Internet Explorer 11 is not supported

For optimal browsing, we recommend Chrome, Firefox or Safari browsers.

How Government Can Nurture the Nudge

Behaviorally informed interventions can improve outcomes. Louisville's approach shows the wisdom of starting small.

In implementing behaviorally informed interventions known as "nudges" to improve program outcomes, the city of Louisville has followed an important precept: You have to learn to walk before you can run.

Kentucky's largest city is one of many jurisdictions and agencies that have been experimenting with nudges, approaches derived from behavioral science that encourage preferable behaviors by manipulating individuals' "choice architectures" -- the social, psychological and physical contexts in which they make decisions. Louisville has taken incremental steps to integrate nudging into its larger data-driven portfolio, building upon smaller, common sense initiatives.

Mayor Greg Fischer has demonstrated a consistent commitment to implementing data-driven tools that improve Louisville's responsiveness and efficiency. The city recently partnered with Bloomberg Philanthropies' What Works Cities initiative, which helps cities across the U.S. use data and evidence to improve policy and program outcomes through technical assistance from expert organizations.



What Works Cities paired Louisville with the Behavioural Insights Team (BIT), which was started by the British government and now helps other public-sector partners develop behaviorally informed policies. When partnering with cities via What Works Cities, BIT conducts randomized control trials in which a subset of service recipients receives an intervention and their behavior is compared to that of a control group that doesn't receive the nudge.

In an early randomized control trial, for example, Louisville and BIT redesigned the city's communications to residents with unpaid parking tickets. The new communications use social norms to encourage residents to pay up, notifying those with unpaid tickets, for instance, that three-quarters of the people in their neighborhood had paid their fines on time in the past year. The new letters more than doubled payments, generating a net return of $4.53 per letter.

 
Louisville was able to promote citywide buy-in for nudging in part by picking an area where there was a demonstrable, easily quantifiable problem: $1.1 million in parking fees and fines had gone unpaid in the second half of 2015, and it was clear that the old way of prompting payments -- letters 7 and 14 days after the ticket simply notifying residents that they needed to pay -- was not working. Louisville wisely "picked the low-hanging fruit," says Elspeth Kirkman, a BIT senior vice president based in the organization's Brooklyn office. Because the mayor and the What Works team were able to point to the results of the randomized trial, it was easier to get the city's Parking Authority to change its practices.

"There will always be barriers, and it helps to be able to show success," says Kirkman. As every mayor knows, it can be difficult to convince departments that have followed the same procedures for years or decades to change their ways. But by providing a tested solution to a demonstrable problem, Louisville's parking nudge encouraged other departments to pursue behavioral interventions. "It forces other folks to start looking at their forms," Mayor Fischer explains. "These government forms have been the same language for 20 years."

The success of the parking-ticket trial encouraged Louisville to pursue nudging on its own. Soon after working with BIT on the parking-ticket nudge, Kirkman says, Louisville conducted a randomized trial to encourage city employees to donate to charity. That nudge turned out not to be effective, but the city proved to itself that it had the tools to implement and test nudges independently, spurring further behavioral interventions down the road.

Louisville has since greatly expanded its nudging agenda, using them to reduce false alarms and improve code compliance. Moreover, the city recently repurposed a position to create a BIT-trained in-house nudging expert in its Office of Performance Improvement and Innovation. Shireen Deobhakta will oversee and design Louisville's nudging campaigns and ensure that city employees are properly trained to conduct randomized control trials. She will also create and sustain university partnerships to expand Louisville's nudging capacity.

Louisville's experience provides a useful roadmap for cities interested in pursuing behaviorally informed policy: Start where you are likely to succeed and build around that success. While nudging cannot solve all of a city's problems, the practice can produce tremendous value when integrated over time into a city's wider data-driven reforms.

Chris Bousquet, a research assistant/writer for Data-Smart City Solutions at the Harvard Kennedy School's Ash Center for Democratic Governance and Innovation, contributed to this column.

Stephen Goldsmith is the Derek Bok Professor of the Practice of Urban Policy at Harvard Kennedy School and director of Data-Smart City Solutions at the Bloomberg Center for Cities at Harvard University. He can be reached at stephen_goldsmith@harvard.edu.
Special Projects