in the Social Security Administration
For two and a half years in the 1970s, I worked for the Social Security Administration, taking claims for benefits and doing the paperwork to pass the claims on to the next agency. A decade later, PhD'd but jobless, I worked for the state agency in Massachusetts that evaluates disability claims for SSA. That lasted for eight months until I joined the post-doc crowd and left the bureaucracy behind.
During my post-doc in psychology and law, I began to analyze the disability evaluation process from a different perspective. I read a lot , discovering a huge literature on disability evaluation, frequently critical of the supposed objectivity of it all. I did some informal interviews. I wrote a long paper, focusing on the evaluation of mental disability, different versions of which I presented at a couple of conferences. But the paper didn't seem to fit any journal's scope, and it remained (and remains) unpublished.
This somewhat shorter version did fit into the Journal of Social Behavior and Personality, in a special issue on disability in 1994. Unfortunately, the article's focus was disability in general rather than mental disability. The longer paper, updated in '94, is available as Governmental Ideology and Bureaucratic Reality in Mental Disability Evaluation.
Note: This version may not exactly match the published version!
Flawed assumptions underlying the Social Security Administration's (SSA's) disability adjudication process conflict with the reality of bureaucratic decision making. SSA insists that objective decision making can be ensured by a variety of managerial procedures. However, the notion that disability can be objectively determined is questionable. Furthermore, even if disability were objectively determinable in theory, bureaucratic reality makes SSA unlikely to attain such objectivity in practice. Consequently, people with disabilities would be served best not by merely reforming SSA's disability adjudication procedures (as SSA now proposes) but by fundamentally rethinking the inadequate linkage between the disability system, other social welfare systems, and the nature of work in modern society.
Americans with disabilities that prevent them from working frequently find themselves at an office of the Social Security Administration (SSA) seeking benefits under the Social Security Disability Insurance and Supplemental Security Income programs. Unfortunately, SSA's decision making process often leads to negative decisions or unacceptable delays for people who cannot readily prove to the agency's satisfaction that they are unable to work. Although SSA has made repeated efforts to improve its evaluation of disability, the subjective nature of disability and the vagaries of bureaucratic functioning make it impossible to resolve this problem satisfactorily within the narrow confines of SSA's traditional approach.
I base these observations about the disability evaluation process partly on my own work experiences. In the mid-1970s, I worked for two and a half years as a claims representative in Social Security District Offices, interviewing claimants for disability and other benefits. A decade later I worked for eight months as an examiner in a state agency making disability decisions for SSA. In both positions I observed from the front lines how the agency carried out Congressional and judicial directives. I later interviewed SSA examiners, doctors, psychologists, psychiatrists, managers, policy makers, and others in several states and at SSA headquarters in Baltimore as well as attorneys and paralegals representing individuals seeking benefits.
With these experiences behind me, I was not surprised to learn that Franz Kafka discovered the nature of bureaucracy while working in a disability benefits agency (Mashaw, 1983). In my view, even SSA's most recent tentative proposals to make disability adjudication less Kafkaesque (Process Reengineering Program, 1994) are likely to fail because the inherent limitations of any disability system ensure that many individuals who cannot work will remain unable to demonstrate that inability. Meeting their needs requires wide-ranging changes in the disability system, in other social welfare systems, and in the nature of work in modern society.
In theory, a well-functioning government bureaucracy consists of an efficient hierarchy ensuring a high degree of internal control. Employees at all levels and at all locations enforce legally mandated regulations in a consistent manner. There is little room in this ideal world for discretionary action. Individual workers apply standardized rules so that the agency's clients are treated exactly the same regardless of which particular worker they encounter.
Outside observers, and bureaucrats themselves, recognize that this idealized view bears little resemblance to reality (Hummel, 1987). Individuals working in any agency differ in their abilities, sympathies, politics, career goals, and other traits. To reduce the internal inconsistency resulting from these differences and to decrease discretionary action, managers give detailed instructions, improve training, increase quality control, and so on. Despite these efforts, however, even under the best of circumstances the tension between bureaucratic imperatives and individual inclinations remains. Under circumstances that are less than best--characterized by factors such as ambiguity, role conflict, power imbalances, political cross-pressures, and turf battles--the tension causes serious difficulties for many who come under the bureaucracy's jurisdiction either as clients or as employees.
