1

Enemies of the State: Interdependence Between Institutional Forms and the Ecology of the Kibbutz, 1910-1997

Tal Simons Carnegie Mellon University

&

Paul Ingram Columbia University

October 18, 2002

The order of authorship was randomly determined. Yael Parag provided resourceful research assistance. We are grateful to David DeVries, John Freeman, Richard Harrison, Ray Horton, Ira Katznelson, Dan Levinthal, Joel Podolny, Joyce Robbins, Chuck Tilly, Elisabeth Wood, Ezra Zuckerman as well as participants in the Organizational Behavior Seminar at the University of California, Berkeley, the Organizations and Competition Seminar at the University of Chicago, the Economic Sociology Seminar at Princeton University and the Contentious Politics Seminar at Columbia University for helpful comments on this paper.

2

Enemies of the State: Interdependence Between Institutional Forms and the Ecology of the Kibbutz, 1910-1997

The kibbutz was once lauded as an exemplar of the potential of utopian organization. More recently it has been criticized as yet another illustration that socialist arrangements are inferior to capitalist ones. In this paper, we test a number of explanations of “What happened to the kibbutz?” using an analysis of the founding rate of the kibbutz population. We find support for the most popular accounts of the stagnation of the kibbutz. Competition from the capitalist economy is indicated by a strong negative effect of the number of corporations on kibbutz founding. Challenges to the form of community on the kibbutz are reflected in competition from the moshav, a type of agricultural cooperative that is like the kibbutz in terms of production, but maintains more traditional patterns of consumption and family life. We also find a strong negative impact of the Israeli State on the kibbutz. Tension between the new State and the kibbutz in the years immediately following 1948 is well documented, but in subsequent decades the salience of that conflict has faded, and the State is generally characterized as supporting the kibbutz. In contrast, our models indicate that the State’s efforts to substitute for the roles of defense, settlement and absorption of immigration that the kibbutzim filled under the British Mandate affected a drastic shift in the evolutionary trajectory of the population. We use evidence from the kibbutz case to build more general theory about the interdependence between organizations and states.

As much as any twentieth-century organizational form, the kibbutz has captured the

imagination and attention of the public and the research community. Thousands of books, papers

and theses in fields such as psychology, sociology, economics, anthropology, political science,

and education have focused on the kibbutz. Volunteering on a kibbutz has been a rite of passage

for tens of thousands of young people, Jews and Gentiles, from around the globe. The political,

military, and economic history of Israel has given a starring role to the kibbutz, at least until

recently. All of this attention derives from the status of the kibbutz as a great experiment in

utopianism, the extension of the control of a democratic organization to almost all elements of

social and economic life. But the interpretation of the results of the experiment has shifted

radically over time. Once a model that was emulated around the world, the kibbutz now is

criticized and occasionally ridiculed within Israel, and mainly seen elsewhere as yet another

failed socialist model.

In this paper we present the first ecological analysis of the rise and fall of this storied

organizational form. We organize the large set of contending explanations into three categories

according to the alternative models of social control they represent: market, community and

state. Market explanations claim that the kibbutz floundered due to competition from capitalist

organizations. Community explanations address the limits of the kibbutz approach to

consumption and family relations. Explanations focusing on the state highlight the changing role

of the kibbutz as a function of the political transition between the British Mandate for Palestine

and the State of Israel.

We analyze the founding rate of the kibbutz population for evidence of the influence of

these alternatives. We find support for the market and community explanations for the arrested

development of the kibbutzim. Evidence for the former is a strong negative influence of the

number of corporations in the economy on the kibbutz founding rate. Evidence for the latter

comes from negative influences on kibbutz founding from the moshavim, organizations that are

similar in economic activities to the kibbutz, but different in the structure of consumption and

family relations. The above findings confirm dominant theories on the development of the

4

kibbutz. More surprising are the substantial effects of the state on kibbutz founding. In this

regard, we find evidence that the Israeli State exerted a negative influence on kibbutz founding,

both directly by substituting for the defense and immigrant absorption roles that the kibbutzim

held under the British Mandate, and indirectly by promoting alternative forms of settlement.

The interdependence between the kibbutz and the state is also the source of the most

original theoretical implications of our analysis. The interdependencies between market,

community and organization are fairly prominent in organizational theory. Ideas about the

relationship between organizations and states, however, are less developed. Although the state is

never absent from accounts of the rise and fall of organizational populations, it is treated mostly

as an exogenous force that bestows or withholds favor for unknown or unanalyzed reasons. In

contrast, the kibbutz case suggests that the role of the state can be at least partially explained by

considering its autonomy and capacity, and the strength of the organizational populations with

which it interacts. In the next sections we focus on the kibbutz and then distill ideas for a more

general theory of state-organization relations.

INTRODUCING THE KIBBUTZ

The first kibbutz, Degania on the shores of Lake Galilee, was established in 1910 by Jewish

immigrants from Germany, Poland, Galitzia and Russia. It was among a number of

organizational forms that contended to address the challenges Jews faced in Palestine. Its

emergence as an archetype for settlement is attributable to practical and ideological factors.

