How Is It To Be Done? Building a Social Science of Radical Reform

Abstract

The science of inequality has delivered robust evidence on the depth and severity of inequality in the contemporary United States. Yet despite the accumulation of evidence highlighting the consequences of profound inequality for individuals and communities, our inequality policy is largely focused on incremental interventions. Faced with this disconnect, sociologists reasonably ask what can be done to forge closer links between research and policy. In this commentary, for the Socius “Special Collection: Sociology’s Role in Responding to Inequality,” I consider just how such links might be forged. I argue that we must invest in infrastructures that support sociologists in developing policies that reflect the ambition and insights of our discipline. Of special importance is building institutions that increase cross-disciplinary research communication, improve access to administrative data, and increase the influence of sociologists in the public sphere.

What is to be done? We acknowledge the question with a wry smile, a recognition that radicalism is in the air. The question operates as a touchstone and a balm. It allows sociologists to believe that the aim of changing society is close and provides reassurance that our work matters.

Over time, sociologists have provided ever bigger and ever smaller answers to the question. The bigger answers weave narratives of oppression, exploitation, and power to paint a picture of obstacles that, if not insurmountable, are certainly immense. Although such narratives help sociologists in understanding the origins and persistence of inequality, the implied visions of future societies have a dream-like quality. These dreams may be important in inspiring the public into action, but they are not oriented toward the business of everyday—or even once-in-a-generation—policy-making. The smaller answers present estimates of a given policy’s effects on a given outcome in a given context or provide estimates of causal effects that might one day be amenable to policy intervention.

Sociologists with an interest in producing substantial change must find a path to combining expansive visions with scientifically grounded proposals that take seriously problems of viability, likelihood of successful outcomes, and implementation. But there will be little chance of spurring substantial change if success depends on urging individual sociologists to act or, indeed, urging those outside of sociology to listen to us. Our energies would be better spent on investing in infrastructures that support sociologists in developing policies that reflect the ambition and insights of our discipline. Put simply, we need to build a science of radical reform.

Where to Begin?

Our current policy infrastructure is characterized by incrementalism. For the most part, policy interventions introduced in the inequality arena are oriented toward preserving existing institutional structures and providing amelioration for the range of hardships that emerge in deeply unequal and high-poverty societies. Our inequality policy provision is subsequently limited in both size and scope. With respect to size, the extent of the inequality problem far outweighs the policies designed to mitigate its effects. For example, education policies do much to improve the educational opportunities of underprivileged children (e.g., Johnson and Jackson 2019), but they do not equalize the educational experiences of underprivileged and privileged children. With respect to scope, even if educational experiences were to be equalized, underprivileged children’s opportunities would still be limited because those opportunities depend on far more than just the quality of educational institutions and personnel. Because privileged children are provided with high-quality child care, enveloped in supportive extracurricular activities, live in safe and pristine neighborhoods, and attend the very best primary and secondary schools, their childhood and adolescent years amount to a cocoon of privilege (Jackson 2020). The effects of growing up in such cocoons of privilege will not be undermined by a few nudges, an “information intervention,” changes to “mindsets,” or any other of the small-scale interventions that have been proposed as solutions in recent years.

It would be impossible to survey our current policy infrastructure and conclude that current interventions are sufficient to ensure that children from privileged and underprivileged families face equal constraints on achieving educational and occupational success. Certainly no social scientist could believe that current policy could undo all of the harms produced by our social institutions. Yet it is seen as imprudent, and perhaps even indecent, to note that contemporary “opportunity policy” will never provide equal opportunity for all children. For much good work can be done in the area that lies between perfect equality of opportunity and laissez-faire (e.g., Wright 2010). It is here that the best of social-science-informed policies, such as those identified by DiPrete and Fox-Williams (2021) and Nalani, Yoshikawa, and Carter (2021), are to be found. It is here that significant improvements to a child’s life chances can be made.

