© The Author 2014. Oxford University Press and New York University School of Law.
All rights reserved. For permissions, please e-mail: [email protected]
Two ironies about American
exceptionalism over social rights
Jeff King*
It is often assumed that the United States of America is exceptional in according poor
legal status to socio-economic rights, and that its judges show a distinct reticence towards
enforcing positive entitlements. This article challenges this view by identifying two ironies
in the American brand of exceptionalism about social rights. It argues that, notwithstanding America’s exceptionally low public commitment to social spending, its judges have often
enforced positive duties to provide social services in a manner that is in fact strikingly interventionist by international comparison. It also shows how this more robust involvement in
welfare rights adjudication has often not produced the results desired by advocates of social
rights, and may even in some cases have been regressive. The article explores the causes and
consequences of the experience, and suggests what lessons it may hold out for other countries.
1. Introduction
This article makes three basic points, one of which is familiar, the other two both
surprising and ironic. The familiar one is that, among countries with similarly high
levels of economic development,1 the United States is exceptional for demonstrating a low public commitment to securing social welfare and eliminating economic
inequality (Section 2). The first surprising point is that despite this, American judges
often enforce obligations to provide social welfare services in a manner that is staggeringly interventionist by comparison with most other wealthy states (Section 3).
The second is that although these strong remedies are effective for claimants, the
impact on bureaucracies more generally has drawn highly ambivalent reactions
among informed commentators who appear generally sympathetic to the social need
for a robust welfare state (Section 4). These are the two ironies about the American
* Senior Lecturer, Faculty of Laws, University College London, and Attorney, Bar of New York. Email:
[email protected]. The author would like to thank Caroline Daly and Alma Mozetic for outstanding
research assistance, and Murray Wesson and an anonymous reviewer for helpful advice and criticism.
1
Throughout this article, reference will be made to “comparators” or “natural comparators.” By these
I mean those countries having similarly high levels of economic wealth, and long-standing democratic
institutions with mature welfare states. If pressed for a precise definition, I would put the “EU 15” countries belonging to the European Union prior to the expansion in 2002, as well as Australia, Canada, and
New Zealand into this category.
I•CON (2014), Vol. 12 No. 3, 572–602doi:10.1093/icon/mou049
Two ironies about American exceptionalism over social rights
573
exceptionalism over social rights. American judges have been more assertive than
many suppose, and delivered less than most dared to hope for, and may even have
inadvertently caused regression in some instances. I consider some explanations for
these two ironies and reflect on what lessons the American experience holds out for
other countries (Sections 5 and 6). I suggest there that the two ironies may well be
explained by America’s tendency to treat rights as trumps that are resistant to limitations, the demand-pressures for legal relief generated by America’s also exceptional
chronic underfunding of public services, as well as by the radical decentralization of
some social services. However much they may go towards justifying and explaining
it, these demand-led pressures for legal accountability cannot change the underlying reality of how difficult it is for judges to manage broad-level social welfare policy
under a hostile political climate.
2. American exceptionalism about social welfare
The United States is exceptional in its low public commitment to securing adequate welfare provision to its citizens. This widely believed claim is best understood by comparing
figures on public social spending (Figure 1) and relative poverty (Figure 2). The degree
of public social spending no doubt has some relation to relative poverty rates, which are
defined by the Organisation for Economic Co-operation and Development (OECD) as the
share of individuals with disposable income less than 50 percent of the median income
of the whole population. On both measures, America stands out as exceptional. More
surprising perhaps are differences in social attitudes towards poverty. In his study on
Figure 1. Public social spending as percentage of GDP 1980–2007.
Source: OECD, Social expenditure: Aggregated Data (OECD Social Expenditure Statistics Database)
(2014), http://www.oecd.org/statistics/. 574
I•CON 12 (2014), 572–602
Figure 2. Relative Poverty Rates in the Mid-2000s
Source: OECD, Poverty rate after taxes and transfers, poverty line 50% (OECD Income Distribution
and Poverty Database) (2014), http://www.oecd.org/statistics/. (Note: All figures 2004, except
Denmark (2005).)
inequality, John Hills, reports a survey that asked respondents to indicate why people live
in need, by choosing between “laziness and lack of willpower” or because “society treats
them unfairly.”2 In the US, 61 percent of respondents chose the former, whereas a similar
answer was produced, within fifteen comparable countries in Europe, in numbers that
averaged at 18 percent of respondents. This fact is further illuminated by studies concerning anti-poverty rhetoric in America, where comments about “welfare queens,” often had
racist overtones towards African Americans and where both racial minorities and singlemothers have often been stigmatized as part of the “undeserving” or “unworthy” poor.3
America’s exceptionalism in this regard is not simply the product of malign conservative and corporate forces. Many have bona fide commitments to republican and
libertarian values. The Constitution accords limited power to the federal government
under the Tenth Amendment, and the Lochner era was not only a result of doctrinal freewheeling and reactionary judicial politics.4 Some aspects of the American
approach to welfare also reflect some distinctly American beliefs about equality of
opportunity. Protections offered in the area of equality of access to education, educational performance,5 protection of children, and disability rights for students and
John Hills, Inequality and the State 69 (2004).
See Joel F. Handler, The Transformation of Aid to Families with Dependent Children: The Family Support Act in
Historical Context, 16 N.Y.U. Rev. L. & Soc. Change 457, 459 (1987); Gwendolyn Mink, Welfare Reform in
Historical Perspective, 26 Conn. L. Rev. 879, 880 (1994); See also Elizabeth Bussiere, (Dis)Entitling the Poor: The
Warren Court, Welfare Rights, and the American Political Tradition 73–76, 83 (1997); Bertrall L. Ross II & Terry
Smith, Minimum Responsiveness and the Political Exclusion of the Poor, 72 Law & Contemp. Prob. 207 (2009).
4
The conventional account of the New Deal period is discussed and criticized in G. Edward White, The Constitution
and the New Deal esp. chs. 1 and 4 (2000). See also James O. Freedman, Crisis and Legitimacy: The Administrative
Process and American Government (1978). See also Martha Derthick, Agency Under Stress 19 (1990).
5
OECD, PISA 2009 Results: Executive Summary 8 (2010) (showing US average 15-year-old student per­
formance in mathematics, reading, and sciences to not be statistically different than the OECD average,
and performing ahead, notably, of Germany, France, and Sweden).
2
3
Two ironies about American exceptionalism over social rights
575
workers appear to have been reasonably robust (and extraordinarily well-protected by
judges), as the discussion in Section 3 below will reveal. No country spends more on
health care, and university participation is comparatively quite high despite high fees
in elite institutions.6
Even so, these caveats do not undo the basic point that the US compares poorly
internationally in terms of public support for the less well off. Even the Aid for
Families with Dependent Children (AFDC) program (1935–1996), which was itself
eviscerated by welfare reform in 1996,7 was described by R. Shep Melnick as “ungenerous by international standards and uneven in [its] coverage across the states and
population groups at risk.”8 Disability rights, even, do not always compare well,9
and the absence of universal health care coverage is a problem that continues even
after the Patient Protection and Affordable Care Act’s passage in 2010. Worst of
all for the American dream, social mobility (measured by the correlation between
father and son’s income) is by far the lowest among eight comparable countries,
including the UK, Germany, and Canada, and social immobility has more than tripled since the 1980s, the time of America’s pronounced libertarian turn.10 So even
with the caveats added, America remains exceptional in its (dis)regard for protecting social rights.
3. The first irony: exceptional judicial enforcement of social
welfare rights
Perhaps incredibly, one can find many examples of US courts intruding more deeply,
and more often, into welfare bureaucracies than do the courts of perhaps any other
comparable country. This is evident upon examination of the situation under the federal constitution, state constitutions, and federal statutory law enforced nationwide in
the federal and state courts.
The data are somewhat complex on this point. Compare OECD, Education at a Glance 2011: OECD
Indicators (2011), 306 (Chart C2.1, entry rates), with 60 (Chart A3.1, completion rates), as well as 65
(Chart A3.4, adjusting for foreign student participation). On all measures, the US compares reasonably
well if not favorably with its comparators.
7
Title IV, Social Security Act 1935 (42 U.S.C. § 601 et seq. (1935–1996)) [hereinafter AFDC], repealed by
the Personal Responsibility and Work Opportunity Reconciliation Act of 1996 [hereinafter PRWORA],
Pub. L. 104–193, 110 Stat. 2105, which among other things instituted Temporary Assistance for Needy
Families 42 U.S.C. § 601 et seq. [hereinafter TANF]. See further infra notes 45, 87, and text accompanying
note 97.
8
R. Shep Melnick, Between the Lines: Interpreting Welfare Rights 65 (1994) (confirming that AFDC conforms
to this general assessment of American public assistance programs, which he quotes from Margaret
Weir, Ann S. Orlaff & Theda Skocpol, The Politics of Social Policy in the United States 4 (1988).
9
Cf., however, Jennifer L. Erkulwater, Disability Rights and the American Social Safety Net 235–239 (2006)
(comparing US disability rights protection in part unfavorably with Swedish and German protection, due
importantly to the absence of a more comprehensive welfare state as a support structure).
10
Richard Wilkinson & Kate Pickett, The Spirit Level: Why Equality is Better for Everyone 160 (2010). On the
rise of libertarian thinking, see The New Conservatism: The Rise of the Right in America (Michael J. Thompson
ed., 2007).
6
576
I•CON 12 (2014), 572–602
3.1. The US Constitution
There is a standard story about the rise and fall of a welfare rights vision of the US
Constitution. From judicial obstructionism during Lochner Era, to a progressive brand
of deference afterwards,11 the Warren Court (1953–1969) was established and
it became famous around the world for its protection of marginalized groups in a
range of famous cases.12 Throughout the 1960s, poverty law courses grew exponentially in US law schools, and leading constitutionalists like Charles Reich and Frank
Michelman developed theoretically sophisticated arguments for why the Fourteenth
Amendment was best read as protecting welfare rights.13 In landmark Supreme Court
cases, claimants argued that wealth or poverty was a suspect ground of discrimination under the equal protection clause;14 that certain welfare interests should be recognized as implied rights under the (substantive) due process clause, the interference
with or denial of which attracts stricter judicial scrutiny; 15 and that once welfare or
other benefits were granted, they became akin to a “new property,” the public deprivation of which must respect procedural due process. This view was accepted by the
Supreme Court in respect of welfare benefits determinations in the judicial high water
mark for this brief period of hope. 16
On the conventional view, the high hopes were dashed by the shift from the Warren
Court to the Burger Court in 1969. In Dandridge v. Williams (1970),17 the Supreme
Court announced that “the intractable economic, social, and even philosophical
problems presented by public welfare assistance programs are not the business of
this Court.”18 In San Antonio School District v. Rodriguez (1973),19 the Court held that
“[e]ducation, of course, is not among the rights afforded explicit protection under
our Federal Constitution. . . . It is not the province of this Court to create substantive constitutional rights in the name of guaranteeing equal protection of the laws.”20
It was in DeShaney v. Winnebago County (1989),21 a case where social services failed
to protect a child at obvious lethal risk, where the Court put the point at its starkest:
The [Due Process] Clause is phrased as a limitation on the State’s power to act, not as a guarantee of certain minimal levels of safety and security . . . [and] . . . its language cannot fairly be
Bruce Ackerman, We the People: Foundations (1993); Cass R. Sunstein, Lochner’s Legacy, 7 Columbia L. Rev.
