DOCUMENT RESUME

EA 013 591
ED 202 139

AUTHOR Meyer, John W.

TITLE The Impact of the Centralization of Educational
Funding and Control on State and Local Organizational
Governance.
INSTITUTION Stanford Univ.rCalif. Inst. for Research on
Educational Finance and Governance.
SPONS AGENCY/ National Inst: of Education (DHEW) , Washington, .

D.C.
REPORT NO IFG-PR-79-B20
PUB DATE Aug 79. NN
GRANT NIE7P-79-0086; OB-NIE-G-78-0212,
NOTE 33p.; Paper presented at the HEW School Finance Study
Meeting on Resource Allocation, Service Delivery, and
Schdol Effectiveness (September 1979).
AVAILABLE FROM Publications, Institute for Research on Educational
Finance and Governance, School of Education/CERAS
Bldg., Stanford University, Stanford, CA 94305
($1.00).

EDRS PRICE .
MF01/PCO2 Plus Postage.
DESCRIPTORS *Administrative Organization; *Centralization;
Educational Policy; Educational Research; Federal
Aid; Federal FTograms; Federal Regulation; Federal
State Relationship; *Government School Relationship;
Organizational Change; *Organizational Effectiveness;
Organizational Theories; Power Structure; Public
Policy

ABSTRACT
American education is distinctive in the
decentralization of its funding and control. Despite recent
expansion, the role of the federal government is still restricted to
situations, while both
funding and authority in special programs and
educational purposes,
states and localities have authority to define
programs, and policies. At all levels of the administrative
organization, administrators must consider relationships with groups
parent, community, and
outside the educational hierarchy, such, and funding at the
legislative bodies. Centralization of authority
theoretically reduce\ the power of these outside
federal level would relationships within the
groups and increase the importance of
vertical hierarchy, while simplifying and\ritualizing administrative
authority alone were centralized, it
functions. If. funding alone or
occur. The American
would appear that many of the same results would
featuring unrelated federal
case is one of fragmented centralization,
funding programs processed through several independent channels. The
situation seems to read to a massive middle-level educational
below, little
bureaucracy, poorly linked with the classroom world organized
integrated around broad educational policies or purposes,
funding and
for the function of reporting to a wide, fragmentedrespond to the
control environment, and less and less able to
legitimate authority of local systems. (Author/PGD)
 U.S. OEPARTMENT OF HEALTH.
EDUCATION £WELFARE
NATIONAL INSTITUTE OF
EDUCATION
THIS DOCUMENT HAS BEEN REPRO.
DUCED EXACTLY AS RECEIVED FROM
THE PERSON OR ORGANIZATION ORIGIN-
ATING IT POINTS OF VIEW OR OPINIONS
STATED DO NOT -NECESSARLLY REPRE
SENT OFFICIAL NATIONAL INSTITUTE OF
EDUCATION POSITION OR POLICY

Institute for Research on Educational Finance

anco,vernance
tt

SCHOOL OF EDUCATION STANFORD UNIVERSITY

2
 Program Report No. 79-B20

THE IMPACT OF THE CENTRALIZATION OF

EDUCATIONAL FUNDING AND CONTROL ON STATE
AND LOCAL ORGANIZATIONAL GOVERNANCE

John W. Meyer*

August 1979

Education, Stanford
*The author is Professor of Sociology and, by courtesy,
University, Stanford CA 94305.
Finance StUdy
This paper was prepared for presentation at the HEW School
Effectiveness,
Meeting On Resource Allocation, Service Delivery, and School
Septraber 1979.
the Educational
The research for this report was supported in part by funds from
of Education (Grant
Policy and Organization'Division of the National Institute Education
National Institute of
No. NIE P 79 0086) and also by fUnds from the
do not necessarily
(Grant No. OB NIE G 78 0212). The analyses and conclusions
reflect the views or policies of this organization.
other members of
I am indebted to the suggestions of W. Richard Scott, and to Stanford
the Institute for Reset .1 on Educational Finance and Governance,
University.
 INSTITUTE FOR RESEARCH ON
EDUCATIONAL FINANCE AND GOVERNANCE

The Institute for Research on Educational Finance and Governance is

a Research and Development Center of the National InStitute of Educatioff
(NIE) and is funded under the provisions of Title IV of the Elementary
and Secondary Education Act of 1965. The Institute is administered
through the School of Education at Stanford University and is located
in the Center for Educational Research at Stanford (CERAS).

The Research activity of the Institute is divided into the following

program areas: 'Finance and Economics; Politics; Law; Organizations; and
History. In addition, there are a number of other projects and program's
in the finance and governance'area that are sponsored by private founda-
Center
tions and government agencies which are outside of the special R & D
relationship with NIE.
The Impact of the Centralization of Educational Funding and Control

State and Local Organizational Governance

- Abstract

The actual or potential impact of various forms of organizational

centralization of state and school district4 is discussed. Special
attention is given to the present American system of fragmented cen-
tralization--the rise of central controls through funding in .a. dispa-
rate set of specialized areas, without policy integration or the centra-
generates,
lization of general authority. It is argUed that this system
educational bureaucracies that are increasingly
at lower levels, large
from
organized around environmental demands and increasingly decoupled
actual educational work.
7
 Funding and Control
The Impact of the Centralization of Educational
State and Local Organizational Governance

its funding
American education is distinctive in the decentralization of
has
and control. Unlike other nation states, the American central government
and attempts to construct
little constitutional authority to regulate education,
Even a national Office of
such authority have historically been defeated.
States. And to this
Education came late and-With feW powers in the United
other countries
day we do not have a Ministry of Education, unlike most

(Ramirez and Rubinson, 1979).

weakoor unimportant
This is not to suggest that education is historically
of
in the United States. An expanded system of schools, with very high levels

enrollment, developed early -- long before there was much Federal involvement,
elaborated. their authority in
and considerably before the several states had
through a
education (Meyer,. et al., 1979; Tyack, 1974):This sytem arose

decentralized, process -- a series of social

national, but organizationally
controlled set of decisions.
movements, rather than a unified bureaucratically
with a network of
The school system grew up in the 18th and 19th centuries
only in the latter
local community and parent controls and constituencies, and
organized in some part around state -
part ofthe 19th century did it come to be

level controls and funds.

