The Time to Reinvent our Social Institutions is Now
Arundhati Roy, the prize-winning author and political activist, writes that “historically, pandemics have forced humans to break with the past and imagine their world anew.” Recent weeks and months have raised the possibility that tectonic shifts are looming, or even under way. The coronavirus revealed deep race-based disparities that have long been widely ignored. Then came the vicious police killing of George Floyd, an unarmed Black man in Minneapolis, which ripped the veneer off not only policing practices, but many of our core social institutions. Jelani Cobb explained the weight of the protests that followed: “In death, George Floyd’s name has become a metaphor for the stacked inequities of the society that produced them.” Years of police brutality targeting Americans of color suddenly became visible. Long ago, James Baldwin pointed to the importance of this visibility when he wrote, “Not everything that is faced can be changed, but nothing can be changed until it is faced.”
If you’re an optimist, as I am, you see the potential of fundamental changes occurring in response to our having faced the profound meaning of the race-based disparities in the toll of the coronavirus, and the stacked inequities symbolized by the death of George Floyd. The possibility that big ideas could suddenly take hold might have seemed fanciful even months ago. But a global recession, skyrocketing unemployment, a pandemic that could be with us for years, and an unprecedented degree of racial awareness represent the kind of shock to the system that could indeed be the start of massive rebuilding.
The prospect of the laws, regulations, and practices of police departments throughout the country being changed profoundly in the near future seems plausible. The responses to the economic shocks wrought by the coronavirus pandemic are still in turmoil and their effects unclear. In this essay I will address what we have learned from the past that should inform our efforts in two other domains: What it will take to build a reliable anti-poverty safety net, and to re-imagine our social institutions so they will assure decent life prospects for all, including for those whose life prospects are now grim.
A Reliable Safety Net
We should not have needed the pandemic to learn how little safety our social safety net provides. The safety net, with its Supplemental Nutrition Program, Earned Income Tax Credit, Temporary Assistance for Needy Families, Unemployment Insurance, and Medicaid, has always been surrounded by controversy — how much public funding should go into its various components, how expenditures are allocated, and the conditions that define who qualifies for the benefits. The protection provided by the US safety net has always been weak when compared to that of other rich countries, most of which start with universal health insurance and a minimum cash income for families with children
Despite the strong evidence of the effectiveness of each of its elements, the safety net has been funded far short of its potential. The lasting scars of early deprivation have been clearly documented, with poor children growing into less healthy adults, with lower earnings and higher arrest rates. Each part of the safety net has turned out to be a bargain when its costs are compared to the long-term savings it brings. For example, children with access to food stamps experience a significant reduction in the incidence of obesity, high blood pressure, heart disease, and diabetes in adulthood; for women, food stamps are associated with an increase in economic self-sufficiency.[1] Expanding Medicaid coverage to low-income children has been shown to have significantly increased their high school completion, college attendance, and even college completion. Children whose mothers gained eligibility for prenatal coverage under Medicaid have lower rates of chronic conditions as adults and fewer hospitalizations related to diabetes and obesity.
But the evidence of long-term benefits has quite consistently lost out to easily fanned fears that the safety net could undermine the will to work, and concerns about whether the beneficiaries are truly deserving. The search for the drivers of the animus againstthe protections of the safety net most often finds race at the top of the list. In his March 2020 book, American Poison, Eduardo Porter of The New York Times puts it starkly: “Americans have repeatedly rejected expansions of the social safety net because it inevitably collides with one of the most powerful forces shaping the American experience: uncompromising racism.” Porter’s assertion is backed by research showing the centrality of “racial resentment” in widespread public opposition to welfare programs, although African Americans have never been as high a proportion as White recipients of Medicaid, food stamps, and the other means-tested welfare programs that make up the safety net.
The association of welfare and race has a long history, and was solidified when Ronald Reagan, first as governor of California and then as president, set out to tear down supports for poor people through regular broadsides against the Black “welfare queen.” He painted a dramatic picture of a high-rolling Chicago grifter who used 80 names, 30 addresses, and 15 telephone numbers to collect not just cash welfare payments, but also food stamps, Social Security, and veterans’ benefits for four nonexistent deceased husbands. The legend of the Cadillac-driving fur-draped welfare queen hardened into a stereotype that was deployed for years to delegitimize benefits for the poor, with brutal effectiveness.
Donald Trump picked up the Reagan mantle in associating race and welfare — whether in awareness or ignorance of the facts is not known. NBC reported that in a Spring 2017 meeting between the newly elected president and the Congressional Black Caucus, one of the members mentioned that cutbacks would be detrimental to her constituents receiving welfare payments — adding, “Not all of whom are Black.” The president was reported to be incredulous, saying “Really? Then what are they?”
