INFORMATION TO USERS

T his rep ro d u ctio n w as m ad e from a copy of a m a n u s c rip t s e n t to u s for p u b lication
a n d m ic ro film in g . W hile th e m o st a d v a n c e d tec h n o lo g y h a s b e e n u s e d to p h o ­
tograph a n d rep ro d u ce th is m a n u s c rip t, th e q u a lity of th e rep ro d u ctio n is heavily
d ependent u p o n th e q u a lity of th e m aterial su b m itte d . Pages in any m a n u s c rip t
may have in d is tin c t p rin t. In all cases th e b e s t available copy h a s b een filmed.
The following e x p la n a tio n of te c h n iq u e s is provided to help clarify n o ta tio n s w hich
may a p p e a r on th is rep ro d u ctio n .
1. M a n u sc rip ts m ay n o t always be com plete. W hen it is n o t possible to o b tain
m issin g p a g e s, a n o te ap p e a rs to in d icate th is.
2. W hen copy rig h ted m aterials are rem oved from th e m a n u sc rip t, a note a p ­
p e a rs to in d ic a te th is.
3. Oversize m a te ria ls (m aps, draw ings, a n d c h a rts) are p h o to g rap h ed by sec­
tio n in g th e original, b e g in n in g at th e u p p e r left h a n d c o m e r a n d c o n tin u ­
ing from left to r ig h t in equal se ctio n s w ith sm all overlaps. E ach oversize
p ag e is a ls o film ed a s o n e e x p o s u re a n d is av ailab le, for a n a d d itio n a l
charge, a s a s ta n d a r d 35 m m slide o r in b lack a n d w hite p a p e r form at.*
4. M ost p h o to g ra p h s rep ro d u ce acceptably o n positive m icrofilm o r m icro­
fiche b u t lac k clarity on xero g rap h ic copies m ade from th e m icrofilm . For
a n a d d itio n a l c h a rg e , all p h o to g ra p h s a re availab le in b lac k a n d w h ite
s ta n d a rd 3 5 m m slid e form at. *

*For more inform ation about black and w hite slides o r enlarged paper reproductions,
please contact the D issertations Custom er Services Department.

DM

’XJmMershy

Microfilms
. International

8603457

N a h ra , J o h n S i

LEGISLATIVE ST A FF AND LEGISLATOR ACCOUNTABILITY: AN
EXAMINATION O F LEGISLATIVE EN T E R PR ISE S IN MICHIGAN

P h .D .

M ichigan S ta te U n iv e rs ity

University
Microfilms
International

300 N. Zeeb Road, Ann Arbor, Ml 48106

Copyright 1985
by
Nahra, John Si
All Rights Reserved

1985

PLEASE NOTE:

In all c a s e s th is m a te ria l h a s b e e n film ed in th e b e s t p o ssib le w a y from th e a v a ila b le co p y .
P ro b le m s e n c o u n te r e d w ith th is d o c u m e n t h a v e b e e n id en tifie d h e re with a c h e c k m ark
V

1.

G lo ssy p h o to g r a p h s o r p a g e s ______

2.

C o lo red illu stra tio n s, p a p e r o r p r i n t _______

3.

P h o to g ra p h s w ith d a rk b a c k g r o u n d _____

4.

Illu strations a r e p o o r c o p y _______

5.

P a g e s with b la c k m a rk s, n o t o rig in al c o p y

4- - ~

6.

Print sh o w s th r o u g h a s th e r e is te x t o n b o th sid es of p a g e ________

7.

Indistinct, broken or small print on several pages

8.

Print e x c e e d s m a rg in r e q u ir e m e n ts ______

9.

Tightly b o u n d c o p y with p rin t lo s t in s p i n e ________

10.

^— -

C o m p u te r p rin to u t p a g e s w ith in d istin c t p r in t_______

11.

P a g e ( s ) ______________lack in g w h e n m aterial receiv ed , a n d not a v aila b le from s c h o o l o r
au th o r.

12.

P a g e ( s ) ______________ s e e m to b e m issin g in n u m b erin g only a s tex t fo llo w s.

13.

Two p a g e s n u m b e r e d

. T ex t follow s.

14.

C urling a n d w rin k le d p a g e s _______

15.

D issertation c o n ta i n s p a g e s w ith p rin t a t a slant, film ed a s received

16.

.

O th er

________

University
Microfilms
International

Legislative Staff and Legislator Accountability:
An Examination of Legislative Enterprises in Michigan

By
John Si Nahra

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Political Science

1985

Copyright by
John Si Nahra
1985

ii

ABSTRACT

Legislative Staff and Legislator Accountability:
An Examination of Legislative Enterprises in Michigan
By
John Si Nahra

Legislator accountability is a mainstay of representative
democracy.

In this research,

legislative staffs’ impact on

legislator accountability is analyzed.

Accountability takes

many specific forms, all of which presume that legislators are
responsible for their actions.

The assignment of responsi­

bility requires knowing whose interests exert influence (are
legislators in control) and how these interests combine (is
staff discretion tolerable and manageable).

These concerns are

formalized and tested using data from the Survey of Legislative
Staff in Michigan,

1984, supplemented by data from various

secondary sources.

Emphasis is placed on staff access to

legislators as the indicator of staff use. Access is defined by
the nature and extent of interactions with legislators reported
by staff, and converted into temporal measures.
staff access, attitudes,

The measures of

and assignment, provide the empirical

bases for hypothesis testing. Remarks made by legislators in
personal interviews,

as well as staff comments,

lend credence

to the empirical findings and reaffirm that staff have been
assimilated into Michigan’s legislative environment without
sacrificing legislator accountability.

DEDICATION

To my bride, who knew what she was getting into and
encouraged me to go ahead.
This is ours, too.

ACKNOWLEDGEMENTS

This research could never have been completed without the
assistance and advice of numerous individuals.
Members of the
Political Science Department at Michigan State University
provided the opportunity.
Frank Pinner, my early mentor,
David Rohde, the Department Chairman, and Gary Miller, my
dissertation committee chairman, all gave me the support and
latitude needed to pursue a non-traditional academic
timetable.
The willingness of my employer, Symond Gottlieb,
to support ray academic pursuits was also critical to the
commencement of this project.
The actual conduct of The Survey of Legislative Staff in
Michigan, 1984, required the cooperation of numerous staff who
took time from their other responsibilities to complete the
survey.
Three people in particular were a source of valuable
advice:
Kevin Seitz, then with the House Fiscal Agency,
Patience Drake, an analyst with the House Fiscal Agency, and
William Ryan, former Speaker and Clerk of the Michigan House
of Representatives.
Specific legislators who gave generously
of their time include Representatives Robert Emerson,
Lynn Jondahl, David Hollister, Deborah Stabenow, and Senator
Lana Pollack.
The patient typing and revising of text provided by
Chriss Olivo and the typing of tables by Mona Torp produced
this document.
Their aid was indispensable.
Finally, a special thanks to Arnie Greenfield who provided a
sounding board for ideas that needed to be discussed and
developed before they could be used.

iv

TABLE OF CONTENTS

Introduction
Chapter 1:

Chapter 2:

Legislator Accountability
•

Defining Legislator Accountability

•

Prior Views on Staff and Legislators
- Proponents
- Critics

•

Why Staff Numbers Have Grown
- Growth of Advisors Generally
- Growth of Legislative Staff Specifically

•

Access
- Describing Staff Access
- Relating Access to Accountability
- Control
- Discretion

•

Summary

Describing Michigan Staff:
Their Development
and Interactions with Legislators
•

Staff Development
- Size of State
- Level of Interaction Among Governmental
Entities
- Political Structure
- House
- Senate
- House and Senate Compared
- Joint Agencies

•

Legislator-Staff Interaction
- Staff
- Reported Characteristics of Interactions
- Written Products
- Work Environment
- Attitudes
- Legislators

•

Summary
v

Chapter 3:

Chapter 4:

Chapter 5:

Access and Accountability
o

Access
- Advisory Access Generally
- Staff Access Specifically
- Sunary

o

Formal Argument
- Legislators Control Staff
- Staff Discretion is Tolerable
- Staff Discretion is Manageable
- Summary

Empirical Results
o

Data Analysis Plan
- Measuring Interaction Time
- Linking Interaction Time Levels to
Hypotheses

o

Hypothesis Testing
- Legislator Control
- Discretion Being Tolerable
- Discretion Being Manageable
- Summary

Legislator and Staff Observations
o
o

Chapter 6:

Legislator Comments
Staff Comments

Insights and Implications

Appendix A : Survey of Legislative Staff in Michigan,
Instrument

1984

Appendix B : Survey of Legislative Staff in Michigan,
Codebooks

1984

vi

LIST OF TABLES

Table 1:

A Comparison of Growth in Michigan Governmental
Revenues and Growth of Michigan Legislative Staff
for Various Years Between 1960 and 1983

Table 2:

Michigan Governmental Revenues Characterized by
Source, Actual Amounts and Relative Percents, for
Various Years Between 1960 and 1983

Table 3:

A Comparison of Personal and Committee Staff
Allocated to Republicans and Democrats in the
Michigan State Senate Under a Democratic Majority
(1983) and a Republican Majority (1984)

Table 4:

A Comparison of Professional Legislative Staff
Arrangements in the Michigan House and Senate for
Various Years Between 1960 and 1983

Table 5:

Survey of
Legislative Staff in Michigan,
Results of Survey Distribution

1984:

Tab le 6:

Survey of Legislative Staff in Michigan,
Actual Responses, Usable Responses, and
Corresponding Response Rates

1984:

Table 7:

Survey of
Legislative Staff in Michigan, 1984:
Percentage Comparison of Surveyed Population and
Respondent Population

Table 8:

Survey of
Legislative Staff in Michigan, 1984:
Frequency and Duration of Contacts Between
Legislators and Staff

Table 9:

Survey of Legislative Staff in Michigan, 1984:
Initiation and Nature of Contacts Between
Legislators and Staff

Table 10: Survey of Legislative Staff in Michigan, 1984:
Subjects Discussed in Legislator-Staff Contacts
Ordered by Frequency of Mention and Priority

Table 11: Survey of Legislative Staff in Michigan, 1984:
Staff Tenure Categorized by Prior Experience and
Current Work Arena for All Staff Respondents

vii

Table 12: Survey of Legislative Staff in Michigan, 1984:
Characteristics of Staff Work Environment for
Total and Final Respondents
Table 13: Survey of Legislative Staff in Michigan, 1984:
Summary of Staff Personal Attributes for Total and
Final Respondents
Table 14: Legislator Tenure by Chamber and Party of Actual
Legislator Complement and Legislators Identified
by Staff in the Survey of Legislative Staff in
Michigan, 1984
Table 15: Number of Committees by Chamber and Party for the
Actual Legislator Complement and Legislators
Identified by Staff in the Survey of Legislative
Staff in Michigan, 1984
Table 16: Number of Staff Associated with Legislators by
Chamber and Party for the Actual Legislator
Complement and Legislators Identified by Staff in
the -Survey of Legislative Staff in Michigan, 1984
Table 17: Survey of Legislative Staff in Michigan, 1984:
Views of Staff Work Ranked First and Second by
Final Staff Respondents
Table 18: Calculation of Legislator-Staff Interaction Times
Table 19: Frequency Distributions for the Components of
Legislator-Staff Interactions and Resulting
Interaction Times from the Survey of Legislative
Staff in Michigan, 1984
Table 20: Survey of Legislative Staff in Michigan, 1984:
Number of Legislator Contacts Reported by Staff
and Number of Staff Contacts Associated with
Legislators
Table 21: Survey of Legislative Staff in Michigan, 1984:
Estimated Total Time per Month Spent Interacting
with Staff (for Legislators) and with Legislators
(for Staff)
Table 22: Survey of Legislative Staff in Michigan, 1984:
Legislator Involvement in Staff Hiring and
Retention Categorized by Staff Superiors and Staff
Identification with Legislators

Table 23: Survey of Legislative Staff in Michigan, 1984:
Staff Attitudes Reflecting a Service Orientation

viii

Table 24: Survey of Legislative Staff in Michigan, 1984:
Staff Attitudes Comparing Staff Influence to
Influence of Other Legislative Participants
Table 25: Survey of Legislative Staff in Michigan, 1984:
Comparison of Mean Staff Time Devoted to
Legislator-Staff Contacts Categorized by
Working Relationship
Table 26: Survey of Legislative Staff in Michigan, 1984:
Comparison of Distributions for Staff Time Devoted
to Legislator-Staff Contacts Categorized by
Attributes of the Legislator-Staff Working
Relationship
Table 27: Comparison of Mean Number of Staff Assigned to
Legislators Categorized by Various Attributes of
Legislator Authority
Tab le 28: Comparison of Distribution for Number of Staff
Assigned to Legislators Categorized by Various
Attributes of Legislator Authority
Table 29: Results of Bivariate Regressions for Members of
House and Senate Regressing Number of Staff
Assigned to Legislators by Legislators Tenure
Table 30: Survey of Legislative Staff in Michigan, 1984:
Comparison of Mean Legislator Time Devoted to
Legislator-Staff Contacts Categorized by Various
Attributes of Legislator Authority
Table 31: Survey of Legislative Staff in Michigan, 1984:
Comparison of Distributions for Legislator Time
Devoted to Legislator-Staff Contacts Categorized
by Various Attributes of Legislator Authority
Table 32: Survey of Legislative Staff in Michigan, 1984:
Results of Bivariate Regressions for Members of
the House and Senate Regressing Time Spent by
Legislators in Legislator-Staff Contacts by
Legislator Tenure
Table 33: Survey of Legislative Staff in Michigan, 1984:
Comparison of Mean Times and Distributions for
Time Devoted to Legislator-Staff Contacts
Categorized by Initiator of Contact

Table 34: Survey of Legislative Staff in Michigan, 1984:
Comparison of Mean Staff Time Devoted to
Legislator-Staff Contacts Categorized by Staff
Personal Work Orientations

ix

Table 35: Survey of Legislative Staff in Michigan, 1984:
Comparison of Distributions for Staff Time Devoted
to Legislator-Staff Contacts Categorized by Staff
Personal Work Priorities
Table 36: Survey of Legislative Staff in Michigan, 1984:
Comparison of the Percent of Total Interaction
Time and Percent of Total Legislator-Staff
Contacts Reported by Staff Attributable to Varous
Groups of Legislators Arrayed by Total Time
Devoted to Legislator-Staff Contacts

x

LIST OF FIGURBS

Figure 1:

Organizational Chart for Staff Agencies of the
Michigan Legislature, 1984

xi

INTRODUCTION

Legislative staff retained by American legislatures at both
the state and federal levels have matured from a cause
celebre of post-World War II legislative reformers to become
a burgeoning profession which,
contemporary observers,

according to some

threatens to change the very

legislative process staff were designed to serve.

The rise

and institutionalization of any new cadre of actors in the
legislative arena is a development of considerable import.
Legislative staff are no exception.

To explore the

implications of staff development and use, this research
will describe and analyze interactions between legislators
and staff in one state, Michigan.

Data will come from a

survey of professional staff working for the Michigan House
or Senate during 1984, followed by personal interviews with
legislators having high levels of reported staff contact.

The purpose of this analysis goes beyond a single state.
Michigan's legislator-staff relations will be used to
examine the effect of staff on legislator accountability.
In the process,

a view of staff and an approach to

documenting their function is suggested that,

if useful in

Michigan, may have relevance to other legislative settings.

2

The utility of this approach will be assessed by how well it
furthers understanding of several critical issues concerning
legislator accountability.

Chapter Synopses

in Chapter 1, the definitions and logic supporting the
research are elaborated.

Legislators’ maintenance of their

accountability is seen as central to representative
democracy.

Various forms of legislator accountability are

defined as expectations others place upon the performance of
legislators.

The nature of these expectations are

determined by how decision making is perceived.

Some look

for rationality or predictability in legislative
performance.

The view adopted here is that it may not be

reasonable to always expect rationality or predictability;
but, it is reasonable to expect that legislators are
responsible for creating and maintaining the interactions in
which they participate.

The presumption of such

responsibility is the basis for all forms of accountability.

The literature on legislative staff reveals a concensus that
staff do influence legislator accountability.

However,

sharp differences of opinion exist concerning whether this
influence is positive or negative.

Early proponents

advocated more staff to provide legislators with the
resources needed to carry out their responsibilities and

3

maintain accountability.

Later critics have viewed the rise

of staff with some alarm pointing to their independence and
potential influence as threats to legislators rather than
aids.

Resolution of this dilemma lies at the heart of the

research.

To begin this resolution,

the analysis of legislative staff

must emphasize relevant concerns.

Placing staff in their

proper historical context provides this focus.

Staff are

seen as a contemporary manifestation of the political
advisory function found throughout history.

Once this

historical link is made, features of staff worth stressing
emerge.

Both proponents and critics of staff base their

arguments on staff growth.
misplaced.

I maintain this emphasis is

Staff growth in America’s legislatures is

consistent with the historic growth of political advisors.
Qiven the proper conditions, growth in the numbers and types
of advisors is to be expected and offers little evidence as
to whether staff help or hinder legislator accountability.
Of more direct concern to legislator accountability is the
nature and extent of staff access to legislators.

In Chapter 2, the relevance of staff growth and access to
legislators in the Michigan legislative environment are
reviewed.

The development of staff and their use are

described using primary and secondary data sources compiled
for this research.

Different staff structures are found in

the two legislative chambers.

The House is characterized by

4

centralized staffing while the Senate has a more
decentralized use of staff with staff assignment being
associated with individual legislators.

Chapter 3 integrates access into an analysis of
accountability through the exploration of what legislative
staff scholars term the Legislative Enterprise.

Under this

approach legislators are not viewed as individuals, but as
managers of subordinates

(staff).

Staff access defines the

dimension of each legislator's enterprise.

The nature and

context of these legislator-staff interactions permit an
assessment of who controls the Legislative Enterprise and
how much discretion staff are permitted.

Control and

discretion become the principal indicators used to gauge
legislator accountability.

To the extent that legislators

control their enterprises and the associated staff
discretion,

they maintain their accountability;

to the

extent that legislators lack control and staff discretion
lacks limits,

legislator accountability is threatened.

This

argument is formalized with three assertions, one on control
and two on discretion.

Hypotheses designed to test the

validity of these assertions are formulated.

Chapter 4 begins with a discussion of how access is
empirically defined.

An approach to the measurement of

interaction time for legislators and staff is devised.

This

chapter also offers conclusions about the hypotheses derived
from statistical interpretations of the data.

All three

5

assertions find support indicating that, at least in
Michigan,

legislators are able to control staff and that

they find staff discretion both tolerable and manageable.

Chapter 5 reinforces these empirical findings with
observations made in semi-structured interviews with the top
five legislators most frequently mentioned by staff as well
as comments offered by staff regarding their work.

Chapter 6 concludes the research by summarizing the insights
gained here as well as their implications for future
analysis of legislative functioning.

CHAPTBR 1
LEGISLATOR ACCOUNTABILITY

Representative government rests on the presumption that
legislators are accountable for the performance of their
public duties.

Elections and other electoral devices,

as primaries and recalls,

such

act as the final arbitrators of

legislator accountablity by placing periodic value judgments
on legislators.

These judgements presume that legislators

are (or should be) accountable for the conduct and results
of their public actions.

Other external agents, most notably a watchful press and
media, monitor legislative performance.

This scrutiny can

help maintain legislator accountability by revealing its
absence.

However,

these findings are after-the-fact.

The

ability to correct or redress any perceived short-coming is
limited by the nature of these participants as external to
the formal legislative process.

The ultimate guarantors of legislator accountability must be
legislators themselves and the legislative processes they
devise.

Over the years,

legislators have reformed and

refined the legislative process both at the urging of others
and on their own initiative.

Each of these internal reforms

7

have sought to support, explicitly or implicitly,

the main­

tenance and reinforcement of legislator accountability.

Longer and more frequent sessions as well as lengthened
terms of office were two reforms designed to give legis­
lators more time to carry out their duties.

Adequate time

for the performance of their functions is a key aspect of
legislator accountability.

The evolution of partisan

leadership and the institutionalization of committee
structures are two other administrative reforms legislators
adopted to better structure their tasks through the
delegation of duties and the division of labor.

Here,

changes in the legislative process developed to help improve
efficiency and better define legislator accountability.

The utility of these measures are still debated in terms of
legislator accountability.

Have longer and more frequent

sessions given legislators needed time or opened the door to
extraneous activity?

Have more defined institutional

structures and legislator relations helped or hindered
legislative performance by differentiating certain legis­
lators from others?

Various points of view on these

questions define the debate over legislator accountability.

This research examines a more recent legislative reform,
legislative staff,

to assess its effect on the maintenance

of legislator accountability.
be stated quite simply.

The essence of my inquiry can

Legislator accountability to the

8

public requires staff accountability to legislators.

If

staff are accountable to legislators, then legislator
accountability to the public is maintained even after the
introduction of legislative staff reforms.
accountable to legislators,

If staff are not

then legislator accountability

itself is weakened by staff development.

To begin the research,

this chapter offers a definition of

legislator accountability.

Staff relevance to the

maintenance of legislator accountability is examined by
reviewing prior assessments of staff impact on legislative
functioning.

Most analyses of staff have stressed growth in

staff numbers.
important,

I maintain that this emphasis, while

is incomplete.

A more direct indicator of staff

impact upon legislator functioning and accountability is
staff access to legislators.

Access delineates which staff

interact with which legislators.

By ascertaining who

controls these interactions and the extent of staff
discretion these interactions reflect, direct observations
as to whether or not staff are accountable to legislators
can be made.

The chapter concludes with a discussion of

access, control,

and discretion that will guide the

development of the formal research argument.

DEFINING LEGISLATOR ACCOUNTABILITY

Legislators are accountable to numerous individuals and
interests.

Particular stands on issues, political support,

9

access to influence,

assistance when needed, or patronage

are among the tangible expressions of accountability.
Accountability,

however,

is no one of these things.

Rather,

they can all be subsumed under a more general definition.
Legislator accountability consists of expectations others
place on legislator decision-making.

Such expectations take three general forms.

Expecting

legislators to be responsible for the decisions and actions
attributed to them is one type of accountability.

Predict­

ability [1] in legislative performance is a further
expectation that can be placed upon legislators to gauge
their accountability.

Finally, various types of rationality

in legislator behavior may be looked for as a part of an
accountable legislator [2],
mutually exclusive.

These expectations are not

However,

assessments of accountability

can, and commonly do, emphasize one over another.

I maintain that, at a minimum,
sponsibility.

accountability requires re­

Without some assignment of responsibility,

predictability and rationality lack substance.

It is this

form of accountability that my research explores.

For both legislator and staff,

responsibility,

and the

accountability it supports, has two features which can be
ascertained by answering the following questions:
interests exert influence on legislative decisions;

whose
and how

do these interests combine to produce legislative actions.

10

These facets of responsibility are returned to later in this
chapter as, respectively, control and discretion.

While accountability requires the assignment of responsi­
bility, the two concepts are not synonymous.

This can be

seen by considering how the other forms of accountability
differ from responsibility.

Responsibility for public

policy decisions is often not predictable.

The number of

interests and the variety of their interactions,

especially

in the legislative setting, accounts for the elusiveness of
predictability.

Since policy analysis (and, therefore, legislative
decision-making) is incremental, exploratory, serial,
and marked by adjustments of ends to means, it is to be
expected that stable long-term aspirations (i.e., pre­
dictability) will not appear as dominant critical values
in the eyes of the analyst (or legislator).
[3]
Similarly, responsiblity may exist without obtaining
rationality in decision-making even though rationality may
be a preferred choice.

If he (the legislator) himself actually understands
that, in so far as his opinion has any effect on the
choice, he is a participant in partisan mutual adjust­
ment, at the very least in partisan discussion as one
form of the process (of decision-making) he may wish to
tailor his role to increase his effectiveness, if that
is what he desires.
And, incidentally, he will come to
accept certain divergence between his preferred choice
of method and actual choices made as appropriate, not
necessarily evidence of the foolishness (i.e., lack of
rationality) of political choice.
[4]

11

PRIOR VIEWS ON STAFF AND LEGISLATORS
The relevance of legislative staff to legislator account­
ability permeates the literature analyzing legislative
staff.

Observers of staff agree on two basic propositions.

First of all, some subset of legislator functioning involves
interaction between legislators and legislative staff.
Secondly, as legislative staff numbers grow, this heightens
the potential for legislative staff to influence legislator
performance.

Disagreement sets in as to whether these

statements are to be viewed positively or negatively.

To

begin an assessment of legislative staff and its impact on
legislator accountability requires an appreciation for these
differing points of view.

Staff proponents argue that staff have been an aid to legis­
lators providing them with the means to maintain legislator
accountability.

Critics state that staff have been delete­

rious to that accountability complicating the already
complex tasks legislators face and further removing those
tasks from legislators hands.

These views of staff

proponents and critics are reviewed here to gain an
appreciation for the manner in which staff influence on
legislator accountability has been discussed.

Proponents.

Staff proponents have consistently argued

that staff are a more efficient response to increasing
legislative complexity;

that the technical assistance they

12

provide is needed to exercise legislative supervision of the
executive branch;

and that staff help inform legislators and

reflect the public concensus.

Soon after World War II calls

for increased legislative staffing using such arguments
began to be heard.
fashion,

While not coordinated in any formal

support for staff grew until by 1974 the following

statement could be made without fear of contradiction.

Surveys reveal that the single improvement most
frequently mentioned as necessary by members of state
legislatures is more professional staff.
[5]

Support for legislative staff development did not follow a
set pattern or timetable in the states,
after federal developments.

nor was it modelled

However, similar arguments for

legislative staff based on efficiency of administration,
maintenance of legislative prerogatives, and response to
growing societal demands are found throughout the 1950’s and
1960’s.

Since the focus of this research will be the Michigan legis­
lature, two studies presenting reasons for staff development
in Michigan will be considered.

While not necessarily

representative of what other states did, the arguments
presented are typical of those found in other reports.

Report #11 of the Michigan Joint Legislative Committee on
Reorganization of State Government dated March 1951 covers
the topic of "The Legislature."

This Joint Committee

13

(referred to as the "tittle Hoover Commission" after the
federal Hoover Commission) reviewed the broad array of
Michigan governmental functions recommending changes felt
warranted or desireable.

It was strongly supportive of

increased legislative staff.

At the time of the Joint Committee’s Report #11 Michigan had
a part-time legislature with only ten full-time staff.
Another 96 staff were retained on a monthly or per diem
basis during periods the legislature was in session.

Of the

twelve recommendations in Report #11, five related to staff
or staff agencies.

The Joint Committee clearly felt that to meet growing
demands on the legislature’s committees, more staff were
needed.

With the insistent demand not only for more services but
for more kinds of services, the need for equipping the
state’s important legislative committees with competent
staff personnel becomes increasingly urgent.
By
adequate staffing, committee operations could be raised
to a new level of performance.
[6]

To better monitor expenditures by the executive branch, the
Joint Committee also called for replacing the then elective
post of Auditor General with a legislative appointed and
accountable State Auditor.

14

The determination of whether the money which it (the
legislature) appropriates is being spent for specified
purposes and whether it is being wisely expended
requires competent and adequate staff facilities under
the legislatures own jurisdiction.
[7]

Ten years later in September of 1961, Herbert Garfinkel,
then Associate Professor of Political Science at Michigan
State University, prepared a paper for the State’s Constitu­
tional Convention Preparatory Commission on the subject "The
Constitution and The Legislature".

While not an overt advocate of legislative staffing,
Garfinkel does note two advantages of legislative admini­
stration which implies, at least in part, staff involvement.
The first advantage relates to the legislative process
itself.

Legislatures are perhaps less dependent than other
branches of government on good administration for a
satisfactory product; the substantive quality (rather
than the execution) of statutes is mainly what the
legislature can control.
However, the efficient organi­
zation of the legislature, such that large numbers of
Representatives and Senators can cope with numerous
proposals of considerable complexity, entails many
problems of good administration.
[8]

The second advantage Garfinkel attributes to legislative
administration is communication between the public and
elective officials.

15

Good administration of the legislative operation is more
than a matter of time saving or cost reduction; it is
equally a problem of enabling the legislators to help
inform the public as well as reflect the popular
concensus.
[9]
Throughout his paper,

Garfinkel reviews the pros and cons of

the various issues he reviews.
specifically,

When he does discuss staff

he notes three arguments used to oppose staff:

that legislators would become rubber stamps for a new
bureaucracy;

that merit system employees would not fit in a

political arena;

and that staff costs money.

In balancing these objections,

Garfinkel summarizes many of

the advantages noted by others.

In answer to (the opposition), the proponents maintain
that professional staff aids are necessary precisely
because the legislators cannot hope to personally
research all the numerous proposals they must consider.
This places them at the mercy of only too-eager-toassist private groups, whose staff are certainly not at
the direction of the legislators.
Whatever difficulties
may exist in maintaining control over the legislatures
research aids, that is far easier than attempting, each
legislator for himself, to match the expertise available
to the executive branch of government and the organized
groups in the community at large.
[10]

Arguments such as these tended to support the establishment
and expansion of legislative staff that occurred in the
1960’s and 1970’s.

Much of the early literature on staff

takes the positive attributes of staff as given and focuses
instead on how staff functions are performed or might be
performed better.

16

Such views were not universally held, however.
critics are hard to find but did exist.

