A yardstick should reflect realized financial and non-financial bottom-line performance over the long-term.

Executives should only be rewarded for added value and not for value creation that would have happened anyway.

Simplicity, objectivity, transparency, measurability, non-manipulability, and comparability are important features of a good yardstick.

The first component of a remuneration model is the yardstick to judge performance by. We will discuss the concept of performance, as well as (issues with) metrics for financial and non-financial performance and possible improvements.

What do we consider ‘performance’?

We have used the idea of ‘performance’ rather off-hand and often synonymous with ‘good (societal) outcomes’. In terms of yardstick design, this is insufficient. Good outcomes can also be the result of luck or decisions made by third parties. Conceptually then, we need a more precise definition. We adopt a ‘next-best executive/next-best decision’ approach. Performance is the created value in excess of the value that would have been created in absence of purposeful action by the executive.

From this conceptual point of departure, two empirical questions remain:

  • What metric best reflects value, or:
    • How to account for financial performance?
    • How to account for non-financial performance?
    • How to combine and align financial and non-financial performance?
  • What is the best estimate of value creation that would have resulted without purposeful action by the executive (how to measure the area between the curves in the figure above)?

Continue reading on Financial Performance


Continue reading on Non-Financial Performance



Integration versus parallel targets

An open question in terms of remuneration design is whether targets should be parallel or integrated. Parallel is to mean that I) executives have targets for both financial and non-financial performance, II) that the compensation design puts weights on financial and non-financial performance. Integrated is to mean that executives have a single target that reflects true societal outcomes, and that this measure is automatically reflective of the weights on financial and non-financial performance. In this way, incentives are aligned to create impact for both shareholders and stakeholders.

The parallel approach is principally the same as current practice. Improvements are made through better measurement of these targets, and a governance model that assigns more weight to non-financial vis-à-vis financial outcomes.

The question then becomes how to assign optimal weights to financial and non-financial performance respectively. In deciding on weights, it would be useful if financial and non-financial performance were expressed on the same scale so that (potential) trade-off between financial and non-financial performance is reflected in the same unit of account.

Determining value creation that would have happened anyway

Our concept of ‘performance’ is essentially relative. We deem performance value creation in excess of value creation that would have happened anyway and/or under alternative decisions. Unfortunately, this alternative state of the world is unobserved, and we hence need to estimate it.

There are several ways of dealing with this problem. Companies with equity compensation relative to a peer group already employ one of them: relative performance evaluation. Performance is the share price change between two periods relative to the share price change of the peer. Thus, it is the excess stock return over a peer group. This is akin to what in the econometrics literature is known as a difference-in-difference analysis.

As a matter of principle, this technique is incredibly powerful. For executives, it provides a hedge against common shocks (they can still outperform their peers even in a declining market). For investors, it provides an incentive for executives to perform better than an alternative investment and it supports efficient risk-sharing.

Continue reading