A common approach for supervisors to set ambitious but achievable performance targets is to start with last year’s performance and adjust it for expected future growth. A downside of this approach is that it discourages managers from going above and beyond because good performance today raises the bar tomorrow, making future targets more difficult to achieve. We look at an alternative target-setting approach based on peer performance that retains incentives to perform well.


This alternative approach begins with identifying a peer group of managers who face a similar business environment. The average performance of this peer group then becomes a long-term benchmark for target setting. Managers with performance below the benchmark would see substantial target increases until they catch up with their peers. In contrast, managers with performance above the benchmark would only see modest increases or possibly even decreases in their targets.

Peer-based targets have two key advantages. First, to the extent that the peer group is exposed to similar economic fluctuations, peer-based targets are flexible and adjust upward as well as downward in response to changes in the business environment. This provides some reassurance that the best in a peer group will be rewarded even if adverse economic shocks reduce their performance. Second, best performers are less concerned about greatly outperforming their current target if targets depend more on peer performance than on their own past performance. Conversely, managers lagging behind their peers are under pressure to improve.

The obvious downside to peer-based targets is that there is no such thing as a perfect peer group. Even in homogeneous peer groups, each manager faces a slightly different business environment. Another, bigger downside of peer-based targets is that they could invite competition rather than cooperation among peers. If higher peer performance raises targets for everyone, it can hurt managers who make an effort to help others rather than trying to maximize their own performance.


Does the importance of cooperation affect how supervisors set targets? We conducted a study using data from a company where cooperation is important (for the full study, see Martin Holzhacker, Stephan Kramer, Michal Matejka, and Nick Hoffmeister, “Relative Target Setting and Cooperation,” Journal of Accounting Research, 2019, pp. 211-239). In what follows, we refer to the company as Gamma.

With annual revenues of around $1 billion, Gamma provides testing and certification services for oil wells, pipelines, refineries, and other energy infrastruc­ture. The worldwide operations are decentralized into seven business geographical regions containing 15 business groups, which in turn are comprised of many business units that operate as autonomous profit centers responsible for serving customers in a particular geographical area.

The primary resources used in Gamma’s testing and certification services are qualified technicians and specialized equipment, used in testing for defects such as cracks or corrosion, and certifications of safety and reliability. Gamma faces highly volatile demand, and business unit managers often deal with capacity bottlenecks that make managing capacity utilization critical for profitability.

Business unit managers have some flexibility to adjust capacity through overtime or temporary employment contracts, but these alternatives are costly and only possible if additional demand can be predicted. Given these capacity bottlenecks, managers are very concerned about having to decline a surprise order, which could mean losing a longtime customer. To reduce this risk, business units invest in slack capacity both in terms of labor and equipment, even though it involves significant long-term cost commitments.

Slack capacity can also be shared among business units, and the benefits of such capacity pooling are widely recognized within Gamma. Cooperation via internal transfers is the least costly way to address bottlenecks because transfers are much cheaper than temporarily hiring qualified external personnel.

Cooperation is particularly desirable when, for example, travel distances between business units are small and capacity sharing is less costly, when offered services are similar so workers and equipment can be exchanged more easily between business units, or in situations where demand is predictably asynchronous—that is, when workers and equipment in one business unit are sitting idle while another business unit is scrambling for resources due to a high workload. Maintaining a norm of voluntary cooperation among its business unit managers is critical for Gamma.


Gamma follows a multistage process to set budgetary targets whereby business unit managers provide an initial estimate of next-year earnings and revenues and negotiate targets with business group managers. During these negotiations, business group managers extensively use a business intelligence tool that tracks sales growth and profit margins for each of their peer business units in their business group. The resulting targets are therefore generally based on past business unit performance as well as past performance of peer business units, although the weight placed on both varies from unit to unit.

Collecting multiple years of data on budget target revisions for business unit managers, we examined whether business unit targets are to a lesser extent based on past peer performance benchmarks when cooperation among managers is critical. Our analysis confirms that peer-based targets are used less when cooperation is most beneficial (i.e., travel distances are smaller, services are similar, and demand is asynchronous), resorting to setting targets based more on last year’s performance plus growth.

Yet we also found that cooperation wasn’t the only consideration when budgetary targets were set—for example, supervisors also refrained from setting peer-based targets if peers were relatively dissimilar to begin with.


Calibrating targets based on peer benchmarks can be an effective method to improve managers’ motivation and avoid strategic gaming of the target-setting process. It also encourages a culture that may be appealing to managers who enjoy competition.

On the other hand, using peer benchmarks in target setting may discourage cooperation because it can penalize managers who help others by granting them more difficult future performance targets. In practice, avoiding cooperation can come in various forms: silo thinking, lack of teamwork and knowledge sharing that impairs innovation, or even outright sabotage.

The way supervisors set targets can fundamentally affect workplace climate, so it’s important to be mindful of this impact the next time the annual planning cycle comes around.

About the Authors