Pay, Performance, and the AI Talent Race

Fresh from the stage at NACD’s Leading Minds of Governance & Technology event in Boston, Farient Partner Angela Moe offers a timely perspective on using incentives to drive the right outcomes, choosing key performance indicators that stand up to scrutiny, and competing for scarce AI talent without losing sight of culture. In this conversation, she examines what these challenges mean for boards overseeing compensation, talent, and risk.

How do you explain the connection between compensation design and the risk that employees may stretch, manipulate, or even cross ethical lines to hit a target?

The point of incentives is ultimately to drive behavior. Sometimes, that behavior can be unethical and/or illegal. There are two ways to design incentives to avoid that. The first is upfront. When you design the incentives, be very thoughtful about metric selection because some metrics are more easily manipulated by participants, while others are more objective. Think carefully about the definitions of the measures themselves. Are we setting up the calculations so that we’re judging only the last day of the period, quarter, or year? If so, participants may feel pressured to make their numbers by doing things around the edges. Or are we defining the measurement period as an average over the year? In that case, the results are more likely to be genuine rather than manipulated. The third piece we pay close attention to is the goal ranges, which also tie to payout leverage.

When goal ranges are too narrow, participants are being asked to perform within a very narrow band. That’s hard to do, and in some industries harder than in others. If management doesn’t feel it can perform within that goal range and is operating outside it, that’s not an appropriately designed plan. The payout portion is interesting, too. If, for example, the threshold is set at 50% of target and there is a big cliff, participants who know they are coming in at or very close to threshold performance may engage in unusual behavior.

You distinguish between internal and external KPIs. Why does that distinction matter? What should boards understand about how each can influence behavior?

Companies typically track numerous KPIs internally. The SEC’s Pay Versus Performance rule (effective in 2022) requires companies to disclose measures it considers most important when linking pay and performance. Investors expect to review the proxy, replicate the performance calculations, and clearly understand how results translate into payouts. When we advise on metric selection, we carefully consider which measures are best suited for internal use and which can be disclosed externally with confidence. Investors also value consistency; they generally do not want plans to change dramatically. A consistent framework lets them assess the company’s performance over time and understand its pay-for-performance story.

Once an incentive plan is designed, how should compensation committees assess whether it is driving the intended behavior without creating undue risk?

Every compensation committee should conduct risk assessments. Since 2010, SEC rules have required public companies to determine annually whether their compensation policies and practices create risks that are reasonably likely to have a material adverse effect on the company, and to provide disclosure when that standard is met. We find that it’s really important to have a meaningful assessment done, preferably by an external advisor or by a combination of internal and external advisors working together, to make sure that plans are assessed from a governance perspective, rather than just a competitive one, because a competitive plan is not necessarily an appropriately risk-adjusted plan.

Let’s shift gears and talk about AI and human capital oversight. AI is advancing a company’s ability to gather more—and potentially more meaningful—data faster. What does that mean for boards overseeing talent and compensation?

Competition for AI talent is exceptionally intense, although the underlying dynamic is not unprecedented. Each new technology creates demand for emerging skill sets. Companies are competing for a limited pool of professionals who can integrate the technology and lead enterprise transformation.

Over time, those skill sets will become more widely available, and the market premium will decline. We are currently seeing a premium of approximately 30% for AI-related roles, which is significant. Many companies are also determining how best to integrate AI into their organizations because successful implementation still depends on human expertise. Technology alone cannot advance a strategy; people must lead its execution. Companies may also outsource portions of their AI integration to an external firm or recruit AI leaders from other industries.

One recent example involved a large multinational company based in the Midwest that recruited AI talent from Silicon Valley. The company needed to assess the implications for both its compensation structure and cultural integration. It paid a premium outside its customary pay structure, but the placement ultimately encountered challenges arising from differences in organizational culture. The broader lesson is that companies must consider how AI—and the capabilities required to drive transformation—fit within their strategy and culture.

How does that affect the compensation landscape for everyone when highly sought-after talent is being paid more?

It can cause internal equity issues if you’re not managing it carefully. You might need to build some flexibility into your compensation structures so that salary or long-term incentive (LTI) bands aren’t so rigid. You might also include add-ins and build variability into LTI grants to ensure top talent is motivated and stays.

You eventually connected the AI discussion to succession planning, which I think we all would agree is one of the board’s most important responsibilities. Where do you see boards improving, and where could they do better?

The scope of board responsibilities and stakeholder expectations has expanded considerably. We work most closely with compensation committees, which are increasingly responsible for talent oversight and succession planning as well. Companies have access to extensive workforce data, and most now use some form of electronic human capital management system. The most forward-thinking boards identify and prioritize the information needed to evaluate executives and employees several levels below the C-suite proactively, enabling them to consider the organization’s leadership needs three, five, or 10 years into the future. Relevant information may include retention statistics, compensation history, performance evaluations, work history, expertise, and the areas in which an individual must develop or gain experience to advance to the next role.

As boards navigate these evolving compensation and talent challenges, what should they keep in mind—and how should they account for the growing access to information through AI?

Boards should remember the fundamentals. For example, is there a clearly stated compensation philosophy? Can you explain why you’re making the decisions you are?

And one final AI-related thought: Whenever a board does anything with compensation, assume employees from top to bottom know more—and learn it faster—because of AI.