Nevertheless, it is important to be clear that the majority of policy-oriented work in social science operates at one extreme: in support of incremental policy. This state of affairs is likely the result of several considerations (Jackson 2020). First, social scientists with an interest in making a difference understand that politicians and the general public are often reluctant to embrace far-reaching and fundamental reform. If the scale of proposed reform is inversely proportional to the probability of the reform being implemented, there are clear incentives to aim for incremental reforms that have the potential to do at least some good. Second, those engaged in policy-relevant research might be trying to satisfy several different objectives. Policies that have the potential to fundamentally alter our existing social system might threaten liberty or economic growth at the same time as they increase inequality of opportunity. Third, social science has developed in ways that strongly constrain the types of policies on offer. Increasing specialization within social science has reduced the breadth of our research and our capacity to speak in broad terms about the causes and consequences of inequality. Increasing emphasis on causation has pushed us to exploit techniques that speak to precise mechanisms operating within given contexts. Increasing publication pressures have raised the costs of undertaking expansive projects such that articles reporting narrow tests of well-defined mechanisms offer higher individual returns than they did in the past. These trends have pushed inequality researchers to take on smaller questions, with implications for the types of policies that can claim empirical support. Economists have been at the forefront of these developments, but the trends are also evident in empirical work within sociology. Our inequality policy did not just get small: Social science nudged it in that direction.

It is important to acknowledge that some of the changes that have occurred within the social sciences may have been beneficial to science while they have at the same time undermined the development of expansive inequality policy. When highly specialized researchers work on narrow problems, employing a well-defined and standardized logic of inference, the quality and quantity of research is likely increased (Becker and Murphy 1992; although see also Chu and Evans 2021). In part, the “frame-shifting” research that DiPrete and Fox-Williams (2021) highlight has come to dominate because sociologists and other social scientists have been successful in building a science of inequality that is cumulative, high quality, and robust. To be sure, improvements to this science could be made, particularly with respect to translating the research into actionable policy (see Nalani et al. 2021). But there is much to celebrate in the achievements of inequality researchers in recent decades.

As it stands today, then, the science of inequality has considerable strength. It has delivered a robust body of evidence that emphasizes the depth and severity of inequality. These achievements did not, however, come without a cost. A better basic science does not necessarily have to generate, at the same time, a science that is better at developing policy of the scale and magnitude that we need. Indeed, our science appears to have ushered in a fractured and disconnected policy infrastructure—with a multitude of interventions designed to tackle a multitude of social problems—that is difficult for poor people to navigate (Herd and Moynihan 2019). The mismatch between the development of opportunity policy and the extent of the inequality problem has only magnified in recent years. Although income inequality has been increasing for nearly five decades, the pandemic has introduced new types of health and economic inequalities (Chetty et al. 2020), making small-scale opportunity policy even less attractive. As Boushey and Park (2020) have noted, the United States was not prepared for the pandemic and associated recession, and recovery will require us to invest in large-scale, structural reform.

How is it to be Done?

It is relatively easy to document the extent of inequality and why policy falls short. It is much more difficult to propose scientifically grounded policies that are likely to have large positive effects. The authors in this special issue have challenged sociologists to consider how they might best orient themselves to the task of influencing policy (Gamoran 2021). All sociologists who aim to influence policy should undoubtedly think deeply about their role before they embark on this path. It is, however, unlikely that the discipline of sociology will achieve greater influence in policy-making unless we create new disciplinary structures that support individual scientists in this endeavor. Moral suasion and impassioned tracts are not enough. We instead need to build new institutions. These institutions must complement both the “pure” scientific work that sociologists of inequality engage in today and the radical policy agenda that is needed to address the problems identified in the scientific work.

I will focus in particular on three areas in which structural changes could be made to increase the likelihood that sociologically informed inequality policy could be implemented: interdisciplinary foundations, testing infrastructure, and persuasion.

