Unformatted text preview: erican Politics Research
http://apr.sagepub.com ourt, the Solicitor General, and the Separation of Powers
Timothy R. Johnson
American Politics Research 2003; 31; 426
The online version of this article can be found at:
ttp://apr.sagepub.com/cgi/content/abstract/31/4/426 Published by:
http://www.sagepublications.com rvices and information for American Politics Research can be found at:
Email Alerts: http://apr.sagepub.com/cgi/alerts
Citations http://apr.sagepub.com/cgi/content/refs/31/4/426 AMERICAN POLITICS RESEARCH / JULY 2003
Johnson / EXECUTIVE PARTICIPATION IN THE SUPREME COURT THE SUPREME COURT,
THE SOLICITOR GENERAL,
AND THE SEPARATION OF POWERS
TIMOTHY R. JOHNSON
University of Minnesota Supreme Court justices attempt to rule as closely as possible to their policy preferences, but their
decisions are not unconstrained. Rather, justices pay attention to the preferences of other
actors—including those external to the Court. Whereas most scholars focus on the relationship
between the Court and Congress, this article focuses on the relationship between the Court and
the executive. Specifically, it argues that justices seek information about how the administration
wants them to act because, like Congress, it can sanction the Court for making decisions that
diverge from administration policies. Certainly this information can be gathered in a number of
ways, but this article argues that when not readily available, justices can obtain it by inviting the
solicitor general to appear before the Court as amicus curiae. The findings provide the first systematic evidence that justices actively seek information about the preferences of other actors
during their decision-making process. Keywords: Supreme Court; separation of powers; solicitor general; amicus curiae; executive/judiciary relations When the U.S. government is not a party to a case before the
Supreme Court, it can still make its preferences known to the justices
by participating as amicus curiae. In this capacity, solicitors general
may submit briefs at their discretion (Supreme Court Rule 37.4). This
discretionary participation suggests that the president wants to make
his preferences known to the Court, especially when a case is an
“agenda” issue for the administration (Salokar, 1992, pp. 134-142).
Author’s Note: I would like to thank Richard Herrera, Eric Smith, and Barry Burden for generously sharing their data. I thank Juliet Gainsborough for her comments and criticism on early
drafts of the article and Laura Bishin for her enthusiastic support. Finally, I thank Gary King and
Micah Altman, who provided an unparalleled work environment at the Harvard MIT Data Center, where this project began. FILTER estimates and the standard errors used in this analysis are
available at http://homer.bus.miami.edu/~bbishin/FI.htm.
AMERICAN POLITICS RESEARCH, Vol. 31 No. 4, July 2003 426-451
© 2003 Sage Publications 426
Downloaded from http://apr.sagepub.com at Serials Records, University of Minnesota Libraries on April 23, 2009 Johnson / EXECUTIVE PARTICIPATION IN THE SUPREME COURT 427 Sometimes, however, the Court seeks to obtain information about the
administration’s position even when the solicitor general does not voluntarily file as amicus. In these cases, the justices themselves may
request that the solicitor general participate by issuing the following
order: “The Solicitor General is invited to file a brief in this case
expressing the views of the United States” (Stern, Gressman, Shapiro
& Geller, 1993, p. 381). This invitation is not considered optional, and
“all solicitors general have tacitly acknowledged” this fact (Salokar,
1992, p. 142). Invitations, however, are rare; between 1953 and 1986,
the Court invited the government to participate at the merits stage an
average of only 2.15 times per term (Gibson, 1997).1 The puzzle is
when, and under what circumstances, does the Court invite the solicitor general to participate as amicus?
In this article, I provide one explanation by testing the hypothesis
that the Supreme Court is more likely to invite the solicitor general to
participate when the justices believe they might be sanctioned for
making decisions that are inconsistent with how the executive branch
wants them to act. In so doing, I compare cases of when the Court
invites the solicitor general to file as amicus curiae against those when
an invitation is not issued.
This article makes two explicit contributions to the separation-ofpowers literature. First, this literature usually focuses on the relationship either between the president and Congress (Bond & Fleisher,
1990; Bowles, 1987; Edwards, 1990; Light, 1999) or between the
Court and Congress (Eskridge, 1991a, 1991b; Gely & Spiller, 1990;
Martin, 1997; but see Martin, 2001). Thus, this analysis significantly
increases scholarly understanding of interinstitutional relationships at
the federal level because it focuses on the relationship between the
Court and the president. Second, existing literature often assumes that
justices have complete information about how Congress and the president want them to act. I argue that this is not the case and provide systematic evidence that justices actively seek information about the
preferences of the current administration. In so doing, I delineate
explicit conditions under which the justices should be and are concerned with how the current administration wants them to act. The
findings shed light on the Court’s decision-making process as well as
on the way that U.S. federal institutions interact with one another. Downloaded from http://apr.sagepub.com at Serials Records, University of Minnesota Libraries on April 23, 2009 428 AMERICAN POLITICS RESEARCH / JULY 2003 THEORETICAL FOUNDATIONS Supreme Court justices attempt to rule as closely as possible to
their most preferred goals, but their decisions are not unconstrained
(Epstein & Knight, 1998a; Eskridge, 1991b; Maltzman, Spriggs, &
Wahlbeck, 2000). Rather, as they pursue policy goals, justices pay
attention to the preferences of actors who are external to the Court—
especially those of the current Congress and administration. As
Epstein and Knight (1998a) point out, “To create efficacious law—
that is, policy that the other branches will respect and with which they
will comply—justices must take into account the preferences and
expected actions of these government actors” (p. 138). In other words,
justices on the Supreme Court act strategically when dealing with the
other branches. This section provides a theory of how inviting the
solicitor general to participate as amicus can help the justices learn
what the administration’s preferences are, why they must be cognizant
of such preferences, and when the Court is likely to issue these
THE INVITATION TO PARTICIPATE Existing literature demonstrates the informational role that amicus
briefs play in the Supreme Court’s decision-making process (Caldeira
& Wright, 1988; Epstein & Knight, 1998b; Epstein & Kobylka, 1992;
Songer & Sheehan, 1993; Spriggs & Wahlbeck, 1997). In particular,
Epstein and Knight (1998b) argue that amici curiae provide the justices with information about the preferences of other actors. Furthermore, Caldeira and Wright (1988) posit that “amicus curiae participat i o n b y o rg a n i z e d i n t e r e s t s p r ov i d e s i n f o r m a t i o n , o r s i g n a l s —
otherwise largely unavailable” (p. 1112). For my purposes, signals
about the preferences of the executive branch might come directly
from the administration through amicus briefs filed by the solicitor
Certainly, information about how the current administration wants
the Supreme Court to act is readily available when the government is a
party in a case or when the solicitor general chooses to file as amicus Downloaded from http://apr.sagepub.com at Serials Records, University of Minnesota Libraries on April 23, 2009 Johnson / EXECUTIVE PARTICIPATION IN THE SUPREME COURT 429 curiae.2 Often, however, the government is not involved in either
capacity. Indeed, even when the administration is not a litigant and a