In most nations, especially-
This is all quite unusual in the modern world.
the national state: built
the newer ones, education is a central function of
and funds; and controlled at many
up in the first place by'state decisions
Meyer, et al., 1977).
points by national rules (Ramirez and Rubinson, 1979;
of the nation
Organizationally, education in the modern world is a creature

Its central features .curricula, organizational roles, teacher

state.
usually defined
qualifications, subdivisions of types of schools -- are all
 -2-

of educa-
by a Ministry of Education, which provides a kind of general theory

tiqn underlying the whole system.

Something like this greW up at the state level in the United States,

beginning late in the 19th century (for instance, with the passage of compul-
in
sory attendance laws between 1870 and 1910), and developing'bureauctatically

the 20th century. At the state leveli one finds the emergence of some official

the
rules defining curricula, defining teacher qualifications, prescribing

appropriate properties of school buildings; and so on -- much like the rules'of

national centralization found elsewhere. But even with this, most of the lunding

recently
for even public primary and secondary education has continued until very

to come from local sources.

involved,
Since World War II, the Federal Government has come to be more

but still in a limited way. Only a small fraction of eduCational funds come

from this level. Typically, these funds are attached, not co education in

general, but to various special programs, focused on particular kinds of stu-

handicapped)
dents (the poor. mino:ity groups, the educationally or physically
schools). And
or particular schoolin&situations (impacted areas, segregated

the authority built up has been legitimated not as an expansion of general

Federal educational control, but rathet by very special purposes -- most

1976).
commonly, the restriction of inequalities (Levin 1977; Kirst,
of
The point is central to our'later discussion: With all the expansion
still restricted
the Federal role in education in recent decades, this role is
and situations.
to funding andauthority in various_special educational programs

There is no legitimated Federal or national policy covering main educational

for admisdion to
issues. ,There'are no national curricula, no national teats
 -3-

national
the next level of schooling, no national criteria for achievement, no

definitions of appropriate teaching methods, and no national principles for

accrediting schools or teachers. Thus, there is some centralization of finding,

and of the parallel organizational controls, but no great expansion of legiti-

set'of
mate national responsibility and authority over education, and no
the battle
Federal functionaries, who can authoritatively lay out or integrate

rules of .American education. When the American government reports on education

passively reporting
to the world,, unlike other states it is in a position of

Whit seems to happen in states and localities (UNESCO, 1955-66).

The situation with states and localities is less clear. Both have legiii-

(Kirst$
authority to define educational purposes, programs, and policies
le
mat

1970). State authcrity is in principle extensive, though local decision-making

This distinction is less
often seems to domlnate, at least in the short run.

important for our purposes, which are to discuss the impact of such Federal

centralization as exists on states and localities.

and local educa-
In this paper, we consider the potential impacts on state
We consider first
tional organizations of various forms of centralization.
which would flow from
a hypothetical case -- the organizational consequences
authority. Then we
the centralization of both funding and basic educational
centralization, when
consider the potential impacts of two partial forms of
control but not
both have a,simple structure: the centralization of funding

authority,Hand.the centralization of authority without funds.

Finally, we discUssthe American case -- it involves the centralization

authority, but in a
of funding without the centralization of substantive
-

For the expanded Federal

peculiar way: we call it "fragmented centralization".
--/no single
role in the funding ofeducation has not itself been integrated
 -4-

the'separate funding principles and

Federal office or program brings together

makes them consistent. Rather, a long series of special controls and funding
1
and control system, but
programs have been created, each with its own purposes
1977). Further,
none part_of a larger_ integrated or unified package (Lavin,
also fragmented
this Federal system'is not only fragthented at the source, but is

Some Federal, funds flow to the states, and

in its organizational structuring.
is much sub-
through them down the organizational line (though even here there
unintegrated).
stantive fragmentation, so that different funding programs are
And'still others flow-more directly
Other flow directly to school districts.
fragmentation,'. inci-
toward schools, or subprograms within schools. "(This same'
the separate states fund
dentally, also describes some of the programs by which

districts and schools.)

organizational impact of centralization, we find
As we discuss the general
related issues.' Almost
ourselves at odds with the main body of literature on
with the impact of
the entire American literature on the subject deals, not
specific centralizing programs
centralization in general, but with the impact of
We have many studies of the impact of
(e.g. Berman and McLaughlin, 1975-78).
to aid the youthful poor,
Federal desegregation rules, of the iaact of programs
speciiicjlandicapped students, and so on.
of the impact of programs targeted on
organizational flavor: they are
These s /udies have a substantive rather than
implementation, or if implemen-
concerned with the great failures in educational
They only indirectly concern
tation has gone on of failures in effectiveness.
itself as an organizational
themselves with the'impact of the centralization
the problem. And they rarely
phenomenon, though many side comments are made on
simultaneously operating centralized'
or-never discuss the impact of a system of
here. Consequently, we discuss
funding and control programs -- our main problem
 -5--

than might, otherwise

the problem theoretically, but with'less empirical support

be the case. And we suggest appropriate research designs with which the problem

might be better attacked in the future.

A Theoretical Note on Educational Organizations

The organizational impact of centralization depends heavily on the nature

The centrali-
of the organizational domain- .inwhich centralization is going on.

zation of control over automobile production, for instance, should have quite

different organizational effects from the centralization of educational.control.

do-
The main distinction needed here is between technical organizational

mains and institutional ones (Meyer -and Rowat 1977, 1978; Meyer, et al., 1979).
side
Technical domains are mainly cont olled from the actual work or output

market or other specification are imposed on products. Whatever organizational

controls exist must in-ione.way o another come to terms with, and be partly

justified iterms of, properties of actual work or its outputs. Institutional

domains are those it which environmental forces specify work forms or categorical
of the propriety
processes as proper and legitimate, and provide funds in terms

in institutional terms of the general activities and categories of activities.

Thus, centralization in a technical domain can effectively occur only so long

properties; Mective
as some central control is coordinated with actual output

centralization it automobile production, for instance, usually means the cen-

tralized coordination And control over at least some of the properties of-the

automobiles produced.