Degania was an initial economic success, turning a small profit in its earliest years, in contrast

with the typical experience of Jewish agricultural settlements in Palestine until then. It did so

while reflecting basic tenets of the "Religion of Labor" philosophy promoted by A.D. Gordon,

which held that physical labor was a form of art and that moral elevation through work required

the full attention of a worker who was free from hierarchical supervision. Supervising others

was also taboo. So, the kibbutz emerged as the organizational manifestation of an ideological

position that the Jew be “neither the exploited nor the exploiter” (Gordon, 1938: 63).

5

A distinct set of practices delineate the boundaries of the kibbutz as an organizational

form. Kibbutzim differ from other cooperative enterprises in Israel by their permanence of place

and population, and from other rural settlements by their degree of communalism. All kibbutzim

are permanent settlements based on land leased from the Jewish National Fund. Traditionally

they all had common ownership and democratic management of financial affairs, communal

consumption and child care, and a centralized labor allocation system which emphasized

job-rotation and reliance on member (as opposed to hired) labor. Over time some of these

practices have been relaxed.

On other ideological issues there were persistent splits among kibbutzim, and their

respective positions on these issues formed the basis for the grouping of kibbutzim into a number

of political federations. The major questions had to do with the optimal size of a kibbutz, the

appropriate economic activities, how tradeoffs between Zionism and socialism should be made,

and to a lesser extent, what role Judaism should have on the kibbutz. For much of their history,

kibbutzim self-divided into four federations based on their positions on these questions. These

federations encouraged their member kibbutzim to adhere to ideological principles, facilitated

assistance and exchange between their members (e.g., by establishing schools that were shared

by member kibbutzim), and planned the establishment of new kibbutzim. Recently, the salience

of within-population differences has faded, and federations have merged.

Figure 1 displays the number of kibbutzim, and the total population of the kibbutz

movement over time. Two things about this figure are important for this paper. First, both the

number and population of kibbutzim increased throughout their history. This fact is at odds with

popular perceptions in Israel, which see the kibbutzim as a failed population. Many Israelis

believe that if kibbutzim have not failed outright, they have recently “failed by change”, that is,

by adopting practices (such as differential wages) that are incompatible with the accepted idea of

a kibbutz. In truth, outright failure among kibbutzim is extremely rare, with only thirty-eight

failures in the first seventy-five years of the population’s history, and almost all of those among

gestating organizations that had not yet become full-fledged kibbutzim (Parag, 1999). And

6

according to Israel’s Registrar of Cooperatives, only five to seven percent of kibbutzim have

failed by change, even by the strict definition of a kibbutz that he employs.

Second, the growth (in numbers and population) displayed in the figure comes almost

entirely from the founding of new kibbutzim. We have explained that kibbutzim seldom failed.

It is also the case that the growth of individual kibbutzim, once established, was not significant.

(The correlation between the number of kibbutzim and the total number of kibbutz members is

greater than ninety-eight percent). Different federations developed different positions on the

optimal size of a kibbutz, but kibbutzim almost always grew to the size that was deemed

appropriate, and then stopped. This limit on growth is understandable given that kibbutzim rely

on social control, which requires dense social relationships. With dynamics that are determined

almost exclusively by the process of founding, kibbutzim are, to our knowledge, unique in the

empirical literature that documents the rise and fall of organizational populations. That literature

typically combines analyses of founding, failure, and growth to explain the dynamics of specific

populations (e.g., Carroll and Hannan, 2000; Hannan and Freeman, 1989). Whether or not

kibbutzim are a unique population in this sense, their mono-process dynamics are advantageous

for our purposes. They make it possible for us to describe kibbutzim’s rise by analyzing only the

founding rate. This allows us to present results that are particularly transparent as to what was

good and bad for the kibbutz population.

Before proceeding to specific arguments that explain the pattern of population growth

displayed in Figure 1, it is worth considering other strengths and limitations of the analysis of

kibbutz founding rates. As will become quickly apparent, there are many accounts of the rise

and stagnation of the kibbutzim. Mostly these are supported by rich description of specific

kibbutzim (e.g., Lieblich, 1981; Warhurst, 1998), or of historical trends in the kibbutz population

(e.g., Rayman, 1981; Near 1992, 1997). These qualitative approaches are very informative as to

the mechanisms that contribute to the evolution of the kibbutz, but they are limited in their

capacity to demonstrate the linkage between these mechanisms and the population-level

outcomes that describe the overall fate of the kibbutz, as well as in their capacity to differentiate

7

between rival explanations. A small percentage of the relevant research involves quantitative

analysis which tends to be cross sectional (e.g. Don, 1988; Einat, 1997), and very occasionally

longitudinal, but only over a limited time period (e.g., Barkai, 1977; Simons and Ingram, 1997).

These works are often able to identify specific mechanisms of kibbutz change, and to consider

the simultaneous impact of rival explanations. They are unable, however, to identify the

relevance of any specific mechanism across the span of kibbutz history.

