Nudging Cities: Innovating with Behavioral Science

Lessons learned at the 15th Convening of the Project on Municipal Innovation

Monika Glowacki Grey

By Monika Glowacki • May 17, 2016

It is now common practice for private sector companies to use behavioral science to cue better decision-making among employees. Google, for example, used a behavioral economics intervention – also known as a “nudge” – to promote healthy-eating initiatives in its corporate offices. By reorganizing the fridge with bottled water at eye level and soda at the bottom, water intake increased by 47% and soft drink consumption decreased by 7%.

Cities across the country are beginning to apply the same principle to do low-cost government interventions in the civic realm. At the 15th Convening of the Project on Municipal Innovation on March 18th, mayoral chiefs of staff and leaders in the field discussed how behavioral science can be used as a tool to improve public policy.

The Promise

“Behavioral science promises large-scale outcomes at a low cost – the holy grail of innovation,” said Matthew Klein, Executive Director of the Center for Economic Opportunity in New York City. Behavioral economics drove the design of CEO’s $aveNYC initiative, which nudged low-income households to be more financially responsible. The pilot was designed to make choosing to save convenient, simple, and rewarding by opening savings accounts for New Yorkers at the same time they did their taxes and providing a financial incentive – a 50% savings match on the tax-time deposit if left in the account for one year. While initially half of the program’s participants reported no history of savings, 80% saved for at least one year to receive the match and 75% continued to save thereafter.

Cities are increasingly embracing nudging and applying it to a broad range of government areas, including revenue collection, transparency, public health, civic engagement, police recruitment, and water conservation. If there is a behavior involved, you can nudge. “There is no end to levers or domains because behavioral science can be applied anywhere,” said Ted Robertson, Managing Director of ideas42. “One can apply it at the policy level or the micro-behavioral level or anywhere in between.”

The Takeaways

Over the course of the panel, key considerations emerged regarding how cities can apply behavioral science to better design standard municipal government practices. They include:

  • Start simple. Cities’ first experiment with nudging should focus on a simple behavior to make a small change to something that is already happening. Starting with communications is easy and cheap, and the savings stemming from it can be measured immediately. Elizabeth Linos, Head of Research and Evaluation for the Behavioral Insights Team North America shared a case in which changing the language on letters brought forward $300 million in additional revenue by increasing the number of people who paid their taxes. The intervention came at no extra cost, as these letters were already being sent out.
  • Break down larger social challenges into specific behaviors. Linos said that when government clients come asking for a solution to issues like a high unemployment or murder rate, her team takes the big social program, tries to understand the systemic and institutional barriers that exist, and then maps out specific micro-behaviors to which rigorously evaluated, iterative tweaks can be made to promote better decision-making. “Tweaking at these pinpoints can have disproportionately large effects,” she said.
  • Focus on understanding the target audience of the nudge. It is important to consider the day-to-day lives of a particular group of people in order to design an intervention that resonates with them. One must determine the most effective behavior to anchor to, the right motivational lever to pull, and what language or visual cues to use. For example, people enjoy operational transparency and seeing the behind-the-scenes work on what is being done on their behalf. This insight can be applied to change how people think about government. While initially Boston’s 311 app only showed when open complaints were logged and resolved, it was not until the city implemented a map allowing citizens to track issues submitted by other users that it generated a more widespread positive reaction.

The Open Questions

Though city representatives were excited about putting behavioral design into practice, panelists and chiefs of staff raised some questions about using these insights in government in the long-term, including such considerations as:

  • Sustainability. Even cities that have the expertise, training, and viable partnerships for nudging are apprehensive on whether a behavioral science project can be scaled or institutionalized. The case of Denver’s use of behavioral science suggests some lessons. The city wanted to increase usage of pocketgov, a web app that provides information on city service delivery and allows citizens to report issues like potholes and graffiti to city hall. With funding from What Works Cities, the Denver Marketing Team worked with the Behavioral Insights Team to give its staff behavioral science-based marketing training. They implemented multiple email campaigns with varied language to test which would most effectively drive usage. With this experience and training, Denver’s Marketing Team can now apply the lessons it learned to future initiatives.
  • Changing the norm. Nudges are more effective in bringing people closer to socially desirable behavior than in changing the norm itself, and bad civic habits are often the standard. People are late with tax payments, reluctant to go to the doctor even if they qualify for a free visit, and often unlikely to save. Panelists recognized that it is very difficult to push people away from bad habits to reset the norm itself. It would require moving beyond financial incentives or changes in wording on a letter to a more complex mix of messaging, behavioral design, and policy change to elevate a set of behavioral responses into a new norm.
  • Ethics. Michael Norton, a professor at the Harvard Business School, emphasized that behavioral science is not about coercing people into behaving a certain way, but rather informing them to improve their defaults. It is about putting something in the environment that changes behavior for the better. The panelists agreed that it is important to constantly be mindful of the ethical considerations of behavior design, and to always keep the focus on preserving decision-making power for those being nudged.

Some critics are suspicious of governments acting as choice architects and determining which ways the public should be nudged. However, when thoughtfully and transparently selected and designed, nudges promise to improve lives while maintaining freedom of choice. The conversation at the PMI convening suggests that although behavioral science offers government significant utility at a low cost, there is still work to be done, and plenty room for collaboration and replication across cities. “Cities are asking the same questions and worrying about the same issues,” Linos said. “Everyone is dealing with these challenges at the same time.” The panelists hope that, with time and practice, behavioral science can be elevated to another established administrative tool in the forward-thinking government toolkit.