InAmstitutional domains, this technical constraint is weakened -- and

education is an excellent example. Centralized control means the central defi-

not necessarily
nition of some of the legitimate categories of activity,---.it_does
outputs. Nor
imply the central control over properties of the actual-student

10
 -6-

does it necessarily imply the central control over the actual work processes
work
affecting these outputs -- in institutionalized settings like education,

processes are.frertly delegated beyond the control of the formal organization

itself.

This general property o. educational systems is sometimes called "loose

coupling" -- the tendency of educational organizations to discOnnect policies

from outcomes; means from ends, and structures or rules from actual activity

(Weick, 1976; March and Olsen, 1976; March, 1978). Educational organizations

would,ba.
can be centralized around rule systems that are unimplemented, or

disastrous or inconsistent if implemented. This occurs in technical organiza-

the actual
tional domains too, but is much\more likely to create difficulties in

work processes or outputs that are brought under organizational control.

systems can more easily
It is thus, important to remember that educational
property:' they can more easily.
adapt to Centralization because of this general
avoidance
deal with impossible or inconsistent centralizing constraints by the
control,
of'implementation, the ritualization of implementation or evaluation or.

and so on (Meyer and Rowan 1977, 1978). .One consequence of centralization in

formal
education, for instance, can simply be the further decouPling of the
and
authority or rule system and the actual activities going on in schools

classrooms.

A second loosely-coupled property of institutionalized organizational sys-

tems like education must also be kept in mind. In these organizations, high

dis-
proportions of administrative or organizatiOnal management activity are
closely connected
connected-with the actual work activities of schools, but are
(Pfeffer and
with the political and institutional structures of the environment
\

Salanclie, 1978). Since technical work is managed below the level of inspection
 -7-

organizational
Of the organization's 'structure, much or most management and
This kind of
activity is free to adapt to changing environmental constraints.

organization tends to be closely linked to ,the environment, for the environmental

required
social and authority system is the source of the legitimacy and resources

for organizational action and not through market processes involving output

School and district

inspectionn but through direct dependence relationships.
. ,
N
administrators must be on good terms with their politics' and institutional

constituencies: output issues are much `less critical. No matter how good an

instructional job a school does, if it is disaccredited by the district, state

I ,

Or Federal rule system's, much trouble ensues. And such disaccreditation ordinarily
I

formal
has nothing to do with actual output measures, but rather arises from

institutionalized rules defining proper organization, procedure, and so on.

Therefore, in, institutionalized organizational systems like education,

of technical
Administrative functions ordinarily involve less the management
and conforming lo
work than the management of relations with the environment

institutionally required rituals (see the papers in M. Meyer, et al., 1978).

THE IMPACT'OF,UNITARY INTEGRATED CENTRALIZATION

Suppose,\in this kind of organizational system, control and funds were

\
curriculum, and
completely centralized to the Federal level, with a national
controls and
nationally - defined structure,ld with autonomous local and state

funds completely eliminated. This would be a system Much like those that pre-

vail.in many countries. What organizational impact on the lower levels (state;

district, and school) would we expect.

Compared with a hypothetical system in which there was littl8shigher-level

local community
control of any kind (i.e., one-room schools responsible to the

12
 -8-

involve a great
only,or to parents only) the centralized system would clearly
There would be school
expandion of administrators and administrative levels.
and Federal
administrators-, district administrators, state administrators,
each
administrators, with some intervening layers of supervisors or inspectors,:
ft'and would
layer would be directly and solely responsible to the one above
The result would be a simple
habv authority and control over the one below it.

chain -of- command organizational structure, as many ,countries have.

decentralized, Is
But the present system, which is by no means completely

the relevant comparison. And it is not clear that simple centralization would,
each educe-
.
increase administrative activity over the present system, in which
:,
level,
tionaflevel has to maintain administrative linkages, not only with the

above it, but with many lateral groups. Teachers in the, United States deal,

Principals also
not only with the principal, but with parents and community.

deal with parents and community, often extensively, not only with the district

Superintendents deal with school boards and the community. And state
office.

education.functionariei deal with legislatures, interest groups, and so on

(Kirst, 1970; Wirt,:1975).. The present political culture of American edUaation

have both
is filled prithaiily with such lateral relationships, lateral groups

legitimate authority and funding power in the system.

complete centralization
.
All thissttucture would be undercut by a.shift to

of authority and Lunds. And the administrative prOblem at, each 'educational
N4,

level.wodld be greatly simplified. This is especially true given the institu-

administrative
tionalized character of educational organization, in which primary
linkage
77\,problems involve less the coordination orwork 'activity than effective
environment: Thus:
with the institutionally authoritative (and funding)

4
 Proposition A: Complete centralization of funding and control in
American education in the hands of an integrated Federal authority
would reduce the size of the administrative component at each
lower level.

But not'only size would be reduced. The administrative function would be-

come directed primarily toward the management of conformity at lower levels and

the maintenance of conformity with higher ones. Lateral relations would be of

lesser importance. Several consequences follow:

Proposition B: Complete centralization would increase authority

and hierarchical relations throughout the system, and would de-
crease the organizational extent and importance of lateral relations.

.Proposition C: Complete centralization would lower the number and

complexity of environmental relations for each level, and would
therefore lower (independent of size) the organizational differ-
entiation or complexity.foUnd at each level.

Thus, one of the present characteristics of American school district and

state organisations isithe small number of administrators who have direct line,

authority and the large number of administrators and units with specialized

functions (either in dealing with special aspecta of the environment, or with

special-educational issues). (See Scott,'et al., 1976.) This balance would

radically change with,centralization. With more line administrators, and fewer

differentiated functions, organizational integratiOn would be increased, both

- .
vetical3j. and horizontally. At 'present, vertical relationships in American

education-are weakened:by,the fact that each level in the :system has autonomous

lateral relationshipsi'and so ,evaluation and inspection are very weak (Dornbusch

al.:d Scott, '1975). The coordination of lateral.relagionships among administrative

4t»

Units t3 attenuated by the same faCtors (see Rowan'1979; Davis, et al., 1977).

