Researchers conducted a sprawling field test involving 17,303 wallets distributed across 355 cities in 40 countries to examine how people respond to lost property. The central finding: when a wallet contained cash equivalent to US$13.45, the rate at which recipients reached out to the owner jumped from 40% to 51%. This counterintuitive result, published in Science in 2019 under the title Civic Honesty Around the Globe, upends the conventional wisdom that financial incentive necessarily breeds dishonest behaviour. However, the study measured a specific action—sending an email to the supposed owner—rather than tracking physical wallet returns.

How the experiment was structured

Alain Cohn, Michel André Maréchal, David Tannenbaum and Christian Lukas Zünd created a controlled scenario mimicking a realistic situation. Research staff approached workers at institutions including banks, hotels, cultural organisations, post offices and government offices, presented them with a transparent wallet they claimed to have found, and then departed. The experimental design hinged on a single variable: some wallets contained no money, while others held the local equivalent of US$13.45. Over a 100-day window, the team tracked whether recipients sent an email attempting to return the wallet to its owner.

This approach proved more rigorous than simply asking people whether they consider themselves honest individuals. By altering one element of the scenario and observing the consequences, the researchers could isolate the effect of cash. Yet the setting itself—employees encountering lost property at their workplace—means the findings apply to a specific population rather than representing all people who might discover a wallet.

Understanding the 11 percentage point shift

The University of Michigan's summary of the results describes the overall movement from 40% to 51% contact rates. This represents an 11 percentage point increase in absolute terms, or roughly a 27.5% rise when calculated relative to the original 40% baseline. These two ways of describing the same change convey different impressions and should not be conflated. If 100 wallets were handed over under each condition, the averages would translate to approximately 40 emails sent when no cash was present and 51 when money was included.

Importantly, the experiment does not suggest that most people will automatically contact an owner even with cash present. Substantial numbers of recipients failed to reach out in both scenarios. The study's contribution lies in demonstrating a comparative effect: the presence of money shifted behaviour in a measurable direction.

The psychology behind the shift

Researchers from the University of Zurich proposed that self-image plays a central role. Keeping a wallet may trigger psychological discomfort because the person might view the action as theft. More cash could amplify both the financial temptation and the mental burden of seeing oneself as a thief. Survey participants rated keeping a wallet with more money as feeling more like stealing. The team also tested wallets containing a key—an object valuable to the owner but not typically useful to the finder. These wallets were returned at higher rates, suggesting concern for the owner's welfare influenced decisions.

These explanations illuminate the observed pattern, yet they remain inferences rather than direct evidence of individual thought processes. Behaviour stems from multiple overlapping motivations: concern for others, workplace norms, personal responsibility and habit. The cash was not necessarily the only factor being weighed.

Testing with larger amounts

A secondary experiment increased the cash to the local equivalent of US$94.15 in the United States, the United Kingdom and Poland. The Michigan account notes even higher reporting rates with this larger sum. This extension tests whether the main finding depended on the temptation being minimal. However, it does not establish what occurs at every conceivable amount. The behaviour associated with a wallet holding roughly one hundred dollars cannot be extrapolated to scenarios involving suitcases of cash, accounting fraud or commercial deals, where the form of the opportunity, the context and the stakes all shift fundamentally.

The same limitation applies to geography. While a broad international sample holds value, extending the high-amount test to only three countries remains a three-country extension. Caution is warranted against treating this as evidence that every amount was tested across all 40 countries.

What email contact reveals and conceals

The most accurate interpretation stays grounded in what the evidence actually shows. An email provides a pathway for returning property and represents an observable action that can be compared between experimental conditions. It does not, however, describe every interaction that follows or reveal permanent character traits. Similarly, the absence of an email should not be automatically read as proof that a particular worker kept the money. Institutions may handle lost property through other channels, or messages might simply go unsent. These possibilities constrain how individual cases can be characterised, even though the group-level comparison remains instructive.

For those studying workplace behaviour, the study offers grounds to reconsider oversimplified theories about incentives. Financial gain can reshape how people interpret an action, not just what it pays. The recipient was evaluating more than cash alone. A name and contact information for the owner were also present.

Source: Silicon Canals