A few years ago, It was difficult to come across any local government employee who was well versed in the concept of behavioral economics. The idea of municipalities testing ways to "induce" residents to pay delayed services or register for preventive medical screening campaigns is considered It's still young.
But the application of this concept on the ground has become widespread, and achieve tangible results, These findings were reinforced by the What Works Cities initiative, launched by the Bloomberg Foundation, which aims to help 100 cities in the United States enhance and improve the use of data and evidence to improve services to residents. These cities used behavioral insights to increase arrears in tax payments. Promote diversity in new appointments in the police force. and improve waste collection processes and others, Behavioral insights are one of the most important tools cities can use in their pursuit of innovation.
The Behavioral Insights Team (BIT), It is a British company with social goals that seeks to help cities develop and evaluate ideas to improve government services, One of the most important and prominent partners in the efforts to exploit behavioral insights. Over the past three years, The team carried out 99 experiments in 37 U.S. cities. Within the "What Works Cities" program, These experiments are characterized by their low cost and high accuracy, It is carried out according to the model of randomized controlled trials, It seeks to find out whether amendments to mail, email, text messages and other government means of communication with citizens and residents will make a tangible change in how recipients respond to these messages. While these experiments don't always work, However, their success helps cities adopt these strategies and launch them on a large scale. It is confident that it will achieve results.
To learn the most important lessons learned by cities from these efforts, Bloomberg interviewed Michael Holzworth, Director of the Behavioral Insights Team for North America, and Sasha Tingebov, who directs the team's operations within the What Works Cities initiative in the United States. We also spoke to a group of U.S. city leaders who conducted behavioral experiments, Whether in collaboration with the Behavioral Insights team, in person, or in partnership with others. Here are 7 key insights drawn from these efforts.
1. Just testing and learning is a success in itself
There are many success stories achieved by municipalities in this field, Syracuse, New York, saw a $1.5 million rise in property tax payments due after the municipality sent notification letters to residents in arrears. The municipality placed the letters inside an envelope and wrote on it short, striking handwritten phrases . The number of people of color from Chattanooga , Tennessee, who apply to the police has also increased. And that is after I tried writing new phrases on the recruitment mail that you send to candidates. In turn, The City of New Orleans found that the desire of residents eligible to receive free health care to see a doctor, It increases by 40 percent if they receive a text message saying that they have been "selected for a free doctor's check-up appointment." More case studies and use cases can be found in the Behavioral Insights Team report, Behavioral Insights for Cities .
For New Orleans, A different language has been tried to send text messages to encourage residents to take advantage of free medical check-ups, The message referred to in the middle image above was met, The strongest response from the population.
"We're trying to make simple, cost-effective behavioral shifts," Holsworth says. He noted that proven effective methods could be applied more widely. "Even minor improvements, It can be a great success."
The Director of Performance Strategy and Innovation at Tusla City, Oklahoma, agrees, James Wagner, who previously worked with the Behavioral Insights Team in two experiments, one of which resulted in a 15 percent increase in the number of people paying traffic fines, This is after receiving a text message from the municipality reminding them of the deadline for paying fines.
"The most important lesson I've learned is to test anything before adopting it and implementing it more broadly," Wagner says. Noting that the implementation of the texting experiment cost only $ 100. "The experiment was likely to fail," he said. We could easily have assumed that the experiment would work. Within a short period of time, "We will have spent $15,000 and gone ahead with the project with a supplier without knowing if the project will succeed or not."
2. Failure is the best teacher
Not all experiments achieve the desired success, Many cities share the same reality; while they put a great intellectual effort into redesigning a traditional government mission, for example, It turns out that the new message made no difference, Or the old language of the message was much better in terms of result.
Katie Scheffley says, She is a senior financial analyst at the Portland , Oregon Budget Office. Such results can cause frustration. Portland has previously conducted many experiments on more than 8 departments, Some of which succeeded, While others failed. However, Thinking about evidence of interest and designing an experiment to find such evidence are worth the effort. Scheffley says, "Even when we don't get any results, We are trying to see the great value of collaborating with the program team to help them think about how we can improve the services they provide to people."
Get the latest innovation-related news from Bloomberg Cities! Subscribe to SPARK.
Adria Finch agrees with Chifley. Finch is the director of the Syracuse City Innovation Team , an advisory team based in the municipality of Syracuse with core funding from the Bloomberg Foundation . During her work, Finch has led the implementation of several behavioral experiments, with the last one targeting arrears taxes achieving amazing results, while other experiments have not achieved similar results.
But she thinks this is normal, she says, "We always learn something new from these experiences," she adds, "So, We will continue to conduct them, And we are not discouraged by the failure of one from time to time, Herein lies the joy of these experiences: They don't need much investment, and satisfy our desire to try, If we succeed, That's great! If we don't succeed, We try again."