Our ecological approach responds to the principle limitations of previous methods,

allowing for the comparison of alternative explanations, on a fundamental population-level

outcome, across both the good times and bad times for the kibbutz. Of course these benefits are

associated with costs. One cost is that in translating the implications of market, community and

state control into covariates for a macro analysis, detail regarding the operation of specific

mechanisms is lost. We do, however, rely on symbiosis between our method and past research—

although we can’t demonstrate the operation of micro mechanisms and processes, we can and do

rely on past research that has. A limitation that is tougher to overcome is that the analysis of

foundings cannot effectively account for change within existing kibbutzim. Here, our approach

is to say what we can about how the kibbutz has changed, and how those changes may be related

to interdependence with the market, community, and state. We believe that the analysis of

foundings (and consequently, population growth) addresses important limitations of past

research on the kibbutz, but we understand fully that foundings don’t tell the whole story.

It is also important to say something about a current debate in the study of the history of

Jewish settlement in Palestine and Israel. A “new history” movement is challenging previous

understandings of Zionist institutions, including the kibbutzim (e.g. Shafir, 1996). The debate

concerns the moral character of those institutions, with the new historians contesting previous

characterizations that they were part of a heroic enterprise, and instead emphasizing their role in

the oppression of Palestine’s Arabs. Needless to say, there is deep and passionate disagreement

in this debate. In this paper we take neither side. Our goal is to provide a social scientific

account of the interdependence between kibbutzim and other institutional forms, and we do not

8

attribute anything normative to the founding of a kibbutz. Fortunately for us, historians from

both camps agree on the basic facts upon which our account rests (e.g., that the kibbutzim were a

mechanism of Jewish settlement and that they came into conflict with the State of Israel). It is

likely that representatives of those two camps would make different interpretations of the results

we present here, but we leave it to historians to resolve such a dispute.

INTERDEPENDENT INSTITUTIONAL FORMS: ORGANIZATION, MARKET,

COMMUNITY AND STATE

The utopian model of the kibbutz, emphasizing organizational control in atypical realms (e.g.,

gender relations), reminds us that organization is one of a number of institutional alternatives for

the governance of social behavior. The others, according to Streeck and Schmitter (1985), are

market, community and state. These institutional alternatives serve well to categorize a range of

specific arguments for the pattern of evolution of the kibbutz.

The Kibbutz and the Market

The relationship between organizations and markets is the explicit focus of transaction cost

economics, which characterizes markets and organizations as substitutes. The argument asserts

that the ideal institutional form to govern transactions depends on the level of market

imperfections. As imperfections increase markets become less attractive and organizations more

attractive (Williamson, 1985).

The general idea that markets are better than organizations for governing certain

activities is the most prominent explanation for the struggles of the kibbutz in the second-half of

the twentieth century. To the question “What happened to the kibbutzim?” most would respond

“capitalism”. The popular wisdom relies on the fact that Israel’s capitalist economy began to

thrive at about the same time that the kibbutzim began to flounder. As the capitalist economy

grew, the argument goes, the alternatives to the kibbutz became more apparent and attractive

(Chafets, 1998 provides a journalistic version of the argument, while recent examples from the

9

scholarly literature include Bloomfield-Ramagem, 1993; Ben-Rafael, 1996; Rosolio, 1999a;

Gavron, 2000; Lapidot, Applebaum and Yehudai, 2000.).

The direct explanation for the negative influence of capitalist organizations on the

kibbutzim amounts to coercive isomorphism (DiMaggio and Powell, 1983). Simons and Ingram

(1997) document the erosion in the 1950s and 1960s of the principle that the kibbutz should

employ only the labor of members, and not hired workers. Their kibbutz-level analysis showed

that this principle was more likely to be dropped or relaxed by kibbutzim that were indebted to

capitalist banks. They argued that debt was used to influence kibbutzim away from a principle

that violated the banks’ capitalist ideology. There is evidence from a number of contexts that

capitalist organizations coerce cooperative organizations to change elements of their structure,

threatening to withhold resources that cooperatives need to survive (Mintz and Schwartz, 1985;

Rothschild and Whitt, 1986, Ingram and Simons, 2000). As past evidence has shown, this

ideological competition affected change within existing kibbutzim, notably the transition to hired

labor, but other revisions of their structure as well. It may have also retarded kibbutz founding

by reducing the expectation of potential founders that they would be able to operate the

organizational form of their choice with autonomy.

Indirect arguments for the infringement of capitalism on the kibbutz are even more

common. According to these arguments, which typically rely on an assumption that capitalist

organizations are more productive than cooperative ones, the opportunities of the capitalist

economy lured potential participants away from the kibbutz. Another expression of this idea is

that the favor towards the kibbutz in the early years of Jewish settlement in Palestine was a

function of necessity. Due to the harshness of the political climate these settlers faced, and the

dearth of employment opportunities, cooperative and communal organizations were among a

small set of viable economic options (Near, 1992). Over time, as the capitalist economy grew in

Palestine and Israel, the salience of this alternative must have increased. Its feasibility must have

also increased, as the success of capitalist organizations depends partly on having other capitalist

organizations to exchange with, and on supporting institutions such as lending and stock

10

markets, which themselves depend on a critical mass of client organizations (Mizruchi and

Stearns, 1994). As the salience and feasibility of the capitalist alternative increased, the relative

attractiveness of the kibbutz could be expected to decrease.