case can be considered one of its priorities (Light, 1999), the government voluntarily files amicus briefs in only 8% of all cases (Gibson,
1997). In the remaining 92% of cases (whether or not they are on the
administration’s agenda), the justices might still want information
about the preferences of the executive branch; they can obtain it by
inviting the solicitor general to participate as amicus curiae (Stern
et al., 1993, p. 563). As Stern et al. (1993) argue,
Such an invitation reflects the fact that some governmental interest
might be involved in a case, an interest that is not represented by the private litigants. Or the question involved might be of sufficient public
concern that the views of the government are felt to be relevant to the
Court’s consideration of the case. (p. 381) Thus, when the justices want to be clear about where the administration stands on an issue, and the solicitor general does not voluntarily
file, they are likely to issue an amicus invitation.3
One might wonder why the Court would ever issue an invitation—
if the solicitor general can participate as amicus curiae without an
invitation from the Court, the justices would learn about the administration’s preferences whenever the solicitor general deemed it necessary to provide such information. In other words, because of Rule
37.4, one could argue that the administration does not care about the
outcome of cases in which the solicitor general does not voluntarily
file, and therefore, the justices do not have to be concerned with how
the president will react to their decisions in these cases. However, a
norm exists that stops the solicitor general’s office from participating
of its own volition in too many cases. As former solicitor general Rex
Lee (1986) argues, “It is a mistake to file in too many [cases],” because
in so doing “the ability of the Solicitor General to serve any of the
president’s objectives would suffer” (pp. 599-600). Salokar (1992)
agrees and notes that solicitors general must not “become involved in
so many cases that the Court should begin to expect the government’s
views, and as a result, give them less weight” (p. 141). Therefore, there
are times when the administration might really want to file an amicus Downloaded from http://apr.sagepub.com at Serials Records, University of Minnesota Libraries on April 23, 2009 430 AMERICAN POLITICS RESEARCH / JULY 2003 brief but believes that filing will damage the reputation of the solicitor
The key point for my theory, then, is that the solicitor general’s
office must weigh its reputation against the policy it could achieve by
filing in a particular case. As one former assistant solicitor general
The question is whether you lose some of that credibility by filing
briefs in cases where it is clear to everybody, including the Court, that
the only interest is political, political in the sense that this is this administration’s philosophy. (Salokar, 1992, p. 14) To put it another way, the decision to participate as amicus curiae is
essentially a bidding game for the solicitor general. Although a plethora of cases exist in which the administration wants its views known to
the justices, the solicitor general knows that he cannot express these
views in every case. At the same time, the solicitor general knows that
the Court does issue invitations in some cases when the solicitor general does not choose to participate. As such, this dynamic can be
understood as a game in which the solicitor general and the Court
work together to ensure that the justices know the administration’s
views even when the government does not voluntarily participate as
Overall, given that the solicitor general’s office often does not voluntarily participate as amicus curiae, the justices issue invitations to
solicit the administration’s views. This is consistent with what solicitors general themselves say about amicus invitations. For instance,
Rex Lee argues that “the Court may have . . . wanted to know what
impact the case would have on the government” (as cited in Salokar,
1992, p. 144). Others in the solicitor general’s office explicitly suggest
that the Court ask for their help because the justices sometimes cannot
figure out the federal interest (Salokar, 1992, p. 143). These accounts
are also consistent with Epstein and Knight’s (1998a) conception of
the strategic model. That is, invitations appear to be explicit requests
by the Court for information about the executive branch and its policy
goals.4 Downloaded from http://apr.sagepub.com at Serials Records, University of Minnesota Libraries on April 23, 2009 Johnson / EXECUTIVE PARTICIPATION IN THE SUPREME COURT 431 EXECUTIVE SANCTIONS AND THE COURT Generally, Supreme Court justices account for how the executive
branch might react to decisions and sometimes issue invitations for
the solicitor general to provide this information because the president
can sanction the Court in a number of ways if he, or an agency, does
not agree with its decisions.5 I focus on three sanctions that might
come into play. First, although executive agencies have the power to
enforce the Court’s decisions, they do not have to do so. As Epstein
and Walker (1998) note, “The bureaucracy can assist the Court in
implementing its policies, or it can hinder the Court by refusing to do
so, a fact of which the justices are well aware” (p. 43). Although scholars debate about whether the president fully controls the bureaucracy
and is able to use it for his political advantage, Moe (1982) demonstrates that presidents have some control over independent commissions. Thus, even though a president might not be able to unilaterally
order an agency to disregard a Court decision, the threat is real; it has
been carried out in the past. For instance, Wasby (1994) notes that the
Reagan administration had a policy of “nonacquiescence” (p. 330) for
judicial decisions that it disliked, especially in social security cases.
Although the president might not have absolute control over the
bureaucracy, he can personally sanction the Court by refusing to
enforce its decisions. The most oft-cited example of this behavior is
President Jackson’s response to a Court decision that he particularly
disliked: “John Marshall has made his decision, now let him enforce
it” (Ducat, 1996, p. 110). Other confrontations demonstrate that the
president can and does judge whether the Court has made the right
decision. For instance, President Jackson vetoed a bill that established
a national bank, even after the Court declared such an entity constitutional (Wasby, 1994). Several years later, President Lincoln defied the
Taney Court by refusing to release an alleged traitor, imprisoned while
the right of habeas corpus was suspended, even though the Court
ordered him to do so (Wasby, 1994). This concern about enforcement
is not relegated to the 19th Century. Rather, Ducat (1996) notes Justice
Frankfurter’s concern when the Court decided Brown v. Board of Education (1954): “Nothing could be worse from my point of view than
for this Court to make an abstract declaration that segregation is bad
and then have it evaded by tricks.” Downloaded from http://apr.sagepub.com at Serials Records, University of Minnesota Libraries on April 23, 2009 432 AMERICAN POLITICS RESEARCH / JULY 2003 Beyond refusing enforcement, the administration can support antiCourt action in Congress if the president or an agency disagrees with
the justices’ policy choices (Baum, 1995, p. 159). Two examples illustrate this tactic: President Roosevelt’s Court-packing plan in response
to the justices’ continued rejection of the administration’s New Deal
policies, and President Jefferson’s involvement in forwarding the
impeachment of Samuel Chase (Rehnquist, 1992, pp. 22-23). Finally,
presidents and their advisors can publicly criticize the Court if they
disagree with its decisions (Baum, 1995, p. 159), or they can fail to
support it for decisions with which they disagree. Baum (1995) argues
that President Reagan and his Justice Department often used the former strategy, whereas President Eisenhower used the latter tactic.