3. Start with the easy that can be achieved
The first experiences of behavioral insights in cities focused on government messages and other means of communication with residents, Hollsworth says there are compelling and logical reasons for this; first, City municipalities send a large number of messages to residents, and secondly, Modifying or redesigning the language used in these messages doesn't cost a lot of money. In the third place, Measuring the impact of these adjustments is relatively easy. "We can achieve results quickly," says Hollsworth, whether we're experimenting with service charges, law enforcement or any other type of collection.
He adds, "The results may not be impressive, But don't forget that they are extremely important for the sustainability of cities, We believe it is a good basis to open the way for the expansion of the application."
When the city of Syracuse carried out an experiment involving handwritten graffiti on the envelopes of municipal letters to residents regarding arrears, The value of collections from these receivables jumped.
Achieving some successes and quick gains is important, It enables employees who are uncertain about the importance of applying behavioral concepts to note the value of these experiments and the importance of allocating the time and effort required to conduct them. Brent Stockwell says, Assistant Administrator of Scottsdale, Arizona, "We should always remember that the mere introduction of behavioral insights into local government is an initiative for change. Like any other change initiative, It will take time, so, We have to think about small achievements to pave the way towards big achievements."
4. Leadership support for these efforts is imperative
Due to the newness of this type of work in local governments, Task forces need the endorsement and support of the mayor, the city manager and other senior leaders. who are not required to know how to conduct binary hypothesis tests, It is enough for them to develop a clear vision about the importance of using data and evidence in the work. They should also support change agents who are leading these experiences and conceptualizing fruitful collaboration between different departments.
"These experiments are an advanced and evolving change. Not only has it accelerated efforts to provide services in cities, It created a completely different way of doing the job and providing services." "It's important that the team has the support of a mayor who can clearly state 'we are committed to learning what works and implementing what works.'" This will help solve many of the day-to-day challenges you can encounter during the implementation of these projects."
5. It is necessary to establish a regulatory framework and structure the use of behavioral insights
Cities must find specialized task forces in this field, To be able to extend the use of this approach, so that the members of these teams, In the typical case, Employees within the municipality in locations that enable them to work with various departments and departments, Such as the innovation team or the city manager's office.
In Scottsdale, Cindy Eberheart, which earned the title of "Fighting Ninjas in the Behavioral Insights Team", Position as the leader of the city's behavioral insights team. Having participated in several successful experiences in collaboration with the Behavioral Insights Team, In a two-day training of more than 20 employees, Eberhart has formed several teams within different departments to conduct these experiments more broadly within the organization, Representatives from each department meet monthly to discuss the lessons they have learned from these experiences.
Eberhart says, "The work should be organized within a clear structure and subject to regular periodicity, Even if the experiment had the support of the City Manager's Office, its implementation is not within the work plan of any of the employees who are very busy in their work, We are not yet used to this kind of change in business execution."
6. You do not need a PhD in statistics, You need experience.
The test model used by the Behavioral Insights Team to train teams in municipalities is simple and low-cost. However, Experiments require someone with data experience, This is difficult to find in many local administrations.
Lindsay Mazer says, She has been involved in several behavioral projects in Portland, "Things can get more complicated very quickly, Even if you can find someone within the municipality who has enough experience to help you design a randomized, disciplined experiment, It won't be enough, Because that doesn't fall within his purview." Mazer adds that what Portland needs is a competent employee who acts as a consultant within the municipality. This specialist will assist in clarifying the technical aspects of the design of experiments and in analyzing the results, He will also be keen to keep up to date with the latest research in the field of behavioral insights.
Revenue collection is one area where cities should look for partnerships outside the municipal framework to do. To implement the successful late tax project , the City of Syracuse has partnered with Maxwell X Lab . At Syracuse University, which specializes in helping the public and nonprofit sectors embed behavioral insights into their work environment. A local institution has provided funding for the project.
7. Be sure to involve IT and Legal in this work
The various cities that have taken this work have faced a common obstacle in the form of their data systems, where it hinders the existence of an out-of-date database, Know how successful the intervention was made.
Trigipov says, "One of the lessons we learned was to involve IT in our discussion about the project from the very beginning; at the beginning of our work on What Works Cities, We were saying, 'We'll fetch the data after we finish designing the experiment', in turn, The IT staff were content with brief answers: 'We don't have this data! The system doesn't work like that!"
The same applies to the Legal Section; Efforts to test the design and language users in some messages have encountered legal obstacles, The Legal Affairs Section expressed concern, After changing the design of some messages, A portion of the people will receive messages that are different from the letters that others receive from the municipality. Finch says, "It's important that we have open dialogues with the legal team and ask them what they feel is the best option and if they have any recommendations." She continued, "If you make sure to involve them from the beginning, You will avoid a lot of challenges in the long run."