H1: The kibbutz founding rate will be negatively related to the number of capitalist

organizations.

The Kibbutz and Community

The embeddedness research program introduced community into organizational analysis,

partly as a response to the limited institutional scope of transaction-cost economics (Granovetter,

1985). The focus of embeddedness research is on the interpersonal trust and social cohesion that

derive from interpersonal connections. Close interpersonal relations are argued to affect the form

of feasible organization, generally by favoring small autonomous organizations over large,

comprehensive hierarchies (e.g., Uzzi, 1996). Consider also Putnam’s (2000) account of the

historic substitution between community and organizations in the United States, which identifies

organizational contributions to order that go far beyond the management of economic exchange.

Community governance was particularly germane as an institutional alternative to the

kibbutz’s communal consumption and the organizational structuring of family life. This

contrasts with the market alternative to the kibbutz which addressed the perceived limitations of

cooperative production. However, it must be recognized that the kibbutz is an integrated

organizational model, with linkages between the structure of production and consumption. For

example, the ballyhooed principle of gender-equity in productive work was supported by

transferring the duties of childcare to the communal organization, so as to free women from the

traditional constraints that motherhood placed on economic participation. Similarly, the

organizational approach to child-rearing was designed to socialize young participants to the

model of organizational democracy they would be expected to employ as adults. Consequently,

the children were organized into a microcosm of the kibbutz, a “children’s society” with

11

significant rights of self-governance, and the requisite organizational trappings, including

committees, a general assembly operated by the children, and even “children’s farms.”

These practices had some success (more for the socialization of children than establishing

the equality of their mothers), but they also faced controversy and resistance from within the

kibbutz. Starting with the first two children born on the first kibbutz, mothers struggled with

kibbutz-imposed allo-mothering (Baratz, 1954). This struggle was manifested in the

reintegration over time of children into the family home, and in the redefinition of women’s

work roles to place them more proximate to children, but also into service and child-rearing jobs

that were of lower status on the kibbutz (Ben Rafael, 1988). The link between these trends and

community governance is established by research that attributes familiazation and the re-

establishment of traditional work roles for women to the persistent influence of cultural norms

from the wider society (Schlesinger, 1977; Hertz and Baker, 1983; Leviatan, 1985).

Many of the concrete implications of the pressures towards community governance for

the kibbutz are in the form of organizational changes that will not be reflected in our founding

analysis. Most significant is the establishment in the 1970s of the nuclear family, with wedded

parents and children in the home, on the kibbutz. This change has been accompanied by a

subsequent and less dramatic shift in the pattern of consumption—away from the strict

egalitarianism and asceticism of early kibbutz life to allow for more family discretion on

expenditures on items such as food, clothing and entertainment (Talmon, 1972; Barkai, 1977).

The above changes have occurred late in the life of the kibbutz, and do not completely

reconcile the difference between family relations and consumption on the kibbutz and in the

wider community. The possibility remains that the lure of alternative consumption/familial

arrangements may have suppressed kibbutz founding. It is true that many parents cited

dissatisfaction with these elements of kibbutz life as arguments for leaving the kibbutz. It seems

certain that many others would have been discouraged from joining, or founding, kibbutzim for

these same reasons. But how can the temptation to other models of consumption and family be

operationalized in our analysis?

12

The kibbutzim have co-existed with an alternative organizational form, the moshav, that

is an ideal comparison in that it is like the kibbutz in terms of production, but like the wider

community in terms of consumption (Weintraub, Lissak, and Azmon, 1969). Moshavim are

permanent settlements that employ cooperative principles with regard to work, and were

traditionally focused on agriculture (and, like the kibbutzim, have more recently expanded the

scope of their economic activities). Unlike the kibbutzim, the moshavim have always employed

traditional forms of consumption: members live in nuclear families, in their own homes, and

spend their share of the organization’s profits as they choose. Because these organizational

forms are so similar on other dimensions, we use the number of moshavim to test the lure of

alternatives to the kibbutz model of consumption and family structure.

H2: The kibbutz founding rate will be negatively related to the number of moshavim.

Another common argument of wider community norms impinging on the kibbutz

concerns the influence of Jewish immigrants from non-European countries. The early period of

Jewish re-settlement in Palestine was dominated by immigrants from European countries such as

Germany, Poland and Russia. That pattern changed when the Israeli State was established in

1948. At that point, Jews from Asia and Africa began to pour into Israel, often as refugees from

countries like Morocco, Iran and Iraq. In 1949 alone, more Jews immigrated from Asia and

Africa than in all of the previous half-century. The introduction of substantial numbers of Jews

from Asia and Africa has caused notable changes in the culture, politics and economy of Israel,

and has been suggested as one reason for the stagnation of the kibbutz in the last fifty years.

The argument is that the culture of immigrants from Asia and Africa made them less

partial to the kibbutz than immigrants from Europe (Ben-Zadok, 1985; Cohen-Almagor, 1995).

