Under the radar: Getting social policy done in a divided Washington

The U.S. Capitol is photographed behind a chain fence

Who says nothing gets done in the nation’s capital? While it is true that many major issues, including such basic functions of government as passing a budget and taming the federal debt, go without resolution or even serious discussion, a number of important domestic social issues have been addressed in legislation in recent years or appear likely to be addressed soon – and usually on a bipartisan basis. Here, I review initiatives in various stages of their Congressional journey and speculate about what might reach the social-legislation agenda in a still-divided government after the election.


After enactment of the No Child Left Behind law in 2002 at the behest of the Bush II administration, the federal role in education became even more controversial than usual. Many Republicans found themselves with buyer’s remorse because the legislation was seen as empowering the federal government to impose burdens on states and to violate the longstanding principle of state and local control of education – the opposite of the direction conservatives preferred in trying to limit the reach of Washington. But Democrats came to dislike the law as well, in large part because it imposed achievement goals on states that proved impossible to meet and because of what many – including parents – saw as an excessive focus on testing.

After enactment of the No Child Left Behind law in 2002 at the behest of the Bush II administration, the federal role in education became even more controversial than usual.

Rising popular condemnation of gridlock put pressure on both political parties, producing a window for action on an issue on which Republicans and Democrats shared some common ground. Thus, a major bipartisan education-reform bill was enacted by Congress and signed by President Obama in December 2015. Among the provisions: more flexibility on what tests and standards schools may use to measure the progress of their students, what accountability goals to adopt, and changes in the preconditions for state takeover of failing schools.

Welfare Reform

Education is not the only social issue moving in Congress. This year marks the 20th anniversary of the sweeping 1996 welfare-reform law that replaced the landmark New Deal program, Aid to Families with Dependent Children, with the Temporary Assistance for Needy Families program. The differences between the two are substantial. For example, whereas the older program guaranteed benefits to destitute families with children, under the newer one recipients are required to work or prepare for work in order to get benefits. Similarly, whereas under Aid to Families with Dependent Children, states mostly just sent out benefit checks, Temporary Assistance for Needy Families requires each state to have a welfare-to-work program in which at least half of its caseload is enrolled.

Despite the fact that President Bill Clinton favored the reform and half the Democrats in Congress voted for it, many liberals outside Congress predicted disaster as a result of what they saw as the harsh provisions of the welfare-reform law. In particular, they worried that the tough eligibility requirements would lead to much higher poverty rates. In fact, during its initial years (the late 1990s), the new law was associated with an unprecedented increase in work rates by poor mothers and a major drop in poverty among children in families headed by single mothers.

In those early years, around 70 percent of the mothers who left welfare found work. True, most ended up in low-wage jobs. But other federal programs that had been designed to subsidize low-wage work of families with children and thereby make work more attractive made most of these families better off financially than they had been before welfare reform. Between 1995, the year before welfare reform, and the back-to-back recessions of the early 2000s, the poverty rate among never-married mothers and their children (an extremely poor demographic group) fell from 52 percent to 39 percent, the lowest that had ever been measured for this group.

A recent study by the nonpartisan Congressional Research Service concluded that the work-support programs reduced the poverty rate among all female-headed families by about half, and even held poverty among these families almost steady during the Great Recession. Thus, despite the recessions, low-income mothers still work more today than they did before welfare reform and the poverty rates for them and their children are, as a result, still lower.

But all is not well with the Temporary Assistance for Needy Families program. After two decades, the reforms seem to have run out of steam, and, as the law permits, states are using money from the program for purposes other than providing cash welfare and encouraging work. Meanwhile, an increasing number of mothers and their children, who before 1996 would almost certainly have received Aid to Families with Dependent Children cash, are not receiving benefits under the newer program because of its restrictive provisions and the way states are administering it.

As a result, the number of female-headed families in deep poverty (with income at or below one-half the poverty level, or about $10,000 for a mother and two children) increased, even as the overall poverty rate was falling. These mothers need help in the form of cash welfare or in preparing for and finding jobs, and many states are not doing a good job of providing either.

