bV1ESI_J RETURNING MATERIALS:
PIace in book drop to
LJBRARJES remove this checkout from

“ your record. FINES win
be charged if book is

 

 

 

returned after the date
stamped below.

 

 

 

 

COPING WITH COMPLEXITY: BUREAUCRATIC DECISIONMAKING
UNDER THE NATIONAL ENVIRONMENTAL POLICY ACT

By

Peter Thomas PaIuch

A DISSERTATION

Submitted to
Michigan State University
in partiaI fquiIIment of the requirements

for the degree of

DOCTOR OF PHILOSOPHY

Department of Resource DeveIopment

1985

ABSTRACT

COPING WITH COMPLEXITY: BUREAUCRATIC DECISIONMAKING
UNDER THE NATIONAL ENVIRONMENTAL POLICY ACT

By

Peter Thomas Paluch

The National Environmental Policy Act of l970 mandates a
decisionmaking reform by requiring an environmental impact
statement for any major federal action which significantly
affects the quality of the human environment. The underlying
concept is that an open, public planning process would insure
consideration of environmental factors in implementation.
After over a decade of experience, there is no consensus
about NEPA's performance. The focus of this study is on the
implementation of NEPA and agency compliance.

The primary purpose of this research is to reveal whether
environmental factors were included in agency decision
premises in their program and project planning. An empirical
investigation of 26 route location decisions by the Michigan
Department of Transportation and a case study of one decision
provided basis for this investigation. This study combines
information from these sources to make inferences about the
agency's implicit decision rule. It was assumed that an
agency would simplify the decision process and rely on only
a few factors to reach its decision. It was hypothesized
that an agency would comply where compliance was easiest to

determine, and try to retain control over most other decision

Peter Thomas Paluch
aspects.

The implementation strategy exhibited by the MDOT was
to use the early planning stages to screen and reduce the
final decision alternatives to a very small set of choices
with little variation. There was evidence that these
alternatives all met specific environmental policy require-
ments, while factors in the general spirit of NEPA had a
weak influence. Final choices became technical questions
of transportation and engineering. This placed the public
at a disadvantage because participation has a history of

being strong only in the later planning stages.

 

 

This is dedicated by my Parents
for their unwavering love, faith, and
encouragement.

ii

ACKNOWLEDGMENTS

Just as one never feels that a piece of research or
writing is finished, it is equally as difficult to determine
where an idea originated, and where the line of people who
helped develop that idea begins. Likewise I feel that many
people have contributed in some way to this dissertation
whether it was in terms of ideas, criticisms, encouragement,
or faith. I think it is in order to first thank the
Department of Resource Development and especially, Professors
Raymond Vlasin, Raleigh Barlowe, and the late Ronald Shelton,
for supporting my program of study which departed from the
traditional path. Their energy, enthusiasm, and genuine
concern for how society allocates our natural resources was
an inspiration to me.

This dissertation, as well as my program of study,
combined the substantive field of natural resource policy
and the discipline of political science. To make this
research possible I was fortunate to have had two disserta-
tion committee Chairpersons, Peter Kakela, Department of
Resource Development, and Charles Ostrom, Department of
Political Science. Peter Kakela deserves credit for
accepting to be involved in my graduate program at a late

date and still providing a true interest in my finishing.

iii

Enough words can not be expressed to thank Chuck Ostrom for
his teaching, guidance, and patience. His ideas inspired
my approach and are utilized throughout the analysis.
Completing my committee were Milton Steinmuller, Department
of Resource Development, and Gary Miller, Department of
Political Science. Both contributed several ideas from
their own experiences and very helpful criticisms.

I wish to record my thanks to a number of helpful
colleagues and friends, most of whom provided me with
beneficial advice and counsel: Cleo Cherryholmes,

Harriet Dnank, Bev Fleisher, Margaret Glikison, Robin Marra,
Robert Noll, Harold Prince, Leni Staley, and Carla Moore
Young.

I would like to especially record my gratitude to
Mary Schneider for typing the final product, as well as
earlier drafts a number of times, with skill, intelligence,
and thoughtfulness that only a good, caring friend could
provide.

Finally, Valerie Johnson deserves particular credit
for her patience and understanding during this trial of

perseverance .

iv

TABLE OF CONTENTS

Page
LIST OF TABLES. . . . . . . . . . . . . . . . . . . . viii
LIST OF FIGURES . . . . . . . . . . . . . . . . . . . ix
CHAPTER ONE - NEPA AND ITS IMPLEMENTING REGULATION.

—-l

Introduction. .
Description of the Law.
A Policy for the Environment.
A Policy for Decisionmaking .
Establishing the Council on Environmental
Quality . .
History of NEPA' 5 Administrative .Rulemaking
NEPA' 5 Early Guidelines . . . .
NEPA Implementing Regulations . .
Regulations for Implementing NEPA: A Description.
Early Planning. . . . . . . . . . . . .
EIS Preparation .
Review and Comment.
Record of Decision.
Conclusion.

N—J—J—Jd—A
WKDVVONOKOCD ounce—a

NM
03-h

CHAPTER TWO - RESEARCH PROBLEM.

0000
—l—l

Decision Context.
Level and Type of Decision. . . . . . . . .
Decisionmaker . . . . . . . . . . . . . . . . 34
Decisionmaking. . . . . . . . . . . 36
Boundedly Rational Decisionmaking . . . . . . . . 40
Cognitive Structure . . . . . . . . . . . . . 4l
Decision Premises . . . . . . . . . . . . . . 43
Factual Premises. . . . . . . . . . . . . 44
Value Premises. . . . . . . . . . . . . . 46
Role. . . . . . . . . . . . . . . . 47
Decisionmaking Logic. . . . . . . . . . . . . 49
Generation of Alternatives. . . . . . . . . . El

00
N

Rules of Inference. . . . . . . . . . . . . 53
Agency Compliance with NEPA . . . . . . . . . . . 55
Bureaucratic Politics . . . . 56
A Bureaucratic Strategy for NEPA Compliance . 58
Conclusion. . . . . . . . . . . . . . . . . . . . 6O

CHAPTER THREE - DECISION MODEL: HIGHWAY ROUTE LOCATION
DECISIONMAKING IN MICHIGAN. . . . . . . . . -

The Decision. . . . . . . . . . . . . .
Type of Decision: Highway Route Location.
The Decisionmaker: MDOT . . .

Decision Process. Route Location Planning
Decision Criteria . . .
MDOT Route Location Premises.

MDOT Compliance . .
The MDOT Decision Rule.
Route Location Factors. . .
Transportation, Planning, and Cost.
Associated Effects. . .
Operational Model of MDOT Decisionmaking.
The Dependent Variable. . .
The Independent Variable.
Research Approach .
Screen Factors. . . . . . . . . . . .
Final Decision Factors. . . . . . . . . .
Level of Transportation Service
Reasonable Cost . . . . . .
Degree of Community Compat1b1l1ty . . . .
Conclusion. . . . .

CHAPTER FOUR - ANALYSIS OF EMPIRICAL RESULTS.

Data Analysis: Rated Factors.
Factors Rated with Choice
Model Estimation.

PROBIT. .

Hypothesis Tests.

Conclusion.

CHAPTER FIVE - A CASE STUDY OF MDOT DECISIONMAKING:
ROUTE LOCATION PLANNING FOR INTERSTATE 69 FROM
CHARLOTTE T0 LANSING. . .

Introduction. . . . . . . . . . . . . . . .
Project Study: I-69 Route Location Decision
History and Background. . . . . . . . . . .
Corridor Selection. . . . . . . . . . . . .
Alignment Development . . . . .
Reconsideration of the Approved Alignment . . .
Analysis of the Planning Process. . . . . . . .
Conclusion. . . . . . . . . . . . . . . . . . .

vi

.4
KO

d—J—l—l—J—l—l—l
wNNN—‘d-JO O
-4\OU'INO\OOLO

Page
CHAPTER SIX. . . . . . . . . . . . . . . . . . . . . . l34

Conclusions. . . . . . . . . . . . . . . . . . . . l34
Findings . . . . . . . . . . . . . . . . . . . 135
Research Conclusions . . . . . . . . . . . . . l37
Implementation Approaches. . . . . . . . . . . l39
Types of Compliance. . . . . . . . . . . . . . l4l
Complying with NEPA. . . . . . . . . . . . . . l43

LIST OF REFERENCES . . . . . . . . . . . . . . . . . . I45

vii

Table

aha-w
wN-a—a

LIST OF TABLES

List of MDOT Route Location Factors.
No Significant Difference Ratings.

Number Rated with Selected Alternatives:

Model of Route Location Decisions.

viii

 

LIST OF FIGURES

Figure Page ]
5.l I- 69 Study Area. ll7 !
5.2 I- 69 Alternative Corridors . . 118
5.3 I- 69 Final Alternative Alignments. 123a
5.4 I- 69 Reconsideration Alignments. l26

 

ix

CHAPTER ONE
NEPA AND ITS IMPLEMENTING REGULATIONS

Introduction

 

In l969 Congress passed the National Environmental
Policy Act which has been labeled the first "comprehensive
national commitment to protection of the environment"
(Caldwell, 1979, 5000l). In addition to providing a policy
statement which recognized environmental values, Congress
included a policy for implementing the substantive direc-
tives of this law. These have become known as the "action
enforcing" provisions of the Act. The action enforcing
provisions mandate a decisionmaking reform by requiring
environmental impact statements for any major federal
action which significantly affects the quality of the human
environment (NEPA, Sec. 102, (2)C). In essence NEPA calls
for federal agencies to follow a policy which describes how
they should make decisions. The model of choice, which
forms the basis of the environmental impact statement, is
rational and analytical. It stresses a "systematic,
interdisciplinary" approach which is akin to comprehensive
planning. The underlying concept is that comprehensive

planning would insure the consideration of environmental

factors along with all other variables. In this way better
decisions would lead to a better environment.

After over a decade of experience with NEPA, there is
no clear consensus among students of environmental policy
on its success or failure (Liroff, l980; Caldwell, l979;
Fairfax, I978). This can be in part attributable to the
fact that NEPA's goal, to encourage productive and enjoy-
able harmony between man and his environment, is general
and ambiguous, and that performance toward reaching such a
goal is difficult to measure. While many agree that NEPA's
effect on the landscape is hard to assess, researchers
have not failed to examine federal agency implementation
of NEPA. Some recent criticism has pointed out problems
in carrying out the objectives of the Act. The action
enforcing provisions do not seem to be producing compre-
hensive decisions (Friesma and Culhane, l976; Hill and
Ortolano, 1978). Although critics have been quick to point
out these inadequacies in the way agencies have complied
with NEPA, few have offered any explanation for the
bureaucratic behavior they have observed.

If one wants to determine whether NEPA is effective in
improving the quality of the environment, and if one wants
to improve the implementation of the Act, it is necessary
to know and understand how its present implementation is
reflected in agency decisions. The focus of this study is

on the implementation of NEPA's action enforcing provisions

from the perspective of a bureaucracy which is faced with

an inherently complex task. The basis of this perspective
is that, when the complexity of the human environment and
the limitations of human information processing capabilities
are considered, the assumptions of NEPA's implementation
model are unrealistic (Caldwell, l979, 50003). A gap exists
between what Federal agencies can accomplish and what is
required of them by NEPA. Therefore, what is observed is
the outcome of how bureaucrats cope with these unrealistic
demands. The objectives of this examination of NEPA
implementation are to formulate and test a model of agency
decisionmaking, consistent with the concept of bureaucratic
c0ping behavior, and to assess NEPA's impact on agency

decisionmaking.

Description of the Law

 

The Congressional intent behind NEPA has been the
subject of many articles and books. It is accepted that
one key reason for a national environmental policy is to
establish an umbrella policy statement for the protection
of the environment. Previous legislation had taken a
piecemeal approach. By establishing an umbrella policy
Congress felt that it could promote the consistent and
systematic application of the various pieces of environ-

mental legislation and bolster their consideration in

administrative decisionmaking at all policy levels -
legislative, program, and project. Section l02 which,
among other things, calls for the preparation of environ-
mental impact statements provides the integrating instrument
to satisfy this intent.

From the recollections of staff members and the record
of legislation it is reported that "the impact statement
was not intended merely to provide data on description, but
to force a change in the administrative decisions affecting
the environment" (Dreyfus and Ingham, 1976, 254). In order
to force change, the impact statement requirement brings
administrative decisionmaking under the public eye. The
impact statement provides a forum for public discussion and
administrative explanation of how environmental issues are
resolved in project decisions. The underlying incentive
for bureaucratic compliance is the threat of litigation.
Although it has been admitted by Congressmen and staffers
that the impact of NEPA could not be predicted, the objective
of NEPA is the promotion of a better environment through
better decisions, meaning more comprehensive decisions.
To insure better decisions, the decisionmaking process is
public.

NEPA contains three components: (1) a policy statement
for the environment, (2) a policy statement for how decisions
affecting the environment are to be made, and (3) the

establishment of a Council on Environmental Quality. All

three components make up a legislative package designed to
change previous government practice which had resulted in a

deleterious effect on the Nation's environment.

A Policy for the Environment

 

As an outgrowth of the environmental movement, NEPA
recognized the importance of environmental factors which
had previously been treated only in cursory fashion. Other
environmental legislation left gaps in environmental pro-
tection because they failed to consider the interrelatedness
of environmental affairs and how it could directly impact
project decisions. NEPA is an attempt to cover these gaps
with an umbrella policy and an implementation instrument
which focuses on decisionmaking. In giving recognition to
the impact of man on the natural environment, Congress stated
in NEPA,

...it is the continuing policy of the Federal
Government...to use all practicable means and
measures...in a manner calculated to foster
and promote the general welfare, to create
and maintain conditions under which man and
nature can exist in productive harmony, and
fulfill the social, economic, and other
requirements present and future generations
of Americans (Sec l0l(a)).

In making this policy statement Congress gave standing
to environmental concerns next to that of social, economic,
and technological values in federal government decision—
making. It also gave weight to the consideration for future

generations in the use of the environment. Congress

elaborated on these points by asking that the federal

government strive for beneficial uses of the environment
"without...undesirable and unintended consequences" (Sec.
lOl(b)(3)). To maintain the environment for future genera-
tions, federal decisions should show a ”balancing of
population and resource use which permits a high standard
of living through the enhancement of the quality of renew-
able resources and recycling of depletable resources". In
addition, important historic, cultural and natural aspects
of national heritage are to be preserved. Finally, there
is the added objective that "an environment which supports
diversity and variety of individual choice" is to be
maintained. In so doing the federal government is to adopt

the role of trustee for the environment.

A Poligy for Decisionmaking

 

To work with the policy statement for the environment
Congress included an implementation policy which describes
how planning decisions are to be made. This second compo-
nent of NEPA outlines a planning procedure for federal
agencies to follow. Although Congress left the drafting
of specific rules to the agencies, it established a general
policy that the planning process utilize a "systematic, and
interdisciplinary approach which will insure the integrated
use of the natural and social sciences and the environmental
design arts..." (Sec. l02(2)(A)). Methods and procedures

are to be developed "which will insure that presently

unquantified environmental amenities and values may be given
appropriate considerations in decisionmaking". The proof of
this consideration of environmental values is exhibited in a
planning report called an environmental impact statement.

Congress described in skeletal form the composition of
an EIS which is to accompany federal proposals for both
legislation and action. EIS's must be prepared for all
"major federal actions significantly affecting the quality
of the human environment” (Sec. l02(2)(C)). The EIS must
include the following information:

(i) The environmental impact of the proposed action,

(ii) Any adverse environmental affects which cannot
be avoided should the proposal be implemented,

(iii) Alternatives to the proposed action,

(iv) The relationship between local short-term
uses of man's environment and the maintenance
and enhancement of long—term productivity,
and

(v) Any irreversible and irretrievable commitments

of resources which would be involved in the
proposed action should it be implemented
(Sec. l02(2)(C)).

The final EIS also is to include comments of other
federal agencies which have jurisdiction by law or expertise
with respect to the environment involved, as well as of the
general public (Sec. l02(2)(C)). Although NEPA required the
preparation of an EIS, the law is not clear whether this
activity should be integrated into federal agency planning

procedures or be separate from it. Whichever is the case,

the purpose of EIS preparation is to demonstrate federal
agency consideration for environmental values in their

decisionmaking.

Establishing the Council on Environmentalgguality

 

In addition to passing a national mandate for the
protection of the environment Congress provided a body to
assume leadership for directing this policy. Title Two of
NEPA establishes the Council on Environmental Quality in
the Executive Office of the President. Its principle
purpose is to advise the President and report on the state
of the nation's environment. The Council's other tasks are
to review and appraise federal programs with respect to
NEPA policy, to recommend national policies for the quality
of the environment, and to conduct studies relating to
environmental quality. Finally, the CEO is one of the
designated recipients of agency EIS's.

In summary NEPA declared a national policy for the
environment. This policy supported existing federal
environmental legislation, and provided "all federal
agencies with a mandate to weigh environmental impacts in
their decisions" (Liroff, 1980, 154). In order to encourage
agency compliance, Congress instituted a "series of action-
forcing mechanisms". The principle mechanism is the
environmental impact statement, which is required of all

major federal actions significantly affecting the quality

of the human environment. The process, leading to the
preparation of the EIS, has been characterized as NEPA's
"internal reform" of administrative decisionmaking. Like—
wise, NEPA's other requirement of interagency review and
public comment is an external reform which pressures
administrative compliance (Liroff, 1980). Generally
speaking, NEPA is designed around these two reform ideas,
but whose implementation, as NEPA's early history shows,

presented difficult problems in compliance.

History of NEPA's Administrative Rulemaking

 

With the signing of NEPA by President Nixon in 1970,
not only did the United States gain an environmental policy,
but it also established the Council on Environmental
Quality. One of the Council's first and most important
duties was the development of administrative guidelines for
the implementation of NEPA. The direction and emphasis of
the CEO guidelines were a key factor in how federal agencies
perceived the Act and their duties to comply with it. As
the record of experience with NEPA grew, the CEO further
revised its guidelines in reSponse to agency behavior and
to judicial review. The Council was at the interface. In
his analysis of the CEO guidelines, Herbert Stevens states

that the guidelines

10

...are not simply parallel to and separate

from the judicial interpretation of NEPA,

but are instead a kind of hybrid creation -

an administrative-judicial gloss on the

statutory language of NEPA. Given the

advisory position of CEO with respect to

implementation of NEPA, as contrasted with

the position of those mission agencies of

the federal government, the Guidelines

draw their strength from their consolidation

of important cases under NEPA...(Wichelman,

1976, 275).

Relying on the NEPA case history where need be, the

CEO slowly and consistently tried to bring federal agency
behavior in line with the spirit and goals of NEPA. The
period from 1970 to 1978 shows the evolution of guidelines
from being narrowly focused on the "detailed" statement to
a broader concern about NEPA's integration in agency
decisionmaking. In 1978 the CEO issued Implementing
Regulations for NEPA which became effective in July, 1979;
these regulations were the culmination of this first period
of NEPA's history. Before a detailed description of the
Implementing Regulations, a brief history of the CEQ's
efforts will be given in order to put the Regulations in

their proper context.

NEPA's Early Guidelines

 

The history of NEPA's guidelines begins with Executive
Order No. 11514 in 1970 in which President Nixon delegated
the responsibilities given to him by the Act to the CEO.
The Executive order restated the requirements of the Act,

but emphasized the timely access to information for public

11

review. It implied "a policy...of increased public partici-
pation before the process of recommendation, review, and
approval was completed" (Andrews, 1976, 28). Second, it
emphasized the preparation of the "detailed statement",
required by Section 102 (2)(C). This point was stressed,
while no mention was made concerning the other requirements
of the Act. Andrews suggested that,

Perhaps in part because of this emphasis,

however, both CEO and the other agencies

focused from the onset almost exclusively

upon the procedures for producing these

statements and virtually i nored most of

the law's other provision Il976, 29).

Following the lead of the Executive order "interim"
guidelines were issued in April, 1970, which covered in
detail only the preparation of environmental statements.
These advisory rules dealt with the procedural aspects of
preparation and gave little guidance on substantive
compliance. This responsibility was left to the line
agencies, themselves. The interim guidelines were thought
to represent CEQ's policy strategy of "accommodation" rather
than confrontation with other agencies, and maybe explains
its lack of more specific direction (Andrews, 1976, 29).
Late in 1970 a subcommittee of the House Merchant Marine
and Fisheries Committee held congressional oversight hearings
on the implementation of NEPA. Some problems in NEPA

implementation were identified, and the Council's policies

dealing with disclosure of environmental statements and

12

comments were criticized. As an outgrowth of these hearings
the Council was forced to revise its guidelines "to require
disclosure of environmental impacts a reasonable length of
time before taking administrative actions" (Andrews, 1976,
33). Revised Guidelines were adopted in April, 1971.

