Download Statutes with Multiple Personality Disorders PDF

TitleStatutes with Multiple Personality Disorders
File Size6.3 MB
Total Pages111
Table of Contents
                            Statutes with Multiple Personality Disorders: The Value of Ambiguity in Statutory Design and Interpretation
	Joseph A. Grundfest
	Adam C. Pritchard
	Recommended Citation
Document Text Contents
Page 1

University of Michigan Law School University of Michigan Law School

University of Michigan Law School Scholarship Repository University of Michigan Law School Scholarship Repository

Articles Faculty Scholarship


Statutes with Multiple Personality Disorders: The Value of Statutes with Multiple Personality Disorders: The Value of

Ambiguity in Statutory Design and Interpretation Ambiguity in Statutory Design and Interpretation

Joseph A. Grundfest
Stanford Law School, [email protected]
Adam C. Pritchard
University of Michigan Law School, [email protected]

Available at:

Follow this and additional works at:

Part of the Courts Commons, Legislation Commons, and the Securities Law Commons

Recommended Citation Recommended Citation
Pritchard, Adam C. "Statutes with Multiple Personality Disorders: The Value of Ambiguity in Statutory
Design and Interpretation." J. A. Grundfest, co-author. Stan. L. Rev. 54, no. 4 (2002): 627-736.

This Article is brought to you for free and open access by the Faculty Scholarship at University of Michigan Law
School Scholarship Repository. It has been accepted for inclusion in Articles by an authorized administrator of
University of Michigan Law School Scholarship Repository. For more information, please contact
[email protected]
mailto:[email protected]

Page 2


Statutes with Multiple Personality Disorders:
The Value of Ambiguity in Statutory Design

and Interpretation

Joseph A. Grundfest* & A.C. Pritchard**
I. INTRODUCTION .................................................................................................. 628
II. INTERPRETATION, AMBIGUITY, AND EQUILIBRIUM .......................................... 637

A. Ambiguity and Mixed Strategies ......................................................... 637
B. Legislative Incentives for Ambiguity .................................................. 640
C. Judicial Responses to Ambiguity: The Formalist Perspective ............ 642
D. Judicial Responses to Ambiguity: The Institutional Perspective ........ 644
E. Equilibrium ......................................................................................... 649

A. Background to Legislative Action ....................................................... 650
B. The Reform Act and the "Strong Inference " Standard ...................... 652
C. The Uniform Standards Act ................................................................ 661

* William A. Franke Professor of Law and Business, Stanford Law School;
Commissioner, United States Securities and Exchange Commission, 1985-1990.

** Assistant Professor of Law, University of Michigan Law School. Professor
Pritchard was previously Senior Counsel at the Securities and Exchange Commission and in
that capacity wrote the Commission's amicus briefs in two of the cases discussed in this
Article: I re Silicon Graphics Inc. Sec. Litig., 183 F.3d 970 (9th Cir. 1999) and In re
Comshare, Inc. Sec. Litig., 183 F.3d 542 (6th Cir. 1999). The views expressed here,
however, are those of the authors alone and do not represent the views of the Commission or
its staff.

The authors are grateful to Evan Caminker, Steven Choi, Hanoch Dagan, Jesse Fried,
Tracey George, Rick Hills, Rick Lampert, Jane Schacter, Elliott Weiss, and participants at a
Fawley Lunch at the University of Michigan Law School, a faculty workshop at the Stanford
Law School, and the Judging Business conference at the University of Michigan Law School
for comments on earlier drafts of this Article. Susan Ehr provided extraordinary assistance
in constructing our District Court Database. Victoria Stodden and Charles Mathis, Ph.D.
candidates in statistics at Stanford University, were instrumental in helping execute our
statistical analyses and in constructing our Statistical Appendix, which can be found at Eric Cielaszyk, Tim Foley, Susan French, Kurt
Kauffman, Daniel Spies, and Jeff Wu provided additional invaluable research assistance.