Congress is now considering ways to repair Temporary Assistance for Needy Families. This is a good year to do so because the program needs to be reauthorized this year. (In writing legislation, Congress often requires it to be reauthorized every several years, in order to force itself to assess the impacts of the legislation and to give relevant committees an opportunity to hold hearings and draft amendments.)

Surprisingly, Republicans and Democrats on the House Ways and Means Committee reached bipartisan agreement on a reform bill last year that included more education and other provisions that may have induced states to pre-emptively improve their work programs. But many conservative Republicans strongly objected to the bipartisan bill because they believed it weakened the work requirement and was too narrowly focused. The Republican Study Committee, an influential House group that speaks for the right wing of the Republican Party, held out for a much broader bill that included reforms of other welfare programs and that tightened – not weakened – the work requirement in those other welfare programs such as food stamps and housing.

The bipartisan bill written by the Ways and Means Committee was thereby doomed. And Republicans on Ways and Means decided to take the path of least resistance and drop nearly all the reforms in order to ensure that reauthorization for Temporary Assistance for Needy Families would pass this year. Failing to reauthorize the law would mean that the federal government could not give states the $16.7 billion in annual grants authorized by the 1996 law, an outcome that would be exceptionally embarrassing to Republicans and potentially disastrous to some of the poorest people in the nation.

The reauthorization bill that has now been approved by the House by voice vote (with the reluctant acceptance of Democrats) and awaits Senate action has only two provisions besides the language that reauthorizes spending. The first sets aside $100 million for states to conduct research on their programs – a welcome change in that it reflects a new emphasis on rigorous evaluation of social programs. (I return to this topic below.) The programs to be evaluated are ones that aim to help welfare recipients prepare for, find and keep jobs.

In addition, the research provision would create a clearinghouse at the Department of Health and Human Services that would make the results of the state evaluations available to other states and to the public. These clearinghouses are a fundamental part of the evidence-based movement that, although still modest, is increasingly influential in federal and state policymaking. The Temporary Assistance for Needy Families clearinghouse would provide detailed information about state and local welfare-to-work programs that have been shown to succeed, that have been tried but have not been well evaluated and that have been tried and shown to fail.

Whether this provision will lead to change remains to be determined. A problem preventing states from trying innovative programs is that many have already committed their Temporary Assistance for Needy Families dollars to other functions, such as child protection and abstinence-based sex education, which they were allowed to do under the 1996 reforms. The money is thus filling gaps in other parts of the states’ budgets; trying to get it back to ensure benefits for eligible families will set off a conflict between the state agency that controls the money and the state’s Temporary Assistance for Needy Families program. Place your bets on the agencies that currently control the funds.

The second provision, on social-impact financing (also called “pay for success”), is designed to promote a clever way to finance government social programs. The idea, developed over the past decade, is to use money supplied by private entrepreneurs (including foundations) to conduct intervention programs that would save government spending. If the government saving is actually realized, as determined by a high-quality evaluation, the investor is paid back with interest. The approach involves two of the fundamental tools of the current emphasis on evidence-based policy. Investors use their own money to support programs they have reason to believe are going to succeed and thereby save government money, which harmonizes with the goal of getting government to spend its funds on programs that produce results. In addition, rigorous evaluation is a central part of the approach because paying for success requires a reliable way of measuring outcomes.

These provisions on developing and testing programs that work to help poor parents and children, plus the reauthorization of the $16.7 billion in Temporary Assistance for Needy Families spending, allowed the bill to sail through the fractious House. There appears to be some opposition in the Senate to the way the House bill finances the pay-for-success provision. But there is a good chance that, before Congress adjourns in the fall, House and Senate negotiators will work out a final bill that closely resembles the House’s bipartisan bill. Then it will be up to states to serve their function as the laboratories of democracy and cure what ails the Temporary Assistance for Needy Families program.