In general, although rarely invoked by the executive branch, the
sanctions delineated here might decrease the Court’s power as the ultimate arbiter of the law. It is easy to see why. If an administration
refuses to enforce the justices’ decisions, then the Court is impotent to
make or affect policy. Similarly, public criticism or anti-Court measures can erode the Court’s legitimacy. Thus, Supreme Court justices
must, on occasion, account for how the executive branch might react
to their decisions, and ensure that they do not stray too far, too often
from its preferred policy goals. In other words, justices “act strategically, anticipating the wishes of the executive branch, and responding
accordingly to avoid a confrontation” (Epstein & Walker, 1998, p. 43).
SANCTIONS, INVITATIONS, AND PRESIDENTIAL POLITICAL CAPITAL The question, however, is, When will the Court seek information
about the preferences of the executive branch and specifically issue
invitations to do so? My theoretical argument focuses on the president’s power, or what Light (1999) calls political capital. For Light, a
president’s strength is composed of partisan support in Congress, his
approval ratings, and his margin of victory in the most recent election
(p. 32). When each of these factors increases, the president gains political capital and is, therefore, more likely to garner congressional support for his domestic agenda.
Although Light’s (1999) argument speaks directly to the relationship between the president and Congress, a similar argument can be Downloaded from http://apr.sagepub.com at Serials Records, University of Minnesota Libraries on April 23, 2009 Johnson / EXECUTIVE PARTICIPATION IN THE SUPREME COURT 433 made about the relationship between the executive branch and the
Court. Just as the president is more successful at fulfilling his domestic agenda when he has more political capital, the administration
should be more successful at influencing the Supreme Court’s decisions when the president is politically strong. In other words, justices
know that an administration with a strong president is more likely than
an administration with a weak president to sanction the Court, because
the former has more political capital to expend. Thus, administrations
with stronger chief executives should be more successful at ensuring
that the justices do not stray too far, too often, from executive policy
More specifically, Light’s (1999) measures of political capital parallel the sanctions delineated in the previous section. For instance,
when the president enjoys high public approval ratings, he is more
likely to consider not enforcing Court decisions that are contrary to the
administration’s policy goals. He is also more likely to publicly criticize the Court when he enjoys high levels of public support. That is,
when the public supports the president, he can put pressure on the
Court. This conforms to the literature that finds an indirect effect of
public opinion on the choices justices make (Marshall, 1989; Mishler
& Sheehan, 1993, 1996). A similar argument applies to the president’s
use of anti-Court action in Congress. When the president has strong
support on Capitol Hill, he is more likely to utilize Congress to force
the Court to take action consistent with the administration’s policy
objectives. Generally, it is easier for the executive and the legislative
branches to coordinate their efforts against the Court when the president has strong party support in Congress.
Overall, a gap exists in our understanding of the relationship
between the executive branch and the Court. This gap needs to be
filled because it has become increasingly clear that justices consider
how the administration wants them to act when making decisions.
Furthermore, past solicitors general argue that the Court explicitly
uses amicus invitations for this purpose. Invitations to participate,
then, might be strategic tools that help justices decide cases in line
with their own preferences while avoiding sanctions from the
administration. Downloaded from http://apr.sagepub.com at Serials Records, University of Minnesota Libraries on April 23, 2009 434 AMERICAN POLITICS RESEARCH / JULY 2003 HYPOTHESES Based on the theoretical foundation, my general hypothesis is that
the Supreme Court is more likely to invite the solicitor general to participate as amicus when the president possesses high levels of political
capital. Specifically, I posit that these invitations are tied to five key
factors of presidential power.
First, conventional wisdom suggests that the president is strongest
when he has strong party support in Congress (Bond & Fleisher, 1990;
Edwards, 1990; Light, 1999). That is, when the president enjoys high
levels of support in Congress, he has more power to forward antiCourt measures, and the administration is less likely to face retribution
from Congress if the president, or an agency, refuses to enforce the
Court’s decisions. This leads me to the
Congressional support hypothesis: The Supreme Court should issue more
amicus invitations to the solicitor general when the president has support in both the Senate and the House of Representatives. Second, when the president enjoys high public approval ratings he
is more powerful, and the Court should, therefore, be more willing to
defer to the preferences of the executive branch. Indeed, when a president is more powerful he should be more willing to sanction the Court
for decisions with which he disagrees. This leads me to the
Public support hypothesis: Amicus invitations to the solicitor general
should be more frequent when the president enjoys high public
approval ratings. Third, incumbent presidents know that they have less capital during
election years and, therefore, often try to postpone debate on certain
issues until after an election (Brace & Hinckley, 1992, p. 46;
Cameron, Cover, & Segal, 1990). Therefore, I hypothesize the
Election year hypothesis: Invitations to the solicitor general should be
more prevalent in years without a presidential election. Downloaded from http://apr.sagepub.com at Serials Records, University of Minnesota Libraries on April 23, 2009 Johnson / EXECUTIVE PARTICIPATION IN THE SUPREME COURT 435 Fourth, several scholars (Bond & Fleisher, 1990; Brace & Hinckley,
1992) suggest that the executive branch is strongest during a president’s honeymoon period. In this time period Congress often defers to
policies on the president’s agenda, and the Senate usually confirms
initial cabinet nominees with little controversy (although there are
exceptions). Given these findings, I hypothesize the
Honeymoon period hypothesis: The Court should invite the solicitor general to participate as amicus more often during a president’s first year
in office. Fifth, Light (1999) argues that presidents have more capital to
expend in Congress when they win their elections by wide margins.
For instance, President Johnson enjoyed more capital than did President Kennedy because he won the 1964 election by a much larger margin than did Kennedy in 1960. As with the other measures of capital, I
argue that a similar logic applies to the relationship between the president and the Court. Thus, I hypothesize the
Margin of victory hypothesis: Invitations to the solicitor general should be
more prevalent as the president’s margin of victory in the most recent
election increases. In addition to these variables that specifically tap the president’s
political capital, I expect the Court to respond to the relative importance of a case. Specifically, given the likely impact of politically
important cases, the justices are more likely to want to be clear about
the administration’s stand on them. In short, because the stakes are
higher in salient cases, ensuring that the Court does not stray too far
from the president’s preferences becomes particularly important. As
such, I hypothesize the
Case salience hypothesis: The Court is more likely to issue invitations to
the solicitor general in cases of greater salience. Downloaded from http://apr.sagepub.com at Serials Records, University of Minnesota Libraries on April 23, 2009 436 AMERICAN POLITICS RESEARCH / JULY 2003 DATA AND METHOD To test these hypotheses, I utilize data from several sources but rely
primarily on Gibson’s United States Supreme Court Judicial Database, Phase II: 1953-1993 (1997). Gibson’s data are the first to include
variables about all amici curiae participation and position-taking
before the Supreme Court. More important for this analysis is the fact
that these data also include cases when the solicitor general participates as amicus curiae at the request of the Court. Thus, this database
provides a unique opportunity to test hypotheses concerning the strategic use of invitations.