Again, it can be said that the CEO dwelled on the
procedural aspects in the Revised guidelines. This time
the emphasis was on trying to build NEPA implementation into
agency decisionmaking. To meet this objective, agencies
were ordered to assess environmental impacts "as early as
possible, and in all cases prior to agency decision". In
addition, agencies were asked to specify the review processes,
and points within these processes where environmental
statements would be made available. Finally, at the urgings
of Congress a minimum time period for public review was
established. The review period for a QLEEE environmental
statement is ninety days and it is thirty days for a £1331
statement. In so doing the CEO tried to bring environmental
policy to a forefront position in agency decisionmaking, as
well as to alert interests outside the agency to the proposed
action.

The Guidelines were revised again in August, 1973.
During the intervening period, the CEO drafted a series of
memoranda which continued their attempt to improve imple-
mentation through more detailed procedures, and the growing

legal history. The CEO was consistent in moving toward

13

procedures which integrated environmental policy into agency
decisionmaking and new emphasis was placed on the other
requirements of the Act. Stressing NEPA's specific policy
objectives, the Revised Guidelines strengthened the concept
that agencies should utilize in systematic fashion the
natural and social sciences in their planning and decisions;
that environmental assessments should be conducted at the
same time as technical and economic studies; that where
competing factors, which are balanced with environmental
impacts, are involved in the overall decision, they are to
be mentioned within the environmental statement. This last
requirement broadened the idea of the detailed statement
beyond that of an objective, factual declaration. As
Andrews points out, "the language of NEPA required bgth an
objective declaration 33d evaluative judgment about the
merits of what was declared: a detailed statement not only
of the impacts and alternatives pg: 53, but also of the
"adverse effects" of a "proposed action" (1976, 37)". In so
doing, the Council expanded the scope of the environmental
statement to include more of an agency's decision calculus,
which likewise brought more information of agency decision-
making under public review.

The Revised Guidelines of 1973 remained in effect until
final regulations were issued in July, 1979. These final
regulations were the result of an eighteen month rulemaking

process which was initiated by President Carter's Executive

14

Order No. 11991. This order directed the Council "to reform
the NEPA process so as to reduce paperwork and make the
statutorily required environmental analysis more useful to
Federal decision-makers and the public" (Environmental Law
Reporter, 1979, 10005). In large part the Executive order
was another response to problems which arose in NEPA's
implementation history. Most of the NEPA related litigation
was due to agency noncompliance with environmental statement
preparation or inadequate preparation. Once the question of
which agency decisions require statements was resolved,
adequacy and actual utilization of EIS information in agency
decisions were the chief concerns of the CEO.

While the success or failure of NEPA in influencing
agency decisions has been near impossible to determine, some
are of the Opinion that:

The assumption that environmental quality
could be enhanced simply by increasing
information and agency disclosure, however,
has proved fallacious. Change in type and
amount of information available, without a
corresponding change in the manner in which
it is evaluated and used, simply has not had
significant influence on agency action (Yale
Law Journal, 1979, 599).

Those critics who adopt this view argue that up until
1979 the environmental statement was just a procedural tool,
and that there was nothing to influence agencies to alter
their mission orientation. For others, NEPA's performance

was a gray area. Certain case studies have shown NEPA to

be a determining influence, but for the most part, NEPA's

15

impact on the federal bureaucracy was seen to be gradual

and modest through procedural innovation (Wichelman, 1976,
263-300). The evidence of NEPA success, that is cited by
policy analysts, is usually in terms of project delays,
public controversy, and litigation. While modest changes

in agency decisions can be attributed to the reasons cited
above and other factors internal to an agency, such as new
personnel, special offices, and compatible agency missions,
it seems as though there was an absence of a strong link
between environmental information, generated by the NEPA
requirement, and the design of agency action early in the
planning process. A CEQ study of the first six years of
NEPA found that many decisions on the design of alternative
choices were based on formal and informal assessments before
an EIS was prepared (CEQ, 1976). This absence was compounded
by a lack of any follow-up review or monitoring of the
implemented action with respect to the predicted impacts

in the environmental statement. Although there was no
conclusive evidence on the impact of NEPA on agency decisions,
the impression that was left by assessments of the law's
implementation was that the process of EIS preparation was
not integrated into agency planning, and therefore remained

less effective in bringing about NEPA's substantive goals.

l6

NEPA Implementing Regulations

 

In trying to improve on the shortcomings of the Revised
Guidelines of 1973 which focused only on Section lO2(2)(C)
of the Act, the CEO issued Regulations in 1978 which had a
broader application of NEPA's procedural requirements. The
new CEO Regulations became effective on July 30, 1979.
Their stated purpose was to make the process more useful
through the reduction of paperwork, avoiding unnecessary
delays, and reaching better decisions. The CEO Regulations
attempted to respond to the criticism of NEPA implementation.
There were three major changes in the new CEO Regulations.
First, the Regulations were made binding on all federal
agencies. Previously, the Guidelines were only advisory;
now one set of standards is established. Second, the
Regulations apply NEPA policy to a wider range of agency
decision activities than just EIS preparation. The NEPA
policy extended from early planning through post-decision
reporting. Last, for simplicity and clarity federal
agencies are required to use a standard terminology and
format, as well as follow standard procedures (McDermott,
1979, 92-9). Standardization allows administrative parti-
cipants to know the schedule of events, and reviewers to
oversee the decision process. By formulating these three
changes, the CEO continued to strengthen its commitment to
the spirit of NEPA which calls for a systematic, inter-

disciplinary planning and decisionmaking approach.

17

Regulations for Implementing NEPA: A Description

 

The CEO Regulations are described in order of the
decisionmaking events of the NEPA process, but it should be
recognized that the process is flexible and allows for
planning adjustments in response to changing situations.
The Regulations can be divided into five stages: (1) early
planning, (2) draft EIS preparation, (3) review and final
EIS preparation, (4) record of decision, and (5) monitoring
implementation of the decision. They are discussed in that

order.

Early Planning

 

One of the stated purposees of CEQ's new regulations is
to "insure appropriate consideration of NEPA's policies" in
agency decisionmaking. To accomplish this purpose and also
to reduce paperwork and time delays in the decision process
a number of early planning or pre-EIS activities are
performed. Once an agency has a proposed project it must
determine whether an EIS is necessary. The decision to
prepare an EIS can be made in two ways: (1) by classification
of projects which normally require EIS's, and (2) by preparae
tion of an environmental assessment (EA). Implicit in both
methods is the determination of what is a "major federal
action", what are "significant affects" on the environment,

and what is considered the "human environment". Some

18

guidance in defining these forms is offered, but leave the
specifics to the particular agencies. The definitions do
not eliminate ambiguity.

If the proposed action "normally" requires an EIS, the
decision to prepare an EIS is simple. On the other hand,
if the pr0posed action does not fall in regular procedural
categories, the federal agency prepares an environmental
assessment (EA). An EA is a public document that serves to
discuss the need for the proposal, the alternatives to be
considered, and the environmental impacts. It is produced
with the consultation of other appropriate agencies. The
result of preparing an EA is to determine the need for an
EIS. If it is found that an EIS is not necessary by means
of categorization or by preparing an EA, the agency drafts
a "finding of no significant impact". Where an EIS is
required, the process of "scoping" is commenced.

With the start of the "sc0ping" process a lead agency is
designated, as well as cooperating agencies for the project.
The lead agency assumes the primary responsibility for
preparing the EIS. The sc0ping process is an early planning
step. This process begins with the invitation to "affected
federal, state, local,...and other interested persons" to
participate (40 CFR 1501.7). This procedure tries to answer
questions of who should be involved in the process, what
issues should be discussed in the EIS, how should the EIS be

prepared, and when will NEPA process and related decisions

19

be made. It is also possible for revisions to answers to
these questions to be made if changes are offered in the
proposed action. The project of the scoping process is a
plan of action for the preparation of the EIS and its use.
This plan should indicate the responsibilities of the lead
agency and cooperating agencies. It should also show how
this NEPA process activity fits with other proposals and
existing projects. Finally, all requirements for review
and consultation should be identified and analyses planned
for meeting them.

As a result, the scoping process can reduce paperwork
by avoiding extraneous material, involve interested parties
early in the process, and reach better decisions through
greater participation in the review procedure (McDermott,

1979, 103).

EIS Preparation

 

The CEO Regulations specify the form and contents of
environmental impact statements. Consistent with the spirit
of NEPA, one of the main purposes for preparing an EIS is
to:

...provide full and fair discussion of
significant environmental impacts and

shall inform decision-makers and the

public of the reasonable alternatives

which would avoid or minimize adverse

impacts or enhance the quality of the
human environment (40 CFR 1502.1).

The boundaries of this discussion are set during the

sceping process. EIS preparation is to provide the

20

information and analyses on which the discussion is to be
supported. The regulations stress, as an end purpose of the
EIS, that it will be used by federal officials "...to plan
actions and make decisions" (40 CFR 1502.1). Thus, the EIS
is an integral part of the NEPA decision-process because it
is a platform for discussion of pertinent issues to the
decision at hand, as well as a subject of review and
criticism.

Agencies first prepare a Draft EIS according to the
plan which is developed in the scoping process. The Draft
EIS is circulated for review and comment to other agencies
and the public. A Final EIS is written which includes
responses and alterations as raised in comments and
criticisms on the Draft EIS.

The outline of an environmental impact statement is
general. Its major sections are: (a) purpose and need,

(b) alternatives, including the proposed action, (c)
affected environment, and (d) environmental consequences

(40 CFR 1502). "Purpose and need" is a brief explanation

of the problem to which the agency is responding in its
proposal. The Regulations call the "alternatives" section
the "heart of the EIS" because it describes the alternatives
under consideration and their impacts in comparative form.
It should be written “...sharply defining issues and
providing a clear basis for choice among options by the

decisionmaker and the public". The information and analyses

21

for comparing alternatives comes from the remaining two
major sections. The third section described the environ-
ment of the areas to be affected or created by the proposed
action. Finally, "environmental consequences" section forms
the scientific basis for impact determination and alternative
comparison. The nature of the impacts discussed in this
section are outlined in Section 102(2) C of NEPA while the
new Regulations specify that the scope shall include:
(a) Direct effects and their consequences.
(b) Indirect effects and their significance.
(c) Possible conflicts between the proposed
action and the objectives of federal,
regional, state and local...land use
plans, policies and controls for the

area concerned.

(d) The environmental effects of alternatives
including the proposed action.

(e) Energy requirements and conservation
potential of various alternatives and
mitigation measures.

(f) Natural or depletable resource require-
ments and conservation potential of
various alternatives and mitigation
measures.

(9) Urban quality, historic and cultural
resources, and the design of the built
environment...

(h) Means to mitigate adverse environmental
impacts (40 CFR 1502.16).

As an examination of this list shows the NEPA require-
ments are an attempt to be comprehensive. The sc0ping

process of early planning should have narrowed the bounds

22

of this discussion and investigation, as well as identified
what will be emphasized. These four sections make up the
body of the EIS document. In addition to these sections,
the EIS is rounded out by a summary statement, list of
preparers, list of agencies, organizations, and persons to
whom copies are sent, and appendices.

The CEO Regulations also offer some guidance for certain
situations which can be faced by preparers of an EIS.
First, an EIS need not duplicate the discussion of a higher
level (broader) EIS which has already been prepared.
"Tiering" is acceptable. The lower level EIS need only
summarize the higher level EIS discussion. Second, when
relevant information is lacking or uncertainty exists, the
agency shall acknowledge this point. If the needed infor-
mation is essential and the costs of obtaining it "are not
exorbitant", the agency shall include it. On the other
hand, if the costs are exorbitant or the means of securing
the information are unknown,

...the agency shall weigh the need for
action against the risk and severity
of possible adverse impacts were the
action to proceed in the face of
uncertainty. If the agency proceeds,
it shall include a worst case analysis
and an indication of the probability
or improbability of its occurrence
(40 CFR 1502.22).
Third, the EIS must contain explicit references to the

sources which are relied upon with respect to scientific

methodology and accuracy. Fourth, while the NEPA process

23

does not require it, cost-benefit analysis can be incorpo-
rated in the statement by reference or appended. If it is
prepared, the regulations ask that there is a discussion of
the relationship between the cost-benefit analysis and any

analyses of unquantified environmental impacts.

Review and Comment

 

The Draft EIS is the first substantive document of the
NEPA process that is subject to outside agency review. The
purpose of the Draft EIS is to help garner comments about
the lead agency's proposal for action. The lead agency
must obtain comments from any federal agency which has
jurisdiction by law, or expertise to any environmental
impact involved (40 CFR 1503); and it is the duty of these
agencies to comment. Likewise, it can request comments
from state and local agencies and the public. The Regula-
tions ask that the comments be specific and offer alterna-
tive approaches to the ones that are criticized. Thus, it
becomes a requirement of both the lead agency and reviewing
agencies to participate actively in the NEPA process.

The preparation of the Final EIS begins once all
comments on the Draft EIS are received. All the comments
are assessed by the agency and a response in the Final EIS
is made. Responses can be made in a number of different
ways. They range from develOping and evaluating new

alternatives to explaining why certain comments do not

24

warrant any further agency response. After consideration
and responses are given to the comments generated by the
Draft EIS, the Final EIS, which includes these responses,
is circulated in the same way as the Draft. Either through
agency review or other contact, interagency disagreement is
always possible. In such a case where an agency finds the
proposed activity is "unsatisfactory from the standpoint of
public health, welfare, or environmental quality" and
resolution of this conflict with the lead agency can not be
reached, referral is made to the Council of Environmental
Quality. The Council takes authority to bring this situa-
tion to resolution (40 CFR 1504).

Record of Decision

 

In the early guidelines there were no provisions for
the utilization of EIS's in agency decisionmaking. The new
regulations expand to include decisionmaking and implementa-
tion. They explicitly state that agencies shall follow
procedures to make decisions in accordance with NEPA and by
requiring that:

...relevant environmental documents, comments,
and responses accompany the proposal through
existing agency review processes...
..The alternatives considered by the decision-
maker are encompassed by the ranoe of alter—
natives discussed in the relevant environmental
documents and...a1ternatives described in the
environment impact statement (40 CFR 1505.1).

Upon making a decision the agency prepares a record of

decision. The record of decision states the outcome of the

25

decision process. In addition the agency must communicate
the alternatives which were considered, and which one is
environmentally preferable. The agency shall:

...identify and discuss all such factors...

which were balanced by the agency in

making its decision and state how those

considerations entered into its

decision (40 CFR 1505.2).

While the Regulations do not require that the agency
select the environmentally preferable alternative, it has
to explain why this course of action was not chosen if
another option was selected (McDermott, 1979, 107). A
statement in the record of decision must also be made to
demonstrate that an attempt was made to minimize harm to
the environment in the selected alternative.

In order to follow through with mitigation measures and
other conditions proposed in the EIS the Regulations demand
that a "monitoring and enforcement program be adopted and
summarized" as part of the public record. The lead agency
is supposed to impose "appropriate conditions in grants,
permit, or other approvals" (40 CFR 1505.3) to assure
implementation according to the plan.

In summary, the record of decision is an attempt to
bring agencies to show how the EIS is used in their
decisionmaking, and it is a document to which they can
be held accountable in post-decisions actions.

The history of the CEO Regulations shows the evolution

of the formal operating procedures for implementing NEPA.

26

The direction of this evolution was to bring more and more
of federal agency planning and decisionmaking, which
affected the environment, into the public domain in order
to force the integration and utilization of NEPA policy in
that process. The major implementing innovation for this
policy is the EIS; the EIS is strengthened by the require-
ment of another innovation, the "record of decision”. In
addition to these required decision documents the Regula-
tions stressed consideration of NEPA policy early in the
planning process to influence the design and analyses of
alternative choices. Altogether the history of the
Regulations demonstrates a clear movement by the CEO to
incorporate NEPA into every stage of federal bureaucratic

planning and decisionmaking.

Conclusion

 

What can be gleaned from the history and criticism of
federal agency compliance with NEPA is that agency planning
and decisionmaking has been perceived by environmentalists
to have fallen short of expectations. This perception may
have been based on the public's and court's interpretation
of NEPA's requirements as a demand for a comprehensive
planning process and for the EIS to serve as evidence of
compliance. This is consistent with one of the underlying

concepts of NEPA - that decisionmaking would consider

27

environment factors in the decision calculus, and this would
result in better decisions and in a better environment.
While there is no consensus on what constitutes better
decisions or a better environment, critics have pointed
out, and it can readily be observed by reading an EIS, that
the results of agency planning under NEPA are seemingly
short-sighted. Criticism has been leveled at both the
substantive inadequacies of EIS's and the procedural
failures of some agencies in preparing them in a compre-
hensive and integrated fashion. Thus, the implementation
strategy of the Act, of internal reform of decisionmaking
and external oversight, is not producing the results as
effectively as drafters and environmentalists had hoped.
One of the reasons that has been offered to explain
part of NEPA's unimpressive performance is that the assump-
tions of the decisionmaking model are too hard to meet in
dealing with problems with environmental complexity.
Agencies like individuals have limited capacities and
resources to process all the necessary information for
reaching optimal decisions. Thus, no matter how effective
the decision strategy is, the result will be by definition
suboptimal. Any agency, which must adhere to NEPA, confronts

this implementation dilemma - federal agencies are under

 

pressure to comply with NEPA while it is impossible to do so.
From over a decade of experience with NEPA one can see

that the federal bureaucracy has responded to the

28

difficulties brought about by this dilemma. As the history
of NEPA guidelines and regulations show, the CEO has gone
through an iterative process of incrementally revising
administrative procedures. Each iteration had two purposes:
(1) to reduce ambiguity, and (2) expand guidelines to
include more of NEPA's policy - to show links between
pre-decision planning and the decision stage. By the last
revision in 1978 the NEPA Regulations fully specified the
decisionmaking process from initiating a study to evaluation
of the resulting project implementation. In spite of these
revised rules and also a decade of litigation in the court
system, the model of choice is still the same, but it is
tempered by the "rule of reasonableness". While this may
have lessened the real demands of the decisionmaking
assumptions of the Act, it has not eliminated the complexity
and ambiguity inherent in planning with the environment in
mind.

Some critics feel that there are better policy instru-
ments to promote a national environmental policy; more
powerful regulations, allow the court system to determine
policy, and others (Fairfax, 1978). While the debate
continues about the benefits and costs of NEPA's performance,
there has been little attempt to explain the results of
NEPA policy in terms of agency decisionmaking behavior -

to examine the decisions themselves.

29

The implementation dilemma, imposed by NEPA, presents
agencies with an impossible task. This raises important
questions: how have agencies responded, and what are the
policy consequences of NEPA's mandate? The impossibility
of compliance with NEPA pushes agencies beyond simple rule
following to developing strategies to deal with the
conflicting pressures faced during implementation. Agencies
can be expected to develop strategies which promote organi-
zational stability and their ultimate survival (Bartlett,
1973). Thus these strategies become significant determi-
nants of the overall success or failure of NEPA.

The research, which addresses these questions of policy
implementation of NEPA. is developed in the following
chapters. Chapter Two defines the research problem involved
in studying bureaucratic coping behavior, and outlines the
theoretical foundation for model development. Chapter Three
describes the planning procedures of the Michigan Department
of Transportation, and their decision process under NEPA
for highway route location projects. This example of MDOT
planning and decisionmaking provides an empirical base from
which a model of agency decisionmaking can be tested. The
data, collected from MDOT documents, concerns highway route
location decisions which are subject to NEPA requirements
to qualify for federal funding. Chapter Four discusses the
results of the hypothesized decision model with respect to

predictability and compliance with the spirit of NEPA.

30

Chapter Five is a case study of one controversial highway
route location project. The planning and decision process
of the MDOT on this one case is followed from beginning to
end from the perspective of the agency's planners and
engineers. This case study provides a description of the
bureaucratic environment within which decisions are made,
and a foundation for the assumptions of the empirical model.
Finally, Chapter Six draws on the case study and empirical
model results to make a statement about compliance in the

NEPA era.

CHAPTER TWO
RESEARCH PROBLEM

Decision Context

 

The guiding purpose of this study is to add to our
understanding of agency performance in situations which
require compliance with the National Environmental Policy
Act. While assessing the impact of the Act on the state of
the national environment may be an insurmountable task,
describing its impact on decisionmaking and the response of
the agencies may not be. This study is concerned with NEPA's
implementation policy for decisionmaking reform, and its
consequences for substantive policy. This study would like
to make statements about how decisions are made by imple-
menting agencies, and about how environmental variables are
used in planning and decisionmaking. The research task is
to formulate and test a positive model of agency performance
in the implementation of NEPA.