Child-Protection Reform

Few people dispute that the focus of the program, helping poor single mothers support their families, should be a priority. But another critical issue is coping with child abuse and neglect, preventing it where possible and minimizing the consequences where it isn’t. Each year, around 700,000 children are reported by states to be victims of abuse or neglect. More than 1,500 children die each year from abuse or neglect, about 80 percent of them at the hands of their parents.

To deal with this problem, every state has an elaborate program supported by federal, state and sometimes local dollars and usually called child protective services. That program’s offices investigate reports of maltreatment and then offer a range of services and treatments to the adults and children who are involved in confirmed cases. Children are often removed from their parents’ homes (about 400,000 children are in foster care at any given moment). Most child protective services offices also offer prevention services, which are much preferred because they often allow children to stay with their parents and avoid the collateral damage of foster care.

Washington spends about $7.6 billion on these programs. But this money is used inefficiently. Because of the way the 1980 law that established the major federal programs was drafted, most federal dollars can only be spent once children have been removed from their homes and families – in spite of the fact that researchers and administrators have come to the view that many more abused and neglected children should be kept at home while their parents participate in treatment. On numerous occasions over the last quarter century, Congress has considered amending the law to permit more of the federal dollars to be spent on prevention and treatment. But the legislation has never passed.

Now the Obama administration and Congressional leaders are very close to agreement on changes that would give states more money for prevention and treatment. The legislation passed the House by voice vote. Interestingly, the House worked with the Obama administration to develop the legislation and to convince Democrats to support it. Now the House and the administration are working together to convince the Senate to put the bill on the president’s desk. As in the case of the education bill and reauthorization of Temporary Assistance for Needy Families, this shows that bipartisan action is still possible – in this case, on legislation that many previous Congresses were unable to pass.

Evidence-Based Policy

The most basic goal of the rapidly expanding evidence-based-policy movement is to focus government funds on social programs that work, if possible by taking money from programs that don’t. Although this goal seems so obvious that many people assume that government has always operated this way, nothing could be further from the truth. Most decisions in Washington and in state capitals are based on shifting mixtures of ideology, the personal views of elected officials and important constituents, the interests of contractors, the availability of funds to pay for the programs and so forth. Evidence of whether a program works has often been a minor consideration, especially when the federal government distributes billions of dollars to states through programs with broad goals, such as “to improve education,” “improve health” or “reduce child abuse and neglect.” Such goals are so general that evaluation is meaningless, and in many cases nobody even bothers to try. So policymakers and the public have no way of knowing whether the programs work. The heartwarming anecdote is often king of policymaking in legislative bodies.

But this is changing. The movement for evidence-based policy draws energy from the fact that it is difficult for elected officials to deny that they should want good evidence of whether a program is producing the intended impact – and that the evidence should be used to shift spending from ineffective programs to effective ones. And it’s about time: most social programs – including widely heralded programs such as Head Start and Drug Abuse Resistance Education (DARE) – don’t consistently produce good results.

One measure of the efficacy of evidence-based policy is the share of federal and state social spending that goes to programs with strong evidence of success. But the evidence-based movement has a quiver full of additional arrows. One of the most popular is the pay-for-success approach, mentioned above. Although still in the experimental stage, with more than 30 projects now going on across the country, pay for success is closely tied to rigorous evaluation of programs. Without it, entrepreneurs won’t have accurate guides to where success is possible – and administrators won’t know whom to reward and by how much.

The evidence-based movement is not something dreamed up by policy wonks that’s being foisted on the federal government from the outside. Rather, for nearly a decade now, both Congress and the executive branch have been buying into ever-expanding uses of evidence of program success or failure. In 2009, in a move to both expand evidence-based policy and test its effectiveness, Congress (with the approval of the White House) passed several pieces of legislation that established programs requiring those applying for cash to show they were using model programs with evidence of success, to employ high-quality evaluation procedures to test their ongoing effectiveness and to make the results of their evaluations public. Bills further expanding evidence-based policy are in the works.