Early

Today, the staff

literature reflects a much more critical perspective on
staff and its influence on legislator accountability.

It is

to this side of the debate that we now turn.

Critics.

Collectively,

critics of staff point to the

following issues associated with increased staff
development.

Staff adds to and does not reduce legislative

work and complexity;

legislators become dependent on staff

who influence conclusions by structuring options to be
considered;

and, staff represent a barrier blocking direct

legislator-constituent interaction.

In 1967, Norman Meller published the first systematic
critique of legislative staff.

His article in the Western

Political Quarterly "Legislative Staff Services:
Specific,

Toxin,

or Placebo for the Legislature's Ills" remains one

of the better attempts to assess the implications of legis­
lative staff growth.

Meller assesses three assumptions made

by "advocates of staff augmentation" which he finds
questionable and "goes far to undermine the advisability of
staff expansion".

These assumptions are:

1.

Augmented staff lightens the legislative burden;

2.

Expanding objective expert staff of committees
increases legislative efficiency; and,

3.

A large personal staff for the legislature is a
cure-al1.

17

Regarding the first assumption, Meller argues that per­
ceiving staff as a means to lighten the legislative burden
is an error.

In fact,

the reverse is likely to occur.

To the extent that reference bureaus, bill drafting
agencies, and other comparable agencies are enabled to
meet increased legislative demands by augmenting their
staffs, their product adds to rather than reduces the
volume of the legislators work.
Indeed, there is even
the suspicion that Parkinson’s Law governs and work
requested of legislative service agencies expands to
meet growth of staff, rather than the reverse.
[11]

Rather than viewing the expert committee staff as improving
legislative efficiency, Meller sees them as a danger.
Legislative staff cannot hope to duplicate anew executive
branch budget formulations and policy initiatives;

but,

they

can provide a critical review of executive branch actions
that lead to the formulation of the legislative position.
This could place staff in positions of influence.
controlled,

If not

this influence could be used in ways not derived

from legislator demands.

This is a danger to be reckoned with, and to be guarded
against by adding committee experts only as they can be
kept accountable.
[12]

As to the assumption that larger personal staffs will serve
to improve legislative performance, Meller retorts as
follows.

13

The indiscriminate augmentation of personal staff could
lead to the institutionalizing of the legislator and
eventually to each legislator becoming the captive of
his own staff.
[13]

A final assessment of staff points to the risks of: inter­
jecting between constituents and legislators a semiautonomous bureaucracy;

raising debate above the common

understanding of citizens;
representative contact.

and blocking constituent-

These could " . . .

eventually

subvert the fundamental processes of the legislature it
(staff) serves"

[14].

Other authors have elaborated one or more of Meller’s
concerns.

We will consider two:

Nachmias and Rosenbloom’s

Bureaucratic Government USA as well as Michael Malbin’s
Unelected Representatives.

Nachmias and Rosenbloom view staff development in the
context of a general bureaucratization of governmental pro­
cesses.

Staff, for them, are a bureaucratic response to a

complex legislative system with many veto points.

The

influence of staff is conditioned by several factors such as
circumstances and staff attributes (e.g., tenure, speciali­
zation, expertise).

Nevertheless,

staff exercise great power which, these

authors feel, threatens to bury the legislature in paper.
At a federal level, they see staff as part of a circular

19

phenomenon:
more staff,

more bills,

leads to more committees,

leads to more bills, etc.

leads to

This link between

staff and committees and the basic legislative bureaucratic
structure carries over into executive-legislative
relations.

For Nachmias and Rosenbloom,

committee

specialization leads to a coziness and accomodation between
the executive and legislative branches.

They find no

evidence that staff has made legislators (in Congress, at
least) less dependent on the executive branch or interest
groups.

Staff simply compile existing information although

this does afford them some influence.

By structuring the

choices legislators consider via advice, alternatives, and
information, staff can influence legislative conclusions.

The final author we will consider here is Michael Malbin.
As the title of his book, Unelected Representatives,
implies, Malbin takes the role of staff seriously.

In a

series of richly detailed case studies, Malbin examines the
impact staff have had not only on individual members of
Congress but the legislative branch as a whole.
positive results are identified.

Certain

More, and more active,

staff have enabled the legislator to maintain the role of
initiating policy; much executive oversight is fostered and
conducted by staff; and, staff afford an openness to nontraditional groups and ideas.
These benefits have not been without cost.

For Malbin,

essence of the legislative branch is deliberation.

the

20

For a process of legislative deliberation to function
reasonably well, at least three distinct requirments
must be satisfied.
The members need accurate informa­
tion; they need time to think about that information;
and they need to talk to each other about the factual,
political, and moral implications of the policies they
are considering.
The new use of staff undercuts each of
these. [15]

Among the new uses of staff Malbin identifies are staff
members persuading bosses to adopt their ideas; staff nego­
tiating for members in various stages of the legislative
process;

staff to committee chairs acting on behalf of the

chair to control the flow of information to committee
members;

staff judgments effecting what information commit­

tee chairs receive; and staff acting independently to get
information to the press [16].

In short, staff acting in an

independent fashion has complicated the legislative process
and weakened its deliberative nature.
The member (of Congress) . . . is becoming more of a
Chief Executive Officer in charge of a medium-sized
business than a person who personally deliberates with
his colleagues about policy.
[17]
No staff critic has offered a satisfactory alternative to
further staff development.

Nachmias and Rosenbloom see

staff as one expression of a pervasive modern dilemma;
takes bureaucracy to control bureaucracy.

it

Meller offers as

his response a reduction in the legislative workload by
delegating some legislative demands to other governmental
entities

[18].

Over a decade after Meller, Malbin is still

calling for ways to limit the legislative agenda [19].

21

The prospect of legislators surrendering their current res­
ponsibilities and resources appears remote.

Legislative

staff are effectively a permanent part of legislative
functioning in America.

They need to be understood and

integrated into our thinking on legislative processes.
begin this integration,

To

the claims of staff proponents and

the charges of their critics need to be carefully
scrut inized.

WHY STAFF NUMBERS HAVE GROWN
Prior attempts to assess the implications of staff have
centered on analysis of staff growth.

Staff proponents saw

more staff as the resources required to deal with modern
governmental complexity,

assure legislator responsibility

for legislative acts and thereby maintain legislator
accountability.

Critics viewed the rise of numerous, new,

potentially independent participants in the legislative
process as actually making performance of legislative duties
more difficult,

further reii >ving legislative activity from

responsible legislators and weakening their accountability.

An appreciation for staff growth begins with the recognition
of historical parallels.
occurrence,

Staff are not a peculiarly modern

they are simply a contemporary manifestation of

the political advisory function found throughout history.
This insight helps place staff in perspective and makes
many of the features associated with staff familiar.

We

22

will be examining contemporary advisors (legislative staff)
who advise identifiable decision makers (legislators) in a
particular context (the Michigan State Legislature).
However,

the specificity of our subject should not obscure

the broader context of our inquiry.
across civilizations,

advisors have been an integral part of

political decision-making.

Accordingly,

Throughout history and

[20]

consideration of staff growth can be best

understood by proceeding from the general to the specific,
reviewing first observations about advisors overall, and
then legislative staff specifically.

Growth of Advisors Generally. There are certain
preconditions that affect the development,
of advisors.

[21].

growth and role

Changes in these forces have contributed

to the increasing numbers of staff.
forces is the size of the state.

The first of these

Small, homogeneous states

may be governed relatively easily; but, as the state grows
in size and diversity additional forms of advice and
advisors are necessary to manage its governance,
communications, and administration.

The number of polities involved in political discussions
also influences the need for advisors.

A state existing

apart from others does not require the same type of advisors
as one that must consider relations among states.

As the

23

number and frequencies of these interactions increases, so
too do the demands for advisors.
Finally,

the type of political structure has a direct

bearing on the number of advisors.

Monolithic polities

develop advisory functions differently than states with
constitutional provisions for separating governmental
functions and authorities.

In the latter instance, each

constitutional entity is an arena within which advisors act.
Each of these forces can be seen at work encouraging the
growth of the political advisors being examined here,
legislative staff.
Growth of Legislative Staff Specifically. Regarding
growth in the size of the state, historic increases in
governmental budgets as well as greatly expanded arenas of
legislative activity and interest are all indicative of the
growth conducive to the increase in staff advisors.

These

features are hallmarks of American government at both the
federal and state levels during the last generation.

Not

only have governmental budgets grown but also the purposes
for which governmental funds are allocated have expanded.
Direct payments to individual citizens on a contractual,
merit, or need basis greatly increased governmental outlays;
public subsidies for a host of economic and social purposes
are now common; and the fostering or protection of rights
for various individuals and interests has become an expected
role of government.

24

Justifying these various undertakings;

garnering support for

their enactment; monitoring their performance;

and amending

their operation were among the tasks required of America’s
legislatures.

Each of these tasks led to greater

requirements for legislative assistance in the form of more
staff.

Turning to the level of interactions among governments,

the

American federal system fosters intergovernmental contacts
among states as well as between states and the federal
government.

This interaction has been furthered by federal

government initiatives cutting across state lines as well as
the demands by municipalities and substate entities for
legislation or assistance from state legislatures.

This web

of governmental interactions requires the attention of
increasing numbers of staff advisors to monitor events and
help formulate actions.

The rise of single-issue interest

groups and their legislative demands lead to more points of
interaction that also require staff attention.

Finally, America’s political structure with its separation
of powers encourages legislative independence and the
inclination to balance bureaucratic agency personnel with
legislative staff.
partisan divisions,

[22]

American legislatures with their

committee structures,

leadership roles,

and direct legislator-constituent relations further
fragments the governmental structure and establishes
numerous niches for staff advisors.

The fact that the

25

American experience with the growth of legislative staffs is
not found in European parliraentary systems is the most
direct evidence of this phenomenon.

[23]

The nature of partisan competition in the American political
system has been found to contribute to staff growth.
Greater levels of partisan competition give rise to caucus
staff and may be related to greater decentralization of
staff.

[24]

Relatively weak party organization in single­

member districts may cause staff to be used as substitutes
for party agents.

[25]

The importance of committees and subcommittees to legis­
lative functioning encourages the growth of committee staff.
It is here that the policy-relevant expertise that staff can
provide is most needed.

Such expertise has been regarded as

indispensable to American legislators aspiring to attain a
more professional status.

[26]

The power of legislative

leaders has also shaped staff development.

[27]

Taken together, America’s growth of government generally,
the myriad of interactions between its governments,

and the

structure of the governmental apparatus have all encouraged,
if not required,

the recruitment of increasing numbers of

legislative staff.

26

ACCESS
Earlier assessments of staff stressing growth in staff
numbers are incomplete.
expected;

Not only is staff growth to be

growth only measures the magnitude of the staff

phenomenon, not what staff do or how they influence
legislator accountability.

In this research,

I pursue a

different tact by arguing that the impact of staff on
legislator accountability is not reflected solely, or even
principally , in rising staff numbers.

It is how staff are

used that determines their impact on accountability.

The use of staff influences and is influenced by the
legislative mileiu in which it occurs.

Staff cannot be

viewed alone but must be integrated into an appreciation for
the priorities and processes relevant to other legislative
actors.

The chief legislative actor is the legislator.

Staff access to legislators is the most important attribute
of staff use and is central to understanding the
significance of staff.

An exploration of staff access will

permit an assessment of issues raised regarding the
maintenance of legislator accountability.

Access is the key to appreciating staff’s impact on
legislator accountability because it affords the most
relevant indicator of staff activity and influence within the
legislative process.

By documenting access, networks of

27

legislators relating to staff and staff relating to
legislators can be observed and analyzed.
What remains is to approach the discussion of staff access in
a manner that can be related to legislator accountability.
For this, we will draw upon works from three areas of
political science.

Robert Salisbury and Kenneth Shepsle’s

work on what they term the Legislative Enterprise provides
the basis for understanding access.

Barry Weingast,

in his

examinations of legislators relations with executive agency
bureaucrats identifies control and discretion as the
principle attributes for appreciating how advisory access
impacts accountability.

Futher, he stipulates criteria for

assessing how control is exercised.

Charles Lindblom in his

works examining individual responsibility in political
decision-making defines the parameters of discretion
associated with advisory access.

Describing Staff Access. To fully appreciate staff
access I will argue, with Salisbury and Shepsle,

that

legislators can no longer be viewed simply as individuals
elected to an office [28].

Rather,

they head "Legislative

Enterprises" composed of ". . . personnel differentiated by
political function and legal status, but nevertheless
subordinate to the specific member."

[29]

While others have

recognized this broader legislative role,

[30] Salisbury and

Shepsle give it the most in-depth elaboration.

28

Their article on staff turnover in the American Political
Science Review states their point of view well.
Bach member (of Congress) has come to preside over a
system of greater or lesser complexity in scope.
The
member of Congress may be best understood, therefore,
as an enterprise and the analysis of phenomena . . .
must be analyzed in terms of the dynamics within those
member enterprises. . . our basic premise is that the
House (Senate) is no longer best conceived of as a body
of 435 (100) individuals or even as a structurally
differentiated (committees, subcommittees) collection
of legislators, but rather as a collection of
organizations or member-centered enterprises [31]
(emphasis in original).

In a subsequent publication,

they define their frame of

reference as "analytically superior to extant alternatives"
[32] a contention they support as follows:
The principal advantage of this perspective is that it
allows us to incorporate the phenomena of congressional
staff systematically with the analysis of Congress
rather than awkwardly appending it to a discussion of
congressmen as discrete individuals. [33]
Our analysis accepts that contention as applicable to state
legislators as well.

To it, we add the observation that

access defines the composition of each legislator’s
enterprise.

Those staff having access to a legislator are

part of that Legislative Enterprise regardless of their
formal ties to a legislator.
relations,

Employment, partisan

committee emphases, even House or Senate

affiliation offer less reliable indicators of legislatorstaff relations when compared to access.

With access as the

criteria for deliniating each Legislative Enterprise the
network of staff contacts a legislator utilizes can be

combined;

legislators use of staff need not be limited to

those they hire or have assigned to them.

Similiarly, staff

can be viewed as simultaneously participants in multiple
enterprises offering their time, services, and expertise to
all legislators with whom they have access.

Using access to define the Legislative Enterprise also
enriches our understanding and appreciation of access.

The

Legislative Enterprise provides the appropriate focal point
for assessing dispersed instances of staff access to
legislators.
duration,

The components of access such as frequency,

initiation and purpose can be meaningfully grouped

and analyzed within the context of the Legislative
Enterprise.
Relating Access to Accountability. To be used in the
assessment of legislator accountability,

the ability to

analyze access and the Legislative Enterprises it defines
requires refinement to identify those aspects of access
relavant to accountability.

Our earlier discussion of

accountability centered on the presumption of responsibility
upon which all forms of accountability rest.

The assignment

of responsibility was said to be a function of two factors:
whose interests exert influence;

and, how those interests

combine to produce action.

When discussed in terms of access,
responsibility become:

these aspects of

1) control (the party in control

30

will be the party whose interests exert influence);

and, 2)

discretion (how interests combine in the interactions
resulting from access will be determined largely by the
amount of discretion allowed the party lacking control).
An examination of control and discretion in legislator-staff
access provides the basis for assessing the presumption of
responsibility upon which legislator accountability rests.
To begin the examination of these features of responsibility
we will draw upon work done by Barry Weingast regarding the
relations between legislators and bureaucrats in the
regulatory policy arena.

[31]

Weingast’s research focuses on

understanding the relations between federal legislators and
bureaucrats.

Despite the specificity of his subject,

the

observations drawn from his work are relevant to the
analysis of the Legislative Enterprise just discussed.
noted earlier in discussing staff growth,
advisory function takes many forms.

As

the

Viewed generally,

executive agency bureaucrats in Weingast’s regulatory
context are a form of legislative advisor.

Weingast

stresses the importance of committee functions.

This

emphasis is premised on the recognition of links between
legislator activities and larger legislative processes, a
recognition central to the Legislative Enterprise.

Further,

his concern with appreciating how legislators and
bureaucrats relate is consistent with our attempts to
understand legislator-staff interactions.

In short,

Weingast’s analytic framework is felt to be generalizable to

31

other aspects of legislative functioning,

specifically the

Legislative Enterprise.

In his paper prepared for the Fifth Carnegie Conference on
Political Economy held in Pittsburg during 1983 [35]
Weingast,

drawing on economic principal-agent theory, makes

several observations of direct relevance to our exploration
of responsibility and the accountability it supports.

Control.

Regarding the first facet of responsibility,

whose interests exert influence, Weingast notes that direct
legislative oversight is not always possible or desirable.
But, controls do exist to assure agencies are responsive to
legislators just as successful agents are responsible to
their principals.

Similarly, staff may frequently function

outside direct legislative purview without negating the
exercise of legislator control over staff.

Among the devices available to ensure legislator control,
Weingast cites four:
1) legislators having influence over placement and
promotion which can afford ex post sanctions (e.g.,
firing) as a real threat that leads to ex ante
incentives;

32

2) advisors being dependent on the acceptance of
legislators for their influence conditions advisory
behavior and constrains options;
3) legislators being in contact with other parties
whose input provides automatic checkpoints on the work
of advisors;
4) legislators favoring advisors with whom they share
electoral or other legislative

priorities.

Bach of these mechanisms helps assure that access to
legislators by their advisors generally and staff
specifically are controlled by legislators.

With these

incentives and disincentives controlling access,

legislators

can structure their enterprises with responsive
advisors/staff.

There is a clear superior/subordinate quality to the
relations between legislators and staff.

The staff

relationship to legislators is ultimately one of dependence
[36] reinforced by unwritten norms of staff deferrence and
loyalty to legislators.

[37]

Many of the staff are hired and

fired by individual legislators.
stand for election,
peers;

Legislators cast votes,

and relate to fellow legislators as

staff do not.

experience, knowledge,

Regardless of individual skills,
or salary,

legislators are in a much

stronger position to control staff and their access than
vice-versa.

33

Discretion.

The second feature determining

responsiblity entails understanding how interests combine to
produce action.

While legislators may exert influence and

control over their interactions with staff, this does not
negate the presence of staff interest nor opportunities for
independent staff actions that might impact legislator
responsibility and the accountability it supports.

As

Weingast notes in his paper on congressional-bureaucratic
relations,

the likelihood that some discretion (or

"shirking" in economic terms) follows from the delegation of
tasks to bureaucratic agents (or,
here advisors and staff).

in the terminology used

This is a common phenomenon, but

not necessarily intolerable or delegation would not be as
readily accepted in various economic and political
transactions.

Weingast notes that legislators delegate when

the benefits derived (e.g., completing the task, acquiring
information,

or extending influence via a surrogate) exceed

other alternatives (e.g., performing the task, or exercising
continuous oversight) even though delegation will entail
discretion or shirking.

We will look for a similar

benefit-versus-cost logic to govern legislator delegation to
staff.

To ascertain where opportunities for staff discretion exist,
as well as boundaries delineating the limits to this
discretion, we refer to a series of works by Charles
Lindblom.

34

While using different terms, three works by Lindblom present
the most in-depth discussion of individual discretion in
political decision-making.

The beginnings of his thought

are found in a book co-authored with David Braybrooke, A
Strategy of Decision

[38].

This work is largely concerned

with refuting a centralized decision-making model as un­
realistic.

His alternative,

disjointed incremental ism,

emphasizes the role of individuals and their interactions as
more appropriate for understanding public policy
development.

This emphasis on the individual is extensively pursued in The
Intelligence of Democracy [39].

Here Lindblom elaborates his

thoughts on partisan mutual adjustment. While his intent is
to contrast this perspective to calls for centralized
governmental control,

his arguments explore the various

nuances of individual discretion in political interaction.
His final work in this area The Policy-Making Process

[40],

synthesizes his views of decision-making into a succinct
examination of what he terms governmental politics.

Lindblom divides political interactions between individuals
(X and Y) into five general categories:
1.

Partisan discussion:

[41]

X changes Y ’s perception of

given advantages and costs of a course of action,

35

2.

Bargaining:

X alters advantages and costs to Y

contingent on Y ’s response,
3.

Reciprocity:

4.

Coercion:

X obligates Y to respond,

X unconditionally alters advantages and

costs to Y of a course of action of
5.

Command:

Y,

X authoritatively prescribes to Y.

These situations delineate control and discretion.
instance,

In each

X is the controlling party because it is X ’s

interests that exert influence on decisions.

The five

relationships between X and Y express how X uses this control
and the type of discretion available to Y.

For legislators to be responsible,

they must

be party X or

minimize the instances in which they are party Y.

How

responsib1ity is exercised will depend on the manner in
which legislators combine their interests with others using
one or more of the five scenarios outlined above.

Under the first scenario, partisan discussion,

individual X

changes Y ’s perception of the advantages and costs of a
course of action. Here, either the legislator or staff may
be party X.

While legislators can control staff access,

once access is gained,

staff may attempt to change a

legislator’s perception of a given situation.

Legislator

receptivity to such persuasion may vary, but assuming some

36

openness to discussion, staff may exercise a fair degree of
discretion in the conduct of these discussions.

The second form of political interaction between individuals
is bargaining.

Lindblom defines bargaining as X altering

the advantages and costs to Y contingent on Y ’s response.
Staff acting alone have limited resources that might alter
legislator response.

However, staff acting on behalf of one

legislator may be in a position to bargain with another
legislator.
discretion,

While this may constitute a degree of staff
it is discretion with definite limitations.

First of all, the staff person is bargaining as a surrogate
for a legislator, not as an independent political actor.
The legislator on whose behalf the staff first bargains is
the true party X.

Should the staff person exceed his

bargaining authority this legislator may renege or change the
conditions of the bargain.

A further limit on staff

discretion in a bargaining situation is that legislator Y
may supercede any bargain with a staff person by going to
the legislator whose interests are involved.

At this

legislator-to-legislator level, staff discretion in
bargaining can easily become peripheral to the bargaining
between legislator peers.

Reciprocity, the third form of political interaction,
when X obligates Y to respond.

occurs

There are few situations

where staff action will place a legislator in a position

37

where a response is required. More often than not,
legislators may ignore staff actions with few,
negative consequences.

if any,

With the exception of the bargaining

situations discussed above (where staff acting on behalf of
a legislator obligates that legislator to respond by either
supporting, renegging,

or modifying the bargain staff struck

with another legislator) staff discretion in the area of
reciprocity is limited.

While staff may not be able to obligate legislator response,
it may be possible for staff to unconditionally alter the
advantages and costs to a legislator of a particular course
of action.

This is the form of political interaction that

Lindblom terms coercion.

For example,

information provided

to the media or another third party may enable staff to
force a legislator to delay, avoid,
contemplated action.

This would constitute an extreme

instance of staff discretion.
probably is, exercised;
use.

or reformulate a

Such discretion can be, and

but, staff must be judicious in its

Coercive acts may prompt legislators to exercise their

control over staff by limiting further staff contacts or
seeking the termination of staff employment.

Unless the use

of coercion on a legislator is sponsored and supported by
another legislator (in which case that legislator and not
the staff person is the coercing party) staff may find
repercussions and reprimands result from their coercive
acts.

This undoubtedly tempers such staff discretion.

38

The final form of political interaction that Lindblom
discusses is command where X authoritatively prescribes to
Y.

Staff ability to command legislators goes well beyond

discretion.

It negates the exercise of legislator control

and renders nonexistent the maintenance of legislator
responsibility and the accountability it supports.

Not even

the severest critic places staff in a position of commanding
legislators.

Staff have no basis of authority which enables

them to supercede and prescribe to legislators.

Therefore,

the ability to command legislators must be placed outside
the purview not only of staff discretion, but staff
capacity.

To recapitulate,

this review of the various forms of

political interactions have shown that opportunities for
staff discretion do exist; but, this discretion is exercised
within definite limits and under specific conditions.

Staff

acting on behalf of a legislator have more discretion than
staff relating directly to a legislator. When acting on
behalf of a legislator,

staff discretion must be kept to

limits tolerable to the sponsoring legislator.
support,

the staff person may discuss, bargain,

on that legislator’s behalf.
legislator,
legislator.

Assuming such
and coerce

When dealing one-on-one with a

limits to staff discretion are set by the
In either instance,

legislators have the

ability to curtail staff discretion.

Whether they choose to

use this ability will depend upon whether they view staff
discretion as tolerable and subject to their influence.

39

SUMMARY
This chapter presents the principal components of the
analysis to follow.

The purpose of the analysis is to

ascertain whether staff are accountable to legislators
thereby supporting legislator accountability to the public.
Accountability is defined as expectations placed on
legislative decision-making.
accountability,

The most basic form of

responsibility,

is said to be a function of

control (whose interests exert influence) and discretion (how
interests combine).

Prior attempts to assess the implications of staff have
stressed staff growth.

While growth in staff numbers is

important for understanding the staff phenomenon,

it is a

predictable response to historical circumstances and reveals
little about how staff use impacts legislator accountability.

Staff access to legislators is a more direct indicator of
staff use.

Occurrences of access can be grouped to reveal

Legislative Enterprises defined by staff and legislator
interactions.

Of even more importance,

access can be related

to control and discretion which are central to an
appreciation of responsibility and the legislator
accountability it supports.

Barry Weingast’s analysis of

legislative-bureaucratic relations identifies features of
control that can be applied to occurrences of access defining

40

Legislative Enterprises.

Charles Lindblom in his discussion

of individual decision-making delineates the parameters of
discretion that determine whether staff discretion is
tolerable for legislators as Weingast*s work would suggest.

Each of these considerations are elaborated in the chapters
that follow.

Chapter 2 details the growth of staff in

Michigan as a function of historic forces identified here.
In addition,

it describes the process used for documenting

legislator-staff interactions and the Legislative Enterprises
they define.

Chapter 3 structures the concepts of control

and discretion around considerations of access to develop a
formal research argument that is empirically tested in
Chapter 4.

Chapter 5 reinforces the empirical findings with

comments drawn from personal interviews with five Michigan
legislators and written comments from that state’s staff
members.

CHAPTER 1:

References

1
Expectations of predictability are used as a basis for
determining accountability.
Individual predictability is
sought from legislators in the form of consistent loyalties
to a party or interest.
Predictability at this level is
usually gauged by voting records and underlies much of the
literature on roll-call analysis.
Organizational
accountability relies heavily on predictability as the basis
for assessing decision-making at this level of analysis.
An
in-depth treatment of how this model relates to Congressional
budgetary decision-making is Aaron Wildavsky’s The Politics
of the Budgetary Process. Little, Brown & Co., Boston, 1974.
2

Attempts to "rationalize" the budgetary process by
explicitly structuring and scheduling the various decision
points involved is one example of how rationality is sought
in the legislative processes.
Formal legislative oversight
of executive actions based upon agreed-to roles and
objectives is an effort in a similar vein.
The obstacles to
the implementation and on-going operation of such reforms
speak to the difficulties associated with instilling
rationality as the basis for legislator accountability.
3
Braybrooke, David and Lindblom, Charles, A Strategy of
Decision:
Policy Evaluation as a Social Process. The Free
Press of Glencoe, London, 1963, p. 102.
4
Lindblom, Charles, The Intelligence of Democracy:
Decision Making through Mutual Adjustment, the Free Press,
New York, 1965, pp. 299-300.
5
Rosenthal, Alan, Legislative Performance in the States:
Explorations of Committee Behavior. The Free Press, New
York, 1974.

6
Staff Report to the Michigan Joint Legislative
Committee on Reorganization of State Government, The
Legislature. Report # 1 1 . March 1951, p. 11-54.

7
Ibid., p. 11-58.

41

42

8
Garfinkel, Herbert, Michigan Constitutional Convention
Studies, The Constitution and The Legislature.
Constitutional Convention Preparatory Commission, State of
Michigan, September 1961.
9
Ibid., p . 42.
10
Ibid.. p. 51.
11
Meller, Norman, "Legislative Staff Services:
Toxin,
Specific, or Placebo for Legislature’s Ills", The Western
Political Quarterly. Volume XX, No. 2, Part 1, June 1967,
p. 385.

12
Ibid.. p. 387.
13
Ibid.. p. 388.
14
Ibid., p . 389.
15
Malbin, Michael J., Unelected Representatives:
Congressional Staff and the Future of Representative
Government. Basic Books, New York, 1980, p. 242.
16
Ibid., p. 8.
17
Ibid., p . 6.
18
Meller, Op. Cit.. p. 388.
19
Malbin, Op. Cit.. p. 251.
20
Goldhamer, Herbert,
1978.
21
Ibid.

The Adviser. Elsevier, New York,

43

22
Patterson, S.G., Lowenberg, G., Jewell, M.E., "Editor’s
Introduction", Legislative Studies Quarterly. Volume VI, No.
4, November 1981, p. 491.

23
Ibid.
24
Simon, Lucinda S., Understanding Legislative Staff
Development;
A Legislator’s Guide to Staffing Patterns.
National Conference of State Legislatures, Denver, August
1979, p. 65.
25
Patterson,

et al., Op. Cit.

26
Ibid.
27
Simon, Op. Cit.
28
Salisbury, Robert H. and Shepsle, Kenneth A.
"Congressional Staff Turnover and the Ties-that-Bind"
American Political Science Review, vol. 75, no. 2, June
1981.
29
Salisbury, Robert H. and Shepsle, Kenneth A.
"U.S.
Congressman and Enterprise" Legislative Studies Quarterly,
vol. VI, no. 4, November 1981, p. 561.
30
See, for example, Malbin, Michael
Unelected
Representatives. Basic Books, New York, 1980; or, Loomis,
Burdett A. "The Congressional Office as a Small (?)
Business” Publius 9, 1979.
31
Salisbury and Shepsle,
Op. Cit.. p. 382.