It is certainly true that the recruitment systems of the kibbutz federations, which included

training and socialization of Jewish youth, were focused in Europe, and in other countries with

substantial populations of Ashkenazi Jews, such as Australia and South Africa. It is also true

13

that Sephardic immigrants were prominently represented in alternative settlement types, such as

the moshavim (Lipshitz, 1998), and the development towns (Spillerman and Habib, 1976). But

did they forego the kibbutz? Our research indicates that they did not. Of the ninety-six

kibbutzim founded during the greatest wave of sephardic immigration, between 1948 and 1952,

forty-two of them (44%) had immigrant groups from Asia or Africa (excluding South Africa) as

a substantial part of their founding cohort. This evidence casts doubt on the idea that the

Sephardim scorned the kibbutz. Nevertheless, the idea is prominent, and therefore our models

will test it by including measures of the flow of Jewish immigrants from Africa and Asia.

The Kibbutz and the State

Organizational theorists have not given the same attention to the third interdependence.

There is no explicit recognition in the organizations literature that the type and number of

organizations depends on interdependence with the state. This is not to say that organizational

analysis ignores the state. The state figures prominently in explanations of the rise and fall of

organizational populations, as a source of endorsements and surety (Barnett and Carroll, 1993;

Baum and Oliver, 1992; Ingram and Simons, 2000; Wholey, Christianson and Sanchez, 1992),

defining the rules of competition (Barron, West and Hannan, 1994; Dobbin and Dowd, 1997;

Wade, Swaminathan and Saxon, 1998; Silverman, Nickerson and Freeman, 1997), as a source of

isomorphic pressure (Carroll, Goodstein and Gyenes, 1988) and as a source of environmental

change (Dobrev, 1999). The theme of all this work, however, is the dependence of organizations

on the state, not the interdependence between the two forms.

Although there is no comprehensive theory of state-organization interdependence, there

are two ideas, one new and one old, that serve as promising building blocks. The new idea is

that the strategy the state employs for its interactions with organizations depends on its relative

strength, vis-à-vis those organizations. Dobbin and Sutton (1998) argue that weak states may

employ a range of indirect mechanisms, including their influence over legitimacy, to control

organizations. The old idea is Durkheim’s forceful claim that states rely on symbiosis with

14

organizations, due to the basic inadequacy of the state to promote cohesion. In the conclusion to

Suicide (1951: 389) he describes the state as being “as intrusive as it is impotent. It makes a

sickly attempt to extend itself over all sorts of things which do not belong to it, or which it grasps

only by doing them violence.” In the preface to the second edition of The Division of Labor in

Society (1984: liv), he calls a modern state that governs unorganized individuals “a veritable

sociological monstrosity”, and claims “[a] nation cannot be maintained unless, between the state

and individuals, a whole range of secondary groups are interposed.”

The concept of symbiosis effectively captures the relationship between the kibbutzim and

the state until 1948 - the British Mandate. Britain occupied Palestine in 1917, and in 1920 was

granted a mandate to rule there by the League of Nations. From the beginning British rule

struggled (McTague, 1983: 164). The frustrations of persistent Arab-Jewish conflict combined

with poor prospects for economic gain by the British resulted in a de facto abdication of many of

the responsibilities of governance (Migdal, 1988; 2001; Shalev, 1992; Biger, 1994; Segev,

2001). The kibbutz federations and their member kibbutzim stepped into the institutional void

left by the British (Grinberg, 1993), acting in response to collective problems of the Jewish

population of Palestine (the Yishuv). According to Near (1992: 2), there were “three functions

which the kibbutzim had always taken on themselves – absorption [of immigrants], settlement,

and defense” which were almost universally seen as the priorities of the Yishuv. The most

significant element of Yishuv defense during the mandatory period was the Haganah, which was

the first and largest of a number of underground armies that defended the Yishuv. Largely,

kibbutz members staffed it and kibbutzim were the sites of its training and weapons storage.

Defense and settlement were closely tied for the Jews of Mandatory Palestine. In describing the

rationale for settling in a specific location, a kibbutz member observed that “the mere existence

[of the kibbutzim] was determined, sometimes as the primary consideration, based on their

ability to defend the ‘home front’ of the Yishuv” (Eilat, 2000: 157).

As for immigration absorption, the kibbutzim actively aided immigrants, be it by finding

work in one of the traveling work groups that often became the seeds for new kibbutzim or by

15

directing the immigrants to settle on an existing kibbutz (Aharoni, 1991). The kibbutzim were

also a keystone of the system of workers’ schools, hosting and staffing many of the schools. By

the end of the Mandate, these schools educated half of the children of the Yishuv, and more than

seventy percent of the children of new immigrants.

The linkage between kibbutz service on these dimensions and their founding is illustrated

by the kibbutzim’s response to the outbreak of the Arab Revolt of 1936. This revolt included a

general strike by Arabs that paralyzed Palestine’s government and public transportation services

and violence, including assassinations and terrorist attacks on cities, buses, and public facilities.

The Yishuv response included a rapid increase of militarization, including an expansion of the

Haganah, and modernization of its weapons (Horowitz and Lissak, 1978). Demonstrating a

conscious symbiosis, many of the new weapons were provided by the mandatory government in

an effort to leverage its own defense resources. The Yishuv reaction also included a deliberate

increase in the establishment of new settlements, to reinforce previously isolated Jewish

settlements, and to establish initial footholds in areas that were considered to be strategically

important (Near, 1997; Rozenman, 1997). Throughout the disturbances of 1936-1939, forty-

three settlements, mostly kibbutzim, were founded in remote and sparsely populated (by Jews)

areas, such as the Western Galilee (Orren, 1978; Weintraub, Lissak and Azmon, 1969).