I analyze every formally decided case from 1953 through 1985 (N =
3,778).6 The dependent measure is whether the Court invites the solicitor general to participate as amicus curiae in a given case (1 = an invitation was issued; 0 = otherwise). Normally, logit analysis is an appropriate modeling choice for a dichotomous, dependent variable.
However, invitations to the solicitor general are quite rare, which
might be a problem when modeling this phenomenon because logistic
regression underestimates the probability of a rare event occurring
(King & Zeng, 2001a, 2001b; Tomz, King, & Zeng, 1999). In other
words, the coefficient estimates in rare events are biased downward,
which affects the constant term and the remaining coefficients as a
result. Given this problem, Tomz et al. (1999) propose a correction
that lowers the mean square error of a model. Although this solution is
effective in many contexts, it is particularly useful “when the number
of observations is small (under a few thousand) and the events are rare
(under 5% or so)” (King & Zeng, 2001b, p. 158). Because my data
meet these conditions, I employ this technique in conjunction with
Stata 7.0 (Tomz et al., 1999).7
To determine whether invitations are issued only in specific types
of cases, I conducted two separate analyses on this variable. First, one
might expect the Court to seek the administration’s views only on
issues of presidential power, policies important to the current president, or procedural issues about which the solicitor general can help
the Court decide. Table 1 indicates that this is not the case, as no particular issue area dominates this variable. The most invitations, 20%,
occur in civil rights and voting rights, whereas about 10% of the invitations are issued in federal preemption cases. Thus, although two Downloaded from http://apr.sagepub.com at Serials Records, University of Minnesota Libraries on April 23, 2009 Johnson / EXECUTIVE PARTICIPATION IN THE SUPREME COURT 437 TABLE 1 Invitations by the Court to the Solicitor
General by Issue Area (1953-1985)
Civil rights/voting rights
Election of remedies/liability
Federal securities regulation
Federal consumer protection
Supreme Court jurisdiction
Review of nonfinal order
Federal state ownership dispute
Intergovernmental tax immunity
State boundary dispute
Nonreal property dispute between states
Total Number of Cases With Invitations
74 issue categories have significantly more invitations, the Court seems
to offer invitations across the issue spectrum defined in Gibson’s
Second, one might question whether invitation cases are different
from cases in which the solicitor general decides to file an amicus brief
without an invite. My analysis indicates that there is also little difference among the characteristics of cases with invitations, cases with no Downloaded from http://apr.sagepub.com at Serials Records, University of Minnesota Libraries on April 23, 2009 438 AMERICAN POLITICS RESEARCH / JULY 2003 solicitor general participation, and cases in which the solicitor general
chooses to participate. For instance, 21% of all cases without the solicitor general are complex, 22% of cases in which the solicitor general
participates voluntarily are complex, and 23% of cases with invitations fall into this category. Second, only 3% of all cases in which the
solicitor general participates are reargued (regardless of how it joined
the case). Finally, although 8% of the cases are on the president’s
agenda when the solicitor general volunteers to participate, 6% are in
this category when the Court invites the administration’s participation. In short, there is little systematic difference between invitation
cases and cases in which the solicitor general’s office joins a case on
INDEPENDENT VARIABLES The model includes several independent variables, six of which
measure the political power of the president and executive branch.
They, along with the control variables, are summarized in Table 2.
First, I include two measures to determine the president’s power to
forward anti-Court action in Congress. In the Senate, I calculate the
president’s ideological distance from the filibuster pivot—the greater
the distance, the less likely the president will be able to push antiCourt measures in the Senate.8 For the House, I simply calculate the
percentage of seats held by the president’s party.9
Third, to account for the popularity of the president as well as for
indirect public opinion effects, I use Edwards’s (1990) measure of
presidential approval. There were several ways that this variable could
be measured, but I settled on approval during the month prior to the
Court’s term. Although a month-to-month measure is intuitive, the
majority of cases are docketed toward the beginning of the term, so
this measure is appropriate.10
Fourth, I create a dummy variable to capture whether the Court
term includes an election year. Cases in terms that include a presidential election take on a value of 0, whereas cases in terms without such
an election are coded 1. This is also a difficult variable to measure
because Court terms overlap presidential election years as well as
nonelection years. Because I know when cases were argued before the
Court, I utilize the calendar year for this variable. During the 1959 Downloaded from http://apr.sagepub.com at Serials Records, University of Minnesota Libraries on April 23, 2009 Johnson / EXECUTIVE PARTICIPATION IN THE SUPREME COURT 439 TABLE 2 Variables Affecting the Supreme Court’s Decision to Invite Amicus
Participation by the Solicitor General’s Office (1953-1985)
Does the Court ask the solicitor
general to participate?
Distance of president from
Senate filibuster pivot
Percentage of House seats held
by president’s party
Presidential approval prior to
current Court term
Case occurs in nonelection year
Case occurs in president’s first year
President’s margin of victory in
most recent election
Ideological distance between
president and Court median
Case has multiple legal provisions
Case was reargued
Case issue on president’s agenda Mean Minimum Maximum SD Hypothesized
Direction 0.02 0.00 1.00 0.13 0.28 0.12 0.47 0.10 – 0.47 0.15 0.68 0.13 + 53.42
+ 8.95 0.00 23.16 8.34 + 0.19
+ SOURCE: Light (1999); Gibson (1997); Edwards (1990); Epstein, Segal, Spaeth, and Walker
(1996); Wayne, O’Brien, Cole, Mackenzie, and Mackenzie (1996).
NOTE: Because of missing values, the N for the final model is 3,778. October term, for example, cases argued beginning in January 1960
are considered in an election year. Likewise, all cases heard from
October to December 1960 (the 1960 October term for the Court) are
also in the election year.11 I take a similar tack for the other remaining
political capital variables. Fifth, I include a variable to determine
whether a Court term encompasses the first year of a president’s term
in office. Cases heard during these years are coded 1 and 0 otherwise.
Sixth, I include a measure of the president’s margin of victory in the
most recent presidential election (Light, 1999).
Finally, to determine whether more invitations are issued in highprofile cases, I control for the political salience of a case (Maltzman et
al., 2000, p. 46). Like Maltzman and his colleagues (2000), I argue that
a good measure of political salience is the number of amicus briefs
filed in a case. However, given that amicus participation has dramati- Downloaded from http://apr.sagepub.com at Serials Records, University of Minnesota Libraries on April 23, 2009 440 AMERICAN POLITICS RESEARCH / JULY 2003 cally increased over the terms included in my sample, I calculated
term-specific z scores to determine whether a case had more amicus
filings than the average case heard during a term. Thus, this variable is
composed of the z score for each case.12
I also include several control variables to test competing hypotheses. To test the attitudinal explanation (Segal & Spaeth, 1993) that
invitations are more likely to be issued when the president is ideologically aligned with the Court, I calculate the absolute ideological distance between the president and the issue-specific median justice.