"Congressional Staff Turnover"

32
Salisbury and Shepsle,
p. 562.

"U.S. Congressman" Op. Ci t .

33
Ibid.. pp. 562-563.

34
Malbin, Op. C i t . p. 21.

44

35
Patterson, Samuel L . , "The Professional Staffs of
Congressional Committees", Administrative Science Quarterly,
vol. 15, no. 1, March 1970.
36
Weingast, Barry E. "A Principal-Agent Perspective on
Congressional-Bureaucratic Relations", Fifth Carneigie
Conference on Political Economy, Carneigie-Mellon University,
Pittsburgh, June 1983.
37
Ibid.
38
Braybrooke and Lindblom, Op. Cit.
39
Lindblom, Op. Cit.
40
Lindblom, Charles, The Policy-Making Process. PrenticeHall, Inc., Englewood Cliffs, N.J., 1968.
41
Lindblom, The Intelligence of Democracy. Op. Cit.,
p . 63.

CHAPTER 2
DESCRIBING MICHIGAN STAFF:
THEIR DEVELOPMENT AND
INTERACTIONS WITH LEGISLATORS
Michigan’s State Legislature is the site selected to assess
staff impact on legislator accountability and
responsibility.

Before elaborating the formal research

argument and reporting the findings it yields,

the

legislative setting chosen for the analysis needs to be
described.

This description will build upon the themes of

growth and access identified in the preceding chapter.

Conditions contributing to the growth of political advisors
such as staff are related to the specific experience of
staff development in Michigan.

A political structure

stressing strong partisan leadership is found to be of
particular relevance.

The exercise of partisan leadership

in both the House and Senate are described and compared in
some detail.

Staff access to legislators is empirically defined by
documenting the key features of legislator-staff
interactions (i.e., frequency,
setting).

In addition,

duration,

initiator,

and

attributes of staff and legislators

relevant to their interactions are also quantified.

Staff

are used as the source for identifying and describing their

45

46

interactions with legislators.

The process of surveying

staff and the descriptive results from the data gathered are
reported.

Legislator characteristics were obtained from

various secondary sources which are reviewed and summarized.

STAFF DEVELOPMENT
In 1960, Michigan’s legislative staff numbered 120.

Their

numbers grew steadily for the next twenty years, reaching a
high point of 921 persons in 1980.
fallen to 815 individuals.

By 1983, their ranks had

Forces influencing staff

development have been grouped under three headings:
in the size of the state,

growth

increased levels of interactions

among polities, and the type of political structure
developed.

The changes observed in Michigan’s legislative

staff complement can be related to each of these factors.
Of the three, political structure has the most direct
bearing on staff.
Size of State
The relationship between the growth of government and the
growth of legislative staff is more aptly described as
symbiotic rather than causal.

Each benefits from the

other’s growth without being the sole, or even principal,
determinant of that occurrence.

Growth of government

affords greater resources that can be used to develop staff;
although decisions to develop staff will be based on more
than the availability of funds.

Similarly,

greater staff

capacity can facilitate legislative handling of a larger

47

governmental apparatus; but,

the political dynamics

associated with the desirability of governmental growth goes
well beyond staff’s ability to influence.
This symbiosis can be observed in Michigan’s history.
Table 1 displays the growth in dollars available to the
State of Michigan with the actual numbers of staff employed
in the state legislature for various years between 1960 and
1983.

[1]

To permit comparison, both indicators are

converted to a percent change from the previous reported
year.

What emerges from a comparison of the percentages is

a pattern of strong growth for both measures between 1960
and 1970 followed by increasing moderation to that growth
throughout the 1970’s.

In 1983, staff numbers declined

while total revenues maintained a pattern of growth,
although at a greatly reduced level.

Whether the decline in staff is a temporary aberration or a
harbinger of further declines in state resources cannot be
ascertained.

But, the historic pattern shared by the growth

in the size of the state and the size of the legislative
staff complement is clear.
numbers grow;

As the state grows, staff

and, conversely,

as the size of the state

moderates, staff growth slows and eventually declines.

Level of Interactions Among Governmental Entities
Michigan’s legislature,

like all legislatures in America,

exists as part of a complex mosaic of interactions between

48
Table 1
A Comparison of G r o w t h in M i c h i g a n G o v e r n m e n t a l
Revenues and Grow t h of M i c h i g a n L e g islative
Staff for Various Y ears Between
1960 and 1983

Governmental Revenues
Actual Dollars
% Change
(in 000)
from Prior
Period

N u mber of Legis l a t i v e S t a f f 2
Actu a l
% Change
Numbers
f r o m Prior
Pe riod

1960

$1,107,927

_

121

1970

$3,224,408

191.0%

329

172.0%

1973

$5,021,810

56.0%

566

72.0%

1977

$7,174,696

43.0%

874

54.0%

1980

$9,518,130

33.0%

921

5.0%

1983

$11,158,538

17.0%

815

-11.5%

Sources:
^State of M i c h i g a n Financial Statements
l e g i s l a t i v e Staff Directories,
professional and support.

includes all staff

-

49

different branches of state government;

other levels of

government comprising the federal system;
public and quasi-public bodies;

a variety of

and a seemingly endless

array of interests ranging from private individuals to
highly developed and well-financed organizations.

An

integral part of staff responsibilities requires an
awareness of, and an ability to participate in, the varied
relationships that comprise the legislative environment.

Quantifying these interrelationships cannot be done with any
precision.

However, some measure of the rise in

governmental interactions and its implications for staff
development can be gained by looking at the components of
Michigan’s finances since 1960.

In Table 1, the total

amount of revenues available to Michigan state government at
various times since 1960 is reported to show how growth in
the size of government was reflected in staff changes.
Table 2 arrays those total revenues by their component parts
(taxes,

federal agency monies,

local agency monies,

and

other) to demonstrate how the growth in government brought
new levels of governmental interaction that supported staff
growth.

When the components of Michigan’s revenues are viewed in
relative terms, as a percent of total revenues,

the most

noticeable change involves the proportions associated with
taxes and federal agency monies.

In 1960, taxes represented

Table 2
Mi c hi g a n G o v e r nmental Revenues Charact e r i z e d by Source,
Actu a l Amounts and Relative Percents, for
Various Years Between 1960 and 1983

Federal Agen c y

Total

Taxes

1960
Dollars (000)
% of Total

1,107,927
100.0

887,220
80.1

120,429
10.9

1970
Dollars (000)
% of Total

3,224,408
100.0

2,282,917
70.8

566,723
17.6

14,470
0.4

360.298
11.2

1973
Dollars (000)
% of Total

5,022,810
100.0

3,323,894
66.2

1,202,562
23.9

19,018
0.4

476,336
9.5

1977
Dollars (000)
% of Total

7,174,696
100.0

4,760,007
66.3

1,840,934
25.7

29,184
0.4

544,571
7.6

1980
Dollars (000)
% of Total

9,518,130
100.0

6,126,400
64.4

2,452,370
25.8

52,137
0.5

887,223
9.3

1983
Dollars (000)
% of Total

11,158,538
100.0

7,333,434
65.7

2,768,773
24.8

142,491
1.3

913,840
4.2

Source:

State of M i c h i g a n Financial Statements
*0ther includes service fees, licenses,

permits,

Local A g ency

and miscellaneous.

Not
Reported

Other *

100,278
9.0

51

80% of total revenues; by 1983,
15% to 65.7%.

this level had fallen almost

During the same period,

federal agency monies

rose as a percent of total revenues by an almost equivalent
percentage;
1983.

from 10.9% of total revenues in 1960 to 24.8% in

The fact that growth in federal monies going to

Michigan follows a pattern akin to that of staff growth
found in Table 1 is not surprising.

The increased

availability of federal funds brought with it new
requirements and opportunities that staff could help
legislators monitor and exploit.

Absolute growth for all of the revenue components also
occurred during the period reviewed.

While some portion of

this growth may be attributable to a larger economic base or
inflation, new sources of revenue or changes to existing
sources are also factors behind this growth.

The magnitude

of this growth invites interaction among interested parties.
Those upon whom these assessments fall will become more
concious of government’s receipt of their monies.

Those

seeking to benefit from government growth will utilize the
public programs and devices financed with these additional
dollars that did not exist in 1960.

In both instances,

legislators must be able to acknowledge and respond to the
various interests.

In an almost inevitable sequence of

events, growth in government leads to increased demands and
expectations of government that are expressed through
various forms of interactions between, within, and among

52

governmental bodies.

To keep abreast cf these developments,

legislators in Michigan as elsewhere have turned to staff
whose numbers grew accordingly.

Political Structure
In Michigan,

two words describe the political structure

governing staff development,

"partisan leadership".

Lucinda

Simon in her review of legislative staff in the fifty states
uses this term to describe staffing patterns not only in
Michigan but also in Massachusetts,
Pennsylvania.

New York, and

Her characterization of partisan leadership

fits Michigan well.
The partisan leadership model of legislative staffing
is very hierarchical with most personnel decisions
emanating from the principal legislative leaders.
Generally, the majority party leaders ultimately
determine the amount of salary monies that are made
available to individual members, committees, other
leaders and the minority party.
The party leaders also
decide how many staff will be employed, in which jobs
and at what salaries . . . In all of the states in this
grouping, staff is usually available to committess and
members.
Party leaders generally allow members and
chairman flexibility to hire their own staff, but in
some instances personnel choices may be dictated by the
leadership.
The management style is authoritarian
rather than collegial.
Legislative leaders tend to
rely on principal staff advisors rather than other
leaders for management decisions.
The offices of House
Chief Clerk or Senate Secretary are in most cases
important adjuncts of the majority party, overseeing
large staffs of chamber, managerial and support
personnel . . . The non-partisan status of most joint
agencies is almost an anomaly in these highly partisan
environments.
The budget and fiscal analysis staffs in
these states are the only offices that are somewhat
independent from the overall partisan heirarchy.
In
all four states, the money committees are assisted by
separately-budgeted staff offices . . . In Michigan,
the house and senate fiscal agencies are non-partisan
offices overseen by bi-partisan governing boards. [2]

53

The Speaker of the House and the Senate Majority Leader are
the partisan leaders who govern staff in Michigan.

The

influence these partisan leaders exert on staff development
and use is considerable.

However,

their authority over

staff does differ and in neither case is it absolute.

Of

the two, the Speaker exercises greater centralized authority
over staff development while the Senate Majority Leader
finds staff prerogatives dispersed to all State Senators.
House and Senate rules define the power of these partisan
leaders over staff.

These same rules also delimit and

qualify this power.

Specific legislative and constitutional

provisions govern certain staff agencies'
partisan leadership control.

relations with

The overall organizational

structure that results is diagrammed on the next page.

Its

details are discussed in the following paragraphs.
House.

Growth of House staff and the Speaker’s

centralized partisan control over these staff is generally
attributed to Representative William Ryan, a Democrat from
East Detroit who served as Speaker from 1969 to 1974.
During his tenure as representative and Speaker increased
staff was one of a series of reforms pursued to move
Michigan from a part-time,
full-time,

citizen’s legislature to a

professional legislative body.

Speaker Ryan, while continuing as a state representative,
stepped aside as Speaker to be succeeded by his protege,
Representative Bobby Crim, who held the Speaker’s position

Figure 1: Organizational Chart tor Staff Agencies of the Michigan legislature, 1984
HOUSE OF REPRESENTATIVES

Speaker
Majority Legs.,

Pers­
onal
Staff

Minority Ldr.

Appro.
Conmittee

Minority legs.

House
Fiscal
Agency

Corrm.

Clerk
Staff

House
Anal.
Sec.

Pers­
onal
Staff

Maj.
Caucus
KEY:
Srgt.atArms

Figure 1 (continued)
SENATE

Senate
Anal.
Sectior

Full
Senate

Judiciary
Coimittee

Sec. oi
Senate

Gen.
Counsel

Majority
Leader

Fiscal

Minority
Leader

Majority
Members

Srgt.-

atArms

Pers­
onal
Staff

Pers­
onal
Staff

Caucus
Staff

Caucus
Staff

KEY:

Staff

56

from 1975 to 1979.

Speaker Crim furthered the development

of centralized partisan staff by increasing the number of
staff devoted to the Democratic and Republican caucuses.
These staff, while serving the various members of their
respective partisan caucuses,

are accountable principally to

the leadership elected by each caucus.

This method of

staffing serves to reinforce partisan leadership control
over staffing in the House.

The current Speaker of the Michigan House, Representative
Gary Owen,

a Democrat from Ypsilanti, became Speaker in

1980. Once in power, he showed the continuing role of
centralized partisan control over House staff.

While not

changing the overall structure and composition of staff,
Speaker Owen took two administrative actions of significance
to staff functioning.

The first affected the Democratic

caucus staff which represents the largest single
concentration of House staff resources.

Speaker Owen

instituted greater reporting by these staff to his office,
placing the Speaker in a position of greater control over
staff actions.

To reinforce this control,

the Speaker took

a second step that altered all staff access to the floor of
the House while in session.

Previously,

staff had

relatively unfettered access to the floor.

This was changed

with the requirement that staff presence on the House floor
had to be approved in advance by the Speaker.

The

imposition of these requirments are recent examples of the

57

influence partisan leadership exercises over staff in the
House.

House rules reinforce the role of partisan

leadership in controlling staff development.
House Rule 6 reads as follows:
The Speaker shall appoint all committees except where
the House shall otherwise order.
He shall appoint
employees of the House, except as otherwise provided,
and such appointments and responsibility shall be
entered in the Journal.
All employees appointed by the
Speaker shall be subject to his orders and to summary
removal by him on failure to properly perform the
duties assigned them; the reason for such removal to be
reported forthwith to the House.

There are formal exceptions to this general authority of the
Speaker.

Those noted in the rules include delineation of

certain functions assigned to the Clerk and the
Sergeant-At-Arms to facilitate administration of the House.
Regarding committee staff,

the Clerk may appoint assistants

". . . by and with consent of the Speaker"

(Rule 19).

The

Clerk may prescribe the duties of committee clerks with the
approval of the respective committee chairs who are
responsible for the work of all committee aides (Rule 31E).

Public Act 412 of 1965 Section 16A establishes the House
Fiscal Agency as staff to the House Appropriations
Committee.

These staff are not governed by the Speaker but

by a committee of three members from the appropriations
committee,

including the committee Chair and two members

appointed by the Chair, one of whom must be a member of the
minority party.

53

Informal exceptions to the Speaker’s authority over staff
also exist.

Most notable of these are the delegation of

minority committee appointments and staff assignments to the
minority leader;

and, the administration of staff agency

functioning by agency heads.

The fact that both these

informal exceptions may be suspended by the Speaker serves
to underscore the strong role played by partisan leadership
over staff development in the House.

Senate.

While partisan leadership’s influence over

staff development is operative

in the Senate, its exercise

is far more decentralized than

in the House.

This point

is

well illustrated in the more detailed attention given to
staffing responsibilities in the Senate Rules.

Senate Rule 1.15 A under the heading of Senate
Organization reads as follows.

"The majority leader shall

assign duties to Senate employees not specified by other
rules and shall approve all expenses for the operation of
the Senate except as designated by statute."
rules" alluded to here include

The

a Secretary of the

"other
Senate who

is an elected officer of the Senate (Rule 1.6) with
responsibilities enumerated throughout the Rules.

Among the

Secretary’s duties is the oversight of Parlimentary
Procedure (Rule 1.7),

the maintenance of Senate journals

(Rule 1.8), and the orderly introduction and processing of
bills and resolutions

(Rule 1.9), along with a variety of

59

other administrative tasks (Rule 1.15 G-J).

Supervision of

the Sergeant-At-Arms under the direction of the majority
leader is a further task assigned to the Secretary.

Rule 1.18 provides for a General Counsel appointed not by
the majority leader but by the Senate upon recommendation of
the Judiciary Committee.

The General Counsel serves the

Senate by providing legal advice and assistance to any
Senator,

the Senate’s presiding officer,

or the Secretary of

the Senate as requested.

In preparing an annual budget for the Senate,

the majority

leader must discuss it with the minority leader (Rule
1.15C).

Further each Senator is alloted a separate budget

for staff and office operation with committee Chairs
allotted additional budget amounts for their committee
operations (Rule 1.15E).

These conditions on the majority leader’s authority
involving the minority leader and individual Senators are
repeated in the portion of the Rules dealing with Senate
employees.

Rule 1.39 gives the majority leader power to

appoint staff; however, minority staff employee
appointments must be made from a list submitted by the
minority leader.

Specific provision for each member

appointing staff directly responsible to that Senator is
found in Rule 1.42.

Committee Chairs appoint their clerks

60

from members of their staff (Rule 1.43); and, with
authorization from the majority leader, may appoint
additional staff (Rule 2.7).

Compensation of Senate employees must follow guidelines set
by the majority leader (Rule 1.40).

But, the majority

leader’s ability to terminate Senate employment is
substantially limited by Rule 1.41, which excludes employees
elected by the Senate (the Secretary and General Counsel) or
employees designated under other rules (personal and
committee staff).

A staff agency that is relatively independent from the
majority leader is the Senate Fiscal Agency.
counterpart,

Like its House

this non-partisan entity enjoys separate

legislative authority.

The Senate Fiscal Agency is

established by Public Act 412 of 1965 Section 16, which sets
its governance as the responsibility of the Chair of the
Appropriations Committee and two Appropriations Committee
members appointed by the Chair,

one of whom must be a member

of the minority party.

The fact that the Senate’s partisan leadership’s control
over staff is decentralized in no way negates its exercise
and influence over staff development.

This point was well

illustrated by the impact on staff brought about by a recent
change in the Senate’s partisan control.

In the aftermath

61

of a politically divisive battle over increased state income
taxes,

two Democratic state Senators were recalled and

replaced in special elections by Republicans.

The change in

these two seats was sufficient to alter the majority status
in the 38-member Senate from Democratic to Republican
control.

This shift in the Senate’s partisan leadership

resulted in changes to committee composition and Chairs, as
is expected.

It also showed the influence of partisan

leadership on staff assignments.

Democratic members

experienced reductions in their office staff (as well as
relocation of their physical office space).

Committee staff

also changed with committee Chairs.

The net effect of these changes on staff are shown in Table
3.

Numbers of staff (personal and committee) assigned to

Democratic and Republican Senators are compared between
1983, when Democrats held majority status,

and 1984, when

Republicans gained partisan leadership control, for those 36
Senators serving through the period of recalls and special
elections.

The staff shift is noticeable.

Republican

Senators each gained an average of one additional staff
person after the shift in partisan leadership.

Democratic

Senators lost an average of one-half staff person per
Senator.

The general distribution of Senators by number of

staff in their offices reflects this overall shift.

Most

Republican Senators now have three or more staff members
after the change in leadership as opposed to most having
only one or two when they were the minority party.

Of the

62
Table 3
A Compar i s o n of P ersonal and Committee Staff
Al l o cated to Repub l i c a n s and D e m o c r a t s
in
the M i c h i g a n State Senate U nder a
Democratic M a j o r i t y (1983) and a
Republi c a n M a j o r i t y (1984)

Nu mber of Senators

Number of Staff

1983

Democrat
1984

1

2

0

0

0

2

12

4

6

6

3

2

9

5

9

4

2

3

3

3

5

0

2

3

0

2.2

3.2

3.3

2.8

Mean Number
of Staff

Source:

Repu b l i c a n
1983
1984

Legisl a t i v e Staff Dire c t o r i e s

63

eighteen Republican Senators,
transition.

thirteen gained staff in the

For the Democrats, seven experienced a net

reduction in staff, eight maintained their staffing levels
through the change in leadership,

and three did manage to

acquire an additional staff member.

House and Senate Compared.

The exercise of centralized

partisan leadership control in the House and decentralized
partisan leadership control in the Senate has yielded
different staffing patterns between the two chambers.

The

historic development of various House and Senate staff
entities are compared in Table 4.

In 1960, the Senate had

more staff than the House; by 1970 this relative position
had reversed.

Throughout the 1970’s, House staff developed

more rapidly and sustained a growth in staff numbers longer
than the Senate.

Senate staff numbers peaked in 1977 while

the House staff complement continued to grow through 1980.

Within these overall levels,

the allocation of staff to

various functions show the influence of centralized versus
decentralized partisan leadership on staff development.

The

majority of Senate staff can be related to individual
Senators’ personal or committee staff reflecting the
decentralized character of Senate staffing.

In the House,

fewer than half of the staff complement are associated with
specific legislators;

however,

a much larger number are

allocated to Democratic or Republican caucuses.

This

64
Table 4
A C o m p a r i s o n of P r o f e s s i o n a l * Legi s l a t i v e Staff
Arrang e m e n t s in the M i c h i g a n H ouse and Senate
for Various Years Betw e e n 1960 and 1983

1960

1970

1973

9

30

-

Analy s i s Section
Fiscal A gency

Senate
Person a l / C o m m i t t e e
Caucus
- Democr a t i c
- Republican

Secretary
General Counsel
Total

House
Personal/Committee
Caucus
- Democratic
-Repub l i c a n

1977

1980

68

128

127

101

-

9
6

24
12

20
11

17
17

-

-

-

-

5

5

-

-

12

21

26

22

18

4

7

13

4

3

-

1

1

3

4

3

27

35

103

201

197

168

10

27

98

118

125

117

1

14
8

13
19

59
28

53
39

53
39

2

9

8

7

-

1983

A n a lysis Section

-

-

Fiscal A gency

-

-

13

17

24

24

Clerk

11

12

16

22

20

20

Total

22

61

161

253

269

260

Source:

Legisl a t i v e Staff Dir ectories
*Excludes all secreta r i a l and support staff

65

distribution supports the centralized partisan emphasis of
the House.
size.

Both chambers have fiscal agencies of comparable

The House Analysis Section was established prior to

the Senate Analysis Section.

Administratively,

the House

Analysis Section is an adjunct to the Speaker’s office,
while the Senate Analysis Section is the responsibility of
the Secretary of the Senate,

an arrangement that echoes the

centralized versus decentralized staffing patterns that
distinguishes the two chambers.

Only the Senate allocates staff for a General Counsel.
Differences between the staff devoted to the Clerk of the
House and the Secretary of the Senate are difficult to
compare due to changes in reporting of those staff over
t ime.

Joint agencies.

In addition to staff developed for

either the House or Senate, Michigan’s legislature is also
assisted by several joint staff agencies.

The Legislative

Service Bureau is the largest joint staff agency.

Its

history can be traced back to 1941 when it served as a
legislative reference library.

Over the years,

its

structure and responsibilities were redefined and expanded.
Public Act 412 of 1965 placed the Legislative Service Bureau
under the ageis of the Legislative Council established
through Article 4, Section 15 of the State Constitution of
1963.

Composition of the Council is defined by law in a

manner that emphasizes partisan leadership control.

Both

66

the House and Senate are given eight positions,

the majority

leader and seven of the majority leader’s appointees (3 of
whom must be from the minority party).

Current functions performed by the Legislative Service
Bureau include legal editting and research, a legislative
science office,

library and research facilities,

and a

variety of administrative divisions including secretarial
services, and printing and mailing capacities.

Public Act 46 of 1975 established the Office of the
Legislative Corrections Ombudsman as a further staff agency
overseen by the Legislative Council.

This office has a

small staff of six whose responsibility is to investigate
complaints against the Department of Corrections.

The Joint Committee on Administrative Rules, established by
Public Act 306 of 1969, consists of ten members,
the Senate and five from the House.

five from

Committee staffing and

expenses are handled jointly through appropriations from
each of the two chambers.

Its purpose is to review rules

and regulations proposed by executive agencies to assure
adherence to legislative intent.

Article 4, Section 52 of the State Constitution of 1963
provides the legal basis for a legislatively appointed
Auditor General.

Earlier state Constitutions had provided

67

for a popularly elected Auditor General.

Concern over the

autonomy from legislative control an elected Auditor General
might exercise had been a principal concern of legislative
reformers in Michigan during the 1940’s and 1950’s.

[3]

By

setting the term of Auditor General for eight years;
providing for his removal only by a two-thirds vote of each
chamber;

granting civil service status to most of the

auditor’s staff;

and prohibiting appointment or election of

the Auditor General to other state office for two years
after the end of his term,

the conduct of post-audits on all

state financial transaction was buffered from the influence
of electoral politics and partisan leadership.

68

LEGISLATOR-STAFF INTERACTION
Much of the theoretical and empirical arguments to follow
utilize staff access to legislators.

The most tangible

expression of such access is said to be the degree of
interaction between legislators and staff.

To document such

interaction a survey of legislative staff in Michigan was
undertaken.

Data derived from this survey described key

characteristics of these interactions and various attributes
of staff.

Additional data describing features of

legislators involved in these interactions were garnered
from secondary sources.

The process of staff and legislator

data acquisition is outlined and descriptive results
summarized in the remainder of this chapter.

Staff
In seeking to document legislator-staff interactions as the
most tangible expression of access,

a choice had to be made

between the use of legislators or staff as the primary
source for identifying and describing their interactions.
Staff were chosen for three reasons.

First of all, staff

responses were felt to be more reliable.

There are fewer

and better-known legislators for staff to remember than the
multiple and sometimes obscure staff that legislators meet.
Secondly, staff were more likely to provide valid responses.
Staff attentiveness to and ability to distinguish legislator
contact with staff from other contacts is likely to be

69

better than a legislator who is approached by multiple
individuals,

interests and constituents.

Finally,

acquiring

staff participation in a survey was more feasible.

Response

to surveys by legislators is at times deferred complicating
completion or delegated to staff complicating
interpretation.

The list of staff to be surveyed was drawn from the Michigan
Legislative Telephone Directory for 1983.

Support staff

such as secretaries, maintenance personnel,

and others in

solely administrative positions were deleted from the total
staff roster.

A universe of 428 individuals eligible for

survey participation was identified.

Surveys were hand

delivered to each staff person or their offices in January
of 1984.

Telephone follow-up was pursued for each House

staff member not responding by the February 1 return date.
When warranted, post card reminders were sent to those staff
indicating an intention to respond.

Table 5 details the various phases of survey administration.
Of the potential population of 428 individuals, 32 could not
be found at the location given in the Legislative Telephone
Directory and 25 were no longer employed,
potential set of 371 respondents.

leaving a

Table 6 reports the

response patterns derived from these potential respondents.
A total of 74 people attempted to complete the survey for an
overall response rate of 19.9*.

Incomplete or missing data

70

Table 5
Survey of Legisla t i v e Staff in Michigan,
Results of Survey D i s t r i b u t i o n

Eligible

Not
D eli v e r a b l e

1984:

No L o n g e r
Employed

Potential
Re s ponses

Senate
Person a l / C o m m i t t e e

101

10

0

91

Caucus
- Democratic

17

0

0

17

-Republican

17

0

0

17

5

0

0

5

22

0

0

22

Secretary

3

0

0

3

General Counsel

3

0

0

3

168

10

0

158

117

18

11

88

Caucus
-Democratic

53

4

6

43

- Republican

39

0

7

32

7

0

0

7

Fiscal A g e n c y

24

0

0

24

Clerk

20

0

1

19

Total

260

22

25

213

59

0

0

59

428

32

25

371

Analysis Section
Fiscal A g e n c y

Total
House
P e r sonal/Committee

Analysis

Section

Joint Agencies
Grand Total

71

Table 6
Survey of Legisl a t i v e Staff in Michigan, 1984:
A c t u a l Responses, U s able Responses, and
C o r r espon d i n g R e sponse Rates

Potential
Responses

Actual
Responses

Actual
Response
Rate

Usable
Final
Responses Response

Senate
Person a l / C o m m i t t e e

91

15

Caucus
-Democratic

17

-Repub l i c a n

16.5%

15

16.5%

0

0.0%

0

0.0%

17

4

23.5%

4

23.5%

5

3

60.0%

0

0.0%

22

0

0.0%

0

0.0%

Secretary

3

0

0.0%

0

0.0%

General Counsel

3

0

0.0%

0

0.0%

158

22

13.9%

19

11.3%

Personal/Committee

88

19

21.6%

18

15.4%

Caucus
- Democratic

43

11

25.6%

8

18.6%

-Republican

32

3

9.4 %

2

6.3%

7

3

42.9%

2

28.6%

Fiscal A g e n c y

24

8

33.3%

7

29.2%

Clerk

19

2

10.5%

0

0.0%

Total

213

46

21.6%

37

14.2%

59

6

10.2%

4

6.8%

371

74

12.9%

60

16.2%

Analysis

Section

Fiscal A g e n c y

Total
House

Analysis S e c tion

Joint Agenc i e s
Grand Total

72

caused the deletion of 14 responses for a final pool of 60
responses or 16.2% of the potential population.
information is detailed for each staff agency.

Similar
The timing

of the survey (immediately following the recall election of
1983) precluded follow-up to encourage response in the
Senate which may account for the lower response rate when
compared to the House where follow-up was done.

Table 7 compares the percent of staff in each agency for all
371 potential respondents,

the 74 actual respondents,

the 60 responses used in the analysis.

and

Despite the low

response rates the distribution of responses in the House is
fairly representative of the actual agency staff
complements.

In the Senate,

the largest group of staff

(personal and committee staff) are also appropriately
represented.