Throughout his term, President Trump has chipped away at the social safety net, proposing budgets that gutted housing assistance, food stamps, and health insurance for the poorest Americans. When Congress rejected those cuts, the Trump administration enacted rules to make it harder to qualify for federal benefits, pushing states to make eligibility dependent on beneficiaries finding work. In January 2018, the Centers for Medicare and Medicaid Services announced that it would allow states to implement work requirements as a condition of receiving Medicaid coverage. If a beneficiary doesn’t meet the work requirements, he or she is locked out of Medicaid coverage.
Increasingly, the Trump Administration’s safety net was made contingent on work, weakening the protection to the most disadvantaged families as well as the defense against even mild economic downturns, not to speak of the shattering economic collapse from the coronavirus pandemic. Amid the shuttered businesses and workers laid off, evicted, and getting sick, even conservatives noticed the failings of the safety net. “The crisis has profoundly changed the playing field and has made the need for benefits much more visible,” according to Columbia University professor of social work, Jane Waldfogel. She adds that the crisis has also made the people now receiving expanded benefits, including sick leave and family medical leave “much more sympathetic.” It has also become clear that the roots of the disparities that have resulted in COVID-19 infecting and killing African Americans at far greater rates than any other group are deep, and they are embedded in the institutions and arrangements that we must now reinvent.
It is hardly a big leap to ask whether the time might be right for building a strong and lasting post-coronavirus safety net.
Darren Walker, president of the Ford Foundation, while applauding the philanthropists who are playing an essential short-term role in ameliorating the crisis, calls attention to the ideas emerging from grass roots advocates. Many of these proposals, like guaranteed sick leave, increases to the minimum wage, and even steady cash transfers to low-income households, would go far to strengthen the safety net permanently. He points out that these ideas were once deemed marginal but are today being given serious consideration by policymakers in Washington and state capitals. By adding universal health insurance, and expanding food stamps and housing vouchers, we would have built an economic floor with proven long-term benefits for all of us — not just the direct recipients.
Are the prospects of such momentous changes realistic? Pew surveys have been documenting over the years the steady growth in Americans’ political divisions. While partisan differences on governmental help to the needy grew steadily throughout much of the 1990s and 2000s and were highest in 2019, recent post-coronavirus numbers show dramatically increasing bi-partisan support for doing more to help the needy, even if it means increasing the national debt. Not only does the body politic seem more receptive to change, but with so many people plunged into unemployment and health crisis it becomes ever clearer that we won’t be able to recover our national strength without building massively on the short-term reforms already underway.
Re-Imagining Our Social Institutions
The imperative to re-imagine our social institutions that provide education, health care, early childhood and family supports, juvenile justice, and safe and supportive housing and neighborhoods, comes from two sources. First, there is unprecedented pressure to reinvest at least some of the money now spent on policing to design, in Michelle Alexander’s words, “alternative approaches to poverty drug abuse, mental illness, trauma and violence that would do less harm than police, prisons, jails and lifelong criminal records.” Second, and closely related, there is a new determination to assure that our social institutions will function in ways that will eliminate disparate outcomes by race and class.
The data that illuminate disparities in a full range of outcomes have long been clear, but never as striking as they became in the context of the toll taken by the coronavirus. The roots of the disparities that have resulted in COVID-19 infecting and killing African Americans at greater rates than any other group are deep, and they are embedded in the institutions and arrangements that we must now reinvent.
African Americans who live in cities are most likely to live in neighborhoods of concentrated poverty. Their jobs are at the bottom of the economic ladder and leave them vulnerable. Their children have been going to schools where they are less likely to succeed, beginning in kindergarten and extending through elementary and high school. African American families tend to be relegated to the least sanitary, most substandard, crowded, segregated housing, and they are exposed — at much higher rates than the rest of the U.S. population — to lead-poisoned contaminated pipes and toxic waste, to polluted air and the constant threat and stress of evictions. They are more likely to suffer from the “underlying conditions” that are correlated with more serious and deadly cases of coronavirus, such as diabetes and high blood pressure. And as Black people are laid off from work at higher rates, their families are subject to the threats of eviction, food insecurity, homelessness, and the attendant scourges of poverty.
These conditions are not the result of personal choices but of policies that go back generations. The Federal Housing Administration, the New Deal creation celebrated for expanding home ownership, refused to back loans in predominantly Black neighborhoods, or for Black people altogether. New Deal labor codes allowed businesses to offer Whites a first crack at jobs and authorized lower pay scales for Blacks. In their first incarnation, Social Security and the Fair Labor Standards Act excluded domestic and farm jobs, which employed two out of three Black workers.[2]
Probably as important as any of these in explaining how African Americans have suffered disproportionately from the coronavirus is the long history that civil rights historian Bryan Stevenson identifies as “the narrative of racial difference,” created to legitimate slavery and, continuing to the present, to justify discrimination against people of color, with its “presumption of dangerousness that weighs on you.” Physicians and public health epidemiologists cite research which shows that the “wear and tear of racism” combines with the failures of our health care system to add up to the physical price that being Black exacts on the bodies of African Americans, and has made them so susceptible to being infected by and dying from COVID-19.
