Why Fair Targets are Unfair: Setting Meaningful Growth Targets

Setting meaningful growth targets in an uncertain future is challenging, especially when the ambition and overconfidence typical in executive board situations can lead to overly optimistic goals. Most targets are set too high, as the exercise intuitively conveys, leading to underpaying and a bias toward downsizing because growth targets are underrepresented in pay systems. This article explores the problem of meaningful growth targets and proposes a solution for fairer pay based on indexed performance.

Stakeholder: Executives, Shareholders

Risk: Wrong Incentives

Management: Be careful with growth targets, caps, and floors. Better index pay, as we show here

Last week, we established that caps and floors in bonus systems are notoriously hard to set. For the business simulation we did, we knew the average growth rate of large companies was 8%. Based on that assumption, we challenged ourselves to set sensible caps and floors for a variable compensation plan. If you haven’t done the simulation, do it here before reading this second episode.

Negotiated Targets Too High

Today, we ask ourselves what meaningful growth targets are if we have historical information and know little about the future, which is typical for most companies. We do this in two exercises.

First, we think about growth targets more generally. Let’s assume the industry history for the construction sector is plotted below. What you see in the chart is an actual case. The numbers are genuine and established for one of Obermatt ’s clients.

In the graph, we can see that the construction industry's median growth fluctuates. Sometimes it is higher, sometimes lower, and sometimes even in negative territory. We are now in the last year and assume that we already know this year’s growth will be negative -4.5%. We also see the median of 1.9% and the average of 0.9% over all the years plotted in the graph and displayed in the upper left corner. Based on this historical information, the task of this simulation is to set the target for next year.

Please do it for yourself. Where would you set the growth target if you were a board member of a construction company? Write down this number before you continue.?

We now provide more information. Below, we have the history of a construction company. This is again an actual case from the peer group of one of Obermatt ’s clients.

In the graph, we can see again that growth fluctuates. Again, it is sometimes higher, sometimes lower, and sometimes even in negative territory, albeit it fluctuates less than the industry if compared year by year. Let’s assume again that we are in the last year and already know that last year’s growth will be negative -3.3%. We know the median of 1.0% and the average of 0.2% over all the years. Based on this historical information about the company’s growth rate, the task of this second simulation is again to set the target for next year.

Please answer this question for yourself and write it down.

You should now have one or two growth targets for the subsequent year, and you may have made the experience that it was hard to settle on a reasonable value. When I ask my students this question in the classes I teach (for instance, at the St.Gallen Board Programmes | University of St.Gallen for the ESG seminar), they sometimes can’t even come up with an answer. It is simply too difficult. We don’t know the future; and from the past, we know it is very uncertain. Setting a future growth target seems less reliable than asking the Oracle of Delphi.

Most students come up with an answer as you might have. The answer is typically in the 3% to 6% target range. If you have arrived at a target, you may also have a similar figure.

The problem with such a target is that it is above the median, making it difficult to achieve. If we only know the history of the industry and the company, you can see in the following table how difficult a certain target level actually is:

The table shows the percentiles for different target levels. A target of 3.7% is in the 70th percentile of the history. That means 70% of the available historical years were below that target. It is quite similar for both the entire industry and the company itself.?

If the target is 3.7%, the company is unlikely to surpass it. In 70% of cases, it will be below the target. This means that executives will often be disappointed in themselves and their variable pay.

If you want to provide a 50% chance of winning this “bonus game,” you have to set the target at around 1.0% to 1.9%, a figure you see in the second lowest row in the table. In this design, executives have a 50% chance of winning and a 50% chance of underperforming. Not even this is good enough because we know from the phenomenon of loss aversion (in prospect theory) that humans are far more emotionally affected by losses than by wins. Even a 50:50 chance will eventually be felt as unsatisfactory pay.

Fair Targets Unfair

Will you set the target at the median level? Hardly, even if all your executives, board members, and shareholders are statisticians. For all other humans, it will be pretty strange.?

A one to two percent growth target will look ridiculous to most professionals in commercial enterprises. The target is far too low to be taken seriously. Executives will shrug their shoulders and expect a big bonus, boards will feel unease, and shareholders will object or even get furious at what they think are “soft targets”. If you are responsible for target setting, you cannot set a growth target as low as you should for a payout probability at par.

Even more problematic is that most executives are ambitious, sometimes even suffering from overconfidence, the fact that humans tend to be overly optimistic. Combine ambition and overconfidence, and most growth targets are north of the top quartile in the real world.

The reasons stated above may explain why only a minority of growth targets are included in actual compensation plans, as shown here, even though growing is the core goal of every profit-oriented enterprise, and the other metrics provide too much incentive to downsize.

There is no solution for fair targets, but there is a solution for fair pay?

There is no right growth target. However, there is a right compensation plan. In the graph below, we have combined the company's historical performance with the industry's history. You can now see when company growth outperformed industry growth and when it underperformed. The graph also shows the borders of the lower and upper quartiles. The blue shaded areas show where the companies in the middle, from the 25th percentile rank to the 75th percentile rank are. The blue-shaded areas represent 50% of all companies in the industry, and the outliers are above and below.?

We call this graph the Operating Index, and we have introduced it in the LinkedIn Stakeholder Risk blog already a couple of months ago: Indexing Operating Performance for the Helicopter View

With the Operating Index above, we can pay for growth without having to set - typically - overly ambitious growth targets. We simply pay more if performance is above the industry median and less if it is below. The basic idea: Median pay for median performance.

The Pay Index Motivates

While we know how difficult it is to define targets, caps, and floors in target-based pay designs, index-based designs are easy and intuitive. The graph below shows how indexed pay works:

The horizontal axis shows the performance as percentile ranks. The vertical axis shows pay. The worst performance is the zero percentile rank, for which there is zero pay. At the first quartile, the 25th percentile rank, pay is 0.5 times target, and the median pays target pay, which is typically based on industry median pay benchmarking. The maximum pay of two times the target pay is reached at the best performance, the 100th percentile rank.?

The benefit of moving from target-base pay to indexed pay is threefold:?

1. No underpay bias from typically ambitious targets
2. No confidential and overly risky targets, caps, and floors
3. Motivation in every situation - effective both in downturns and booms

What boards should know

Today's exercise of setting meaningful targets, the exercise from the last blog for setting sensible caps and floors, and the issues of profit-based incentives have shown that absolute targets are problematic in executive pay, typically leading to unintended underpay and a bias of downsizing. To avoid these pay-for-performance traps, the following guidelines help boards:

1. Measure performance for executive pay systems always relative to peers by indexing performance.
2. Distrust absolute targets in pay systems. While they are important for aligning executives on strategic goals, they have many undesired side effects on executive pay.
3. Measure top-down by selecting peers for the group and then for divisions and business units, aligning these organizational entities with their markets.

As a board member, you may encounter some hostility from executive management. Indexed pay is transparent, and transparency is not always in the interests of executives. However, indexed pay means that executives get more pay because their overconfidence misleads them to accept targets that are too ambitious, in other words, above average, and thus leads to below-average payouts.

Last but not least, always use percentile ranks to calculate payouts, discarding the method of calculating the difference between public indexes and industry benchmarks. We will show you why in two weeks.

In the meantime, subscribe to this stakeholder risk newsletter so you don’t miss it.