Transparency Can Backfire
Transparency is usually treated as an unquestionable virtue. In markets and public policy alike, the assumption is simple: the more people know, the better their decisions will be. But in places where trust in institutions is fragile and information scarce, transparency can have unintended consequences. However, a new study published in The Review of Economics and Statistics (“When Transparency Fails: Financial Incentives for Local Banking Agents in Indonesia”) by Erika Deserranno (Bocconi Department of Economics), Gianmarco León-Ciliotta (Universitat Pompeu Fabra, Barcelona) and Firman Witoelar (Australian National University) shows that revealing how much local banking agents are paid can actually undermine trust and stall the adoption of beneficial financial products.
Across much of the developing world, access to banking depends on local agents: shopkeepers and small business owners who act as intermediaries for banks. Indonesia embraced this “branchless banking” model in 2014 to bring savings accounts and digital wallets to rural areas. Agents earn commissions for each customer they sign up, creating a classic policy challenge: how to motivate effort without eroding trust.
As the study notes, “technology adoption depends crucially on the local delivery agents being motivated enough to promote the product […] but also on the community having enough trust in the local agent.”
A simple experiment, a surprising result
To understand how incentives and trust interact, researchers ran a large-scale field experiment in 401 villages in East Java. They varied two things: the level of commissions paid to agents (low or high), and whether those commissions were kept private or disclosed to potential customers. When commissions were high and kept private, the results were striking. Adoption of new banking products more than tripled. Agents worked harder, approached more people, and customers who opened accounts actively used them, increasing savings and transactions. But when the same high commissions were made public, the effect disappeared. Despite higher effort by agents, take-up and usage did not increase at all.
When incentives become signals
The reason lies in how people interpret incentives. In communities with little prior exposure to formal finance, compensation acts as a signal of intent: learning that an agent was paid a generous commission often triggered suspicion rather than reassurance. In the authors’ words, “financial incentives send a negative signal to potential clients about the reliability and trustworthiness of the product.” High disclosed incentives were interpreted as a sign that agents were motivated by money rather than by customers’ interests. Trust declined not only in the agent, but also in the bank and the product itself. Once trust eroded, extra persuasion no longer translated into adoption.
The opposite pattern emerged for low incentives. When modest commissions were disclosed, trust increased. Customers perceived agents as more credible and less opportunistic, and demand rose relative to cases where low incentives were kept private. As the paper explains, high incentives can signal that “the agent is primarily motivated by earning money (…), and hence more likely to take advantage of an uninformed consumer”, while low incentives may signal honesty or pro-social motivation.
Why effort isn’t enough
A key insight is that incentives operate on both sides of the market. They affect supply, by motivating agents, but also demand, by shaping beliefs. In this setting, transparency didn’t discipline behavior but perceptions instead. When incentives were private, higher pay translated cleanly into higher effort and higher adoption. When incentives were public, higher effort collided with lower trust. Transparency neutralized the policy. The effects were strongest among people with less information (those unfamiliar with banking or the agent), that is precisely the sections financial inclusion policies aim to reach.
Lessons beyond banking
Although the study focuses on Indonesia, the lesson travels well. Many sectors rely on commissioned intermediaries, from insurance and healthcare to education and development programs. Wherever people ask, “Why is this person recommending this to me?”, incentives become signals. The uncomfortable conclusion is that transparency is not always welfare-enhancing. In low-trust environments, revealing incentives can backfire. High incentives may work best when kept private; low incentives may build trust when disclosed. Combining high pay with full transparency can be the worst of both worlds because sometimes, what people infer matters more than what they are told.