To simplify this research task, the decisionmaking
environment must be focused, and its characteristics
described. This study will be based on one particular
planning and decisionmaking setting, which has characteris-
tics that are typical of the agencies which face the

impossibility of complying with NEPA.

31

32

Level and Type of Decision

 

As a national policy for the environment, NEPA applies
its mandate on all departments and at all levels of federal
activity which may significantly effect the environment.

This study focuses on planning and decisionmaking which

takes place at the project level. The project level decision
is below program level decisions and above operational
decisions which implement a project. This is a choice
between competing alternative solutions to an agency problem.
Alternative solutions are the product of an agency's planning
process. Each solution is a program of action which is
designed to bring about a desired effect. In cases where
NEPA applies, these solutions have impacts on the human
environment in addition to the desired effect. The project
level decision is the last planning decision before an agency
implements a course of action.

Project level decisions are selected for this study for
a number of reasons. The first reason for studying them
is that these decisions have a direct impact on the physical
environment, and they represent the final policy outcomes
for which NEPA was designed to influence. A second reason
is that they are, more or less, independent of the confound-
ing influence of factors other than the consideration of
alternative solutions (i.e. independent of the choice of
which projects to pursue). Thus, project decisions by their

level and independence can be used to demonstrate how NEPA

33

is integrated in the decision calculus of federal agencies.
While a project level decision is not irreversible, it
represents a commitment on the part of an agency to a
specific course of action which will have an impact on
people, property and the state of the environment. A
project decision is perceived separate from higher level
policy and program decisions, and usually receives a different
public response. Project level decisions consider more
specific courses of action. Likewise, consequences and
impacts of action alternatives are more explicitly described.
The greater detailed nature of agency planning information
makes it easier for critics and outside parties to trace the
agency planning effort, and more importantly, to perceive
potential benefits and costs of project actions. NEPA
strengthens public, as well as interagency access to project
planning information. Public informational meetings,
public hearings, and EIS's give those outside the agency the
opportunity to follow and contribute to agency planning and
decisionmaking. This heightens the visibility of the project
decision process to the public at large, and especially
those parties that are likely to be affected by the project
decision.
This greater visibility and opportunity for comment
and criticism, while designed as an incentive for government
agency compliance with NEPA, makes most project level

decisions the center of controversy. While controversial

34

issues are raised at all levels of decisionmaking, those
raised at the project level, usually concern, in addition to
general questions of policy, specific aspects of planning
decisions. These are brought up by local group interests
and from a variety of individual perspectives. Each, it can
be assumed, is concerned about ones own self-interest. As a
result, the project level decision process is closely
scrutinized. This public pressure on the project sponsoring
agency not only inhibits it from ignoring its mandate to
plan comprehensively, but makes avoiding controversy almost
impossible. The project level decision has to be made in a
situation of inherent conflict (West, 1984, 341). Thus, the
agency must adopt a strategy in its planning and decision-
making which tries to minimize controversy and maximize its
planning goals. This provides a third reason to study
project level decisions; that is, a project level decision
is highly visible to special interests, and the decision
maker is held publicly accountable. Thus, what one observes
is a choice at the interface of an agency and the public

sphere.

Decisionmaker

 

Project decisions are made by federal officials with
the power delegated to them by their department head. In
cases where the sponsoring body is not a federal agency

(i.e. state and local government agencies), project decisions

35

are made by the appropriate decision body with the under-
standing that federal approval would be forthcoming. Here,
there is usually a close working relationship between the
federal funding agency and the project sponsor. The federal
agency assiststhe sponsor in following federal administrative
guidelines and qualifying for funding. Whichever is the
project sponsoring body, the decisionmaker is an agency head
official. An agency official must rely on a planning staff
to carry out the planning tasks leading up to the project
decision. Given the magnitude and complexity of most project
decisions, a planning staff makes numerous planning decisions.
It shapes the alternative choices and the estimates of their
consequences which are considered in the final decision.

The attitudes, preferences, and beliefs of those making these
planning decisions influence the way problems are solved.

At the same time, the structure of an agency's organi-
zation imposes some degree of control over the operations of
a planning staff to coincide with the wishes of the decision-
maker(s). It would be difficult to separate these two
influences and how they contribute to the final decision
without extensive interviews of all the individuals involved.
This approach has been tried by other researchers in an
attempt to reconstruct decision situations and to identify
determinants of organization behavior (Roberts and Bluhm,
1981). For the purpose of this investigation, the organiza-

tion, an agency, will be treated as the decisionmaker. The

36

project decision is viewed as the product of an agency
effort. In order to make this assumption, this study further
assumes the agency structure is stable, or any change is slow

and incremental.

Decisionmaking

 

The determinants of agency behavior are grounded in the
theoretic principles of how humans make decisions. It is
assumed that an agency acts in a rational way in seeking its
goals. Planning and implementing projects meet this purpose.
Project implementation, in the cases which will be examined
here, require approval that is subject to compliance with
NEPA. One, therefore, expects an agency to attempt to comply
with NEPA in the best way it knows how in the process of
meeting its goals. This means that an agency simplifies the
decision task into a manageable process, and it simplifies
based on its perception of the political situation it finds
itself in.‘ The agency defines its situation in terms of its
survival (Downs, 1967; Niskanen, 1971; Bartlett, 1973).

The project decision is made from the information
generated by an agency's planning process. It tries to
determine the costs and benefits of each alternative solution
under consideration. While an agency may consider as many
factors as it feels necessary for it to make a project

decision, NEPA requires that it consider environmental

37

factors along with social, economic, and technical factors.
The NEPA guidelines and regulations have identified the
content of these sets of factors; it also asks that they be
comparable. NEPA also requires that all information used to
make a decision be reported. This includes other supporting
documents, along with an EIS. In order to meet these
requirements an agency needs to translate its goals and
knowledge of the human environment into operational terms

for reporting and decisionmaking. It is further assumed that

a.) I... ;_T m.“~—-—-—g‘m

a correspondence exists between an agency's general statements
and its operational terms. Finally, given that agencies are
rational actors, it is expected that a relationship exists
between the project decision and characteristics of the
selected alternative.

The decision and decision documents are the outgrowth
of agency planning and reflect agency policy. More impor-
tantly, this set of information demonstrates how an agency
coped with the complex tasks of planning and complying with
NEPA. The word "coped" is used here to emphasize the point
that an agency faced difficult tasks in completing its
decision process. It is confronted with a complex problem,
and it has limited capacity and resources to allocate to
find the best solution. As Roberts and Bluhm point out,
"Reality is unpredictable: most actions do not have exactly
foreseeable consequences." This is due in part to the

inherent "randomness" involved in the behavior of social

38

and natural systems, and to "our ignorance of the world"
(Roberts and Bluhm, 1981, 18). Imperfect knowledge of the
real world impairs the human decisionmaker's ability to
locate the set of all possible alternative solutions to a
problem. Similarly, man's limited capacity to process
information is dwarfed by “the full range of conceivable
relevant consequences of most decisions" (Roberts and Bluhm,
1981), and by the enormous number of calculations necessary
to determine the shape and magnitude of these consequences.
Herbert Simon has aptly described this human decisionmaking
predicament as the principle of bounded rationality:

The capacity of the human mind for formu-

lating and solving complex problems is

very small compared with the size of the

problems whose solution is required for

objectively rational behavior in the

real world - or even for a reasonable

approximation to such behavior (1957,

198).

The circumstances of complexity and limitations in
information processing hinder the human decisionmaker from
reaching the requirements assumed by classical rational
behavior. In the case presented here, an agency, faced with
a complex problem, copes; that is, it adopts a strategy
against complexity which dictates how it allocates its
decision capacity and resources, and how it selects a
solution. If one wants to predict the outcome of agency

behavior, one needs to understand that nature of its coping

strategy in light of theories which take the limitations of

39

the decisionmaker and the complexity of the problem into
account. This situation is better described by Simon:
The classical theory is a theory of man
choosing among fixed and known alterna-
tives, to each of which is attached known
consequences. But when perception and
cognition intervene between the decision-
maker and his objective environment, this
model no longer proves adequate. We need
a description of the choice process that
recognizes that alternatives are not given
but must be sought; and a description that
takes into account the arduous task of
determining what consequences will follow
on each alternative (Simon, 1959, 272).
Understanding agency behavior is enhanced by a model of
man which considers human limitations in processes of search,
computation, and choice.

Despite the complexity of the problem situation, a
project sponsoring agency carries out a planning process and
makes a decision. How does an agency come to terms with
complex problems? This study adopts the position that an
agency develops a strategy. which is consistent with the
principle of bounded rationality, for c0ping with the
ifluaossibility of NEPA, and which is shaped by the political
arwena of the agency. In the following discussion a descrip-
t1<>n of boundedly rational behavior and a description of

bur~eaucratic politics are presented.

40

Boundedly Rational Decisionmaking

In addressing his criticism at the rational actor model
of decisionmaking, Herbert Simon set the stage for a modified
rational model of man. His approach concerns itself with how
man operates in decisionmaking, rather than how he should.

As a consequence of the principle of bounded rationality a
decisionmaker is required to simplify the real decision
situation in order to deal with it. In so doing Simon
introduced the notion of a "real life decision":
A real life decision involves some goals
or values, some facts about the environ-
ment, and some inferences drawn from the
values and facts. The goals and values may
be simple or complex, consistent or contra-
dictory; the facts may be real or suppressed,
based on observation or on reports of others;
the inferences may be valid or spurious. The
whole process may be viewed, metaphorically,
as a process of "reasoning," where the values
and facts serve as premises, and the decision
that is finally reached is inferred from these
premises (Simon, 1959, 273).

A real life decision is a broader everyday sense of
rationality, which includes more than calculating the results
of a given set of alternatives. This broader concept is
equally concerned with the process of reasoning. The
process of reasoning can be characterized by the cognitive
structure of the decisionmaker, the formulation and content
of the decision premises, and the logic of the inference
process (Ostrom and Job, 1982, 16). This characterization

of the problem is managed, and choice simplified by the

 

41

human decisionmaker consistent with his limitations.

Cognitive Structure

 

One of the conditions of operating under the principle
of bounded rationality is that alternative solutions are not
given. They must be sought. Thus, how a decisionmaker
perceives his environment and goes about simplifying it
take on important meanings in understanding decision
behavior and in the quality of the decision. To begin with,
Simon states that decisions tend to be made in terms of the
highest goals that are operative:

The operative goals provide the seed around

which the (decisionmaker's) simplified model

of the world crystallizes. He takes into

account those matters that are reasonably

directly related to these goals and discounts

or ignores others (1976, XXXV).
The decisionmaker exercises selective perception; he collects
a limited amount of information with a minimum of depth.

A project sponsoring agency establishes lines of
investigation and communication consistent with its selective
perception. These channels of information bring particular
pieces of the decision problem to the agency's attention,
while shielding it from others. The complexity that still
gets through to the agency by way of these selective channels
is analyzed only partially and incompletely (Simon, 1976,
xxxv). Selective perception supplies the input which under-

goes the process of simplification. The agency, as decision-

maker, realizes that the world is made of "more stuff" than

42

it perceives, but it is,
...content with this gross simplification
because (it) believes that the real world
is mostly empty - that most of the facts
of the real world have no great relevance
to any particular situation (it) is facing
and that most significant chains of causes
and consequences are short and simple
(Simon, 1976, xxix-xxx).

The process of simplification, as mentioned above,
involves representing the decision problem in very general
terms that are an incomplete accounting of the real world.
In addition, the decisionmaker uses and monitors only a
small and relatively fixed number of factors. The choice
of which factors are utilized is determined by the experi-
ence, knowledge, goals, and outside pressures of the
decisionmaking agency. In an agency working under NEPA,
administrative regulations spell out a general set of factors,
while outside interest groups may mention those and pin-point
others of a more specific nature. At a bare minimum, the
agency must report on those cited in the law, but the
principle of bounded rationality would lead one to believe
that factors actually vigilantly monitored are fewer in
number. Through simplification, an agency tries to elimi-
nate, as much as possible, any inherent variety in the
problem situation (Steinbrunner, 1974, 56), and maintain a

stable cognitive structure which conditions project

decisions.

43

Decision Premises

 

A decisionmaker's cognitive structure provides a way to
avoid or reduce the complexity of the problem situation,
and lessens the amount of resources used to gather infor-
mation. It also offers a clue to the form of decision
premises. They are composed of a small set of essential
variables that are measured in a simple way. The decision-
maker monitors only this set of variables until, for some
reason, action taken in response to these variables fails
consistently. These variables are the factors that
influence the decisionmaker's evaluation of the problem
at hand.

Simon identifies two types of premises that underlie
decision behavior. They are factual, and 1313; premises.
In organizations, one is also concerned with the concept of
a rglg, and the premises that make it up. Knowing a
decisionmaker's premises is almost tantamount to predicting
behavior.

Before hypothesizing the content of decision premises,
this study looks to NEPA and its guidelines for decision
premises, and evaluates their congruence with the prin-
ciple of bounded rationality. In theory, Congress, by
enacting NEPA, tries to control decisionmaker behavior by
specifying the decision premises for federal agencies.

This discussion proceeds by reporting what NEPA offers with

respect to factual and value premises.

44

Factual Premises. NEPA Regulations, "Part 1505 - NEPA

 

and Agency Decisionmaking," instruct agencies to adopt
procedures to ensure that decisions are made in accordance
with the Act. These procedures shall designate, "the
major decision points for the agency's principal programs
likely to have a significant effect on the human environ-
ment and assuring that the NEPA process corresponds with
them" (40 CFR 1505.1 (b)). The two important phrases are
"significant effect" and "human environment." The Regula-
tions define the human environment to be:

...interpreted comprehensively to include

the natural and physical environment and the

relationship of pe0ple with that environment

...When an environmental impact statement is

prepared and economic or social and natural

or physical environmental effects are inter-

related, then the environmental impact

statement will discuss all of these effects

on the human environment (40 CFR 1508.4)

While the Regulations are written in very general terms
the environment it speaks of is grossly characterized by
natural physical, social, and economic features. Agencies,
who make project decisions, are interested in changing the
state of the human environment to a better position.
Subsequently, their concern is with the outcomes or
"effects" of project choices. Parties, outside of these
agencies, are likewise interested in project "effects" for

different reasons; on how project outcomes alter their

welfare.

45

Project effects add a second dimension to the environ-
mental features. The Regulations state that "effects" is
used synonymously with impacts. There are two categories
of effects: (1) direct - caused by the action and occur
at the same time and place, and (2) indirect — caused by
the action and are later in time or farther removed from
the action not only by physical distance, but by distance
in the chain of interactions. Effects can be beneficial
or detrimental. Thus, the factual premises of the project
sponsoring agency structure information about the problem
environment in terms of natural (biological), physical,
social, and economic systems, and whether the project
action will have direct or indirect effects on each system.

From what has already been mentioned about cognitive
structure and human limitations, these factual premises
present a fairly large set of information requirements to
the decisionmaker, more likely a rational actor model. The
NEPA Regulations identify fairly general premises, and
instruct the agencies, themselves, to construct specific
procedures to articulate the factual premises in more
operational terms, geared to their particular missions.

This study assumes that the essential variables are
among those for which information is gathered and reported
in environmental impact statements and in subsequent reports
which include EIS results. Information is reported on a

number of factors. These factors are operationalizations

46

of the effects or impacts on the physical, biological, and
social systems which characterize the environment. It is
expected that an agency focuses on only a few factors which
are perceived to be important impacts.

Value Premises. While factual premises established the

 

nature of the information a decisionmaker collects and
previews, value premises are necessary in order that this
information can be compared and evaluated. Again, if one
looks to NEPA and its implementing regulations for guidance
on value premises, one finds that an agency shall:

Identify and develop methods and procedures

...which will insure that presently

unquantified environmental amenities and

values may be given appropriate considera-

tion in decisionmaking along with economic
and technical considerations; (Sec. 102.

(2)(B)).

The law does little to reduce the ambiguity about how
values should be incorporated in the decision process, but
points out that environmental factors (meaning natural and
physical) should be raised to be on a par with other consi-
derations. It should be remembered that NEPA was proposed
as a decisionmaker reform. Although the Regulations are
ambiguous on this point, it tries to alter agency value
premises which historically adopt a mission orientation.

A review of NEPA and NEPA Implementing Regulations does
fig; reveal fact and value premises in a form conducive to
easy agency utilization. NEPA does not clarify how an agency

is expected to comply with the law. The factual premises

47

that can be garnered from the law run the gamut of possible
areas, i.e., natural, social, economic, and built environment,
in terms of both direct and indirect effects. In other

words, something must be said about any perceivable impact.
Likewise, NEPA provides little guidance on the equally

crucial value premises. NEPA demands only the consideration
of environmental factors with other factors. For an
implementing agency this can be interpreted to mean equal
effort of investigation of environmental impacts with

other impacts.

The Implementing Regulations of 1978 try to narrow the
focus of the factual premises procedurally. This is
accomplished through the "scoping process," which is
described in Chapter One. The success of the sc0ping
process is dependent on the assumption that the partici-
pants in the process have some notion of the spirit of
NEPA and some idea of how to decompose the decision
situation consistent with the spirit of NEPA and the
mission of the project sponsoring agency. Clearing the
ambiguity of the value premises is in the hands of the
project sponsoring agency and the EIS reviewers.

3913. A federal agency is part of the government
structure; it is appr0priate, then, to picture it in an
organizational context. An agency's behavior is influenced
by the organizational environment. The nature of this

influence has been incorporated by administrative theorists

48

in the concept of "role." Each agency of the federal
government has a role that has been designed in legis-
lation and executive orders. A role is an organization's
way to evoke particular or desirable behavior from its
members. "A role must be understood not as a specific,
stereotyped set of behaviors but as a program...For
determining the courses of action to be taken over the
range of circumstances that arise" (Simon, 1976, 269). It
specifies some of the fact and value premises which are
used by an agency to make its decisions. As it has already
been pointed out, NEPA introduces and establishes a new
set of decision premises from those previously held by
federal agencies. By definition NEPA altered federal agency
roles to reflect the national environmental policy mandate.
Thus, all federal agencies and "designated" agencies operate
with fact and value premises, and a particular agency in
the federal structure will hold the same premises over time.
From the perspective of this study, the concept of
"role" brings with it the characteristic of stability to
understanding agency behavior. By assuming that an agency
ad0pts, one can expect the agency to respond to similar
situations of complexity in the same manner. Further, one
expects the agency to use only those factors that are
considered essential. Finally, the relative importance of
these essential variables will remain constant. As a

result of assuming an agency maintains its role, any

49

differences that one might ascribe to personality changes
in an agency are, in fact, a result of differences in the
perceived environment (Ostrom and Job, 1982, 23-24). Thus,
agencies play a role in the structure of government which
leads to the expectation that they are relatively stable
over time. How agencies perceive their role and the
premises defining that role depend on their relationship

with the political environment.

Decisionmaking Logic

 

The "process of reasoning" is culminated by the selec-
tion of a course of action from a set of alternative
solutions. With consideration given to the complexity of
the decision problem and his own limitations, the decision-
maker makes a choice. Operating according to the principle
of bounded rationality, the decisionmaker has formulated
rules of inference which make the choice process manageable.
It can be called a strategy of decision, which, more
importantly, describes how decision premises affect
decisionmaking. By definition, changing the decision
premises, as NEPA tries to do, influences the whole process
of reaching a decision. If it is successfully implemented,
NEPA alters all agency decision activities leading to
choice. In contrast to the "classic" rational actor, the
boundedly rational decisionmaker must search for or design

alternative solutions, in addition to using a decision rule.

50

The policy literature, for the most part, has made an
artificial distinction in treating the search process as a
separate step in decisionmaking; where in reality, it is
"usually performed simultaneously with the estimation of
the effect and merit of each alternative" (Brewer and
DeLeon, 1983, 82). Search and choice are interrelated.
Alternatives are generated by a limited search process,
and a decisionmaker usually follows simple rules. As
alternatives are found, they are evaluated. When a solu-
tion meets the decision criteria, the choice is accepted,
and the search is stopped. The number of requirements
relied on is small. Based on the decision premises, one
focuses on a limited set of factors and ignores all others.
Relatively few alternatives are produced by the search
process, and the search is not exhaustive (Janis and Mann,
1977, 29-30).

Satisfying replaces the goal of Optimizing which
governs the "classic" rational actor. A decision rule,
consistent with the goal of satisficing, establishes a set
of requirements that a possible solution must surpass to
be selected. The decisionmaker limits attention to finding
out whether an alternative solution falls above or below
minimum cutoff points for each requirement (Janis and Mann,
1977, 30). In the event that a satisfactory solution is
not found, attention is directed to modifying the criteria

that the solution must meet. The criteria is relaxed so a

51

solution can be found (March and Simon, 1958, 197-180).