Interdisciplinary Foundations

The high degree of specialization in contemporary sociology creates barriers to the development of a more-radical inequality policy because it reduces the extent of discussion and collaboration across specialisms, weakens coherence at the disciplinary level, and raises the costs of entry. Taken together, these barriers make it more difficult for voices outside of the policy mainstream to be heard and more difficult for sociologists to propose policies that address multiple threats to equality.

Developing structures that support sociologists in conducting research that spans several specialist fields should be a priority. DiPrete and Fox-Williams (2021) pointed to the role that Annual Review of Sociology can play in building bridges between descriptive and explanatory research and academic and policy research. In addition, the inequality field would benefit from frequent and comprehensive reviews of the specialist subfields of inequality research and from articles that survey recent important developments that have consequences for the field as a whole. It is probably not feasible for the Annual Review of Sociology to take on these additional responsibilities: The field of inequality is so large that the journal would soon be overwhelmed. Instead, we need an Annual Review of Inequality, an interdisciplinary journal that publishes systematic reviews of each inequality subfield, relevant methodological and theoretical work, and recent policy developments.

An Annual Review of Inequality would allow inequality scholars to obtain a greater breadth of knowledge and help to promote cross-fertilization among the subfields of inequality research. There can be no question that such a journal would need to be interdisciplinary. State-of-the-art research on inequality comes from across the social sciences: Economics, political science, psychology, and other disciplines produce research relevant to sociologists’ understandings of inequality and the policy tools that might be used to address inequality. Researchers working in the inequality field already have a responsibility to take account of relevant work from other disciplines, but this comes with significant costs, and it is a responsibility that is not always honored. Social media complaints about inadequate cross-disciplinary citation practices are legion, with economists coming in for particular criticism (e.g., Cirone 2021). Research on interdisciplinary citation practices shows that economists cite other social sciences at a lower rate than anthropology, political science, and sociology, but economists’ rates of interdisciplinary citation are increasing over time (Angrist et al. 2020: Figure 1). An Annual Review of Inequality would simplify the process of finding relevant work in different disciplines, encourage cross-disciplinary citation, and promote interdisciplinary research.

Building structures that synthesize literatures and connect inequality researchers to one another would have positive benefits for the science of inequality, and it would also lay the foundations for a more-radical policy agenda. First, new researchers would be more easily integrated into the science of inequality, and the costs of entry to new fields would be lowered. In turn, this could broaden the set of voices represented in inequality research and broaden the set of policy ideas discussed by researchers. Second, sharing knowledge across specialized subfields and disciplines is likely to raise the status of those areas of inequality research that are currently underappreciated by fellow inequality scholars and policy-makers. If the current state of inequality policy can be attributed to the strong influence of a particular subset of social scientists on policy discussions, highlighting key contributions from researchers across the range of social-science disciplines will ensure that the best research from multiple disciplines feeds into the policy conversation. Finally, collaboration across disciplines is more likely to occur when researchers are aware that relevant research is being carried out in other disciplinary contexts. Such collaboration makes more expansive and ambitious cross-disciplinary research possible, which in turn provides a strong scientific basis for more expansive and ambitious policy.

Testing Infrastructure

Ambitious policy proposals are difficult to subject to empirical test. One reason why nudges took over the policy world was because of a body of experimental work suggesting that large effects could be secured for relatively minimal cost (e.g., The Economist 2021). Similarly, quasi-experimental analyses have assumed greater importance in policy development; such analyses promise to establish the causal effects of given policies without the costs and inconvenience of a large-scale randomized controlled trial (Athey and Imbens 2017). If we take as given that policy-makers will not implement new—or more radical—policies without evidence that such policies will work to reduce inequality, our focus must be on developing an infrastructure that will allow us to test promising ideas and demonstrate the efficacy of proposed policies.

