How to think about social science and governing
How to think about social science and governing
The things I wish I'd known in graduate school
My day job is a social scientist who studies how government works. It’s a great gig, and I see others who do the same as my peers. But how should we think about our work? What is the purpose of the applied study of government?
I've been thinking about these questions of late. Earlier this year I received the 2022 Herbert Simon Award from the Midwestern Political Science Association for my research about bureaucracy. Much of Simon’s career centered on thinking about issues of design — “concerned with how things ought to be, with devising artifacts to attain goals” — and positioning himself as a scholar to lead such design. He was a political scientist who started out in public administration, but reinvented himself to play foundational roles in the fields of both behavioral science (for which he won a Nobel prize) and artificial intelligence (for which he won the Turing Award).
The Simon award comes with a lecture, which I used to advance the argument that scholars like myself should be engaged in a project of embedding institutions of public sector change. My hope is that this discussion is useful to academics, especially younger scholars, but also for anyone trying to change things in government. It also offers a window for non-academics curious about how scholars think about their work.
You can read the full Herbert Simon lecture at Perspectives on Public Management and Governance.
Perhaps some of us came to do research on public sector institutions, or other applied areas of social science, with a very clear notion of the normative purpose of our efforts. I didn’t. I had studied public administration as an undergraduate at the University of Limerick, seeing it as a useful domain of knowledge, although I originally assumed I would engage in practice, mostly likely as a civil servant in my native Ireland. I moved to the United States thanks to a scholarship that allowed me to complete a Master of Public Administration (MPA) at the Maxwell School of Citizenship and Public Affairs at Syracuse University.
In the years that followed, whatever vague sense of purpose I had was overwhelmed by the demands of the next professional hurdle—becoming a PhD student, passing the methods classes and preliminary exams, completing a dissertation, stressing about the academic job market, publishing, and getting tenure. The effort in overcoming each hurdle tended to blur rather than sharpen the vague sense of purpose that originally drew me to the field.
Inattention to normative intent does not mean that the normative effects of our research disappear. Questions about the purpose and professional norms of an applied academic field are also necessarily interrogations about the role of those asking the questions, and how the two things interact. We cannot answer who are we? without relating our identity to our function: what purpose do we serve? Answering this question, in turn, depends upon both a mixture of normative concerns what sort of government do we want? what public values shall it serve? and empirical questions how do we achieve those values?
So, what is the purpose of the applied study of government?
This is, or should be, a question that all students of governance ought to explicitly consider and debate, and the answer they come up with should shape how they approach their work. To engage in this topic first means recognizing publicness and public values, and considering what obligations and constraints they pose for governing. It next compels us to probe and incrementally revise those obligations and constraints to devise compatible institutions.
The particular nature and constraints of government pose both a normative and design challenge that researchers cannot ignore. Public organizations lack market pressures for improvement. Democratic forces for change may miss the target, focused on political goals, some of which may themselves be antidemocratic or prioritize dysfunction over performance. Hierarchically imposed solutions, especially in authoritarian regimes, often lead to disastrous outcomes, as they miss the necessary feedback loops and experimentation needed for government to succeed. The applied study of government, properly understood, is able to not just bear witness to such problems but play a role in resolving them.
Academics can (but often don’t) play an essential role in helping to identify and test institutions of change. They can perform this role better if they aware of and purposeful about it. Such engagement means more than doing good empirical work. It poses two other challenges.
First, that we reflect explicitly about how our research relates to normative goals. Some are uncomfortable with this, believing that it turns us from dispassionate researchers into advocates. But research about the role, structure, and effectiveness of government is inherently normative, tied to often unstated assumptions of the role of democracy and markets, individual rights and the limits of state power, equality and how we evaluate organizational goals. It is just a matter of whether we are explicit about this fact or not. The case for being more explicit has become stronger as taken-for-granted normative goals come under more open challenge, via, for example populist or antidemocratic movements.
The second challenge is that we think about the world in design terms: not just to offer accounts of how it is, but how it could be. This must mean something more than a tagged-on discussion of policy implications. How might we better orient ourselves to do the hard work of fixing problems, and not just recording them?