Generation of Alternatives

 

Consideration of the generation of alternative solutions
has received relatively little research attention by
decision theorists in comparison to other decision stages,
but no one doubts the vital role this stage plays in the
performance of decisionmaking. The area has been investi-
gated mostly by cognitive psychologists and later by those
concerned with human information processing. Likewise, in
the study of agency implementation of NEPA the design of
options can not be overlooked because it is interwined with
the final outcome.

The focus of attention in the search process follows a
general sequence. The search and design process must have
a starting point. With rare exceptions problem solvers
begin by using commonly held or accepted ideas. Agencies
have a history from which precedents can be drawn. With
respect to substance, attention is paid first to "variables
that are largely within the control of the problemsolving
individual or organizational unitJ'(March and Simon, 1958,
197-180). Second, if an acceptable solution is not found,
attention will be directed to changing other variables
that are not under direct control. In order for the
process to be carried through, some type of evaluation

must be made to determine acceptability. It may be

52

intuitive or informal. Although it is quite obvious, it

is important to realize that analysis in the search and
design process preempt formal evaluation in decisionmaking.
This is in contrast to the ideal rational model in which
the development of alternatives is separate and distinct
from the succeeding stage of formal evaluation, and in
which judgment is deferred during search.

By moving away from this ideal, understanding decision
behavior involves consideration of all earlier decision
activities leading up to formal evaluation and choice. In
a pilot study on the design of alternatives in organiza-
tions, Ernest Alexander conceptualized and presented
qualified findings on the interaction of decision stages.
Alexander's study found that the process of alternative
development is inhibited in generating a broad range of
options for formal evaluation. An intuitive, or semi-
formalized evaluation process occurs before formal evalua-
tion even begins. A preliminary review is linked with
design. The surviving alternatives come through a process
of blending or elimination until they reach a domain of
acceptability. As one might expect, the alternatives
produced by this process are likely to be only marginally
different. This process involves a series of decisions
which seem to be based on a single important value, rather
than a formal tradeoff (Alexander, 1979, 398). Importantly,

this supports the idea that an early review of alternatives

53

narrows down the options which are finally reviewed.
Accepting the idea that the decisionmaker narrows or
focuses the range of alternatives before any systematic
analysis is done, it is important to know how this process
is accomplished. Alexander reports that there is a rapid
convergence of options, both in number and range, before
evaluation. In the review process, criteria without
cumbersome data needs or complex judgmental demands get
more attention; judgmental factors tend to be undervalued
(Alexander, 1979, 396-398). Therefore, any discussion of
decision rules should incorporate the inferences used by
the decisionmaker in both the design and formal evaluation

stages.

Rules of Inference

 

By stepping back from the picture of decisionmaking
shown above, one can take a complete view of the process
offered by theorists of bounded rationality and the
findings of Alexander. The selection process begins with
the search for alternatives and ends with a formal evalua-
tion. In a truly Cybernetic model of choice there is no
distinct evaluation, but one is made here because the
evaluation stage is recognized formally in administrative
rules and procedures. Consistent with boundedly rational
behavior, a decisionmaker, facing a complex problem, can

go through a number of iterations of "search and modify

54

expectations" before finding a satisfactory solution (Mack,
1971, 175-9). The iterations before final selection are
done to either reduce a prohibitive number of or to increase
the availability of options. In an organizational context
decisionmakers usually enter the formal evaluation stage
with a set of choices to consider. By this time, as
Alexander reports, the alternatives are in a domain of
acceptability. Thus the final formal evaluation is the

last iteration to select a solution.

The rationality in the process lies in the consistency
of the inferences made in each iteration leading to the
final choice. At each point in the process the decision-
maker limits attention to finding out whether an alternative
meets a set of requirements. Thus each decision to elimi-
nate or blend an alternative is made according to a
decision rule, as is the final selection.

The findings of Alexander's pilot study substantiate
the importance of pre-selection activities, especially
search, to the final outcomes. Therefore, it is expected
that pre-selection activities have a significant impact
on the implementation of NEPA by project sponsoring agencies.
NEPA, in its attempt to impose rationality on agency
decisionmaking, puts the project decisionmaker in a
peculiar situation. Although NEPA imposes an "ideal"
rational decisionmaking process on decisionmakers,

decisionmakers respond with decision strategies consistent

55

with the principle of bound rationality. The successful
implementation of NEPA would be in the form of a decision
strategy which would include environmental factors in the

search and design stage, as well as in formal evaluation.

Agency Compliance with NEPA

 

It was pointed out earlier that Congress was fortunate
to have specified an implementation mechanism as part of
NEPA's mandate. The primary component of that mechanism
is the environmental impact statement which tries to bring
comprehensiveness and rationality to the administrative
process. While the purpose of institutionalizing this in
the process is better decisionmaking, NEPA's mandate is
also one way to control the discretion delegated to public
agencies (West, 1983, 326). Its intuitive appeal is its
emphasis on objectivity and efficiency. The mandate
assumes that the statutory goals are clear, and the imple-
mentation is "tractable" (Mazmanian and Sabatier, 1983).
If this is not the case, not only is implementation going
to pose a problem for the implementing agency, but it poses
an equally difficult problem for the overseer to determine
compliance. Thus, the implementation dilemma, associated
with NEPA and faced by agency bureaucrats, carries over to

affect the review and determination of compliance itself.

56

The discussion of boundedly rational behavior suggests
a model of how agencies are expected to operate in carrying
out NEPA, but it assumes that the agencies' goals are
specified. What is being offered here is the idea that
agencies simplify the complexity of their decision problem
in a boundedly rational fashion, and they do it according
to premises derived from their goals. Since NEPA fails to
provide useable premises, a theory is needed to prescribe
their origin. Agencies do not work in a vacuum but in a
very political environment. The discussion turns to
describe the bureaucratic politics behind implementation
to determine the make-up of agencies' essential variable

set which governs, in part, how agencies cope.

Bureaucratic Politics

 

Until now the discussion has centered on the way agencies
identify, break-down, and generate solutions to problems
they face, and how NEPA attempts to reform this process.
This organizational approach, based on the principle of
bounded rationality, placed emphasis on routines or programs
that an agency has for problem solving and implementing
courses of action. These programs bring information to
decision points in the process of generating alternatives
and analyzing the consequences of proposed actions. What
this organizational approach fails to consider is the

political environment of implementing agencies which

57

determines the nature of the decision premises. This is
especially crucial in the case of NEPA in which the
decision premises are ambiguous.

Without clarity and feasibility of policy, NEPA's
implementation and the decisions, which it is designed to
affect, can not be made by a single, rational choice, but
"by the pulling and hauling that is politics” (Allison,
1971, 144). From this point of view agency compliance is
not the result of reaching a pre-determined solution, but
more a result of agency compromise and conflict resolution
among the many possible interests effected by the decision.
The decisions that are made and the action which follows
are not based solely on supporting reasons or organiza-
tional routines, but also on the power and skill exercised
by the actors in the decision process.

The important questions, asked by a bureaucratic
politics approach, are offered by Allison:

1. What are the existing action channels
for producing actions on this kind of

problem?

2. Which players in what positions are
centrally involved?

3. How do pressures of job, past stance,

and personality affect the central

players on this issue?

4. Where are foul-ups likely? (1971, 257)
These questions show that the important determinants of

choice involve greater details of individual leaders or

58

agencies of government and the politics among them.

A Bureaucratic Strategy for NEPA Compliance

 

The political bureaucratic picture just described can
be perceived within the structure of how organizations
make decisions in a boundedly rational way. Project
sponsoring agencies can be seen as actors with specific
desires and goals which they would like to attain. These
agencies bargain with other agencies which are likewise
pursuing their interests. Goal conflict is inevitable
given government's shared responsibilities. Governmental
choices are the result of bargaining along regularized
channels (Allison, 1971, 162-180). The bargaining game is
played by actors with different goals, different positions
and authority, different demands, and unequal levels of
influence. Given this environment, what can agencies be
expected to do.

Perceiving an agency as an actor in a bargaining
situation, an agency is expected to serve its own best
interest. In the private sector this amounts to trying to
maximize profit. Because the public sector does not work
to profit, another objective is necessary. Given that the
bureaucracy serves at the grace of the public, governmental
agencies have been thought to try to enhance and promote
their survival. Thus agencies can be expected to develop

a strategy or plan of action which protects their survival

59

in the process of implementing policy.

In the case of being called upon to implement NEPA,
some agencies ignored the law until they determined that
failure to do so would be detrimental to their budget,
on-going projects, and their very existence. Being pressured
by the political environment, agencies had to face the
impossibility dilemma described earlier. Forced to comply,
it is hypothesized that they responded in the best way they
knew, given their self-interest in survival.

Agency response can be characterized in the following
way:

1. Agencies try to maintain control over
as much of the decision process as possible;

2. Agencies try to reduce conflict by
anticipating and avoiding problem issues;

3. Agencies try to comply with parts
of the law that are easiest to meet;

4. Agencies try to restrict choice in
decisions which they can not control.

These characteristics of agency behavior allow the agency
to meet its own mission, and Survive. What an agency is
expected to do is develop in its own way a strategy of
decisionmaking which exhibits these tendencies of goal

achievement with minimal conflict.

60

Conclusion

 

The principle of bounded rationality and the work of
Simon, March and others in developing this principle has
influenced ideas of human decisionmaking to consider broader
dimensions of this human activity, while relaxing the
assumptions of the classical rational actor model. A
decisionmaker, contemplating a "real life“ decision, adopts
simplifying rules and procedures to cope with the complexity
of the problem situation. The way a human decisionmaker
simplifies the complexity of the problem is rooted in the
makeup of his cognitive structure and in his decision
premises.

Administrative agencies of the federal government face
the complexity every day in planning and decision activi-
ties. Agencies simplify the problem of choice like indivi-
dual decisionmakers. The enactment of NEPA is an attempt
to alter agency decision premises to reflect a national
policy for the natural environment. In essence, it
redefined the roles of federal agencies. Not only do
agencies strive to meet their specific agency mandates, but
now they must do so in a manner which enhances nationally
desirable qualities of the environment for future genera-
tions. By redefining agency roles, and thus changing
decision premises, the drafters of NEPA expect to observe

decisions which are based on the integration of this

61

broader concern with agency goals. To back up this change,
NEPA promotes decisionmaking which is comprehensive and
rational in the implementation of its substantive policy.
Although this decisionmaking process is being imposed
by NEPA, implementing agencies operate in boundedly
rational fashion. Given an understanding of "real life"
decisionmaking, one expects to find a relationship between
agency decisions and characteristics of the selected
alternatives. That is, agency decisions should be a
function of the social, economic, and environmental
attributes of the solutions; all attributes are not
considered by a boundedly rational agency but only a small
number of essential, but simply defined factors. Putting
the agency in its political environment, suggests that
these essential factors are determined by compromise
between the actors involved in the process. Compliance is
bargained in the political arena. The essential factors
reflect the strategy adopted by the decisionmaking agency.
Simple algorithms are used to minimize the calculations
of consequences whether informally in searching for solu-
tions or formally in evaluation. There is a correspon-
dence between the essential factors and an agency's set of
goals. Thus, agencies use standard operating procedures
but they are strategic, politically directed. The successful
implementation of NEPA mandates would mean the inclusion of

environmental factors in an agency's goals and represented

62

in its essential factor set.

This study attempts to develop a model of agency
decisionmaking under NEPA which reveals the nature of this
relationship--the underlying decision criteria. In the
chapter that follows, this model of agency decisionmaking
under NEPA is further developed and operationalized. An
estimation procedure is described for estimating an agency's
decision criteria based on the public information from

environmental impact statements and supporting documents.

CHAPTER THREE
DECISION MODEL: HIGHWAY ROUTE LOCATION
DECISIONMAKING IN MICHIGAN

In the preceding discussion it is argued that one must
consider the complexity of the decision problem and human
limitations in order to understand human decisionmaking
behavior. Grounded on the theoretical principles of how
humans make decisions, characteristics of a general model
of decisionmaking are offered. Assuming that a project
sponsoring agency acts like an individual decisionmaker,
these characteristics can help formulate a model of agency
decisionmaking in the implementation of NEPA. This study
provides an operationalization of these characteristics in
a model of highway route location decisionmaking by the
Michigan Department of Transportation (MDOT) and a tech-
nique for its estimation. Specifically, answers are sought
to the following questions: (1) How does the MDOT make
route location choices? To what extent is a route decision
determined by a simple decision rule? and (2) What weight
is given for social, economic, and environmental factors

reported in agency decision documents?

63

64

The Decision

 

A general model of agency decisionmaking under NEPA is
presented in Chapter Two. To test this model, this study
focuses on the decision behavior of the MDOT in developing
and maintaining the quality of transportation in Michigan.
Because a large portion of highway construction is financed
with federal-aid highway funds, the MDOT must follow, among
other procedures, the requirements of NEPA if it wishes to
qualify for this aid. Federal-aid high funds are adminis-
tered by the Federal Highway Administration (FHWA), which
is located in the U.S. Department of Transportation. The
FHWA assists the MDOT in applying for aid and in meeting
federal requirements.

Although NEPA guidelines, and later Implementing Regu-
lations, interpret how to carry out the law, the FHWA took
the initiative to draft their own administrative regulations
with respect to NEPA. NEPA's mandate is tailored to the
peculiarities of transportation planning by the FHWA (see
23 CFR 771). The M001, designated as the "highway agency",
puts into action these administrative procedures from the
FHWA. The implementation of FHWA rules likewise require
the M001 to develop their own procedures to instruct its
staff. As required by Section 136(b) of the Federal-Aid
Highway Act of 1970, the Federal Highway Administration

developed process guidelines for considering social,

65

economic, and environmental effects of highway projects in
Policy and Procedures Memorandum 90-4. PPM 90-4 was
designed to aid the various state highway agencies in
identifying and reviewing their present organizations and
processes that are related to these impacts. The MDOT
Action Plan was developed to meet the requirements of

PPM 90-4. It is a "...through description of the organi-
zation and processes to be followed in the development of
federal-aid highway projects, from their conception in the
systems planning to the point where all significant
planning alternatives have been eliminated." (MDOT, l974,i).
As one can see, the operationalization of NEPA involves a
hierarchy of interpretations from the statute to adminis-
trative action. This study relies on the final interpreta-

tion at the project level to test the general model.

Type of Decision: Highway Route Location

 

The MDOT activities include planning and development of
all modes of transportation in the state. Highway planning
considers decisions on route, interchange, and structure
location. For the purposes of this study, the primary
focus is highway route location. This usually requires the
completion of the NEPA planning process, or an environ-
mental assessment which argues for exempting a particular
project from further NEPA requirements. The interest here
is in those projects which demand the execution of the NEPA

process.

66

MDOT's highway planning provides an excellent example
for the study of decisionmaking under NEPA for a number of
reasons. First, the nature of MDOT's mandate to provide
safe and efficient transportation requires a commitment to
continuous planning. Given this need to plan, the MDOT has
formalized its planning and decision activities in standard
Operating procedures, or rules of thumb for reoccurring
situations. Variation in the factors of decision should
reflect the complexity of the problem situation more than
changes in the makeup of the Department. Second, the
scope and nature of highway transportation planning includes
the full gamut of social, economic, and environmental
issues. Although some other types of federal programs may
require compliance with NEPA, they may only face a few
environmental issues or ones that are limited in scope.
Highway transportation planning is multi-dimensional; it
must deal with many issues at both the macro and micro
levels. Highway planners can not avoid the complexity of
the environment. Third, more so than other agencies, the
MDOT must confront the implementation problems associated
with NEPA in its everyday activities. Since the inception
of the law, it has develOped a record of decisions which

demonstrate its response or manner of coping with complexity.

The Decisionmaker: MDOT

 

The authoritative decision body of the MDOT is the

Michigan Transportation Commission. The Commission is made

67

up of four members, appointed by the Governor with advice
and consent of the State Senate. No more than two members
can be of the same political party. The members serve
staggered four-year terms. The Commission commands a
staff which, historically, has developed an expertise in
highway planning and engineering. Starting from its highway
orientation, the scope of responsibility of the M001 was
expanded in 1972 to include other modes of transportation -
airports, railways, public transportation systems, and
bikeways. Given the MDOT's vast responsibility, the
Commission must rely on the staff experience and knowledge
in each field of planning. Almost as a rule, the Commis-
sion accepts the departmental staff recommendations of
project level choices. Project decisions are assumed to
be the outcomes of organizational processes and procedures.
The MDOT, as any specific function department, has a
set of goals that govern their behavior. The primary
concern of the MDOT is maintenance and expansion of trans-
portation service of the state. The emphasis is expected
to vary between maintaining and expanding, depending on
the nature of travel demand in Michigan and on the resources
of the state. It is also in the best interest of the MDOT
to qualify for federal matching funds to build its projects.
Therefore, the MDOT makes the effort to have its routes
designated as part of the Interstate System or US network.

While it can be argued that Michigan, like any other state,

68

wants to get its fair share of federal aid, the objective

is to try to get as much as possible.

Decision Process: Route Location Planning

 

Highway route location planning begins with the recog-
nition of transportation problem areas. These areas or
trouble spots are located by, among other things, a rating
system which identifies deficiencies in transportation
service. The condition of the transportation system is
used at the statewide planning level to establish needs
for the regions of Michigan. Once a set of needed highway
transportation projects is identified, a number is selected
each year to study for possible improvement. The MDOT
commences planning procedures for each selected project.
Depending on the nature of a route location project,
alternative transportation corridors may be considered
first. Corridor analysis is followed by an examination of
alternative routes within a chosen corridor. In another
study of highway planning, the number of alternatives which
are considered seems to be a function 01: the physical
constraints present in the corridor, expected cost of the
project, and the extent of "community distributional
effects" (McFadden, 1975, 404). Information about the
effects of each alternative route is organized and presented

in a Draft EIS. The Draft is reviewed closely by the FHWA,

69

and upon approval it is distributed to the public and to
governmental agencies.

After the completion of the first Draft EIS, a series
of public meetings, and later hearings, are conducted to
generate any information which may have been overlooked,
as well as interagency and public review and criticism.
Public hearings, more importantly, serve to establish a
record of comments on the planning process, and Of the
preferences of citizen, business, and other interest
groups. Upon completion of public hearings all information
and comments are reviewed by the MDOT staff. Where
necessary, answers are developed to questions raised
during the public hearings. If additional analysis is not
required, the final EIS is written, and a recommendation of
a highway alternative is made to the Transportation Commis-
sion. The results of the MDOT planning and environmental
analysis is reported to the Commission in a Project
Engineering Report. After reviewing this report, the
Transportation Commission makes its decision on the fate
of the project.

Once the Commission makes its route location decision,
the M001 applies for final federal approval for highway

funds.

Decision Criteria

 

The M001 presented and evaluated thirty-one route

location factors in its Project Engineering Report during

70

the time period of this study. While this in itself is a
substantial reduction from the universe of factors which
make up the human environment, the principle of bounded
rationality suggests that the decisionmaker manages the
complexity of the problem at hand by focusing on only a
handful of important or essential factors. Likewise, the
MDOT processes the information reported in the Project
Engineering Report by focusing its attention on what it
deems to be essential decision factors. The way it
processes the information from the essential factors
depends on the choice criterion. It is assumed that there
exists an implicit choice criterion that governs MDOT
behavior. In order that some understanding can be brought
to the question of MDOT decisionmaking, it is necessary to
hypothesize the makeup of the essential factor set.
Although the M001 reports on thirty-one route location
factors, it does so in accordance with FHWA guidelines.
These decision factors fall into eight general categories
(23 CFR 790.3(c)). The MDOT, on the other hand, uses only
those variables which it feels are important. What is
considered important is shaped by its perspective,
experience, and external pressure. MDOT's history, like
that of an individual, represents a record of trial and
error. From this record, a set of standard procedures is
developed and fine tuned. These standards allow the M001

to deal with difficult decision situations. Certain simple

71

patterns of behavior or rules of thumb are relied on by

the organization. In the pre-NEPA period highway planning
by the MDOT consisted of a series of sequential steps which
processed inputs from the staff and department programs,
and which made outputs in the form of recommendations
”...in relative isolation from outside influence.“ (MDOT,
1974, v). The external costs to the local economy, social
cohesion, and environment were secondary to transportation
goals.

The introduction of NEPA called for a change, a new
response. Part of the MDOT's response is the development
of its Action Plan and subsequent reorganization to implement
the plan. The Action Plan opened avenues of communication
and review between the M001 and other federal and state
departments during the project planning process. While
this was a significant achievement in itself, this study
is interested in the decisions which are the outcome of
this reform in the decisionmaking process.