--Proposition D: Complete centralization would increase vertical

integration (through inspection' and evaluation) and. the hori-
zOntal integration of each level (through aUthoritative.admini7'
strative 'coordination)
 -10-

Incomplete Unitary Centralization

1.
funding base
Suppose the creation of an integrated Federal authority and

did not replace, but was addedion to, the present system,with its authority

and funding bases operating simultaneously at several levels.

authority rela-
Some of the propositions above might still hold: vertical
also
tions would be enhanced (Proposition B), and vertical integration might
probably not occur.
be increased (Proposition D, part 1). But the other effects would
Administrative' size ancFcomplexity at any level (Propositions A and B) would probably
would add complexity to the relation-
increase -- this form of centralization

ships of each level with its environment. And integration at any level (Propo-

sition D,.part 2), would be' made more difficult, not less. difficult.

Obviously, the overall organizational effects of a unitary centralization

dependmeryheaVily on the extent to which authority and funds are completely

centralized, and on what other sources of power are left in the system.

Ef ts'of Centralization on Teaching Work

instructional
V have considered above mainly organizational effects, not

ones, of centralization. But clearly effective centralization would tend to

standardize curricula and instructional practices in school's, and tend to immunize

O
controls, tastes,
these aspects of teaching from the variance introduced by local

and so on.

Standardization would also result from other than .organizational proCesses.

in
The legitimation of.a unified central authority would involve 'a reduction
nationwide
the puralism of American. educational culture -- the collection of

eliteg-TaiiTideas about education that clearly plays an extremely important Tole

in the present system. At 'present, this culture -- which seems to,manage the

Hansot, 1979; Meyer,.

system more than organizational decisions do (Tyack and
and ideologies flow
et al., 1978) -- is quite diVersified, aa-educational ideas
Centralized authority would turn
through a decentraliied cultural marketplace.
who now
the intellectuals and innovators-and educational cultural authorities,

15
 for the status of advisors
compete in a dispersed market, more into competitors

to the prince. Even now, in American education, the social right and capacity
of status and
to give testimony to the Congress becomes an increasing source

legitimacy for putative innovators and advisors.

Research Designs on Unitary Centralization

Little unitary centralization has gone on at the Federal level in the United

States. But it would be posSible to investigate effects with the following

strategies:

1. Comparative cross-sectional studies could examine variations

in lower level organizational structures across countries,
t to see if they are affected by centralization.
in
2. Comparative longitudinal studies could investigate changes
lower-level organization as they are produced by-centralization.

3. Comparisons could also be made among AmeriCan States, which

vary considerably in their degree cf Ltary centralization,
organization. Some
to look for effects,on district and
cross-sectional data which would lend'themselves to this kind
of study, are already available (Abramovitz and Tenenbaum, 19.78).
But longitudinal studies -- looking far the impact of increases
over time in state centralization would be highly desirable.
Of course, state studies could only., really investigate the im-
pact of partial centralization, since rarely,in America does
the state assume such a commanding role as to: eliminate auto-
The most highly
nomous authority and funds at lower levels.
centralized state -- Hawaii --.should.obviously be given special
research attention.

FRAGMENTED CENTRALIZATION:. TWO SIMPLE CASES

Suppose only one of the two basic.forms of educatiOnalpowerauthority-

14hat conseq-
and funding -- is centralized, with the other left decentralized.

the question is of some practical importance, becaUse

.

quences doei this haVe?

American case. Federal funding
in some ways this describes recent changes in the
legitimate authority over
.
has increased, but there has been little Shift In.
 -12-

central educational issues to the Federal level. In reality, though, the nmeri-

can case is more complex, as we will discuss below.

Centralized authority, but at least1partially decentralized funding, is a

relatively common situation in other countries. Given the general character of

in the
educational organization, with its inclination to become loosely-coupled

relation between organizational authority and actual work patterns, the conse-

quences of this situation are fairly clear.

Proposition E: The centralization of authority but not funds in-

creased the decoupling of authority and actual practice. Practice
is organized in varying ways, depending on funds, but is integrated
in a common categorical scheme built up by the centralized authority.

have
This situation is readily visible in-the AmeriCan states, many of which
. ,

built up a ,unified, educational authority

without complete control over funds.

high
In this Situation,jall schools must formally meet standard criteria. All
instance,
schools,; for instance, adopt the same category scheme (all offer;'for
almost all have
a 12th grade, almost all offer a long list of standard courses,,
Criteria) : But
the basic curricula, and almost all the teachers meet state.
.

(recall that
they vary greatlyjn practice, depending on student constituencies
Some
these schools are highly dependent on the local community) and resources:
Some of the schools
of the 12th grades are actually doing 10th grade work.
Some of the
.have teachers who arevastly:better qualified in- their subjects.
This
Chemistry courses are by no stretch of the imagination' real Chemistry.

situation is even more.caMmon in developing countries.

Nevertheless; the simple centralization of formal authority, may produce some

propositions. The
of the administrative consequences indicated in our earlier
administrative bur-
weakening of local authority probably simplifies and reduces.
Conceivably horizontal integration at each
dens and organizational complexity.
 -13-

level is also increased, though we have argued that vertical integration is if

anything lowered (Proposition E): in a system with centralized authority, it

becomes crucial to minimize information about local nonconformity and variability,

and vertical relations tend to be.ritualized.

Consider now the other alternative -- the centralization of funding without

authority, as in the American Federal case, and as in the earlier history of

some of the states. Here we have the ritualization of:local authority in meaning-

less policy decisions, policies that cannot be implemented, and so on. And we

who
have the rise of what.may broadly be called the accountants -- the personnel
The central
manage the funding and reporting relations with central power.
k

functionaries do not have the direct authority to set policies, and so justify
Sometimes these are
their expenditures through narrower technical rules.

budgetary rules that restrict the possibilities for expenditure. Sometimes

.
they are accompanied bybureaucraiic educational rules --,e.g., technicaL

definitions as to what is. and is not properly fundableSchooling. And sometimes

appropriate'educational out-,
they are, accompanied by technical definitions of
of attendance,
puts: (a), Thus the.American eMphaSig on the proper accounting
of attend-
Since funds are dependent upon ADA,.and the ignoring Of .the character
. . .
. .

.
.
.
. .