This variable indirectly tests the hypothesis that the president’s most
important influence on the Court is the power to appoint like-minded
justices to the bench. For the president, I employ DW-NOMINATE
scores, and for the Court I use the percentage of cases in which the
median justice voted liberally for each issue area per term.13 Both measures are scaled from 0 to 1 so that they are comparable.
Case characteristics might also lead the Court to invite the solicitor
general’s participation. First, the Court might be more likely to issue
invitations in complicated cases. This follows Drew Days’s argument
that the Court often asks for the solicitor general’s help in multidimensional cases (Pacelle, 2003). As such, I include a variable that measures whether a case covers multiple legal provisions. Cases with multiple legal dimensions are coded 1 and 0 otherwise. Similarly, cases
might be reargued because they are complicated (Hoekstra & Johnson, in press), which means that justices might be more likely to ask
for the solicitor general’s expertise in these cases. Thus, I include a
dummy variable that is coded 1 for reargued cases and 0 for all other
cases. Additionally, I determine whether the justices are more likely to
ask for the solicitor general’s input in cases that are on the president’s
agenda. To do so, I match agenda items from each president included
in the sample (Light, 1999) with the issue of each case in the sample.
Cases that match items on a president’s agenda are coded 1, whereas
all other cases are coded 0.
RESULTS The results of the analysis are presented in Table 3, and they are
compelling.14 When the president’s party controls more seats in the Downloaded from http://apr.sagepub.com at Serials Records, University of Minnesota Libraries on April 23, 2009 Johnson / EXECUTIVE PARTICIPATION IN THE SUPREME COURT 441 TABLE 3 Rare Event Logistic Regression Model of Supreme Court’s Decision to
Invite Amicus Participation by the Solicitor General (1953-1985)
Distance of president from
Senate filibuster pivot
Percentage of House seats
held by president’s party
Presidential approval prior
to current Court term
Case occurs in nonelection
Case occurs in president’s
President’s margin of victory
in most recent election
Ideological distance between
president and Court median
Case has multiple legal
Case was reargued
Case issue on president’s
agenda Coefficient Robust SE Significance
(one-tailed test) Expected
Direction? –7.82 1.38 .00 1.16 1.62 .24 No 3.24 1.22 .00 Yes 0.03 0.01 .04 Yes 0.83 0.37 .01 Yes 0.33 0.31 .15 Yes –0.03
Yes –0.29 1.16 .40 No –0.01
Yes 0.22 0.57 .35 Yes Valid N = 3,778 House (p = .00), when he enjoys high public approval ratings (p = .04),
when a Court term does not include an election year (p = .01), and
when a case is highly salient (p = .00), invitations are significantly
more likely to be issued to the solicitor general. The honeymoon
period variable is in the right direction, but it does not reach an acceptable level of statistical significance. Finally, note that the president’s
margin of victory in the most recent election and the distance of the
president from the filibuster pivot are signed incorrectly.15 Despite
these two unexpected results, these data indicate that Supreme Court Downloaded from http://apr.sagepub.com at Serials Records, University of Minnesota Libraries on April 23, 2009 442 AMERICAN POLITICS RESEARCH / JULY 2003 justices are generally able to differentiate between strong and weak
None of the alternative explanations has an impact on the Court’s
decision to invite the solicitor general to participate. Of particular
note, the ideological congruence between the Court and the president
(an appealing alternative hypothesis) fails to predict when invitations
are likely to be issued to the solicitor general. This variable is signed
incorrectly and is not even marginally significant, which indicates that
the president’s potential support on the Court has little influence over
whether the solicitor general is invited to appear as amicus curiae. Furthermore, neither the complicated nature of a case nor the fact that it
includes an agenda issue for the president plays a significant role in
the Court’s decision to issue such an invitation. This is even stronger
evidence that this decision is a strategic one for the justices.
Because it is difficult to interpret the substantive effects of the coefficients in Table 3, Table 4 delineates the predicted probabilities of the
Court’s decision to invite the solicitor general to appear as amicus.
When all of the variables are held at their sample mean or modal values, the probability of such an invitation is only .01. In other words,
invitations really are rare events. Using this baseline, I compare the
predicted probabilities for strong and for weak presidents. Interestingly, none of the presidential capital variables, in isolation, has a
large substantive effect on the Court’s decision to invite the solicitor
general to participate as amicus curiae. For instance, when all of the
variables are held at their sample mean or mode, and party support in
the House is at its maximum (68%), the probability of an invitation
being issued only increases to 2%. Similarly, when presidential
approval is at its maximum (79%), the probability also reaches only 3%.
However, considering realistic combinations of these variables
yields clear substantive effects. For instance, if a president has high
approval ratings (e.g., Kennedy enjoyed 79% approval before the
1961 term began), if the Court term does not include an election year,
and if the president enjoys a high level of party support in the House
then this probability jumps to .05. Although this is still a small probability, it is a five-fold increase from the baseline. This suggests that the
president’s political strength does affect the justices’ desire to learn
about the administration’s preferences. Furthermore, when the above
factors hold and the case is politically salient, the probability jumps to Downloaded from http://apr.sagepub.com at Serials Records, University of Minnesota Libraries on April 23, 2009 Downloaded from http://apr.sagepub.com at Serials Records, University of Minnesota Libraries on April 23, 2009 TABLE 4 Predicted Probabilities That the Supreme Court Invites the
Solicitor General to Participate as Amicus Curiae (1953-1985)
All variables held at sample mean or mode
President the strongest
President the strongest and salient case
President the weakest
President the weakest and not salient case Strong party support in House = .68 (maximum);
nonelection year; presidential approval = 79% (maximum)
Strong party support in House = .68 (maximum); nonelection year;
presidential approval = 79% (maximum); salient case
Strong party support in House = .24 (minimum); election year;
presidential approval = 30% (minimum)
Strong party support in House = .24 (minimum); election year;
presidential approval = 30% (minimum); not a salient case p (y = 1) 95% Confidence
Interval .01 Description .01 to .02 .05 .02 to .15 .16 .07 to .28 .00 .00 to .01 .00 .00 to .01 NOTE: The predicted probabilities are computed using the RELOGIT estimates from Table 2. Continuous variables are held constant at their mean values and
dummy variables are held constant at their modal values. 443 444 AMERICAN POLITICS RESEARCH / JULY 2003 16%. In short, the substantive effect of my key variables is clear:
When the president is strong, and the justices know the case might be
particularly important to the administration, the Court is much more
likely to issue an invitation.
The opposite effect holds when the administration is politically
weak, which further supports my argument. The probability of an
invitation occurring during a presidential election year, when there is
weak party support in the House, and when the president suffers from
low approval ratings (e.g., Carter was at 30% during the 1979 term) is
less than 1%. The fact that no amicus curiae briefs are filed in a case
does not drop this probability any lower. In short, it is the political
strength of an administration that has the greatest effect on the decision to invite the solicitor general’s participation.