H3: Under the British Mandate, kibbutz founding will be greater when there are greater

opportunities to provide defense, settlement and absorption.

The symbiotic relationship the kibbutzim had with the British Mandate did not survive

the transition, in 1948, to the Israeli state. Prime Minister David Ben-Gurion and other early

leaders of Israel pursued a policy of statism1 which called for the centralization in the state of

1 The Hebrew word mamlachtiyut, which we translate as statism, has semantic overtones that cannot easily be reflected in English. Mamlachtiyut implies “something less precise but also broader and more powerful than the state” (Cohen, 1987: 203), hinting at “the grandeur of the biblical kingdom of Israel” (Near, 1997: 184, fn. 31).

16

previously dispersed mechanisms of governance (Cohen, 1987). National objectives that had

previously been pursued by the kibbutzim, and other organizations such as the Histadrut (a

coalition of cooperative organizations that included the kibbutzim) were seen as within the

legitimate domain of the state (Horowitz and Lissak, 1978). As Ben-Gurion put it, “all national

initiatives will be exclusively controlled by the State’s apparatus” (Ben Chorin, 1983:8).

The kibbutzim were among the first organizations that Ben-Gurion attempted to

subordinate to the state, and they became the main targets in his struggle to establish the state's

authority (Etzioni, 1966; Rosolio, 1999b). Rhetorically, the shift in power away from the

kibbutzim to the state is illustrated by the use of honorific terms such as pioneer and volunteer,

which Ben-Gurion and others started applying to civil servants (Near, 1997: 184-185), thus

shifting prestige from kibbutz members to state bureaucrats (Cohen, 1987: 218). Also

illustrative is a speech to the Israeli parliament in 1950 where Ben Gurion claimed to be

“humiliated and ashamed” by his previous association with the kibbutzim. (He had been a

kibbutz member and a leader in the kibbutz movement during the Mandatory period).

Attacks on the kibbutzim were not only rhetorical. They included absorption by the state

of their military and education systems. The independent workers' education system and the

pioneering youth movements, two critical recruitment and socialization mechanisms for the

kibbutzim, were incorporated into the state-controlled educational system. The Haganah and the

Palmach (an elite element of the Haganah, also dominated by the kibbutzim) were disbanded,

and the Israel Defense Forces were formed (Yanai, 1982). Similarly, the strategic planning of

settlement locations was undertaken by the state (Ben-Zadok, 1985).

Why did the state move against the kibbutz? The perception of many was summarized

in a kibbutz member's recollection that Ben-Gurion "feared our strength, so he had to break us

up. He didn't want any strong autonomous organizations, because he considered them a threat to

the new state" (Lieblich, 1984: 119). This threat even manifested itself in rumors that some

kibbutzim were secretly accumulating weapons and sustaining an underground for the purpose of

overturning the State. Kibbutz members had their fears also, and some resisted giving up their

17

weapons, claiming: “we won't be protected because we present some sort of a political

opposition (Eilat, 2000: 158).” Members also realized that the rhetorical attacks and structural

changes imposed by the new state undermined the roles of defense, absorption and settlement

that the kibbutzim had played under the British Mandate (Ben Chorin, 1983; Tzur, 1984).

H4: Under the State of Israel, there will be a lesser effect on kibbutz founding of opportunities to

provide defense, settlement and absorption.

While the state gained substantial organizational apparatus from the kibbuz, it also

developed new mechanisms to govern. A very important example is the ‘development towns’,

peripheral, non-agricultural settlements established by the state in response to the need to provide

a home to a large number of immigrants and to occupy strategic areas of the country (Spilerman

and Habib, 1976). The development towns were central to the state’s effort to substitute for the

settlement and absorption efforts previously provided by kibbutzim. Most of the forty towns that

currently exist were established between 1948 and 1957 in locations that at the time of their

establishment were remote and rural, often close to the 1948 borders of Israel. These were

similar to the areas that up to that point had been settled almost exclusively by kibbutzim

(Weintraub, Lissak and Azmon, 1969). Immigrants were directed by the state to these towns and

many of them never moved away.

H5: The kibbutz founding rate will be negatively related to the number of development towns.

Interdependence between market, community and state

So far we have considered the dyadic relationships between the kibbutzim and the other

institutional alternatives, but we recognize that those alternatives are themselves inter-related. It

has been argued that the modern capitalist firm depends heavily on institutions provided by the

state (Miliband, 1969) and there is evidence that the Israeli State established policies that

18

smoothed exchange between independent organizations and thereby improved their life-chances

(Ingram and Simons, 2000). The Israeli State also influenced the growth of the moshavim, for

example by directing a disproportionate amount of the substantial external funds that were raised

in the early years of the state to moshavim (and development towns) as opposed to kibbutzim

(HaKibbutz HaMeuchad, 1967: 79). To the extent that the Israeli state promoted increases in the

number of corporations and moshavim, our analytic strategy may understate the total influence

of the state on the kibbutzim, and overstate the relative influence of market and community.