Reported Characteristics of Interactions.

The actual

survey instrument staff were asked to complete is included
as Appendix A.

The instrument is divided into eight parts.

Part 1 asks for information detailing staff interactions
with legislators.

Interactions were defined as work-related

discussions, involving direct conversations between a staff
person and a legislator related to the conduct of staff
responsibilities.

Specifically excluded were discussions

during committee hearings, non-work-related personal
conversations with legislators,

or legislator contact that

was social or recreational in nature.

Staff were asked to

73

Table 7
Survey of Legisl a t i v e Staff in Michigan, 1984:
Percentage Compa r i s o n of Surveyed P opulation
and Re s p o n d e n t P o p u l a t i o n

Potential
Respondent
Popula t i o n

Actual
Respondent
P opulation

F inal
R espondent
Population

Senate
P e r s onal/Committee

21.2%

20.3%

25.0%

Caucus
- D emocratic

4.0%

0.0%

0.0%

- Republican

4.0%

5.4%

6.7%

Analy s i s

1.2%

4.0%

0.0%

5.1%

0.0%

0.0%

Secretary

.6%

0.0%

0.0%

General Counsel

.6%

0.0%

0.0%

36.7%

29.7%

31.7%

P ersonal/C o m m i t t e e

20.5%

25.7%

30.0%

Caucus
-Democratic

10.0%

14.9%

13.4%

-Repub l i c a n

7.4%

4.0%

3.3%

Analysis

1.6%

4.0%

3.3%

Fiscal A gency

5.6%

10.8%

11.7%

Clerk

4.4%

2.7%

0.0%

Total

49.5%

62.2%

61.6%

13.8%

8.1%

6.7%

100.0%

100.0%

100.0%

Section

Fiscal Agency

Total
House

Section

Joint Agencies
G rand Total

74

identify by name those legislators with whom they had at
least one such work-related discussion per month during the
months of September 1983 to December 1983.

If staff had

such discussions with more than ten legislators during this
time, they were asked to select and name the top ten;

few

staff identified a total of ten legislators.

For each legislator named,

staff were asked to describe five

features of interactions with that legislator:

1.

Frequency of their discussions (ordinal response
categories were daily, more than once per week,
once per week, more than once per month, once per
month).

2.

Duration of their discussions (ordinal response
categories were less than 15 minutes,

15 to 30

minutes, 30 minutes to one hour, one to two hours,
over two hours).

3.

Usual initiator of their discussions

(nominal

response categories were the staff person, the
legislator,

4.

or third party).

Nature of their discussions (nominal response
categories were private talks or group
discuss ions).

75

5.

Subjects covered in their discussions (nominal
response categories were given for twelve different
subject areas pretested on staff).

These are the basic features of access.
For the first four features describing their discussions,
staff responses were constrained to selecting the one
response that provided the best overall description of their
typical interaction with that legislator.

In reporting the

subjects covered during their conversations,

staff were

asked to rank order the various topics by frequency of their
occurrence with one being most frequent,
frequenty,

etcetera.

two the next most

Staff were also allowed to specify and

rank other subjects of discussion although few took this
opportunity.

Table 8 summarizes the ordinal aspects of the interactions
with legislators reported by staff.

A total of 313 distinct

legislator-staff interactions were identified by the 60
staff respondents,

or an average of 5.22 legislators

reported per staff person.
contacts (56.5%) were short,

Most of the legislator-staff
lasting less than 15 minutes;

the most common level of frequency reported was more than
once per month (but less than once per week) which
encompassed 31.3% of all legislator-staff interactions.
Overall,

the frequency of meetings are more evenly

distributed across the available response categories;

while

Table 8
Survey of L e g islative Staff in Michigan, 1984:
Frequency and Du r a t i o n
of Contacts Between L e g islators and Staff

Frequency
Once per
Month

More than
Once per M o n t h

Once
per W e e k

#

40

74

34

%

12.8

23.6

10.9

Estimated D u r ation

Under 15 Minutes

15 - 30 minutes

I - 1 hour

1 - 2

hours

Over 2 hours

Total

More than
Once per We e k

24

Daily

5

Total

177

7.7

1.6

56.5

#

8

16

19

31

13

87

%

2.6

5.1

6.1

9.9

4.2

27.8

#

3

4

4

8

30

%

1.0

1.3

1.3

3.5

2.6

9.6

#

0

4

0

6

6

16

%

-

1.3

-

1.9

1.9

5.1

#

0

0

0

0

3

3

%

-

-

-

-

1.0

1.0

#

51

98

57

72

35

313

%

16.3

31.3

18.2

23.0

11.2

100.0

11

the reported duration of the typical meetings between
legislators and staff was short with only 6.1% lasting over
one hour.

Table 9 interrelates the two nominal features of the
reported contacts,

the nature of the meetings and who

usually initiated these discussions,
of legislator-staff interactions.

to describe the setting

Meetings between

legislators and staff are most often characterized as
private meetings initiated by staff.

Legislators initiate a

substantial minority of the discussions;

these too are

typically private meetings.

The subjects discussed at these meetings tend to focus on
the legislative process more than constituent activities or
relations with the executive branch.

As Table 10 shows,

backgound on legislation is the most frequently mentioned
topic of legislator-staff discussion and ranks highest in
priority of all the subjects mentioned.
strategy,

committee activities,

Legislative

and policy development all

receive a top priority mention in over 10% of the 313
reported discussions;

however,

they are not mentioned as

topics in over half of the legislator-staff interactions.

Written Products.

Part two of the survey attempted to

quantify a second form of legislator-staff interaction,
written communications.

Unfortunately,

attempts to have

78

Table 9
Survey of L e g i s l a t i v e Staff in Michigan, 1984:
Initia t i o n and N a ture of Contacts Between
Leg islators and Staff

Initiator
Staff

Legis l a t o r

#

130

107

20

257

%

41.5

34.2

6.4

82.1

22

22

8

52

7.0

7.0

2.6

16.6

//

154

131

28

313

%

49.2

41.9

8.9

100.0

Nature

Private
M e eting

Group
Discus s i o n

Total

Other

T otal

Table 10
Survey
of Legis l a t i v e Staff in Michigan, 1984:
Subjects D i s c u s s e d in
Legislator-Staff Contacts Ordered by Frequency of M e n t i o n and Priority

Priority
Subjects Discussed

1. Background on
L egislation

2. Legislative
Strategy

3. Committee
Activities

4. Policy
Development

5. Constituent
Requests

6. Background on
Budget

Frequency

1st

2nd

3rd

Below
3rd

Not
M e n t ioned

Total

#

76

50

21

26

140

313

%

24.3

16.0

6.7

8.3

44.7

100.0

#

52

43

19

36

163

313

%

14.0

13.7

6.1

14.1

52.1

100.0

#

42

30

33

31

177

313

%

13.4

9.6

10.5

10.0

56.5

100.0

#

38

30

17

23

205

313

%

12.1

9.6

5.4

7.4

65.5

100.0

#

29

14

13

29

228

313

%

9.3

4.5

4.2

9.2

72.8

100.0

#

22

19

18

17

237

313

%

7.0

6.1

5.7

5.5

75.7

100.0

Table 10 —

Continued

Priority
Subjects Discussed

7. Bill Drafting

8. Home District

Frequency

Not
M e n t ioned

Total

27

239

313

5.4

8.7

76.4

100.0

11

7

22

258

313

3.5

2.2

7.1

82.4

100.0

1st

2nd

3rd

#

12

18

17

%

3.8

5.7

#

15

%

4.8

Below
3rd

co
o

9. Executive Agency
Oversight

10. Interest Group
Relations

11. Office Admin.

12. Speech Writing

#

5

12

10

24

262

313

%

1.6

3.8

3.2

7.7

83.7

100.0

#

3

9

29

265

313

%

1.0

2.2

2.9

9.2

84.7

100.0

#

2

8

6

10

287

313

%

0.6

2.6

1.9

3.2

91.7

100.0

#

4

1

8

297

313

%

1.3

0.3

2.5

94.9

100.0

7

3
1.0

81

staff describe the types of written products proved to be
too complicated for a mail survey.

Many respondents noted

difficulty in adequately describing the range of written
products they produce and felt uncomfortable giving equal
weight to seemingly similar endeavors

(e.g., preparation of

background reports) that may entail widely divergent
expenditures of time in research and writing.

Further

complicating responses to this section of the survey was an
attempt to associate volume indicators with the various
types of communications listed.

For example,

letters may

dominate the volume of written products generated by staff
but these may be similar letters to common constituent
requests that require little additional effort to produce.

Given these shortcomings,

data from this portion of the

survey were deleted from the analysis.

The only tangible

insight to be gained from an attempt to quantify staff
written products was that this form of staff activity was
generally requested by a legislator for his or her use.
Further assessment of this form of staff activity would
require some type of content analysis or a review of actual
staff written products to be of value.

Work Environment.

Descriptions of their working

environment were elicited from staff in parts three and four
of the survey.

Such description is integral to

understanding the context in which legislator-staff
interactions occur.

82

One of the more important attributes of any work setting is
its stability.

Table 11 shows that most staff (30) came to

their present job from outside government and have been
employed as staff just over three years.

Some internal

promotion possibilities are suggested by the 24 staff
members who reported their prior position as another staff
position in the Michigan legislature.

These respondents

were employed somewhat longer in their present capacity than
their colleagues.

The only groups reporting appreciably

longer lengths of service were those entering with other
governmental experience outside Michigan’s legislature;
however,
small.

the number of respondents in these categories are
House employees, both partisan and non-partisan,

report longer tenure than Senate staff members.

Attributes of the staff work setting concerning their
relationship with others is summarized in Table 12.

To

better understand the staff population being analyzed,

the

total pool of 74 respondents is compared to the final set of
60 staff members.

The 14 responses eliminated from final

analysis were predominantly those with no legislator
interactions in the form of work-related discussions to
report.

By comparing the total respondents with those able

to report legislator-staff interaction some inkling of
differences between staff with and without access to
legislators can be obtained.

83

- Table 11
Survey of L egislati v e Staff in Michigan, 1984:
Staff Tenure Categorized by Prior E xperience
and Current W o r k A r e n a for
All Staff R e spondents

N umber of
Staff

M e a n Tenure
in M o nths
in Years

Prior Experience
24

49.2

4.1

M i c h i g a n Government
(non-legislative)

5

75.0

6.2

Other Government
(non-Michigan)

6

63.0

5.2

30

39.2

3.3

E d u c ation

5

44.4

3.7

M i s cellaneous

4

52.2

4.3

Mi c h i g a n Legislature

Non-Government

Current W o r k Arena
House,

overall

18

51.9

4.3

House,

Democrats

25

45.9

3.8

House,

Republ i c a n

4

44.5

3.7

5

41.4

3.4

Senate, Democrats

9

43.0

3.6

Senate,

8

11.2

.9

5

38.6

3.2

74

42.4

3.5

Senate,

overall

Republican

House and Senate
Total

84

Table 12
Survey of Legis l a t i v e Staff in Michigan, 1984:
Characteristics of Staff W o r k E n v ironment for
Total and Final R e s p o n d e n t s

To t a l R e s p o n d e n t s

#
Superior
-Legislator
-Staff

%

Final R e spondents

#

%

33
41

44.6
55.4

32
28

53.3
46.7

50
9
9

67.6

41

12.2
12.2

8
8

6

8.0

3

68.4
13.3
13.3
5.0

Partisan Activity
-Expected
-Not Expected

32
42

43.2
56.8

29
31

48.3
51.7

Legislator Identification
-Specific Legislator
-Particular Legislator
-None

33
23
18

44.6
31.1
34.3

31
19

10

51.7
31.7
16.7

44

21

59.4
28.4

9

12.2

39
7
14

65.0
11.7
23.3

Number of Subordinates
None

1 - 2
3 - 5
over 5

Legislator Influence
on Employment
-Most Important
-Others More Important
-Not Important

85

The final set of respondents are more likely to report
directly to a legislator which lends credence to associating
the lack of access to legislators with the elimination of
the 14 respondents.

The two groups are similar in terms of

their relationships with subordinates;

and, a majority of

both groups are not expected to participate in partisan
activity as part of their job.

The final group of

respondents are more likely to identify with a specific
legislator,

reinforcing the stronger presence of access in

their ranks when compared to the total set of respondents.
A comparison of the reported influence of legislators on
each staff person’s employment shows that for the final
respondents legislators are either the most important
influence or are not involved.

Legislators discussing

hiring/firing decisions with others is more common in the
total respondent set.

Table 13 completes the description of the staff work
environment by summarizing various personal attributes (age,
education,

and salary levels) for all respondents and the

final set of 60 respondents.
part seven of the survey.

These data were requested in

No substantial differences

between the two respondent groups are found.

Staff are

young (most are under 34 years old); well-educated (85-90%
are college graduates,
and most

over 40% have post-graduate degrees);

(56.7%) earned under $24,000.

86

Table 13
Survey of Legis l a t i v e Staff in Michigan, 1984:
Summary of Staff P e r s o n a l Attri b u t e s for Total
and Final R e s pondents

Total Respondents
%
#

Final Respo n d e n t s
#
%

Age
Under 25
25 - 29
30 - 34
35 - 39
40 - 44
45 - 50
Over 50
Missing

9
18
16
11
7
6
5
2

12.2
24.3
21.6
14.9
9.5
8.1
6.8
2.6

7
17
13
8
7
4
2
2

11.7
28.3
21.7
13.3
11.7
6.7
3.3
3.3

Education
High School
Some College
College
Some Post Graduate
Masters
Doctorate
Law
Missing

2
9
22
11
23
2
4
1

2.7
12.2
29.7
14.9
31.1
2.7
5.4
1.3

6
20
8
21
1
4
1

10.0
33.3
13.3
35.6
1.7
5.0
1.7

Salary Levels
Under $15,000
$15,000 - $19,999
$20,000 - $24,999
$25,000 - $29,999
$30,000 - $34,999
$35,000 - $39,999
Over $40,000
Missing

6
16
20
12
7
3
7
3

8.1
21.6
27.0
16.2
9.4
4.0
9.5
4.0

4
14
16
11
3
3
6
3

6.7
23.3
26.7
18.3
5.0
5.0
10.0
5.0

87

Attitudes.
instrument

The remaining sections of the survey

(parts five and six) pose questions designed to

reflect staff attitudes.

Most of these items are used

directly in the testing of the formal argument regarding
staff and legislator accountability found in Chapter 4.
Their analysis is deferred until then.

LEGISLATORS
Data on legislators was compiled from secondary sources,
principally the Legislative Telephone Directory for 1983,
legislative rosters,

and committee listings produced for

legislators and their staff.

Each legislator was assigned a

number reflecting whether the person was a Representative or
a Senator,

party affiliation,

and the unique identifier used

to link all references to a specific legislator by any of
the 60 staff respondents.

Common information on all

legislators was obtained reflecting tenure, number of
committee assignments and types of positions held on those
committees,

and the number of staff assigned to a

legislator’s office.

These variables relate not only to

legislative authority and position, but also, to
availability of staff, both important factors influencing
levels of legislator-staff interactions.
positions held were also noted; however,

Leadership
these data proved

too difficult to interpret because most legislators in
Michigan hold some leadership title and differentiating the
relative authority associated with these titles is not
poss ible.

88

Data on legislator tenure,

committee assignments,

and office

staff size are summarized in Tables 14, 15, and 16
respectively.

Each table compares absolute and relative

numbers of legislators between all elected legislators and
those mentioned by at least one staff respondent.

Overall differences are not large between the total
legislator complement and those identified as a source of
interaction by staff.

All but thirteen House Democrats and

thirteen House Republicans are mentioned at least once by
staff.

On the Senate side omissions are smaller still with

only three Democrats and one Republican not included.

The

proximity of these total legislator counts do not allow for
great differences between the two populations being compared
and little difference is observed.

Table 14 shows that for

House Democrats there is a higher proportion of short tenure
(under two years) legislators among the legislators
mentioned by staff than legislators overall.

For Senate

Democrats there are fewer legislators included for analysis
in the high tenure category (over ten years)
Senate Democrats. Committee assignments
comparable for both sets of legislators.

than among all

(Table 15) are
Regarding reported

office staff (Table 16) only House Democrats show any
appreciable difference between the two groups.

Relatively

more legislators identified by staff had no explicitly
identified office staff.

Table

14

L egislator Tenure by Chamber and Party of A c t u a l Legislator Complement
and L e g islators Identified by Staff in the Survey of
Legislative Staff in Michigan, 1984

House

Tenure
Less than
2 Years

2 - 4
Years

4 - 6
Years

6 - 8
Years

8 - 1 0
Years

Over 10
Years

Actual

Democrats
Identified

Senate
Republicans
Identified
Actual

Actu a l

Democrats
Identified

Republicans
A c tual
Identified

#

20

19

23

15

4

3

11

10

%

31.7

38.0

48.9

44.1

22.2

20.0

61.0

58.8

#

7

5

7

5

0

0

0

0

%

11.2

10.0

14.9

14.7

-

-

-

-

#

9

5

6

5

4

4

5

5

%

14.3

10.0

12.8

14.7

22.2

26.7

27.8

29.4

#

6

6

3

1

2

2

1

1

%

9.5

12.0

6.4

2.9

11.1

13.3

5.6

5.9

#

6

5

4

4

3

3

0

0

%

9.5

10.0

8.5

11.8

16.7

20.0

-

-

#

15

10

4

4

5

3

1

1

%

23.8

20.0

8.5

11.8

27.8

20.0

5.4

5.9

Table 14 —

Continued

House

Actual

Total

Democrats
Identified

Senate
Republicans
Actual
Identified

Actu a l

Democrats
Identified

Republicans
Actual
Identified

#

63

50

47

34

18

15

18

17

%

100.0

100.0

100.0

100.0

100.0

100.0

100.0

100.0

vo
o

Table 15
Number of Committees by Chamber and Party for the A c t u a l L egislator
Complement and Legislators Identified in the Survey
of Legislative Staff in Michigan, 1984

Senate

House
Number of Committees
Actual

1

Democrats
Identified

Republican
Actu a l
Identified

Actu a l

Democrats
Identified

Republicans
Actual
Identified

#

15

12

6

5

4

4

6

6

%

23.7

24.0

12.8

14.7

22.3

26.7

33.3

35.3

#

2

1

1

1

6

5

11

10

%

3.2

2.0

2.2

32.9

33.3

3.3

61.1

58.8

#

2

1

8

5

6

4

1

1

%

3.2

2.0

17.0

14.7

33.3

26.7

5.6

5.9

#

9

8

27

19

2

2

0

0

%

14.3

16.0

57.4

25.9

11.1

43.3

-

-

#

35

28

5

4

0

0

0

0

%

55.6

56.0

10.6

11.8

-

-

-

-

#

63

50

47

34

18

15

18

17

%

100.0

100.0

100.0

100.0

100.0

100.0

100.0

100.0

2

3

4

5

Total

Table 16
Number of Staff A s s o c i a t e d W i t h Legislators by Chamber and Party
for the Actu a l
Legis l a t o r Complement and L e g islators Identified
by Staff in the Survey of Legislative Staff in M i c h i g a n - 1984

House
Number of
Staff

A ctual

Democrats
Identified

Senate
Republicans
Actual
Identified

Actual

Democrats
Identified

Republicans
A c tual
Identified

#

30

21

1

1

0

0

2

1

%

47.6

42.0

2.1

2.9

-

-

11.1

5.9

#

12

9

0

0

6

5

11

11

%

19.0

18.0

-

-

33.2

33.3

61.1

64.7

#

1

1

1

1

5

4

3

3

%

1.6

2.0

2.1

2.9

27.8

26.7

16.7

17.6

//

0

0

0

0

3

2

2

2

%

-

-

-

-

16.7

13.3

11.1

11.8

#

2

2

0

0

3

3

0

0

%

3.2

4.0

-

-

16.7

20.0

-

-

#

18

17

45

32

0

0

0

0

%

28.6

34.0

95.8

94.2

_

5 or 6

Unknown

Table 16 —

Continued

House
Number of
Staff

A c tual

Democrats
Identified

Senate
Republicans
Actu a l
Identified

Actual

Democrats
Identified

Republicans
A c tual
Identified

#

63

50

47

34

18

15

18

17

%

100.0

100.0

100.0

100.0

100.0

100.0

100.0

100.0

Total

94

Given the close correspondence between these legislative
populations, observations about legislators as a whole would
appear to be applicable to the legislators included in the
analysis of legislator-staff interactions.

Their

characteristics reinforce the importance of majority
position in a legislature.

Democrats have more legislators

with longer tenure than Republicans in both the House and
Senate (Table 14); Democrats serve on more committees than
Republicans in both the House and Senate (Table 15); House
Democrats are more likely to have at least one staff person
than their Republican colleagues;

and, at least prior to the

two recall elections discussed, more Senate Democrats had
large staffs than Senate Republicans (Table 16).

The difference between staffing patterns in the two chambers
(Table 16) are pronounced.

Most House Republicans have no

staff specifically assigned to their office and a
substantial minority of the House Democrats are in a similar
situation.

State Senators, however,

all have at least one

office staff member, with most having two or more.

These

differences echo the earlier discussion of centralized
versus decentralized partisan leadership control in the
House and Senate.

Partisanship and the power of the

majority leadership permeate not only considerations of
staff development but legislators themselves.

SUMMARY
The examination of growth and access is elaborated for
Michigan specifically in this chapter with consideration of
staff development and legislator-staff interaction in the
State legislature.

Staff development in Michigan is found

to be related to factors influencing advisory staff growth
generally.

A political structure emphasizing partisan

leadership control conditions much of staff development in
Michigan.

Differences between the House and Senate showed

different ways partisan leadership is exercised.

In both

instances, whether it is the centralized House approach or
the Senate’s decentralized use of partisan control,

its

presence influences staff.

Interactions with legislators reported by staff are
documented as the most visible and tangible expression of
access.
nature,

The dimensions of access (frequency, duration,
initiator,

described.

and content) are quantified and

Staff attributes are also reported to help place

the interactions reported in a context.

While overall staff

participation in the survey used to garner these data is
low, the distribution of respondents is felt to be
adequately representative of the House and Senate’s largest
staff groupings.

Information on legislators is obtained from secondary
sources to describe the key features of the legislators
participating in legislator-staff interactions reported by

96

staff.

Almost all legislators are included among the

various interactions reported.

Themes of partisanship and

majority control permeate legislator characteristics.

All data gathered on staff from the survey instrument and
legislators from secondary sources were computerized using
the facilities of the Michigan Terminal System at Wayne
State University.

Data were entered and converted for use

with the data base management package known as MICRO.

Code

books created from this process are found in Appendix B.
Initial analyses of data were performed using MICRO.

More

detailed analyses were done using the Michigan Interactive
Data Analysis System (or MIDAS) software package.

The

statistical procedures available through MIDAS provided the
technical basis for using the data described in this chapter
to empirically test the formal argument about legislator
staff and legislator accountability presented in the next
chapter.

CHAPTER 2:

References

1
The selection of these years was arrived at as
follows:
1960 and 1970 offer historic reference points to
guage the magnitude of current trends; 1983 is the most
current year of data available for all indicators and the
year reflected in staff survey responses; 1973, 1977 and
1980 are selected as roughly equivalent intervals which
reflect various phases of Michigan’s cyclical economy.
For
consistency, use of these years is maintained throughout for
historical analyses.
2
Simon, Lucinda Understanding Legislative Staff
Development:
A Legislator’s Guide to Staffing Patterns.
National Conference of State Legislators, Denver, August
1979, p. 65.
3
Staff Report to the Michigan Joint Legislative
Committee on Reorganization of State Government, The
Legislature. Report # 1 1 . March 1951, pp. 11-54.

97

CHAPTER 3
ACCISS AND ACCOUNTABILITY

Despite the significance of staff development it does little
to further our understanding of legislator accountability
and the responsibility it presumes.
issues,

To address these

staff cannot be considered apart from legislators.

The nature and extent of staff access to legislators is
considered central to understanding legislative staff
activity and its implications for legislator accountability.

ACCESS
The emphasis placed on access is warranted for at least
three reasons.

First of all,

levels of access reflect

consumption of limited legislator time.

The amount of time

spent with staff will bear on a legislator’s priorities and
the ability to attend to other duties such as collegial
deliberations.

Secondly,

amount of access,

use of staff,

as evidenced in the

is indicative of the allocation and

acquisition of scarce staff resources.

Ascertaining who

receives the greater amounts of available staff time
demonstrates who benefits from legislative staff use.
Finally,

how staff relate to legislators during these

interactions permits examining the functioning of this new
cadre of legislative actors and how staff have been
assimilated into the legislative process.

98

99

The raison d ’etre of staff requires then to relate directly
or indirectly to legislators’ individual or collective
needs.

Legislator-staff interaction or, sore simply,

access

is the key to understanding how legislators use their added
staff resources.

Advisory Access Generally
Just as historical parallels assisted in understanding the
growth of staff,

so too,

important insights regarding access

cone from considering other advisors.

The advisory relation is decisively effected by the
nature of the access that the advisor has to his
principal. [1]

The dimensions of access cited to define its nature include;
fora (use of the spoken and/or written word),
(brief or long),

frequency (continuous or sporadic),

ness (with or without
initiates),

duration

intermediaries),

genesis

direct­

(who

and exclusivity (principal/preferred source or

multiple competing sources).

Opportunities to (advise) will depend on where in the
spectrum the advisor is situated with respect to each
of these dimensions of access.
These variations in
access also affect almost all other aspects of the
advisor-advisee relation and more especially the
strength of the advisor's influence and the likelihood
that his advise will be implemented.
[2]

Other authors, while using different concepts and terms have
identified additional factors that influence interactions

100

between advisors and their principals.

For example,

identi­

fiable patterns of behavior have been associated with indi­
viduals in advisory positions.

Lerner

[3] in controlled

laboratory analogues of expert behavior identified three
behavioral types.
by Meltsner

[4]

These correspond closely to groups found

in his examination of policy analysts in the

federal bureaucracy.

These and other typologies include:

technicians who perform a job,

relating chiefly to

professional peers in their advisory activity;

climbers

who appeal to an immediate superior or client for support
often aimed at job advancement;

and entrepreneurs who seek

change in public policy and the allocation of resources
towards some preferred end.

The differing priorities inherent in these types points to
different styles and purposes associated with
advisor-principal

interactions.

The presence of these

differing attitudes among staff in Michigan was documented
as part of the staff survey conducted for this research.
Staff were asked to rank three statements reflecting the
triad of opinions towards their work.
results.

Table 17 arrays the

Relatively few ranked the technician role highly;

most placed priority on a climber identification although
the entrepreneurial emphasis was a close second.

The specificity of client demands upon advisors goes far
toward determining the nature of advisory activity.

Merton

101

Table 17
Survey of L egislat i v e Staff in Michigan, 1984:
Views of Staff W o r k R a nked First and Second
by Final Staff Respondents

Statement R a nked First
Statement R a nked Second

Techni c i a n

#

Technician

-

%

Climber

#
%

Entrep r e n e u r

#
%

Total

#
%

7

Climber

Entrepreneur

5

10

8.3

16.7

-

13

11.7

21.7

5

20

8.3

33.3

12

25

23

41.7

38.3

20 . 0

-

Total

15
25.0

20
33.3

25
41.7

60
100.0

102

hypothesizes an inverse relation between specificity and
advisor latitude.

The earlier in the continuum of decision that the
(advisor) operates, the greater his potential influence
in guiding the decision.
[5]

Similiarly,

the degree of technical complexity influences

the advisory relationship.

Lerner hypothesizes a

curvilinear affect with advisor influence waning beyond some
point of technical

intricacy set idiosyncratically by those

receiving the advise.

[6]

Organizational position also has a bearing on advisor
access.

The position of advisors within a hierarchical

organization defines their autonomy in the sense that higher
units may consider options presented by lower units or take
a lower unit's reaction to a decision into account;
larly,

simi­

lower units may anticipate reactions of upper units.

In short,

organizational situation conditions the advisor's

involvement with a problem and its solution.

[7]

Attributes of both advisor and principal have also been
found to influence their relationship.

The activities of

advisors will differ depending upon professional training
and education as well as the positions and titles they
assume in an organization.
perspective,
reasons;

[8]

Viewed from the principal's

different clients seek advise for different

they also differ in their capacity or inclination

to use and learn from advise.

[9]

103

Staff access specifically.
The descriptive literature on legislative staff is permeated
with observations on access.

Staff are dependent on legis­

lators from whom they derive their influence

[10] and by

whom that influence is defined [11] and limited

[12].

The

legislator-staff relation has been variously described as
one of "love/hate"

[13] or "feudal".

specific characterization,

[14]

Whatever the

all connote a bond between

legislators and staff that underscores the import of staff
access to and interaction with legislators.

The import of access becomes even more clear when staff
activity is placed in the context of the overall legislative
system.

Staff are just one set of actors

in a competitive

market of ideas and services seeking legislator attention.
Forces outside the legislature in the executive branch,
political parties,

and both organized and unorganized

interest groups vie for legislator support.
legislature,

Within the

legislative colleagues and legislative staff of

various descriptions interact and compete with each other.

Various examinations into legislators'

use of information

sources show signficant staff utilization;

although this

performance varies across aspects of the legislative
environment.

In 1954,

a survey of 27 congressmen found

office staff,

committee staff, and staff agencies to be

among the top ten sources of information in terms of

104

frequency of use.

