Human behavior – the next frontier in agency innovation

Understanding human behavior could save agencies money and make their programs more effective, according to a recent White House memo. That theory is currently ...

Innovation is not just about ideas. It’s about how you implement those ideas, identify the ones that work and spread those successful ideas throughout your organization.

Jitinder Kohli, director of Federal Government Performance at Deloitte, says that’s the key takeaway from the July 26 White House memo, Next Steps in the Evidence and Innovation Agenda.

“A lot of people think that innovation is just about generating ideas,” he said. “And actually, if you look at some of the things that the administration has said over the last few years, you could be forgiven for thinking it’s all about crowdsourcing, big data and those sorts of things.”

Instead, the memo is about agencies trying new methods, establishing whether they work, gathering data about their effectiveness and promoting the methods that are the most effective.

Jitinder Kohli, director of Federal Government Performance at Deloitte
The memo came from Director Sylvia Burwell of the Office of Management and Budget, Director Cecilia Munoz of the Domestic Policy Council, Director John Holdren of the Office of Science and Technology Policy and Chairman Alan Krueger of the Council of Economic Advisers.

The memo introduces the idea of agencies using scientific, behavioral insights to improve the way employees do their jobs and increase the effectiveness of agency programs.

“Science has told us a lot about how people make decisions, the way in which we’re wired and how we sort of decide which way to go,” Kohli told The Federal Drive with Tom Temin and Emily Kopp on Aug. 5. “And what it’s saying is that agencies use these behavioral insights to find programs that are both cheaper and more effective.”

Deloitte developed six white papers that provide examples of how agencies could apply behavioral insights.

“It’s absolutely essential that agencies take risks, and I think that’s exactly what the White House is saying here,” Kohli said. “But it’s going further than that. It’s really saying, ‘Don’t just take a risk and not work out what impact it had. Actually, constantly measure the effectiveness of the new approach that you’re taking. Make sure your grant programs are designed in a way that your grantees are incentivized not only to try new and different things, but to establish the relative effectiveness of those things and always keep an eye on what works and what doesn’t work and use that information to scale the things that are most effective.”

One of the principles of innovation is being clear about what it is you’re trying to achieve, Kohli said. Also, you should look far and wide for the best ideas and then build simple protocols that allow you to get timely, useful data about the effectiveness of those strategies.

“I don’t think that requires you to have large, new departments or different kinds of people,” Kohli said. “And, indeed, a lot of work is not done in Washington. It’s done out in the states and among the grantees, so often you might have to help grantees and build that capacity in the space.”

For example, in a grant program that has 100 grantees, some will always get “more bang for their buck” than others, Kohli said.

“Often, federal agencies don’t even know which ones are which, let alone use that information in real time to help drive improvements in the program,” he said. “So, you can imagine an agency that’s more focused on innovation knowing which grantees are more effective and which get more bang for the buck, for their dollars, and then using that information to help those grantees at the other end of the spectrum really learn from the best grantees. That’s sort of innovation in action.”

Using behavior insights can help save money

Kohli said the British government has used behavioral insights to get more citizens to pay their taxes on time.

Her Majesty’s Revenue and Customs — the British equivalent of the Internal Revenue Service — changed the color of the envelope it used to communicate with taxpayers from brown to white and added a handwritten message on the envelope that said: “This is important. Please open it.” In the text of the letter itself, it added the phrase “Nine out of 10 people in your neighborhood pay their taxes on time.” With these simple changes, the British government saw the taxpayer compliance rate jump from 68 percent to 83 percent, which resulted in $45 million in savings.

“People seem to feel that they want to be part of the norm,” Kohli said. “People don’t want to be the odd man out. So, being very clear with them that they would be seems to have this effect. It’s hard to argue that you want people to pay their taxes on time. If you want people to pay their taxes on time, then a letter that’s well designed is clearly the right answer.”

According to the White House memo, the IRS is in the process of making similar changes.

“They’re also changing letter design, and they’re also checking to see which wording is most effective,” Kohli said. “And then, they’ll scale the ones that are the most effective and sort of grow those approaches.”

One of Deloitte’s white papers pointed to the Department of Agriculture as one agency that could also benefit from behavioral insights to reduce childhood obesity. The report described how Google applied behavioral principles to create “smart lunchrooms”:

  • Moving water bottles to eye-level increased water consumption by 47 percent
  • Moving the water bottles also reduced soda consumption enough to decrease caloric intake from beverages 7 percent
  • Moving M&Ms from transparent to opaque containers led to a 9 percent drop in caloric intake from candy in one week.

USDA could apply similar behavior insights in establishing new guidelines for school lunchrooms and thereby increase the effectiveness of one of its health initiatives.

Changing behavior through behavior insights

Another Deloitte white paper examined how behavior insights could be used to reduce the number of traffic-related fatalities on the nation’s highways.

The British Department for Transportation recently conducted research to identify the characteristics of the drivers who caused crashes. Many younger, urban drivers did not consider driving 40 mph to be more dangerous than driving 30 mph. But, statistics showed that a pedestrian struck by a car at 30 mph had an 80 percent greater chance of survival than a pedestrian struck at 40 mph.

The research also reported the younger drivers said that if they struck a pedestrian, they would experience extreme guilt. The British DFT launched an advertising campaign targeted at younger drivers using this behavior insight.

“As a result, DFT was able to decrease child deaths by 10 percent more than their target, two full years earlier. … By targeting the beliefs and motivations of particular groups at risk for negative behavior, DFT focused on what worked, rather than wasting energy and resources convincing the inconvincible,” the white paper said.

In the end, the trick is to find programs where applying behavior insights can help the agency improve its process throughout the organization.

“There’s a detailed scientific literature here that needs to be sort of translated into policy program language, and then, deploying three or four insights, trying some things and then constantly trying to establish which ones are being more effective and which ones aren’t,” Kohli said. “Then, once you know the approaches that are most effective, then you can scale them up and then really roll them out across your program.”

Read all six of Deloitte’s white papers on behavioral insights.

The sentence that can save government $45 million

Behavioral insights for responsible financial behavior

Behavioral insights for sustainable behavior

Behavioral insights for healthy behavior

Behavioral insights for transportation

Deploying behavioral insights

Copyright © 2024 Federal News Network. All rights reserved. This website is not intended for users located within the European Economic Area.