METHODS

Foundings in a population are counts of events over a discrete period (typically a year). Poisson

regression is often an appropriate method for modeling dependent variables which are event

counts (King, 1988). With Poisson regression as the starting point, we considered three

additional methodological concerns to choose a modeling strategy.

The first concern was which unit of analysis was appropriate. Organizational foundings

are often analyzed at the level of the country, but are sometimes analyzed at the level of sub-

country regions (e.g., Swaminathan, 1995). For the analysis of kibbutz foundings, there are a

number of factors that favor regions as the unit of analysis. First, the kibbutz federations

sometimes made founding decisions with consideration to what they perceived as the optimal

number of kibbutzim in a given geographic region (Katz, 1995; Rayman, 1981). So, we would

expect that the influence of the number of existing kibbutzim on the founding rate would be

more pronounced in smaller regions than in the country as a whole (exploratory analysis

supported this expectation). Second, a key resource for kibbutz founding, land suitable for

settlement and agriculture, is distributed unevenly throughout the country. Modeling the

availability of land for new kibbutzim requires measures of the amount and type of land and of

existing settlements, particularly the moshavim and development towns, that may compete for

that land. The only meaningful way to do this is by operationalizing those variables in

geographic regions that are small enough to be reasonably consistent as to the type of land they

19

represent. Third and related, the area and type of land available for settlement changed during

the period we studied as the borders of Israel changed. This is difficult to deal with if the

country is the unit of analysis, but much easier with smaller regions, because they can be added

to or dropped from the analysis depending on their feasibility for settlement.

Given these arguments, we decided to use regions as our unit of analysis, and our

dependent variable was defined as the number of kibbutzim founded in a region in a year2. The

next step was to decide what those regions should be. Here we were guided by the practical

demands of coding data at the regional level. Recording borders, settlements, land quality and

weather conditions at the regional level required using numerous maps. So, we defined our

regions using the most common map-grid used in maps of Mandatory Palestine and Israel. This

grid divided the territory into ten-by-ten kilometer squares which became our regions. In the

early stages of our analysis we used a different grid because it was superimposed on the first map

we found showing all of the kibbutzim. Results of that early analysis were comparable to results

using the ten-by-ten kilometer grid, so we do not believe that our analysis is biased by the

particular grid we used to define the regions.

The second methodological concern emerges from the use of multiple regions as our unit

of analysis. This approach produces a panel data structure, with repeated annual observations of

each region. A potential problem is that observations of the same region may not be mutually

independent. When they are not independent, conventional Poisson models (and mixed-Poisson

models such as the negative-binomial model) are inappropriate because they are based on the

assumption of independence (Guo, 1996). In response to this problem we added a gamma-

distributed region-specific random effect to our Poisson models (Guo, 1996).3 This approach, 2 A few readers have mistakenly assumed that a regional analysis is equivalent to multiplying the number of observations in our analysis by the number of regions, and thereby artificially increasing the statistical power of our models. While the regionalized analysis increases the number of units that may experience an event, it does not increase the number of events. Instead, the aggregate number of events are divided among units, and the overall risk of an event does not increase or decrease. This is the spatial equivalent of temporal spell-splitting in an event-history analysis—a refinement of the unit of analysis that allows finer specification of some covariates, but should not (and in our analysis, does not) affect the significance of covariates that do not vary across units. 3 An alternative approach is to add a fixed effect for each region. The fixed-effects approach requires fewer assumptions than the random-effects approach, but cannot be used if the explanatory variables include some that do

20

called a negative-multinomial model, makes explicit allowance for interdependent observations

by modeling unobserved influences shared by all the counts of a region.

The third methodological concern comes from another assumption of the Poisson model,

one of equality between the conditional mean and the variance of the dependent variable. Often,

as in our data, the variance exceeds the conditional mean, resulting in what is called

overdispersion. The negative-multinomial model actually accounts for overdispersion. It is

essentially a negative-binomial model (a variant of the Poisson model which is commonly used

to deal with overdispersion) with an overdispersion parameter that varies across regions. We

also estimated a similar, but more restrictive version of this model, the negative-binomial model

with random effects proposed by Hausman, Hall and Griliches (1984). That model starts with

the negative multinomial model and makes additional assumptions about the distribution of the

parameters of the gamma distribution, which effectively layers a random group effect across a

negative-binomial model. Results for the two models were almost identical, so in our analysis

we used the less restrictive negative-multinomial model:

where 8ij is the predicted foundings in region i in year j ,2i is a gamma-distributed random effect

for region i, and xij∃ represents the vector of independent variables and coefficients for region i

in year j.

Data and Variables

We built the data from a number of historical and archival sources. Most useful were a large

number of maps that identified the location and founding dates of kibbutzim, moshavim,

development towns, and other towns; the boundaries of the Jewish population of Palestine and of

not vary by region. In our analysis, land quality and rain are important variables that do not vary by region, which influenced us to choose the random-effects model. Fixed-effects models run on our data without the rain and land quality variables produced results that were comparable to the random-effects models reported here.