Disability claims can be adjudicated in a variety of ways. How will disability be defined? Will the agency recognize partial disability or only total disability? Temporary disability or only permanent disability? Will claims be allowed for alcoholism and drug addiction? For mental disability? For physical disability caused by risky behavior? What kind of evidence will be required? Should the state of the job market be relevant? Most fundamentally: What does it mean to be disabled? Political decision makers generally answer these questions.
It is the Social Security Administration's mandate to implement the politically determined answers. Among the many possible alternatives, SSA has established its own approach. SSA's most important assumption is that disability is caused by a medical condition that can be established by objective evidence. In SSA, a state Disability Determination Service (DDS) makes that determination. Although federal law binds every DDS, DDS workers are not federal employees but state employees subject to their own state's political culture. This accounts for much of the state-by-state disparity in procedures, examiner qualifications, and allowance rates.
The actual disability decision comes in two parts, reflecting SSA's definition of disability as:
the inability to engage in any substantial gainful activity by reason of any medically determinable physical or mental impairment which can be expected to result in death or which has lasted or can be expected to last for a continuous period of not less than 12 months. (Social Security Act, 1992)
First, a DDS medical doctor examines the evidence in the file and rates the degree to which an individual is "impaired." The doctor bases this rating on "objective" physical signs and laboratory tests (e.g., blood tests, x-rays) rather than on "subjective" symptoms (e.g., pain, fatigue). Second, a vocational-disability examiner applies to the claimant's rated impairment the "vocational factors" of age, education, and work history and officially declares the claimant to be "disabled" or "not disabled." Neither the doctor nor the examiner examines the claimant.
Critics question both the assumptions behind this approach and the actual methods used. Even experts within SSA often disagree about the nature of an impairment and the extent to which an impairment renders a particular individual incapable of working. As a consequence, SSA acknowledges the necessity of "professional judgment." (Ironically, however, the agency considers certain categories of impairment automatically disabling even if the individual can work, such as in the case of blindness.) The inadequacy of the process is demonstrated by the fact that more than 60 percent of claimants denied by DDSs successfully appeal (see generally Cofer, 1985, 1987; Diehl, 1990; Dixon, 1973; Liebman, 1976, 1985; Mashaw, 1983; Mezey, 1988; Mills, 1993; Process Reengineering Program, 1994; Stone, 1984; Taibi, 1990; Weaver, 1989).
Decision making problems are compounded when SSA applies the assumptions and methods developed for claims of physical disability to claims of mental disability (Fox, 1994). Clinical experts routinely disagree about diagnosis, treatment, prognosis, and, most important, work ability. In 1938 the Social Security Board originally excluded mental impairment from the proposed disability program "because it would be too hard to determine" (Stone, 1984, p. 72). But mental disability claims became part of SSA anyway, to be determined by the usual objective evidence.
In the mid-1980s, after outside criticism, SSA drastically revised its procedures. It increased the focus on work-related functional limitations and reduced its reliance on one-shot evaluation of symptoms (Collins & Erfle, 1985; General Accounting Office, 1986; Mezey, 1988; Novack, 1987; Rubenstein, 1985). Unfortunately, the new rules did not resolve the underlying problems. From my position as a vocational-disability examiner in a large urban Disability Determination Service, it was clear that the new rules were vague, the new training inadequate, and the new workloads unmanageable. Examiners, agency doctors, supervisors, and managers reacted in a variety of ways. The fate of an individual's claim increasingly depended on the examiner or doctor assigned to the case. Some evaluators, for example, willingly allowed mental disability benefits even if the impairment could not be "objectively determined" because of a belief that "this person will never work" and that it didn't really matter if support came from SSA or from unemployment or welfare, so long as it came from somewhere. On the other hand, some workers denied claims for those they saw as having "caused their own problems" when additional evidence might have resulted in allowances.
Discussions with DDS and SSA workers in several parts of the country revealed that these problems were widespread, a situation confirmed by the General Accounting Office (1986). SSA itself noted that widely disparate allowance rates and the high proportion of incorrect denials quickly escalated (Office of Assessment, 1987). Despite years of effort, SSA has not fully recovered even today. Long delays in disability evaluation continue for mental as well as for physical disability (Delfico, 1992; Process Reengineering Program, 1994).
Not surprisingly, SSA's difficulties have led to a great deal of outside interest. SSA is under constant judicial pressure to liberalize standards and procedures according to legal concepts rather than medical or bureaucratic ones. The agency traditionally resists such efforts, preferring instead its own interpretations of legislative mandates.
