Understand Glass Lewis’ Approach to Peer Groups

The submission period for proxy filings from February 1 to July 31 is now closed. An additional submission window will be open in July & August for filings made in the second half of the year. Please check back for updates.


Effective January 1, 2020, Glass Lewis’ Pay-for-Performance Model introduced an enhanced peer group methodology that is proprietary to Glass Lewis and leverages the global compensation data and analytics tools of our global partner CGLytics. The rest of the model continues to operate as it did previously.

The new methodology to create Glass Lewis’ peers has an eight-fold increase in the number of independent tests applied to a company’s self-disclosed peers, resulting in a 14% reduction in overlap between a company’s self-disclosed peers and the peers that Glass Lewis uses. We believe this new approach to peer selection will provide you a higher level of confidence in the integrity and independence of both our peer assessment and our Pay-for-Performance model.

What are the specific changes to Glass Lewis’ peer methodology?

The retired peer methodology is based on the strength of connectivity between groups of corporate peers (known as a peer-of-peers methodology), using a company’s self-disclosed peers as a starting point. Although we believe the company’s self-disclosed peers and peer-of-peers are important inputs when evaluating pay, we have become increasingly uncomfortable with the limitations and groupthink created by this method.

We believe a truly independent and robust comparison should also consider investor views of factors such as industry, country and company size. By incorporating the investor view, we can avoid the “echo-chamber” effect and market-wide ratcheting on executive compensation levels that is encouraged by peer-of-peers methodologies that rely exclusively on how companies reference one another in their disclosures.

Our new peer methodology addresses these issues in a measured way. Beginning with a company’s self-disclosed peers, Glass Lewis then includes investor views on both industry-based and country-based peers, in addition to the company’s peers-of-peers. The new methodology then scrutinizes this larger pool of potential peers by introducing additional screens based on corporate revenue, market capitalization and assets; weightings for peers based on the source and frequency of confirmation; and peer rankings based on a strength-of-connection approach that considers all potential peers, not just those resulting from the network effects of corporate disclosures.

As an independent advisor, Glass Lewis has designed this new peer methodology to give clients and companies the highest level of confidence in the independence of our peer assessment, our Pay-for-Performance Model and our Say on Pay recommendations.

Will these new peers change Glass Lewis’ recommendations on Say on Pay proposals?

For some companies, yes. The changes to the Glass Lewis peer methodology may change individual outcomes, which will influence our recommendations in positive or negative directions for some companies.

At a market-wide level, there will be no material change to the distribution of grades awarded or the number of against recommendations as a result of the new peer methodology because our Pay-for-Performance Model distributes grades evenly. Glass Lewis expects to support the same level of Say on Pay proposals it has in recent years.

Will there be any other changes to Glass Lewis’ Pay-for-Performance Model for the 2020 season?

No. We will continue to assess the rigor and independence of our models going forward, as we do with all our policies, with continued research and extensive engagement with our clients and the public companies that they own.

Why did Glass Lewis make these changes?

Glass Lewis decided to change our peer methodology to address the important concerns we noted above that were raised during extensive consultation with our clients and the market in general, as well as our own governance experts who analyze pay and performance at some 18,000 companies every year. This consultation included more than 3,000 engagements with companies, substantial conversations with compensation consultants and advisors, and countless discussions with our investor clients.

The final step of the consultation was the investor-only client survey Glass Lewis conducted in October 2019, which confirmed that relying solely on a peer-of-peer methodology was considered the least effective peer group for measuring pay and performance from an investor’s perspective. Investors were overwhelmingly in favor of industry-based peer groups, followed by a mix of multiple peer groups, country-based peers and a company’s self-disclosed peers. Unsurprisingly, our engagements with public companies has indicated a strong preference for their self-disclosed peer groups and some level of comfort with the peer-of-peer methodologies commonly advocated by their consultants.

Glass Lewis has tried to balance these myriad views by more rigorously testing a company’s self-disclosed peers against peers that are determined independently from the subject company’s disclosure. As always, Glass Lewis’ research will continue to consider companies’ self-disclosed peers in our Say on Pay analysis, but we believe our newly enhanced peer group methodology will provide a greater level of independence and objectivity to the quantitative portion of our research.

Can you summarize the changes above visually?