In Chapter Two the point is made that in spite of the
NEPA mandate for a systematic and comprehensive decision
process human decisionmakers are boundedly rational and
use simplifying strategies to deal with complex problems.
Consistent with the discussion of boundedly rational
behavior, it is argued here that the MDOT adopts a planning
procedure which reduces decision alternatives to a final

set of acceptable ones which are then given a formal

72

review. The final set of acceptable alternatives are also
in compliance with NEPA and other statutes. It seems
unreasonable for the MDOT to even consider alternative
route locations which might fail to comply. Thus, although
compliance is not granted until the final Commission
decision is made, the MDOT uses its experience and the aid
of the FHWA administrator to produce options during the
design stage. The questions are which route location
factors are part of the MDOT's essential variable set, and

are the environmental factors part of this set?

MDOT Route Location Premises

 

The decision criteria of the MDOT are consistent with
the Department's responsibilities, aspirations, and view of
the world -- its perceived role. The MDOT, like all
agencies of government, is given executive and legislated
mandates which shape its role in the state bureaucracy. At
the heart of the MDOT's perceived role are the fact and
value premises which guide it in assessing and solving
problems it must answer. It is hypothesized that the
MDOT's factual premises instruct it, in making route location
decisions, to focus on calculating the level of transporta-
tion service, reasonableness of cost, degree of community
compatibility, and compliance with federal requirements.

The Department would like to design and select highway

routes that serve a number of purposes. First, and maybe

73

foremost, it is in the interest of the MDOT to demonstrate
its planning and engineering expertise by improving the
transportation system of the state. From an engineering
perspective, objective standards exist to determine the
degree of improvement. Second, an alternative, just like
a project itself, needs to be justified. Transportation
alone is not a reason for a project, but the benefits
accruing to the communities serviced by the transportation
system are. The degree of "fit" between the proposed
alternative route and existing and future community needs
and service demands describes how the transportation
benefits will be distributed. It seems important that the
MDOT select routes that benefit the Department's perceived
clientele. Third, the MDOT, while not willing or able to
find the most cost effective alternative route meeting its
goals, would still try to show its cost consciousness.
Being a public agency, the MDOT can not appear to be
unconstrained with public money. At the same time there
usually is very little outside, public expertise which can
determine if there are more beneficial or less costly
alternatives than the ones designed and analyzed by the
MDOT.

Finally, MDOT wants routes which qualify for federal
highway funds. Thus factual premises must reflect compli-
ance with FHWA administrative regulations. The federal

regulations, of which NEPA is part, put constraints on the

74

selection of a route. Given the scope and nature of NEPA's
mandate, compliance is not a single objective determination,

and it will be discussed further.

MDOT Compliance

 

In Chapter Two compliance is defined as a determination
by an overseer that the implementation of a program or
plan of action has or has not been followed by the imple-
menters. This determination depends on the information and
knowledge available about implementation and the strategy
of the implementers. MDOT compliance with NEPA through
the FHWA regulations depends on the implementation plan
established by the MDOT and approved by the FHWA. NEPA's
implementation requires both procedural and substantive
components.

In the initial years of NEPA the implementation plan
was generally described in 001 Policy and Procedure
Memorandums. In 1974 the highway decisionmaking process
was specifically outlined by the MDOT, as well as by all
states applying for federal highway funds, in its Action
Plan. For the most part, the Action Plan spells out the
types of studies which should be performed and the impacts
which should be discussed. It provides a program for
collecting, coordinating, and integrating information
from the public and other parts of government. The concern

of this study is centered on the application of the

75

substantive component of NEPA. It is assumed that the
MDOT would avoid the risk of losing federal funds or of
project delays from judicial action due to procedural
non-compliance. In addition, the incentive to comply
procedurally is greater because it is relatively easy, and
also failure to comply is likewise easy to detect.

Compliance could be characterized by the overseer
exercising both control and management activities. It is
hypothesized that the MDOT under the direction Of the FHWA
follows a strategy. A strategy, leading to compliance, is
one where attention is paid to meeting easily verifiable
standards and where satisfying ambiguous factors is nego-
tiated. The MDOT is expected to use as factual decision
premises the statutory requirements, such as air, noise,
and water pollutiOn standards, which are under the NEPA
umbrella. These and similarly developed standards serve as
constraints within which solutions are sought. On the
other hand, general and non-specific environmental factors
which are in the spirit of the Act, such as general
ecology, wildlife habitats, and community cohesion, are
weakly defined and poorly understood. Whether or not these
factors influence decisionmaking depends on the managing or
problem-solving attitude of the FHWA and MDOT response.

The way in which the MDOT combines, blends, or elimi-
nates policies pertinent to highway route location deter-

mines the nature of the value premises. Value premises

76

are adopted for which the conflicting interests in each

transportation project are minimized.

The MDOT Decision Rule

 

The M001 relies on rules of inference to reduce the
universe of choices and provide a set of appropriate
alternatives within which selection can be made. The
number of choices is reduced by assuming that the MDOT
gives attention to a limited number of variables. It is
assumed that the M001 selects the alternative solution
which yields most perceivable net benefit in terms of its
goals. The MDOT adds the weighted variables of its
essential set. The weights for the essential variables are
the value premises which are derived from the interaction
of the M001 and its political environment. In this case,
the MDOT combines the weights attached to the set of route

location factors.

Route Location Factors

 

The choice of an alternative solution to a transporta-
tion project by the MDOT culminates the highway planning
process. Outcomes of MDOT planning are a function of an
underlying choice criterion. The empirical analysis of
this study attempts to determine the underlying choice

criterion from MDOT's history of decisions.

77

The empirical analysis is based on data collected from
route location environmental impact statements and correspon-
ding project engineering reports. The period of investiga-
tion is 1970 to 1980. During this period there were
twenty-six projects reported. These are the primary
decision documents of the M001. These documents summarize
the process and all information available to the MDOT, as
well as comments and criticisms generated through the
public review process, on alternative routes.

The description of the general model of decisionmaking
suggested that a decisionmaker focuses only on a few
essential variables in order to make a decision in complex-
ity. This study assumes that these variables are reported
in the MDOT decision documents.

An examination of Project Engineering Reports reveals
that the MDOT assesses and reports on social, economic, and
environmental effects by using thirty-one factors. The
decision factors are presented in Table 3.1. The informa-
tion that is reported in the text on each factor ranges
from specific enumerations to short descriptive comments.
In addition, each report summarizes the social, economic,
and environmental effects for each alternative in terms of
the degree of advantages each is evaluated to possess.
Advantages are rated as "major", "minor", or "no significant
difference". These are ratings against the other competing

alternatives in the set being evaluated. These judgments

78

Table 3.1
List of MDOT Route Location Factors

 

REGIONAL AND COMMUNITY GROWTH

a. General plans and proposed land use

b. Total transportation requirements

c. Status of the planning process

d. Fast, safe and efficient transportation

CONSERVATION AND PRESERVATION

Soil erosion and sedimentation
General ecology of the area

Park and recreational facilities
Wildlife and watersowl areas
Historic and natural landmarks

(DQOO'OJ

PUBLIC FACILITIES AND SERVICES

Religious facilities

Health facilities

Educational facilities

Public utilities

Fire protection and other emergency services
National and civil defense

OMMUNITY COHESION

Residential and neighborhood character and stability
Minority and other specific groups and interests
Local tax base

Property values

ISPLACEMENT OF PEOPLE, BUSINESSES, AND FARMS
Displacement

Relocation assistance

Replacement housing

Economic activity and employment

IR, NOISE AND WATER POLLUTION
Air pollution

Noise pollution

Water pollution

DUO!) QOU'OIU 0.0an ‘thOU'Q’

AESTHETIC AND OTHER VALUES
a. Visual quality
b Joint development and multiple use of space

PROJECT AND RELATED FACILITIES COSTS

a. Engineering, right of way and construction costs

b. Maintenance and operating costs

c. Operation and use of existing facilities and other
transportation facilities during construction and
after completion

 

79

about the degree of advantage are point by point evaluations
of alternative routes on each factor. Many of these
judgments have to consider non-measurable, as well as
measurable impacts. It is assumed that these point by
point ratings are true reflections of MDOT analyses of the
social, economic, and environmental factors required by
NEPA and other legislation.

The essential variables are found, it is hypothesized,
in the reported factors in Project Engineering Reports.
The route location factors are discussed in two general
categories. The first category is transportation, planning
and cost. It represents those factors that are closest to
the MDOT's mission or goal. The second category covers all
other factors which provide a more specific picture of the

impacts associated with a transportation project.

Transportation, Planning, and Cost

 

Providing direct benefits through a transportation
improvement, such as a highway, is the primary goal of the
MDOT. Meeting this goal justifies the allocation of
project funds, as well as demonstrates MDOT's engineering
competence. Transportation benefits are accounted from
those factors that lead to fast, safe, and efficient
transportation service. Benefits can be accrued by
improved service which can be assessed two ways. First,

transportation, measured by user cost and travel time, can

80

be thought to represent the cumulative direct benefit of a
proposed alternative route. It is measurable and easily
evaluated by the MDOT. One would expect the M001 to use
this variable to determine the acceptability of alternative
routes. Second, the compatibility of the transportation
alternative with local and regional land use plans is both
transportationally and politically significant. Present
and future land uses are one of the sources of traffic
demands for which the improvement is proposed by handle.
While a transportation improvement can be a solution to a
problem, it can also act as a generator of opportunities
and promote the direction of an area's growth. For this
reason, as well as for considering local and regional
political interests, one would think that the MDOT pays
attention to the status of planning in the project area.
Project costs have many desirable properties to aid
MDOT decisionmaking. Direct costs are easy to identify and
to calculate. Reliance on the price system provides a
convenient common denominator for the assessment of alter-
natives. Given the assumptions of the market, cost figures
tell the decisionmaker about opportunities given up to
attain the proposed benefits of an alternative route.
Construction costs give an indication of two features of
a route: (1) the physical constraints of the natural
environment over which it travels, and (2) the amount of

resources necessary for its completion. On the other hand,

81

right-of-way costs describe, in part, the level of indirect
costs associated with a particular choice. To the planner
and engineer, the sum of these two types of cost into total
project cost may be representative of opportunities lost.
While some might use this figure, it is generally accepted
that there are indirect costs which are missing in this
calculation. Supplemental measures are needed. Failure to
seriously consider indirect costs along with direct costs
to the environment in previous decisionmaking was one of
the reasons behind the enactment of NEPA. This flaw in
the use of project costs does not outweigh its importance
in public sector choice.

The following category, for the most part, describes
specific, non-aggregated types of impacts due to the

proposed transportation improvement.

Associated Effects

 

The list of factors reported by the MDOT in accordance
with the FHWA regulations is rounded out by those which
provide the details of effects or impacts not central to
the mission of the Department of Transportation. These
associated factors fill the gaps left in the assessment of
benefits and project costs. The range of associated factors
covers every conceivable impact. Although this is compre-
hensive, determining the effects of some factors is

difficult. Some are based on weakly defined and poorly

82

understood concepts. They involve dynamic systems (bio-
logical, social, economic, and physical), which make
calculations very complex. Subsequently, only very simple
measures are sought by the decisionmaker.

The MDOT had to deal with such factors even before the
enactment of NEPA. Highway projects by design always had
associated effects which required attention. With the
passing of NEPA, emphasis has been placed on increasing
the attention given to these factors, as well as adding to
the list. The associated effects are composed of three
groups of route location factors: (1) public facilities
and service, (2) environment, and (3) social and economic
systems. The factors are grouped in this way in Table 3.1.

Associated effects of an alternative route can be
either beneficial or costly. Benefits of this type are
outside the bounds of the highway planner's primary interest.
One would expect the MDOT to do little accounting of them in
decisionmaking. This may not be the case for negative or
costly effects. It is assumed that negative associated
effects decrease the acceptability of a route and the
chance that it would comply with the legal mandates.
Consistent with boundedly rational behavior, the highway
planner can be thought to perceive negative associated
effects as design constraints. The planner attempts to

avoid these factors as much as possible in the early design

 

stage. This strategy eliminates or modifies unacceptable

83

features of optional routes to simplify the final choice.

Operational Model of MDOT Decisionmaking

 

The Dependent Variable

 

The dependent variable is the MDOT's evaluation of an
alternative route for a given highway route location
planning project. Because the true evaluation index is
unobservable, it is only known whether a route alternative
is accepted or rejected. Thus, the dependent variable is
given a value of "1" if it is the route recommended by the
staff in the Project Engineering Report; it is assigned a

value of "0" if not selected.

The Independent Variables

 

In constructing a model of MDOT route location decision-
making independent variables representing the hypothesized
factual premises are devised from the reported route
location information. All reported factors are initially
examined for what they may contribute. In operationalizing
the independent variables the MDOT evaluation of advantage
is utilized. Although the MDOT rates on a major, minor,
and no significant difference scale, there is no information
which would suggest that the interval between levels is
unitary. Lacking this knowledge, major and minor advantages

are rated simply as an advantage.

84

Research Approach

 

The process to be followed in analyzing MDOT route
location decisions is composed of two steps which are
consistent with the theory of decisionmaking in the
previous chapter. First, the data will be analyzed to
find evidence of screening or filtering by the MDOT in
the design stage of the planning process. Second, an
estimation will be made using the factors which appear to
have been used by the MDOT to differentiate their alterna-
tive choices. This two-stage approach recognizes that
the whole decision process must be examined to have an
accurate picture of MDOT planning. This is especially
true if specific factors are used as design constraints in
the process. The second stage of this analysis can not

stand alone without misleading results.

Screen Factors

 

It is hypothesized that specific, legislated policy
leads to "protected" factors in route location. The MDOT
minimizes conflict and project delay by screening out as
many of these factors in the generation of alternatives as
possible. Laws exist for historic and anthrOpological
sites, endangered species, clean water, noise, coastal
zones, wetlands, clean air, and federally subsidized
parks (see 49 CFR 771.103). If factors representing these

pieces of legislation are screened out, the data analysis

85

should show them as not being rated very often because each
alternative would treat them the same way, and thus no one
alternative would be given an advantage. HISTORIC, AIR
POLLUTION, NOISE POLLUTION, WATER POLLUTION, PARKS, and
WILDLIFE are expected where possible to be screened. In
situations where they cannot be screened out, they should

coincide with the alternate choice.

Final Decision Factors

 

Although the screened factors are an important part of
the decision process, their lack of use in the final decision
stage would prohibit them from being statistically signifi-
cant in model estimation. In this second stage of analysis,
those factors which were used, that is rated, in half or
more of the 26 project decisions will be estimated as part
of the model. It is hypothesized that these factors will
show the MDOT to be following its own goals. If the MDOT
has developed a good strategy of minimizing conflict and
focusing the decision into its area of expertise, trans-
portation and transportation related factors should be
dominant. There are three categories of factors which are
transportation related and largely within the control of
the MDOT. They are: level of transportation service, degree

of community compatibility, and costs.

86

Level of Transportation Service

 

There are four factors which the M001 can use to assess
the level of service provided by an alternative route. They
are TRANS REQ, TRANS STATUS, TRANS EFFIC, and EXIST RDWY.
TRANS RED and TRANS STATUS establish the priority of the
project route and identify the existence of local and
regional government plans in the communities in the project
study area. These are informational factors which conform,
as they all do, to FHWA regulations. There are two design
factors. The first in EXIST RDWY which assesses the
disruption to existing transportation facilities associated
with an alternative during and after construction. Second,
TRANS EFFIC is short for "fast, safe and efficient trans-
portation”. This variable stands for transportation
benefits, and it is usually referred to in terms of ''user
cost". In many cases user cost is a function of route
length. The shorter route length means user time saved.
The greater the degree of user cost saved, the more likely

the MDOT is to select the route.

Reasonable Cost

 

The M001 reports two types of cost associated with a
project: PROJECT COST and OPER COST. Operating cost is
very closely related to the length of the alternative
route. By trying to achieve transportation efficiency,

the MDOT almost automatically reduces Operating cost. 0n

87

the other hand, project costs are more complicated and are
determined by natural and social environment encountered by
a route. The MDOT is expected to demonstrate its concern
for efficient transportation and cost-effectiveness by
favoring the least cost alternative route, all other things

being equal.

Degree of Community Compatibility

 

While it could be argued that community compatibility
is a function of a number of specific impact factors,
GEN PLAN not only considers present compatibility but also
the future. By rating each alternative route according to
its fit with government land use and development plans,
the MDOT evaluates the distribution of transportation
benefits against the perceived needs and desires of local
governments. The route which enhances this fit increases

its likelihood of selection.

Conclusion

 

The purpose of this chapter is to describe how the
theory presented in Chapter Two can be applied to the case
of an agency which must implement and comply with NEPA,
and to specify how questions about implementation and
decisionmaking under this Act can be investigated empiri-

cally. The M001 is an excellent example to study because

88

their decisions are the type which were the focus of
criticism during the environmental movement which led to
the passage of NEPA. At the same time they have established
a record of implementing the law with all its ambiquities.
Using this record of coping and decisionmaking, this
study, relying on the boundedly rational model in a
political environment, tries to test the hypothesized
decision premises used by the MDOT. By examining the
whole decision process, as opposed to just final choice,
the influence of NEPA can be evaluated in fairer way than
by previous research. The significant step in this exami-
nation is the analysis of the operational model results,

using PROBIT, and of the data itself.

CHAPTER FOUR
ANALYSIS OF EMPIRICAL RESULTS

As the history of modifications of the CEO Guidelines
and the final NEPA Implementing Regulations indicate, the
early planning stages of a program or project should be
just as important as the final choice stage in integrating
NEPA's mandate. Thus the whole process should be examined
in the evaluation of implementing this policy. Following
the description of MDOT decisionmaking already presented,
this study relies on an empirical analysis to develop
evidence to determine the MDOT's essential variable set.
This analysis, consistent with the need to look at the
whole decision process, is divided into two parts. The
first part discusses the utilization of route location
factors and makes inferences about their role in early
planning. The second part of the analysis is the interpre-
tation of the estimated decision model. Together these
parts provide a picture of the MDOT decisionmaking behavior.

The following empirical analysis of MDOT decisionmaking
is based on 26 route location decisions made during the

period 1970-1980.

90

Data Analysis: Rated Factors

 

An examination of the data collected from MDOT Project
Engineering Reports offers some clues to the pre-decision
activities of a highway project. Table 4.1 shows the
number of decisions in which each factor was given a "no
significant difference" rating. This rating means that the
factor in a project was not significant in the final
decision stage because it did little to differentiate the
alternatives being considered. There can be a number of
reasons for a factor repeatedly receiving a NSD rating. The
factor can occur only on rare occasions, be too general or
difficult to calculate, or can be a planner's or engineer's
design constraint. The greater the number of times a
factor does not separate the alternatives the lower its
chances of being a determining variable in the final decision
stage.

Consistent with the literature on bureaucratic decision-
making, it is hypothesized that the MDOT, like any agency,
goes through a process which is a preformal analysis of the
choice consequences to filter out unacceptable solutions.
The decision problem is simplified by the interaction of the
design and analysis stages of the process. After design is
completed, a set of alternatives are left which are perceived
to be feasible and acceptable. They are acceptable in the

sense that they meet some minimum operational standards.

91

 

 

Table 4.1 "No Significant Difference" Ratings
Factor Number of N50 Ratings
Project cost 4
Displacement 5
Transportation efficiency 6
General plans 7
Soils 8
General ecology 9
Residential character 10
Operating cost 10
Local tax base 10
Wildlife 12
Education 13
Visual 13
Existing roadway 15
Parks 16
Water pollution 17
Economic activity 18
Religious 20
Public utility 20
Noise pollution 20
Transportation requirement 20
Transportation status 21
Property values 23
Fire service 23
Air pollution 24
Relocation 24
Replacement 24
Defense 24
Health 24
Minority interest 25
Historic 25
Joint development 25

 

Number of project decisions = 26

92

In the case of the MDOT, it is theorized that final alter-
natives surpass some minimum transportation requirement and
comply to specific mandates in the law. The final and
formal decision is really one of selecting the best fitting
alternative to the remaining and important planning factors
which are not handled completely by the discretion of the
transportation designer. If this description is correct,
the failure of factors to affect route location selection
should be examined to determine if their weakness is due to
their status as design constraints or to their insignifi-
cance.

If certain factors, especially environmental factors,
are design constraints in the alternative generation process
they would not be significant in the final choice because
all alternatives in a project set would have met these
design requirements. In examining Table 4.1, four of the
environmental factors, NOISE POLLUTION, AIR POLLUTION, JOINT
DEVELOPMENT, and HISTORIC, are NSD in twenty or more
decisions. Of these four all but JOINT DEVELOPMENT have
specific administrative regulations behind them. On the
other hand, SOILS, GENERAL ECOLOGY, WILDLIFE, and VISUAL are
factors which lack specific statutory backing. These
factors are definitely in the spirit of NEPA, and they are
clear enough to differentiate alternatives as demonstrated

by the number of times they received ratings.

93

Factors Rated with Choice

 

The number of times a factor was used to report that an
alternative had an advantage and was then selected provides
more information about the planning process and potentially
important factors. Table 4.2 displays this information.