Page 55


tendency to adopt pro-defendant interpretations of the pleading standard.
This behavioral pattern could result either because judges who are

frequently exposed to securities fraud litigation learn to be skeptical of
plaintiffs' claims on the merits (the "familiarity breeds skepticism" hypothesis),
or because these judges have an incentive based on judicial economy to clear
their dockets of frequently arising claims (the "docket control" hypothesis).
We view the data as more consistent with the "familiarity breeds skepticism"
hypothesis because variables that measure the intensity and frequency of
securities fraud litigation and litigation against technology issuers are
systematically and meaningfully significant, unlike other variables designed to
measure for general docket-control incentives. Further, judges who are
interested in economizing on resources might often be better served by issuing
pro-plaintiff rulings that could quickly drive the litigation toward a settlement.
These pro-plaintiff strategies could consume fewer judicial resources in
comparison with the judicial effort required to write complex pro-defendant
opinions on motions to dismiss that might only serve to prolong judicial
involvement at the trial and appellate levels.

Other variables are sporadically significant in our analysis, but none tell as
compelling or comprehensive a story. The identity or political affiliation of the
president who nominated a district court judge is occasionally significant.
Judges nominated by Democratic presidents tend to display pro-plaintiff
tendencies, but this pattern is hardly uniform or pervasive in the data. At the
appellate level, there also appears to be a weak tendency on the part of judges
nominated by Democratic presidents to adopt more pro-plaintiff interpretations.
We do not, however, view these patterns as sufficient to support a strong
conclusion that political inclinations are here systematically correlated with
judicial behavior. The data are, at most, suggestive.

Presence in the Northern District of Illinois or in the Southern District of
New York is correlated with pro-plaintiff rulings. These tendencies appear to
reflect judicial perspectives that are not systematically related to any other
significant variables measured in our analysis. If those court-specific variables
were more consistently significant in a larger number of regressions, we would
suggest that plaintiffs have an unexplained advantage in those two districts.
The data are not, however, strong enough in our view to support that

We also find that judges with backgrounds as prosecutors tend toward pro-
defendant interpretations of the statute. This finding is contrary to our initial
expectation that judges with prosecutorial backgrounds would tend to give
credence to plaintiffs' allegations of fraud. Again, given the large number of
regressions we have run and variables we have tested, we caution against
overinterpretation of this finding.

What are the implications of these results? At the narrowest level, they
suggest that the questions presented by the "strong inference" debate are ripe

[Vol. 54:627

Page 56


for Supreme Court review in order to resolve a split among the circuits and
confusion in the district courts. A Supreme Court decision could establish a
consistent nationwide interpretation of the "strong inference" standard, but we
are not confident that any such standard would generate a consistent set of
results on the resolution of motions to dismiss. Again, this is not to suggest
that a Supreme Court decision interpreting the "strong inference" standard
would have no effect on the resolution of the underlying motions. The
implication of our findings is, instead, that it is easy to exaggerate the degree of
consistency likely to result from any such ruling.

More broadly, however, our findings can inform the larger controversy
over methods of statutory construction and the analysis of the strategic
interaction between the legislative and judicial branches. Recall our earlier
observation that legislators seeking compromise will rationally prefer
ambiguity in the interpretation of statutory language. Without such ambiguity,
legislators operating in a rational expectations equilibrium would be unable to
implement mixed-game strategies as part of the legislative process, and
legislators with conflicting views would be unable simultaneously to claim
victory over the same legislative language. Ambiguity is thus a valuable
lubricant in the legislative process.


Judges can also value ambiguity to the extent that it allows them greater
latitude to exercise discretion, more room within which to compromise with
colleagues, and increased opportunity to avoid resolutions that they view as
unjust or incorrect by whatever metric they might apply. Judge Posner, for
example, suggests that judges often vote "their policy preferences and personal
convictions," within the confines of the "rules" of judging, as part of the
judging "game.' '179 Judges express these preferences particularly when "cases
cannot be decided by reasoning from conventional legal materials. Such cases
require a judge to exercise a legislative judgment, although a more confined
one than real legislators are authorized to exercise." 180 Statutory language that
is successfully ambiguous cannot, by construction, be "decided by reasoning
from conventional legal materials." It follows that to the extent that judges
value their ability to vote "policy preferences and personal convictions" as part
of the "game" of judging, judges should also support a technology of
interpretation that is not unduly constricting and that permits a material,
principled degree of dispute over the pragmatic meaning of words.181 The