The level of the administrative law judge (ALJ) hearing has drawn much attention (Cofer, 1985, 1987; Mashaw, et al., 1978; Parmele, 1987; Process Reengineering Program, 1994). A central focus has been the appropriate degree of independence of ALJs, who work within SSA's Office of Hearings and Appeals but see themselves as independent. There is a common view within SSA that the ALJs simply "don't follow the rules," that ALJs in particular allow cases without having to justify their decisions. One manager told me that the high reversal rate was a "shameful statistic," and that ALJs, who are "incredibly poorly trained in SSA regulations and medicine," incorrectly "see themselves as members of the judiciary rather than as the administrative personnel they really are."
In contrast to this common SSA view, Cofer (1985) pointed out that ALJs base their decisions on Congressional statutes and judicial precedent rather than on SSA's internal instructions, despite SSA's insistence that its own instruction manual presents the correct interpretation. Cofer uncovered several important differences between statutes and SSA guidelines that add to the high reversal rate of DDS denials. Although many ALJs regard the quality of DDS case development as poor, "if the states and ALJs continue to use essentially different standards of judgment, all of the care given to case development will be for naught" (Cofer, 1985, p. 184). A 1983 lawsuit filed by the Association of ALJs reflected this continuing conflict. The ALJs claimed that SSA "had directed administrative law judges to ignore court decisions favorable to claimants and had put pressure on them to deny claims" (Liebman, 1985, p. 38).
There are other reasons for the disparity between DDS and ALJ decision making. Perhaps most importantly, Stone (1984) noted, the ALJ sees and talks to the claimant in person, unlike DDS personnel. This allows a focus on the individual's subjective experience rather than on the "average person" approach of the formal rules. In addition, Stone pointed out, legal and medical notions of causation and proof differ. To oversimplify, the doctor/scientist may allow a claim only if 95 percent "medically certain" that the underlying impairment fully causes the inability to work; the lawyer/judge may award benefits if a preponderance, or 50.1 percent, of the evidence shows that the impairment is a contributing cause. Finally, the courts are more likely to interpret the disability program as a contract between the insured and the insurer "whose provisions are somewhat fuzzy" (Stone, 1984, p. 159) and, thus, to be interpreted more liberally than the stringent official definition.
Although SSA typically regards the Administrative Law Judges as too lenient, when ALJs do affirm denials of disability claims, federal judges often overrule them. Ironically, judges frequently consider ALJs too tightly linked to SSA. The resulting incentive to go to court has led to frequent lawsuits as well as to proposals for specialized Social Security courts (Lauter, 1987).
SSA's usual response to case inconsistency and delay, employee variability, state discrepancies, and high reversal rates is to revise its methods. In fact, the Administration is now considering a new set of wide-ranging reforms designed to "fundamentally rethink and radically redesign SSA's work processes to achieve dramatic improvements in critical measures of performance" (Process Reengineering Program, 1994, p. 18,188). For example, SSA's Reengineering Team has proposed personal interviews between DDS evaluators and claimants, a policy long advocated by outside critics as well as by the General Accounting Office (Delfico, 1991). These kinds of reforms, however, rarely resolve the difficulties faced by clients with disabilities. Such resolution requires more widespread changes.
SSA policy makers generally assume that the failure to improve the evaluation process "constitutes a problem of implementation, not problems in the concept of impairment itself. Hence, the SSA responds to criticisms of its determination process . . . by providing more training, more guidelines, and more supervision" (Stone, 1984, p. 127). Although attempts to enhance bureaucratic control are occasionally useful, often they are counterproductive. As Mashaw (1983) noted, "a demand for objectivity, without changes in the resources devoted to processing, tends to produce stringency" (p. 176) and a reduced allowance rate.
Adherence to objective scientific medical diagnosis fails on at least two counts. First, "medical criteria cannot be more objective and scientific than the body of knowledge and practice on which they rely" (Mashaw, 1983, p. 110). Many facts will always remain unknown, for reasons ranging from fragmentary medical records and individual differences in interpreting X-ray reports to bureaucratic preferences for certain kinds of information (such as the discounting of evidence submitted by chiropractors, social workers, and other "nonmedical" sources). In the case of mental impairments, "misdiagnosis is common" (Rubenstein, 1985, p. 720) and can affect how the agency evaluates the claim.
Second, and of crucial importance, there are large individual differences in people's ability to function given the same degree of objective impairment.