[15]

Saauel Patterson's "The

Professional Staffs of Congressional Committees"

[16] showed

legislators relying heavily on office staff for legislative
research but turning to their colleagues in preparation for
floor debates and aeetings.
Representatives
specialist

Robert Z w i e r ’s study of U.S.

[17] drew a sharp distinction between

legislators who place great importance on staff

in their information search and non-specialist who depend
aore on colleagues or constituents.

Using the index score

developed by Zwier, Robert Bradley [18] found Nevada state
legislators placed their staff bureau a close second to
committee hearings as the most useful source of information
on technological

Staff are used;
factors.

issues.

but that use is conditioned by a variety of

There are functional constraints on access in the

sense that certain functions are called upon as needed while
others Bay entail aore regular contact.

Normative con­

straints also define operating rules that channel staff
activity and contact with legislators.

[19]

Staff access is frequently discussed in organisational terms
with organization expressed as staff position relative to
legislators.
services

Meller in his review of legislative staff

[20] suggested placing staff in a "legislative

service matrix" defined by two axes,
identification.

clientele and personal

The clientele axis expresses whether a

staff person identifies with a specific legislator at one

105

extreme or the legislature as a whole at the other.
polls of the personal

identification axis range fron

personal involveuent to anonyaous objectivity.
[21] groups staff into two categories.

staff are other staff agencies.
three staff office structures,
[22]

Leonard Saxe

Direct staff are

those who work for a aeaber or coaaittee chair;

and individualistic.

The

support

Fox and Haaaond docuaent
hierarchical,

coordinative,

Bach is defined by staff

proximity to legislators.

Attributes of legislators and staff also condition staff
access.

Legislator ability to delegate to and then aanage

professional staff is a key feature.

[23]

Further,

legis­

lators' perception of the role they play will condition the
use aade of staff.

Staff types have been repeatedly identi­

fied as explaining why staff seek access to legislators.
These types echo the characterizations of Lerner and
Meltsner reviewed earlier.

The professional or technical

staff person emphasizes performance of a task in a neutral
fashion;

access for these staff depends upon the need for

their expertise.

Entrepreneurial staff seek out legislators

in hopes of realizing certain objectives.

Policy entrepre­

neurs seek implementation of public policies they support;
political entrepreneurs seek career furtherance for both
themselves and a legislator.

106

SUMMARY

Both staff advocates and critics have based their
observations on staff growth.

Advocates called for aore

staff to aid beleaguered legislators in carrying out their
responsibilities.

Critics witnessed the rapid rise in staff

nunbers with alarm viewing the influx of these new actors in
the legislative arena as an additional burden to legislators
already hard-pressed to meet their responsibilities.

The

growth of advisors such as staff is a recurring historical
phenomenon that is a response on the part of decision makers
to the growth in the size of states,
interactions across polities,
political structures.

and the complexity of

All these factors have been operative

in post-World War II America,
legislative staff.

the level of

contributing to the growth in

Increases in the number of staff were to

be expected and are not indicative of staff's success or
failure in responding to legislator's needs.

The impact of

staff depends on how staff have been used.

The most direct and meaningful

indicator of staff activity

is the level of access staff have to legislators.

To either

positively or negatively impact legislator responsibility
requires staff to have access to legislators.

The

importance of access to the advisory function has been
stressed by various authors and is a common observation
found in the descriptive literature on legislative staff.

107

The various dimensions of staff access
frequency, directness,
defined.

genesis,

In this research,

(fora,

duration,

and exclusivity)

can be

they will be quantified,

used in an analysis of legislator responsibility,

and

and the

accountability it supports.

FORMAL ARGUMIMT
To empirically examine staff access to ascertain staff
effect on legislator accountability requires development of
a formal argument from which hypotheses can be derived and
tested.

The essential question being researched is whether

staff have helped or hindered legislator accountability.
Legislator accountability takes many forms based on various
expectations of legislator performance.

All these

expectations of accountability are said to rest on the
presumption that legislators are responsible for their
official acts.

To determine if staff negate this

presumption of responsbility (and,
legislator accountability)

therefore,

undermine

certain assertions about key

features of legislator-staff relations are put forward for
examination.

The first feature of lagislator-staff relations relevant to
this inquiry is control.

A strong indicator that

legislators maintain their accountability in the presence of
staff is legislator control of staff.
is that legislators control staff.

The first assertion

The second feature of

108

legislator-staff relations is the extent of discretion
available to staff.

Bven if staff are found to be

controlled by legislators,

opportunities for staff

discretion are likely to occur.

How such discretion occurs

is less important than its impact on responsbility and the
accountability it supports.
that staff discretion exists,

In this regard it is asserted
but is 1) tolerable and 2)

manageable by legislators.

To see if these assertions are credible,
hypotheses are outlined.

a series of

If the hypotheses prove correct,

the three assertions will be supported which,

in turn,

reinforces the presumption of legislator responsibility upon
which legislator accountability rests.
false,

Should they prove

credence would be given to concerns that staff nay

not further legislator accountability because they undermine
the assignment of responsibility for legislative acts which
accountability requires.

Legislators Control Staff
The first set of hypotheses regarding the assertion of
control are based upon our earlier review of the devices for
control Weingast suggests are available to legislators.
assertion and its associated hypotheses can be stated as
follows:

The

109

ASSERTION 1.

Controls exist to assure that staff are

responsive to legislators.

HYPOTHBSIS 1.
staff,

If legislators control a responsive

then legislators will be the aain source of

decisions to hire and retain staff.

HYPOTHESIS 2.
staff,

If legislators control a responsive

then staff attitudes will reflect a service

orientation.

HYPOTHBSIS 3.
staff,

If legislators control a responsive

then staff attitudes will reflect a lack of

staff control.

HYPOTHESIS 4.
staff,

If legislators control a responsive

those staff having aore direct working

relations with legislators will have greater
levels of access to legislators.

Hypotheses 1 through 3 entail descriptive analyses of the
legislator-staff eaployaent relationship,

as well as staff

attitudes towards their work and their relative influence.
Hypothesis 4 is the first direct

inquiry involving staff

access and the Legislative Enterprises it defines.

By

Joining legislators and staff according to access patterns
and quantifying the levels of access,

these levels can be

statistically conpared to see if significant differences are
observed between staff with direct and indirect working

110
relationships with legislators.

Several dimensions of

direct or indirect legislator-staff working relations will
be examined including:

reporting responsibility (staff

reporting directly to legislators versus those reporting
through an intermediary);

personal

identification (staff

identifying personally with a specific legislator versus
those without such identification);

and, partisanship (staff

who routinely participate in partisan politics as part of
their job versus those who do not).

Staff Discretion is Tolerable

The second assertion and its associated hypotheses begins to
address staff discretion.

Discretion is an expected part of

the legislative-staff relation.

The initial question about

discretion is not whether it exists,
tolerable.
interests,

but is its existence

If we assume legislators act in their best
staff discretion can be said to be tolerable if

legislators use staff in a way that indicates staff benefit
outweighs staff cost

(including the costs of staff

discret io n).

Legislators in positions of authority can use that authority
to acquire more staff; because of their authority these
legislators are also more likely to be sought out by staff.
Should these legislators take advantage of their ability to
acquire staff assistance,

then their view of staff will be

characterized as tolerating the costs associated with staff

Ill

discretion.

Should they not capitalize on their

opportunities to use staff,
is due,

at least in part,

it aay be that this reluctance

to the consequences of staff

discretion being intolerable.

This arguaent can be stated

as follows:

ASSERTION 2.

Legislator behavior will be consistent

with the point of view that the benefits

legislators

derive froa staff outweigh the cost of staff
discretion.

HYPOTHBSIS 5.
best interests,

Assuaing legislators act in their
if the benefits of staff outweigh

the cost of staff discretion,

then the nuaber of

staff allocated to legislators will relate to
existing sources of legislative authority.

HYPOTHESIS 6.
best interests,

Assuaing legislators act in their
if the benefits of staff

discretion outweigh the cost of staff discretion,
then the aaount of access legislators perait staff
will relate to existing sources of legislative
authority.

Hypothesis 6 eaploys the nuaber of staff foraally assigned
to a legislator as the basis for defining legislator-staff
relations.

Hypothesis 6 utilizes the Legislative Enterprise

defined by legislator-staff access as the basis for its

112

measurement.

The hypotheses are identical

in their logic,

but these differing aeasureaent techniques will permit a
comparison between results from the traditional approach of
counting staff as compared to findings derived from an
examination of the Legislative Enterprise as defined by
access.

With both hypotheses those legislators in positions

with authority will be statistically compared to legislators
who lack or have lesser authority to ascertain whether
levels of staff use are significantly different across these
various groups.

Sources of legislator authority that will be

examined include those belonging to the majority party,
legislators holding at least one committee chair,

and

legislators arrayed by tenure with authority presumably
increasing as tenure increases.

Authority was not defined by formal legislative offices for
two reasons.
Michigan)

First of all, most legislators

(at least in

hold some title either as partisan leaders,

operatives,

floor

or caucus members making any comparison group

too small for useful statistical analysis.

Secondly,

no

reliable method could be found for differentiating the
relative authority associated with various titles (e.g.,
does a Majority Floor Whip have more or less authority than
the Caucus Chair;

or,

do all Whips have equal authority).

While analysis of authority in these terms proved
problematic,

the expressions of authority included in the

analysis should be adequate for assesing Assertion 2.

113

Staff Discretion is Manageable
Assertion 3 continues with the examination of staff
discretion.

Given the existence of discretion,

exercise subject to legislator influence.
answers this query affirmatively.

is its

Assertion 3

The two hypotheses

associated with this assertion each posit an instance of
staff discretion and the countervailing expression of staff
management.

Both are based upon legislator enterprises

defined by legislator-staff access.

ASSERTION 3.
but,

Opportunities for staff discretion exist;

are tempered by legislator influence.

HYPOTHBSIS 7.

Both legislators and staff initiate

their contacts reflecting some degree of staff
discretion;

but, more access will be attributable

to contacts

initiated by legislators.

HYPOTHBSIS 8.

Staff discretion is evidenced in

personal work orientation;

but,

these varying

orientations will be unrelated to levels of staff
access.

Hypothesis 7 recognizes that one form of staff discretion is
staff ability to initiate contacts with legislators.

One

management response available to legislators to temper this
form of staff discretion is limiting the duration of staff

114
initiated interactions and devoting lore tine to those that
the legislator initiaties.

Statistical coaparisons of tiae

devoted to staff-versus-legislator initiated contacts can
ascertain whether this aanageaent technique is utilized.

Hypothesis 8 derives froa the body of literature that
identifies various personal aotivations underlying advisory
and staff behavior.
of staff discretion.

[12]

These aotivations represent foras

Three types are identified:

technicians who want to perfora a specific job;

cliabers who

want to use their present job as a aeans for professional
advanceaent;

and entrepreneurs who seek to take advantage of

their job situation to influence public policy.

While staff

aeabers will prioritize these roles differently,

such

discretion will not influence legislator availability or
access.

Legislators will

interact with staff in a aanner

unrelated to personal staff priorities.

SUMMARY
Barlier discussions of legislator accountability are focused
on several key concepts which are then structured into a
formal argument amenable to empirical analysis.
staff access and the Legislative Enterprises

Legislator-

it defines

provides a focus for the examination of staff's impact on
accountability.

Maintenance of legislator accountability is premised on
legislators being responsible for their actions.

115

Legislators are responsible for their enterprises if they
exercise control over a responsive staff. Even with
legislative control,
likely to occur.

none degree of staff discretion is

Such discretion has U n i t s

and need not

negate legislators being responsible for their actions and
the actions of their enterprises provided the discretion is
tolerable (i.e.,

its costs do not outweigh the benefits of

staff use) and subject to legislative influence.

A series of three assertions
being tolerable,
put forth.

(one for control,

discretion

and discretion being nanageable)

are then

Hypotheses related to each assertion will be used

as the basis for supporting or refuting the assertions and
the observations regarding responsbility and the legislator
accountability they support.

The next chapter gives these hypotheses enpirical expression
that can be statistically tested within the context of the
Michigan State Legislature.

CHAPTBR 3:

References

1
Goldhaner,

Herbert,

The A d v i s e r . Elsevier,

New York,

1978.

2

Ibi^.
3
Lerner, Alan W . , The Politics of Decision-Making:
Strategy. Cooperation, and C o n f l i c t , Sage Publications,
Beverly Hills, 1976.
4
Meltsner, Arnold J., Policy Analysts in the
Bureauc ra cy. University of California at Los Angeles,

1976.

5
Merton, Robert K . , Social Theory and Social St r u c t u r e .
Free Press, Glencoe, 111., 1957, p. 215.

6
Lerner, Op. C i t .. pp.

134-136.

7
Meltsner, Op. C i t .. p. 6.

8
Ib id .. p. 4.
9
Ib i d . . p. 6.

10
Hartnark, Leif, "The Role of the Legislative Budget
Staff in the Budgetary Process of Wisconsin", The Political
P u rse strings. ed. A.P. Balutis and D.K. Butler, Sage
Publications, Beverly Hills, 1975, p. 87.

11
Fox, Harrison W. and H a m o n d , Susan W . , Congressional
Staffs?
The Invisible Force in American L a w a k i n g . Free
Press, New York, 1967.

12
Farnuw, Bugene, "The Legislative Fiscal Staff's Role in
the Budgetary Process in Michigan", The Political
P u rse strings. Op. C i t . . p. 87.

116

ll7

13
BeVier, Michael J., Politics Backstage:
Inaide the
California L egislature. Teaple University Press,
Philadelphia, 1979, p. 175.

14
Malbin, Michael J., Unelected Representatives:
Congressional Staff and the Future of Representative
G o v e r n a e n t . Basic Books, New York, 1980, p. 21.
15
Hattey, Lowell and Hofheiaer, Susan, "The Legislators'
Source of Bxpert Inforaation", Public Opinion Quart erl y.
Fall 1954, pp. 300-303.
16
Patterson, Saauel C., "The Professional Staffs of
Congressional Co aa ittees” Adainistrative Science Q u a rte rl y.
Vol. 15, No. 1, March 1970, pp. 22-37.
17
Zwier, Robert, "The Search for Inforaation:
Specialists and Non-Specialists in the U.S. House of
Representatives", Legislative Studies Qu a r t e r l y . Vol.
No. 1, February 1979, pp. 31-42.

IV,

18
Bradley, Robert B., "Motivations in Legislative
Inforaation Use", Legislative Studies Q u a r t e r l y . Vol.
3, August 1980, pp. 393-406.

V, No.

19
Patterson, O p . C i t .. p. 29.

20
Meller, Noraan, "Legislative Staff Services:
Toxin,
Specific, or Placebo for Legislature’s Ills", The Western
Political Q u a r t e r l y . Vol. XX, No. 2, Part 1, June 1967.
21
Saxe, Leonard and Koretz, Daniel, "Bditor's N o t e ” ,
Making Evaluation Research Useful to C o n g r e s s . Evaluation
Research Society, 1982.

22
Fox and Haaaond,

Op. Cit.

23
BeVier, Op. C i t . . p. 92.

CHAPTER 4

EMPIRICAL RESULTS

Data on the Michigan Legislature will be used to empirically
test the hypotheses laid out in Chapter 3.

The sources of

these data include the Survey of Legislative Staff in
Michigan,

1984,

conducted as part of this research and

supplementary data from secondary sources.

These data,

described in Chapter 2, provide the basis for assessing
staff impact on legislator accountability in one legislative
setting.

To test the hypotheses and assess accountability
necessitates the development of a measurement technique for
gauging legislator-staff contact.

By using summary

descriptions provided by staff of the frequency and duration
of their meetings with legislators,

an estimated amount of

interaction time (expressed as hours per month)

that

legislators devote to staff and staff spend with legislators
is derived.

The calculation of this measure is explained in

the data analysis plan that follows.

118

119

The value of interaction tine as the aeasure of access and
the quantifier of the Legislative Enterprise is shown in the
analysis of data derived from the hypothesis testing.

The

first three hypotheses use staff descriptions of attitudes
to assess control;

the regaining five use one of three

levels of analysis the measure of interaction time affords.
How this measure of time complements more traditional
methods of legislator-staff linkage are shown in a direct
comparison of statistical results derived from the two
measures when Hypotheses 5 and 6 are analysed.

DATA ANALYSIS PLAN

Measuring Interaction Time

The preceding discussions of access have stressed its
importance to accountability and its utility in defining the
Legislative Enterprise.

Interaction is proposed here as the

empirical expression of access and the quantifier of the
Legislative Enterprise.
and calculations
time measures.
in Michigan,

Table 18 details the conversions

involved in the development of interaction
As part of the Survey of Legislative Staff

1984,

staff were asked to identify by name

those legislators with whom they had work-related
discussions more than once per month in the preceding three
months.

For each legislator staff named they characterized

the frequency and duration of their interactions with

Table 18
Calculation of L e g islator-Staff Inte r a c t i o n Times

Aspect of Interaction

Original Survey Que s t i o n

Respondent-Selected
Ordinal Scores

Frequency

For each legislator listed,
indicate the overall
frequency of your
discussions.

5
4
3
2
1

=
=
=
=
=

D uration

For each legislator listed,
indicate the usual length
of time your discussions
lasted.

5
4
3
2
1

=
=
=
=
=

Daily
More than once p/w.
Once per w e e k
More than once p/m.
Once per month

Over two hours
One to two hours
One-half to one hour
Fift e e n to thirty min.
Under fifteen minutes

M i n i m u m legislator-staff interaction
(1 day per month) x (10 minutes)
or .17 hours per month.

T e mporal C onversion

30
8
4
3
1

days per m o n t h
days per m o n t h
days per m o n t h
days per m o n t h
day per m o n t h

120 minutes
90 minutes
45 minutes
22 minutes
10 minutes

= 10 minutes per m o n t h

M a x i m u m legislator-staff interaction
(30 days per month) x (120 minutes)
or 60 hours per month.

= 3600 minutes per m o n t h

121

legislators by checking one of five autually-exclusive
ordinal scores.

Frequency of staff interaction with

legislators was scored according to days per aonth or week;
duration scores took the fora of ainutes/hours devoted to
the "typical" interaction.

Teaporal expressions for

frequency and duration were devised.

Frequency descriptions

were standardized to the approxiaate nuaber of days per
aonth a particular staff person and a particular legislator
had a work-related discussion.

The usual length of these

interactions are expressed in ainutes.
frequency tines duration,

By aultiplying

an estiaate is derived for the

nuaber of ainutes per aonth consumed by each
legislator-staff pair.

Division by sixty converts the

ainutes into hours per aonth,
interaction tine.
ainutes

the final eapirical aeasure of

The ainiaua interaction tiae is ten

(or .17 hours) per aonth;

ten ainutes per nesting.

one day per aonth tines

The aultiplication of thirty days

per aonth by 120 ainutes per meeting results in 3,600
ainutes (or 60 hours) per aonth devoted to interactions by a
legislator and a staff advisor as the aaxiaua amount of
interaction tiae.

The distributions of the frequency and duration measures as
well as the interaction tines froa their aultiplication are
shown in Table 19.

Frequency of aeetings shows anple

distribution across all five points while 84.3% of all
duration responses are in the two lowest points
or ten ainutes per interaction).

(22 ainutes

The coabination of these

122

Table 19
Frequency Distri b u t i o n s for the Compon e n t s of
Legislator-Staff I n t e r a c t i o n s and Re s u l t i n g
Interaction Times f r o m the Survey of
Legislative Staff in Michigan, 1984

F requency

Duration

#

%

#

%

30 days per month

35

11.2

120 minu t e s

3

1.0

8 days per month

72

23.0

90 minu t e s

16

5.1

4 days per month

57

18.2

45 minu t e s

30

9.6

3 days per month

98

31.3

22 minu t e s

87

27.8

1 day per month

51

16.3

10 minutes

177

56.5

313

100.0

313

100.0

Total

Legislator-■Staff Inte r a c t i o n T imes

(in Hours per Month)
Time

#

%

3.00

4

1.3

2.6

4.50

4

1.3

74

23.6

5.00

5

1.6

.67

34

10.9

6.00

11

3.5

.75

3

1.0

11.00

13

4.2

1.10

16

5.1

12.00

6

1.9

1.30

24

7.7

22.50

8

2.6

1.47

19

6.1

45.00

6

1.9

2.25

4

1.3

60.00

3

1.0

2.93

31

9.9
Total

313

100.0

Time

#

%

.17

40

12.8

.37

8

.50

123

two temporal score* into interaction tiae* result* in 50.9X
of the total 313 interaction* between legislators and staff
consuming under one hour per aonth.

At the initial level of the 313 individual interactions
reported by sixty staff with 118 legislators,

certain

features of the Legislative Enterprise are revealed.
obviously,

Most

they are results of selective interactions.

Bach

of the 60 staff respondents were able to report up to a
total of ten legislators with whoa they interacted on at
least a monthly basis.
legislators,

If each respondent had named ten

600 interactions would have been identified;

each had named but one,
identified.

if

only sixty would have been

The fact that just over half the potential

number of interactions were identified indicate that staff
have ready access to more than one legislator but that some
staff have access to a broader spectrum of legislators
either by their own selection or the selection of
legislators.

Table 20 displays the distribution of the sixty staff and
legislators according to the actual number of reported
legislator-staff interactions with which they were
associated.

Overall,

each staff member interacted with an

average of 5.22 legislators although the largest proportion
of staff are at the two extremes of contact levels (28.3%
reported one contact,

21.7% named nine legislators).

Legislators,

tend to concentrate at the low end of

however,

124

Table 20
Survey of L egisl a t i v e Staff in Michigan, 1984:
Number of Legi s l a t o r C o ntacts R e p o r t e d by
Staff and N u m b e r of Staff Contacts
Associ a t e d w i t h Legi s l a t o r s

N umber of Contacts

R e p o r t e d by Staff
(with Legislators)

#

%

In v o l v i n g L e g islators
(with Staff)

#

%

1

17

28.30

43

36.44

2

3

5.00

29

24.58

3

2

3.33

18

15.25

4

2

3.33

16

13.56

5

8

13.33

2

1.69

6

2

3.33

3

2.54

7

3

5.00

2

1.69

8

7

11.67

1

.84

9

13

21.67

-

-

10

3

5.00

4

3.38

Total

60

100.00

118

100.00

N u mber of Contacts

5.22

2.65

125

contact frequency;

61* were Mentioned by only one or two

staff people which is close to the average for each
legislator of 2.65 reported staff interactions.

When the 313 distinct interaction tines are sunned for each
legislator and each staff person the interaction tines
arrayed in Table 19 and the distribution of legislator and
staff contact reflected in Table 20 conbine to forn two
further levels of interaction tines described in Table 21;
each legislator's interaction tine with all sixty staff and
each staff n e n b e r *a total interaction tine with the 118
legislators identified.
than staff,

Because there are nore legislators

interaction tines are typically shorter for

legislators than for staff.

The sunnary statistics reported

at the botton of Table 21 support this observation as does
the percent distribution of legislators by their interaction
tines when conpared to the distribution of staff.

Linking Interaction Tine Levels to Hypotheses

The ability to identify three levels of analysis fron a
neasure of interaction tine is one of the innediate benefits
associated with this approach to quantifying access and the
Legislative Enterprises they define.

Bach of the three

levels is used to test at least one of the eight hypotheses
put forward for testing.

126

Table 21
Survey of L egisl a t i v e Staff in Michigan, 1984:
Estimated Total Time per M o n t h Spent
I nteracting w i t h Staff (for Legislators)
and w i t h L e g i s l a t o r s (for Staff)

Overall
I n teraction Time

Less than 1 Hour

Legislat o r s w i t h Staff
#
%

Staff w i t h Legislators
#
%

34

28.8

4

6.7

1 - 2

Hours

18

15.3

6

10.0

2.1-3

Hours

9

7.6

5

8.3

3.1-4

Hours

3

2.5

1

1.7

4.1 - 5 Hours

5

4.2

5

8.3

5.1 - 6 Hours

5

4.2

2

3.3

6.1 - 7 Hours

5

4.2

1

1.7

7.1-8

Hours

3

2.5

1

1.7

8.1-9

Hours

0

-

1

1.7

9.1 -

10 Hours

2

1.7

10.1 -

15 Hours

11

9.3

8

13.3

15.1 -

20 Hours

5

4.2

9

15.0

20.1 -

30 Hours

5

4.2

5

8.3

30.1 -

40 Hours

3

2.5

4

6.7

40.1 -

50 Hours

5

4.2

3

5.0

50.1 -

60 Hours

2

1.7

1

1.7

60.1 -

70 H o u r s

2

1.7

1

1.7

70.1 -

80 Hours

1

0.9

0

-

3

5.0

118

100.0

60

100.0

Over 80 Hours
Total

0

0

127

Table 21 —

Continued

Summary Statistics
Minimum

10 m i n u t e s / m o n t h

20 m i n u t e s / m o n t h

Maximum

77 h o u r s / m o n t h

163 h o u r s / m o n t h

Mean

10 h o u r s / m o n t h

20 h o u r s / m o n t h

3 hours/month

15 h o u r s / m o n t h

Median

128

Level 1, the 313 individual legislator-staff interactions,
permits testing of Hypothesis 7 which coaparea interaction
tiaes according to who initiates the interactions.

Level 2,

the accuaulated interaction tiaes for each legislator is
used to defined Legislative Enterprises froa the
legislator's point-of-view and assess the relationship of
legislator authority to the tiae devoted to interactions
with staff (Hypothesis 6).

Level 3 quantifies the

Legislative Enterprise froa s t a f f s ’ perspective by suaaing
interaction tiaes for each staff person;

this serves as the

basis for testing whether staff working relationships with
legislators

(Hypothesis 4) and staff work priorities

(Hypothesis 8) are related to levels of interaction tiae
staff spend with legislators.

Use of these measures,
1 through 3 and 5,
chapter.

as well as other data for Hypotheses

is pursued in the remainder of this

With half the hypotheses dependent on the measures

of interaction tiaes,

the utility of this measurement

approach to the quantification of the Legislative Enterprise
and the assessment of legislator accountability is evident.

Because each of the levels of interaction tiaes have
decidedly non-noraal distributions,

assumptions underlying

the use of paraaetric statistics for hypothesis testing aay
be violated.
T-Teat,

The parametric statistics used (the Two-Sample

regression,

and ANOVA F-Test)

for observing

differences between means for various groups are rather

129

robust and nay be sufficient for the analysis.

However,

to

fully assess the data and the hypotheses they test,
non-paraaetric tests

(Mann-Whitney U, Kruskall-Wallis,

and

the Median Test) which use the distribution of interaction
tines as their criteria,

are also employed to augment the

■ore traditional parametric tests.

HYPOTHBSIS TBSTING

Legislator Control

In Chapter 1, four devices available to legislators to
ensure their control over staff were identified based on
W e i n g a s t ’s observations regarding legislator relations with
executive agency bureaucrats.

Hypotheses

1 through 4 each

express one of these fonss of control.

Hypothesis 1 states that one fora of legislator control is
their being the main source of decisions concerning staff
hiring and retention.

Data in Table 22 tend to support the

existence of this fora of control.

A majority of staff

(65.OX) report in response to survey item IV.C that
legislators are the aost iaportant influence on eaployaent
decisions affecting staff in their positions.

Iaportant

exceptions to this overall pattern are observed when
legislator involveaent in staff hiring and retention is
categorized by staff superiors

(itea III.A.3) and staff

130

Table 22
Survey of Legisl a t i v e Staff
Legisl a t o r Inv olvement in
Re t e ntion C a t e g o r i z e d by
and Staff Ide n t i f i c a t i o n

in Michigan, 1984:
Staff H i r i n g and
Staff Superiors
with Legislators

L e g i s l a t o r I n v olvement in Staff H i ring
and R e t e n t i o n
Staff Superior

L egislator

Other Staff

L egislator Mo s t
Important

O ther M o r e
Important

L e g i s l a t o r Not
Important

Total

#

28

2

2

32

%

71.8

28.6

14.3

53.3

#

11

5

12

28

%

28.2

71.4

85.7

46.7

Staff Identification
with Legislators
Identified w i t h
Specific
Legislator

#

27

2

2

31

%

69.2

28.6

14.3

51.7

Identified w i t h
Group of
Legislators

#

10

0

9

19

%

52.6

-

47.4

31.7

Not Identified
w i t h Any
Legislator

//

2

5

3

10

%

20.0

50.0

30.0

16.7

#

39

7

14

60

% 100.0

100.0

100.0

100.0

Total

131
identification with legislators (itea IV.D).

Those staff

with non-legislator superiors are sore likely to be reaoved
froa direct legislator influence on their eaployaent.

Staff

identifying with groups of legislators as opposed to a
specific legislator have a higher proportion of their
nuabers with no reported legislator involveaent

in their

hiring and retention.

The second feature of control highlights staff dependence on
legislators for their influence.
of dependence,

If staff are in positions

their attitudes should reflect a service

orientation to their work.

A series of Likert-type

questions were posed to staff concerning their level of
agreeaent/disagreeaent with several stateaents about staff
activity (section 3 of the survey).
shown in Table 23.
service orientation.

Results obtained are

Staff attitudes are consistent with a
When given stateaents reflecting a

lack of service orientation (i.e.,

staff hinders

legislator

deliberation and staff are able to act on their own) over
80k of the respondents disagreed with sizable ainorities
noting strong dis ag ree ae nt.
positive questions,

In response to aore neutral or

staff attitudes reflected support for

providing useful inforaation but not necessarily new ideas.
They agreed by quite a aajority (73.4k) that staff case work
helps legislators be aore responsive to their constituents.

