The author of this article writes that benchmarking can be used as a tool for de-escalation as well as escalation in pay issues.

Ken Charman, CEO of uFlexReward, takes a personal look at current issues in this field of executive compensation. (This article was also published in Executive Compensation Briefing in July.)

Without travelling all the way back to 1997 and the Cadbury Committee, it’s worth remembering that the UK corporate governance code was introduced following John Major’s hint that regulation might be needed to respond to public disquiet over executive pay. The Cadbury code led to the establishment of independent remuneration committees, presumably with the intention of restraint. Benchmarking is the “club” that battered that intent. Instead of keeping the lid on it, it has grown into a very sophisticated and advanced process that perpetuates what looks like an arms race to activist investors and wider society.

Pay is now a lot more complex and is heavily incentivised towards performance. Benchmarking proves (at great expense from data providers) that executives are paid what is normal for peers in the market.

The authors of the new (2018) corporate code seem to have sighed and come up with another attempt to flatten the ballistics and tie things down to something more terrestrial.

Based on the theory that publication promotes change, the Financial Reporting Council uses gender pay gap analysis and executive pay ratio reporting to encourage responsible behaviour. The 2018 code asks remcos to take account of whether companies live up to their rhetoric on equal pay, when approving executive remuneration. 

It is early days yet, but lawyers and auditors are advising firms to take this seriously. That means that benchmarking is now expanding from external pay matching into an internal reference of pay across the firm.

In late 2020, the CIPD reported here that 34 per cent of companies link metrics on equality to incentives for executives, while 100 per cent use external pay benchmarking.

This will surely grow under pressure from fair pay activism and investors who are linking fair pay to their interest in ESG. The consequences for not taking this into account are costly. The FT referred in July 2020 to a five-year study by Morgan Stanley that showed share price lagged by 15 per cent in the 18 months after a pay revolt.

So, what are companies showing to remcos? And what should remcos be asking for to satisfy critics and eliminate risk when benchmarking pay across the organisation against the stated policies towards fair pay and equality? 

Especially with companies even more keen to align following the tragic case of George Floyd in the US and increased social pressure centred around ethnicity pay gaps.

The guidance does not stipulate a methodology. Even the fairly lax guidance applied to gender pay gap and executive pay ratios is absent. So internal pay benchmarking is open to interpretation and companies can make do with a number of methods to jump the hurdle. They can use statistically defensible samples (rather than the whole population of employees) and they can step back from using the whole of total reward (including all benefits and shares) because they can claim that equal pay refers to core compensation, bonus and benefits and, anyway, valuing LTIs, stock and some other complex benefits is too complex. 

The risk with this approach is that it might be covering up inequalities that the remco should be able to see and take into account when setting executive pay. The inference here is that the only way to be certain is for internal pay benchmarking to be based on all employees and all forms of reward. In  days gone by this was a data nightmare as the details for total reward were managed in separate systems that were impossible to consolidate without complex manual data extraction and one-off calculations. In this data landscape, annual gender pay gap analysis, even when based on a sample population and a subset of rewards, could take a team of analysts months to produce. Expanding that to include all reward for every employee and then analysis by ethnicity and other factors would be crippling.
 

Fortunately, or unfortunately, if your underlying data shows that you are not as equal in pay as you had hoped, this technical limitation has been solved by digitisation. For less than $10 per employee, per annum, the consolidation and normalisation of all forms of reward data for all employees in a huge global firm can be digitised (very quickly) and delivered in real time. So a remco can ask for pay comparisons of any mix of employee profiles (gender, ethnicity, location) in any grade or job.

Companies which supply data to remcos, should be making it clear that they now supply complete data and not a synthesis or subset. The risk of unintended inequalities being concealed by anything less than this is potentially very costly, not only because of loss of reputation but also because it would reveal that the company could perform better, and be a better citizen, if it was, truly, an equal payer. Now that total reward is digitised there are no technical excuses.

Finally, in this provocative, but supportive paper on benchmarking for remcos, here is another point to ponder. This month I gave a talk about digitising pay at the UK Defence Academy. At dinner the night before, a general asked me how executives could lead organisations where they might earn 25 times or even 100 times the average pay of their employees. He told me his pay ratio at eight times the average seemed acceptable to colleagues who are asked to go in harm’s way. Why do people follow? How is it justified?

Placed on the spot, I rattled off the usual defences; it is benchmarked against peers, it reflects the value added, remcos are obliged to set pay at these levels. Valid, but it sounds hollow said out loud in such company and it probably doesn’t address the concerns of investors, society, government and regulators who would like companies to take more account of normal pay for terrestrial observers.

Bearing this in mind, maybe benchmarking should refer to the work of Nobel prize winner Daniel Kahneman and the tactics of Tom Sawyer in Huckleberry Finn.

They both saw the power of psychological anchoring in referring to benchmarks which influence psychological judgement. As well as paying consultants to benchmark pay packages to match the current, highly complex norms of peers in similar organisations, should benchmarking set anchors that refer to dissimilar organisations and peers? Perhaps not nurses (the populist benchmark) but maybe a general or an NHS brain surgeon? Or a Prime minister…? 

Benchmarking can be used as a tool for de-escalation as well as escalation.

The author
Ken Charman FRSA CEO, uFlexReward.
Visiting fellow, Institute of Ageing, Department of Medicine, Newcastle University.
uFlexReward is a consolidated HR and rewards data platform providing organisations with a granular view of their total labour costs and a technology framework for managing the workforce of the future.