CONCLUSION Supreme Court justices need to account for how the executive
branch wants them to act when making decisions. If they do not, then
they risk having sanctions leveled against the Court in the long run.
Although it is sometimes easy to obtain this information, there are
cases when the government does not participate or when the president’s public statements are unclear. As such, there are times when the
Supreme Court invites the solicitor general to present the government’s position so the justices can ensure that they have information
about the administration’s preferences. The key finding is that invitations are most likely to occur when the administration holds an abundance of political capital. This has implications for the literature that
addresses the separation of powers and the relationship between the
Court and public opinion.
First, the findings add further weight to the notion that to make adequate predictions about justices’ behavior, scholars should consider
how the Court interacts with the other branches of government (Cohen
& Spitzer, 1994; Ferejohn & Weingast, 1992; Gely & Spiller, 1990).
As Martin (1997) argues, “The Supreme Court decides disputes in
light of the facts of the case vis-à-vis the ideological attitudes and values of the justices, and the political context” (p. 23). What Martin and
most separation-of-powers scholars assume, however, is that political Downloaded from http://apr.sagepub.com at Serials Records, University of Minnesota Libraries on April 23, 2009 Johnson / EXECUTIVE PARTICIPATION IN THE SUPREME COURT 445 context mainly means the president’s preferences or the preferences of
the pivotal member of Congress.
My findings suggest that political context also includes the justices’
perception of the likelihood that the president will sanction the Court.
As my theory suggests and the findings demonstrate, a president without the political capital to sanction the Court will be less likely to use
such a tactic. Indeed, although Supreme Court justices need to be concerned with the president’s preferences, if the administration does not
possess the political power to sanction the Court, then the justices
should be less worried about straying from how the administration
wants them to act. Thus, measures of political context should also
include the political capital enjoyed by the current administration or
by the controlling party in Congress.
An even more important implication focuses on the assumption
made by much of this literature (Martin, 1997, being a key exception)
that justices have complete and perfect information about the administration’s preferences (Cohen & Spitzer, 1994; Eskridge, 1991a,
1991b; Ferejohn & Weingast, 1992; Gely & Spiller, 1990). In contrast,
my theory points out that justices might not always have complete and
perfect information about how the president wants them to act. At the
same time, the findings provide evidence that a mechanism exists for
Supreme Court justices to obtain this information when the information they do possess is imperfect and incomplete. In other words, the
findings are systematic evidence that justices actively seek out information about the administration’s preferences in cases when such
information is not readily available. If this information were not available, the justices would have a much harder time making decisions
that satisfy their own preferences as well as the president’s preferences. For these reasons, this article makes a key contribution to the
Furthermore, the findings here contribute to the literature that
focuses on the relationship between public opinion and Supreme
Court decision making. It is clear that when the president’s approval
rating increases, the Court is more willing to seek out the administration’s views of a case. This comports with the literature that finds both
a direct and an indirect link between public opinion and the Court’s
decisions. This is an equally important finding because few studies Downloaded from http://apr.sagepub.com at Serials Records, University of Minnesota Libraries on April 23, 2009 446 AMERICAN POLITICS RESEARCH / JULY 2003 have provided evidence that justices think and care about how the public might react to their decisions.
At the end of the day, this article indicates that Supreme Court justices strategically utilize invitations to the solicitor general by
accounting for the strength of the administration when determining
whether they want additional information about how it wants them to
act. Specifically, when the president is weak, the justices are less
inclined to seek out information about the executive’s policy objectives. This has clear implications for our understanding of the relationship between the Court and the president as well as for our more general understanding of strategic interaction between institutions. That
is, actors are more likely to account for other actors’ preferences when
they are strong rather than weak.
1. The informal norm on the Court is that an invitation will be made at the request of any
three justices (Stern, Gressman, Shapiro, & Geller, 1993, p. 563). Note that the solicitor general
is actually invited many more times than this number reflects because the Court issues invitations at the certiorari stage as well (Salokar, 1992). However, for two main reasons, I only analyze cases when the solicitor general is invited to participate at the merits stage. First, as Spriggs
and Wahlbeck (1997) note, it is at the merits stage when policy is set, so I am more interested in
this part of the decision-making process. Second, data are much more difficult to obtain about
solicitor general participation at the certiorari stage, and they are not included in the Gibson
(1997) data set. Although there is little written about the actual timing of invitations, Stern and
colleagues (1993) indicate that the justices either discuss doing so during conference and then
vote at the same time, or the discussion and voting take place via internal memoranda after certiorari has been granted in a case.
2. Clearly, the solicitor general represents the views of the president and administration. For
instance, Meinhold and Shull (1998) demonstrate that the solicitors general are responsive to the
ideological preferences of the presidents who appoint them. Furthermore, Deen, Ignagni, and
Meernik (1998) argue that the solicitor general “is generally viewed as responsible for advancing
the President’s agenda in the legal system” (p. 4). Pacelle (1999) agrees with this assessment and
suggests that even though solicitors general are independent, “Most . . . believe it is proper to use
the office to contribute to the President’s broader agenda” (p. 2). Empirically, Epstein and
Walker (1998) demonstrate that President Clinton’s solicitor general (Drew Days III) rewrote at
least four briefs that had already been filed by President Bush’s solicitor general to reflect the
preferences of the new administration.
3. Of course, the justices can do so through other means as well. For instance, Martin (1997)
points out that because the president is so high profile, justices can learn about his preferences by
following his public statements.
4. Despite this intuition, scholars who have studied this phenomenon argue that the Court’s
invitations are often meant simply to provide the justices with additional legal arguments and Downloaded from http://apr.sagepub.com at Serials Records, University of Minnesota Libraries on April 23, 2009 Johnson / EXECUTIVE PARTICIPATION IN THE SUPREME COURT 447 information about a case (Salokar, 1992). Pacelle (1999) posits that invitations are best viewed as
instances when the Court seeks help from the top lawyers in the country. He suggests that in most
cases of invited participation, “The [Solicitor General] serves as an officer of the Court, rather
than as an advocate” (p. 12). Elsewhere, Pacelle (2003) argues that in landmark cases, the Court
might issue invitations to figure out what the office wants them to do. In other cases, however, he
suggests that the administration has no interest in the case and so acts only as a fifth clerk. The
problem is that neither of these analyses tests the propositions, nor do they provide systematic
accounts of when and why invitations are issued.
5. Note that this theory does not hinge on the fact that the justices fear being punished in any
one particular case. Rather, consistent with the strategic model (Epstein & Knight, 1998a), my
theory rests on the idea that the justices want to ensure that they do not upset the executive in too
many cases. As such, they have an incentive to invite the solicitor general’s participation if the
government has not explicitly made its views known in a case.