873 (1987).
12
See, e.g., Brown v. Board of Education, 347 U.S. 483 (1954); Baker v. Carr, 369 U.S. 186 (1962); Miranda
v. Arizona, 384 U.S. 436 (1966).
13
Charles A. Reich, The New Property, 73 Yale L.J. 733 (1964); Frank I. Michelman, The Supreme Court,
1968 Term—Foreword: On Protecting the Poor Through the Fourteenth Amendment, 83 Harv. L. Rev. 7
(1969).
14
Shapiro v. Thompson, 394 U.S. 618 (1969).
15
Lindsey v. Normet, 405 U.S. 56 (1971).
16
Goldberg v. Kelly, 397 U.S. 254 (1970).
17
Dandridge v. Williams, 397 U.S. 471 (1970). Though Goldberg and Dandridge are sometimes contrasted as
the high and low of the period, Dandridge was handed down less than three weeks after Goldberg.
18
Id. at 487.
19
San Antonio School District v. Rodriguez, 411 U.S. 1 (1973).
20
Id. at 33, 35.
21
DeShaney v. Winnebago County, 489 U.S. 189 (1989).
11
Two ironies about American exceptionalism over social rights
577
extended to impose an affirmative obligation on the State to ensure that those interests do not
come to harm through other means.22 Stephen Gardbaum has argued convincingly that claims about American exceptionalism about constitutional rights litigation are often exaggerated, including those
concerning social rights.23 Even so, it is right to acknowledge that there are subtle ways
in which US courts are more hostile to social rights claims than are their foreign cousins. One difference is that non-American apex courts often reject watertight distinctions between socio-economic and civil and political rights, leave open the question of
whether social rights could be justiciable nationally, and are less wary of extending
positive human rights obligations into general social services duties.24 In its Z v. United
Kingdom (2001) decision, for instance, the European Court of Human Rights (ECtHR)
held, contrary to the approach of the US Supreme Court in in De Shaney, that a failure
by social services to protect children could amount to “cruel, inhuman and degrading treatment,” and that it did in that case.25 These subtly different judicial attitudes,
which have allowed piecemeal inflation of social rights friendly readings of classical
civil rights, are part of an expanding European and Commonwealth political accept­
ance of social rights as human rights, and potentially constitutional rights if the right
political steps are taken. The fruit of that tree is found in the adoption of social rights
in the EU Charter of Fundamental Rights, the evolution of social rights under the
German Constitutional Court’s jurisprudence, and even serious parliamentary consideration of a bill of social rights in the United Kingdom.26 Even so, and in fairness to
Gardbaum, it would be quixotic to consider a door left open to be a truly remarkable
difference at this point. This is all the more so when one contemplates the other side of
the federal constitutional story in the United States.
That other side is that there are at least four areas in which the US courts have
given considerably more protection under the US Constitution to welfare interests
than have the courts in any other comparable country. The first is in protecting substantive equality in education. In Brown v. Board of Education (1954),27 the Court ruled
unanimously that racially segregated educational facilities are inherently unequal.
Thereafter, the federal courts went to astonishing lengths in order to make desegregation orders and superintend their implementation. Perhaps the most (in)famous
Id. at 195.
Stephen Gardbaum, The Myth and Reality of American Constitutional Exceptionalism, 107 Mich. L. Rev.
391, 446–453 (2008).
24
Airey v. Ireland, Eur. Ct. H. R., 2 EHRR 305, ¶ 22 (1979–80) (“there is no watertight division separating [social and economic rights] from the field covered by the Convention”); see further A.R. Mowbray,
The Development of Positive Obligations under the European Convention on Human Rights by the European
Court of Human Rights (2004). See also Irwin Toy Ltd. v. Quebec [1989] 1 S.C.R. 927, 2003–4 (Can.);
R v. SSHD, ex parte Limbuela [2005] UKHL 66, ¶ 920 (Lord Brown). See also Bundesverfassungsgericht
(BVerfG) (Federal Constitutional Court), Feb. 9, 2010, BVerfGE, 1 BvL 1/09 (“the Hartz IV Decision”)
(Ger.) (Existenzminimum) and BVerfG, July 18, 2012, 1 BvL 10/10, 1 BvL 2/11 (“Asylum Seekers Benefits
Act Case”) (Ger.) (Existenzminimum).
25
Z v. United Kingdom, 34 Eur. Ct. H.R. 97 (2001).
26
For background, see Jeff King, Judging Social Rights ch. 1 and 279–281 (2012).
27
Brown v. Board of Education, 347 U.S. 483 (1954).
22
23
578
I•CON 12 (2014), 572–602
was Swann v. Charlotte-Mecklenburg Board of Education (1971), which found that there
would be “a presumption against schools that are substantially disproportionate in
their racial composition” and an affirmative state obligation to “eliminate from the
public schools all vestiges of state-imposed segregation.”28 The remedial dimension
became even more famous: that busing students to schools outside their residential
zone was an appropriate desegregation remedy. Swann had a very considerable impact
in other cases. In Adams v. Richardson (1972), for example, a DC District Court accepted
that the National Association for the Advancement of Colored People’s (NAACP) Legal
Defense Fund’s class action case represented “students, citizens, and taxpayers” of
seventeen different states.29 Having shown that Swann’s presumption had not been
rebutted, Judge Pratt ordered the US Department of Health, Education and Welfare
to commence the termination of funding to 127 school districts where the proportionality principle set out in Swann had been violated. The Department thereafter
commenced busing plans in these cities.30 In perhaps the most remarkable decision,
Missouri v. Jenkins (1995), the Supreme Court approved the judicial imposition (by a
lower court judge) of tax increases to support over $2 billion in school improvements
that were required to desegregate Kansas City schools.31 This remedy cuts further
across contemporary notions of the separation of powers than any I have yet come
across in another social rights case. Furthermore, key educational equity cases were
not all about desegregation. In Lau v. Nicholls (1974), the Supreme Court found that
the failure to provide second-language learning services to a large group of Chinese
students who had limited English language proficiency amounted to discrimination
on grounds of national origin.32 The Court found that the school districts must “take
affirmative steps to rectify the language deficiency in order to open its instructional
program to these students.”33 The case led to important new administrative guidelines
known as the “Lau remedies.”34 These forays were as much about education as they
were about discrimination, and they played a crucial role in enabling the subsequent
explosion in educational finance litigation.35
The second noteworthy form of protection was the Court’s jurisprudence on
administrative due process in benefits decision-making. In Goldberg v. Kelly (1970),36
the Supreme Court ruled that a person is entitled to a hearing before a decision to
terminate his or her welfare benefits under the AFDC program. The first issue was
whether deciding entitlements under AFDC engaged due process at all. Here the Court
was unequivocal:
Swann v. Charlotte-Mecklenburg Board of Education, 402 U.S. 1 (1971).
Adams v. Richardson, 351 F. Supp. 636 (D.C. Cir. 1972).
30
Adams v. Richardson, 356 F. Supp. 92 (D.C. Cir. 1973). See further, Joshua Dunn, Complex Justice: The Case of
Missouri v. Jenkins 28–29 (2008). Ultimately, the Supreme Court curtailed the use of busing as a remedy
to situations of de jure discrimination: Milliken v. Bradley, 418 U.S. 717 (1974).
31
Missouri v. Jenkins, 515 U.S. 70 (1995). A definitive study of it is found in Dunn, id.
32
Lau v. Nicholls, 414 U.S. 563 (1974).
33
Id. at 568.
34
See the discussion infra Section 4.1(c).
35
Melnick, supra note 8, at 144.
36
Goldberg v. Kelly, 397 U.S. 254 (1970).
28
29
Two ironies about American exceptionalism over social rights
579
Such benefits are a matter of statutory entitlement for persons qualified to receive them. Their
termination involves state action that adjudicates important rights. The constitutional challenge cannot be answered by an argument that public assistance benefits are “a ‘privilege’ and
not a ‘right.’”37 Such constitutional protection has not been recognized in Canada, which omitted
“property” (and thus “new property”) from its Charter, and provides no freestanding right to a fair trial (though does give substantive and procedural protections to
rights to life, liberty and security of the person).38 Similarly, English judges have been
slow to conclude that welfare benefits decision-making determines “civil rights and
obligations” under the European Convention on Human Rights (ECHR),39 and the
Strasbourg Court (ECtHR) has crept along slowly, braving strong dissenting judgments about “civil rights” not intended to mean “public law rights.”40
The second main issue was what process was due, and here the Supreme Court
went well beyond what the ECtHR has been willing to hold.41 The Court in Goldberg
remarked that due process did not entitle the recipient of benefits to a full judicial
trial. However, it did find only six years later that Goldberg’s requirements “closely
approximat[ed] a judicial trial . . . .”42 It is not hard to see why. The case secured constitutional entitlements, prior to termination of benefits, to (a) timely and adequate
notice; (b) an oral hearing; (c) the effective opportunity to confront and cross-examine
any adverse witnesses; (d) representation by counsel (though not state-funded counsel); (e) an impartial decision-maker (who could nonetheless be another department
official who did not participate in the initial decision); and (f) a decision based solely on
evidence and legal rules adduced at the hearing.43 It is true that, in Matthews v. Eldridge
(1976), the Supreme Court refused to apply the Goldberg requirements to Disability
Insurance (DI) determinations, finding that welfare benefits were based on financial
need and are fundamental, whereas DI benefits (which are contributory, insurancebased) were not necessarily either.44 Nonetheless, Goldberg has remained both good
law45 and an important landmark in administrative justice, and was comparatively
quite interventionist.
The third notable area of social protection under the US bill of rights concerns
rights to state-funded medical assistance for prisoners. In the recent case of Brown
Id. at 262–263 (citations and footnotes omitted).
See Robert J. Sharpe & Kent Roach, The Canadian Charter of Rights and Freedoms, ch. 13 (4th ed. 2009).
39
Ali v. Birmingham [2010] 2 AC 39 (UKSC) (statutory rights to benefits-in-kind (as opposed to cash benefits) are not “civil rights” within the meaning of the Convention).
40
Feldbrugge v. The Netherlands, 8 E.H.R.R. 425 (1986) (contributory benefits engage “civil rights”; cf.
dissent at ¶ 23); Salesi v. Italy, 26 E.H.R.R. 187 (1998) (non-contributory benefits included; cf. strong
dissent).
41
See, e.g., the cautious approach in Bryan v. U.K., 21 E.H.R.R. 342 (1995); Tsfayo v. U.K., 48 E.H.R.R. 18
(2009).
42
Mathews v. Eldridge, 424 U.S. 319, 333 (1976).
43
Id. at 266–271.
44
Mathews, 424 U.S. at 340–342.
45
The welfare reform measures in the PRWORA have diminished Goldberg’s reach. See Cynthia R. Farina,
On Misusing “Revolution” and “Reform”: Procedural Due Process and the New Welfare Act, 50 Admin. L. Rev.
591 (1998).