, .

stddy, continuation,
ance (which has-produced, the most interesting types of work-
in school);
and alternative schoolarrangements for.studenta not actually present
evaluational
(b) and thus thefrequen American emphasis on some kind of technical
needs.of
data on%student outputs ---data that can suppoit the justificatory

central functionaries who do not haVe authority to operate on their own judge-
,

ments (McLaughlin 1975; David, 1978); ('c) Arid: thus the American emphasis oni

counting graduates, and treating dropout as a-uniformly negative characteristic.-,

Thus:
 Proposition F: The centralization of funding but not authority
statis-
generates organizational controls through accounting and
Administrative work is deflected from policy
tical mechanisms.
and authority and toward accounting systems.

centralization. Much
It is difficult td discuss other effects of funding
by which they
depends on the way the flunds are organized, and the criteria
centralization of funding
are allocated.: In gener l, we may suppose that the
) levels and
cuts off power relations between eadh of the lower organizational
administrativeexpan-
its lateral constituencies, and thus lowers the need for

sion to coordinate with them. Thus, many of our initial propositions (A t rough

D) may hold.

Research Designs, 1

Here again, it would be useful to engage in cross -national longitudinal/

The states differ greatly

studies, or in studies of. American states over time.
educational authority.
in their relative centralization of funding and of formal
The opportunity
And there have been many changes over time on such dimensions.
changes for local and
for time-series studies of the. consequences of these

school organization is substantial.

FRAGMENTED CENTRALIZATION: THE AMERICAN FEDERAL CASE

There has been relatively little

Ile American case-is still more complex.
There has been more funding
centralization of direct educational authority.
There:is'no system'
centralization, but- even this. has been-highly,fr4mented.
and these funds are
for the integrated disbursal of Federal educationai-funds,
organizational mechanisms. The
organized in terms of disbursal through varying

important points are two: First, there are many-unrelated funding programs. '

the System of American education

Second, fundsliom.theselprograms go through
and
'through different channels,. Some funds go rather directly, to schools,
 -15-

and states'. Other

direCtly undercut the intervening authority of districts
Still other go to states, and
funds go to districts on a rather direct basis.
It may be said in brief
are allocated to lower levels through state authority.
and something of
that the system is an organizational theorist's nightmare,
send and receive a blizzard
a bad dream for administrative practitionerswho must
The practical literature
of reports to and from distinct reporting agencies.
organizations have risen up
is filled with complaints, and especially the state
burden." We may assume
to resist various aspects of the "Federal reporting
integration in the sys-
that in the future at least one aspect of organizati
educatlonal funds flow
tem will be enhanced -- that increasingly,,Federal
But eveia this is not entirely
through the states, and not directly to subunits.

clear: 1

structural-consequences
The organizational literature is clear on the main:
integrated one:
of such a system, when contrasted with a more

Proposition Administrative size and complexity will expand

vari-
in subunits as those subunits are exPosed to an increasing
environments
ety of funding and Authority relatiohs with their
SalaCik,
(Emery and last 1965, Terreberry, 19.68; Pfeffer and
1978; Udy, 1970; Deal, et al. 1977). .

structural,.
And,giveryhe 'inclination of educational organizatiohs to decouple

levels from each.other, and frOm work activity:

Proposition H: Subunit horizontal integration and internal coor-

dinationyill decline 'with an increased variety of distinct
(Meyer and
funding and authority relations with the environment
fRowan, 1977).
.

Proposition:I: Subunit' organization will shift

.

___
in E/tructure from
those.d.tVerentiated to match
forms designed tocoordinate.work to
and-Salancik, 1978;
the environmental funding structure (Pfeffer
Thompson; 1967; Deal et al.,1977).
 -16-

increasingly
Thus the structure of state Departments of Education should
exigencies of educa-
reflect the structure of Federal funding programs, not the

tional coordination. So also with the structure of school district organiza-

So also, even, with schools. And the capacity for

tions (see Rowan, 1977).
1979).
internal lirikage should decline (Davis, et al., 1977; Rowan,
integration too,
But this'hoids true of the capacity for general vertical

especially given educational loose-coupling:

and funding.
,' Proposition J: Faced with diverse external authorities
requirements and lose
sources, subunits/organize around reporting their sub-
their more general capacity to exercise authority over
1;
ordinate levels./

following fact:.
Further,, this general inclination is-increased 'by the
. / .

much room'for internal

The unintegrated system of Federal controls leaVes open
for inconsistency
inconsistency in.ita'requirementa, and also leaves open room

with state or local requirements. Thus:

increased
Proposition K: Vertical and horizontal loose coupling is
funding agencies pose conflicting and
as environmental'groups_and
inconsistent requirements (Meyer anURowan, 1978).

superintensiegt.
Consider the practical situation of a school principal or
for handicapped stu-
The state will provide, extra funds for a special,program
if there is no.special
dents: the Federal government Will provide further funds

program (i.e., for' mainstreaming). The parents insist that funds be managed

and federal governments

equitably within the school. and district: butboth'state 1-

ON
few schools, or even for
provide special funds which must be spent only within a

Or.more simply:: the,reporting.and budgetary, /1.

a few students within.a school. 4

ForPrograms.D, E,'and
deadline for Programs A, B, and C, is July 1.
awareness(and
Is February 1: Yet each budget and each report itiOiltlie made in

1
 -17

And each budget and report

reported awareness) of each other budget or report.
conflicting, categories.
must be organized in terms of different, and sometimes
The answer is s ple: have a differentiated
What is the administrator to do?
subunits report as
subunit for each funding or authority program, let th se
aving the subunits brought
beat they can in conformity with requirements, avoid
inconsistency),
in contact with each other (so. as' to avoid explicit conflict or
various programs, reports,
and remain in ignorance of the exact content of the
rather than one of
and budgets (so as to maintain a posture of incompetence,
picture of
dishonesty). The ideal administrator, in this situation, will be a
nominally under'
ineptitude: ignorant of the most obvious aspects of the reality
'This posture will
supervision, and tolerant of the most aggregious mistakes.
to funding and program.
have the additional organizational banefit.of confirming
the:Urgency of their).
.officers.at higher levels,the importance of theitfunction,
41

beneath them for their

work; and the despaTate need of.theeducational world

reforms,
And the literature
This is the world suggested- by' the prOpositions above.
far frabvreality..The old line
is full of.suggestIons that it is not so
much gonel, and
authoritative and-charismatic local administrator is pretty
[

called negotiation (TyaCk

replaced by,administrators skilled in wh4t might be
/'

provided by centralized programs are indeed spent,

.
.