One important area in which institutional investments would be welcome is in facilitating access to administrative data. The records that federal, state, and local governments collect are immensely valuable to social scientists with an interest in policy, but gaining access to such records is far from straightforward. In the absence of a national register, access to administrative data is secured via personal contacts with government gatekeepers, membership of research teams, or (in some cases) applications to individual agencies. The hurdles to accessing administrative data limit the numbers of social scientists able to exploit these high-quality, large-n data-sets and limit the voices represented in discourse about policy. Traditionally, economists have been much more successful in gaining access to administrative data and sociologists less successful. But this means that the highest profile studies of inequality using administrative data are disproportionately found in economics and that economists have disproportionate power in shaping the public’s understanding of these data. Given the differences in sociological and economic approaches to the study of class, race, social institutions, and test scores (to name but a few), sociologists must prioritize gaining equivalent access to administrative data if our insights are to be properly represented in the public sphere.

Work is underway to improve scientists’ access to administrative data. The Commission on Evidence-Based Policymaking (2017:1, 21) concluded several years ago that “improved access to [administrative] data . . . can lead to an increase in both the quantity and the quality of evidence to inform important program and policy decisions” and that “government must improve its existing infrastructure and legal frameworks to enable more and better evidence building.” If a National Secure Data Service were to be implemented in line with the Commission’s recommendations, this would immediately expand access to administrative data for qualified social scientists and dramatically improve our capacity to evaluate and develop creative inequality policy (Hart and Potok 2020). ¹ To ensure that such initiatives are brought to fruition, individual sociologists can lobby both their elected officials and federal agency workers. The American Sociological Association and other professional bodies also have an important advocacy role. Although access to administrative data is a relatively dry cause compared to many of the other causes that might demand the attention of radically-minded social scientists, these data hold the promise of allowing us to conduct ambitious and large-scale research projects and to test ambitious and large-scale policy ideas.

It is important to underline that radical policy will require both qualitative and quantitative research evidence. Administrative data linkage is a priority for quantitative and qualitative studies; administrative data can be used to situate qualitative findings in context and to make inferences to population-level processes. Both DiPrete and Fox-Williams (2021) and Nalani et al. (2021) highlighted the contributions of qualitative research to policy development, and it is clear that this is a type of research evidence that sociologists are uniquely qualified to provide to contemporary policy-makers.

Persuasion

Without influence, all is lost. The intellectual work needed to develop creative and far-reaching policy, the empirical research needed to show positive policy effects, and the planning needed to implement policy are all worthless without attention from those with the power to make policy.

Two powerful groups are of special importance: The policy elite and the general public. The policy elite includes government bureaucrats at the national and local levels, nonprofit workers, and politicians. Nalani et al. (2021) pointed to the importance of developing research-practice partnerships to translate research into policy and pointed to the work of the National Academy of Sciences with Harvard’s Center on the Developing Child as an exemplar of this approach. Opportunity Insights at Harvard, the Center on Poverty and Inequality at Stanford, and the Institute for Research on Poverty at Wisconsin are also examples of policy-engaged research centers in the inequality and mobility field. Nonprofit organizations such as Equitable Growth, Brookings, and the William T. Grant Foundation provide additional opportunities for scholars to speak to policy-makers. As this extremely partial list demonstrates, there are multiple organizations that draw policy-makers into conversation with academic researchers, and no doubt much good is achieved through these partnerships. What is missing, though, is an umbrella organization that pulls together the work from these diverse organizations and summarizes the main themes for the policy elite.

Just as the Annual Review of Inequality would synthesize research for academics in the inequality field, an umbrella organization would synthesize important policy-relevant findings and circulate this research to the appropriate policy-makers. Although the American Sociological Association has resources to connect individual researchers with policy-makers and will write letters to policy-makers to advocate on behalf of sociologists, it does not play this synthesizing role. The organization that is closest to playing the role is the National Academy of Sciences (NAS), through their consensus study reports (which include the child poverty report discussed by DiPrete and Fox-Williams 2021). Possible approaches, then, to providing synthesized research reports for inequality policy-makers include developing similar initiatives within the American Sociological Association, founding a new organization that takes on this task, or expanding the number of inequality-focused NAS consensus study reports. Synthesizing research within such reports will make it possible for policy elites to engage with contemporary inequality research while removing the burden of “engagement” from individual researchers whose efforts might be better spent elsewhere.