37
38
580
I•CON 12 (2014), 572–602
v. Plata (2011), the Supreme Court held, reaffirming the precedent set in Estelle
v. Gamble (1976),46 that “[a] prison that deprives prisoners of basic sustenance,
including adequate medical care, is incompatible with the concept of human dignity
and has no place in civilized society.”47 In Brown v. Plata, part of an established judicial
practice of reforming prisons,48 the Court ruled unconstitutional a decades-long practice of extreme overcrowding in California’s prisons. It issued a remedy which Scalia
J described in dissent as “perhaps the most radical injunction issued by a court in our
Nation’s history: an order requiring California to release the staggering number of
46,000 convicted criminals.”49
The final and most important aspect of the “other side” of the federal constitutional
story is the remedial innovation of the structural injunction.50 This remedy, roughly
speaking, is one in which a judge issues an order to a defendant institution to undertake comprehensive structural reforms. The judge retains supervisory jurisdiction,
requiring the defendant to report back to the court on success in satisfying judicially
imposed benchmarks and timelines. The judge typically orders the appointment of
an official (normally, a “special master”) who has technical proficiency in the area at
issue.51 The role of the master is to help devise and later supervise implementation of
the decree, and to report to the court. They often supervise negotiations between claimants and defendants, steering them to an agreed remedial plan that the judge subsequently turns into a binding public law obligation by means of a “consent decree,” the
violation of which will constitute contempt of court. Many such decrees remain in
effect for several years, in some cases decades (on which, see infra Section 4), and they
are supplemented from time to time by court orders that may be aimed at aspects of
administration or the legislature itself. The cases are episodic, better described as “litigations” rather than court cases. Foreign courts other than those of India have largely
been quite cautious about using these remedies; they are for the most part miles away
from anything remotely as interventionist as the practice in America.52 While structural injunctions grew out of the institutional reform litigation following Brown, they
have been used subsequently in hundreds (or more) cases, many concerned with education, disability, and mental health.
Estelle v. Gamble, 429 U.S. 97 (1976).
Brown v. Plata, 131 S. Ct. 1910, 1928 (2011).
48
Malcolm Feeley & Edward Rubin, Judicial Policy Making in the Modern State: How the Courts Reformed
America’s Prisons (2000).
49
Brown, 131 S. Ct. at 1950. This claim oversimplifies the Court’s actual remedy. It gave two years to reduce
overcrowding in other ways, and emphasized that the lower court may extend the deadline if necessary.
50
See Abram Chayes, The Role of the Judge in Public Law Litigation, 89 Harv. L. Rev. 1281 (1976); Colin
S. Diver, The Judge as Political Powerbroker: Superintending Structural Change in Public Institutions, 65 Va
L. Rev. 43 (1979); Judith Resnik, Managerial Judges, 96 Harv. L. Rev. 374 (1982). For a more recent discussion, see Charles Sabel & William Simon, Destabilization Rights: How Public Law Litigation Succeeds, 117
Harv. L. Rev. 1015 (2004).
51
See Donald L. Horowitz, Decreeing Organizational Change: Judicial Supervision of Public Institutions, 32 Duke
L.J. 1265, 1272–1276, 1297–1302 (1983).
52
See King, supra note 26, at 271–275 for a comparative overview. The situation in some Latin American
countries has begun to change markedly. See Bruce M. Wilson, Institutional Reform and Rights Revolutions
in Latin America: The Cases of Costa Rica and Colombia, 2 J. Pol. Latin Am. 59 (2009).
46
47
Two ironies about American exceptionalism over social rights
581
3.2. State constitutions
The most remarkable instances of social rights litigation are for the most part found
under state constitutions, on which there is a sizeable literature.53 Most state constitutions have some provisions related to welfare or education, and many have been
litigated. Helen Hershkoff has authored a range of articles that synthesize and restate
the main trends, arguing that the deference seen under the federal constitution is neither apposite nor observed by state courts.54 She explains that judges are often elected
and state constitutions more easily amended, and the public law litigation paradigm
has allowed judges to fashion more manageable standards of adjudication. One can
have a better appreciation of the range of social rights provisions and adjudication by
considering a few areas.
(a) Constitutional rights to welfare and housing
According to William C. Rava, twenty-three state constitutions “recognize that someone or something in the individual states will provide for those in need.”55 Some
provisions fall short of a duty; they indicate a power, or that an institution “shall be
established.” However, four states have an affirmative duty: Alabama, Kansas, New
York, and Oklahoma.56 The clause in New York’s Constitution (art. XVII(1)), introduced by amendment in a constitutional convention in 1938, provides a good example:
The aid, care and support of the needy are public concerns and shall be provided by the state
and by such of its subdivisions, and in such a manner and by such means, as the legislature
may from time to time determine.
A similar article is provided in the Kansas Constitution (art. VII(4)):
The respective counties of the state shall provide, as may be prescribed by law, for those inhabitants who, by reason of age, infirmity, or other misfortune, may have claims upon the aid of
society. The state may participate financially in such aid and supervise and control the administration thereof.
In Tucker v. Toia (1977), the New York Court of Appeals struck down an attempt by
the state legislature to deny eligibility for federally subsidized AFDC benefits to those
young claimants who had not obtained a disposition (legal order) against a relative
who would ordinarily be responsible for care. However, in the subsequent case of
Bernstein v. Toia (1977), the Court of Appeals rejected a claim that a state policy of a
For a small sample, see Burt Neuborne, Foreword: State Constitutions and the Evolution of Positive Rights,
20 Rutgers L.J. 881 (1989); Jonathan Feldman, Separation of Powers and Judicial Review of Positive Rights
Claims: The Role of State Courts in an Era of Positive Government, 24 Rutgers L.J. 1057 (1993); William
C. Rava, State Constitutional Protections for the Poor, 71 Temp. L. Rev. 543 (1998); Elizabeth Pascal, Welfare
Rights in State Constitutions, 39 Rutgers L.J. 863 (2008).
54
Helen Hershkoff, Positive Rights and State Constitutions: The Limits of Federal Rationality Review, 112
Harv. L. Rev. 1131 (1999); Helen Hershkoff, “Just Words”: Common Law and the Enforcement of State
Constitutional Social and Economic Rights, 62 Stan. L. Rev. 1521 (2010).
55
Rava, supra note 53, at 551. The constitutional provisions that I cite or quote in this section are reproduced in Appendix A to Rava’s article.
56
Id.
53
582
I•CON 12 (2014), 572–602
fixed, flat rate for shelter allowances that allowed no consideration of special circumstances was unconstitutional.57 Nevertheless, there remains a range of other welfare
cases under New York law where statutes or policies have been held to violate article
XVII of the Constitution.58
In welfare cases, the courts of Kansas largely followed the more deferential line
of authority established in New York, and the courts in Alabama deferred in 1953
and have not revisited the issue since.59 Montana provides an interesting contrast,
as it previously had an affirmative duty to secure welfare to the needy in its constitution. In Butte I, the Montana Supreme Court declared a regulation unconstitutional
which severely curtailed eligibility for state assistance to able-bodied persons under
the age of thirty-five.60 In Butte II, the Montana Supreme Court also rejected the
legislative response, which was a law limiting state assistance to able-bodied persons without children to no more than two out of any twelve consecutive months.61
Thereafter, the electorate voted to amend the constitution to replace the affirmative
duty with a permissive grant: namely, “the legislature shall provide such economic
assistance . . .” was replaced with “the legislature may provide such economic assist­
ance . . . ,” which effectively closed the door on state constitutional welfare rights
cases in Montana.62
There appears to be no explicit state constitutional rights to housing as such.
However, some courts have construed welfare provisions as securing shelter in some
circumstances.63 New York again leads the way. In Callahan v. Carey (1979), the judge
found that the City was constitutionally bound to provide a homeless shelter to certain
Bernstein v. Toia, 373 N.E.2d 238 (N.Y. 1977).
For a non-exhaustive sample, see Hudson v. Sipprell, 76 Misc. 2d 684, 351 N.Y.S.2d 915 (Sup. Ct.
1974) (Commissioner of social services had no power under the social services law to promulgate regulation requiring withholding public assistance from welfare cheats until amount withheld equaled amount
wrongfully obtained); Young v. Toia, 93 Misc. 2d 1005, 403 N.Y.S.2d 390 (1977), mod. on other grounds
66 A.D.2d 377, 413 N.Y.S.2d 530 (4th Dept., App. Div.) (form of extremely low-paid workfare declared
unconstitutional as involving both a form of involuntary servitude and failure to provide public relief
to needy); Lee v. Smith, 43 N.Y.2d 453 (Ct. App. 1977) (law excluding persons receiving federal supplemental security income (inter alia) from eligibility for home relief); Minino v. Perales, 79 N.Y.2d 883
(Ct. App. 1992) (holding unconstitutional a “deeming” provision of social services law rendering aliens
ineligible for certain state assistance); Aliessa v. Whalen, 181 Misc. 2d 334, 694 N.Y.S.2d 308 (Sup. Ct.
1999) (law tying state-funded Medicaid coverage for lawfully resident aliens to definition of “qualified
alien” under federal welfare reform law violates constitutional obligation to provide aid to needy persons);
Aliessa v. Novello, 96 N.Y.2d 418 (Ct. App. 2001)) (social services law violates letter and spirit of Art
XVII § 1 by imposing, on legal aliens, overly burdensome eligibility conditions having nothing to do with
need). In New York, the Supreme Court is the court of first instance. Appeals lie to the Supreme Court,
Appellate Division, and from there to the NY Court of Appeals, which is the final court of appeal on state
law matters.
59
Bullock v. Whiteman, 865 P.2d 197 (Kan. 1993); Atkins v. Curtis, 66 So. 2d 455 (Ala. 1953). But see
Health Care Authority v. Madison County, 601 So. 2d 459 (Ala. 1992) (the constitution requires that a
county must bear the costs of mental health treatment for the indigent).
60
Butte Community Union v. Lewis, 712 P.2d 1309 (Mont. 1986) [hereinafter Butte I].
61
Butte Community Union v Lewis, 745 P.2d 1128, 1131–1133 (Mont. 1987) [hereinafter Butte II].
62
Rava, supra note 53, at 565–566.
63
Norma Rotunno, Note: State Constitutional Social Welfare Provisions and the Right to Housing, 1 Hofstra L. &
Pol’y Symp. 111 (1996).
57
58
Two ironies about American exceptionalism over social rights
583
men, and entered a consent decree to that effect.64 In McCain v. Koch (1986), the Court
found that New York City’s practice of shuttling the families between local welfare
offices, requiring them to sleep in such offices, or in squalid hotels, could be a violation, inter alia, of the constitutional right to emergency shelter secured by article
XVII.65 The case was a motion for a preliminary injunction granted by the Supreme
Court, Appellate Division, which on this issue concluded as follows:
It is also likely that plaintiffs will succeed on their claim that NY Constitution article XVII
obligates defendants to provide emergency shelter for homeless families. New York State has
made the care of its needy residents a constitutional mandate. . . . The framers of the State
Constitution intended article XVII to require the State to take positive steps to assist the needy,
rather than to voice aspiration towards ideal social policy.66
In related follow up cases regarding the right to emergency shelter, the New York
courts found there to be clear obligations in non-constitutional New York law to furnish emergency shelter immediately. When a trial court judge found that consent
decrees against social services had been continuously left unfulfilled and that the
practices continued, she found the officials in contempt, ordered money damages
against the City, and actually ordered four senior administrative officials (including
the Deputy First Mayor) to spend a night sleeping in the offices with the applicants for
emergency housing! That last remedy was not upheld on appeal, but the highest court
in the state did find that such a remedy would in fact be available “for particularly
egregious conduct or willful inaction.”67
(b) Constitutional rights to adequate education
The extraordinary standout—and consistent with America’s oft declared commitment to equality of opportunity—are state constitutional rights or duties in respect of
education. Nearly every state constitution has an education clause that requires the
state to establish a public school system, and many of these also indicate that some
level of adequacy is constitutionally required.68 One bout of litigation in particular
illustrates the form, content and judicial interpretation of such provisions.