Table 23
Survey of Le gislative Staff in Michigan, 1984:
Staff Attitudes
Reflecting a Service O r i entation

Staff hinders
legislator deliberation

Staff able to
act on own

Casework helps
constituent and
legislators
Staff should provide
useful information

Disagree

No
Opinion

Agree

Strongly
Agree

Missing

#

25

27

3

1

2

2

%

41.7

45.0

5.0

1.7

3.3

3.3

#

12

37

3

5

1

2

%

20.0

61.7

5.0

8.3

1.7

3.3

6

2

#

0

28

11

13

Z

—

46.7

18.3

21.7

10.0

3.3

#

2

5

3

33

14

3

%

3.3

8.3

5.0

55.0

23.3

5.0

#

2

4

7

34

10

3

%

3.3

6.7

56.7

16.7

5.0

11.7

Total
60
100.0
60
100.0
60
100.0
60
100.0
60
100.0

132

Staff should
generate n e w ideas

Strongly
Disagree

133
Hypothesis 3 speaks to legislator control by examining lack
of staff control.

As only one participant in a crowded

legislative arena,

staff ability to influence legislators

has liuits.

When asked to coapare staff influence to that

of other parties involved in the legislative process
VI.H) staff lack of control coaes into clear view.
presents these attitudes on relative influence.

voters,

and lobbyists,

in that order.

Table 24

Those

perceived by staff respondents as aore influential
legislative staff include other legislators,

(itea

than

the governor,

Opinions of relative

influence were split on the press and interest groups.
Staff of both party organizations,

executive agency

personnel and the governor's staff were perceived as less
influential than legislative staff.

The first use of the interaction tiae aeasures discussed
previously in this chapter is found with Hypothesis 4.
Weingast's observations that legislators will favor advisors
w ith whoa they are in accord is said to find expression in
the aaount of tine staff spend with legislators being
related to their working relationship with legislators.
Those staff with a direct legislator working relationship
(itea III.A.3), those identifying with a specific
legislator or legislators

(itea IV.D),

and those for whoa

partisan activity is an expected part of their job (itea
IV.A) were used to define close legislator-staff working
relationships indicative of accord between the two parties.
Weak support for Hypothesis 4 is found.

Table 25 presents

Table 24
Survey of Legislative Staff in Michigan, 1984:
Staff Attitudes
Comparing Staff Influence to Influence of Other
Legislative Participants

Influence Compared to Staff
Much
Less

Other Legislators

1

Less

5
1.7%

Voters

3

9

1

12

1

Democratic Party
Staff

13

Republican Party
Staff

15

Bureaucracy

18

6

12

13

20
13.3%

1

15

1

14
25.0%

60
100.0%

5.0%

60
100.0%

6.7%

60
100.0%

13.3%

60
100.0%

5.0%

60
100.0%

3.3%

60
100.0%

3
1.7%

1
33.3%

5.0%

8
1.7%

25.0%

60
100.0%

4
3.3%

8.3%

21.7%
15

33.3%

2

5

3.3%

3
8.3%

21.7%

20.0%

36.7%

5

13

60
100.0%

3
11.7%

30.0%

16.7%

31.7%
22

8

18

10

19
25.0%

7

2
1.7%

Total

5.0%
2

18.3%

35.0

30.0%

30.0%

10.0%
Governor's Staff

18

3

11

21

No
Opinion

25.0%

38.3%

26.7%

25.0%

21.7%

23

16

15

15
41.7%

20.0%

20.0%

1.7%

25

12

Much
More

More

18.3%

15.0%

1.7%
Interest Groups

11
8.3%

5.0%
Lobbyists

Same

Table 24 —

Mu c h
Less

Comparison Groups

Press

8

Less

17
13.3%

Governor

2

Same

11
28.3%

7
3.3%

Continued

18
18.3%

11
11.7%

4
30.0%

27
18.3%

Much
More

More

2
6.7%

11
45.0%

No
Opi n i o n

60
3.3%

2
18.3%

Total

100.0%
60

3.3%

100.0%

136

Table 25
Survey of L e gisla t i v e Staff in Michigan, 1984:
C o m p a r i s o n of M e a n Staff Time D e v o t e d to
Legis l a t o r - S t a f f C ontacts C a t e g o r i z e d by
W o rki n g R e l a t i o n s h i p

Attributes of
Legisla tor-Staff
Working R e l a t i o n s h i p

To w h o m Staff
Report
- Le gislator
- Staff

b. W i t h w h o m Staff
Identify
- Specific
L e gisl a t o r
- Group of
Le gislators
- No specific
Legisl a t o r

Measure s R e l a t e d to M e a n s
N

Mean

Variance

■ Test-Statistics

Std. Dev.

60

T = 1.38
25.05
14.76

741.06
943.01

31

24.45

793.31

28.17

F = 1.31

19

20.15

1,302.50

36.09

Sig. = 28

10

7.41

63.74

7.98

32
28

6 F = 58
Sig. = .17

60

c. P a r tisan A c t i v i t y
Expe cted of
60
Staff
- Yes
29
- No
31

T = 1.43
25.75
15.10

1,572.30
185.09

4 F = 58
Sig. = .

137

the parametric results obtained froi comparisons of group
means formed by the three variables defining legislatorstaff working relationships.

Group means are in the

expected direction with those staff having more direct
reporting relationships,

identification,

or partisan

involvement with legislators being those staff with higher
mean interaction times.

However,

statistically significant.

the differences are not

When the distributions of the

interaction times for these staff groups are compared using
non-parametric statistics significant differences are
observed (Table 26).

Staff reporting to legislators are

more likely to have interaction times above the median than
staff reporting to other staff.

Differences in the

distribution of staff interaction times according to staff
identification with legislators approaches significance at
the .11 level.

Based on the results of Hypotheses 1 through 4, the forms of
legislator control identified by Weingast are found to be
operative in the legislator-staff relationships observed in
Michigan.

138

Table 26
Survey of Legisl a t i v e Staff in Michigan, 1984:
Compar i s o n of D i s t r i b u t i o n s for Staff Time
Devoted to L e g i s l a t o r - S t a f f Contacts
C ategorize d by Attrib u t e s of the
L e g i s l a t o r - S t a f f Work i n g
R e l a t ionship

Attribute of
L e g i s l a tor-Staff
W o rking R e l ationship

N

a. To w h o m Staff
Report
- Legislator
- Staff

32
28

b. W i t h w h o m Staff
Identify

60

- Specific
Legislator
- Group of
L egislators
- No Specific
Legis l a t o r
c. P a r t i s a n A c t ivity
E x p ected of Staff
- Yes
- No

Avg.
Rank

Distributions
N=
N<.
N>

60
36.00
24.21

8
18

19
10

Median

11.00

Mann-Whitney
U = 272.00
Sig. = .01
Sig. of M
Test = .00

16.00

KruskallWallis = 4 . 3 9
Sig. = 11

5

0

21

34.08

10

17

4

19
10

29.74
20.85

9
7

9
3

1
0

60
29
31

30.74
30.27

14
12

11
18

4
1

Test
Statistics

Sig. of
M Test =

11.00

.11

Mann-Whitney
U = 442.50
Sig. = .92
Sig. of M
Test = .31

139
Discretion Being Tolerable

The existence of none degree of staff discretion is taken
as a given.

The question is whether such discretion is

both tolerable and nanageable.
whether discretion is tolerable;

Hypotheses 5 and 6 exaaine
hypotheses 7 and 8 assess

evidence of aanageaent response to opportunities for staff
discretion.

Hypotheses 5 and 6 take two conpleaentary view of
legislator-staff relations.

Hypothesis 6 uses a technique

of linking legislators with staff by foraal assignment.
This approach reviews

legislative staff rosters (in this

instance the Michigan Legislative Telephone Directory) to
identify those staff assigned to a legislator's office
either as personal or coaaittee staff.

A count of the

number of staff assigned to each legislator is derived.
This is the aore traditional approach used by various staff
researchers to aeasure legislator-staff relations.

[1]

Hypothesis 6 eaploys legislator interaction tines as the
dependent variable.

Both hypotheses array their respective

aeasures according to various attributes of legislator
authority.

The logic behind both these hypotheses is as follows.
assuae legislators act in their best interest,

If we

those in

authority positions will use their authority to acquire aore

140

staff resources

(either in teras of numbers of staff or

aaount of interaction tiae) only if staff discretion is
tolerable (i.e.,

the benefits of staff use outweight the

cost of staff use such as discretion).

For both hypotheses,

the expressions of legislator authority used in the analysis
include aajority/ainority standing,

coaaittee position held

(defined as Coaaittee Chair, Co-Chair, Minority Chair,

or

Appropriations Subcoaaittee Chair) and tenure.

Results for Hypothesis 5 using the nuaber of staff assigned
to legislators are suaaarised in Tables 27 through 29.
Table 27 presents paraaetric results for the first two
attributes of legislator authority with Table 28 displaying
the coaparable non-paraaetric measures.

In both tables

findings are consistently significant only for Senators.
This pattern is attributed to the dependent aeasure used
(nuaber of staff assigned to legislators).

Chapter 2

contrasted the centralized nature of staffing in the House
with the decentralized nature of Senate staffing.
centralized staff structure, aany individual

In a

legislators,

such as aeabers of the Michigan House, will not have
specifically identified staff foraally assigned to them.
This eliminates many representatives from the analysis
dropping the N to 35 or less in the House for both tables.
The Senate,

however, with its decentralized assignment of

staff to specific legislators shows a pattern of staff

141

Table 27
Compar i s o n of M e a n Numb e r of Staff A s s i g n e d to
Legis l a t o r s Categ o r i z e d by Various
Attributes
of Legis l a t o r A u t h ority

Attributes of
Legislator
Authority

Measures Rela t e d to Means
N

Mean

Variance

a. Majori t y / M i n o r i t y
Standing
- House M a j o r i t y
- House M i n o r i t y

33
2

1.56
2.00

1.06
2.00

- Senate M a j o r i t y
- Senate M i n o r i t y

15
17

3.40
2.35

1.83
.62

- Total M a j ority
- Total M i n o r i t y

48
19

2.15
2.32

2.00
.67

Test

Std. Dev.

T = .56
«fF = 33
Sig. = .58

- .72
T =
<Jf = 30
01
Sig.
T =

Committee Posit i o n
- House Chair
27
Co-Chair
1
M i n o r .Ch.
0
Appropr.
1

1.70
1.00
-

1.00

1.22
-

.49

cf F = 65
Sig.

b.

Statistics

1.10
-

-

-

-

-

- Senate Chair
Co-Chair
Min.Ch.
Appropr.

9
1
11
5

3.78
2.00
2.27
3.00

2.19
-

1.48
-

.62
1.00

.79
1.00

- Total Chair
Co-Chair
Min.Ch.
Appropr.

36
2
11
6

2.22
1.50
2.77
2.67

2.33
.50
.62
1.47

1.49
.71
.79
1.21

63

F = .38
Sig. = .69

F = 3.20
Sig. = .04

F = .41
Sig. = 7 5

142

Table 28
Co m p a r i s o n of Dist r i b u t i o n for N u mber of Staff
As s i g n e d to Legislators C a t egorized by Various
Attributes of Legisl a t o r A u t h ority

A t t r ibute of
Legislator
Authority

A. M a j o r i t y / M i n o r i t y
Status
-House M a j o r i t y
-House M i n o r i t y

N

Avg.
Ra n k

Di stributions
N <
N>
N=

Median

1.00
33
2

17.76
22.00

0
0

12
1

21
1

-Senate M a j o r i t y
-Senate M i n o r i t y

15
17

20.43
13.03

0
1

10
5

5
11

-Total M a j o r i t y
-Total M i n o r i t y

48
19

31.81
39.53

21
2

13
6

14
11

b. Committee P o s ition
- House Chair
Co-Chair
M i n o r i t y Chair
Appropriations

27
1
1

15.44
9.00
9.00

0
0
0

12
0
0

15
1
1

-Senate Chair
Co-Chair
M i n o r i t y Chair
Appropriations

9
1
11
5

18.06
7.50
9.77
14.70

2
1
8
2

7
0
3
3

0
0
0
0

-Total Chair
Co-Chair
M i n o r i t y Chair
Appropriations

36
2
11
6

26.35
19.25
31.50
34.42

15
1
1
1

10
0
3
3

11
1
7
2

Test
Statistics

Mann-Whitney
U = 25.60
Sig. = 1.00
Sig. of M
Test = 1 . 0 0

2.00
Mann-Whitney
U = 68.50
Sig. = .02
Sig. of M
Test = .53
2.00
Mann-Whitney
U = 351.00
Sig. = 12
Sig. of M
Test = .01
1.00

-

KruskallW a l l i s = 1.07
Sig. = .59
Sig. of M
Test = 1.00
2.50
KruskallW a l l i s = 6.54
Sig. = .09
Sig. of M
Test = .11
2.00
KruskallWallis = 2 . 4 7
Sig. = .48
Sig. of M •
Test = .17

143

Table 29
Results of Bivarlate R e g r e s s i o n s for Mem b e r s
of House and Senate R e g r e s s i n g N u mber
of Staff A s s i g n e d to L e g islators
by L e g islators Tenure

Variable

Partial

Coef f i c i e n t

Std.

Error

T

Sig.

House
Constant
Tenure

.25

1.28

.28

4.63

.00

.04

.03

1.45

.15

F
2.12

Senate

T
Constant
Tenure

.29

Sig.

R2

.15

.06

Sig.

2.48

.30

8.18

.00

.07

.04

1.64

.11

F
2.67

Sig.

R2

.11

.08

144
assignment consistent with the hypothesized relationship to
authority.

Those Senators with greater authority have more

staff.

Table 29 presents the results of the regression testing the
relationship between the number of staff assigned
legislators and the final feature of legislator authority
examined,

tenure.

No significant relationship is observed

for either the House or Senate.

Tables 30 through 32 document findings for Hypothesis 6
using interaction time as the dependent measure.

The amount

of interaction time spent on members of the majority in the
House is significantly different from minority members
interaction time for both parametric (Table 30) and
non-parametric

(Table 31) indicators.

House committee

position also shows significant differences in interaction
time across both statistical approaches.

Support for the

relationship in the Michigan House between legislator
authority and amount of interaction time is also found in
Table 32.

Here interaction time is regressed against

legislator tenure.

The constant,

regression are all significant;

coefficient,

however,

and total

the amount of

variance explained is small indicating the presence of
factors other than tenure influencing levels of legislator
interaction time.

145

Table 30
Survey of Legislat i v e Staff in Michigan, 1984:
Comparison of Me a n L e g i s l a t o r Time D e v o t e d to
Legislator-Staf f Co n t a c t s C a t e g o r i z e d by
Various Attribu t e s of L e g i s l a t o r A u t h o r i t y

M ea s u r e s R e l a t e d to Means
Me a n
Variance
Std.Dev.

Attributes of
Legislator Authority

N

a. M a j o r i t y/Minority
Standing
-House Majority
-House M i n o r i t y

50
34

12.87
2.20

302.68
9.27

15
17

11.91
17.67

306.65
461.13

-Senate Majority
-Senate M i n ority

Test
Statistics

T = 3.53

b F = 82
Sig.

T = .82

J F = 30
Sig.

-Total M a j ority
-Total Minority

65
51

12.65
7.36

278.98
207.90

Senate Chair

= .42

T = -1.76

b F = 114
Sig.

b. Committee Position
-House Chair
Co-Chair
Mi n ority Chair
Appropriations

= .00

= .08

27
14
23
1

15.88
3.93
2.47
76.76

335.86
21.51
11.98
—

18.33
4.64
3.46
—

F = 15.77
Sig. = .00

126.61

11.25

F = 2.92

9

6.62

Co-Chair
M i n o r i t y Chair
Appropriations

1
11
5

60.00
16.62
11.83

506.28
178.82

22.50
13.37

-Total Chair
Co-Chair
M i n o r i t y Chair
Appropriations

36
15
34
6

13.57
7.67
7.04
22.65

294.97
229.54
206.55
845.75

17.17
15.15
14.37
29.08

-

-

Sig. =

.06

F = 2.06
Sig. = .11

Table 31
Survey of L e g islative Staff in Michigan, 1984:
C omparison of
Distributions for L egislator Time Devo t e d to
L e g islator-Staff Contacts C a t egorized by
Various A ttributes of L egislator A u t hority

Attributes of
Legislator Authority

Avg.
Rank

50
34

52.20
28.23

18
24

32
10

0
0

- Senate Majority
- Senate Minority

15
17

15.47
17.41

8
8

7
9

0
0

- Total Majority
- Total Minority

65
51

47.28
67.30

2331

39
18

3
2

27
14
23
1

41.78
31.64
22.13
65.60

8
8
16
0

18
6
7
1

1
0
0
0

9
1
11

11.06
25.00
13.82
14.90

6
0
6

3
1
5

0
0
0

a. M a j ority/Minority Standing
- House Majority
- House Minority

N<M

D i stributions
N?M
N=M

N

Test Statistics
Median

2.84

M a n n - W h i t n e y U = 365.00
Sig. = .00
Sig. of M Test = .00

5.61
M a n n - W h i t n e y U = 112.00
Sig. = .56
Sig. of M Test = .50
2.93

b. Committee Position
- House Chair
Co-Chair
Minor i t y Chair
Appropriations
- Senate Chair
Co-Chair
M i n ority Chair
Appropriations

Ma n n - W h i t n e y U = 1085.50
Sig. = .00
Sig. of M Test = .01
2.75
K r u s k a ll-Wallis = 16.36
Sig. = .00
Sig. of M Test = .03

5.21

K r u s k a ll-Wallis = 3.37
Sig. = .34
Sig. of M Test = .29

Table 31 —

A ttributes of
Legislator Autho r i t y

N

Avg.
Ra n k

36
15
34
6

53.07
44.50
36.51
61.08

N 4.M

Continued

D i stributions
N>M
N=M

Test Statistics
Median

2.93
Total Chair
Co-Chair
Mi n ority Chair
Appropriations

13
8
21
1

22
6
12
5

1
1
1
0

K r u s k a l l - W a l l i s = 8.97
Sig. = .03
Sig. of M Test = .07

148

Table 32
Survey of L e gisla t i v e Staff in Michigan, 1984:
Results of Biva riate R e g r e s s i o n s for Mem b e r s
of the House and Senate R e g r e s s i n g Time Spent
by Legislators in L e g i s l a t o r - S t a f f Contacts
by L e g i s l a t o r Tenure

Variable

Partial

Coefficient

Std.Error

T

Sig.

House
Constant
Tenure

.23

5.20

2.19

2.37

.02

.61

.29

2.15

.03

F

Sig.

R2

.03

.05

4.64
T

Constant
Tenure

-.07

Sig.

16.45

5.18

3.17

.00

-.27

.70

-.39

.70
_.

n o

JL

Sig.

R2

.15

.70

.00

149

The ability to analyze legislator-staff relationships in the
centralized staff setting of the Michigan House is aade
possible by the quantification of legislator-staff
interaction time.

This measurement technique complements

the more traditional approach of assigning staff to
legislators that perforas well in decentralized staff
structures such as is found in the Michigan Senate.

Both

lend support to the hypotheses and reinforce the assertion
that the presence of staff discretion is tolerable.

Discretion Being Manageable

Given the existence of staff discretion and the presence of
legislator controls over staff,

the question arises as to

whether legislators attempt to influence or aanage staff
discretion.
discretion,
discretion.

If legislators do seek to aanage staff
this aay help explain the tolerability of staff
Alternatively,

staff aay teaper their use of

discretion and aake it tolerable for legislators without
necessitating legislator management intervention.

Docuaenting the presence or absence of legislator management
of staff is difficult especially when the data available
reflect only staff perceptions.

However,

some sense of the

issue is obtained by positing instances of staff discretion
that staff describe and looking for evidence of legislator

150

influence indicative of management.

That ia the tack

pursued in Hypotheses 7 and 8.

Hypothesis 7 notes as the instance of discretion staff
ability to initiate contacts with legislators.

Data

reported earlier noted the close split between the nuaber of
contacts initiated by legislators and staff, so it is known
that staff do initiate soae portion of their legislator
contacts.

Legislator exercise of aanageaent control over

their staff sessions aay be looked for in the length of tiae
devoted to their interactions with staff.
aanage staff interactions,

If legislators

the aaount of tiae devoted to

interactions the legislator initiates can be expected to be
greater than the aaount of time devoted to staff initiated
contacts.

Table 33 shows this to be the case.

The average

level of monthly interaction between a legislator and a
staff person is 2.5 hours longer if a legislator initiates
the interactions than if they are usually initiated by
staff.

The non-paraaetric comparison of distributions is

less conclusive but not contradictory to this finding.

A second manifestation of staff discretion is the differing
orientations and priorities they associate with their staff
work.

These orientations were initially defined in a manner

consistent with an historic triad of advisory roles
discussed in Chapter 1.

Their prioritization was

incorporated into the survey questioning (item V).

The

association of a specific legislative management response to

151

Table 33
S urvey of Legisl a t i v e Staff in Michigan, 1984:
C o m p a r i s o n of M e a n Times and Di s t r i b u t i o n s for
Time D e v o t e d to L e g i s l a t o r - S t a f f Contacts
Categ o r i z e d by Initiator of Contact

M e a s u r e s Related to M eans
N

Mean

Variance

Legislator

131

5.56

135.60

Staff

154

3.06

48.11

Initiator

Std.Dev.

N

Avg.
Ra n k

Statistics

T = 2.24
</F = 283
Sig. = .03

Di stributions
Initiator

Test

Test

N

N

N=

M e d i a n Statistics

1.10
Legislator

131

146.03

58

62

11

Staff

154

140.42

77

73

4

MannWhitney U :
9689.50
Sig. = .56
Sig. of M

152

each staff priority is not possible since the staff
priorities are an artifact of this research.

However,

the

presence of a countervailing legislative aanageaent response
aay be inferred if it is found that staff priorities bear no
relation to the aaount of interaction tiae consuaed by
staff.

This is the thrust of Hypothesis 8 the results of

which are reported in Tables 34 and 35.

The lack of

significance found for both paraaetric and non-paraaetric
tests supports the hypothesis and lends credence to the
notion that staff relationships with legislators aust
consider aore than staff priorities and orientations.

Suaaary

All three assertions put forward in Chapter 3 are supported
in the eapirical analysis reported here.

Legislators

exercise control over staff in a variety of foras.
with this control,
however,

Even

soae staff discretion is likely to occur;

levels of staff discretion are both tolerable and

subject to legislator aanageaent.

These suaaary observations are derived froa paraaetric and
non-paraaetric tests for a series of eight hypotheses.

In

addition to direct descriptive and attitudinal questions
posed to staff,

a aeasure of legislator-staff interaction

tiae was developed and used to quantify Legislative
Enterprises froa both the legislator and staff perspective.

153

Table 34
Survey of L egisla t i v e Staff in Michigan, 1984:
Comparison of M e a n Staff Time Dev o t e d to
Legislator-S t a f f C o ntacts C a t e g o r i z e d
by Staff Pe r s o n a l W o r k O r i e n tations

A ttrib u t e s of
Staff P e r sonal W o r k
Orientations by
P riority

Measures R e l a t e d to Means
N

Mean

V a riance

Test

Std.Dev.

S tatistics

-Climber
E n trepreneur

30

18.33

798.51

28.26

F = 1.01

-Entrepreneur
Climber

13

24.72

770.47

27.76

Sig.

-Entrepreneur
Technician

10

35.25

2170.40

46.59

-Technician
Climber

7

10.50

176.88

13.30

-Technician
E n trepreneur

5

10.72

88.97

9.43

-Climber
Technician

5

9.41

88.96

9.43

= .42

154

Table 35
Survey of Legisl a t i v e Staff in Michigan, 1984:
Comparison of Di s t r i b u t i o n s for Staff Time
Devoted to L e g i s l a t o r - S t a f f Contacts
Ca tegor i z e d by Staff Personal
W o r k Priorities

Attrib u t e s of Staff
Personal W o r k Orientatations by Priority

N

Avg.
Rank

Distrib u t i o n s
N<
N>
N=

Median

Test
Statistics

11.00

_ Climber,
Entrep r e n e u r

20

28.05

9

9

2

- E n trepreneur
Climber

13

33.69

5

7

1

- E n trepreneur
Technician

10

40.35

2

8

0

- Technician,
Climber

7

25.14

5

1

1

- Technician,
E ntrep r e n e u r

5

25.90

2

2

1

- Climber
Te c h n i c i a n

5

24.40

3

2

0

Kr u s k a l l W a l l i s = 5 . 62
Sig.= .34
Sig. of M
Test = .40

155

Legislator interaction tiaee were found to be a
superior method for gauging staff use in the centralised
staff structure of the House when compared to the
traditional approach of linking legislators to staff by
formal assignment.

The traditional approach did prove of

use in the setting of the Senate with its decentralised
staffing patterns.

CHAPTBR 4:

References

1
E.g. Salisbury, Robert H. and Shepsle, Kenneth A.
"Congressional Staff Turnover and the Ties-that-Bind",
Aaerican Political Science R e v i e w , vol. 75, no. 2, June
1981; or, Shiff, Steven H. and Saith, Steven S.,
"Generational Change and the Allocation of Staff in the U.S.
Congress", Legislative Studies Q u a r t e r l y , vol. VIII, no. 3,
August 1983, pp. 457-467.

156

CHAPTBH 5
LEGISLATOR AND STAFF OBSERVATIONS
The empirical findings reviewed in the preceding chapter are
buttressed by personal observations aade by both Michigan's
legislators and nenbers of their staff.

Legislator consents

were obtained through a series of seni-structured interviews
with the five legislators nost frequently nentioned by staff
in the Survey of Legislative Staff in Michigan,

1984.

Staff

renarks were voluntarily offered by nost of the 74 staff
respondents who conpleted an open-ended request

for

additional observations that concluded the Survey of
Legislative Staff in Michigan,

1984.

Consents fron both groups are sunnarized here to lend
credence to the argunent explored in this research that
legislators control staff and that staff discretion is both
tolerable as well as nanageable.

In the process,

the

priorities that legislators and staff associate with their
interactions cone in to view and point out new directions
and issues in need of exploration.

Legislator Connents
Each of the five legislators contacted was a willing
participant in the interview process and gave generously of
their tine.
one hour.

The interviews lasted between 45 ninutes and
To each of the legislators

157

(Representatives

158

Robert Baerson,

Lynn Johndahl,

David Hollister,

Stabenow as well as Senator Lana Pollack)

and Deborah

I extend my thanks

and appreciation for their cooperation.

Although the five legislators contacted had little
foreknowledge as to the content of the interviews,

it was

evident that all had given staff functioning and their
relationships with staff considerable thought on their own.
None of the five interviewees were surprised at their
selection.
question:

This was evident in their responses to the first
"Based on ay survey of legislative staff,

you are

one of the top five aost frequently contacted legislators
aentioned by staff;

does that surprise you?"

Their reactions to their reported levels of staff use
revealed a conscious effort to utilize staff resources.
Specific coaaents spoke to the iaport of staff recruitaent
("I take staff recruiting very seriously.
need to know everything,

Legislators don't

but they aust be able to use staff

resources at their d isposal.” ); an openness to staff contact
("I know aany staff and have aaintained those relationships;
therefore,

I tend to use thea.

I know what they can do and

find it easy to interact with thea.";

"I know aost staff so

I see thea and talk to thea and call on them for
background.");

and the connection between being an active

legislator and using staff ("Legislator-staff relations are
a two-sided exchange.

I use staff as auch as possible.

159

Staff who are interested in substance look for interested
and hard-working legislators.")

When asked to assess Michigan's current legislative staff
coapleaent,

coapliaents to the coapetency of staff were the

doainant theae in all five interviews.

I have a good deal of respect for staff.
They have a
deeper and broader knowledge in their area of
specialization than legislators on their own.
I look
to staff for inforaation and background but not
political judgeaents.
I depend on staff expertise and the skills of people I
view as capable and credible.
I try to get the aaxiaua
effort froa people and find the liaits of their
ability.
Staff are an iaportant resource for inforaation and
inforaation is power for a legislator.
I use staff I have confidence in and rely on staff that
I think are coapetent.
Overall, staff are excellent and coapetent thanks
largely to the efforts of Speakers starting with Bill
Ryan.

The only serious aisgiving about current staffing patterns
expressed by this group of legislators was the need for sore
staff generally:
effectiveness.";

"If I had two aore staff I could double ay
or for aore staff with specific skills "I

could use access to aore legal staff.”

"I think each

legislator should get one aore staff person for constituent
work.”

160

Most of the time devoted to these interviews was spent
discussing the fornal argument explored in this research.
Each of the five legislators were first asked to react
generally to the potential inpact staff nay have on
legislator accountability.

In each interview the legislator

was asked to indicate whether the following statenent had
any relevance to Michigan's legislature.

"Sone observers of

the federal Congress feel that the growth of staff in
Washington has conplicated the federal

legislator's job.

They say that staff have becone too independent and nay be
eroding legislative accountability."
statenent applied to Michigan,

None felt this

although several noted that

it nay be true for Congress given the different legislative
dynanica at the federal level.