The factors are separated into two groups. Those in the tOp
group were used in over half of the project decisions. The
ratio is the number of times a factor was rated out of 74
alternatives in the denominator, and out of that number,

how many times it was selected in the numerator.

The route location factors in the top group of Table 4.2
were used more often, and they provided comparative measures
for the MDOT. Given their more frequent use, the top
group of factors can be tested for their contribution
to the final choice at the decision stage. On the other
hand, attention should not be cast away from the bottom
group of location factors.

If the information on Table 4.1 and 4.2 are examined
together, it appears that whenever some of the lesser rated
factors are present that they match with the choice a high
percentage of the time. For example NOISE POLLUTION was
involved in six decisions, and six times the selected
alternative had an advantage rating for that factor.
Although Table 4.2 shows this factor's ratio of choice to

total rated to be six out of 13, it did better on a decision

94

Table 4.2 Number Rated with Selected Alternative

 

 

Factor Rated with choice/Total rated
Project cost 16/30
Displacement 13/32
Transportation efficiency 13/32
General plans l7/25
Soils 12/22
General ecology 12/23
Residential character 13/22
Operating cost 11/18
Local tax base 11/25
Wildlife 10/19
Education 11/18
Visual 9/20
Existing roadway 6/19
Parks 9/14
Water pollution 7/13
Economic activity 7/16
Religious 5/9
Public utility 4/8
Noise pollution 6/13
Transportation requirement 5/9
Transportation status 3/8
Property values 2/6
Fire service 3/7
Air pollution 2/7
Relocation 2/5
Replacement 2/5
Defense 2/3
Health 2/2
Minority interest 1/2
Historic 0/2
Joint development O/l

 

Number of project decisions = 26
Number of alternatives = 74

95

by decision basis. The reason that there are 13 alterna-
tives rated in only six decisions is due to the fact that
some projects included more than two alternatives. This
allows for the possibility of more than one being rated
for the same decision. This information suggests that
route location factors in the bottom group of Table 4.2
which match the final choice a high percent of the time
are important to the MDOT but appear infrequently. This
infrequent rating can be due to their rare occurence, or
as mentioned earlier, their status as design constraints.
Caution must be exercised here for although it is sugges-
tive, it is only a correlation.

The finding of the first part of this data analysis is
that a number of location factors, dominated by "protected"
ones, appear to have been avoided in the designing of
project alternatives. By "avoiding" it is meant that the
designer or engineer developed options which circumvented
critical areas or, in some way, made the options, being

considered, equal on that particular factor or attribute.

Model Estimation

 

The second part of this empirical analysis considers
the factors affecting the MDOT's final project choice. A

statistical model is described for meeting this purpose.

96

A series of hypothesis are estimated in order to
examine suggested hypothesized decision criterion. A
discrete choice model, probit, is used to generate decision
factor coefficients. Coefficient estimates can be inter-
preted in a similar manner as in linear regression. A
positive coefficient for a variable implies an increase
in the predicted probability that an alternative route
will be chosen. There are a few summary statistics which
serve as a measure of goodness of fit. One is the log
of the likelihood function, and it approaches zero for a
perfect fit (Hensher, 1981, 49-52). Another is the likeli-
hood ratio statistic, -2LLR, which is distributed approxi-
mately chi-square (McFadden, 1976, 59).

The percentage of correct project decision predictions
is also a useful measure. This percentage is computed
by first using the estimated results to predict for each
project that alternative route with the maximum estimated
probability. This is compared to the actual MDOT choice,
and the number correctly predicted is divided by the total
number of decisions.

The selection of PROBIT is based on its compatibility

and convenience for this research.

97

PROBIT

In studying the influence of NEPA on agency project
level decisions one is interested in finding the statistical
distribution of decision rules which explains the observed
patterns of choice (McFadden, 1975, 402). Given that the
attributes of each alternative choice, and the actual
project selection are known, one should be able to predict

project decisions of an agency, assuming the information

1P..Aam-.— gnu--“.-_..__{1

on the attributes are part of the agency's essential
variable set. More specifically, this analysis attempts
to make a statement about the likelihood that a particular
alternative choice will be selected. In so doing the
relationship between a set of attributes, describing an
alternative, and the probability that the alternative will
be chosen can be found (Pindyck and Rubinfeld, 1981, 274).
Statistical models suitable for this type of analysis
are called qualitative choice models. For qualitative
choice models the dependent variable involves two or more
qualitative choices. In this case the behavioral responses
associated with agency decisions are either "accept" or
"reject" for an alternative. A review of the econometric
literature offers a number of statistical models for
consideration. Two models of this type receive most of
the attention; they are the linear probability model, and

the "constrained" linear probability model. The "constrained"

98

linear probability model has two common specifications
called probit and lpgit. These models are based on
alternative assumptions about the probabilistic nature

of the decision process. Given the nature of the dependent
variable, the assumptions of the decision process, and the
desire for predictive capacity, the "constrained" linear
probability model will be used in this analysis - specifi-
cally, the probit model.

The empirical analysis of project decisions by federal
agencies begins by assuming that the probability of an
alternative solution being selected is a linear function
of the attributes of that solution. These attributes are
the essential variables which are monitored by an agency.
Let "A" be the choice set of recognized competing alterna-
tive solutions, and let the agency's evaluation of alterna-
tive "a", which is an element of A, be given by Va =

V (X 8, ea), where X6 is a vector Of alternative a's

a,

comparative attributes, B is a vector of coefficients,

is the disturbance term. X contains the essential

and ea a

variables. These correspond to the requirements of NEPA.
An alternative solution "a" is selected if Va is above the
agency's critical cutoff value, V *. The decision can be

a
represented as,

1, if va>va*

*

accepts "a", Ya:
agency { . .
rejects "a", Ya=0’ 1f Va>Va

99

What must be noted is that the value of Va is
unobserved. It is assumed that the decisionmaking agency
has an index, Va’ which reflects its likelihood to accept

an alternative solution. The index, V is assumed to be

a,
a continuous variable which is random and normally
distributed. What is known is only the observed choice,
that is Ya=0 or 1. The underlying model in regression form

is,
V = 01 + BX 1' e
a a a

Since one is interested in the probability of an
alternative being accepted or rejected, it is necessary to
transform the model such that predictions will lie in the
(0,1) interval for all X. Further, one

...would also like the transformation to
maintain the prOperty that increases in
X are associated with increases (or
decreases) in the dependent variable for

all values of X (Pindyck and Rubinfeld,
1981, 280).

The transformation for the probit model is provided by
using the cumulative normal probability function. Thus,

the probability can be represented by,

Pa = F (a + exa) = F (va)

where F is the cumulative normal probability function.
The probit model assumes that Va* is a normally
distributed random variable; the probability that Va*

100

is less than or equal to Va can be computed from the cumu-
lative normal probability function. "The cumulative normal
function assigns to a number (V) the probability that any
arbitrary (V*) will be less than or equal to (V)" (Pindyck
and Rubinfeld, 1981, 282). The standardized cumulative

normal function is,
-32/2
P = F (Va) = 2'.” e dS

where "s" is a random variable which is normally distributed
with mean zero and unit variance. By construction, Pa will
lie in the (0,1) interval. Pa represents the probability
that alternative "a" will be selected by the agency. Since
this probability is measured by the area under the standard
normal curve from -w to Va, the alternative is more likely
to be selected the greater the value of the index Va.

In order to obtain an estimate of the index Va’ it is

necessary to apply the inverse of the cumulative normal

function.
=a+8Xa

The probability Pa is interpreted as the conditional
probability that the agency will select a particular alter-
native given the alternative's attributes. This is
equivalent to the probability that a standard normal

variable will be less than or equal to a + BXa, which is

‘n Jiifil‘fizmmm

(“3"; I n

101

the sum of the weighted attributes.

Because the cumulative normal transformation, used in
the probit model, is nonlinear, ordinary least squares can
not be applied. Maximum-likelihood estimation can obtain

values for a and B. These parameter estimates will be

'-‘.-1

consistent and the apprOpriate statistical tests can be

' -‘nfiifyfii

performed (Pindyck and Rubinfeld, 1981, 283). Given the

nature of the estimation algorithm, it is necessary to

 

offer an explanation of the probit parameter estimates.

The unobserved dependent variable, V is a standard normal L

a’
variable. The maximum likelihood coefficients represent
the change in the standard normal variable that occurs with
each unit change in the independent variable, holding all
other variables constant. Since Va has a unit variance,
the change in the dependent variable can be interpreted in
standard deviation units (Ostrom and Job, 1982, 34).
Further, because the specification of the relationship is
nonlinear, the change in P associated with a change in X is
dependent on the particular value of X selected.

The amount of influence, that individual variables have
on the probability that alternative route is selected,
depends on how close Va is to Va*; the closer it is to Va*
the more impact the variable has.

The probit model is used to estimate the decision

model described earlier.

102

Hypothesis Tests

 

Three estimates of a model of MDOT route location
decisionmaking under NEPA are presented in Table 4.3.

These estimates examine the factors which were used in over
half of the project decisions. It was hypothesized that,
by the final decision stage, the transportation and related
factors under the direction of the MDOT would be strongly
related to alternative choice.

Estimate I is composed of all factors in the top group
from Table 4.2. Only two factors are significant; GEN PLAN
and PROJ-COST. This form of the model predicts well, 84.5%,
but lacks other significant variables. There is a suspected
problem with multicollinear variables. This would make
significance tests meaningless. The result of a Farrer-
Glauber test, regressing each independent variable on the
others, shows that to be the case.

Estimate II combines seemingly collinear factors into
combined variables to reduce the degree of multicolline-
arity. Factors were combined which made substantive sense.
ECOL is composed of SOILS, WILDLIFE, and GEN ECOLOGY. CHAR
equals RES CHARACTER and EDUCATION; CHANGE combines
DISPLACEMENT and LOCAL TX BASE. This estimate performs
very much the same as the first. ECOL and CHAR are close

to being significant.

vuoxlh .- A _a'qn

"f.s .1.» aim-'1" ..1

103

 

 

 

Table 4.3 Model of route location decisions
Factors I II III
Constant -2.09* -2.057* -l.920*
GEN PLAN 1.998* 1.813* 1.554*
TRANS EFFIC - .494 - .334 - .319
PROJ COST 1.342* 1.332*
OPER COST .457 .332 f
MONEY l.lll* '
EDUCATION .636 _
SOILS .711 i
GEN ECOLOGY .225 y
WILDLIFE .095 1
ECOL .641 .622* E
VISUAL .189 t
CHAR .672 .702* l
RES CHARACTER -.173
LOCAL TX BASE .158
DISPLACEMENT -.279
CHANGE - .304 - .205
% Predicted 84.6 84.6 80.8
Log Likelihood -27.58 -28.37 -30.20
-2X LLF 40.77** 39.19** 35.54**

 

Note: ECOL = (
CHAR = (
CHANGE =
MONEY =

OILS, WILDLIFE, GEN ECOLOGY)

S
RES CHARACTER, EDUCATION)

(DISPLACEMENT, LOCAL TX BASE)
(PROJ COST, OPER COST)

Data: MDOT Project Engineering Reports, 1970-1979.
N = 74 alternatives; 26 decisions.

Key: * Significant at the .05 level for one-tail test.
**Significant at the .01 level, chi-square test.

104

Estimate III treats the collinearity of the cost
variables by combining PROJ COST and OPER COST into one
variable, MONEY, and discarding very weak variables. This
final estimate continues to show GEN PLAN and MONEY to be
significant and heavily weighted, while ECOL and CHAR are

significant, but very weak.

Conclusion

 

The objective of the empirical investigation in this
chapter is the determination of an implicit choice cri-
terion, if one exists, of the MDOT which is used to select
highway route locations. The results of the estimated
model reveal a choice criterion which explains the outcomes
of this agency's decisions. The variables of the decision
model are derived from the information reported in the
MDOT documents which are part of the public record. The
success of the estimated model supports the assumption
that MDOT Project Engineering Reports provide the bulk of
information utilized by the agency in their decisionmaking.

The overall findings of the probit model estimations
lead one to believe that the MDOT places more emphasis on
tailoring its projects to the growth and development plans
of the communities in the study area, demonstrating
relative cost consciousness, and, to a much lesser extent,

heeding the impact on residential character and the natural

105

environment than on the other location factors. It must
be remembered that these results represent an average
weighting of factors in decisions taken from a statistical
distribution Of decision rules which explain patterns of
choice (McFadden, 1975, 402). This means that the MDOT
may not have relied on one decision criterion in making
all its route location selections, given the complexity and
available information on any one project. The results of
the estimated model represent the "average rationale" of
the MDOT in making highway route selections.

Even though the MDOT staff collect and generate a
large amount of information in studying each project, no
visible attempt is made to do a comprehensive benefit-cost
analysis (BCA). This is due, in part, to the Department's
recognition of the pitfalls of BCA, especially in counting
indirect benefits and costs. Instead of reducing all
highway benefits and costs into an aggregate measure, the
MDOT reports them descriptively and rates each alternative
on each factor. Variables, ECOL, CHAR and CHANGE represent
impact factors which have indirect benefits and costs
associated with them. Based on the model results, the only
indirect effects weighed are ECOL and CHAR and only just
significant.

The impact of the environmental factors on route
location is significant, but weak. This reveals that,

even though it processes environmental information about

106

the impact of proposed transportation improvements on the
natural environment in its analyses, the MDOT is not very
sensitive to these general concerns. There are two
possible ways to look at the weakness of the environmental
factors. First, in order for a factor to be used, the
information must be available and the decisionmaker must
perceive that the information is valid. Given the infancy
of environmental assessment, it is very likely that the
MDOT failed to use environmental impact information because
it suspected the validity of the findings. Second, the
MDOT may question the importance of the NEPA policy in
highway planning, and only comply with the procedural
requirements of the Act. This is what Congress anticipated
might be the bureaucratic response, and what led to the
strengthening of the public participation in the decision
process. It is also possible for an agency to meet the
ends of "procedural compliance only" by claiming the
factors lack substantive validity. This is a political
strategy to avoid opening up the process.

The picture that this empirical analysis gives of MDOT
decisionmaking behavior under NEPA is one where Egg
attainment of departmental Objectives seems to predominate
over other factors of consideration, especially environ-
mental benefits and costs. The strength of the MDOT's
sensitivity to community plans and proposed land uses is

a demonstration of that point. Land use plans represent

 

107

a community's future development direction and preferences.
Similarly, compatibility with these plans means meeting
future transportation needs and demands. That is a
planning and engineering bias. At the same time it should
not be forgotten that the MDOT, like all public agencies,
is subject to public and political pressure. The MDOT
gives recognition to the preferences of communities in the
project area by taking their plans into consideration.

An unanticipated finding is the insignificance of the
transportation factors in the final decisionmaking stage.
Although this seems inconsistent with the attainment of
the Department's objectives, it is argued here that this
is consistent with a broader consideration of the decision-
making process. This broader point of view recognizes that
decisionmaking is more than the selection of the best
alternative. It involves the search or design of the
alternatives in addition to the analysis Of the alterna-
tives. The manner by which the decision alternatives are
generated has a significant influence on the outcome or
consequence. The weakness of the transportation and
environmental factors suggests a closer examination of the
data in line with the "strategic" planning notion presented
earlier.

The information presented in Table 4.1 suggests
questions for which answers can not be found in the Project

Engineering Reports and FHWA documents. Generally

-1 C Li \.f.)FZ~"._n W

 

108

speaking, what are the design premises used in the
generation of alternatives? How is environmental policy
reflected in these premises? Are environmental factors
considered in some "zone of acceptance?" To find answers
to these questions the next chapter discusses the planning

process in one MDOT highway route location study.

0.. 9f HEN-54.1“”
. _ .

T'K'Z' ..-2‘. L‘."‘.

CHAPTER FIVE
A CASE STUDY OF MDOT DECISIONMAKING:

ROUTE LOCATION PLANNING FOR INTERSTATE 69
FROM CHARLOTTE TO LANSING

Introduction

 

In the previous chapter a set of route location
decisions made by the MDOT were studied to determine
whether environmental factors were part of its decision
criterion. The purpose of this chapter is to focus on one
project decision in order to supplement the conclusions of
the empirical investigation. A case study of one route
location planning process can show how the MDOT carries out
the route location process, how it reports and integrates
location factors, and how it responds to conflicting
policies in decisionmaking.

The conclusions of the empirical investigation imply
that the implementation of NEPA follows a mixed strategy of
implementation. Likewise, the meaning of agency compliance
has both control and management type components. Route
location factors, which are easily measured, are thought to
be used in the design stage of the planning process. For
other important factors a different style is adopted to
learn how it can be included in the process. In order to
bring more evidence to support these conclusions, a study

of one significant and controversial MDOT route location

109

 

project is presented.

Project Study: I—69 Route Location Decision

 

History and Background

 

A review of the Interstate network in Michigan in the
middle 1960's found that the existing system "lacked the
integration necessary for complete effectiveness ...due in
large part to the termination of I-69 at I-94 near
Marshall." (MDOT, 1971, 3). As a result, the MDOT requested
that an extension of I-69 from Marshall to Port Huron be
included as part of the additional 1,500 miles allocation
authorized in the Federal-Aid Highway Act of 1968. The
US Department of Transportation approved an extension from
Marshall to I-75 at Flint. The route location project
which is the focus of this case study covers part of this
extension to the Interstate system; I-69 from Charlotte to
Lansing is the final link in the freeway between Indiana—
polis and Lansing and Flint. Other sections of this route
had been approved or contracted for construction.

Subsequent to receiving federal approval for the
extension of I-69 into the Interstate system, the MDOT
commenced with corridor planning between the major trans-
portation control points. A corridor study was completed
by the M001 staff for the segment from Charlotte to

northwest Lansing in 1972. Accompanying the corridor study

110

111

was a Corridor Environmental Impact Statement which outlined
the impacts of the project and compared the alternative
corridors. The outcome of the analysis covered in these

two documents was the adoption of a generalized location
corridor. This corridor went north from Charlotte at
Temporary I-69/US-27 and US-27BR intersection and dog-legged
to the east to meet I-96, southwest of Lansing. The
corridor also included the area from the 1-96/1-496 inter-
change northerly for three miles in order that two more
lanes could be added to the existing I-96. This approved
corridor sets the boundary for the development and analysis
of alternatives route alignments.

In 1976 the Final Alignment EIS was approved. In the
process leading up to the Final EIS, the MDOT examined two
detailed alignments in the chosen corridor in a Draft
Alignment EIS and it held the required public hearings to
discuss the information contained in that report. With the
completion of the Final Alignment EIS the Federal Highway
Administration granted location and design approval in
March, 1977.

After FHWA approval had been granted for the project,

a group called "Citizens Concerned About I-69" (CCAI)
questioned the validity of the approved alignment. Their
major concern was that the MDOT did not give full considera-
tion to the impact of the approved alignment in the taking

of prime agricultural land. CCAI lobbied successfully to

112

have the Eaton County Board of Commissioners request that
the Michigan Environmental Review Board (MERB) review the
approved route with respect to the impact on agricultural
land. The request of the County Board gave State Senator
Richard Allen an Opportunity to enter the conflict in
support of local agricultural interests. As a result of
the Senator's efforts and those of local agricultural
groups, a review was conducted by the Michigan Senate
through a series of oversight hearings by its Transporta-
tion Committee. Appeals for a review were also made to
the Governor's Office. The Governor brought other State
Departments to carry out an in-depth review of the issues
being raised. Using the Departments of Agriculture and
Natural Resources in this review, it was concluded by the
Governor's Office that further study was not necessary.
The M001 Commission reanalyzed the information generated
during this process and reaffirmed its decision on the
approved alignment (MDOT, 1981, l; Kibby, 1984, 20-23).
Following its failure to get the MDOT to change its
alignment decision, the CCAI turned to judicial action as
a way to delay the project in order to change the result.
In 1980 the CCAI filed suit first in Ingham County Circuit
Court, and later in Eaton County Circuit Court, seeking
an injunction against proceeding with the 1-69 project.
Both courts ruled against the injunction. The MDOT could

move toward the development of the project.

 

113

During the review process and the attempts to attain
an injunction, the CCAI kept the staff of Congressman
Wolpe informed of the situation. The Congressman set up a
meeting with representatives of the USDOT, FHWA, repre-
sentatives of CCAI, and the chair and director of the
Tri-county Regional Planning Commission. The discussion
focused on the removal of prime farm land in the course of
the I-69 route. At this time the MDOT had begun to let
bids for the project, but federal approval of the bid
amount was required (Kibby, 1984, 25). While the M001 was
waiting for bid approval, an amendment to the USDOT and
Related Agencies Appropriation Act of 1981 was enacted
which had the effect of freezing funds for I-69 between
Charlotte and Lansing for 90 days. During this holding
period, the USDOT reviewed the issues and conducted a public
hearing on the project on November 24, 1980. Although the
USDOT can review the evidence behind the MDOT decision, it
has no power to select alignments; the final decision on
route location rests with the MDOT Commission.
As a result of the review activities of his department,

the Secretary of Transportation, Neil Goldschmidt, signed a
Letter of Findings which stated:

The State of Michigan has validly received

approval for the current alignment and has

expended considerable time, effort, and

money in project development.