The Campaign for Fiscal Equity v. New York litigation commenced when the Campaign
for Fiscal Equity (CFE) filed a claim in 1993, and ended in 2006 when the NY Court
of Appeals directed the State to spend an additional $1.93 billion per year to remedy educational inadequacy in New York City’s public school system. The education
Callahan v. Carey, N.Y. L.J., Dec. 1, 1979, at 10 (Sup. Ct. N.Y. Dec. 11 1979), aff ’d 118 A.D.2d 1054 (1st
Dept., App. Div. 1986); see also Eldredge v. Koch, 98 A.D.2d 675 (1st Dept., App. Div. 1983) (finding the
same duty in respect of women).
65
McCain v. Koch, 502 N.Y.S.2d 720 (App. Div. 1986), rev’d on other grounds, 511 N.E.2d 62 (Ct.
App. 1987).
66
Id. at 730 (citations omitted). The issue dropped out of the picture (to the satisfaction of the plaintiffs) as
the litigation carried on, as was noted in McCain v. Koch, 511 N.E.2d 62, at 65–66 (Ct. App. 1987).
67
The story is told in McCain v. Dinkins 84 N.Y.2d 216, at 230–231 (1994).
68
Allen W. Hubsch, The Emerging Right to Education Under State Constitutional Law, 65 Temp. L. Rev. 1325,
1343–1348 (1992); Peter Enrich, Leaving Equality Behind: New Directions in School Finance Reform, 48
Vand L. Rev. 101, 105, 166 (1995).
64
584
I•CON 12 (2014), 572–602
clause of the constitution is not unequivocal about a constitutional right to adequate
education: “The legislature shall provide for the maintenance and support of a system of free common schools, wherein all the children of this state may be educated”
(art. XI(1) NY Constitution). However, in the first major case, CFE I (1995), the New
York Court of Appeals found that the educational article “requires the State to offer
all children the opportunity of a sound basic education,” consisting of “the basic literacy, calculating, and verbal skills necessary to enable children to eventually function productively as civic participants capable of voting and serving on a jury,” as well
as “minimally adequate physical facilities and classrooms which provide enough light,
space, heat, and air to permit children to learn,” “minimally adequate instrumentalities of learning such as desks, chairs, pencils, and reasonably current textbooks,” and
“minimally adequate teaching of reasonably up-to-date basic curricula such as reading, writing, mathematics, science, and social studies, by sufficient personnel adequately trained to teach those subject areas.”69 The case was thereafter remanded for
trial, during which the Judge Leland Degrasse Jr. heard 72 witnesses over the course
of seven months of testimony, and in which 4,300 exhibits were filed with the court.70
After wading through enormous amounts of (often conflicting) expert-testimony, the
judge concluded in 2001 that New York had violated the Education Article. In CFE II
(2003), the Court of Appeals found that a sound, basic education, as set out in CFE I
and elaborated by the trial judge, meant an “opportunity for a meaningful high school
education,” though refused to peg the meaning to any particular standard.71 It also
upheld the trial judge’s factual findings. It further ordered the state of New York to
ascertain what the cost of a sound, basic education was in the city of New York (rather
than the entire state), and ordered a compliance deadline of thirteen months from the
date of the decision.
After CFE II, the governor of the state ordered the creation of the New York State
Commission on Education Reform.72 The Commission, after much work and just
over four months short of the deadline, found and reported that the “spending gap”
for New York City schools was $1.93 billion per annum. Governor Pataki thereafter
promised to meet and in fact exceed this amount, but he could squeeze not more than
a miserly additional $300 million from the legislature. When the deadline expired and
the matter came again before the trial judge, the judge appointed a panel of three
referees (similar to special masters, discussed above). The referees heard from many
witnesses, including the governor, and then decided that the Commission’s methodology and thus conclusions were flawed. The referees found, and the trial judge
Campaign for Fiscal Equity v. State of New York, 86 N.Y.2d 307, 316, 317 (Ct. App. 1995) [hereinafter
CFE I]. The sequence of quotations is taken from Campaign for Fiscal Equity v. State of New York, 8 N.Y.3d
14 (Ct. App. 2006) [hereinafter CFE III]. On court hierarchy in New York, see supra note 58.
70
Campaign for Fiscal Equity v. State of New York, 719 N.Y.S.2d 475 (Sup. Ct. 2001). This case is not
reported as one of the “CFE” cases, as that abbreviation is reserved for the Court of Appeals cases. The
figures are reported by Chief Judge Kaye in Campaign for Fiscal Equity v. State of New York, 100 N.Y.2d
893, at 902 (2003) [hereinafter CFE II].
71
CFE II, 100 N.Y.2d at 906, 908.
72
The entire story as discussed here is told in CFE III, 8 N.Y.3d 14.
69
Two ironies about American exceptionalism over social rights
585
accepted, that the spending gap was in fact $5.63 billion per annum. This decision
was set aside on appeal to the Appellate Division, which found that the trial judge had
wrongly set aside a determination that was “arguable and reasonable.”73 Even though
much of the Appellate Division’s judgment concerned the need to show judicial deference—saying, for example, “[i]t is not for the courts to make education policy”—
in fact it “directed” the legislature to “consider” an annual funding increase of “at
least” something between $4.7 billion and $5.63 billion, and “that they appropriate
such amount, in order to remedy the constitutional deprivations found in CFE II.”74
And for good measure, it ordered them to provide $9.179 billion for a one-off capital improvement spending program! (Two of five judges dissented, arguing in strong
terms that this remedy did not go far enough.) On appeal, in the final episode of the
saga, the Court of Appeals reversed and found that it was appropriate to defer to the
Commission’s figure of $1.93 billion as reasonable (with the Chief Justice dissenting
on this point), and further that the capital improvement program was no longer necessary.75 There was a great deal of sophistication to the litigation over these several
cases, as well as sage campaigning out of court by CFE and other groups, which helped
to marshal expertise and public support.76
How out of the ordinary was the CFE litigation? One critical author called it the
“peak” of the “adequacy movement,”77 but other extraordinary cases are not too hard
to come by. In Montoy v. Kansas (2005), the Supreme Court of Kansas ordered the
legislature to provide an additional $285 million in school funding over the subsequent two years, which in per capita terms (and presumably revenue implications)
is probably of comparable importance for Kansas as the CFE litigation was for the
state of New York.78 In Abbott v. Burke XXI (2011), the New Jersey Supreme Court
ordered an increase in funding that would amount to roughly $500 million, though
that was only one judgment in a long saga in which the courts ordered legislative
action that doubtless resulted in a far greater outlay.79 Quite apart from these standout remedies, the volume of litigation itself is remarkable. Claimants have challenged
the constitutionality of school-finance systems in at least forty-three states.80 Scholars
working in the field of educational litigation frequently refer to three waves, roughly
75
76
73
74
77
80
78
79
Campaign for Fiscal Equity v. State of New York, 29 A.D.3d 175, 184 (1st Dept., App. Div. 2006).
Id. at 187, 189–191.
CFE III, 8 N.Y.3d 14.
See Michael A. Rebell, Courts and Kids: Pursuing Educational Equity Through State Courts 97–103 (2009).
Rebell was counsel in the CFE litigation.
See Alfred A. Lindseth, The Legal Backdrop to Adequacy, in Courting Failure: How School Finance Lawsuits
Exploit Judges’ Good Intentions and Harm our Children 33, at 34 (Eric A. Hanushek ed., 2006). Lindseth was
counsel for the state in the CFE litigation.
Montoy v. Kansas, 279 Kan. 817, 845; 112 P.3d 923, 940 (3 June 2005).
Abbott v. Burke, No. M-1293-09 (N.J. May 24, 2011).
See Sheila E. Murray, William N. Evans & Robert M. Schwab, Education-Finance Reform and the Distribution
of Education Resources, 88 Am. Econ. Rev. 789, esp. 791–794 (1998). The website of the campaign org­
anization the National Education Access Network maintains a range of figures, including current “litigations.” As of July 2012, it reported that fifteen “litigations” (i.e., several hearings each) were in progress
and that only five states (of 50) had never had lawsuits. For details, see National Education Access
Network, Legal Developments, http://schoolfunding.info/legal-developments/.
586
I•CON 12 (2014), 572–602
divisible between those cases challenging educational equity (i.e., de facto or de jure
discrimination), and those challenging adequacy.81 The first and second waves were
equity cases under the federal and state constitutions respectively, the latter being the
preferred route after San Antonio v. Rodriguez (1973) largely cut off the former. Seven
state constitutional equity (second wave) cases between 1973 and 1988 succeeded in
invalidating school finance systems, with a further fifteen being unsuccessful.82 The
third wave concerned adequacy cases under state constitutions, and they have had
the highest success rate. State supreme courts have since 1998 to present overturned
the school-finance systems in at least twenty-one states on adequacy grounds, rejecting only eleven challenges.83
3.3. Federal statutory law
The US welfare state was greatly expanded in the 1960s, through a range of legislative developments concerning educational equity, fair housing, social security rights
for the disabled, and much more.84 When the Supreme Court rebuffed welfare rights
claims under the federal constitution, they unwittingly pushed welfare rights advocates towards federal statutory law. And they were remarkably successful there.
From the judicial perspective, statutory interpretation did not raise the same countermajoritarian problem, and judges could nudge the law along incrementally instead of
dramatically in each case. Congress, for its part, in many cases obliged by increasing
the number of vague statutory rights, sometimes including “citizen suit” provisions
that invited private instead of purely agency action for enforcement against states.85
A few examples illustrate the role of courts in this new order.86
(a) Aid to families with dependent children
AFDC began as a relatively small and unimportant program under the Social Security
Act of 1935 (as Aid to Dependent Children), and expanded through amendments in
the 1950s and 60s into the key program for delivering means-tested non-contributory
benefits to single-parent families with dependent children, until it was abolished in welfare reform in 1996.87 Until the 1960s, it was presumed by states, federal government
See William E. Thro, The Third Wave: The Impact of the Montana, Kentucky, and Texas Decisions on the Future
of Public School Finance Reform Litigation, 19 J.L. & Educ. 219 (1990); Michael Heise, State Constitutional
Litigation, Educational Finance, and Legal Impact: An Empirical Analysis, 63 U. Cin. L. Rev. 1735 (1995);
John Dinan, School Finance Litigation: The Third Wave Recedes, in From Schoolhouse to Courthouse: The
Judiciary’s Role in American Education 96 (Joshua M. Dunn & Martin R. West eds., 2009).
82
Dinan, id. at 97–98.
83
Id. at 98; see also, for earlier data, Murray et al., supra note 80, at 789, 791–794; Rebell, supra note 76, at
2–3.
84
See Melnick, supra note 8, at 27, 41 (for a partial list).
85
See Ross Sandler & David Schoenbrod, Democracy by Decree: What Happens When Courts Run Government
(2003).