Beyond Programs and “Scientific” Evidence
A critical examination of past efforts to strengthen our social institutions suggests that they have been largely unsuccessful because our goals, our vision, and our investments have been far too modest. We have been trying to fix isolated pieces of the disparity problem with circumscribed, disjointed and underfunded remedies, which have contributed only marginally to better outcomes.
The distinguished sociologist, William Julius Wilson, has observed that “Years of piecemeal approaches to social problems and an unrealistic appreciation of the interconnections of economic distress and social maladies in the inner city gave rise to inadequate social services and the fragmented infrastructure of providers in these neighborhoods.”
The fundamental changes now required mean going beyond yesterday’s piecemeal thinking — and the well-intentioned, but ultimately pernicious, pursuit of “scientific” evidence of effectiveness that has produced the fractured approaches of the past. These piecemeal measures have been part of the overarching focus on programs, even though a clear picture of past efforts would show that single programs — even collections of isolated programs — don’t bring about the depth of change that is now needed. The program-centric focus of the last several decades has resulted in significant part from the practice of funders trying to minimize risk by investing in isolated programs that have been shown by randomized experiments to have worked in the past.
Both government and philanthropic funders have prioritized the replication of programs that had been found effective in randomized experiments. These interventions qualified for the “evidence-based” label using a narrow definition of evidence. They landed in “what works” directories because they were conceptually and operationally neat and could be standardized, often in manuals. They had an easily identified causal relationship to a readily defined outcome. They were reliably shown to work somewhere, even though they wouldn’t work elsewhere. They lent themselves to randomized trials because they stood alone, allowing evaluators to avoid the complexities of measuring the impacts of initiatives whose effectiveness lay in the complex ways they interacted and were embedded in the community.
The appeal of randomized program evaluations comes from the certainty they bring. The idea that information about “what works” has to be certain to be useful has been misapplied in the process of transferring from the biomedical sciences to social interventions. The allure of certainty is also bolstered by the hope that the principles of rational management will discipline otherwise “soft” social policy. The narrow view of reliable evidence benefits as well from claims that the alternative is to rely on opinions, prejudices, anecdotes, and weak data, and to privilege hunches over facts. The current climate in the U.S., in which hunches do seem frequently to be trumping facts, makes it tempting to embrace the use of over-simplified metrics with the certainty they seem to bring.
But there has been a massive misunderstanding: the degree of certainty that is required is not the same for every purpose. We need the certainty that comes out of randomized experiments when we are looking for a vaccine to prevent COVID-19, and to determine which drugs are safe and effective.
While there is a small sub-set of social interventions that are simple enough to be accurately assessed by randomized experiments,[3] when it comes to designing and assessing complex responses to complex social problems, the cost of certainty is too high. In the realm of social problem solving, any knowledge base that consists only of randomized program evaluations is too narrow, and severely limits the universe of strategies that policy makers, funders, and communities can draw on, because it does not include the complex interventions, or the policies and strategies that may hold the greatest promise of improving important outcomes.
Because, over the last several decades, funding went primarily to the programmatic social interventions that were shown to work by randomized experiments, our interventions have typically consisted of highly circumscribed slivers of help. We have sent trained home visitors to provide support to first-time mothers of infants, and mandated that fragile families attend parent education classes to address gaps in their knowledge of child development — even though more comprehensive interventions would have found that many of these children would have been better protected by income supports, housing subsidies, and health care. We have parachuted incentive programs into school districts to attract more effective teachers to work in high poverty schools, leaving untouched the dismal working conditions that caused them to lose their best teachers in the first place. We have implemented new curricula to raise academic achievement and self-esteem among imprisoned young delinquents only to see their impact obliterated by more powerful messages that the young people heard every day, that they were dangerous, feared, worthless, and had no real future.
The time for reinventing our social institutions is propitious. There is a growing enthusiasm for bold and assertive change at federal, state, and community levels. There is a fresh readiness to systematically capture evolving knowledge and new learning. There is an eagerness to broaden our understanding of “what works” and what will work in complex efforts to improve lives and neighborhoods, with their multiple, interacting causes and complex, multiple outcomes. We can build on contemporary real-world examples that it can be done by expanding how we think about what constitutes credible evidence.
