carpenter, D. P. 2010. Reputation and Power: Organizational Image and Pharmaceutical Regulation at the FDA. Princeton: Princeton University Press.

Carpenter, D. P., and G. A. Krause. 2012. Reputation and Public Administration. Public Administration Review 72 (1): 26–32.

Karpoff, J. M., D. S. Lee, and G. S. Martin. 2008. The Cost to Firms of Cooking the Books. Journal of Financial and Quantitative Analysis 43 (3): 581–611.

James J. Blann is an assistant professor of accounting at the Georgia Institute of Technology’s Scheller College of Business, and Roger M. White is an associate professor of accounting at Arizona State University’s W. P. Carey School of Business. This post is based on their recent article, “How Does the SEC Respond to Reputation Shocks?,” available here.