86
See Melnick, supra note 8, at Pt. II, and Sandler & Schoenbrod, id., esp. at 22–23, for a much broader survey.
87
See Melnick, supra note 8, at Pt. II for an in-depth discussion to which my own discussion is heavily
indebted. See also Handler, supra note 3, at 457, 480–483 (for a description of the program in context).
81
Two ironies about American exceptionalism over social rights
587
and agencies alike that rules of eligibility and overall administration would fall principally within state discretion, even though it was a federal program administered
under federal statutes and later under the aegis of a federal agency (presently called
the Department for Health and Human Services (HHS)). Yet since it was a program
under which states became eligible for federal matching funds, it fell into a relationship that came to be described as “cooperative federalism.”88
The case of King v. Smith (1968), Melnick explains, “announced a novel interpretation of the AFDC title of the Social Security Act that effected major changes in welfare
practices.”89 The case struck out what is colloquially known as the “man in the house
rule.” In the history of AFDC administration, “suitable home” eligibility requirements
gave discretion to officials to deny benefits where officials found the home “unsuitable”
for the child. This was often cover for racist and sexist moralizing by states.90 When the
federal government took action to rein in these tactics, states devised workarounds,
one of which was the “substitute parent” rule at issue in King v. Smith. The Alabama
rule deemed as a “substitute parent” any person who “lives in” or “visits frequently”
the child’s home “for the purposes of cohabiting” with the child’s adoptive or natural
mother. If so, the income of the “substitute parent” was deemed available to the child
and benefits were ceased, regardless whether the individual actually supported the
child. As to “frequency,” one official testified in King v. Smith that he believed once every
six months was sufficient.91 In that case, Mrs. Smith, an African American widow and
mother of four, was allegedly visited by Mr. Williams on weekends. Though he was
deemed the “substitute father,” he was in fact married, living with his wife, and supporting nine children of his own. The Supreme Court struck down the Alabama “man
in the house” law, because it “breached [Alabama’s] federally imposed obligation to
furnish ‘aid to families with dependent children . . . with reasonable promptness to all
eligible individuals . . . .’”92 The only persons who could be considered parents were
those legally responsible for providing for the child. The tenor of King v. Smith was that
Congress intended that families with dependent children have a needs-based right to
assistance. When added to Goldberg v. Kelly, decided in 1970, a bona fide welfare rights
vision of AFDC seemed complete. And even when the Supreme Court attempted to
row back from this vision in a confusing climb down in 1973, lower courts continued
to affirm the idea of a needs-based right in the AFDC program in a way that proved
nearly impossible for the Supreme Court to control.93
King v. Smith was also notable for finding that federal courts had jurisdiction both to
hear AFDC claims and to review state eligibility rules. This started what would become
an important stampede of case law in federal courts. In the seven years following King
v. Smith, the Supreme Court heard eighteen AFDC cases and the lower courts heard
King v. Smith, 392 U.S. 309, at 316 (1968).
Melnick, supra note 8, at 65.
90
Handler, supra note 3, and King v. Smith, 392 U.S. at 320–327.
91
King v. Smith, 392 U.S. at 314.
92
Id. at 333 (citation omitted).
93
New York State Department of Social Services v. Dublino, 413 U.S. 405 (1973). See the analysis in
Melnick, supra note 8, at 92–104.
88
89
588
I•CON 12 (2014), 572–602
“hundreds of decisions touching nearly every aspect of the program, from residency
requirements and man-in-the-house rules to work expenses calculation, pregnancy benefits, and cost-of-living adjustments.”94 This appears to have inflated costs significantly,
but whether it did or not, it was certainly perceived that way by important Congressmen.
One of the fiercest congressional advocates of welfare reform and workfare in the 1980s,
Senator Russell Long, Chairman of the Senate Finance Committee, was among the most
vocal critics of the courts in Congress, at a time when Congress also reversed numerous judicial holdings that extended eligibility.95 The political dynamics that followed this
burgeoning AFDC litigation read like the pre-history of the welfare reform movement
that culminated in the Personal Responsibility and Work Opportunity Reconciliation Act
of 1996 (PRWORA), the welfare reform law that abolished AFDC altogether, replaced it
with the much more draconian Temporary Assistance to Needy Families (TANF).96 This
program overtly declared that assistance was not an entitlement, and imposed a cap of
five years as the maximum time a state could continue to use TANF funds to supplement
their own assistance programs for any individual.97 The scale of the political backlash is
one indication of the “major changes” the courts had on the administration of welfare.
(b) Social Security rights for the disabled
The Social Security Act Amendments of 1972 gave the federal government chief
responsibility for administering aid to the aged, blind, and disabled, and, unlike AFDC,
gave clear jurisdiction to the federal government to define eligibility and benefits levels.98 The Social Security Administration (SSA) is the agency that administers disability
benefits. Disability Insurance, is insurance-based (or contributory), and Supplemental
Security Income (SSI) provides needs-based and means-tested benefits for the disabled
who have not paid towards insurance. The latter program was attacked politically
from time to time and especially in the 1980s, prompting courts to intervene.
And intervene they did. Unlike AFDC, SSI was enacted in 1972, in the midst of
the welfare rights movement. Litigation shortly began in earnest, extending Goldberg
v. Kelly pre-termination hearing rights to SSI.99 By 1977, Derthick reports that 143
lawsuits had been brought against the Social Security Administration on behalf of
SSI recipients, 105 of which were class actions. And when President Jimmy Carter
sought to restrain the courts by limiting the grounds of judicial review over SSI determinations, Congress refused.100 The courts would take on critical importance during
the disability review initiated by President Reagan in 1981.101 As part of Reagan’s
96
97
94
95
98
99
100
101
Melnick, supra note 8, at 83.
Id. at 128–130.
Id. at ch. 6, for a masterly analysis of the Congressional response.
42 U.S.C. §§ 601, 608. See generally Joel F. Handler, Welfare Reform in the United States, 35 Osgoode Hall L.J.
289 (1997).
See the discussion in Derthick, supra note 4, at 21–22.
Cardinale v. Mathews, 399 F.Supp. 1163 (D.D.C. 1975).
Derthick, supra note 4, at 137–138.
The framework for the review was contained in the Social Security Disability Amendments of 1980, and
the process outlined in Derthick, supra note 4, at 33–46.
Two ironies about American exceptionalism over social rights
589
promise to take welfare reform to Washington,102 it consisted of an order to the SSA
to review SSI recipients for eligibility to weed out non-meritorious cases, to be done in
short order under a political climate that more or less promised massive cuts. In the
first few years of the review process, of 1,203,000 cases reviewed, 495,000 had their
benefits terminated.103 These were delivered by a range of techniques: by assessing
files on the basis of new medical assessments only (rather than the person’s entire
file); inadequate notice; revision of eligibility standards in policy guidance; and a general political atmosphere of retrenchment that both cut people from the rolls and augmented refusal rates for new applicants by nearly 25 percent in the space of only a few
years.104 It was clear both at the time and admitted on all sides later that this process
yielded gross injustices, and so the courts were flooded with cases.105 For example, by
1982, Administrative Law Judges and the federal courts were overturning approximately 60 percent of the cases under review and accounted for nearly 30 percent of
the cases ultimately deemed eligible for SSI benefits.106 The courts thus helped to slow
down and humanize the disability review process.
Martha Derthick observes with respect to judicial review of the SSA that “it is tempting to conclude that [judges] do not in general defer to agency judgment or attach much
weight to administrative costs . . . .”107 Two stories help illustrate why. One concerned
the judicial rejection of the SSA’s policy of reassessing disability status based on current
medical evidence only (a practice which deliberately disregarded the patient’s history of
treatment). The Court of Appeals for the Ninth Circuit held that the SSA must instead
show a proof of improvement in the recipient’s medical condition prior to termination,
a policy the SSA opposed vigorously.108 The other concerned the Supreme Court case of
Sullivan v. Zebley (1990),109 the complete story of which is told by Jennifer Erkulwater.110
The central issue in the case was how to define childhood disability. The Social Security
Act provided that children would be considered disabled if they experience a disability
that is of “comparable severity” to one experienced by an adult disabled person under
the Act. Since adult disability was defined by reference to conditions affecting the capacity to work, there was an obvious difficulty in comparing the two conditions. The SSA
adopted a standard whereby it considered a child disabled if it satisfied the medical criteria set out in specific listings it published (the “listings approach”). Advocates argued
that the SSA should rather apply a functional approach in which they would determine
104
105
106
107
108
For a discussion in the AFDC context, see Melnick, supra note 8, at 124–130.
These figures are taken from Derthick, supra note 4, at 5, and also the discussion at 33–46.
Id. at 44.
Erkulwater, supra note 9, at 127.
Or, of “allowances” in the local lingo. Derthick, supra note 4, at 40 (together with the Appeals Council).
Id. at 151. However, Derthick argues that this would not be a safe conclusion.
Patti v. Schweiker, 669 F.2d 582 (9th Cir. 1982). It built on a similar finding in Finnegan v. Mathews, 641
F.2d. 1340 (9th Cir. 1981).
109
Sullivan v. Zebley, 493 U.S. 521 (1990).
110
Erkulwater, supra note 9, at 131–137 (Supreme Court decision in context); 181–193 (implementation and impact); and 206–209 (legislative reversal that left the functional test intact). See also Amber
R. Anderson, Disabled without Benefits: The Impacts of Recent Social Security Reforms on Disabled Children,
41 B.C. L. Rev. 125 (1999).
102
103
590
I•CON 12 (2014), 572–602
the child’s ability to undertake certain activities, adjusted for age-appropriateness rather
than work-functionality (the “functional approach”). For adults, the SSA applied a workappropriate functionality test if the applicant did not qualify under the listings approach.
Using the listings approach on its own had several problems, including that there were
no listings for a number of severe childhood disorders such as Downs syndrome, autism
and AIDS, and that it was not adaptable.111 The Supreme Court found (in a 7:2 judgment
whose majority included Scalia J) that by refusing to adopt a functional approach with
age-appropriate individualized assessments, the agency had acted “manifestly contrary”
to the statute’s “comparable severity” requirement. It was no surprise that Zebley would
have a major impact on child disability determinations—it was a class-action certified
as representing 300,000 children. But the scale of the impact was surprising. In the ten
years prior to Zebley, the number of children that became eligible under SSI rose from
228,000 to 340,000. However from 1990 to 1996 the number rose from about 350,000
to over 935,000, with roughly one third of children qualifying under the Individualized
Functional Assessment procedure mandated by the Zebley decision.112 The resulting situation prompted strong political backlash, and ultimately the Zebley decision was reversed
in the PRWOR Act of 1996. Though the functionality dimension was retained in the new
statutory framework, the severity threshold was raised and the primacy of listings was
reaffirmed.113 These two bouts of litigation may well have been extraordinary instances,
but so was the overall extent of litigation. As Erkulwater reports, “[b]y the fall of 1983,
the SSA was defending itself against 34,400 lawsuits, including 95 class action cases.”114
4. The second irony: mixed results
The second great irony is that this comparatively interventionist judicial approach
has not brought unequivocal success. It has generated quite ambivalent reactions by
numerous well-informed commentators who see the value of a stronger American
welfare state. A comprehensive survey of the studies is beyond the scope of this article.
A snapshot can portray, however, their general tenor.
4.1. Potentially good results
Some outcomes in the various cases discussed above seemed good.