6. I use the U.S. citation instead of the docket number as the unit of analysis. I do so because
in consolidated cases, the solicitor general is only invited to appear once rather than for each
docketed case. Additionally, I exclude all cases that were not orally argued (thus eliminating all
summary judgment cases). Accounting for both of these factors decreases the universe of cases
from 7,157 to 4,236. Additionally, I exclude all 413 Rule 37.4 cases (those when the solicitor
general files as amicus without an invitation from the Court) because the Court would have no
opportunity to invite the solicitor general’s participation. I also exclude cases in which the
United States is a party to the case. Finally, several cases are excluded from the final analysis due
to missing values. These values include 7 cases with missing values in the case complexity variable and 38 cases in which the issue area is miscellaneous, so I cannot determine the ideological
distance between the president and the Court. After accounting for these factors, the total cases
analyzed equals 3,778.
7. The ReLogit program “estimates the same logit model as the logit command, but with an
estimator that gives a lower mean square in the presence of rare events data for coefficients, probabilities, and other quantities of interest” (Tomz, King, & Zeng, 1999, p. 1). I also estimated the
model using the more familiar logit command. Note that the results of either estimation procedure are substantially similar—neither my substantive interpretation nor levels of statistical significance are affected by the decision. Because the ReLogit technique is more appropriate given
the distribution of my data, I chose to present those results.
8. Krehbiel (1998) supports this operationalization when he argues that the senator located
at the filibuster pivot is key to the legislative process. I employ DW-NOMINATE scores because
they are directly comparable with scores across Congresses. A case could be made that I should
use Poole and Rosenthal’s Common Space scores (1998) instead of the DW-NOMINATE
scores. However, I am more comfortable with the DW-NOMINATE scores because they are
based on a 3-year trend of a legislator’s votes, and therefore can change over time, whereas the
Common Space scores are static. That is, each legislator has one score that is used for his or her
entire career. Theoretically, the dynamic nature of the DW-NOMINATE scores is more appealing. Additionally, given that these two measures correlate at .94 or higher for all Houses and Senates (see http://voteview.uh.edu/page2a.htm), the DW-NOMINATE scores are just as effective
as the Common Space scores.
9. I tried several other operationalizations for these variables, and all produced similar
results. These included the percentage of seats held by the president’s party in both houses of
Congress combined, the percentage of seats in each individual chamber, the president’s legislative victories in each chamber, and a dummy variable for divided government. Although I am
confident that any of these measures could be appropriate, the two I employ are the most theoretically pleasing. Downloaded from http://apr.sagepub.com at Serials Records, University of Minnesota Libraries on April 23, 2009 448 AMERICAN POLITICS RESEARCH / JULY 2003 10. Stern and colleagues (1993) support this tactic when they argue that the Court disposes
of the greatest number of petitions for certiorari at the beginning of the term (p. 575). For
instance, in the 1991 term, the justices considered almost 1,400 petitions in the initial
11. I use the date of argument rather than the date of the decision because I am modeling the
Court’s behavior prior to making a decision. That is, I am interested in the decision-making process prior to oral arguments.
12. Certainly the reader could argue that this variable leads to circular reasoning. That is,
more amicus participation means the case is more salient when the dependent variable actually
measures amicus participation itself. However, the theoretical justification for testing the
salience hypothesis is that when more amici curiae participate, the justices are more likely to
view the case as important. In turn, when they see both this signal and that the government is not
yet involved in the case, the justices should be more likely to issue an invitation, because I would
expect them to want the administration’s views in salient, rather than typical, cases. Given that
other measures of salience are post hoc—meaning they measure salience after a case has been
argued (see, e.g., Epstein & Segal, 2000) — I am willing to risk a circular argument in favor of an
ex ante measure. Note, however, that using Epstein and Segal’s (2000) measure produces very
similar results. Thus, my choice of the amicus measure is purely theoretical. Finally, note that the
count of amicus participation excludes the solicitor general because I would not want this measure on both sides of the equation.
13. Note that I tried other operationalizations of this variable. Specifically, I calculated the
number of justices I would expect to be ideologically aligned with the president. To do so, I determined whether the president was a Democrat (Republican). I then calculated the percentage of
justices who voted liberally (conservatively) more than 50% of the time for each issue area. I
used this percentage as the measure of ideological compatibility between the president and the
Court. Doing so does not change the results. That is, this variable is also in the wrong direction
and statistically insignificant. Thus, I am even more confident that the ideological congruence
between the president and the justices does not affect the Court’s propensity to invite the solicitor
14. The reader should be aware of two points about the model. First, King and Zeng’s
(2001b) RELOGIT output does not provide goodness of fit measures. However, when I ran the
model with a normal LOGIT estimation, the results were identical (although with negatively
biased coefficients, as King & Zeng suggest), and the log-likelihood ratio was 60.35 (df = 11).
This is statistically significant ( p = .00), which means the model as a whole performs well. Second, because I have clear expectations about the directionality of the relationship between the
dependent and independent variables, I use one-tailed tests throughout the analysis. This follows
Blalock (1979), who explains that “whenever direction has been predicted, one-tailed tests will
be preferable” (p. 163).
15. Because these two variables are signed in the wrong direction, the reader might think
that I should use the more stringent two-tailed test for them. Even using this standard, neither
reaches statistical significance. Indeed, for margin of victory, z = .07, and for the distance from
the filibuster, z = .47.
16. Given that I use multiple indicators of presidential capital, the reader might wonder
whether collinearity affects my results. It clearly does not, as the highest pairwise correlation
between these variables (the filibuster distance and a president’s margin of victory in the most
recent election) is only –.39. Thus, I am confident that collinearity does not affect my model. Downloaded from http://apr.sagepub.com at Serials Records, University of Minnesota Libraries on April 23, 2009 Johnson / EXECUTIVE PARTICIPATION IN THE SUPREME COURT 449 REFERENCES
Baum, L. (1995). The Supreme Court (5th ed.). Washington, DC: Congressional Quarterly Press.
Blalock, H. M., Jr. (1979). Social statistics (2nd Ed.). New York: McGraw-Hill.
Bond, J. R., & Fleisher, R. (1990). The president in the legislative arena. Chicago: University of
Bowles, N. (1987). The White House and Capitol Hill: The politics of presidential persuasion.
New York: Oxford University Press.
Brace, P., & Hinckley, B. (1992). Follow the leader: Opinion polls and the modern presidents.
New York: Basic Books.
Caldeira, G., & Wright, J. (1988). Organized interests and agenda setting in the U.S. Supreme
Court. American Political Science Review, 82(4), 1109-1127.
Cameron, C., Cover, A., & Segal, J. (1990, June). Senate voting on Supreme Court nominees: A
neoinstitutional model. American Political Science Review, 84, 525-534.
Cohen, L., & Spitzer, M. (1994, spring). Solving the Chevron puzzle. Law and Contemporary
Problems, 57, 65-110.