Three characteristics of a problem-driven and solution-oriented approach to studying government
I discuss three characteristics of a problem-driven and solution-oriented approach to the study of governance. First, we have enormous agency in purposeful problem framing. Second, we can choose how to understand behaviors and empirical outcomes. Finally, we can design (and implement) solutions. None of these ideas are entirely new. But it may be useful to suggest both that they are central to our work and that we should think of them as part of a coherent professional framework.
Purposeful problem framing
The framing of problems is an exercise in choosing and prioritizing public values. Sometimes this is done implicitly, or barely done at all, prioritizing a description of gaps in prior research over the values at stake. Incremental academic improvement is important but being explicit about values makes it more likely that we purposefully consider what frame we are using, and why we are using it.
Framing involves asking what is the problem to which we should direct our attention? Why should we care about this? In a world of competing issues, why focus on this one, and this particular understanding of the issue? The frame both shapes what topic we look at and how we look at it.
Understanding behavior and empirical outcomes
The idea that researchers should empirically analyze behaviors and outcomes is the least contentious claim I could make, and so does not bear extended discussion. But it is worth explicitly stating, prompting us to ask of any particular problem framing not just “what are the practical implications?” but also “how shall we study these claims?” It involves making falsifiable claims about human behavior and public outcomes.
Designing solutions
We might assume that there is general acceptance of the idea that providing useful and relevant work to the study of government is relatively uncontroversial. But we have many more theories than widely agreed upon solutions to practical problems, reflecting institutional and cultural academic norms to raise rather than resolve questions.
The design of solutions can take multiple forms. In the context of government, it implies developing a proposed change in existing governance processes. It should explicitly rest upon problem framing. It can both flow from existing research (based on what we already know, here is a design that should work) and be the precursor to such research (let’s see if it works).
An example: the study of administrative burdens
To give an example of how these approaches to the study of government work, I draw from my own research on administrative burden, which examines the causes, natures, and consequences of people’s experiences of friction in processes of policy implementation.
Purposeful problem framing
The ideas of frictions are not novel in social science, or even in the study of government in particular. But they have been far from a central feature of analyzing how citizens experienced the state. The gradual emergence of the administrative burden framework over the last decade has changed this, arguing that burdens are a central feature of citizen–state interactions, that they have large effects, and that they often serve to reinforce patterns of inequality.
Academic research in this domain drew belated analytic attention to an aspect of government people intuitively understand. After all, everyone has experienced burdens in their daily lives. By distinguishing between types of costs (the learning costs of finding out about programs and services, the compliance costs of time and money spent on completing administrative demands, and the psychological costs from such interactions, including stress, frustration, and even fear), the framework allowed researchers to identify how specific types of frictions exist, can be measured, and generate specific types of outcomes.
In offering this framework with Pamela Herd and coauthors, we were explicit about normative goals, arguing that people’s actions with government should be simple, accessible, and respectful. This was not meant to ignore other values, but to call for an explicit consideration of value tradeoffs that would be informed by empirical evidence. In other words, policymakers should ask, and have answers to, questions like: if I add a work requirement, how many eligible program recipients will lose benefits? If I make voting easier, will it lead to significant increase in fraud? How much do documentation requirements or administrative fees discourage those with fewer resources from seeking help? The value of such analysis might seem obvious, but it has not been a feature of either social science or government practice.
Understanding behavior and empirical outcomes
The framing of administrative burdens both drew from and generated empirical research. Much of our initial work was to pull together research by sociologists, political scientists, public administration scholars, and economists within specific policy domains who largely did not think of their work as connecting to some broader project. The framework of burdens clarified the connections and common interests in such work, establishing a generalizable set of claims about how frictions worked in public settings and launching a new wave of research that consciously identified with and contributed to the framework.
Designing solutions
The example of administrative burden illustrated two pathways to solutions. First, it offered an opportunity to work directly with government in specific design alternatives to reduce burdens. As an example of the first pathway, I’ve looked to partner directly with governments and nonprofits. Second, policymakers took the administrative framework to institutionalize processes of solution-seeking within government. It helped that the framing itself was simple and intuitive, able to rest on relatable examples that all could understand.
Lessons for embedding institutions of change
Here are some general lessons about how researchers and institutions might engage to embed institutions of change.
Lesson 1: Both an institutional base and mechanisms of change are needed
Institutions provide the stability and familiarity necessary to embed routines and reshape culture. In non-market environments these are essential tools. Over time, institutions become self-reinforcing as associated norms become widely shared and accepted. The downside to these qualities are the risk of staleness, and maladaptation, as institutions fail to keep up with external demands placed upon them.