(a) Increased funding
The highly rigorous and comprehensive study by the economists Murray, Evans, and
Schwab examined adequacy litigation by looking at education spending within fortysix states over a period of five years. They concluded that “court-ordered reform raises
spending in the poorest districts by 11 percent, raises spending in the medium district
Sullivan, 493 U.S. at 539.
Anderson, supra note 110, at 138; see also id. at 131–133 on the IFA procedure. See also Erkulwater, supra
note 9, at 184 et seq.
113
See Erkulwater, supra note 9, at 206–209; Anderson, supra note 110, at 133–137.
114
Erkulwater, supra note 9, at 124.
111
112
Two ironies about American exceptionalism over social rights
591
by 8 percent, and leaves spending in the wealthiest districts unchanged.”115 Those
states with school finance cases had tended to increase funding available for schooling
without reduction for other social services.116 Litigation in the area of special educational rights for handicapped children had an immediate and profound impact on
funding in that area, doubling state spending within a period of three years after a
few landmark cases, which themselves were the direct cause of the passage of a major
statute.117 The same trends are visible in the holdings of educational adequacy cases
reviewed in Section 3 above, and others have noted that a number of state legislatures
have raised taxes in response to adequacy judgments against them.118
(b) Compelling the enactment of reform legislation
Many of the cases discussed above prompted new legislation. Melnick shows how the
drafters of the Education of all Handicapped Children Act (1975) themselves admitted
to being “influenced and instructed” by court decisions, and viewed the law they had
passed as having “codified the rights already spelled out in earlier court decisions.”119
John Dinan records that “political resistance to school finance rulings has in nearly
all instances been overcome by state courts that eventually compelled enactment of
reform legislation.”120 The same was evident in the disability and AFDC litigation discussed above. The CFE, Montoy, and Abbott v. Burke litigations discussed above, for
example, all resulted in reform legislation.121 This type of impact answers some of the
concerns of those who are skeptical of judges’ capacity to carry out more systematic
reform. Legislatures provide the flexibility, diverse input, and processes that facilitate
the complex task of building welfare programs. If judgments merely set these faculties
in motion, then the competence objection is at least partly met.
(c) Administrative collaboration
As discussed above, the Lau decision on second language education rights expressly
interpreted and enforced agency guidelines on the subject. The agency, in turn, amplified the court’s holding by issuing revised guidelines known as the “Lau remedies,”
though these were abandoned under President Reagan in 1981.122 When Reagan’s
government introduced the disability reviews in 1981, a number of state governors
defied the SSA and Derthick found that they “would not have . . . had they not been
encouraged, and even in some instances commanded, to do so by the federal courts.”123
Murray et al., supra note 80, at 804.
Id.
117
See Melnick, supra note 8, at Pt. II, and for facts mentioned here, ch. 7 and p. 156.
118
Dinan, supra note 81, at 105.
119
Melnick, supra note 8, at 135 (quoting Senator Robert Stafford).
120
Dinan, supra note 81.
121
See also Rebell, supra note 76, at 100–101 on the Education Budget Reform Act of 2007–2008 in the
state of New York.
122
Patricia Gándara, Rachel Moran & Eugene Garcia, Legacy of Brown: Lau and Language Policy in the United
States, 28 Rev. Res. Educ. 27, 29–30, 38 (2004).
123
Derthick, supra note 4, at 45.
115
116
592
I•CON 12 (2014), 572–602
In a number of disability cases, the courts effectively buttressed persons within agencies or other bureaucracies that deplored the practices considered illegal by plaintiffs,
but felt powerless to counteract other players in the system.124 Furthermore, structural reform cases bring a range of diverse resources and staff to bear on the problem.
Melnick was struck more by the similarity than by the differences between court-mandated educational reforms and those initiated by superintendents of schools.125
(d) Reduction of educational inequity
The study by Evans, Murray, and Schwab found that court-mandated reform of
school-finance systems “reduces within-state inequality in spending by 19–34 percent” and that “the gains in equality are obtained by increasing spending in the poorest districts while leaving spending in the richest districts unchanged.”126 Although
results are not uniform on the point, even studies that are skeptical of adequacy cases
tend to acknowledge a modest improvement in equity.127
4.2. Failure to improve
Sometimes the evidence suggests that reform litigation has led on to little positive
change, and it fits well within the claim advocated famously by Gerald Rosenberg that
courts offer a “hollow hope” for significant social change.128
(a) Poor administrative response
Two of the leading experts on school funding reform litigation find that “there is scant
evidence that the numerous school finance judgments issued by state courts since the
1970s have measurably improved student outcomes.”129 Melnick expresses similar
views about the success of school reform litigation, in contrast with that concerning reform of mental hospitals and prisons.130 And Jerry Mashaw’s findings (along
In AFDC cases, consider Melnick, supra note 8, at 71–72, 140–145; Feeley & Rubin, supra note 48, at
307–308. One classic exposition of this absence of bureaucratic unity is found in Michael Lipsky, StreetLevel Bureaucracy: Dilemmas of the Individual in Public Services (1980). See also Jerry L. Mashaw, Bureaucratic
Justice: Managing Social Security Disability Claims esp. 21–49 (1983).
125
R. Shep Melnick, Taking Remedies Seriously, in From Schoolhouse to Courthouse, supra note 81, 17, at 31. See
also Feeley & Rubin, supra note 48, at ch. 7.
126
Murray et al., supra note 80; see also David Card & A. Abigail Payne, School Finance Reform, the Distribution
of School Spending and the Distribution of Student Test Scores, 83 J. Pub. Econ. 49 (2002); Bradley
W. Joondeph, The Good, the Bad, and the Ugly: An Empirical Analysis of Litigation-Prompted School Finance
Reform, 35 Santa Clara L. Rev. 763 (1995).
127
Dinan, supra note 81, at 105 (claiming results are “mixed,” and see esp. notes 69 and 70 in Dinan’s chapter); Melnick, supra note 125, at 27 (‘[D]esegregation litigation has generated enormous controversy but
produced little racial integration of improvement in the education of minority children.’); Christopher
R. Berry, The Impact of School Finance Judgments on State Fiscal Policy, in School Money Trials: The Legal
Pursuit Of Educational Adequacy 213 (Martin R. West & Paul E. Peterson eds., 2007) (finding modest
equity improvements).
128
Gerald N. Rosenberg, The Hollow Hope: Can Courts Bring about Social Change? (2008) (1991).
129
Joshua M. Dunn & Martin R. West, The Supreme Court as School Board Revisited, in From Schoolhouse To
Courthouse, supra note 81, 3, at 96.
130
Melnick, supra note 125.
124
Two ironies about American exceptionalism over social rights
593
with colleagues) concerning the impact of judicial review on disability determinations
were damning: “the tens of thousands of judicial review proceedings that have been
held since the disability program’s inception have either had no perceptible impact on
its functioning or have made it worse.”131 Erkulwater found that the Americans with
Disabilities Act “failed to live up to its expectations” due in part to the fact that employment of the disabled did not rise despite the economy otherwise flourishing.132
(b) Achievements blunted by political backlash
Some cases might best be characterized as one step forward, and either one or two
steps back. An egregious example may be the introduction of the PRWORA (TANF)
in 1996, which replaced AFDC with TANF, curtailed eligibility for welfare benefits by
imposing a five-year lifetime cap for receipt of AFDC benefits, restrained judicial control of eligibility requirements, and legislatively reversed the Zebley decision. It is hard
to know the extent to which court cases contributed to the adoption of PRWORA, but
the evidence suggests they played some role, possibly a strong one. And in state constitutional law, it has not been uncommon for states to amend their constitutions if they
disliked a particular judgment from its own state court.133
(c) Problem of scarcity and polycentricity
The results reported by Evans, Murray, and Schwab were encouraging because they
claimed to show that increases in spending were not accompanied by decreases in
other social services spending. If true, it is not always so. Mashaw reports that the
costs of Goldberg hearings for ineligible persons were ultimately paid for by restricting benefits levels for the eligible recipients.134 There is evidence also of paying for
expanded AFDC eligibility with decreased benefits levels.135 The overriding concern
is of course that the “unfunded mandates” imposed by courts are ultimately paid for
either by contracting eligibility or shifting money around within social services spending. Tracking such things is incredibly difficult.136
4.3. Potentially bad results
One might argue that a failure of institutions to create significant social reform
is not a damning verdict if they manage to help some people along the way. But
a “half a loaf is better than none” argument still must contend with evidence of
setbacks.
Mashaw, supra note 124, at 7. This conclusion seems unduly pessimistic in light of the role played by
courts in Reagan’s disability review.
132
Erkulwater, supra note 9, at 235.
133
Hershkoff, Positive Rights and State Constitutions, supra note 54, at 1162–1163; John Dinan, CourtConstraining Amendments and the State Constitutional Tradition, 38 Rutgers L.J. 983 (2007).
134
Mashaw, supra note 124, at 4. However, he also observes, in Jerry L. Mashaw, Due Process In The
Administrative State 260 (1985), that the SSA itself proposed to adopt an even more cumbersome set of
procedural guarantees than those adopted in Goldberg.
135
Melnick, supra note 8, at 97.
136
King, supra note 26, at 258.
131
594
I•CON 12 (2014), 572–602
(a) Defensive behavior and other litigation costs
Litigation costs comprise the costs of attorneys, disruption to administration, diversion of resources, and uncertainty in program management. According to West and
Dunn, a national survey conducted in 2004 found that up to 82 percent of teachers and 77 percent of principals engage in “defensive teaching to avoid legal challenges.”137 Robert Kagan’s study of what he calls “adversarial legalism” in America
chose the welfare state as one site to expose the phenomenon, and he summed up
his conclusions on it thus: “[i]n a regime of adversarial legalism, litigation bred more
litigation, conflicting decisions, legal complexity, and legal uncertainty.”138 Erkulwater
argues that the Americans with Disabilities Act manifests the “drawbacks of a policy
premised on litigation.”139
(b) Lengthy litigation periods with unclear political accountability
The length of litigation itself creates problems because the bureaucracy under order is
often no longer subject to effective political control. Sandler and Schoenbrod go so far as
refer to the team of attorneys, acting together with judges, as “the controlling group”
of the bureaucracy.140 Were structural reform cases of short duration and taken on the
heels of comprehensive bureaucratic or political failure, this objection would have little
bite. Yet it gains some when the litigation stretches on. In the Jose P. Anbach case in New
York City, a special education case under a federal statute, the litigation endured at least
24 years.141 Similarly, the Abbott v. Burke school finance litigation stretched from its first
ruling in 1985 to the most recent in 2011 (called “Abbott XXI”).
(c) Undue influence of attorneys
Due to the length and complexity of structural reform “litigations,” attorneys often
play a crucial role in strategizing and managing information. In the context of a New
York special education case, Sandler and Schoenbrod quote an educational staff
member as saying “[t]he lawyers would make education decisions, and the educators
would say, ‘we can live with that.’”142 The first duty of an attorney is to her client
rather than the public at large, and she will have incentives for zero-sum victories
rather than the messy compromises that are indispensible in politics. Yet the more
worrying findings are in cases where the attorneys diverge from the interests even of
the groups they claim to represent. According to Joshua Dunn’s book-length analysis of the complex desegregation case of Missouri v. Jenkins, “a local attorney found
a handful of parents willing to let their children serve as plaintiffs and went on to be
recognized by the court as the representative for every student in the district.”143 Dunn
139
140
141
142
143
137
138
Dunn & West, supra note 129, at 3.