Innovative social problem solvers are collecting and putting together evidence that enables them to define a problem and to design solutions where there are no models to replicate. Upstream USA, for example, is filling an infrastructure vacuum that occurred when a new means of contraception became available (known as LARCs) but were not widely used. Upstream creators worked with front-line providers, funders, existing systems, and potential users to learn why, and to put in place the complex, multifaceted links that had been missing. They helped local and state agencies to “re-engineer” the structure of health care in ways that would enable all women to conveniently access the full range of contraceptive methods, including the most effective — IUDs and implants. They were able to expand economic opportunity by empowering women to reach their own goals of becoming pregnant only if and when they wanted to.
Innovative social problem solvers are designing solutions to unsolved problems by generating the evidence they need from networks that include their target populations, content experts, and the day-to-day experience of practitioners. The Carnegie Math Pathways is an early example of an organized network using multiple sources of evidence to deal with a very specific problem: the distressing reality that each year, about half a million students in the U.S. hoping to enter community college are so far behind in their understanding of math that they are assigned to remedial math classes as a prerequisite to taking college-level courses. Over the last several decades, more than three quarters of the students in these remedial courses never completed or passed them. This meant their access to higher paying 21st century careers would be fundamentally foreclosed. Determined to “transform mathematics from a gatekeeper to a gateway,” the Carnegie Foundation leadership knew it couldn’t be done by adopting just one more proven program from a “what works” directory. It would take a combination of people with direct experience and diverse talents and perspectives, joining together to construct new approaches to solving the problem in ways that no one had done in the past, and none of them could have done alone. Today, with a collection of similar networks across a large number and variety of contexts, students are earning college level math credit at triple the rates of comparable students and in half the time.
Other innovative social problem solvers are identifying where the needs are greatest and the solutions least clear. Researchers in several parts of the country are exploring the long-term effects of family and neighborhood poverty, high levels of violence, incarceration, homelessness, and environmental degradation. They have demonstrated that cleaning up toxic waste dumps may be more effective than treating the damaged lungs of neighborhood children, and that reducing mass incarceration may be a better child welfare investment than training more case workers. One team of researchers under the auspices of Opportunity Insights, founded in October of 2018 by Harvard University’s Raj Chetty and Nathaniel Hendres, and Brown University’s John Friedman, are gathering, analyzing and disseminating a wealth of “big data” to shed light on the causes of increasingly limited economic mobility. While most previous work on racial disparities has studied inequality within a single generation, these studies clarify how racial gaps change (and do not change) across generations, and how neighborhoods are affecting the futures of children of all backgrounds, including those that have the best chances of getting ahead and those with the greatest chances of being left behind.
As we look forward to the day that the devastation of the coronavirus pandemic has receded, the economy has recovered, and policing responsibilities have been redesigned — and as we prepare for a uniquely critical presidential election — those of us who consider ourselves social change agents are accumulating a long To Do list. I am hoping that some of us will include a focus on expanding our understanding of the evidence that will be most useful in reinventing our social institutions. We would broaden our definition of “what works” and what will work in complex efforts to improve lives and neighborhoods, with their multiple, interacting causes and complex, multiple outcomes. We would illuminate the evidence that has been hidden, sidelined, or altogether missing because it lacks certainty.
By expanding our capacity to generate, disseminate, and apply a full range of evidence, we will be able to build the social structures that would result in all of us living in a nation of universal decency.
Now that we have been enveloped by crises filled with so much sorrow and chaos, a break with the past seems entirely plausible. The thoughts here come out of the conviction that each of us can and must contribute ideas from our particular vantage point on exactly how we might imagine the world anew — and how we can make it so.
[1] At the current moment of crisis, SNAP, the food stamp program, can furnish the emergency rescue potential we now need. By expanding eligibility based on economic, individual and community need, SNAP could empower hungry American families to buy milk, cheese, bread, cereal, fruits and vegetables and basically all the other food the average American buys at the grocery store. We would no longer have to witness the thousands of people in cars lined up at food banks that don’t have the infrastructure to handle the pandemic level influx of people and food they are facing in the absence of an expansion of food stamps.
[2] The irony here is that Lyndon B. Johnson’s Great Society reforms of 1964, which ended the exclusion of people of color in the network of rights created by FDR in the 1930s, may also, according to Eduardo Porter of The New York Times, have “undermined public support for the safety net altogether.”
[3] The minority of social interventions that are best understood by randomized experiments have the following attributes:
● They are conceptually and operationally neat and simple enough to be standardized
● They can be replicated with fidelity to the original because they have a linear, tightly coupled, causal relationship to the outcome of interest
● They can be implemented without requiring changes in practice or in the systems environment within which they will operate.
