Does the culture of the public sector actually cause people to behave more honestly? In this blog post, Raanan Sulitzeanu-Kenan draws on a series of randomized controlled trials across five countries, to challenge long-held assumptions.
For centuries, human societies have held onto a normative ideal: that those who serve in government possess a higher moral compass than those in the private sector. From the 6th-century BCE Chinese administrative ethics of Confucius to Plato’s prescriptions for Athenian officials, honesty has been championed across cultures as a defining virtue of public servants. To the extent that this characterisation is true, public organisations are expected to place a premium on honesty, encouraging workers to align their behaviour with that ideal, holding other factors – both corrupt and benevolent – constant.
But does the culture of the public sector actually cause people to behave more honestly? In a recent paper co-authored with Markus Tepe, Saar Alon-Barkat, Florian Erlbruch, Omer Yair, Michael Jankowski and Christine Prokop-Scheer,we put the fabled “public sector ethos” to the test. Our findings challenge the received wisdom, and carry profound implications for how we govern and regulate public institutions.
The Challenge of Identifying the Effect of Culture on Honest Behavior
To understand the novelty of this study, we first have to appreciate the difficulty of identifying the roots of honesty. Traditionally, scholars assumed that public sector culture – the unique set of social norms and values internalized by government workers – placed a high premium on fairness and honesty, contrasting with the private sector’s typical focus on efficiency and profit.
Identifying the true causal effect of this culture is a methodological challenge. If we simply observe that public sector workers act more honestly than corporate employees, we run into a “chicken-and-egg” problem: Does the public sector make people honest, or do naturally honest people simply self-select into public service jobs? Further, public and private sector employees differ in countless unobservable ways, meaning any simple comparison could result in a spurious and misleading correlation.
To bypass these hurdles, we utilized a psychological tool: identity priming. Self-Categorization theory suggests that we all hold multiple identities based on our religion, nationality, profession, etc., each coming with its own set of behavioral norms. By having a random treatment group answer specific questions about their public sector work experience, we temporarily activated their “public sector identity.” This allowed us to observe how heightened cultural salience directly influenced their immediate behaviors compared to a control group.
Measuring Honesty
With the participants’ public sector identities successfully primed, we introduced monetary incentives to see if the civil servants would cheat. In four countries – Germany, Israel, Sweden, and Italy – participants played the “dice in mind” game to measure individual honesty.
The rules were simple but tempting: participants privately predicted the outcome of a die roll, rolled a virtual die, and reported whether their prediction was correct. Every “correct” guess earned them real money. Because true predictions are completely unobservable by the researchers, participants could safely lie to increase their payout without fear of getting caught (the game was played for 12 rounds). Under perfect honesty, participants should only guess correctly about one in six times (roughly 17% of the time). This allowed us to measure aggregate cheating based on the group’s deviation from this statistical expectation.
The study didn’t stop at individual cheating. Because real-world corruption is rarely a solitary act – often requiring collusion between bureaucrats – we also tested collaborative cheating. In a fifth study conducted in the UK, public employees were paired up in a sequential die-rolling game. If both players reported the same number (a “double”), they both received a monetary reward. This tested whether activating a public sector identity would suppress the human tendency to collaborate on a profitable lie.
A Surprising Disconnect
The results of these trials are highly revealing. If public sector culture actively promotes honesty, activating that identity should have caused cheating rates to plummet. Instead, we found absolutely zero evidence that public sector culture increases individual honesty. Across Germany, Israel, Sweden, and Italy, the effect of priming the public sector identity was negligible and statistically insignificant. It did not matter whether the country historically demonstrated high or low levels of public integrity – the cultural prime did nothing to restrain individuals from lying for extra cash.
Figure 1: Effects of priming identity priming on honesty and values in Italy, Sweden, Germany and Israel
Source: Sulitzeanu-Kenan, R. et al. (2026)
The collaborative cheating results from the UK were even more surprising. Far from reducing dishonesty, priming the public sector identity had no significant effect on the number of reported “doubles” and actually produced a slight increase in cheating regarding the reported value of the dice rolls.
Figure 2: Effects of priming identity priming on collaborative cheating and values in the UK
Source: Sulitzeanu-Kenan, R. et al. (2026)
Perhaps the most fascinating twist is that while the prime failed to make the civil servants act honestly, it did successfully increase their self-reported Public Service Motivation (PSM). After the tasks, primed participants reported a higher dedication to public service values. This indicates an ‘ethical blindspot‘: individuals can consciously maintain adherence to ethical values while simultaneously rationalising dishonest behaviour behind closed doors.
What This Means for the Future of Governance
These findings deliver a sobering reality check for public administration. The notion that the public sector possesses a unified, shared culture that inherently enforces honesty finds no empirical support. While it remains possible that specific organizational cultures (of a particular department or agency) affect honest behavior, we find no such aggregate effect across the public sectors of these five countries.
For policymakers, the message is clear: relying on an abstract “culture of integrity” is an ill-advised strategy. If we want a clean, honest, and efficient government, we cannot rely on assumed virtues. Instead, we must rely on verifiable guardrails. Governments should prioritize robust institutional monitoring, rigorous financial audits, and strict meritocratic recruitment processes rather than assuming the “public sector ethos” will do the work for them.
You can read the full paper here.
Raanan Sulitzeanu-Kenan is Professor of Public Policy at the Hebrew University of Jerusalem. His research focuses on behavioural public administration, political behaviour, and empirical legal studies. He is the former head of the Federmann School of Public Policy (2017-21).