In an even more striking example of the rise of evidence-based policy, Paul Ryan, the Speaker of the House, a Republican, and Senator Patty Murray, a Democrat, sponsored legislation creating a bipartisan commission charged with helping government and private researchers gain access to administrative data while ensuring that privacy is protected. Although increasing the use of this source of data may seem like a bureaucratic detail, there is now widespread agreement that data sets maintained by the Social Security Administration, the Internal Revenue Service and other federal agencies are gold mines of evidence about vital national issues. Especially important here is that access to the data can greatly reduce the cost of research and produce much more reliable results because of the huge sample sizes involved.

I’m convinced that the potential for evidence-based policy is enormous. Broader use, supported by a bipartisan coalition, could lead to almost immediate improvement in government programs that transcend the many previous efforts to improve their efficiency and effectiveness.

Prospects for the New Administration

The beginning of a new administration is usually a time of intense legislative activity. New presidents, knowing that a short honeymoon period is likely, are anxious to enact as much of their agenda as possible. Donald Trump has announced few, if any, social policies that he would implement if he were elected, unless trade protectionism, mass deportation and the exclusion of immigrant groups are seen as solving social problems. By contrast, Hillary Clinton has proposed a host of initiatives. Her campaign website lists 32 areas of domestic policy in which she favors specific legislative changes. If elected, she is likely to focus on a handful of them that stand at least some chance in a polarized Congress.

The beginning of a new administration is usually a time of intense legislative activity. New presidents, knowing that a short honeymoon period is likely, are anxious to enact as much of their agenda as possible.

My guess is that early-childhood education would be near the top of the list. And here, there is a real opportunity to find common ground with a majority in Congress.

Clinton is proposing a robust early-childhood agenda with three core goals: expand Early Head Start for children under age 3, ensure that all 4-year-olds have access to high-quality preschools and provide scholarships for preschools to low-income families. Liberals and most conservatives agree that the federal government has a role to play in equalizing economic and educational opportunity. And they agree that children from poor families fall behind during the preschool years, thus providing a solid rationale for beefing up federal involvement in preschool programs.

Starting with Lyndon Johnson, every president has expanded the federal commitment to early childhood development, either by creating new programs, expanding old ones or both. As a result, the federal and state governments now spend nearly $34 billion annually on these programs. But there are some children, including poor ones, who do not receive any federal support during the preschool years and many more who get federal support but have access only to substandard programs. Research has repeatedly shown that only high-quality programs boost child development in a way that is likely to improve children’s school performance and achievement in the long term.

Clinton has called for doubling expenditures on Early Head Start and in campaign speeches has said she would also increase Head Start spending. Although Clinton does not specify how much she wants to spend on the latter or how it would be financed, her proposals would undoubtedly cost many billions per year. If Clinton wants to capitalize on the fact that many Republicans see the value of preschool and accept federal involvement in preschool funding, she will need to work closely with Republicans on Capitol Hill to develop her proposals, which will have the effect of greatly reducing the scope of what she is now proposing.

Given the years of hostility between Hillary Clinton and Republicans, it seems obvious that no matter what the issue, Republicans will be wary and some will be hostile at the onset of a Clinton administration. And some will never change their reflexive rejection of all things Clinton. The new president and her aides will thus need to display great patience and thick skin in the initial days of the administration to get anything done – except, perhaps, in the unlikely event that a Clinton landslide ushers in Democratic majorities in both the House and Senate.

On Reflection

It would be Pollyannaish to pretend that business as usual has not been unusual in Washington for more than a decade, or that the partisan divide hasn’t taken a major toll on the quality of government. Even so, on some important social issues – especially those that can safely remain below the radar – Congress has been engaging in normal debate and discussion, generally following routine Congressional procedures and sometimes passing important legislation. We don’t know whether or how a Trump administration would develop working ties with a divided Congress, at least in part because we don’t know what Trump would want to do beyond his signature initiatives on immigration and trade. But the prospects for a constructive relationship – albeit, one at arm’s length – between a Clinton White House and the Republican Congressional leadership seem at least possible.