The big news in Olivia Golden’s book is that child welfare can be reformed. Her proofs: three entire systems, Alabama, Utah¹, and the District of Columbia, although, as we shall see, D.C.’s reforms come with question marks. Golden’s book is not merely a collection of three uplifting stories. It is also analytic, comparative, and prescriptive. It is an impressive achievement.

Golden’s success stories are big news because they contradict the prevailing pessimism, a pessimism so entrenched that few writers tackle the subject of systemic reform. When they do, it is mostly to lament the chronically dismal quality of child welfare services.

Years of failure prove that failure is easy, but even one success shows that reform is possible. Golden’s prime example is the District of Columbia, where the child welfare agency had a long and shocking history of failure. Sued in 1989, operating under a consent decree since 1993, the agency was consigned to a court-ordered receivership in 1995. It was still mired in crisis in 2001, when Golden arrived. Immediately, she was confronted with emergencies that had become routine, such as dozens of children confined overnight amidst the agency’s desks and file cabinets. Most of the children on the agency’s books never saw a social worker.

The barriers to reform were uniquely formidable. D.C. residents are mostly wealthy Anglo Americans or impoverished African Americans. About half of the District’s foster children lived in Maryland, with all the potential for jurisdictional infighting inherent in interstate placement.

Golden had several levers that helped her achieve success. First, the 1989 lawsuit was still active. Without that pressure, it’s unlikely anything would have been achieved.

Second, Golden was high-powered and well-connected. She came to D.C. directly from her position in the Clinton administration as Assistant Secretary of Health and Human Services for Children and Families. In that capacity, Golden headed up the federal agency responsible for oversight of all state child welfare agencies.

Third, Golden reported to a mayor, Anthony Williams, himself a former foster child, who made child welfare a top priority.

Fourth, the litigation gave Golden the power to bring in people of her own choosing to take on the most important positions in the agency.

Fifth, Golden concentrated immediately and continually on improving data management. The systems she inherited reported average caseloads of 35, but in fact many caseworkers had 90 to 100 cases. After months of digging, it became apparent that some caseworkers had secured positions where they had few cases or none. Others were given few cases or none because of “perceived incapacity,” perhaps a euphemism for incompetence. Ultra-low caseloads for some meant impossibly high caseloads for others. “[N]o investment,” Golden states, “is more important than ensuring accurate, relevant data,” a judgment she backs up with convincing evidence.

Finally, when Golden took over, the agency was hobbled by poisonous conflicts with other bureaucracies, especially the courts. Golden invested many, many hours in working with the judges, repairing relationships, and ultimately convincing the judges to reorganize themselves, creating a family law court.

One implication of the D.C. example is that improving the system required the simultaneous pressure of lengthy litigation and the high-powered, well-connected Olivia Golden. What is most ominous is that the improvements during Golden’s years as director have been at least partially lost since her departure in 2004. The agency had four directors in the subsequent five years. The backlog of overdue investigations “skyrocketed,” reaching 1,600 by 2008. Legal action has resumed, with the plaintiffs asking the court to hold the District in contempt.

Golden’s coverage of Alabama and Utah is much less detailed.

Reform efforts were successful, according to Golden, in all three jurisdictions. In Utah, positive results included:

“doubling of the number of caseworkers and reducing caseloads to 13-15 cases per worker; extensive caseworker training, adoption of key practice skills that focus on addressing unique family circumstances and needs, dramatic improvements in data collection and assessment; and high-quality provision of health services to foster children.”

The implications of Golden’s three examples of reform are both positive and negative. On the positive side: systemic reform is possible. It would be hard to overemphasize the importance of that finding. The literature on American child welfare is profoundly discouraging and discouraged. If the prognosis for child welfare systems were dismal, reform efforts would be pointless. But Golden reports that D.C., Alabama and Utah improved, and they improved a lot.

Some cautionary implications of Golden’s three examples are easy to see. All three required multi-decade litigation. In D.C., Dr. Golden, with her influence and connections, seems to have been an essential ingredient. Alabama and Utah also benefited from a talented leader, and expert on child welfare reform, Paul Vincent. In Alabama, Vincent was, as the child welfare agency’s administrator, directly responsible for leading the agency through the reforms.  In Utah, Vincent worked with defendants and plaintiffs to foster and implement reforms. Vincent then went on to found and direct the Child Welfare Policy and Practice Group, a nonprofit based in Montgomery, Alabama, which provides technical assistance to states and organizations involved in child welfare system improvement.   

In the most important section of her chapter on research, Golden brings to bear lessons of reforms in systems other than child welfare. Errors are inevitable in complex systems like child welfare. Research in medical settings has identified systemic characteristics that encourage errors and systemic changes that reduce errors, changes such as simplification of complex procedures, promotion of safety as an organizational value, and creation of venues where staff can report errors without risking their careers. Such research in medicine has led to major improvements: fewer hospital-acquired infections, fewer medication errors, and much fewer catastrophic errors in anesthesia. In child welfare agencies, analogous research hasn’t yet been done. It is time to do it, Golden says.

The second half of Reforming Child Welfare moves from state-by-state narrative to a practical, topic-by-topic analysis, beginning with “Vision, Strategy, and Action.” In Alabama, Utah, and D.C., the reformers articulated their visions again and again. In D.C. the goal was that “children should live in families (including kin, foster, and adoptive families) and not in shelters or group homes.”