Approaches designed to improve connections between researchers and the policy elite are unlikely to have impact on the general public, who are also key actors in the policy process. Frame-shifting research is largely oriented to shifting the frames of elites rather than the general public even while the general public knows less about policy and there is much potential for shifting understanding. Research on public opinion shows that the vast majority of Americans believe that society should do “what is necessary” to deliver equal opportunity to everyone (Rosentiel 2011). Social scientists know that “delivering equal opportunity to everyone” is well beyond our current policy infrastructure, but we are generally reluctant to state this in such stark terms (Jackson 2020). Communicating the basic facts of inequality to the public has the potential to increase support for the more-radical policies that would be necessary to deliver equal opportunity.

Three paths to increasing public support for sociologically informed policy would appear to be particularly promising. First, increasing support for sociological research on the science of persuasion would be beneficial. During the Coronavirus disease-19 pandemic, we have seen how difficult it can be to persuade the public of the individual and collective benefits of vaccination despite the demonstrated safety of the vaccine and its efficacy in preventing infection and serious illness. Clearly, investments in research on persuasion would have public health benefits, and we might expect spillover benefits for research on inequality reduction. Second, investments in the supports required for successful scientific communication would benefit all inequality researchers. Such supports might include association-wide training in translating research findings, network-building initiatives to connect sociologists to journalists, and graduate training in data visualization, social media communication, and public-oriented writing. Third, sociological perspectives on inequality are pushed out of the public sphere by perspectives rooted in economics. Expanding the reach of sociology through undergraduate training has the potential to increase the influence of sociological ideas in public discussion. At a time when young people are particularly attuned to inequality-reducing social movements, investing in undergraduate education is likely to have payoff.

What Next?

Pope Francis (2021) recently spoke of the need to develop an inequality policy that is sufficient to match the extent of the inequality problem:

Right now our brains and hands are not enough, we also need our hearts and our imagination; we need to dream so that we do not go backwards. We need to use that sublime human faculty which is the imagination, that place where intelligence, intuition, experience and historical memory come together to create, compose, venture and risk. Let us dream together, because it was precisely the dreams of freedom and equality, of justice and dignity, the dreams of fraternity, that improved the world.

Many sociologists of inequality would agree. We are living in a period of profound inequality, and there are few indications that the situation is likely to improve in the short or medium term. Our current policy infrastructure is insufficient to deliver on equal opportunity, and sociologists’ voices are largely crowded out of the policy arena. In this context, it is right that we ask: What is to be done?

But dreaming is not enough. We must ask how we can translate dreams of freedom and equality into policy. Dissatisfaction with current sociological influence on inequality policy should lead us to ask: How is it to be done? How can we design and implement sociologically-informed policy?

Discussions of sociological contributions to policy frequently deteriorate into grumbling about relative lack of influence or urging individual sociologists to act. This edited collection is to be welcomed for its constructive, practical approach to the problem of influence, and in this commentary, I have aimed to continue in the spirit of the prior contributions. I have focused not on the actions that each of us might take as individual researchers but, rather, on how we might develop new disciplinary structures that build connections between research and policy. There are structural reasons why sociological influence on inequality policy is relatively weak, and as important as individual motivations might be, it is only by investing in structural reform that we are likely to see substantial improvements. Many of the reforms that I have identified would contribute to building better science, as well as better policy. Sociology provides powerful tools for understanding contemporary inequality, and our inequality policy would be better for sociology’s influence. Focusing on the how in addition to the what is essential if we are to harness the power of our discipline to reduce inequality.