That the clinical concept should fail as a restrictive mechanism is not surprising, because in an important way it sidesteps the key issue of disability: what is it that prevents people from working? In most instances, the cause is not some identifiable physical phenomenon but a complex set of interacting factors involving individual and family history, the state of the economy, and cultural and psychological as well as biological factors. . . . Pain, fatigue, shortness of breath--and, one might add, anxiety--are all real and very powerful subjective phenomena, but they defy measurement. (Stone, 1984, p. 134)
One frequently suggested institutional reform is to abandon the joint SSA-DDS system in favor of a centralized system directly run by SSA. Either federalization or another possibility, contracting out, would give SSA greater control "and, in theory, offer greater possibilities of increased uniformity" (GAO, 1985, p. 3). The General Accounting Office concluded, however, that abandoning the present system would meet much opposition and would probably produce neither better decisions nor financial savings. Of course, if federalization did result in better decisions according to federal standards, the cost would be a lower allowance rate, since SSA frequently defines "better" as denying rather than allowing borderline claims.
Tinkering with the current system, while often well intentioned and sometimes useful in the short term, is unlikely to go far enough. As with many kinds of social problems, the difficulties faced by people with disabilities are not confined to a single agency. Effective solutions must be systemic and comprehensive rather than narrowly focused (Fox, 1991). Unfortunately, even when proposing to rethink the disability process completely ("In other words 'how would we design a process if we were starting over?'"--Process Reengineering Program, 1994, p. 18,188), SSA restricts its vision: "The scope of the assignment to the disability reengineering project team did not include making any changes to the statutory definition of disability" (Process Reengineering Program, 1994, p. 18,188).
Much more must be done. As a start, SSA should abandon its underlying assumption that it can define and determine disability while ignoring the job market. Instead, disability evaluation should take into account the likelihood that an impaired person will find a job when competing with unimpaired individuals for scarce openings.
Such a change would not be unprecedented. Stone (1984) detailed the close historical connection between the labor market and the legal definition of disability. The 14th Century English vagrancy laws, for example, were a response to the labor shortage caused by the Black Plague. To prevent workers from traveling in search of higher wages, the authorities made it illegal to give money to beggars who could work; only people legally recognized as disabled could leave town. In modern times, with labor surplus rather than labor shortage, the trend in Western Europe "has been to extend eligibility under programs traditionally designed for the retired or the disabled as a way to alleviate unemployment" (Reno & Price, 1985, p. 24).
Even under Social Security, despite the official policy, the government occasionally has altered the criteria to respond to a changing labor market. Thus, the institution of early retirement benefits in 1961, during an economic recession, "was seen as a way to reduce unemployment among older workers. The change was in part a response to long-term technological unemployment, recognizing that persons who lost their jobs at older ages might never find other work" (Reno & Price, 1985, p. 30; see also Taibi, 1990). Similarly, changes in SSA's vocational guidelines in 1979 significantly eased the disability criteria for unskilled workers over age 50, making it possible for older workers to drop out of the overcrowded work force (Stone, 1984). Hence, although formally linking the disability definition to the job market would be a major change, it should no longer be beyond discussion.
Merely broadening the definition of disability, which necessarily is arbitrary to begin with, can be done within the existing SSA framework. A more far-reaching approach would be to rethink the rationale for maintaining separate disability, unemployment, welfare, and tort systems. The distinctions among the systems are increasingly artificial and wasteful, with "gaps and overlaps in the existing network of compensation institutions" (Abraham & Liebman, 1993, p. 77). Now that the Clinton Administration has proposed broad changes in the national health care and welfare systems, policy makers might pay more attention to the links between the systems.
Further steps would be even more beneficial. Because "virtually everyone is capable of doing something productive" (Taibi, 1990, p. 919), the institution of a shorter work week and the creation of meaningful part-time work at higher wages and benefits would allow many people now defined as disabled to work in their communities. People with disabilities are especially disadvantaged in a society that has more people than jobs and where it is problems of distribution rather than of production that are most severe.
In the long run, a transition to more cooperative workplaces, more liveable cities, and an economic system not requiring destructive competition would bring further benefits. The degree to which centralized bureaucratic reality is inherently at odds with meeting human needs requires us at least to consider solutions that ultimately go beyond minor reforms (Fox, 1985, 1993; Wineman, 1984). If we do, we will all be better off.
References combined with those from a longer paper.
some political, most not
Page updated August 5, 2009