and Hansoti 1979). The funds

fact been implemen -.
but much question arises ahout,whether'the programs'havein . ,

1975; Levin, 1977). Reporting

ted (Berman and McLaughlin, 1975-78; McLaughlin,
apd,evaluationrequiremeniaare7ritualized (David, 19,78) and often evaded

entirely._
with the fragmenta-
*f,the state level, we have fragmentation, isomorphic
,

Large state staffs are differentiated, not only

-ArounOtate r
tion above.
 -18.-

distticts. Large propor-

but around Federal program structures, just as with
funded with Federal money
tions of State Depattment of Education employees are

(as with district and.schoole, too). -

,Conflict and Dissatisfaction

considered its
In discussing the impact of centralization, we have not
the most common vari-
impact on stress in the organizational system -- perhaps
1

able conventionally discussed.

technical domain of
Clearly, fragmented centralization, if applied to.a
As-disconnected .

organizational life, results in much stress and conflict.

instance, a
.central controls are imposed on the automobile manufactUrers,for

manufacturers, must satisfy consumer demands for

great deal of-Atress occurs:
\.
that insure. that theywillnot start
cars that start, but.also pollution rules
tastes for large and heavy
smoothly; EPA rulesabout gasOline. mileage,_,but also
efficient interiors, but also various
Cats; customer-'desires for attractive and

cumbersome safety .rules. .

domains, it is not so clear.
But in 'education, or othet institutionalized
actual work processee,
None;oftherules really specify and control in detail
of the actual student outputa. Conflicting
*,'and almobt none looks :at the.properties
be simultaneously inCorpor- '

andincensietent environmental deMands can therefore .

. ,"___ .

(e.g.,:lack.of communication)
ated, so long as adequate-organizational buffering
/
. e : ,

institu-
an organizational solution to
-,

is maintained.: Loose coupling, thus, is ,,

An automobile - manufacturer
tional inconsistency.in the -educational sydtem.
that his cars have -had fifteen units
cannot satisfy the government by reporting -
consumer-by a special program.in
of instruction in mileaga,ind.satIsfy the
A school can, by and large; do so.
heaviness and solidity ofconstruction:
a
that -- if properly 'covered -=
It cam-incorporate in its formal curriculum topics
 -19- \*

would consume much more than the school day, just as we professors ordinarily
In a
incorporate exaggerated depictions of the work in our formal' syllabi.

processes and products are institutionally invisi-

,

systeM in.which actual

ble, and presented to the world as a set of myths and rituals (Meyer and Rowan,

1977, 1978), inconsistency is'often easy to.deal with. Any effective school

administrator can honestly assure,some constituencies that the school has sex

education (as a regular part of family life, or even biology, instruction),

and can assure other constituencies that it doesn't. So also with special

versus mainstreamed education. Or separate and funded programs for minority

students, and yet completely integrated ones.

There are Clearly iimitshere, and sometimes -- reality in the classroom

required'
*entirely aside -- there are direct conflicts between environmentally

category and accounting systems. But these can often be resolVed by the reten7

.tion Of buffered.parallel category structures (e.g., vocational and college

by the maintenance
preparatory curricula with .the same formal course titles), or
response, to differ-.
Of several sets of accounting books simultaneously for use in

etit environmental demands::'

.

of
Indeed, we can go further: whatever problems are created by the system

fragmented,centralization' may be over-8h domed by,the organizational resources

created by the Same system. Each organizational functionary now has additional
-
supervisor
"_constituents_to_play off againsi eac h _other: the special education
of:the program result
can tell the superintendent that the unsatisfactory aspects
/
constituents.
from Federal requirements, and can so'put off community and parent
thajocals by clalming
The state 'functionaries can increase their leverage over
And the locals, even,
1

'that Federal coercion is responsible for the new pressures.

Once we see educational
can resist various state eifeorts on the same grounds.
 -20-

organization as decoupledfromeducationsl work and outputs, it is a kind of sha-

dow structure: any given situation can be seen as having only one technical

reality, but it'can have gldny shadows . Thus:

'Pro Osition IL: Fragmented centralization does net, overall;

tend tp'increase conflict and reduce satisfaction in the
organizational system.

What, though, about the components of the system that are not organiXational

the collection of parent and community and interest group constituencies that

surrounds the organization. In many respects, their power to penetrate the

organization is reduced as centralizatiomccurs. We may expect the'following

general consequences:.

Proposition M:'Fragmented Federal centralization. decreases the

power and satisfaction of those constituencies organized pri-
y. at lower, leVels,of the system (i.e., parentsand'communi-.
marily
ties), but may increase:the power and satisfaCtionOf those
constituencies that can mobilize at the Federal LeVel.

Proposition' N: Fragmented Federal centralization increases the

inclinatia of constituencies to mobilize at the Federal level.

Some Additional SpecUlations On.Effects of the Present System

The present system of control and funding in American education is quite

I

distinctive, and a number of disconnected ideas about its effects may be worth

noting: .

1. Effects on the political culture of'American education:, A) Charisma:

As with all systems in which funding is more_ centralized than authority charisma

tends to be drained from.the lower levels, but not to shift to highei ones,

Local and state administfators are increasingly in the position of budgeting

functionaries, not'edUcational leaders. But the Federal fUnctionaries do.not

gain' much generaliied authOrity -- they are in the position of controlling special

funds and programs. This ipart of a historic pattern in which patterns of

 -21-

attributed "educational leadership" have flowed up and out of the educational

organizational system itself, and have come to be located in external rational

elite groups and figures. The heroes of American education are neither the old

local and stateiadministrators (Tyack and Hansot, 1979), nor the new Federal

officers. They are.now outside the organization almost entirely -- intellectuals

'and reformers and professors with missions of reform, and often with justifi-

cation phrased in terms of research (sometimes research not yet done) or

scientific authority (Tyack, et al., 1979). Some recent superintendents of

schooling in America, for instance, have included James Conant, James Coleman,

and Coons and Sugarman.

