Judges Jeffrey Wolfe and Dale Glendening have made a major contribution to the debate over the Social Security disability decisionmaking process in their response to my Fall 2011 article “What Should We Do about Social Security Disability?” in which I urged major changes to that process. Their description of the history of the process and their detailed description of the manner in which it functions today will be valuable in identifying constructive changes in the process.

The judges and I agree on two important points:

  • The decisionmaking process has evolved in ways that have created an unsustainably generous disability program.
  • We need to engage in a “top-to-bottom” review of the process to identify needed changes.

The judges describe a decisionmaking process in which applicants for disability benefits have the opportunity to obtain a “yes” decision at any of five stages in the process: the initial decision by a state agency, reconsideration by a state agency, decision by an administrative law judge (ALJ), decision by a district court, and decision by a circuit court of appeals. Each stage in the process is characterized by two asymmetries that create systemic bias in favor of applicants:

  • The applicant can appeal a “no” decision to the next level, while the government cannot appeal a “yes” decision.
  • Some 80 percent of applicants are represented by attorneys, while the government is never represented by an attorney unless it goes to district court.

We differ with respect to our proposed ways of addressing the problem, however. I urge elimination of the ALJ stage of the decisionmaking process. In the alternative, I urge reinstitution of the ALJ quality-control programs the Social Security Administration implemented in the 1970s.

Important “coercion” | The judges acknowledge that elimination of the ALJ stage of the decisionmaking process would “solve” the problem, in the sense of reducing significantly the number of decisions that grant benefits. But they argue that the cost of such a change would include an undue sacrifice of our values. They object to my proposal to reinstate the ALJ quality-control programs on the basis that those programs interfere with the independence of ALJs. They ask, “Does any reasonable, fair-minded person sincerely believe that potentially subjecting every ALJ to discipline or removal for decisions for which there is not an ‘adequate explanation’ is not coercion per se?”

I am one of the many “reasonable, fair-minded” people who believe that the kind of “coercion” created by the ALJ quality-control programs of the 1970s is essential to further our values. Decisionmaking by SSA ALJs is a classic illustration of a recurring problem. We sometimes encounter conflicts between two goals of due process. The first goal is to provide every litigant an opportunity to obtain a decision from an independent judicial mind. The second goal is to construct a decisionmaking process in which like cases are resolved in like manner. When we have encountered this conflict in the past, we have often chosen to allow the pursuit of the second goal to trump pursuit of the first.

Thus, for instance, in 1984 Congress decided that there was an intolerable degree of variation among federal judges in the sentences they were giving to defendants who were convicted of similar crimes in similar circumstances. Congress responded by creating the Sentencing Commission, an agency that is empowered to issue rules that bind judges in the sentencing context. Several district courts and circuit courts held that the commission was unconstitutional because it interfered with the independent decisionmaking power of federal judges. The Supreme Court upheld the statute, however, on the basis that the statute’s interference with the independent decisionmaking power of federal judges was justified by the need to assure that like cases are resolved in like manner.

A second example is familiar to all law professors. Over the last 50 years, almost every law school has adopted a binding grading curve applicable to first-year classes. Before schools took this step, it was common for first-year students to be assigned randomly to one of two sections of a required course, where one professor regularly gave 70 percent “A” grades while the other regularly gave 30 percent “A” grades. I have taught at several schools at the time they made this change. There were always professors who objected on the basis that a binding, institutionally prescribed grading curve interfered with their academic freedom to base grades on an independent evaluation of each student’s performance. In every case, however, the argument that carried the day was the argument that it was fundamentally unfair to students to subject them to “independent” decisionmakers who varied significantly in their patterns of decisions. I am delighted to be “coerced” to adopt grading norms that coincide with those of my colleagues, knowing that the coercion is essential to creating an environment that is fair to students.

The enormous variation in the patterns of decisions of SSA ALJs makes this another context that is an ideal candidate for allowing the goal of obtaining like decisions in like cases to trump the goal of providing every litigant with access to an independent judicial mind.

Other reform ideas | I am open to alternative means of addressing what the judges and I agree are serious problems in the Social Security decisionmaking process, but the changes they urge would not even move us in the right direction. First, they urge that we eliminate the obvious imbalance between the lawyer-represented applicant and the unrepresented government by assigning an attorney to represent the SSA in each case that is heard by an ALJ. As I pointed out in my article, this “solution” has a fatal cost: it would increase significantly the time required to hear each case. In order to avoid adding to the already long delay to get a hearing, the SSA would have to add many ALJs and many staff attorneys. Thus, for instance, if the conversion of ALJ hearings from ex parte proceedings to full adversarial proceedings tripled the average time required for each hearing, the SSA would need to hire an additional 3,000 ALJs and 4,500 staff attorneys to avoid adding more delay to the decisionmaking process. I do not believe that such a massive increase in staffing could be justified by the potential beneficial effects of such a change or that Congress could be persuaded to increase appropriations to the SSA to the extent needed to implement such a change.

The first goal is to provide every litigant an opportunity to obtain a decision from an indepen- dent judicial mind. The second goal is to construct a decisionmaking process in which like cases are resolved in like manner.

Second, Wolfe and Glendening urge an expanded role for ALJs, creating “an independent corps of ALJs” who would “review administrative decisionmaking in the executive branch.” In addition to the well-documented policy-based disadvantages of such a system, it would violate both the Appointments Clause and the Take Care Clause of the U.S. Constitution. Any ALJ who has the power to make final decisions is an “Officer of the United States” who can only be appointed through the process of nomination by the president and confirmation by the Senate, as the D.C. Circuit concluded in its 2000 opinions in Landry v. FDIC. Moreover, no officer of the United States can be insulated from presidential control by two or more layers of for-cause limits on the president’s removal power, as the Supreme Court held in its 2010 opinion in Free Enterprise Fund v. Public Company Accounting Oversight Board. Implementation of the judges’ proposal to create an independent ALJ corps would violate both of those constitutional commands.