178. See supra Part II.
179. POSNER, supra note 9, at 131. See generally supra Part II.
180. Id. (internal quotation marks omitted).
181. Much in the same vein, Judge Posner has observed that
[t]he interpretation of texts is not a logical exercise, and the bounds of "interpretation" are so
elastic... as to cast the utility of the concept into doubt.... [P]ragmatists will ask which of
the possible resolutions has the best consequences, all things that lawyers are or should be
interested in considered, including the importance of preserving language as a medium of

Apr. 2002]

Page 110


D. Discussion ofMultivariate Results

We suggest that this statistical analysis of the district court data counsels
caution more powerfully than it supports any sweeping assertions about
patterns in the behavior of district court judges. As is apparent from the
univariate and multivariate analyses alike, the significance of individual
explanatory variables is highly contingent on the specification of the dependent
variable and the regression technique employed. Because there is no generally
accepted theory that argues for the superiority of any one model over the
others, any emphasis on regressions that tend to display statistically significant
relationships can reflect data mining rather than analytically meaningful and
robust statistical relationships.

With this very significant caution firmly in mind, we call attention to a
small number of patterns that are apparent in the data, while simultaneously
urging readers not to overinterpret these findings. In particular, the
multivariate data suggest that the more often a judge rules on cases in our
sample, the more likely the judge will issue pro-defendant rulings. This
tendency supports the observation that "familiarity breeds skepticism" when it
comes to resolving motions to dismiss. The data are not, however, as strong in
supporting the conclusion that the selection of a pleading standard is
statistically significant in determining the outcome of the litigation. Thus,
consistent with our analysis of the appellate court data, there appears to be
significant plasticity in the application of standards to facts, but there is a
tendency for more stringent interpretations of the pleading standard to be
correlated with pro-defendant outcomes.

When it comes to explaining the selection of the interpretive standard it
seems that judges with more senior status, having been appointed by Presidents
Carter, Ford, or Nixon, tend to adopt more pro-plaintiff interpretations. The
pro-defendant Northern District of California effect appears to be present only
in the All Observations sample because a large number of the most pro-
defendant decisions in that district were generated by a small number of judges.

Focusing on the Individual Judge sample, however, suggests that judges
who have busier dockets and who serve in districts with more technology cases
per judge tend to adopt pro-defendant interpretations. We thus suggest that the
popular perception that the Northern District of California tends to be pro-
defendant in interpreting the "strong inference" standard is actually a reflection
of the fact that the Northern District tends to have a large technology docket
and a busy civil docket, rather than a reflection on the inclinations of the
individual judges who sit on that bench.

The fact that the RptJdg variable is statistically significant and negative we
believe reflects an interaction effect with the Tech/Judge and Docket per Judge
variables. Thus, rather than refuting the "familiarity breeds skepticism"
conclusion suggested by the univariate analysis, the RptJdg variable, which

Apr. 20021

Page 111

736 STANFORD LA WREVIEW [Vol. 54:627

would also capture a Northern District of California effect because of the
disproportionate presence of "repeat judges" in that district, acts to moderate
the effects of the Tech/Judge and Docket per Judge variables.

If this interpretation is correct, it suggests that judges in districts that have a
large number of technology cases per judge, and with a large docket load per
judge, tend to adopt pro-defendant interpretations of the standard. These
findings are consistent with the "technology is different" and "docket control"
hypotheses suggested earlier in the analysis.

These multivariate results also suggest that many variables that appeared to
be significant in the univariate analysis do not retain their significance in
multivariate regression. While this result is hardly surprising from a statistical
perspective, it is valuable to underscore here because much of the "common
wisdom" about judicial tendencies in the area of securities fraud litigation is
driven by impressions that are, at best, consistent with a subjective univariate
impression of the data. As the preceding analysis suggests, those impressions
are probably deceptive.

Similer Documents