Construction of I-69 in the approved

corridor, while it would have certain
advantages, would lead to direct and

 

114

indirect loss of substantial amounts of
prime agricultural land.

Selection of an alternative of expansion

and upgrading of US-27 would minimize

impacts on agricultural lands, and would

utilize an existing highway corridor which

is the most direct connection between

Lansing and Charlotte. Although the

US-27 corridor would involve more resi-

dential and business displacement, I

believe that careful design can hold

such displacement to a minimum (MDOT,

1982, 3)
In so doing the MDOT is asked to consider alignments which
use the existing right-of-way of US-27 as alternatives to
the approved alignment. The US-27 corridor was considered
earlier in the corridor planning phase and rejected by the
M001.

In the Spring of 1981 the MDOT Commission directed its
staff to develop and analyze a series of alternative align-
ments for a rural interstate freeway on the existing US-27
right-of-way. Six rural freeway alignments were developed.
The CCAI remained active in this process as well. They
wanted additional alternatives to be considered. A consulting
firm was retained by the CCAI to help develop and analyze
these additional alternatives. After completing their
studies the consulting firm presented an urban freeway
design for the length of the US-27 corridor. Later the firm

combined their proposed design with the MDOT's rural freeway

and offered a rural-urban cross-section option.

115

Under directions from the Commission, the MDOT drafted
a report which covered the new proposed alternative align-
ments, included those presented by the CCAI consultant. In
April, 1982 the MDOT received approval for a Draft Supplement

to the Final Environmental Impact Statement from the FHWA.

A public hearing was held on July 21, 1982 to receive F
comments on the document. After reviewing the comments and E
criticisms from the public and interagency review, the MDOT !
Commission selected one of the alternative alignments ‘
addressed in the report on December 15, 1982. The Commission L

chose a rural freeway alignment in the US-27 corridor, and
it rejected the previously approved alignment. Following
the decision and slight corrections in the Draft Supplement,
the Final Supplement to the EIS was submitted and granted
approval from the FHWA on August, 1983. In so doing the
MDOT met the September 30, 1983 deadline, after which the
project would no longer be eligible for the 90% federal
matching funds because construction of the Interstate system
was coming to an end nationally. The MDOT must also meet a
September 30, 1986 date to award construction contracts for
the I-69 project.

In the discussion that remains each planning phase
leading to the Final Commission project decision is described
with respect to the alternatives that were developed and the

issues that were treated.

116

Corridor Selection

 

The corridor planning process begins with the selection
of a general study area which includes terminal control
points or end points for the route in question. The general
study area for I-69 is shown in Figure 5.1. The control
points are identified in this figure; in addition the
intersection of I-96/I-496 was treated as an intermediate
control point. These points were selected ”...so as to
maintain or improve trunk line integration and efficient
travel movements." (MDOT, 1972, 27). By connecting the
control points any new improvement will effectively utilize
the existing Interstate system and meet current and projected
travel demands.

The M001 designed five initial study corridors within
the study area. They are labeled Corridor A, B, B-l, B-2,
and C in Figure 5.2. Some of the important goals which
guided the development of these corridors were:

1. Serving large volumes of long distance traffic;

2. Integrating with existing and proposed trunk line

highways and major county roads;

3. Serving principal traffic generators and land use

features;

4. Serving the highest incidence of population;

5. Integrating with regional and local development

plans; and

6. Minimizing disruption to existing ecological

117

 

 

11.3-1 . _ 205:“.
_ .
_ u
_
_
_
_
_
_
_ 1111111111 1.
_
_
_
_ 19
_ E
_
_
— lllllll
_
_
_
_
_
.
_
_
T lllllllll
_
_ m
_ w.
H
_ w
_ 13:5.
_ _w
_rL
_ .
_ 8|-I-BEE-
Tdu 452.-
_
. m N
_ L E
_ m1-) w
W. .. ,..., W
N

5

    

|
|
|
I
1
I
-J.
l

     

. - , MASON
/\
l M.

l
1

   

--'-___

  

DIMONDALE

,____.___L.____

____ -2._.__..

I-69 Study Area

Figure 5.1

118

 

Figure 5.2 I-69 Alternative Corridors

119

systems (MDOT, 1972, 39).
These goals were translated into more specific objectives
for the I-69 route location project. Corridors were
designed to provide maximum service to traffic destined for
the Lansing/East Lansing area and to through traffic demand;
to integrate with the I-96 and I-496 freeways; and to
consider social, economic, and environmental effects of a
transportation facility.

In its preliminary discussion the M001 found that
corridors B-2 and C had significant disadvantages. While
these corridors have the advantage of separating I-69 traffic
from I-96 traffic, both would cost more money and treat
relatively light traffic for the added distance that they
require. Other disadvantages pointed out by the MDOT were
that Corridors B-2 and C would also require an intersection
with I-496 and would then parallel I-96 going north to the
control point at Grand River Avenue. This configuration
has the disadvantage of causing greater social and economic
disruption on the area, as well as trapping and reducing
the potential use of land between two highways. The elimi-
nation of these weakly conceived corridors was based mainly
on transportation grounds. Most of the MDOT analysis
centers on Corridors A, B, and B-1.

The M001 Location Study Report combined the transporta-
tion studies of the project with the findings of the Draft

Corridor EIS. It was recommended that I-69 be located in

120

in Corridor B. This showed the MDOT balancing transporta-
tion goals. The MDOT felt that although Corridor B was

less effective in serving the high volumes of long-distance
through and terminal travel demands than B-1 and C, it would
do a better job in enhancing and supporting regional
planning objectives. This was to outweigh the added user
costs and implementation costs attributed to Corridor 8.
Specifically, Corridor B provided better service to the City
of Grand Ledge, and it was thought to relieve the highly
traveled M-43 (Grand River Avenue) on the northwest side of
Lansing.

The evaluation of the corridors in the Draft EIS also
came to the same conclusion; "...a final route alignment
located within Corridor B will exert the least adverse impacts
in relative terms upon the total environmental system of
the project area..." (MDOT, 1971, 18). To summarize the
findings of the Draft EIS, the MDOT felt that the corridors
were relatively equal in impacting most Of the environment,
but that Corridor B-l had a higher impact on the natural
drainage pattern and would exert a higher impact on agri-
cultural productivity. The determination Of its impact on
agriculture was due to its diagonal location which went
against the grain of land parcels, and its effect on a large
muck area. Corridor B had an advantage because it traversed
the landscape with the land tenure patterns; that is, it

proceeds north and then directly east. Finally, in the

121

choice between Corridors B and A, the issues seemed to be
social and economic benefits and costs.

In comparing Corridors A and B the MDOT preferred the
rural route with less overall impact on factors of social
and economic value. Corridor A had the "...highest degree
of disruption of social and economic value patterns."
(M001, 1971, 18); the higher degree of disruption was
attributable to the fact that the corridor goes along the
existing US-27 right-of-way where development has already
developed. Thus, it could be concluded that the MDOT
selected the least disruptive corridor in considering all
impacts, and also opted for the corridor which provided for
more potential regional growth and moderate long-distance
service.

The issue of taking of prime agricultural farm land out
of production was addressed only in passing in the MDOT
corridor studies. The loss of short-term agricultural
productivity was mentioned, but this was countered with the
notion that the land might be lost to urbanization in the
near future anyway. Finally, the point was made that the
number of acres in production was effected to a greater
extent by the economic climate of agriculture than by
availability.

In summary the corridor planning process was successful
in quickly reducing the universe of potential routes to a

small number by adopting one corridor. Although the analysis

122

was all encompassing, the topics and depth of discussion
were very general. It should be pointed out that this
analysis was done in the early years of NEPA's implementa-
tion, and the MDOT studies seem to be little more than what
they were doing previous to the passage of the law. Both
major documents, the Location Study Report and Draft EIS,
were relatively short by recent standards; they were 60
pages and 18 pages respectively. The limited treatment
given to the issues in the corridor planning stage of the
I-69 project eventually contributed to the public debate

and reconsideration of the corridor.

Alignment Development

 

Given the approved Corridor B in which to develop
alternative route alignments, the task of the MDOT was not
very difficult. By February, 1973 a Draft Alignment EIS
had been prepared and ready for distribution and review.
In the Draft EIS the MDOT engineers had identified two
alternative alignments, and retained the options of main-
taining the US-27 existing right-of-way and doing nothing.
These last two alternatives were used to establish bench-
marks for comparative purposes, and this had become a
standard operating procedure. Both were dismissed as
impractical given the transportation objectives of the State
and of the Federal Interstate system. Attention was

focused on two alignments for a new highway.

 

123

The discussion of environmental impacts covered the
required questions as stated in the administrative regula-
tions, but on comparative grounds, each alignment was
described on a township by township basis on a small set of
factors. The factors reported were length of new construc-
tion, farm acreage required for right-of-way, number of
buildings removed for right-of-way, number of farms
segmented, woodlot acreage removed, environmentally sensi-

tive areas or zones crossed, distance from closest building

. ii‘fl -' ~— a’T-m. -.Jl-I_ n~'1u xv._ I.”

to nearest traffic lane, and distance from near traffic

lane at which minimum 70.0 dBA occurs. Where in the
corridor stage the task may have been nuigeneral, it was
practical in this stage to collect information and calculate
specific levels of impact.

The proposed alternatives, Alt. 1 and Alt. 2, are
presented in Figure 5.3. They differ in the first full mile
north of Charlotte, but then remain parallel and within a
half mile of each other until the corridor turns east.

After the turn, the alternatives head toward the I-496
interchange with Alt. 2 about one mile south of the other.
For the most part, the close proximity of the alternatives
to each other reduced the differences among them. Alt. 2
was the MDOT recommendation; in looking at the impacts
this alternative minimized, not only length of the route
but all other factors to some degree, except the number of

farms segmented. In addition the MDOT avoided a buried

123a

 

Figure 5.3 I-69 Final Alternative Alignments

124

outwash deposit and a navigation facility located in the
northern part of the corridor by selecting Alt. 2.

Alt. 2 became the "approved" alignment for receiving
federal funding. Upon completion of the planning process
the MDOT proceeded to move toward final engineering designs
and the purchase of right-of-way along the approved align-
ment.

Although it was in the alignment planning stage that the
specific impacts were considered, the selection of one
corridor and examining only options within that corridor
appeared to have greatly reduced what was actually considered.
From the public's or any conscientious reviewer's perspective,
the corridor stage lacked specific information to evaluate,
and the alignment stage did not leave much choice. It
would have been possible to have more discussion on align-
ments if there was more variety in design. The decision
process as carried out in this project demonstrated how a
potential conflict situation was reduced to almost a tech-
nical exercise in following procedures.

It does appear that the MDOT considered environmental
factors in the choice of an alignment, specifically the
impact on water recharge areas. The effect of this consi-
deration amounted to minor alterations in the alignment
route, curving to avoid these sensitive areas. The larger
issue of prime farm land received little assessment after

the corridor decision.

125

Reconsideration of the Approved Alignment

 

In response to the public criticism and political
pressure the M001 directed its staff to develop and analyze
new alignments in the US-27 corridor using the existing
right-of-way. The staff reported back to the Commission
in April, 1981 with its findings. There was enough evidence
in this report to have the Commission request that the MDOT
do further detailed studies of the new alignments. The
detailed analysis became the Supplement to the Final EIS.
Seven alternative alignments were examined in the Supplement,
and a public hearing was conducted on these new choices.
After the public hearing, all input was reviewed, and an
alternative alignment was recommended. The previously
approved alignment of new rural highway was rejected. The
recommended alternative was selected from the US-27 corridor.

The new alternatives, in addition to the original
approved alignment, are presented in Figure 5.4. In
Figure 5.4 the MDOT designed alignments which were rural
freeway cross sections. This alternative was considered in
segments. Thus there were six alignment combinations:

A-l, B-l, C-l

A-l, B-l, C-2

A-l, B-2, C-l

A-l, B-2, C-2

A-2, B-l, C-l

A-2, B-l, C-2

I”?

SEGMENT!

Figure 5.4

126

I-69 Reconsideration

 

Alignments

1mm.L"-nn a‘s—P

127

The last alternative alignment covered in the Supplement was
the combined rural-urban freeway which is shown in Figure
5.4. This alternative was offered by the consulting firm
hired by the CCAI. The M001 also reported that there were
some alternatives which were not being investigated after
the preliminary analysis. These were the rural freeway

options A-2, B-2, C-1, and A-2, B-2, C-2 because their

ml‘flnnra .1.me

"geometrics" would yield an unsatisfactory transportation

structure. Finally, the urban freeway alternative in a

p ‘4 I KI"..-

rural area was deemed impractical, costly, and in poor

design practice. The M001 discussion of I-69 in 1982
centered on the original alignment and these new alterna-
tives.

First, the original approved alignment was rejected.
There were four primary reasons for this decision which
reflected a change in attitude from one of accommodating
and promoting growth to minimizing disruption and encoura-
ging growth near existing developed areas. Alignments
using existing US-27 kept traffic and development within an
existing transportation corridor. The amount of farm land
required by the approved alignment was 905 acres compared to
the alignments along US-27 which needs only 140 to 215
acres. Rejecting the approved alignment also favored the
efficiency of servicing through travel over other factors.
Finally, it was recognized that the approved alignment would

have greater adverse wetlands impacts, both in terms of

128

acres affected and in degree. In the past the Fish and
Wildlife Service of the US Department of Interior and the
MDOT worked out acceptable mitigation measures for these
impacts when the Service recommended denial of permits.
Given that the MDOT, in its reanalysis, had other viable
alternatives which avoided wetland impacts, it was thought
that permits for the approved route would be difficult to
get.

Second, a new alignment had to be recommended from the
remaining seven proposed. After their assessment of the
alternatives, the MDOT selected the alignment, A-l, B-l,
C-l. The reasons given for his choice concentrated on
displacement of homes and businesses, costs, tax base loss,
agricultural impact, wetland impact, and construction impact.
The selected alignment was significantly better in minimi-
zing these impacts. Only in segment C-l was a wetland impact
identified and outweighed by other advantages. Mitigation
measures were developed to meet the objectives of the US
Fish and Wildlife Service.

The recommended alignment, that was finally adopted by
the M001 Commission on December 15, 1982, represented a
significant change by relocating the proposed freeway away
from prime agricultural land to a location near US-27.
Although it appeared that the MDOT made many adjustments as
a result of extensive public participation, it did not give

in to every demand. The Commission decided on a rural

 

129

interchange in segment B-l against local opposition which
supported an urban interchange. Local interests argued
that an urban interchange would take less farm land and
serve their community businesses better. The Commission

chose in favor of safety in rejecting the design.

Analysis of the Planning Process

 

The M001 carried out a two-stage planning process in
reaching a final decision on the location of I-69. By
design, the process of first selecting a corridor and then
an alignment within that corridor is a reasonable way to
narrow down the many possible decision Options. At each
phase the MDOT attempted to be as comprehensive as possible
in its treatment of route location issues, and an environ-
mental impact statement was drafted and reviewed. It was
evident in the description of the corridor and alignment
location decisions for I-69 that the way early decisions
were made affected the factors which played a part in the
final choice.

The corridor location decision is supposed to be the
principal place where social, economic, and environmental
impacts receive attention. In the corridor analysis of
I-69 there was a discussion of these impacts along with the
transportation objectives of the project, but the discussion

was very general. The general nature of this discussion

130

hindered the calculation of potential consequences, and on
specific issues, the MDOT put the discussion off until the
alignment phase. In contrast, there was much more informa-
tion given about the transportation benefits and costs of
various corridors. It was easier for the transportation
factors and related regional growth benefits to dominate the E
discussion and eventually to be part of the MDOT decision 1
calculus. .

Environmental factors were considered during the corridor

 

stage, but not as primary issues. Easily identified
environmental factors were used to eliminate a corridor.
It appeared that the M001 used environmental factors when
they were not in conflict with their agency goals. Where
some conflict existed, the conflict was left to the align-
ment stage and an engineering solution.

By the time a project reached the alignment stage, the
number of possible alternatives was greatly reduced. The
M001 offered only two alternative alignments for in depth
analysis. Somehow the jump in decision level from corridor
to alignment turned a potentially interesting and compli-
cated decision into a transportation engineering exercise.
The two alignments lacked differences, and the choice
was based on technical design factors and cost. Specific
environmentally sensitive areas were avoided where possible.
For the most part, those environmental factors remaining

received minor consideration, or if they happened to require

131

outside permits, mitigation of damage was treated and
resolved.

The I-69 route location project history pointed out the
problem inherent in the jump from corridor to alignment
planning. Public criticism and debate finally pressured
the M001 to consider the significant tradeoff that was made 1
in the selection of Corridor B and the construction of a .
new freeway over an existing roadway right-of-way. Although I
the M001 seemed to give in to the agricultural interests in

its reconsideration of I-69, the rejection of the approved 1

 

alignment was also based on transportation grounds. The
MDOT was not or did not want to be convinced that prime
farm land could alone determine route choice. Finally, the
preservation of prime farm land, as a policy, was given
weight by the passage of the Farmland Protection Policy Act
by Congress in December, 1981. This policy made it easier
for the MDOT to shift the premises behind its original
corridor choice from promoting growth to serving existing

develOpment and the agricultural economy of the area.

Conclusion

 

This description of the I-69 route location planning
and decision process showed that the M001 adopted a specific
strategy to reach a project decision in order to meet its

goals and comply with NEPA. The corridor - alignment, two

132

stage approach, is not new, but the MDOT implemented it in
such a fashion as to quickly reduce the number of choices,
and avoid the problem of debating the merits of conflicting
policies. Not only did the rapid narrowing of alternatives
avoid debate, but it kept the project decision within the
boundaries of the M001. Specifically, it was in the best
interest of the M001 not to provide a public forum for the
discussion of prime farm land policy if it could avoid it.
The better able the MDOT was at shaping the choice in

technical considerations, the better was the chance of

(PF-u“ 11F“-‘*.-1-‘—rg—.-q

project approval and compliance.

The reconsideration of I-69 pointed out what could,
environmentally speaking, be gained from a planning and
decision process which assesses more than one corridor or
has a larger set of alternative choices. With the existence
of other alternatives both the highway engineer and the
outside reviewer are not locked-in to accept essentially
one choice. For example, it was mentioned in the Final
Supplement that the US Fish and Wildlife Service usually
agreed to grant a permit for situations where environmental
damage could not be avoided by the recommended alignment
upon the development of a mitigation plan. If other alter-
natives are being studied which avoid environmental damage,
the permit is not so easily granted.

One of the major implementation mechanisms in NEPA is

the public participation and interagency review process.

133

Although procedural compliance was never an issue, critical
review was weakened by the narrow discussion on technical
transportation issues. The public must get involved early
in the process in order to have an impact. Experience has
shown that this tends to be contrary to human nature.
Finally, an overall evaluation of the MDOT's planning
decisionmaking performance under NEPA shows that specific
environmental features, which are mandated by protective
legislation, are avoided, while less specified environmental
factors do not fair well against the goals of the Depart-
ment. Compliance is reduced to procedural concerns and

following standard design practices.

CHAPTER SIX

Conclusions

 

 

D-
NEPA is one of the first laws to specify an implemen-
tation mechanism which described how decisions are to be
made in its policy mandate. Since its inception in 1970, <
L

the environmental impact statement (EIS) has become a given
operating procedure for planning any major federal or
federally funded project. Although its popularity is wide-
spread, NEPA and the EIS have been criticized by some
environmentalists as a step backwards; it diverts energy,
attention, and other resources away from major environmental
policy questions toward the sometimes mundane tasks of
collecting and analyzing project by project impacts and
decisions. Supporters of the NEPA approach cite the
benefits from having some way to review and "watch dog"
zealous public agencies from making narrow-minded choices
which significantly damage the natural environment. Both
critics and supporters alike have lacked the evidence to
substantiate their views convincingly.

The NEPA mandate is a decision reform. The reform could
be considered successful if agency decisionmakers adopt

environmental factors in their decision premises. The

134

135

primary purpose of the research presented here is to reveal
whether environmental factors are included in an agency's
essential variable set derivated from its premises. An
empirical investigation of one agency's record of decisions,
and a case study of one project decision were conducted to
develop an understanding of this agency's actual decision
behavior since the passage of NEPA.