Deen, R., Ignagni, J. & Meernik, J. (1998, September). Explaining presidential support on the
Supreme Court: Ideology, politics, and the law. Article presented at the meeting of the American Political Science Association, Boston, MA.
Ducat, C. R. (1996). Constitutional interpretation (6th ed.). Minneapolis, MN: West.
Edwards, G., III. (1990). Presidential approval: A source book. Baltimore, MD: Johns Hopkins
Epstein, L., & Knight, J. (1998a). The choices justices make. Washington, DC: Congressional
Epstein, L., & Knight J. (1998b). Mapping out the strategic terrain: The informational role of
amici curiae. In H. Gillman, & C. Clayton (Eds.), Supreme Court politics: Institutional perspectives (pp. 215-235). Chicago: University of Chicago Press.
Epstein, L., & Kobylka, J. (1992). The Supreme Court and legal change. Chapel Hill: University
of North Carolina Press.
Epstein, L. & Segal J. A. (2000, January). Measuring issue salience. American Journal of Political Science, 44, 66-83.
Epstein, L., Segal, J. A., Spaeth, H., & Walker, T. (1996). The Supreme Court compendium.
Washington, DC: Congressional Quarterly Press.
Epstein, L., & Walker, T. (1998). Constitutional law for a changing America: Rights, liberties
and justice (3rd ed.). Washington, DC: Congressional Quarterly Press.
Eskridge, W. N., Jr. (1991a, November). Overriding Supreme Court statutory interpretation
decisions. Yale Law Journal, 101, 331-417.
Eskridge, W. N., Jr. (1991b, May). Reneging on history? Playing the court/Congress/president
civil right’s game. California Law Review, 79, 613-684.
Ferejohn, J., & Weingast, B. (1992, February). Limitation of statutes: Strategic statutory interpretation. Georgetown Law Review, 80, 565-587.
Gely, R., & Spiller, P. T. (1990). A rational choice theory of Supreme Court statutory decisions
with applications to the State Farm and Grove City cases. Journal of Law, Economics, and
Organization, 6(2), 263-300.
Gibson, J. L. (1997). United States Supreme Court judicial database: Phase II (ICPSR version).
Houston, TX: University of Houston (producer), ICPSR, Ann Arbor, MI (distributor).
Hoekstra, V., & Johnson, T. (in press). Delaying justice: The Supreme Court’s decision to hear
rearguments. Political Research Quarterly. Downloaded from http://apr.sagepub.com at Serials Records, University of Minnesota Libraries on April 23, 2009 450 AMERICAN POLITICS RESEARCH / JULY 2003 King, G., & Zeng, L. (2001a). Explaining rare events in international relations. International
Organization, 55(3), 693-715.
King, G., & Zeng, L. (2001b, spring). Logistic regression in rare events data. Political Analysis,
Krehbiel, K. (1998). Pivotal politics: A theory of U.S. law making. Chicago: University of Chicago Press.
Lee, R. (1986, summer). Lawyering for the government: Politics, polemics, and principle. Ohio
State Law Journal, 47, 595-601.
Light, P. (1999). The president’s agenda: Domestic policy choice from Kennedy to Clinton. Baltimore: Johns Hopkins University Press.
Maltzman, F., Spriggs, J. F., II, & Wahlbeck, P. (2000). Crafting law on the Supreme Court: The
collegial game. New York: Cambridge University Press.
Marshall, T. R. (1989). Public opinion and the Supreme Court. Boston: Allen & Unwin.
Martin, A. (2001, June). Congressional decision making and the separation of powers. American
Political Science Review, 95, 361-378.
Martin, A. (1997, November). Designing statistical tests of formal theories: The separation of
powers and the Supreme Court. Paper presented at the meeting of the Law and Society Association, St. Louis, MO.
Meinhold, S. S., & Shull, S. A. (1998, June). Policy congruence between the president and the
solicitor general. Political Research Quarterly, 51, 527-538.
Mishler, W., & Sheehan, R. S. (1996, February). Public opinion, the attitudinal model, and
Supreme Court decision making: A micro-analytic perspective. Journal of Politics, 58, 169200.
Mishler, W., & Sheehan, R. S. (1993, March). The Supreme Court as a countermajoritarian institution? The impact of public opinion on Supreme Court decisions. American Political Science Review, 87, 87-101.
Moe, T. (1982, May). Regulatory performance and presidential administration. American Journal of Political Science, 26, 197-224.
Pacelle, R. (1999, April). Between law and politics: The solicitor general and civil rights, gender
discrimination, and reproductive rights. Paper presented at the meeting of the Midwest
Political Science Association, Chicago.
Pacelle, R. (2003). Between law and politics: The solicitor general and civil rights, gender discrimination, and reproductive rights. College Station: Texas A&M University Press.
Poole, K.T. (1998, July) Recovering a basic space from a set of issue scales. American Journal of
Political Science, 42, 954-993.
Rehnquist, W. H. (1992). Grand inquests. New York: William Morrow.
Salokar, R. M. (1992). The solicitor general: The politics of law. Philadelphia: Temple University Press.
Segal, J., & Spaeth, H. (1993). The Supreme Court and the attitudinal model. New York: Cambridge University Press.
Songer, D., & Sheehan, R. (1993). Interest group success in the courts: Amicus participation in
the Supreme Court. Political Research Quarterly, 46(2), 339-354.
Spriggs, J. F., III, & Wahlbeck, P. (1997). Amicus curiae and the pole of information at the
Supreme Court. Political Research Quarterly, 50(2), 365-386.
Stern, R., Gressman, E., Shapiro, S., & Geller, K. (1993). Supreme Court practice. Washington,
DC: Bureau of Government Affairs.
Tomz, M., King, G., & Zeng, L. (1999). RELOGIT: Rare events logistic regression (Version 1.1).
Cambridge, MA: Harvard University.
Wasby, S. (1994). The Supreme Court in the federal judicial system. Chicago: Nelson Hall. Downloaded from http://apr.sagepub.com at Serials Records, University of Minnesota Libraries on April 23, 2009 Johnson / EXECUTIVE PARTICIPATION IN THE SUPREME COURT 451 Wayne, S.J., O’Brien, D.M., Cole, R.L., Mackenzie, G., Mackenzie, G.C. (1996). The politics of
American government (2nd ed.). New York: St. Martin’s. Timothy R. Johnson received his Ph.D. from Washington University in St. Louis in 1998.
His research interests include judicial behavior, Supreme Court decision making,
Supreme Court oral arguments, executive/judiciary relations, U.S. district court decision making, the impact of norms and rules on judicial behavior, and third-party politics. Downloaded from http://apr.sagepub.com at Serials Records, University of Minnesota Libraries on April 23, 2009 ...
View Full Document
- Fall '10
- American Politics, Supreme Court of the United States, President of the United States, Solicitor General