Robert Kagan, Adversarial Legalism: The American Way Of Law ch. 8 and p. 171 (2001).
Erkulwater, supra note 9, at 232.
Sandler & Schoenbrod, supra note 85, at 61–67.
Discussed in id. at ch. 3.
Id. at 63.
See generally Dunn, supra note 30, at ch. 2. The quote is from Dunn & West, supra note 129, at 6.
Two ironies about American exceptionalism over social rights
595
explains how dissatisfaction with the litigation ultimately “infuriated the black community” and prompted suits challenging both the decree and the representation.144
(d) Collusive or “sweetheart” litigation
Donald Horowitz observes that “[n]ominal defendants are sometimes happy to be sued
and happier still to lose.”145 Melnick quotes an agency official mentioning that sometimes
“named defendants have spent days preparing defences for the suit, and nights assisting
the plaintiffs to prepare their arguments.”146 Why so? Public service delivery is far from
monolithic. States battle the fed, street-level bureaucrats fight management, professionals chafe against bureaucratic imperatives, and agency officials must defend their budgets against the two fronts of professional voraciousness and political retrenchment. The
resulting mix can be an alliance of the virtuous using duplicity to obtain changes that
ordinary politics will not furnish. Some may see that as positive, just as others will see
obtaining more resources through litigation as negative. But that it is duplicitous and
incentivizes gaming the system through litigation is obvious on the face of things.
(e) Fragmentation of policy efforts
Litigation does cause bureaucratic fragmentation, and this is negative if change is
constant and metastatic. That describes the situation faced by the Social Security
Administration: “In 1992, the SSA confronted forty-six threatened or pending class
action lawsuits and thousands of individual lawsuits dealing with issues pertaining to
its disability programs.” As a result of this, “different standards are now in operation
across the country depending on the judicial circuit in which someone lives.”147 This
type of fragmentation, according to agency officials, “flew in the face of everything
Social Security stood for,” because in their eyes equity required equal treatment and
this was a fight they often took to the states in their effort to maintain a just administration.148 Some authors advocate a role for public law as “destabilization rights,”149
though the idea of localism, welfare devolution and experimentation have in the US
probably done more to facilitate welfare retrenchment than to extend public services.
(f) Unforeseen consequences and disruption
One episode in the case of Missouri v. Jenkins illustrates the potential for unforeseen
consequences and the need for adaptation that can sometimes sit uneasily with the
paradigm of legal control. In his effort to desegregate Kansas City schools, Judge Clark
mandated the adoption of a quota by which for every six African-American children
at a school, there must be four white ones. Yet since the district could not even “come
Dunn, supra note 30, at 148, 150. Cf. id. at 165 (continued NAACP support for attorney at the Supreme
Court). For similar observations, see Sandler & Schoenbrod, supra note 85, at 125.
145
Horowitz, supra note 51, at 1294–95.
146
Melnick, supra note 8, at 148.
147
Erkulwater, supra note 9, at 142. See also Melnick, supra note 8, at 108.
148
Erkulwater, supra note 9, at 141. See also Derthick, supra note 4, at 141–143.
149
Sabel & Simon, supra note 50.
144
596
I•CON 12 (2014), 572–602
close” to filling all the “white seats” in the designated schools, many African-American
children became unable to attend the school of their choice despite the availability of
space. By 1989, there were 7,000 such students on waiting lists despite the availability of thousands of seats.150 It is of course easy to cherry pick and parade failures like
this, and the judge did ultimately rescind the plan. Yet as a general matter, the institutional integrity of judging creates inertia, because no judge wants to declare rights on
Monday and abandon them by Friday.151
4.4. Any conclusions?
It is hard to draw firm conclusions from this, and not only because it is far from an exhaustive survey. As Feeley and Rubin show, there is deep disagreement in what counts as a good
or bad outcome in many of these cases.152 Are additional resources a boon or an administrative workaround imposing unseen costs elsewhere? Is a legislative response to a court
order a victory, or does it conceal retrenchment and evasion? And are legislative reforms
the application of legislative democratic faculties, or the grudging political acceptance of
a court-mandated conclusion? Is uncertainty a problem, or does forcing a bureaucracy to
accept the consequences of pervasive injustice and manifest illegality merely force a crisis
to the political fore? How much attorney control is an acceptable cost for the infusion of
millions of dollars of additional resources, discounted by the judicial capacity to vary existing orders and hear from additional interveners? These questions defy easy or perhaps any
answers, and in the absence of answers, sweeping conclusions are untenable. It would be
foolish, however, to disregard empirical findings or to dismiss instrumentalist arguments
either in favor or against court activity altogether. One non-sweeping conclusion that one
can take from the survey above, is that the surprisingly interventionist results have not
been a ringing success. “Ambivalence” is probably the most accurate (if not charitable)
description of the prevailing attitudes among those scholars who evidently support a
strong welfare state and who have studied the question most closely.
5. Causes
It is difficult to be sure about why America has produced this mixture of low public
commitment to social spending, comparatively robust judicial interventionism, and
results that frequently confound the optimists and alarm the skeptics. While a view
from afar can be unreliable, some facts do seem pertinent.
5.1. Rights and litigation culture in America
The United States is a curious legal culture. There is a visible reluctance to engage in
what is often described as judicial balancing of competing factors, especially under a
Dunn, supra note 50, at 147–148.
Further on uncertainty, see Erkulwater, supra note 9, at esp. 144; Melnick, supra note 8, at 108; Derthick,
supra note 4, at 132–135.
152
Feeley & Rubin, supra note 48, at ch. 9.
150
151
Two ironies about American exceptionalism over social rights
597
proportionality doctrine, which is observed in human rights adjudication throughout
Europe and in Canada.153 American judges are rarely willing to call something a right
in one breath, and declare it justifiably limited in the next. But when they do recognize something to be a right, they usually mean business—and for this they deserve
proper credit. This may have two pertinent consequences: it may help to explain why
the federal constitutional jurisprudence has drawn the circle of rights more narrowly,
and also why the state courts, with clearer constitutional mandates, have given the
constitutions a reading that treats the stated rights or duties as generating strong positive obligations. Though admirable for taking rights seriously, it may be that the strong
rights and remedies model, as Tushnet has argued persuasively, is institutionally less
apt than weak-form review for protecting social welfare rights.154
Another feature is the use of the trial in public law litigation. In Montoy v. Kansas
(2005), the Kansas Supreme Court noted that “[t]he district court reached [its] conclusion after an 8-day bench trial which resulted in a record of approximately 1,400
pages of transcript and 9,600 pages of exhibits. Most of the witnesses were experts in
the fields of primary and secondary education.”155 And as already noted, the judge in
the CFE litigation heard 72 expert witnesses over a period of seven months. By contrast,
in the law of judicial review in Britain, for example, it is extremely uncommon to hear
witnesses. And a major judicial review case challenging the Convention-rights compliance of national reforms of the funding of higher education was heard over the course
of two days without any experts (and, predictably, the grander claims were dismissed).156
Canada charts a middle-way, and trials do occur with important constitutional cases,157
but are to my knowledge far less common and protracted. The point is that the modifications of the public law paradigm have, as Abram Chayes noted long ago, created space
for a type of litigation that addresses some of the concerns raised about the courts adjudicating social policy. Of course, the modifications themselves have no doubt contributed to a view (among judges and advocates) that judges can effectively manage complex
institutional reform, a conclusion from which many observers dissent.
5.2. Breakdown in inter-institutional collaboration
In my view, the operating presumption in most wealthy European and Commonwealth
states is that legislatures, administration, and courts generally show comity towards
one another, and seek to cooperate and at times collaborate in good faith in the accomplishment of the same public goals.158 One of the generally accepted goals, in such
Some suggest that there is not only balancing in US law (which I would not deny) but that it is “dominant”: see T. Alexander Aleinikoff, Constitutional Law in an Age of Balancing, 96 Yale L.J. 943 (1987). In
my view, the quite different approach taken in free speech, affirmative action, religious freedom, and
due process cases in the US by comparison with Canada and Europe is at least partially explained by the
textual differences and absence of an express license to balance.
154
Mark Tushnet, Weak Courts, Strong Rights: Judicial Review and Social Welfare Rights in Comparative
Constitutional Law (2008) (and indeed for protecting all rights, see id. at ch. 8).
155
Montoy v. Kansas, 102 P.3d 1160, 1163 (2005).
156
Hurley and Moore v. The Secretary of State for Business Innovation & Skills [2012] EWHC 201 (Admin.).
157
Chaoulli v. Quebec (Attorney General) [2005] 1 S.C.R 791; R v. Spence [2005] 3 S.C.R. 458.
158
King, supra note 26, at 10–12 and 139.
153
598
I•CON 12 (2014), 572–602
states, is the need for a well-functioning, reasonably well-funded, and fairly administered welfare state. Yet these features do not apply clearly in America, in three ways.
First, there is as noted above, no clear political commitment to the idea of general
social rights for the needy. Second, the stories of bureaucratic obstruction evident in
some cases are simply remarkable. The chronic malfunction evident in many of the
cases noted above simply has no defense in principle, libertarian or otherwise. Joshua
Dunn reports the state of the Kansas City schools that became subject to the controversial desegregation in Missouri v. Jenkins as “squalid,” over and above the de facto
segregation: “In some buildings, decayed asbestos fell from pipes, windows fell out
of rotted panes, ceiling tiles hung precariously, and hallways reeked of urine.”159 In
Brown v. Plata, the Court found that prisons designed for 80,000 inmates in fact held
twice that number and thus prevented provision of adequate medical care, and led to
those with serious mental health problems being held in tiny metal cages while awaiting proper hospital beds. In some prisons, nearly 200 prisoners were held in converted
gymnasiums, guarded by two to three corrections officers, and in some cases up to 54
prisoners could be required to share the same toilet. Suicides in the system averaged
about one a week, eighty percent higher than the national average.160 The evidence
was so shocking that the Court appended photographs of overcrowded dorms and
holding cages to its reported judgment.161 These types of cases literally shock the conscience, and pleas for deference to expertise are as inapt as those calling for democratic
restraint.
Furthermore, sometimes when judges have intervened to enforce statutory or constitutional rights, bureaucracy have manifestly denied comity. The most infamous
example was the bad faith evasion of Brown II’s requirement to desegregate with
“all deliberate speed.”162 Another less well-known instance was the policy of “nonacquiescence” adopted by the Social Security Administration in the 1970s. Fearful of
losing control of its programs to federal judges, the SSA explicitly refused to abide by
the courts’ interpretive rulings in cases other than those specific cases in which the
ruling was given.163 The rationale for the policy was the need to maintain national
uniformity, and the impossibility of doing so when each circuit split a different way
and the Supreme Court had not ruled definitively. This response generated hundreds
of motions for contempt of court.164 Ultimately Congress intervened, and legislatively
forced the SSA to adopt the standard for disability reviews mandated by the courts.165
This policy of non-acquiescence was not a one off in the social security context.166
Dunn, supra note 30, at 57, 137 (also mentioning non-functioning heating and air-conditioning).
Brown, 131 S. Ct. at 1924 (Breyer J, for the Court).