Moving fron hypothetical conparisons to observations on the
Michigan legislative environnent,

each interview included

questions on legislator ability to control staff and
assessnents as to whether staff discretion was tolerable and
nanageable.

Regarding legislator control of staff, the views of the five
legislators can be sunnarised with the following
observation.

"Staff are only out of control if legislators

want then to be."

Others elaborated this thene with

connents such as the following.

161

Use of staff is as diverse as legislators.
Soae hold
staff captive; others use staff better.
For example,
the goal of a coanittee Chair sets the tone for the
staff role.
Those who want no conflict set a different
role for staff than those who don't aind conflict at
nestings; but, in these different roles, staff are
responding to legislators.
Coanittee staff reflect the discretion of the Chair.
Personal staff reflect the legislator's predilections.
Partisan staff are not used to their potential because
soae legislators are lazy and use then poorly.

All five legislators acknowledged the existence of staff
discretion,

but such discretion was seen as tolerable.

Staff do know aore on particular subjects but the
discretion this involves is controlled by prior
negotiations between legislators and staff based on
trust.
Discretion has U n i t s set in a variety of ways such as
technical skills, personal rapport, and the history of
the working relationship.
Soae staff aake decisions,
but a ainority.
My staff work on ay agenda and within
ay limits.
Staff do not have great discretion or excessive power.
If staff withold inforaation or give it out
arbitrarily, then staff night have too auch power; but
I don't see that.
The level of discretion is set sonewhat by the Speaker
and who he hires to run the staff agencies.
But, there
are linits to discretion.
Staff know how far they can
go.
Central staff go only as far as the Chair or
legislator wants thea to go.
Discretion is aolded by
how auch the legislator wants to be involved.

To gauge the extent of legislator involvement in controlling
staff discretion,

each of the five legislators was asked

whether they and their peers were able to manage staff
activities.

As a coaaon reference point, management was

162

characterized as a superior-subordinate relationship in
which the subordinate
(legislators) wanted.

(staff) did what the superior
The different nanageaent styles they

use to describe their relationships with staff point to the
nany forns legislator nanageaent of staff Bay take.

Yes, 1 aanage staff in the sense of getting thea to do
what I want; but thirty to forty percent of the tiae
staff coae to ae with ideas.
M y aanageaent is not a
superior-subordinate relation but one of resource
sharing.
Yes, I aanage staff, especially pool (caucus) staff;
but, with ay office staff the personal relationship is
aore iaportant.
There is a superior-subordinate quality to
legislator-staff relations.
Staff reflect legislators.
A weak legislator equals weak staff.
1 aanage ay office staff; with other staff I act aore
as a aeaber of a board of directors.
For exaaple, the
head of the fiscal agency is the president of a
corporation, the legislator is a aeaber of the board.

One surprising observation volunteered by several of the
legislators interviewed was the perception that aost
legislators Bake little use of staff beyond aaintenance of
office correspondence and adainistration.

Soae even went so

far as to hazard an eatiaate of the percentage of
legislators that Bake extensive use of staff.
fell in the fifteen to twenty percent range.

Bstiaates
Table 36 shows

that these educated guesses nay not be too far off the aark.
When legislators are arrayed by the total aaount of
interaction tiae staff attributed to thea, we find that the
top 16* (or 18 legislators)

accounted for 65* of the

163

Table 36
Survey of L egis l a t i v e Staff in Michigan, 1984:
C o m p a r i s o n of the Perc e n t of T otal I n t eraction
Time and Percent of Total Legi s l a t o r - S t a f f
Contacts R e p o r t e d by Staff A t t r i b u t a b l e to
V a r ious Groups of Legis l a t o r s A r r a y e d by
T otal Time D e v o t e d to L e g i s l a t o r - S t a f f
Contacts

Groups of Legisl a t o r s
A r r ange by Total
I nteraction Time

Percent of T otal
Inte r a c t i o n Time

Percent of Total
Contacts

Top 16%

65

27

Next

16%

21

23

Next 16%

9

21

Next

16%

4

14

Next

16%

1

9

Next

16%

^ 1

6

x64

estimated interaction tiae and over one-quarter of all
reported legislator-staff contacts.

The five legislators

interviewed represented 4* of all legislators

identified by

staff, but 14* of reported interaction tiae and 17* of
reported contacts.

This heavy concentration of staff use

aaong a ainority of legislators reinforces the theae
expressed throughout the reaarks garnered in these five
interviews.

Legislator use of staff varies depending on a

legislator’s willingness and ability to use staff resources.
A similar theae is aaong the points aade by staff in their
personal observations.

Staff Coaaents

The last page of the Survey of Legislative Staff in
Michigan,

1984, offered staff the opportunity to aake

whatever coaaents they wished.
option.

Most took advantage of this

Their reaarks covered three subjects:

period of the survey,
staff roles,

1) the tiae

2) the influence of legislators on

and 3) the aaount of tiae staff spend on

non-legislator activity.

Regarding the survey's tiaing,

the period iaaediately

following the recall of two senators is atypical.
one perspective,

that is advantageous.

In addition to

noraal deaands of the legislative calendar,
also one awaiting the executive budget.
coaplicated matters in the Senate,

But, froa

this period was

Recall concerns

but probably heightened

165

legislators'

electoral and constituent concerns.

coincidence of legislative,
activity nay be atypical,
of staff,

budgetary,

The

and electoral

but for reviewing legislator use

it was an ideal tiae.

The iaportance of legislator influence on staff functioning
was stressed by respondents in their coaaents.
such as the following were typical:

Observations

"The role of

legislative staff is entirely dependent on legislators."
"Your duties vary depending on what office you work in and
who you work for.

A staff aide in one office aay have a

coapletely different set of responsibilities than a staff
aide in another office."

"The usefullnesa of staff depends

on the individual legislator's ability and willingness to
delegate responsibility and to trust their staff."
Recognising the key role legislators play in the use of
staff is central to this research.

As noted at the outset,

Maintenance of legislator accountability is ultimately the
responsibility of legislators theaselves.

Several staff emphasized that they devote aost of their
energies to dealings with non-legislators
executive agency staff,
representatives,

(e.g.

other staff,

lobbyists and interest group

constituents,

true that auch (if not aost)

etc.).

It is undoubtedly

of staff tiae is not spent with

legislators.

In fact,

survey results support this

observation.

However,

the intent of this research is not to

166

assess staff activity generally;
staff activity,

but to analyse one facet of

their interactions with legislators and its

iaplications for legislator accountability.

In pursuit of this end, a nuaber of insights were gained
into the analysis of legislative staff that nay prove of use
in future research.
chapter.

These are reviewed in the concluding

CHAPTER 6
INSIGHTS AND IMPLICATIONS

This research centers on a single question.

Have

legislative staff helped or hindered legislator
accountability to the public?

The views of staff proponents

and critics reviewed earlier reveal the dimensions of the
debate about legislative s t a f f ’s effect on legislator
accountability.

Proponents praise staff as the best

response for assisting legislators in their increasingly
complex duties.

Critics warned of an independent staff

bureaucracy that would exacerbate rather than relieve the
complexity of the modern legislator's role.

The resolution to this conflict is surprisingly simple.

It

lies in the recognition that not all legislators use staff
equally.

Those legislators making extensive use of staff do

so consciously and in a manner that leaves them in control
while limiting staff discretion to tolerable levels.

At the

same time,

there are legislators whose priorities,

interests,

and information sources lead them to make little

or no use of staff.

To oversimplify,

it would appear that Michigan's legislators

have headed of the advice Machiavelli gave in The P r i n c e :

167

168

The Prince ought always to take counsel, but only when
he wiBhes, not when others wish; on the contrary, he
ought to discourage absolutely attempts to advise his
unless he asks it . . . It is an infallible rule that a
prince who is not wise hiaself cannot be well advised.

Legislators,

the aodern-day "princes",

take staff counsel;

but when they wish, not as staff dictates.

Not all legislators benefit equally froa the use of staff.
The claias of staff proponents aust be tcapered by this
fact;

any positive results of staff developaent and use

accrue to a ainority of legislators.
who do aake extensive use of staff,
least in Michigan,
aaterialized.

For those legislators
it appears that, at

the fears of staff critics have not

Legislators are in control of staff and keep

staff discretion to acceptable levels through a variety of
approaches to handling the legislator-staff working
relationship.

In either instance,

staff do not looa as a

threat to the assignaent of legislator responsibility and
the public accountability it supports.

In coaing to these conclusions,

other insights were obtained

regarding legislative staff and the legislators they serve.
These insights and their aeaning, not only for this research
but subsequent inquiry,
this chapter.

are reviewed in the reaainder of

169
STAFF AS ADVISORS
The first insight gained into legislative staff is that,
despite their relatively recent
legislatures,

rise in Anerican

they are hardly a new political phenomenon.

Throughout history and across civilizations,
been an integral part of politics.
included tutor and nentor,
counsellor,

soothsayer,

advisors have

The positions held have

personal confidante,

religious

relative of various political

leader,

leaders.

professional

close friend,

and

Their bases of

advice have ranged froa personal charisma to social
standing,

from formal education to practical experience and

training.

[1]

Legislative staff are a contemporary

manifestation of this historic political constant.

The making of this historical
viewed with alarm by some,

link helps place staff growth,

in its proper context.

Advisory

growth generally and staff growth specifically is a normal
political response to growth in the size of government,
frequent occasions for governmental interactions,

more

and the

complexity of political s t r u c t u r e s .

Giving staff a history also gives them relatives.

What is

known about other advisory relationships has potential
applicability to legislative staff;

and,

conversely, what we

learn of legislative staff may be relevant to appreciating
other advisors as well.

Nowhere is the potential for this

exchange of knowledge better Illustrated than in the use of
access to define and interpret advisory functioning.

170

Access to decision aakers is the hallaark of advisory
influence.

Without access,

advisors are peripheral

participants in political discussion;

with access,

advisors

becoae key actors in the dynaaic of political decision
naking.

This research uses as its principal analytic tool

docunenting and analyzing staff access to legislators.

The

insights gained through the analysis of access have iaport
not only for legislative staff in Michigan but other staff
and advisors as well.

THB IMPORT OF ACCESS
Access is used here to join principal-agent theory with
research on Legislative Enterprises enriching both and
providing opportunities for future inquiry.

Principal-agent

theory has its roots in econoaic research seeking to explain
the behavior of firas.

This theoretical approach has been

found to have broader applicability to political
relationships between superiors and subordinates.
W e i n g a s t ’s use of principal-agent theory for exaaining
legislators relations with executive bureaucrats in
regulatory agencies is the aost fully developed exaaple of
applying the tools of principal-agent theory to political
relationships.

[2]

In this research we extend the logic of principal-agent
theorizing to the legislator-staff relationship.

Our

findings can even be restated in terns consistent with

171

principal-agent theory logic.

Legislators will delegate to

staff only when the benefits of such delegation (e.g. aore
coaplete inforaation,
political allies,
aechanisas

(i.e.

opportunities for identifying

etc.)

exceed those of alternative

doing all the work without staff) even

though such delegation iaplies soae staff discretion.
Further,

specialized institutions

eaployaent contracts,
activity,

(i.e. explicit or iaplicit

rewards related to acceptable staff

or norms that temper staff behavior) evolve to

mitigate staff discretion.

However, we do not simply borrow froa principal-agent
theory;

but also,

add to it with the introduction of access

as the aost direct expression of principal-agent
relationships in a political context.

Access provides the

■eans of identifying how principals and agents interrelate
individually and collectively.

In so doing,

it details the

extent of legislator/principal delegation to their
staff/agent and the networks through which various
institutional arrangements function to control or mitigate
staff/agent behavior.

The idea that legislators and staff weave a web of networks
through their various relationships is the basic preaise of
work on Legislative Enterprises.

[3]

However,

use of the

Legislative Enterprise as an approach for analyzing
legislator-staff relationships has been U n i t e d
theoretically and empirically.

The theoretical uses of the

172

Legislative Enterprise widen Markedly once it is recognised
that legislators sitting at the center of a network of staff
and other advisors called the Legislative Enterprise are
simply principals delegating to agents in a fashion similar
to that found in economic firas.

The eapirical handicap of

being able to define a legislator's enterprise only froa
secondary sources
directories)

(usually telephone books or staff

is reaoved with the aeasureaent technique

developed as part of this research for documenting and
quantifying access.

Measuring access and using that measure in analyzing
legislator-staff relations are the major innovations of this
research.

By documenting the main features of access

(frequency of contact,
and usual initiator)

duration of meeting,

typical setting,

legislator-staff interactions can be

converted into statistically meaningful expressions.

The worth of the measure has already been shown.
demonstrated here,

As

it reveals relationships between

legislators and staff not reflected in traditional secondary
sources and does so in a manner that is more comprehensive
and practical logistically than personal observation or
maintenance of logs on legislator-staff interactions.
Standardizing the approach to measuring access also permits
comparing the various

interactions observed.

Finally, we

see that by associating characteristics of legislators and

173

staff with their access patterns, we can begin the process
of characterising and understanding these patterns.

The measure of access also has a potential yet to be
realized.

It represents the first real opportunity for

interstate coaparisons of staff development and use.
Further it can be extended to other legislative agents
Weingast'a regulatory bureaucrats)
legislator access is key (e.g.

(e.g.

or to others for whoa

lobbyists).

Interactions

between others In the legislative arena (e.g. staff-tostaff or lobbyist-to-staff) can be quantified in a similar
fashion yielding the potential for reconstructing fairly
elaborate forms of Legislative Enterprises linked by access
points.

Finally,

the measure can be applied to other

political relationships outside legislatures

(e.g.

relationships within executive agencies).

There are,

however,

limits to this approach that must be

recognized and, where possible,

overcome.

Improvements to

the actual measurement of access itself are possible.

The

current approach is somewhat simplistic since it gives
respondents only one choice among several mutually exclusive
categories describing their interactions.

One desirable

refinement could be to apportion instances of access across
various response categories.

There is also a need to better

characterize what occurs during access.
document overall

Currently,

we can

instances of access, but insights into the

174
import of what occurs is indirect and inferred froa the
magnitude of tine devoted to access.
is acceptable;

but,

As a first step,

this

a conclusive examination of legislator

control and staff discretion or other features of the
workings of the Legislative Enterprise cannot rely solely on
indirect observation.

Ways of attaching value to what

transpires during the interactions associated with access
need to be devised.

This might occur as a refinement to the

survey measurement approach used here;

or,

the survey could

be used as a first step in identifying interactions and
relationships for more detailed observation using other
techniques.

Both the inherent potential and opportunities for
improvement associated with the analysis of access argue
strongly for its further use.
here.

A beginning has been made

Further development will require the interest and

involvement of legislative observers and legislators alike.

CHAPTER 6:

B«f«r«ncM

1
Goldhaaer,

Herbert,

The Advis er . Elsevier,

New York,

1978.
2
Weingast, Barry R. "A Principal-Agent Perspective on
Congressional-Bureaucratic Relations," Fifth Carneigie
Conference on Political Econoay, Carneigie-Nellon
University, Pittsburgh, June 1983.
3
Salisbury, Robert H. and Shepsle, Kenneth A. "U.S.
Congressmen and Enterprise" Legislative Studies Q u a r t e r l y ,
vol. VI, no. 4, Novesber 1981.

175

176

APPENDIX A:

Survey of Legislative Staff in Michigan,

Instrument

1984

Michigan State University
Department of Political Science

To:

Staff of the Michigan

House and Senate

From:

Si Nahra

Subject:

Survey of Legislative Staff

177

As a staff member of the Michigan Legislature, you are invited to par­
ticipate in a first-of-its-kind study of legislative staff. Much has been
written about legislative staff; but, until now, a general survey of a leg­
islature's staff members to find out about their work and attitudes has
never been done.
There are several reasons why I've undertaken this survey. It's most
immediate use will be as part of my doctoral dissertation in political
science at Michigan State University. Further, as a former staff member in
two state legislatures (Michigan and Virginia) I had the opportunity to ob­
serve the workings of staff and gain an appreciation for and continuing in­
terest in the kind of work you do. Finally, and perhaps of most importance,
staff need to be heard. Each year more is written about the role of staff
and its implications for how legislatures work. It is important to add the
staff perspective to these other views. This survey is one way of doing
that.

Please complete the survey and return your response in the enclosed envelope
by February 1, 1984. Send the survey to:
Si Nahra
Department of Political Science
S. Kedzie Hall
M.S.U.
East Lansing, Michigan 48824
Should you have any questions, I can be reached between 9:00 and 5:00 in
Detroit at (313) 963-4990.
Anyone completing this survey is entitled to an executive summary of the re­
sults. If you would like to receive this summary, please write your name and
address in the space provided at the end of the survey.
Thank you for your time and cooperation.

Your participation is entirely voluntary. No one in the legislature
or State government will see your answers or require your response. Use of
the data collected will be limited to my doctoral research. I value your
participation and will treat your answers as confidential.
Each survey has a code number on the last page that enables me to iden­
tify who has and has not responded. I will keep a single master list of
who was mailed which survey. Once data collection is complete (in February
or March, 1984), that list and the last page of all returned surveys will
be destroyed thereby assuring your anonymity.
A group of your fellow staff members already completed the survey. It
took them about 30-45 minutes. The survey is designed so that it need not
be completed all at once. Each section of the survey can be answered inde­
pendently, if that is more convenient for you.

Sincerely,

S -? A / a ^ y \ ^
Si Nahra

INSTRUCTIONS FOR PART I

in this section of the survey you are asked to summarize your work-related discussions with legislators from September 1983 to December 1983 by answering
several questions. To make sure your responses are comparable to other staff members, please read and observe the following definitions and directions.
DEFINITION;

A work-related discussion is a direct conversation between you and a legislator related to the carrying out of your staff
responsibilities. Please do not includes
(1) discussions occurring during committee hearings; (2) personal conversations
not related to work; or (3) contacts with legislators that are social or recreational in nature.

DIRECTION FOR PART A

DIRECTION FOR PART B

DIRECTION FOR PART C

DIRECTION FOR PART D

DIRECTION FOR PART E

DIRECTION FOR PART F

In the first column on
the next page write the
names of those legisla­
tors with whom you had
work-related discussions
at least once per month
from September 1983 to
December 1983.

For each legislator
you listed in Part A,
indicate the overall
frequency of your dis­
cussions from September
1983 to December 1983.

For each legislator
you listed in Part A,
indicate the usual
length of time your
discussions lasted.

For each legislator
you listed in Part A,
indicate who usually
initiated these dis­
cussions. While each
of the three parties
presented as options
might have initiated
some of these contacts,
check the one who did
so most often.

For each legislator
you listed in Part A,
indicate whether these
were more commonly pri­
vate talks or group
discussions.

For each legislator you
listed in Part A, des­
cribe the subjects cover­
ed in your discussions.
Place a 1 for the most
frequent topic discussed,
a 2 for the next most
frequent, a 3 for the
next, and so on. Topics
not covered, leave blank.
Describe and number other
topics in the last column

If you had such discus­
sions with more than ten
legislators during this
time, select and list
only the top ten with
whom you had the most
frequent contact.

It is likely that the
frequency of your con­
tacts varied during
this three month
period. Try to "aver­
age" or summarize these
experiences by select­
ing the one response
that best approximates
your level of contact
with each legislator
appearing in the first
column.

Again, it is likely
that you had conversa­
tions of varying
length. Select the one
response that best typ­
ifies the most common
occurrence.

I.

In this section of the survey you are asked about your work-related discussions with legislators from September 1983 to December 1983.
Please read the definition and directions on the preceding page before filling out this page.
I I IF YOU HAD NO WORK-RELATED DISCUSSIONS WITH LEGISLATORS AS DEFINED ON THE PRECEDING PAGE, CHECK THE BOX AT THE
LEFT AND GO TO THE NEXT PAGE

■£> /

&

A ’A t

<o j /■<?

*Jx XXX X X X X XX- &
O

o^3''%y

-t?

Other:

Please Specify

179

dames of legislators with
whom you had work-related
discussions

II.

In this section of the survey you are asked about the major written products you prepared for legislators or others as part of your legislative
staff duties from September 1983 to December 1983. Such products may include bills, speeches, reports, analyses, letters, memoranda or other
communications.
| |

IF YOU PREPARED NO WORK-RELATED WRITTEN COMMUNICATIONS, CHECK THE BOX AT THE LEFT AND GO TO THE NEXT PAGE.

A.
Please describe the most common types
of written products you produced as
part of your legislative duties

1.
2.

3.

4.

5.

6.

7.

8.

9.

B.
For each item describe
in Part A, estimate
the number of original
communications of this
type you prepared from
9/83 to 12/83

C.
Take the total number from Part B and estimat.e how many of t.his total belon g in the
following categories.
Legislators requested
primarily for their
direct use

Legislators requested
primarily for use by
others

Persons other
than legisla­
tors request­
ed

You initiated
work for use
primarily by
legislators

You initiated
work for use
primarily by
non-legislators

III.

This page of the survey concerns information on you as a professional. Please provide
the information requested in the first column as it applies to your present position
and in Column u tue work you performed prior to this.

A.
In this column please provide the
information requested as it applies
to your
Descriptive information requested

1.

Job
Title

2.

Time in position
(to nearest month)

3.

Title of your
immediate supervisor

4.

Number of persons
reporting to you

5.

Name of office
agency, or organization

present position

B.
In this column please provide the
information requested as it applies
to the job you held
prior to your present position

■

IV.

The four questions on this page provide further background on your present position

A.

C.

Is participation in partisan electoral
activities an expected part of your
job?
(check the most appropriate response)

Which of the following statements best
characterizes legislator influence
concerning the hiring and firing of
persons in your position?
(check the most appropriate response)

Yes

single, most important influence

No

important, but other non-legis­
lators are more important
not involved

D.
Which of the following statements comes
closest to stating your decree of in­
volvement with legislators?
(check the most appropriate response)

I am not identified with any
particular legislator or group
of legislators.
I tend to be identified with a
particular group of legislators.

B.
Where is the principle location of the
legislators (or legislator) for whom
you work?
(check the most appropriate response)
Entire House
House Democrats
House Republicans
Entire Senate
Senate Democrats
Senate Republicans
Both House and Senate

I tend to be identified with a
specific legislator.

V.

In this section of the survey are three possible views of legislative staff work. You are asked to indicate the one you
identify with most closely and the one that is least like your feelings towards your legislative staff work

Place an "x" in each of the two col umns
Indicate here the statement you
identify with the most

Statement 1:
I am a specialist with a particular set of skills and expertise that I use in
my legislative employment. If I did not work for the legislature, I would be
performing the same functions for another organization.

183

Statement 2:
My legislative employment is an important part of my overall career develop­
ment. The experience I gain as a legislative staff member is going to help
me advance professionally either within or outside the legislature.

Statement 3:
The opportunity to influence and shape public policy is what matters to me.
Participating in the legislature's activities enables me to address issues
and ideas of importance to the State.

Indicate here the statement you
identify with the least

VI.

On the following two pages are a series of items concerning your opinions.

(For each statement, check appropriate response)

Statements to Consider
A. In general, legislative staff are able to act on their
own without regard to reactions from legislators

B. in order to talk to their legislator, constituents
find they must first go through staff

C. The time demands staff place on legislators deprives
legislators the opportunity to deliberate with their
colleagues

B. The most important staff functions is to provide leg­
islators with information they would use if they had
the time to find it

E. Most legislative staff react after the fact to exec­
utive agency policies and initiatives

F > The main job of a good staff person is to generate
new ideas for legislative debate

Staff casework helps legislators be responsive to
their constituents without getting into details

Strongly
Disagree

Disagree

No
Opinion

Agree

Strongly
Agree

H.

This question has you compare the overall ability of all legislative staff to influence
legislators with other legislative participants. For each of the comparison groups, pick
the statement that best compares that group's influence to legislative staff influence:

Statements
Comparison Groups

Much More
Influence

More
Influence

About
the Same

Less
Influence

How do you view the work done by legislative staff?
(check the one best response)

Much Less
Influence

of great value
a practical necessity
limited usefulness
irrelevant

Other State legislators

Voters
J.

185

Lobbyists

of great value
a practical necessity
limited usefulness
irrelevant

Special interest groups

State Democratic Party Staff

State Republican Party Staff

Executive Agency Personnel

Governor's Staff

Press

Governor

How do you think Michigan's state Senators view the
work done by legislative staff?
(check the one best response)

K.

How do you think Michigan's State Representatives
view the work done by legislative staff?
(check the one best response)
of great value
a practical necessity
limited usefulness
irrelevant

VIII.

in conclusion, please provide the following background information.

A.

How old cure you now?

186

Age

B.
Please indicate your personal political
identification:
Democrat
Republican
Independent
Libertarian
Other (please describe:

D.

What was the highest grade you completed
in school?

Which category contains your approximate
salary before taxes in 1983

(check the applicable response)

(check the closest approximation)

Grade school or less

Under $15,000

Some high school

$15,000 - $19,999

High school graduate

$20,000 - $24,999

Some college

$25,000 - $29,999

College graduate

$30,000 - $34,999

Some graduate school

$35,000 - $39,999

Graduate degree (Please
specify highest ________

Over $40,000

IX.

Closing Notes

If you have additional thoughts, comments or observations you would like to pass along, I would be most interested in reading what you have to say:

If you would like a copy of the summary of this survey please write your name and address below.
ing the survey:
Name ______________________________
Address ___________________________
City ______________________________

It will be provided free of charge for those complet­

188

APPBNDIX B:

Survey of Legislative Staff in Michigan,

Codebooks

1984

189

FOR: CONTACT
D I C T I OMARV L OCAT I ON:
DATE: JUN
7, 1934

X T U K : CONTACT#

T H I S DATASET CAN BE DESTROYED1.
T H I S DATASET CAN BE REPLRCEDl
USE COUNT:
99
DATA S E T ' D E S C R I P T I O N :
L E G I S L A T O R - S T A F F CONTACTS

N= 3 1 3

ABBR

XTUK/06-06-84

UALUE

DES C RI PT I ON

F <»>

FIELD

F< 1 )

ST A F F . ID

SID

REQUIRED

. STAFF NUMBER
*•
i

F<2)

LEG. ID

LID

REQUIRED

LEGI S LAT OR

F <3>

FREQUENCY
CATEGORIES
DAILY
M T . 1 . UK
ONCE.UK
M T . 1 . MTH
ONCE;NTH

FREQ

F <4>

NAME

DURATION
CATEGORIES
L T . 1 5 . MIN

15.30
30 . 6 0
1 . 2 . HOURS
MT.2.HR
F< 5 >
.

F<6)

INITIATOR
CATEGORIES
STAFF
LEGI S
OTHER

DALY
MTt U
1 . UK MT I M
1 .MH

FREQUENCY OF L E G - ST A F F
5
4
3
2
I

OURA
LT 1 5
1530
30 60
1.2H
MT2H

1
2
3
4
5

LESS THAN 15 MI NUTES
15 TO 30 MI NUTES
30 TO 60 MI NUTES
I TO 2 HOURS
MORE THAN 2 HOURS
IN ITIATOR

1
2
3

NAT

OF CONTACT

STAFF
LEGI SLATOR
OTHER

NATURE
CATEGORIES
PRI UAT E
GROUP

PRI U
GRP

F <7>

DRAFTI NG

DRTG

DRAFTI NG

F <8 )

LEG. BCKGRD

LBGD

LEG I SLAT IUE. BACKGROUND

F <9 )

BUD. BCKGRD

BBGD

BUDGET.BACKGROUND

F<I 0

LEG . ST R A T

LSTR

LEG I SLAT I UE . ST R A T E G Y

F (IU

POL. DEU

POLD

POLI CY. DEUELOPMENT

NATURE OF CONTACT
1
2

CONTAC

D A I L Y CONTACT
MORE THAN ONCE PER UEEK
ONCE PER UEEK
MORE THAN ONCE PER MONTH
ONCE PER MONTH
DURATI ON OF CONTACT

I NIT
STF
LEG
OTH

IDENTIFIER

. P RI UA T E MEETING
GROUP DI SCUSSI ON

190

F<#>

FIELD

F< I 2 )

EXEC. OVER

EXOV

E X E C U T I VE . OV E R S I GH T

F ( 13>

COMM.ACT

COMM

COMMITTEE.ACTIVITIES

F< 1 H)

HOME'. D I S T

HOME

HOME;DISTRICT

F< 1 5 )

CO N S T I T. RE Q

CREQ

CONSTI TUENT. REQUESTS

F < 16 )

I NT . GROUP

INTG

I NTEREST. GROUPS

F< 17 )

SPEECHES

SPCH

SPEECH1. WR'I TI NG

F < 18 >

OFFICE.ADMIN

ADMN

OFFICE.ADMINISTRATION

F< 1 9 )

ADDITIONAL

ADO

OTHER. TASKS. NAMED

F <2 0>

PREDE6IGNATI0N
CATEGORIES
LOR
LO
LR
OR
L
0
R
NONE

DES

CODER'. ASS IGNMENT. O F . TASKS

NAME

ABBR

LOR
LO
LR
OR
L
0
R
NONE

VALUE

I
2
3

H
5
6

7
3

D E S C R I PT I ON

LOR
LO .
LR*
OR
L
0
R
NONE

/
/

FOR: STAFF
D I C T I ON A R Y LO CA T I ON :
DATE: JIJN 1 3 , 1984

191
X T WK : S T A F F *

T H I S DATASET CAN BE OESTROVED.
T H I S DATASET CAN BE REPLACED'.
USE COUNT:
99
DATA SET DES C R I PT I ON :
STAFF SURUEY RESPONSES AND WRI TTEN

PRODUCTS REPORTED N»74

FCO

FIELD

UALUE

F<n

STAFF.ID

SI D

F(2>

NO/ WRI T. PROO

WRI T

F<3)

PERC.LEG.LEG

L.L

PERCENT REQUESTED BY LEG FOR LEG

F<4>

PE RC. L E G. OT H

L.O

PERCENT REQUESTED BY LEG FOR OTHEF

F C5 )

PERC. OTH. OTH

0.0

PERCENT REQUESTED BY OTHER FOR
OTHER

F ( S>

PERC.STF.LE8

S.L

PERCENT REQUESTED

F<7>

P E R C ; S T F . OTH

S. O

PERCENT
OTHER

F< 8 >

TENURE

TEN V

TENURE

F<9)

SUPERIOR
CATEGORIES
LEGI SLATOR
STAFF
OTHER

SUP

STAFF

LEG
3 TFF
OTHR

F (1 0 )

SUBORDINATES

SUB

NUMBER OF SUBORDINATES

F< 1 1 )

WORKPLACE
CATEGORIES
REP. OFF
SEN. OFF
H.REP.STFF
H.DEM.STFF
S.REP.STFF
S.OEN.STFF
HFA
LSB
CONSUN.COUN
LEG.RETIRE
SEN. ANAL
ENROLL; OFF
HSE. ANAL

WKPL

PLACE

NAME

ABBR

R . OF
S. OF
HRSF
HDSF
SRSF
SDSF
HFA
LSB
COCO
LGRE
SAS
ENRL
HAS

D E S C R I PT I ON

STAFF
.