It was assumed that agencies are boundedly rational
actors. Given their limited decisionmaking capabilities,
agencies simplify the tasks in the decision process and rely
on only a few important variables to reach a decision.

The choice of these variables is determined by the imple-
menting agencies Operating in a political context. Vague
mandates, such as NEPA's, which offer no solution to the
controversial issues they raise, are subject to and deter-
mined by the battle of conflicting pressures (West, 1984,
341). If NEPA was successful in influencing agency decision-
making, some environmental factors would be part of this
choice set. Thus, this research began with the idea that

no agency can be perfect in considering all environmental
impacts. The question for which answers were sought is,

are any environmental factors important to overall choice?

Findings
The empirical investigation focused on the final highway

route location decisions of the Michigan Department of

 

136

Transportation. Environmental factors were found not to be
very important in influencing route location if only the
final alignment choice is examined. The proposed trans-
portation facility's compatibility with the regional growth
and development plans, and project costs were strong factors
affecting the route alignment decision. In addition the
MDOT tended to favor routes which traversed new rather than
developed or build-up areas. A closer examination of the
data found that certain environmental factors were in fact
influential, but in the earlier design stages of both
corridor and alignment planning. Because these factors
were design constraintsin the process of developing the
alternatives, they were consistently absent as location
features of the final alignment alternatives. Those environ-
mental factors, which were used as design constraints, are
specified in specific pieces of legislation. These factors
were used to identify environmentally sensitive areas and
protected features which were avoided by the planners and
engineers in drafting alternatives. In contrast, the
ambiguity Of the NEPA mandate for a general environmental
policy made implementation and compliance uncertain.

The implementation strategy exhibited by the MDOT was to
use the early planning decisions to foreclose the final
decision Options to a very small set. Final decisions were
reduced to technical questions for which existing methods,

rules of thumb, were available to find solutions. Problems

137

associated with drainage, soil erosion and sedimentation,
floodplain construction, farm segmentation, relocation, and
others were handled with standard practices. In so doing
the MDOT retained most aspects of the project decision under
its control and authority. It also limited the scope of

the decision which remained outside its bounds. From the 2
project sponsoring agency's perspective, this increased the I
likelihood of reaching a decision with what it considered
to be the desirable results. The approach and subsequent

behavior of the MDOT is not peculiar to transportation

 

agencies, but a common bureaucratic trait to protect the
organization's self-interest.

The overall effect of such an implementation strategy
by project sponsoring agencies is that the general policy
mandated by NEPA, to plan with the harmony of man and nature
in mind, is, for the most part ignored, especially in the
early planning stages of‘a project where its impact could
be greatest. Specific environmental policies fared better.
Bureaucratic behavior in carrying out such an implementation
strategy used organizational processes to control the

political situation where possible.

Research Conclusions

 

There are a number of conclusions that can be drawn
from the findings of this research. First, legislating

rationality, as NEPA attempted to do, through procedural

138

reform is assured to fail unless the requirements for this
decision approach can be met. By requirements it is meant
that policies must be clearly stated and operational.
Experience with NEPA demonstrated that implementing agencies
adOpted the easiest route, that is, procedural compliance,
while lacking in substantive policy.

Second, the situations in which substantive policy 1
compliance was found, are those cases where important
environmental values are protected by specific legislation.
This suggests a future dilemma if every value ends up with
its own law.

Third, the case study pointed out the problem asso-
ciated with simplifying a complex decision. It is unavoid-
able that potentially good alternatives can be eliminated
in early stages of decisionmaking. With respect to I-69,
the corridor decision for a rural highway on a new route
was not based on information which matched the significance
of the decision. Decisionmakers need to weigh these
preliminary decisions on a par with later ones. In addi-
tion, the decision process, if it is going to become a
program or procedure, should have a way to retain variety
in the generation of alternatives and analysis.

Finally, the nature of environmental policy is
complex almost by definition, and it requires a problem-
solving approach in implementation and determining compli-

ance. As evident in the behavior of the MDOT, project

l39

sponsoring agencies strive to reduce the uncertainty
surrounding a policy in a way which can be counterproduc-
tive to environmentally sensitive solutions. Alternative
implementation approaches should be considered which can
handle error and learning in a way that can enhance

responsiveness, as well as rationality.

Implementation Approaches

 

For descriptive purposes two “ideal" implementation

approaches are offered for discussion. The first is

 

 

"programmed" implementation. This approach requires
clarity, precision, and comprehensiveness of policy. If
some of these qualities are lacking, the implementation
planner tries to: develOp detailed objectives, minimize
the number of actors, treat resistence by monitoring,
anticipating contingencies, and using guidelines. Overall
the goal is to minimize discretion at all levels (Berman,
1980, 208-9). The plan is geared toward attaining tech-
nical efficiency. The "programmed approach is similar to
what others have called management control systems. In
this same vernacular, "good management" has come to mean
efficient control procedures. The distinction between
management and control is lost (Landau and Stout, 1979,
149).

Whether it is called programmed implementation or

management control, this approach is politically appealing

140

to public administrators. It provides the public sector
with "accountability", and reduces for the administrator
political liability associated with policy failure. This
is all dependent on the degree of programmed implementation
that is possible. Programmed implementation, or "control",
has certain requirements. The most important requirement
is information.

The extent of our ability to control

is,..., a function of our knowledge.

An understanding of cause-effect

relationships in a specified domain

permits us either to produce or to

prevent some stipulated outcome.

If our knowledge is complete, we can

do so unerringly. (Ibid.)
Thus, the ability to control demonstrates that the "means"
are no longer problematic.

The second "ideal" approach is "adaptive" implemen-
tation (Berman, 1980, 210-213). It is at the other end of
the spectrum. It begins with the assumption that knowledge
vital to implementation is incomplete. Generally speaking,
this approach allows policy to be Specified, modified, and
revised. Policy is adapted according to the interaction
of policy and its institutional setting. It is concerned
with establishing acceptable rules that allow multiple
participants to negotiate and compromise during the course
of implementation. This type of active participation
enhances "problem-solving“. Instead of certainty, policy

is treated as a hypothesis to be tested. Adaptive

implementation recognizes a potential for error and seeks

l4l

knowledge to clarify policy and eventually permit control.
Implementer discretion is encouraged in this discovery
process. In so doing, the "adaptive" approach to imple-
mentation separates, not combines, management from control.
Control and management approaches to policy implementation

suggest corresponding ideas of agency compliance.

Types of Compliance

 

Under programmed implementation, operating procedures
are control routines. Compliance is stressed, and here
it means sticking to rules. Any behavior that deviates
from standard procedures is perceived as an error. Like-
wise, any exercise of discretion equals error, and all
organizational effort goes to protect against error. This
type of compliance is easiest to determine; "...departure
from procedure appears highly visible, clearly measurable
and recognizably blameworthy" (Dunshire, 1978, 99).
Behavior is labeled correct or incorrect. Although this
strict control notion of compliance is an extreme case, it
points out an inherent problem. The problem of which
policymakers must be made aware is the ease by which
policy mgggg can be transformed to policy Eggg. NEPA
presents a good example where this could happen.

NEPA's decision process, as outlined in its admin-
istrative regulations, leads to the final informed agency

decision. The product of the decision process is the EIS.

142

Producing an EIS can, in some instances, become "compli-
ance". An agency just follows the content and review time
requirements to the letter of the federal guidelines.
Given the complexity and ambiguity of NEPA's substantive
policy, more emphasis may be placed on procedural correct-
ness than on the adequacy of the EIS as a decision document. F
Importantly, the hypothetical character of NEPA's policy
direction can be displaced, and the discrepancies between

agency activities and administrative rules can be perceived

 

and labeled "compliance problems" (Landau and Stout, 1979,

lr—‘_"'

153). Compliance would carry a different meaning if
"adaptive" implementation is practiced. Under adaptive
implementation the process of carrying out policy is a
management process. Implementation procedures are
suggested, and variance from these procedures is considered
informational. The overseer learns from the behavior of
the implementer, who, in turn, is learning about the
policy. The assumption that the theory is right is not
made. Appropriate means are sought out. On the value side,
the practice of adaptive implementation realizes that
imposing a change in values is not a simple operation,

but it requires some negotiation. Negotiation takes place
between the overseer and the implementer, and the meaning
of compliance comes out of their relationship; it is a
condition of the relationship (Dunshire, 1979, 23).

Compliance might be described as carrying out a policy in

143

a "zone of acceptance", which is maintained by agreement.
While agencies are not expected to behave according
to either ideal type of implementation, they can be
expected to operate according to a mix of programmed and
adaptive processes. The mix of processes is determined
by the scope and nature of the policy and policy context.
By definition adaptive implementation, which emphasizes
problem-solving and learning, brings a policy's execution
toward more routine and programmed activities. In areas
of policy where there is adequate knowledge and a consensus
on what is acceptable, programmed implementation and control
can be expected to rule. Where either the facts or
values are in dispute, implementation is better served
by an adaptive approach and managing style. In the case
of pursuing a programmed-control approach in face of
uncertainty, a risk is taken. The risk is that such an
instrumental approach may be unacceptable and "inappro-
priate for dealing with many of the broad policy issues
agencies must confront" (West, 1983, 326). It constrains
agency decisionmaking from learning and performing effec-
tively in dealing with the political functions implicit

in administrative mandates.

Complying with NEPA

 

An attempt was made earlier to discern the decision

premises from the law itself and its implementing

144

regulations. This analysis failed to show any specific

or operational fact and value premises. This is not out
of the ordinary for Congressional legislation to leave the
details of implementation to the agencies. It is ironic
though that this is also the case in NEPA which attempts
to impose comprehensive and systematic decisionmaking. W
Having such a general policy in a very complex problem F
area is not helpful to the agencies charged to implement

the law. Any difficulty in putting NEPA's policies in a

 

specified program of action leads to difficulty measuring A
compliance. If the implementation of NEPA is in part
discretionary, determining compliance is not obvious, but
the result of an understanding between overseer and
implementing agency.

Given the scope and nature of NEPA's mandate, a mixed
implementation approach seems most probable. Because
compliance follows the type of implementation strategy
adopted by the agency and approved by the overseer, the
essential variable set will be composed of control factors
which are easily checked and a small number of discretion-
ary factors which are agreed on. This study found evidence
for the essential variable set. It is consistent with the
theory presented above that there is a greater likelihood
that programmed factors from specified legislation are
part of essential set, and they measure compliance; less

likely for others.

LIST OF REFERENCES

LIST OF REFERENCES

Aldrich, John and Charles Cnudde. 1975. Probing the
bounds of conventional wisdom: a comparison of
regression, probit, and discriminant analysis. American
Journal of Political Science XIX (3) August: 571-607.

Allison, Graham. 1971. Essence of decision: explaining
the Cuban missile crisis. Boston: Little, Brown, and
Company.

 

 

Amemiya, Takeshi. 1981. Qualitative response models: a
survey. Journal of Economic Literature 19 December:
1483-1536.

Alexander, Ernest. 1979. The design of alternatives in
organizational contexts: a pilot study. Administrative
Science Quarterly 24:382-404.

 

 

Andrews, Richard N.L. 1976. Environmental policy and
administrative change. Lexington: D.C. Health,

 

 

. 1973. Philosophy of environmental assessment.
Journal of Soil and Water Conservation Sept.-Oct.:
l97-207.

 

Bardach, Eugene. 1977. The implementation game: what
happens after a bil becomes a law. Cambridge, MA:
The MIT Press.

 

 

, and L. Pugliarsi. 1977. The environmental
impact statement vs. the real world. The Public
Interest 49:22-38.

 

 

Bartlett, Randall. 1973. Economic foundations of political
power. New York: The Free Press.

Berman, Paul. 1980. Thinking about programmed and
adaptive implementation: matching strategies to
situations. In H.M. Ingram and 0. Mann. Why_policies
succeed or fail. Beverly Hills: Sage.

 

Brewer, Gary and Peter deLeon. 1983. The foundations
of policy analysis. Homewood, Illinois: Dorsey Press.

 

145

146

Caldwell, Lynton. 1979. Is NEPA inherently self-defeating?
Environmental Law Rgporter 9, 1:50001-50008.

 

. 1975. Man and his environment: policy
and administration. New York: Harper and Row
Publishers.

 

 

 

Council on Environmental Quality. 1976. Environmental
impact statements: an analysis of six year's experience
py seventy federal agencies. Washington, D.C.

 

 

 

Downs, Anthony. 1967. Inside bureaucracy. Boston: Little,
Brown, and Company.

 

Dreyfus, D., and Helen Ingram. 1976. The National Environ-
mental Policy Act: a view of intent and practice.
Natural Resources Journal 16, 2:243-62.

 

Duke Law Journal. 1976. Appropriate scope of an EIS: the
interrelationship of impacts. Duke Law Journal: 623-
649.

 

Dunshire, Andrew. 1978. Control in a bureaucrpcy. New
York: St. Martin's Press.

 

Environmental Law Reporter. 1979. Comments - New rules
for the NEPA process: CEQ establishes uniform proce-
dures to improve implementation. Environmental Law
Reporter 9, 1:10005-10010.

 

Fairfax, Sally. 1978. A disaster in the environmental
movement. Science 199, 4330: 743-47.

Forester, John. 1984. Bounded rationality and the
politics of muddling through. Public Administration
Review Jan.-Feb.:23-3l.

 

Friesema, H.P., and P. Culhane. 1976. Social impacts,
politics, and the environmental impact statement
process. Natural Resources Journal 16, 2:339-356.

 

Hensher, D.A., and L.W. Johnson. 1981. Applied discrete
choice modelling. New York: Halsted Press.

 

 

Hill, W.W., and L. Ortolano. 1978. NEPA's effect on the
consideration of alternatives: a crucial test. Natural
Resources Journal 18:2:285-312.

 

Janis, 1., and L. Mann. 1977. Decisionmaking. New York:
Free Press.

 

147

Kmenta, Jan. 1971. Elements of econometrics. New York:
Macmillan.

 

Landau, M., and R. Stout. 1979. To manage is not to
control: or the folly of type II errors. Public
Administration Review. March/April, 148-156.

 

Lindblom, Charles and David K. Cohen. 1979. Usable
knowledge: social science and social problem solving.
New Haven: Yale University Press.

 

Liroff, R.A. 1980. NEPA - Where have we been and where
are we going? American Planning Association Journal
April: 154-161.

 

Mack, Ruth P. 1971. Planning on uncertainty. New York:
John Wiley and Sons.

 

March, James. 1978. Bounded rationality, ambiguity, and
the engineering of choice. Bell Journal of Economics
Autumn: 587-607.

 

, and Herbert Simon. 1958. Organizations.
New York: John Wiley and Sons.

 

 

Mazmanian, Daniel, and Jeanne Niebaber. 1979. Can ocgani-

 

zations change? Environmental protection, citizen
participation, and the Corps of Engineers. Washington,
D.C.: Brookings Institute.

 

 

, and Paul Sabatier. 1983. Implementation
and public policy. Glenview, Illinois: Scott,
Foresman and Company.

 

 

 

McDermott, James. 1979. Improving NEPA: new regulations
of the Council on Environmental Quality. Boston
College Environmental Law Review 8:89-118.

 

McFadden, D. 1975. The revealed preferences of a govern-
ment bureaucracy: theory. Bell Journal of Economics
and Management 6, 2:401-416.

 

 

1976. The revealed preferences of a
government bureaucracy: empirical evidence. Bell
Journal of Economics and Management 7, 1:55-72.

 

 

Mintzberg, Henry, D. Raisinghani, and A. Theoret. 1976.
The structure of unstructured decision processes.
Administrative Science Quarterly 21:246-275.

 

148

Niskanen, William. 1971. Bureaucracy and representative
gpvernment. Chicago: Aldine-Atherton,

 

Olsen, James M. 1981. Michigan environmental law: a
guide to environmental law. Traverse City: Neahtaw-
wanta Press.

 

 

Ortolano, Leonard. 1983. Integrating environmental
considerations into infrastracture planning. In
Frederick Rossini and Alan Porter ed. Integrated
Impact Assessment. Boulder: Westview Press.

 

 

Ostrom, C.W. 1978. A theory of the budgetary process as
a research program in public administration. Paper
presented at Hickory Corners, Michigan. May 15-17.

, and Brian Job. 1982. The President and
the political use of force. Paper presented at the
1982 Annual Meeting of the APSA. September 2-5.

 

Owen, Bruce and Ronald Braeutigam. 1978. The regulation
game: strategic use of the administrative process.
Cambridge: Ballinger Publishing Company.

 

Pindyck, Robert and Daniel Rubinfeld. 1976. Econometric
models and economic forecasts. New York: McGraw-Hill
Book Company.

 

 

Pressman, Jeffrey and Aaron Wildavsky. 1973. Implemen-
tation. Berkeley: University of California Press.

 

Roberts, Marc J., and J. Bluhm. 1981. The choices of
ppwer: utilities face the environmental challenge.
Cambridge: Harvard University Press.

 

Rossini, Frederick and Alan Porter. 1983. Intggrated
impact assessment. Boulder: Westview Press.

 

 

Sabatier, Paul. 1978. The acquisition and utilization of
technical information by administrative agencies.
Administrative Science Quarterly 23 (September):
396-417.

Simon, Herbert. 1957. Models of man. New York: John
Wiley and Sons.

 

1959. Theories of decisionmaking in
economics. American Economic Review 49:253-283.

 

 

149

. 1976. Administrative behavior. A
study of decisionmaking processes in administrative
organizations. 3rd Ed. New York: Free Press.

 

 

 

 

. 1978. Rationality as process and as
product of thought. American Economic Review 68(2):
1-13.

 

 

Steinbrunner , John. 1974. The cybernetic theory of
decision. Princeton, N.J.: Princeton University
Press.

 

Stout, Russell. 1980. Management or control. Blooming-
ton, In.: Indiana University Press.

 

Wegner, J.W. 1975. Comment: planning level and program
impact statements under NEPA: a definitional approach.
UCLA Law Review 23:124-163.

 

West, William. 1983. Institutionalizing rationality in
regulatory administration. Public Administration
Review July/August: 326-334.

 

1984. Structuring administrative
discretiOn: the pursuit of rationality and responsive-
ness. American Journal of Political Science 28(2):
340-360.

 

 

Wichelman, Allan. 1976. Administrative implementation of
the National Environmental Policy Act of 1969: a
conceptual framework for explaining differential
response. Natural Resources Journal 16, 2:263-300.

 

Yale Law Journal. Implementation of the environmental
impact statement. The Yale Law Journal 88:596-611.

 

LIST OF REFERENCES

Government Documents

 

State of Michigan

 

1.

Michigan Department of Transportation. Draft Corridor
Environmental Impact Statement for I-69. Charlotte
to I-96g Northwest of Lansing. 1971.

 

 

. Final Corridor Environmental Impact
Statement for I-69, Charlotte to I-96, Northwest Of

Lansing. 1972.

 

 

. Location Study Report for I-69,
Charlotte to Lansing, Eaton and Clinton CountiesL
Planning Corridor Analysis. 1972.

 

 

 

Draft Alignment Environmental Impact

 

 

Statement for 1:69, Charlotte to I-96, Northwest of
Lansing. 1973.

Action Plan. 1974.

 

 

Final Alignment Environmental Impact

 

 

Statement for 1-69, Charlotte to I-96, Northwest of
Lansing. 1976.

Proposed Construction of I-69, From

 

the Temporary I-69/US-27BR Intersection in Charlotte
to the I-96/I-496 Interchange, Southwest of Lansing,
Eaton County, MiEhigan. A Report to the Michigan
Transportation Commission. 1981.

Draft Supplement to the Final

 

Environmental Impact Statement for I-69, Charlotte to
Lansing, Eaton County. 1982l

Final Supplement to the Final

 

 

Environmental Impact Statement for I-69, Charlotte
to Lansing, Eaton County. 1983.

150

151

 

 

 

 

 

Federal

l. U.S. Department of Transportation. Environmental
Assessment Notebook Series: Highways. 1975.

2. . Federal Laws, Regulations, and
Material Relating to the Federal Highway Administration.
1970.

3. Review of the Proposed Routing of

 

 

Interstate 69 between Lansing and Charlotte, Michigan.
1980.

CASE STUDY INTERVIEWS

Ataman, M. Tarik. Engineer, Route Location, Design
Division, Michigan Department of Transportation.
Interview, November, 1982.

Chisholm, Philip A. Public Hearings Coordinator,
Public Involvement Section, Michigan Department of
Transportation. Interview, August, 1984.

Hartwig, William. Transportation Planner, Highway
Planning Division. Interviews, 1982-84.

Health, A. Ross. Route Location Supervising Engineer,
Route Location, Design Division. Interviews, 1982-84.

Jones, Ronald. Division Engineer, Michigan Division,

Federal Highway Administration. Interview, August,
1983.

152

 

      

RlES

1

1293

 

M'111111111111llllllllllllll