Solving Public Problems: An Interview with Dr. Beth Simone Noveck

This article was co-authored by Molly Smith and Ellie Vorhaben.

Dr. Beth Simone Noveck is a professor at Northeastern University where she is core faculty at the Institute of Experiential AI, and director of the Burnes Family Center for Global Impact. She is the Director of The Governance Lab, a think-tank with the goal to “improve people’s lives by changing the way we govern.” Dr. Noveck has a long history working for and with governments to increase transparency and improve efficiency and equity with the use of technology. From 2009 to 2011, she served as the Director of President Obama’s White House Open Government Initiative, an initiative signed into law on President Obama’s first day with the goal to increase “transparency, participation and collaboration.” She was appointed a senior advisor for Open Government by UK Prime Minister David Cameron, and was named New Jersey’s first Chief Innovation Officer in 2018. Dr. Noveck sat down with the Chicago Policy Review to talk about her 2021 book, “Solving Public Problems: How to Fix Our Government and Change Our World”. 

This interview has been edited for length and clarity.

Chicago Policy Review: One of the key ideas in your book is a need for upskilling the public workforce – you call it “providing free access to the 21st century.” This skills training wouldn’t just be teaching workers how to code, but implementing a cohesive mindset change and a toolkit that includes analytical thinking, human-centered design, and equitable engagement practices. You’ve had a unique experience of working with both American and global institutions on this idea of skills training. Can you talk a bit about the solutions you’re seeing internationally and how the U.S. might learn from these global programs?

Beth Simone Noveck: What’s very distinct is that other countries have a training strategy for the public sector. The United States government is the largest employer in this country, yet, unlike most private sector employers, the government doesn’t have a clearly articulated training strategy. One reason for that is that the legislative framework for upskilling in the public sector dates back to 1958, and there hasn’t been an executive order updating it since the 1990s.

So, contrast that with other countries. Singapore has a very strong civil service training college that has required coding training for public servants. I think you’re getting the point that this may not be the right strategy. If I had to choose what skill I would teach, everybody wouldn’t be learning HTML or C++. Coding by itself doesn’t translate into better services for citizens. But I do think that placing a bet on what we think people need to know in the 21st century and, especially after COVID, is a great idea.

The German, Canadian, and Argentinian governments all have free online government-wide training platforms. The government of India recently launched a new digital services training platform.

So, there are great examples of how different governments are taking bets on public sector upskilling. The US doesn’t have that at the federal, state, or local level. There are pockets of innovation happening, like California has developed significant programs to teach data science to public servants. There are other pockets teaching performance management or human centered design, but there doesn’t really exist an articulated strategy. For the most part, political leaders view government modernization as the third rail of politics.

CPR: So, politicians don’t want to be associated with pushing for upskilling the public workforce. What can we do to help people realize the importance of modernizing the government?

BSN: Well, this is another area where we can focus on the huge disconnect between the private and public sectors. Private sector publications like Harvard Business Review or Fast Company write every day about the future of work, or skill building among workers, or the mismatch between the skills that workers have and the skills that employers want. I think in the public sector, even among reformers, we tend to be more focused on hiring new people into government. It’s a related question, but you can’t transform how the government works at the margins by hiring new people. We have to do more across the whole enterprise to change how our people work, like learning how to use data to be able to solve problems. According to ProPublica, the US government spent $100 million in the first four months of the COVID-19 pandemic on services from McKinsey. And that’s in part because there aren’t people in government who have those kinds of agile project management and data analysis skills. But they have to start learning these skills before they join the government.

I’m curious, does your program at the Harris School require that you learn data science?

CPR: Yes, as part of our core curriculum in the first two quarters students learn coding in R and take required courses like Statistics for Data Analysis. Ever since the MPP started in 1995, there has been an emphasis on economics, statistics and research methods, although there has recently been a push towards learning coding in R and Python.

BSN: Well, Harris may be unique in that regard, because it is so quant heavy. But I would guess that some of that curriculum is relatively new. A survey completed in 2020 by researchers at Cornell found that 20 of the top 25 public policy schools did not have a requirement to teach data science. Many schools teach quantitative methods in some way and have taught something about data for a long time. But more modern data science techniques like machine learning are still not the norm in public policy education.

We need to start upskilling students enrolled in our public policy schools too, and that’s not just with regard to data. We have to complement both quantitative and qualitative approaches – data analytics versus human-centered design. To help people use these new tools available to us today to solve problems, not just faster and better, but also more equitably.

Another big gap that I see in working with public servants is a long-held culture of closed-door government – people who work in government do not have well-developed skills for engaging with the community. Now there’s a lot of focus within the government on equity and engaging with vulnerable populations, but it doesn’t mean that people know how to do that. So there is still a long way to go in terms of the skills that we need to develop in both the public sector and in universities preparing people for the public sector.

CPR: Picking up on your idea of engaging with the public: evidence suggests that when communities feel more engaged in designing a policy, there will be more buy-in to the final implementation, even if it might not be exactly what they had advocated for. Can you talk about how we can best include communities into the process of policy-making? A particular concern that I hear a lot from the future policymakers here at Harris is, how do we balance advocating for others without paternalistically speaking on their behalf?

BSN: It is a very real concern that you raised – in the end, we want implementable projects that actually improve real people’s lives, and we want to get there with the people who are being affected by the policy understanding the challenges and the solutions we present.

This is a learned set of skills that comes from experience. You should not intuitively know how to effectively do outreach in ways that are designed to tap the collective intelligence and collective action of a community. How do I organize that in a way that is truly inclusive but also respectful of people’s time? You can’t keep people for hours, they have jobs to do and dependents to care for. Designing effective and equitable engagement is a learned skill that requires the knowledge of available platforms and tools to enable efficiency.