XTWKX06-07-84

IDENTIFICATION

NUMBER

NUMBER OF WRITTEN PRODUCTS

1
2
3

1
•?
3
4
5
6

7
3
Q
10
11
12
13

BY STAFF

REQUESTED BY STAFF

IN MONTHS
SUPERIOR

L EGI SLAT OR
STAFF
OTHER

OF WORK

REPRESENTATI VES OFFI CE
SENATORS OFFI CE
HOUSE REPUBLI CAN STAFF
HOUSE DEMOCRATIC STAFF
SENATE REPUBLI CAN STAFF
SENATE DEMOCRATIC STAFF
HOUSE F I S C A L AGENCY
L E G I S L A T I V E SERVI CE BUREAU
CONSUMERS COUNCIL
L E G I S L A T I V E RETI REMENT
SENATE A N AL Y SI S SECTI ON
ENROLLI NG OFFI CE
HOUSE AN A L Y S I S SECTI ON

FOP LEUFOR

192
F<#>

FIELD

F (I2 )

PRI OR. POS
CATEGORIES
OTH. LEG. MI
O T H . G O U . MI
OTH.GOU
HON.GOU
EDUCATION
MI SSING

F<13)

F<14>

F<I 5 )

F<16)

NAME

ELECTI ON. WRK
CATEGORIES
VES
NO
LEGIS.LOCUS
CATEGORIES
HOUSE
HSE.DEM
HSE. REP
SENATE
SEN.DEM
SEN. REP
HSE. SEN

F < I 3)

3
4
5

9
1
2

LOCA

1
2
3
4
5
6

7

■

1
2
3

I

LEG

3

ORIN

1
3
4
5
6

UI.A

I
3
4
5

9

O RI ENT A T I ON

TECH. POL'. ENTRE
T E C H . EN T R E . PO L
POL. EI I TRE. TECH
P O L . TE C H . EN T R E
E N T R E . T E C H . PO L
E N T R E . PO L . TE C H
STAFF

2

IDENTIFICATION

NOT I D E N T I F I E D WITH LEGI SLATOR
I D E N T I F I E D WITH GROUP OF
LEGI SLATORS
I D E N T I F I E D WITH PARTI CULAR
LEGI SLATOR
NELTSNER

2

I N EMPLOVMENT

L EGI SLAT OR IMPORTANT
LEGI SLATOR NOT I MPORTANT
OTHER MORE IMPORTANT
S T A F F - LE G

2

SDI S
DIS
NOP
A6RE
SAGR
MI SS

HOUSE
HOUSE DEMOCRATS
HOUSE REPUBLI CANS
SENATE
SENATE DEMOCRATS
SENATE REPUBLI CANS
HOUSE AMD SENATE
ROLE OF LEG

NOT
GRP

123
132
231
213
312
321

VES EXPECTED
NOT EXPECTED
LOCATI ON OF EMPLOVMENT

HSE
HDEM
HREP
SEN
SDEM
SREP
H.S

IOEN

OTHER HI L EGI S L A T URE
MI STATE GOUERHMENT
OTHER GOUERNM'jENT I I ENT
NON GOUERHMENT
EDUCATION
M I SS I N G
ELE CTI ON WORK EXPECTED

VES
NO

IDEHTIFV
CATEGORIES
NOTID
ID.GRP

ACT. ON.OWN
CATEGORIES
STRG.DIS
0 I SASREE
H O . OP I N
AGREE
STR. AGR
M I SS I N G

I
2

ELEC

LGI M
LGNI
OTIM

D E S C R I PT I ON

PRI OR PO S I T I OH _ H E L D

MI LS' z
MI ST
OTGU
PR I U
'
EDUC
MI SS

BOSS

O RI ENT A T I ON
CATEGORIES
123
132
231
213
312
321

URLUE

PPOS

HIRE.FIRE
CATEGORIES
LEG.IMP
LEG.N.IMP
OTH. IMP

ID.LEG

F< I P I

ABBR

CAN ACT ON OWN

STRONGLV DI SAGREE
DI SAGREE
_
NO O P I N I O N
AGREE
STRONGLV AGPEE
MI SSING

TVPES

193
F <* )

FIELD

“F < 1 2 >

P R I OR . POS
CATEGORIES
OTH. LEG. I l l
OT H . G O U . MI
OTH.GOU
MON. GOU
EDUCATION
MI SSI NG

F< 13 >

F < 14 >

F< 1 5>

F < 16 >

FOP)

FCIS)

NAME

ABBR

PPOS

ELEC

LEGIS.LOCUS
CATEGORIES
HOUSE
HSE. DEM
HSE. REP
SENATE
SEN. DEM
SEN. REP
HSE. SEN

LOCH

1

1
2
3
4
5
9

'

VES
NO

1
2

1
2
3
4
5
6

HSE
HDEM
HREP
SEM
SDEM
SREP
H.S

7

BOSS

WORK EXPECTED

VES EXPECTED
NOT EXPECTED
OF EMPLOYMENT

HOUSE
HOUSE DEMOCRATS
HOUSE REPUBLI CANS
SENATE
SENATE DEMOCRATS
SENATE REPUBLI CANS
HOUSE AMD SENATE
ROLE OF LEG

LG I N
LGNT
OTI M

1
2
3

IDEM
1
2

ID.LEG

LEG

3

IN EMPLOYMENT

LEGI SLATOR IMPORTANT
L EGI SLAT OR NOT IMPORTANT
OTHER MORE IMPORTANT
ST A F F - LE G

OR I N

IDENTIFICATION

NOT I D E N T I F I E D WITH L EG I SL A T O R
I D E N T I F I E D WITH GROUP OF
LEGI SLATORS
I D E N T I F I E D WITH PART I CUL AR
LEGI SLATOR
HELTSNER O RI ENT ATI ON

123
132
231
213
312
321

1
2
3
4
5

6

U I.A
3D I S
DIS
HOP
AGRE
SAGR
MI SS

OTHER HI LEGI SLATURE
MI STATE GOUERNMEHT
OTHER GOVERNMENT 1 1ENT
NON GOVERNMENT
EDUCATION
M I SS I N G

LOCATI ON

NOT
GRP

ACT . ON . D UN
CATEGORIES
STRG.DI3
DI SAGREE
NO. OP IN
AGREE
STR. AGR
M I SS I N G

POS I T I ON _ HE L D

ELE CTI ON

IDENTIFY
CATEGORIES
NOT ID
I D . GR P

O R I E N T A T I ON
CATEGORIES
123
132
231
213
312
321

DESC RI PT I ON

PRIOR

MILG
MI ST
OTGU
PR I U
EDUC
MI SS

E L E C T I O N . HR K
CATEGORIES
VES
NO

HIRE.FIRE
CATEGORIES
LEG. IMP
LEG. I I. IMP
OTH.IMP

UALUE

TECH. POL'. ENTRE
T E CH. ENT RE. POL
PO L . EN T R E . TE C H
PO L . TE C H . EN T R E
ENT RE . TE CH. PO L
ENT RE . PO L . TE CH
STAFF

;

1
2
4
5
9

CAN ACT ON OWN

STRONGLY DISAGREE
DI SAGREE
_
NO O P I N I O N
AGREE
STRONGLV AGPEE
MI SSI N G

TYPES

194

F<*>

' F < 19 >

F<20)

F < 21 )

F(22)

F<23)

-<24>

FIELD

NAME

S T A F F 1. I NTER
CATEGORIES
STRG.DIS
OISAGREE
HO.OP IN
AGREE
STR. AGR
MI SSING
NO.DEHB
CATEGORIES
ST RG. OI S
OISAGREE
N O . OP I N
AGREE
STR. AGR
MI SS I N G
PROU-. INFO
CATEGORIES
STRG.DIS
DI SAGREE
N O . OP I N
AGREE
STR. AGR
MI SSING
RE ACT. AF TER
CATEGORIES
STRG.DIS
DI SAGREE
NO . OP I N
AGREE
M I SS I N G

ABBR

OALUE

U I. B
SD I S
DIS
HOP
AGRE
SAGR
MI SS

CONSTI TUENTS MUST GO THROUGH STA
1
2
3
4

5
9

UI.C
SD I S
DIS
NOP
AGRE
SAGR
MI SS

1
2
3
4
5
9

GEN.IDEAS
CATEGORIES
STRG.DIS
OISAGREE
N O . OP I N
AGREE
STR. AGR
MI SSING

SDI S
DIS
HOP
AGRE
SAGR
MI SS

CASEWORK

01 .G

STRONGLV DISAGREE
DI SAGREE
NO O P I N I O N
AGREE
STRONGLV AGREE
M I SS I N G
MOST

1
2
3
4
5
9

IMP STAFF

FUNCTI ON

IS

INFO

STRONGLV DISAGREE
OISAGREE
NO O P I N I O N
AGREE
STRONGLV AGREE
HISSING
STAFF REACT TO EXEC

UI.E
SDIS
DIS
NOP
AGRE
MI SS

STRONGLV DISAGREE
DI SAGREE
NO O P I N I O N
AGREE
STRONGLV AGREE
MI SSING
STAFF PREUENT LEG D E L I B E R A T I O N

UI.O
SDIS
DIS
NOP
AGRE
SAGR
MI SS

DE SCRI PTI ON

1
2
3
4
9

U I.F

STRONGLV DISAGREE
DI SAGREE
NO O P I N I O N
AGREE
MI SSING
STAFF

1
2
3
4
5
9

JOB

IS

GENERATE MEW I DEAS

STRONGLV DI SAGREE
DI SAGREE
NO O PI N I O N
AGREE
STRONGLV AGREE
M I SS I N G
STAFF

CASEWORK HELPS LEG

195

F<«>

F<25)

F<26>

F<27>

: <23)

r<29)

NAME

ABBR

UALUE

CATEGORIES
STRG.DIS
OI SAGREE
NO. OP IN
AGREE
STR. AGR
MI SSING

SDIS
DIS
NOP
AGRE
SAGR
MI SS

1
2
3
H
5
9

FIELD

OTHER'. LEG
CATEGOR'I-ES
MUCH'. LESS
LESS
SAME
MOREI. I N F L
MUCH'. MORE
MI SSING
UOTERS
CATEGORIES
MUCH1. LE6S
LESS
SAME
MUCH'. MORE
MOREi. I N F L
MI SSING
L O B BYI ST S
CATEGORIES
NUCHi. LESS
LESS
SAME
MORE
MUCH'. MORE
MI SSING
I NT. GROUPS
CATEGORIES
MUCH'. LEGS
LESS
SAME
MORE
MUCH'. MORE
MI SSING
OEM.PTV.STF
CATEGORIES
MUCHl. LESS
LESS
SAME
MORE
MUCH.MORE
M I SS I N G

CLLG
MLES
LESS
SAME
MORE
MMOR
MI SS

1
2
3
H
5
9

1
2
3
5
H
9

I MGR
MLES
LESS
SAME
MORE
MMOR
MI SS
DEM
MLES
LES
SAME
MORE
MMOR
MI SS

H
5
9

OF L O B B V I ST S

US STAFF

OF

IMT

GRPS US STAFF

MUCH LESS
LESS
SAME
MORE
MUCH MORE
M I SS I N G
I NFLUENCE

t
2
3

US STAFF

MUCH LESS
LESS
SAME
MORE
MUCH MORE
MI SSING
I NFLUENCE

1
2
3
H
5
9

OF UOTERS

MUCH LESS
LESS
SAME
MUCH MORE
MORE
MI SSING
I NFLUENCE

1
2
3
4
5
9

US STAFF

MUCH LESS
LESS
SAME
MORE I NFLUENCE
MUCH MORE
MI SS
I NFLUENCE

LOBV
MLES
LESS
SAME
MORE
MMOR
MI SS

STRONGLV DI SAGREE
DI SAGREE
NO O P I N I O N
AGREE
STRONGLV AGREE
MI SSING
OTHER LEGI SLATORS

UOTE
MLES
LESS
SAME
MMOR
MORE
MI SS

DES C RI PT I ON

OF DEM PTV STF US STAFF

MUCH LESS
LESS
SAME
MORE
MUCH MORE
MI SS I N G

/

/

196
F <* ^

FIELD

F <30 )

REP.PTY.STF
CATEGORIES
MUCHl LESS
LESS
SAME
MORE
MUCH.MORE
MI SSING

F< 31 )

EXEC. AGENCY
CATEGORIES
MUCHl.LESS
LESS
SAME
MORE
MUCHl.MORE
MI SSING
GOU. STAFF
CATEGORIES
MUCHl. MORE
MORE
SAME
LESS
MUCH'. LESS
MI SSING

F<32>

:(33)

■

: <34>

NAME

PRESS
CATEGORIES
MUCH'. LESS
LESS
SAME
MORE
MUCH.MORE
M I SS I N G
GOUERHOR
CATEGORIES
MUCHi. LESS
tESS
SAME
MORE
MUCH'. MORE
M I SS I N G

in
V

S E L F 1. OP I N
CATEGORIES
GR. UALUE
PRAC'. NEC
UN ITED
IRP.ELEU
MI SSING

ABBR

UALUE

REPB

I NFLUENCE

MLES
LESS
SAME
MORE
MMOR
MI SS

1
2
3
4
5
9

EXEC
MLEB
LESS
SAME
MORE
MMOR
MI SS

1
2
3
4
5
9

5
4
3
2
1
9

1
2
3
4
5
9

1
2
3
4
5
9

3
4
9

US STAFF

OF PRESS US STAFF

MUCH LESS
LESS
SAME
MORE
MUCH MORE
M I SS I N G
OF GOUERNOR US STAFF

MUCH LESS
LESS
SAME
MORE
MUCH MORE
HISSING
STAFF

1
2

OF GOU STAFF

MUCH MORE
MORE
SAME
LESS
MUCH LESS
MI SSING

I NFLUENCE

SELF
GR. U
PRAC
LMTD
I RRL
MI SS

MUCH LESS
LESS
SAME
MORE
MUCH MORE
M I SS I N G

I NFLUENCE

GOU
MLES
LESS
SAME
MORE
MMOR
MI SS

MUCH LESS
LESS
SAME
MORE
MUCH MORE
MI SS

I NFLUENCE

PRES
MLEB
LESS
SAME
MORE
MMOR
MI SS

OF REP PTV STF US 3TAF

I NFLUENCE OF EXEC AGENCY US STAF,

GOST
MMOR
MORE
SAME
LESS
MLES
MI SS

DE S C RI PT I ON

OPINION

OF SELF

GREAT UALUE
PRAC TI CAL ' NECESSI TY
L I M I T E D USEFULNESS
I RRELEVANT
M I SS I N G

/

/

F< * >
I F< 3^ >

F/<^37)

F<33)

FIELD

NAME

OP I N . S E N
CATEGORIES
GR. UALUE
PRAC'. NEC
LIMITED
I RRELEU
M I SS I N G

ABBR
OSEN

DESCR'IPTTOH
SENATORS O P I N I O N

GR. U
PRAC
LMTD
I RRL
MUSS

1
2
3
4
9

REPS O P I N I O N

GR. U
PRAC
LMTD
I RRL
MI SS

AGE

AGE

AGE

P . ID

PARTV

OREP

RDEM
RREP
IND
LIB
OTR
MI SS

F <4 0 >

SCHL

SCHOOLING
\
CATEGORIES
GR.SCH
SOMELHS
HIGH.SCHL
SOME!. COLL
COLLEGE
SOME!. GRAD

MA
PHD
LAUVER
MI SS I N G
SALARY
CATEGORIES
LT I 5
15.19
20.25
25.30
30.35
35.40
GT40
MI SS I N G

1
2
3
4
9

1
2
3
4
5
9

0
1
0
3
4
5
S

7
S
9

SAL
LT 15
.
15 19
2025
2530 .
30 35
3540
GT40
MI SS

OF STAFF

GREAT UALUE
PRACTI CAL NE C E 6 S I TV
L I M I T E D USEFULNESS
I RRELEUAHT
MI SS I N G
IN

VEARS
IDENTIFICATION

OF EDUCATION

...

ACHI EUED

ELEMENTARY SCHOOL
SOME HIGH SCHOOL
HIGH SCHOOL
SOME COLLEGE
COLLEGE GRAD
SOME GRADUATE SCHOOL
MASTERS
DOCTORATE
LAWYER
MI S SI NG
SALARY LEUELS

'■

.

DEMOCRAT
REPUBLI CAN
INDEPENDENT
LIBERTARIAN
OTHER
HISSING
LEUEL

ELEM
LTHS
HS
SCOL
COLL
SGRD
MR
PHD
LAW
MI SS

OF STAFF

GREAT UALUE
PRACTI CAL NECESSI TY
L I M I T E D USEFULNESS
IRRELEUAHT
MI SS I N G

OPIN.REP
CATEGORIES
GR. UALUE
PRAC). NEC
LIMITED
IRRELEU
M I SS I N G

F ( 3 9 ) • PARTV.ID
CATEGORIES
DEMOCRAT
REPUBLI CAN
INDEPENDENT
LIBERTARIAN
OTHER
MI SS I N G

■F(4I)

UALUE

.1
2
3
4
5
6

y

LESS THAN 15000
15000-19000
20000-25000
25000-30000
30000-35000
35000-40000
OUER 4 0 0 0 0

9

MI SSI NG

.

198
.FOR: LEG I S
D I C T I O N A R V LO CAT I ON :
DATE: JUH 1 3 , 19 34

XTWKsLEGIS*

T H I S DATASET CAN BE OESTROVEDr.
T H I S DATASET CAN BE REPLACED^
USE COUNT:
99
DATA SET D E S C R I P T I O N :
L E G I SL AT O R C H A RA CT E RI ST I C S

N M 46

X T W K m H 5-84

F<#>

FIELD

F ( 1)

LEGISLATOR.ID

LID

LEGI SLATOR

F<2)

CHAMBER
CATEGORIES
HOUSE
.
SENATE

CMBR

LEGISLATIVE

NAME

ABBR

UALUE

DE S CRI P TI ON

i

F <3>

F<4)

HSE
SEN

1
2

IDENTIFICATION
CHAMBER

HOUSE OF REPRESENTATI VES
SENATE

PARTV
CATEGORIES
REPUBLI CAN
DEMOCRAT

REP
DEM

LEADERSHI P

LDSP

LEADERSHI P

DIST

ELECTORAL D I S T R I C T

P r 5 > ■— f H- ST f t WJ -

PRTV

LEGI SLATOR
1
2

NUMBER

PARTV

REPUBLI CAN
DEMOCRAT
POSITION

HELD

F(S)

H. TERM'. BEGI N

HBGN

VEAR H0U6E

F<F)

H. TERM'. END

HEMD

VERR HOUSE TERM ENDS

F <8 )

P.TERM.BEGIN

PBGN

VEAR PRIOR

TERM BEGAN

F (9 >

P. TERMLEND

PEND

VEAR PRIOR

TERM ENDED

F<I 0>

S. TERM'. BEGI NS

SBGN

VEAR SENRTE TERM BEGAN

F <I 1)

S. TERMS. ENDS

SEND

CURRENT VEAR OF SENATE TERM

F<!2>

com m ittee; i

COM1

COMMITTEE

FCI3)

PO SITION .1

POSI

POSITION

F< 1 4 )

co m m ittee;

COME

COMMITTEE

F < 15 >

POSITION.2

POS2

POSITION

F < 16 )

COMMITTEE.3

COMB

COMMITTEE

F < 17 >

POSITION.3

P0S3

POSITION

F < 18 >

COMMITTEE.4

COM4

COMMITTEE

2

TERM BEGAN

1
1
2
2
3
3
4

/

/

199

F<#>

FIELD

NAME

ABBR

F< 1 9 >

POSITION.4

POSH

POSITION

F <2 0 )

COMMITTEE.5

C0M5

COMMITTEE 5

F<21 >

PO SITION.5

P0S5

POSITION

F<22)

COMMI TT EE. 6

C0N6

COMMITTEE

F <23 )

PO SITION.6

P0S6

POSITION

F<24>

NO. COMMI TTEES

NCOM

HUMBER OF COMMITTEE ASSI GNMENTS

F(25)

HO.POSITIONS

NP06

HUMBER OF COMMITTEE

F <2 6 >

NO. STAF F

NSTF

NUMBER OF STAFF
LEGI SLATOR

UALUE

DES CRI P TI ON

4

5
6
6

POSITIONS

MATCHED TO

HELD

BIBLIOGRAPHY

Book*
Allison, Graham T . , Bssence of Decision:
Caban Missile C r i s i s . Little, Brown & Co.,

Explaining the
Boston, 1971.

Beneviste, Guy, The Politics of E x p e r t i s e . The Glendessary
Press, Berkeley, 1972.
BeVier, Michael J . , Politics Backstage:
Inside the
California Legislature. Temple University Press,
Philadelphia, 1979.
Brewer, Gary, Politics. Bureaucracy,
Basic Books, New York, 1973.

and the C o n s u l t a n t .

Braybrooke, David and Lindblom, Charles, A Strategy of
Decision:
Policy Evaluation as a Social P r o c e s s . The Free
Press of Glencoe, London, 1963.
Clapp, Charles, The Congressman:
Anchor Books, New York, 1963.
Coser, Lewis A., Men of Ideas:
Free Press, New York, 1965.

His Work as He Sees I t .

A S o c i o l o g i s t ’s V i e w . The

Davidson, Roger H . , The Role of the C o n g r e s s m a n . Pegasus,
New York, 1969.
Downs, Anthony,
Boston, 1967.

Inside B ureaucracy. Little,

Brown & Co.,

Fox, Harrison W. and Hammond, Susan W . , Congressional
Staff:
The Invisible Force in American L a w m a k i n g . The Free
Press, New York, 1977.
Goldhamer,

Herbert,

The A d v i s e r . Elsevier,

New York,

1978.

House, Ernest R . , Evaluating with V a l i d i t y . Sage
Publications, Beverly Hills, 1980.
Jewell, Malcolm E., The State Legislature:
P r a c t i c e . Random House, New York, 1962.

Politics and

Kofmehl, Kenneth, Professional Staffs of C o n g r e s s . Purdue
University Studies, W. Lafayette, IN, 1962.
Lerner, Allan W . , The Politics of Decision-Making:
Strategy. Cooperation and Co nf l i c t . Sage Publications,
Beverly Hills, 1976.

200

201

Lindblom, Charles E . t The Intelligence of Democracy:
Decision Making Through Mutual Ad.iustaent. The Free Press,
New York, 1965.
Lindbloa, Charles E., The Policy-Making P r o c e s s .
P r e n t i c e - H a l l , Inc., Englewood Cliffs, NJ, 1968.
Malbin, Michael J., Unelected Representatives:
Congressional Staff and the Future of Representative
G o v e r n m e n t . Basic Books, New York, 1980.
Marver, Jaaes D., Consultants Can H e l p . B.C. Heath & C o . ,
Lexington, MA, 1979.
Meltsner, Arnold J., Policy Analysts in the B ureaucracy.
University of California at Los Angeles, 1976.
Merton, Robert K . , Social Theory and Social Structure. The
Free Press, New York, 1957.
Ornstein, Norman J . , Information Resources and Legislative
De ci sio n - M a k i n g . Doctoral Dissertation, University of
Michigan, 1972.
Rosenthal, Allan, Legislative Performance in the States:
Explorations of Committee B e h a v i o r . The Free Press, 1974.
Wahlke, John C.; Bulau, Heinz; Buchaner, William and
Ferguson, LeRoy C., The Legislative System;
Explorations
Legislative B e h a v i o r . John Wiley & Sons, Inc., New York,
1962.
Wildavsky, Aaron, The Politics of the Budgetary P r o c e s s .
Little, Brown fc Co., Inc., Boston, 1974.
Wilson, James Q . , Political Or gan iza ti ons . Basic Books,
New York, 1973.

Periodicals
Bradley, Robert B., "Motivations in Legislative
Information Use", Legislative Studies Q u a r t e r l y , volume V,
number 3, August 1980.
Butler, Warren, "Administering Congress:
The Role of
Staff", Public Administration R e v i e w , volume XXVI, March
1966.
Hatley, Lowell and Hofheimer, Susan, "The Legislator's
Source of Expert Information", Public Opinion Q u ar ter ly ,
volume 15, number 1, March 1970.

in

202

Meller, Noraan, "Legislative Staff Services:
Toxin,
Specific, or Placebo for Legislature's Ills” , The Western
Political Qu a r t e r l y , voluae XX, nuaber 2, Part 1, June 1967.
Patterson, Saauel, "The Professional Staffs of
Congressional Coaaittees", Adainistrative Science Qu a r t e r l y ,
voluae 15, nuaber 1, Mar ch 1970.
Patterson, S.O.; Lowenburg, 0., and Jewell, M.E.,
"Editors' Introduction", Legislative Studies Q u a r t e r l y ,
voluae VI, nuaber 4, Noveaber 1981.
Salisbury, Robert H. and Shepsle, Kenneth A . ,
"Congressional Staff Turnover and the Ties-tha t-B in d",
Aaerican Political Science R e v i e w , voluae 75, nuaber 2, June
1981.
Salisbury, Robert H. and Shepsle, Kenneth A., "U.S.
Congressaan and Enterprise", Legislative Studies Q u a r t e r l y ,
voluae VI, nuaber 4, Noveaber 1981.
Shiff, Steven H. and Saith, Steven S., "Generational
Change and the Allocation of Staff in the U.S. Congress",
Legislative Studies Q u a r t e r l y , voluae VIII, nuaber 3, August
1983.
Zwier, Robert, "The Search for Inforaation:
Specialists
and Non-Specialists in the U.S. House of Representatives",
Legislative Studies Q u a r t e r l y , voluae IV, nuaber 1, February
1979.

Collections of Essays
Evaluation in L e g i s l a t i o n , editor Franklin M. Zevieg,
Sage, Beverly Hills, 1979.
Legislative Staffing:
A Comparative P e r s p e cti ve , editors
Jaaes J. Heaphey and Alan P. Balutis, Sage, New York, 1975.
Making Bureaucracies W o r k , editors Carol H. Weiss and
Allen H. Barton, Sage, Beverly Hills, 1979.
Making Evaluation Research Useful to C o n g r e s s , editors
Leonard Saxe and Daniel Koretz, Evaluation Research Society,
1982.
The Political Purseatringa:
The Role of the Legislature
in the Budgetary P r o c e s s , editors Alan P. Balutis and D.K.
Butler, Sage, New York, 1975.
Role Theory:
Concepts and R e s e a r c h , editors Bruce J.
Biddle and Edwin J. Thoaas, John Wiley & Sons, Inc., New
York, 1966.

203

Monographs
Chadwin, Mario T . , "Legislative Prograa Evaluation in the
States:
Four Case Studies", Bagleton Institute of Politics,
Rutgers University, 1974.
Garfinkel, Herbert, "Michigan Constitutional Convention
Studies, The Constitution and the Legislature” ,
Constitutional Preparatory Coaaisaion, State of Michigan,
Septeaber 1961.
Huwa, Randy and Rosenthal, Alan, "Politicians and
Professionals:
Interactions Between Coaaittee and Staff in
State Legislatures", Center for State Legislative Research
and Service, Eagleton Institute of Politics, Rutgers
University, 1977.
"The Legislature, Report #11", Staff Report to the
Michigan Joint Legislative Coaaittee on Reorganisation of
State Governaent, March 1951.
Siaon, Lucinda S., "Understanding Legislative Staff
Developaent:
A Legislator's Guide to Staffing Patterns",
National Conference of State Legislatures, Denver, 1976.
Weingast, Barry R . , ”A Principal-Agent Perspective on
Congressional-Bureaucratic R elations” , Fifth Carneigie
Conference on Political Bconoay, Carneigie-Mellon
University, Pittsburgh, June 1983.