It’s certainly a rare event, but both ISS and Glass Lewis have recommended voting against a proposal to ratify the appointment of GE’s auditor, KPMG, at the GE annual shareholders meeting. Most often, the issue of auditor ratification is not very controversial—in fact, it’s usually so tame that it’s one of the few matters at annual shareholders meetings considered “routine” (for purposes of allowing brokers to vote without instructions from the beneficial owners of the shares).  Are we witnessing the beginning of a new trend?

In its analysis justifying its negative recommendation, ISS observed that the SEC is currently investigating GE’s revenue recognition practices and internal controls related to long-term service agreements, as well as a $9.5 billion increase in future policy benefit reserves for the GE’s insurance operations. ISS also cites commentators who suggested that GE and its auditors “must have or should have been aware of the issues—particularly the increasing insurance liabilities—for years.” These accounting issues, together with KPMG’s issuance of unqualified reports on the financial statements, were the basis of the recommendation by ISS against ratification of the auditors.  Not to mention that KPMG has been GE’s auditor for a long time—by a “long time,” I mean 109 years!  And notwithstanding major changes in the management team, ISS observed, the board, stressing the benefits of auditor tenure, still reappointed KPMG.

In addition, ISS also saw no discussion in the proxy statement regarding how or whether the board took into account KPMG’s role in GE’s two accounting  problems or any other regulatory issues involving KPMG, including auditor independence allegations (which both ISS and GL indicate were alleged to involve GE) that KPMG settled with the SEC in 2014 or the indictments in 2018 of KPMG employees (see the SideBar below).

Under “ordinary circumstances,” ISS would view the decision about whether to engage a  new auditor to be properly within the purview of the board and the audit committee. However, given the GE accounting issues and the unqualified audit reports by KPMG over a number of years, together with the absence of any insight into the board’s or audit committee’s considerations of the recent legal and regulatory proceedings involving KPMG, “a vote against this proposal is considered warranted.”

Glass Lewis also indicated that it usually supports management’s choice of  auditor except when GL believes the auditor’s “independence or audit integrity has been compromised.” In its analysis, GL raised the same concerns as ISS regarding the SEC investigation of GE and problems at KPMG, noting in particular the large increase in fees to KPMG in the prior year, as well as its long tenure as GE’s auditor, which has “thrown KPMG’s effectiveness and relationship with the Company into question.”

Successful efforts to oust the auditor are rare. According to Audit Analytics, for 2014 through 2016,

“on average, 98% of votes were cast in favor of auditor ratification. The remaining 2% is comprised of votes against auditor ratification (1%) and abstained votes (1%). Similarly, in [Audit Analytics’] last analysis of auditor ratification votes, [it] found that 95 out of 100 times, fewer than 5% of votes were cast against the auditor….There were 8,422 ratification proposals, for example, in which the votes against ratification were between 0% and 1.0% of the total number of votes cast. In all, about 96% of auditor ratification proposals had 5% or less of the votes cast against ratification. On the other hand, just 19 votes (0.2%) had more than 25% of the total votes cast against ratification.”

Audit analytics also found that companies with the highest percentage of votes against auditor ratification have had “going concern, internal control, and disclosure control flags.” Activist Insight cites one example “where both proxy advisers apparently recommended a change, at $1.2 billion market cap company [but] only 15% of shareholders voted against the auditor.”

In this 2017 study, “Auditor Ratification: Can’t Get No (Dis)Satisfaction,” published in Accounting Horizons (your favorite nighttime reading?), the author looked at the impact of proxy advisor recommendations on voting for auditor ratification, the factors used by proxy advisors in making their recommendations, and why there are not more proxy advisor recommendations against ratification of the auditor. The study found that proxy advisor negative recommendations do have a significant effect on shareholder voting, resulting in an increase of approximately 6.7 percentage points in dissenting votes. The author also found that less than 3% of proxy advisor recommendations observed in the study recommended against auditor ratification.

Why were there so few recommendations against auditor ratification relative to expectations based on proxy advisor guidelines?  For that analysis, the author identified examples in the study that she suspected, based on the factors identified in the ISS and GL guidelines, would receive an “against” recommendation and compared that number  against the number that actually did receive an “against” recommendation. She found that proxy advisors were likely to recommend an “against” vote for excessive non-audit services (69%) and legal limitations disclosed in the proxy statement (such as certain damage limitations in the engagement agreement) (85%), while poor audit quality (based on proxy advisors’ own criteria in their guidelines) were much less likely to elicit “against” votes: material restatements (12%) and “aggressive accounting policies,” represented by high absolute value of discretionary accruals (which some associate with earnings management), as calculated under a formula in the study (4%). More specifically, the study identified 512 instances that the author categorized as “suspect” because of aggressive accounting policies, but proxy advisors recommended against ratification of the auditors in only 22 of those instances. In addition, the author found that proxy advisors were less likely to recommend against ratification at companies with stronger returns and companies that disclosed material internal control weaknesses, but more likely to recommend against ratification when the auditor tenure was long or the auditor was a Big 4 firm.

But if audit quality factors are part of the proxy advisors’ guidelines, why didn’t poor audit quality trigger more against recommendations? First, the author suggests that changes may have been made to the auditors prior to the meeting, obviating the need for a negative recommendation. Second, the author suggests that, because “proxy advisors’ guidelines are based on feedback from shareholders[, if] shareholders do not respond to specific signals of poor audit quality, or if shareholders do not place a lot of emphasis on audit quality proxies listed in the proxy advisor guidelines, then proxy advisors will be less likely to issue an Against recommendation for poor audit quality absent obvious signals of audit failure.”

Another reason for the difference, the author suggests, may have been the absence of sufficient information about audit quality issues, leading to “availability bias”: shareholders and proxy advisors may have considered information that was easily retrievable to be “more relevant, and more important for a judgment.’’  As a result, because information about poor audit quality lacked “clear and centralized disclosure,” proxy advisors may have been less willing to issue “against” recommendations based on that factor and shareholders may not have had enough information easily available to justify votes against. The author laments that augmented disclosures previously recommended by the Treasury and the SEC were not adopted. Which, assuming the author’s argument has merit, raises the question of whether the enhanced discussions soon to be required in the auditors’ report regarding “critical audit matters” (ironically, to be prepared by the auditors) might provide some of that clear and centralized disclosure? (See this PubCo post.)

Posted by Cydney Posner