This issue can be studied by simply coding summaries of outgoing national

educational discussions over the last century:to trace evolving centrality in

leadership. Our argument here is that the Shifts upward in funding -but not

authority build up suppOrtfor external charisma. We suppose that authority

1

in the present SyStem tends, to shift outward into the hands of the intellectual
.

and ideological controllers.of national eduCational fashion cyCles.

B) The authority Of parents and community: Clearly the changes we beim been

discussing lower the power :of the4ocal community over its schools. The locals.

are put in the position of having to support their schools in'efforto to get

more state and federal funds, not in the position to exercise control (Deal,.
.4
et al., 1977). -Surveys over time should shoW this effect.
,

But beyond this, lOcal power changes its character: we .expect that local

groups' increabihgly.thobilize, not around local issues, but around evolving",

,Federal controls and national fashions. Movements for competency-based'instruc-,

has ben.
tion or basic education, for instance, should OCcukequally *ilhete there

.
'ncylocal test score problem sakwhere there. has., As the system centralizes,

the'bases for imbilizatidm within it centralize too.

 -22-

flow-through funds:
2. Effects of expanding state control of Federal
gained control over the,
There is some evidence that states have increasingly

internal management of Federal funds. If this occurs, the effects internal to

under the heading of

each state may be a little similar to those we discussed

unitary centralization. State power and authority are increased, and the

(and with it
pluralism of the environment around each district decreases

internal administrative complexity).

takes the
3, Effects of accountant dominance: Because Federal control
justifications tend to be put in
form of fundings, rather,than authority, its
And its impact on
accounting or test - scores or evaluation research terms.

This means that at all levels, we expec

lower levels has this character too.
accountants, evaluators;
to fin\the increasing dominance of technicians --
authorities (David, 1978;
testers, and so on -- rather than broader educational
(a)
Kirst, 1977). Research could study this question in a number of ways:
rise in the system?
What kinds of people, with what kinds of backgrounds,
most at lower levels,
And how is this changed? (b) What types of roles expand

and how has. this changed? (c) To what extent have technocratic ideas about

dominate over substantive ones?

management (e.g., PPBS, MIS, etc.) come to
local and state educational
(d) How have the types of documentation emitted by
produced)? (e).-What types
organizations changed (e.g., the types of paperwork
and for what reporting purposes?
of information are gathered in the system,

And how has this changed?

Research Designs-on Fragmented Centralization

.emphasized longitudinal
In discussing earlier forms of centralization, we
comparativeatudies of national 'societies.. But here we
studies of states, and
have been going on at the national
are dealing with educational changes that
 -23-

Research should attack these issues directly. We

level in the United States.
structure (admini-
need longitudinal studies of local and state organizational
external funding
strative size, administrative complexity, organization around
vertical sytems of control
sources, horizontal coordination in policy terms, and
expansion, and to
and inspection) as they have responded to Federal funding
Independent
some extent Federal authority expansion in specialized areas.
controls on
variables would be properties of the Federal systems, with some
Dependent variables would be such
state and local structure and resources.
It is absolutely crucial
local and state properties as those listed above.
the critical
that such studies should be longitudinal in character, to capture
Derivative studies of which try to
variation in Federal structure and control.
between the Federal system
getat the same thing by looking at varying relations
The
and particular states are open to too many methodological objections.
Federal system
main variance we are after is that occasioned by changes in the

over time, and that is what should be investigated.

of
Of course, it is also possible to study states longitudinally -- many
wish fragmented
them have changed in ways paralleling the Federal changes,
The question here
budgetary allocation systems and hosts of special programs.
(see
would be the effect such changes have on local organizational patterns

Rowan, 1977 for one example).

here too: simple
But another kind of research should receive some priority
Just what types
descriptive studies of the overall organizational situation.
the Federal and state
of fundings and controls from the environment (especially
And how, administratively,
organizations) do typical districts operate with?
How do state Departments of
do they organize in relation to these controls?
And exactly what
Education organize to deal with the Federal Government?

28
 -24-

funding and control linkages do they have? The present situation is an organi-

We know,
zational mess, and poses considerable problems for clear description.

from impressioniStic evidence, that large numbers of educational administrators

connection
at all leirels now deal primarily with one or another programmatic

with the funding environment: often their job titles reflect such linkages.

Evec, individual Schools have specfll administrators for special externally -

funded programs -- the principal no longer always does the job alone. The

same thing Is true even more at district and state levels. The descriptive

question is: overall, what does this system look like?

CONCLUSIONS

We have been working with speculations that at best have some theoretical

basking, and only infrequently with evidence. As we noted at the beginning,'

this results.from the overemphasis in research on the rise of specific centrali-

whole.
zing fundings or controls, rather than on the-organizatidhal system as a
complexities intro-
In this.paper,ve have been concerned with the organizational
The
duced in the overall system by the present world of fragmented controls.

.problem is simple: suppose we know that each of ten Federal programs standing

along could introduc happiness and virtue in each of ten domains in American

of students, special content areas, etc.). Even then we must

schools

also consider what the organizational introduction of all ten programs might do

A second research problem exists: the moralistic research pressure for

contemporaneity builds An a bias against longitudinal organizational studies. 2

will save
Why study the past when Congress has just funded a new program which

educational souls? Why not study the new program and its impact? The studies

advocated here take a more reflective view. They'compare the past organizational
 -25-

evolution of the system in response to various types of centralization over

time. They cover long periods of time during which crucial changes have gone

on. And in some cases, they compare states or even countries to get at the

crucial variation.

.
In any event, we have argued here that fragmented centralization in

AMerican education has in each subunit .ievel expanded administrative size,

increased differentiation and made it more isomorphic with external structures

and. less with internal needs, and lowered vertical and horizontal internal

coordination oz. substantive educational matters. It has, we believe, generated

a massive middle-level educational bureaucracy, poorly linked with the class-

room world below, little integrated around broad educational policies or

purposes, and organized around reporting to a fragmented wider funding and

control environment. And, we argue, it has become less an&less able to respond

to the local systems of control -- one of the main-loci of legitimated educa-

tional authority in the country. Organizationally, the system would be improved i

either by more authoritative and integrated centralization (which would

explicitly undercut the authority of lower levels),, or by a shift.in funding

organization back to a more local or state system. In the absence of these

changes, a simple attempt to integrate and coordinate programs and funding

at the Federal level? fight in itself lower some of the administrative and

reporting burdens.