Alabama, Utah, and D.C. organized their action strategies in three different ways, but with this common theme: they developed their strategies in part via action, for only by making changes could they identify the problems and experiment with solutions. Golden believes this empiricism was essential for two reasons.

First, long delays for planning would have eroded the reformers’ credibility. “Staff and outside observers have become cynical after years of watching visions disconnected from actions and results, and will not take the vision seriously without the accompanying action.” Second, the real problems, and what is and isn’t possible to remedy them, often become visible only in the context of change.

Golden believes reform of information systems is so important that she devotes two chapters to the subject. She observes that information issues are only partially technical. They also reflect maneuvers on the chessboard of organizational power. She writes of efforts to obtain accurate data on foster care home licensing:

“ . . . just as I had concluded earlier that agency managers didn’t seek accurate information partly because they were afraid of what the data would show, I realized that behind the social workers’ mistrust lurked fear that a licensing review was just asking for trouble. If the home could not be licensed . . . yet a child had lived there happily for a long time, where would that child go?”

Golden realizes that data accumulation for its own sake, not carefully designed to mesh with goals, is demoralizing and counterproductive. Performance improved when measurement focused on truly meaningful goals such as “[i]ncreasing visits to children, reducing the backlog of uninvestigated abuse and neglect reports . . . and increasing the number of finalized adoptions.”

Measurement is the foundation of accountability, and holding people accountable—from top administrators to line caseworkers—is a delicate and difficult challenge. If failure to perform has few consequences, performance will fail. But, Golden writes, a too-punitive approach is equally futile:

“In Utah and [D.C.], a punishment only approach had been tried, and it did not work. The more Utah’s administrators punished workers for failing to meet hundreds of standards . . . the more demoralization led to more failures . . . in some [complex] systems, the harder you push, the more resistance you get. . . . Through words and actions, leaders may be able to ease up on a specific measure, while pushing harder on the underlying goals—zigging or zagging in the short run to stay the long-run course.”

Golden emphasizes that senior reformers must involve themselves continually in information systems improvement. Executives need not be information technology experts, but if they are reformers, they must personally direct the information systems reforms. What is written down, how it is aggregated, and who has access to it is central to organizational authority and control. The reformer must gain authority to control the organization’s production and dissemination of knowledge and information.

At the same time, Golden writes, reform of information systems and practices will work only if it is “owned” by people at all levels of the organization:

“ . . . demands that social workers enter data, and threats to punish those who did not, had all failed, but new strategies eventually produced results … [The new strategies] included . . . a new change control committee . . . which gave social workers a voice in setting priorities for possible computer system changes. Social workers soon saw results for children that were traceable to better measurement and tracking, such as reductions in the number of young children in group homes. . . . the information systems team took pains to adapt reports to supervisors’ and workers’ needs … these strategies provided the incentive for better data entry, more trust in the system, and, in turn, [better] data quality.”

Because the role of litigation in child welfare reform is controversial, Golden’s conclusions on that subject are worthy of quotation in some detail. She found periods of “effective court-driven transformation” and periods of “dysfunction” under court supervision in each of the three jurisdictions. Comparing these periods, she:

“found five . . . lessons. First, the courts created moments of opportunity: deadlines that focused the attention of all participants and . . . shocked them out of their old, failed methods of interaction. . . . Second, the parties . . . bought into a vision of reform that was largely shared by the agency, the plaintiffs, and the court. . . . Third, the court had a relatively long-run time frame . . . [which was necessary]. Fourth, the court operated in good faith as an accountability holder: the standards it used . . . were clear . . . . Finally, the court had the power to influence or coerce other important stakeholders . . . so the agency was not completely at the mercy of forces pulling it in opposing directions.”

The chapter on leadership is remarkable for the degree to which it brings into the open the fact that, as Golden experienced it, serious conflict and stress are inevitable components of reform. A sample of her phrases illustrates:

“disasters . . . that I fear could wreck the agency . . . personal responsibility for failures . . . anger . . . internal squabbling, bitterness, power struggle . . . sabotages . . . bureaucratic battles . . . [s]truggles . . . painful . . . damaging battle . . . [c]ontentious issues . . . anger and tears . . . friction . . . individuals can experience great stress as they bear the brunt of others’ fury or personal attacks a new leader . . . fury . . . after years of betrayal, incompetence, and blame. . . . internal and external battles . . . . At the same time, turning around a troubled agency yields tremendous joy.” (Pp. 191-210.)

Golden’s realism about the bruising conflicts inherent in organizational change and their emotional toll is a welcome contrast to conventionally bland treatises on organizational leadership.

Reforming Child Welfare concludes with 22 pages of sound policy recommendations. Unhappily, the political will to implement them is nowhere in sight. The fact that virtually all reformist policies require money is only the most massive of the barriers. Another important obstacle is the complexity of the current hodge-podge of statutes, bureaucratic fiefdoms, and jurisdictional boundaries. This complexity limits discussion to experts and insiders, effectively excluding both the public and its elected representatives. Lawmakers have no choice but to rely on the recommendations of lobbyists and the demands of special interest groups.

Golden’s book is most welcome. For prospective reformers, it is a “must read.” For anyone who would hope to understand the problems of America’s child welfare establishment and the possible solutions to those problems, this slim volume will be a source of major enlightenment.

For an expanded version of this review, please contact eopton(at)youthlaw.org.