Institutionalization lends salience, legibility, and coherence to a management project where people do not have clear ideas about the nature of the problem or what to do about it. That is the current status of efforts to institutionalize within government the ability to routinely detect, evaluate, and minimize administrative burdens on individuals. Compare this to the application of cost–benefit analyses, which have been institutionalized as a means of performing the same function on costs that fall on businesses since the 1980s. Four decades later, current reforms such as a broadened reinterpretation of the Paperwork Reduction Act provide a belated effort to give similar attention to how individuals experience costs in their interactions with government.
Lesson 2: Institutionalization leads to bureaucratization
As ideas of public service governance are institutionalized, they become bureaucratized, i.e., formalized, made legible using bureaucratic roles and processes, and focused on what is technically and politically feasible via bureaucratic processes.
It is likely, for example, that while the research framework on administrative burden is explicit about structural factors, including politically engineered frictions, the application of that framework in administrative practice will focus on improvements that are politically and technically feasible. A bureaucrat may know they cannot change the need for certain eligibility requirements they believe are unfair or unnecessary, but will look for solutions that mitigate the negative effects of those requirements within their domain of discretion.
Just because such work does not address structural factors does not mean it is trivial. There are lots of reasons administrative burdens emerge, and even a bureaucratic approach to addressing them can be useful. The nature of bureaucratic organizations means that by themselves they will often miss the burdens imposed on the public. Status quo bias favors not improving processes as new options become available, and negativity bias, combined with an asymmetric attention to fraud, does not naturally spur efforts to increase access. Processes that correct for these tendencies by pursuing bureaucratic sources of burdens can therefore play a significant role, even if they drain political urgency and conflict from those processes. The tendency for institutionalization to exclude or render invisible certain types of politics underlines the need for other actors, such as academics and reporters, to continue this work.
Lesson 3: The process of framing and reframing never ends
It is useful to contrast the use of purposeful framing to create institutions to reduce administrative burdens with the dire needs to reframe the purpose of a much older set of institutional processes: US public sector personnel systems. The federal civil service system is almost 140 old, unwieldy, and threatened by Republicans.
Such reframing is difficult for researchers to engage in, since it demands either explicitly addressing or ignoring the anti-statist policy agenda of one of the two major American political parties, a departure from nonpartisan neutrality. Academic professional instincts and norms are, contrary to much conventional wisdom or what you see on twitter, not to engage in such explicit partisan criticism in their work. For scholars of the bureaucracy, the last few years have reflected a mismatch between us privately shaking our heads at antidemocratic threats, attacks on state capacity and the career bureaucracy, and our scholarship, which has been largely unreflective of these trends, although this is slowly changing.
Lesson 4: Find, build, and sustain communities of practice
Scholars are privileged with the time and critical training to think more expansively about how to make sense of the world than those working at the front lines of government practice. Of course, no individual researcher can change government. It takes, as they say, a village. Knowledge does not use itself. Frames, empirical research, and design solutions depend upon people to share, question, improve, learn from, and implement them. The existence and nature of those communities matters to the ability of institutions to change.
Engagement with communities of practice helps with every stage of the process of framing, research and devising solutions. It also matters to marketing. Marketing is another crude metaphor, but expressive enough to be useful.
Academic researchers spend a great deal of time generating a product, which is knowledge, and very little relative time and effort in marketing what they found to the audience who could most usefully apply that knowledge.
Engaging with policymakers and bureaucrats helps with marketing. Better yet, if researchers engage with policymakers and bureaucrats in the problem framing stage, they will have greater reassurance that their consideration of problems reflects practitioners’ understanding of public values, and that there will be a customer for the product that results.
Conclusion
My perspective will not be to everyone’s taste, or fit everyone’s brand of scholarship. Some might argue that blurring the lines between science and practice is unhelpful, though this raises the question of the relevance of applied research.
That’s fine.
I’m not trying to tell others that what they are doing is wrong, but rather better elucidate my own thinking about what makes studying government, or any other social topic, both an amazing opportunity and responsibility. Maybe it will resonate with others. If nothing else, I hope it will give both younger researchers and more sage hands a moment to consider what their purpose is.