30
 -26-

References

of Public
Abrathawitz,'Susan and Ellen Tenebaum. High School '7.7:A Survey
Secondary School Principals. National Institute of Education:Washington,
DC: 1978.
Educational
Berman, Paul and Milbrey McLaughlin. Federal Programs Supporting
Change, Vols. 1-8. Rand Corp:Santa Monica, CA,1975 -78.

Park,
DaVid, Jane. Local Uses of-Title I Evaluations.- SRI International:Menlo
CA, 1978.

Davis, Margaret, Brian Rowan and Anne Stackhouse. "Loose and Tight Coupling
Structure of
in Educational Organizations." In M. Davis, et al., The
Technical Report, Stanford Center for Research
Educational Systems.
and Development in Teaching. Stanford, CA, 1977.

"The Early
Deal, Terrence, JoAnn Intili, Jean Rosaler, and Anne Stackhouse.
Implementa-
Childhood Education Program: An Assessment of its Impact and
Stanford,
tion." Stanford Center for Research and Development in Teaching,
CA, 1977.
Evaluation and the Exercise of
Dornbusch, Sanford and W. Richard Scott.
Bass, 1975.
--T Authority. ,San Francisco, CA:Jossey

Emery, Fred and Eric Trist. "The Causal Texture of Organizational Environ--
ments." Human Relations 18 (February), 1965:21-32.
for the Handicapped.
Kirst, Michael, et al. Financing Educational Services
Council for Exceptional Children:Washington, DC, 1976.

After School Finance Re-

Kirst, Michael. "What Happens at the Local Level
form," Policy Analysis, Summer 1977.
the Local, State, and Federal
Kirst, Michael. The Politics of Education at
Levels. Berkeley:McCutcheon, 1970.
Developments in Improving Education and
Levin, Henry. "A Decade of Policy
Training for Low Income Populations." In R. Haveman,(ed), A Decade of
Lessons.
Federal Anti-Poverty Policy:. Achievements, Failures,'and
New York:Academic Press, 1977.
Administration: A Short Analysis."
March, James. "American Public School
School Review, 86 (February), 1978.

Ambiguity and Choice in Organizations. Her-

March, James.and Johann Olsen.
gen:Universitetsforlaget, 1976.

McLaughlin, Milbrey. Evaluation and Reform: The Elementary and Secondary

1975.
Education Act o2 1965, Title I. Ballinger:Cambridge, Mass.

Meyer, John and Bria Rowan. "Institutionalized Organizations: Formal Struc

Sociology, (September),
ture as Myth and Ceremony." American Journal of
1977.
 -27--

-References

Meyer, John and Brian Rowan. "The Structure of Educational Organizations."

In M. Meyer, et al., EnViranments and Organizations. San Francisco:
Jossey -Bass, 1978.

Meyer, John,'Francisco Ramirez, Richard Rubinson,-and John Boli-Bennett.

"The World Educational Revolution, 1950-1970," Sodiology of Education,
50 (1977):242-58.

Meyer, John, W. Richard.Scott, JoAnn Intili, and Sarah Cole; "Instructional

Dpsiensus and Institutional Consensus in Schools." In M. Meyer, et al.,
EnVtionmehts and Organizations. San Francisco:Jossey-Bass, 1978.

Meyer, John, David Tyack, Joane Nagel, and Audri Gordon. "Public Education
Bureaucratization, 1870 -
as Nationbuilding in America: Enrollments and
1930." American Journal of Sociology, November, 1979.
"Institutional and Techni-
Meyer, John, W. Richard Scott, and Terrence Deal.
cal Sources of Organizational Structure: Explaining the Structure of
Educational Organizations." Conference on.human Service. Organizations,
Center for Advanced Study, Stanford, "CA (March), 1979.

Environments and Organizations. San Francisco:

Meyer, Marshall, et, al.
Jossey-Bass, 1979.

Pfeffer, Jeffrey and Gerald Salancik. The. External Control of Organizations:

A Resource Dependence Perspective. New York:Harper, 1978.

Ramirez, Francisco and Richard Rubinson. "Creating Members:: The Political

Incorporation and Expansion of Public Education." in J.'lleyer and M.
Chicago, IL:
Hannan, eds., National Development'and the World System..
University of Chicago Press, 1979.
Schools." Sociology
Rowan, Brian. "The.Loosely Coupled Control System of
of Education Conference, Asilomar, CA (February), 1979.

Enironment: The Case

Rowan, Brian. "Bureaucratization in the Institutional
of California Public Schools, 1930-1970." In M. Davis; et al., The
Strudture of Educational Systems. Technical Report, Stanford Center for
.

Research and Development in Teaching, Stanford, CA, 1977.

Scott, W. Richard, et al. "The Staffing Structure of Districts and Schools."

Schools.
In E. Cohen, et al., Organization and Instruction in Elementary
Technical Report, Stanford Center for Research and Development in
Teaching, Stanford, CA, 1976.

Terreberry, Shirley. "The Evolution of Organizational Environments." Admini-.-

strative Science Quarterly 12 (March), 1968:590-613.

32
 -28-

References

Organizations in Action. New York:McGraw-Hill, 1967.

Thompson, James.
History of American Urban Education.
Tyack, David. The One Best System; A
'Cambridge, Masa:Harvard University Press, 1974.
Managers:
Tyack, David and Elisabeth Hansot. "Social Movements and Professional
Leadership in
An Inquir7pinto the changing Character of Educational
America." Institute for Research on Educational Finance and Governance,
Stanford University, 1979.
"Educational Reform:
Tyack, David, Michael Kirst, and Elisabeth Hansot.
'Retrospect and Prospect." Institute for Research on Educational Finance
and Governance, Stanford University, 1979.
Englewood Cliffs,
Udy, Stanley. Work in.Traditional and Modern Society.
NJ:Prentice-Hall, 1970.
Loosely Coupled Systems."
Weick, Karl. "Educational Organizations as
Administrative Science Quarterly, 21. (March), 1976:1-19.

New Research in Educational

Wirt, Fred (ed). The Polity of the School:
Politics. .Lexington, Mass:Lexington Books, 1975.

N 33