161
Id. at 1949, 1950 (App. B).
162
Brown v. Board of Education, 349 U.S. 294 (1955) (known colloquially as Brown II). See Michael Klarman,
From Jim Crow to Civil Rights: The Supreme Court and the Struggle for Racial Equality chs. 5–6 (2004).
163
The story is told in Derthick, supra note 4, at 135–151; see also Erkulwater, supra note 9, at 124–125, who
reproduces excoriating judicial criticism from the bench in a range of federal court decisions.
164
Derthick, supra note 4, at 140 (contempt), 141–143 (uniformity rationale and reaction to it).
165
Id. at 147. The agency also “surrendered” ultimately: id. at 148–151.
166
Samuel Estreicher & Richard L. Revesz, Nonacquiescence by Federal Administrative Agencies, 98 Yale L.J.
679 (1989).
159
160
Two ironies about American exceptionalism over social rights
599
Where this principle of inter-institutional collaboration breaks down, conventions
of comity between institutions evaporate, and the institutional war of all-against-all
breaks out. This situation sheds some light on the very interesting work done on comparative institutional competence by Neil Komesar. In his book Imperfect Alternatives,
Komesar draws attention to how and why any institutional analysis must compare
the faults of the available alternatives for achieving public goals, arguing that “tasks
that strain the capacities of one institution may be wisely assigned to it if the alternatives are even worse.”167 However, Komesar’s equally important point is that commentators tend to generate demand-style arguments for institutional roles (by indicating
flaws with the alternatives) that are insensitive to the supply-side problems with having other institutions carry out the task.168 In Law’s Limits, Komesar observes that
institutions tend to move together.169 It is the chronic failings of one institution (e.g.,
bureaucracy) that may, ironically, place demand-side pressures that strain the capacities of other institutions (e.g., courts), in turn generating new pathologies. This observation is borne out in the American brand of exceptionalism about social rights. 170 It
would also explain the experience of public interest litigation in India, which shares
important similarities with the experience of structural reform litigation in America
in both causes and consequences.171
5.3. Decentralization
Decentralized arrangements for financing and delivering education and other social
services may also explain the turn to legal relief, and almost certainly are a key part
of the political problem. There is a virtual “consensus in the comparative [welfare
state] literature that federalism is an impediment to welfare state expansion.”172
Legal authority and financing arrangements in the American welfare state are highly
decentralized. In educational policy, for example, responsibility is shared between the
states and the local municipalities, and nearly half of school financing is derived from
local property taxes.173 School funding thus varies radically in line with the real value
of property and local political commitment to the taxation of property to fund education. With welfare, guidelines on the delivery of AFDC benefits (continued under
Neil K. Komesar, Imperfect Alternatives: Choosing Institutions in Law, Economics and Public Policy 6 (1994).
Id.
169
Neil K. Komesar, Law’s Limits: The Rule of Law and the Supply and Demand of Rights 23 (2001), also observing
that ‘all institutions deteriorate as numbers and complexity increase.’
170
See e.g., Karen Swenson, School Finance Reform Litigation: Why are Some State Supreme Courts Activist and
others Restrained? (2000) 63 Albany L. Rev. 1147, 1179: “The states that spend the least per pupil on
public education are more likely to suffer judicial intervention in education policy.”
171
See S. P. Sathe, Judicial Activism: The Indian Experience, 6 Wash. U. J.L & Pol’t 29, esp 65ff. For critique, see
Upendra Baxi, The Avatars of Indian Judicial Activism, in Fifty Years of the Supreme Court of India: Its Grasp
and Reach 156 (Kusum Kumar & S. K. Verma eds., 2000).
172
Herbert Obinger, Francis G. Castles, & Stefan Liebfried, Introduction: Federalism and the Welfare State, in
Federalism and the Welfare State: New World and European Experiences 1, 3 (Herbert Obinger, Stephan
Liebfried & Francis G. Castles eds., 2005). This volume seeks to unsettle that orthodox picture.
173
Michael Paris, Framing Equal Opportunity: Law and the Politics of School Finance Reform 37–38 (2010) and
ch. 2, generally.
167
168
600
I•CON 12 (2014), 572–602
TANF) were largely set by the federal government, but administered by the states, and
were a frequent site of friction between both, who funded the program jointly.174
Decentralization is an important part, but not the dominant explanation, of the
resort to litigation. For one thing, other federal countries are markedly different in
their attitudes to social services. It terms of overall redistribution, the US lags well
behind even its closest federal cousins—Canada and Australia—and is very far off
others such as Austria, Germany, and Switzerland.175 If we focus further on the areas
discussed in this article, we can at the outset set aside the litigation considered above
in respect of disability rights (SSI), which was chiefly against the federal government
rather than states. And though AFDC litigation was largely concerned with state
laws and policies in particular, similar sharing arrangements between federal government and provinces exist in respect of health care and social assistance in Canada.176
Australia and Switzerland, too, are also at least as decentralized in their provision of
social assistance.177 In the United Kingdom, which is a non-federal state but invites
comparison on this subject, nearly 400 local authorities have responsibility for the
delivery and partial financing of social services such as relief for homelessness, education, social housing, and health and social care for the elderly.178 Though these
authorities make up the defendants in most public law claims for services, the English
courts are almost incomparably more deferential on matters of resource allocation
(though under circumstances where spending is higher and institutional failure
much less acute). Indeed, none of these countries have anything like the courtroom
activity found in the US.
Educational policy is truly another story, however. Robert Slavin declares that to
his knowledge the US is the only country to fund primary education based on local
wealth.179 And school finance, rather than school autonomy or state and local control,
is truly the nub of the problem. While school formula funding varies substantially
between countries, a dominant characteristic of most is a guaranteed per pupil spending premium, typically with additional allocations for particular disadvantages.180
This arrangement is largely absent in the US, and shocking inequality of school
finance persists despite various attempts at the state level to equalize.181 But does this
See further Kenneth Finegold, The United States: Federalism and its Counter-Factuals, in Federalism and the
Welfare State, supra note 172.
175
See supra Figures 1 and 2 and for a more comprehensive comparison of federal states, see Federalism and
the Welfare State, supra note 172, at 28.
176
See, e.g., John E. Osborne, The Evolution of the Canada Assistance Plan, in 1–21 Report of the Ministerial Task
Force on Program Review, Final Report, Appendix (Erik Nielsen, Chairman, Supply and Services Canada,
Mar. 1986); Keith Banting, The Social Policy Divide: the Welfare State in Canada and the United States, in
Social Fabric or Patchwork Quilt? The Evolution of Social Policy in Canada 345 (ch. 16) (Raymond B. Blake
& Jeffrey Keshen eds. 2006).
177
Federalism and the Welfare State, supra note 172, at 25.
178
King, supra note 26, at 45–46. See also Housing Act 1996, § 193; Children Act 1989, § 20; National
Assistance Act 1948, § 21.
179
Robert E. Slavin, How Can Funding Equity Ensure Enhanced Achievement?, 24 J. Education Finance 519, 520
(1999).
180
Mihály Fazekas, School Funding Formulas—Review of Main Characteristics and Impacts 9 et seq. (2012).
181
Paris, supra note 173, at 36–39.
174
Two ironies about American exceptionalism over social rights
601
quite atypical arrangement for funding education explain the role of litigation? It is
difficult to tell. Equally atypical are the existence of potent legal obligations set out in
the adequacy provisions of state constitutions and Congressional legislation like the
Education for all Handicapped Children Act 1975,182 as well as structural injunctive
relief developed by courts to enjoin compliance with these positive obligations. These
features too are also unparalleled in most other jurisdictions, and unheard of in the
wealthy ones. Furthermore, the turn to courts is seen not only in education, but is
part of a broader litigation culture found in respect of disability, prison reform, desegregation and much else. Decentralization thus appears to be an important though not
dominant cause of the phenomenon.
6. Conclusion and lessons for other countries
It is hopefully now clear that the attitude of American courts towards judicially
enforceable welfare rights can indeed be exceptional, but in ironic ways. They have
embroiled themselves more deeply into some aspects of welfare administration than
have the judiciaries in comparable countries, and the results have not been what many
of those who advocate greater judicial protection of social rights desire. The causes of
this remarkable constellation are hard to discern, but it seems more than likely that
an abiding faith in rights and courts, when combined with chronic underfunding of
public services, results in a judicial readiness to stand up for values whose principled
implications often take judges deeply into administrative territory.
Despite the uniquely American context, the experience should give foreigners
pause for careful thought (as so much else in its legal tradition does). Some lessons
do emerge. Of the high quality law and politics studies examined above, the skeptical
studies tend to outnumber the favorable ones. Studies by supporters of legal welfare
rights have tended to present narratives and legal victories, rather than studying the
consequences and causes of the litigation. Studies of legal mobilization have been
reinvigorated recently,183 and it may be that examples of particularly successful social
rights litigation will emerge. Reading between the lines in some of these cases, in policy documents and other responses to court judgments, one gets the sense that there
are significant stories to tell here that will even out somewhat the empirical picture.184
At any rate, the experience once again reaffirms the value of empirical legal studies in
developing a better understanding of the dynamics of litigation.
Although we have much with which to distinguish the American experience, it is
also important not to dismiss it as a local peculiarity. There have been benefits, but also
flaws. And many of the flaws betrayed by the record are precisely those highlighted by
See further Melnick, supra note 8, Pt. III.
See, e.g., the studies discussed in Michael McCann, Litigation and Legal Mobilization, in The Oxford Handbook
of Law and Politics 522 (ch. 30) (Keith E. Whittington, R. Daniel Kelemen & Gregory A. Caldeira, eds.,
2010) and Charles R. Epp, Law as an Instrument of Social Reform, in The Oxford Handbook of Law and
Politics, 595 (ch. 34).
184
Feeley & Rubin, supra note 48, is a highly illuminating example of a non-skeptical study that does this.
182
183
602
I•CON 12 (2014), 572–602
critics of judicial review and legalism in the area of social policy. These flaws include
adversarial legalism and defensive administrative behavior, unintended consequences
of litigation complexity, domination of bureaucracies by partisan advocates, the politicization of courtrooms by means of interventions, and the introduction of inflexibility and fragmentation into bureaucracy.185 That these skeptics’ views have received
some support from the experience does not deal a deathblow to arguments in favor
of social rights litigation,186 but it certainly does require advocates to move beyond
poaching cases of daring judicial protection from other jurisdictions with a blind eye
to the greater context. Similarly, they should resist the inclination to treat progressive
objections to social rights adjudication as mere hurdles to clear rather than bona fide
egalitarian concerns about the best institutional method for securing a political endgame that is often not in dispute.
On which see Kagan, supra note 138; Lon L. Fuller, The Forms and Limits of Adjudication, 92 Harv. L. Rev.
353 (1978–1979); Donald L. Horowitz, The Courts and Social Policy (1977); Richard B. Stewart, The
Reformation of American Administrative Law, 88 Harv. L. Rev. 1669 (1975); Carol Harlow, Public Law and
Popular Justice, 65 Mod. L.R. 1 (2002); Mashaw, supra note 124; and Mashaw, supra note 134.
186
In King, supra note 26, I took care to articulate an approach that may be taken in certain countries having
background political conditions not generally present in the United States (see esp. 10–12) and which is,
notably, distinctly crafted to overcome the key difficulties evident in the survey examined above.
185