Politicians happy to blame public servants — but not so keen on praise

By David Donaldson

May 17, 2018

Green dart arrow hitting in the target center of dartboard with modern city and sunset background. Target business, achieve and victory concept .

Cognitive biases — such as focusing on negative performance data — can undermine the potential benefits of increased transparency. They might also help explain bureaucrats’ notorious risk aversion.

Politicians are more likely to attribute responsibility for performance to public managers when presented with poor results, according to a prize-winning academic paper.

“Overall, we find that elected officials become more willing to assign responsibility to bureaucrats if given performance data, but only in cases where data suggests low performance,” write Professor Don Moynihan and Associate Professor Poul Nielsen in the Journal of public administration research and theory.

This bias towards negativity probably helps explain why bureaucracies are typically more concerned with blame avoidance than innovation or high performance — but worryingly it also suggests that by making more data available, policymakers may not be encouraging good results but risk aversion.

“As bureaucrats come to understand these dynamics, their incentives are not to foster high performance, but to avoid negative low performance scores,” they write.

“It is hard not to conclude that performance management often becomes a negativity game that public managers are forced to play.”

The authors also find that an increase in information does not necessarily lead to more nuanced judgements, but can reinforce existing beliefs. Data is often complex and shows multiple trends, or no trend. The same underlying facts can be interpreted differently depending on the ideological frame. Thus, individuals can arrive at very different conclusions from the same results.

But having that data still increases individuals’ sense of certainty in their own position.

“Performance data may provide more information, and consequently more confidence in judgments made, without necessarily providing the causal clarity to justify that confidence,” say Moynihan and Nielsen.

“Part of the appeal of performance data for government is the promise to offer an objective and neutral account of public sector outcomes. But our findings suggest that the cumulative effect of performance data is to encourage polarization among elected officials.”

In addition, they found that politicians are more receptive to narratives about data presented by an advocacy group from their own side of politics.

In the experiments conducted by the academics, Danish left-wing politicians were less likely to attribute responsibility for poor performance to education bureaucrats after being told teachers’ unions believed the data was untrustworthy. Conservatives were unmoved by the teachers’ union’s message, however.

“Prior evidence shows that conservatives are generally more willing than liberals to attribute responsibility to bureaucratic leaders, and we show that data advocacy increases the distance between partisans. This finding arises even under relatively simple conditions of politicians examining a single performance metric.”

While such behavioural biases cannot be eliminated — and indeed in part reflect legitimate political differences — being aware of their existence may help both politicians and bureaucrats improve how they work.

“Practical guidelines and training on performance management emphasize technical skills, but neglect the ways by which politics infuses choices about both the selection of performance measures and their interpretation,” the authors say.

“Practical guidelines are necessary, but just as human resource professionals are sometimes trained to be aware of their implicit biases, policymakers and bureaucrats should be made aware of how negativity biases and ideological preferences affect how they and others use performance data.”

These findings will probably not surprise many public servants, but make a contribution to a field previously little studied in the academy, known as behavioural public administration. Moynihan and Nielsen’s paper recently won best article on public management published in 2017, awarded by the US-based Academy of Management’s Public and Nonprofit Division.


About the author
Inline Feedbacks
View all comments

The essential resource for effective
public sector professionals