You need to design the right process for what you are trying to accomplish. You’ll design a different process if you’re trying to identify problems as experienced by residents, versus if you’re trying to identify solutions to a problem with residents. If you’ve already instituted or legislated a policy and you want to understand if it is actually working – a process called social auditing – that’s going to be a different process. There are ways to engage residents at every stage of the policymaking cycle, from problem identification to solutions implementation to evaluation.

I’ll give you a counterexample. Today I spoke with some civic technologists building a fantastic new site for resident engagement. They’re excited about providing an opportunity for people to participate. But they only thought about it from the resident angle. Is it easy for people to get online? It’s clear that they have not thought about what the institution needs and in what format. So, they’re creating a process which will result in this institution receiving too many comments, so none of it will get used. People will get frustrated, and you will only see further distrust between citizens and government. So there are learned practices that we are developing, but we have to share that learning with people.

CPR: The way that many public frameworks are currently structured encourages public servants to pick and choose metrics to prove an artificial idea of success, rather than proving that a policy was actually impactful. You start the chapter on understanding problems with data with a quote from William Bruce Cameron: “Not everything that can be counted counts, and not everything that counts can be counted.” Can you talk about effectively measuring the success of policies, and the importance of testing and iterating in the service of creating real impact?

BSN: We have wonderful new tools at our disposal to measure impact in new ways, including the ability to collect and gather real-time data that gives us up-to-the-minute analytics to understand, for example, who’s getting a benefit versus who’s not. Previously, a think tank or academic would conduct a study for years after the fact and realize that the service is not working as well as ought to. Now, because so many things are delivered digitally, with the right policies in place to promote and create incentives for sharing and analyzing of data, we can much better understand what’s happening, in real time.

Additionally, we have communications technologies that now enable us to ask people how things are working – this new phenomenon of “social auditing.” We can engage with the communities we’re trying to help in real time to collect feedback.

But, we still need to be clear about what we’re measuring, and ensure that what we’re measuring actually improves people’s lives in an equitable way. During COVID, many people realized that impact has to be measured not just in the aggregate, but broken down by different communities – how does this affect minorities, immigrants, the poor.

We now have powerful tools that can allow us to get granular in a way that we could not have done before. For example, San Francisco has this program called TransBASE, where they look at bicycle fatalities and accidents. They’re able to identify where fatalities are happening, down to the street corner, and determine who lives there. You can gain a much more specific understanding of where a problem is occurring, and therefore measure the impact in those specific communities, rather than doing something wholesale.

CPR: We would be remiss if we didn’t talk about data versus storytelling. As we know, data alone doesn’t talk, and in learning how to run regressions and do statistical analysis, we’re learning how to translate data. What are some of your best practices for communicating data-backed ideas effectively?

BSN: I think your point about merging the narrative with the data analytical, or the qualitative and the quantitative, the stories and the statistic, is extraordinarily important because statistics give us a 10,000 foot view, but stories give us the on-the-ground perspective of why a problem actually matters, who is impacted and where it’s happening. Understanding the problem in context requires the coupling of human-centered and data-driven approaches because the data by itself doesn’t fully explain a cause. You really have to get at the narrative elements to understand both why a problem is happening and why it matters.

There are different ways to capture these stories and what people care about. For example, I chair the state of New Jersey Future of Work Task Force. We knew we wanted to focus on worker rights, safety, and upskilling. We made some assumptions based on data about unemployment, COVID and technology. We used an open-source free online tool to engage with 4,000 workers over two weeks, and what people cared most about was not robots replacing jobs; people were concerned about the cost of upskilling, or affording childcare during a training class, or the cost of healthcare. Any big policy issue, like the future of work or robots taking over, needs to be broken down to understand the root causes that really matter to people.

The best practice here is to say “yes, and.” You have to find the time and skillset to do both. It requires project management and planning to ensure that you have time both to get the data and to get the stories. It requires collaboration and partnership. Some people excel at the human-centered part and some people excel at the data analytical part. Partnering with people who have complementary skills to yours is extraordinarily important. I would not underestimate the value of collaboration and partnership rather than creating the unfair expectation of yourself that you have to be able to do both the stats and stories yourself.

CPR: You talk a lot in your book about not just training, but inspiring the next generation of leaders to solve public problems no matter what sector they choose to work in. We need to stimulate interest in public interest, not just public sector, work. Can you share any advice for future policy workers who want to work on solving these issues?

BSN: Survey data now shows that your generation is interested in doing well by doing good. People care about the values of the companies they work for, they want to make a positive social impact on the world, they care about social change. Of course people also worry about paying their loans back and pleasing their parents. They want to do good in the world and have purpose. That part doesn’t take much convincing.

But most people don’t consider what a career in the public sector could mean. Universities don’t do a good job of selling people on the idea of working for government, or on the idea of mission-driven work. Universities attract other kinds of employment as part of a generational dot com boom and an interest in private sector entrepreneurship. So the infrastructure has been created by universities to measure how many students go to work for tech companies, for example. These metrics are used as the hallmark of success to create alumni who can donate back to universities. I think people aligned more with those values twenty years ago.

We’re behind the times. I’ve seen from surveying my own students that there is an enormous desire to make a dent in the universe and do well by doing good. However, we are not tracking how many people go to work for nonprofits or for government. We’re not creating public sector career fairs, with the exception of public policy schools. Your colleagues in engineering or computer science or marketing all have something to add to a .org or .gov employer, but they do not get exposed to employers from the mission-driven sector.

There are so many options for people to do good in the world, often combining data, technology, and innovations. Universities need to retool the infrastructure by which they measure student outcomes and advise students on these opportunities to make social change.