λ θ β θ ( ) = exp (x ) ,ij i ij i

21

the State of Israel; and land type and amount of annual rainfall. The Statistical Abstract of Israel

(various years) and comparable volumes compiled under the British Mandate provided data on

population, immigration, and number of corporations.

In ecological analyses, counts of the number of organizations of a type are called

densities. Following our hypotheses, our models include densities of corporations (H1),

moshavim (H2), and development towns (H5). Because moshavim and development towns

compete with kibbutzim for suitable land on which to settle and farm, we operationalize those

densities at the local level as the count in a given region. (Comparable, but less powerful results

are obtained when aggregate densities are used for these variables). Corporation density is only

available in aggregate, and in any case, the arguments supporting hypothesis 1 suggest a society-

wide competition between kibbutzim and corporations. The density variables, like all time-

changing variables in our analysis, are updated at the beginning of each year.

Following the norms of ecological analysis, we also include the density of the kibbutz

population. A population’s own density is argued to represent processes of legitimation and

competition (Hannan and Freeman, 1989). Legitimacy increases at a decreasing rate with

density, and increases a population’s founding rate. Competition increases at an increasing rate

with density and decreases the founding rate. These two processes combine to support a

prediction of a non-monotonic effect of density on founding, with founding increasing and then

decreasing with density. Kibbutz density and its square are included in our models to capture

this non-monotonic effect. As noted, kibbutz-federation policy and exploratory analysis

indicated that the effects of kibbutz density were strongest at the regional level so we measured

kibbutz density at the regional level. Following convention, we logged the first-order kibbutz

density measure (Barron, West and Hannan, 1994).

To measure opportunities to provide defense, settlement and absorption (H3) we use two

variables that reflect the near unanimity that the greatest challenges to Jewish society in Palestine

and Israel came from a rapidly growing population, and the threat of political violence (e.g.,

Cohen, 1987; Orren, 1978; Rayman, 1981). The first is Jewish population, the number of Jews

22

in the country at the start of a given year. The second is political violence, an indicator variable

coded one in years of significant political violence or tension between the Jewish population and

surrounding Arab populations4.

We interacted these two variables with a variable that is coded one in years when the

State of Israel is in existence. In this manner our models can reflect differential effects of

governance opportunities under the British Mandate and the State of Israel (H4). This

dichotomous operationalization is meant to reflect that there was a sea-change for the kibbutzim

when the State of Israel was established, a transformation of the state from a symbiotic partner to

a competitor. It is also possible that the increases over time of the size of the British Mandate

and State of Israel affected kibbutz founding. To investigate this possibility, we will also present

models that operationalize the changing state continuously with a variable we call government

size, which is the total number of police and state bureaucrats divided by 100,000 for scaling

(reliable statistics were not available for the period we study for other relevant state participants

such as soldiers and prison staff).

Our models include two variables designed to reflect the attractiveness of a region for

agriculture. Land quality is a continuous variable ranging from 0 to 3. It was created by

measuring the amount of various types of land in each region from appropriate maps. Relying

on the maps’ assertions about the suitability of the land-types for agriculture, in consultation with

an Israeli agronomist, we assigned a value of three to land types that were most appropriate for

agriculture (e.g., the coastal plain and the Yisrael Valley); two was given to land of mixed

quality (e.g. the Galilee Hills); one represented land that was not well suited for agriculture (e.g.,

the Judean Hills); and zero represented desert. Since this variable was based on land area of the

various types, it also reflects the fact that regions had less opportunity for kibbutz settlement if

4 For the mandate period we relied on Haganah (1950), which categorized as representing heightened political violence and tension the years of the first (1920-1921) and second (1936-1939) Arab revolts, the period of World War II, and the years from the end of World War II to the end of the War that began with the UN partitioning of Palestine in 1948. For the Israel period, we take the periods of the Sinai War (1956), the Six-Day War (1967), the 1973 War, the Lebanon War (1982), and the first Palestinian Intifada (1989) to represent heightened political violence and tension.

23

they were less than 100 square kilometers because they spanned borders or bodies of water. The

rainfall variable represents the average annual rainfall in centimeters in the region. Finally, we

include a moving average of the percentage of Jewish immigrants over the preceding five years

that were from Asia and Africa. This is intended to control for the argument that the kibbutzim

floundered due to the changing ethnicity of immigrants. Table 1 presents correlations and

descriptive statistics for these variables.

Results

Table 2 presents nested negative-multinomial models of kibbutz founding. Model 1 is a

base model, to which model 2 adds the interactions with the state variable. Chi-squared tests

indicated that model 2 improves over model 1, so we focus on it in this discussion of results.

In model 2, corporation density has a negative coefficient, supporting H1 that the kibbutz

suffered from competition from the capitalist system. Moshav density has a negative effect, as

predicted by H2. The percentage of immigrants from Africa and Asia has a negative, but non-

significant coefficient. The Jewish population and the political violence variables both have

positive coefficients. The interactions of those variables with the State of Israel variable have

negative coefficients. Wald tests indicated that statistically the coefficient of the main term for

political violence was equal in magnitude to the coefficient of its interaction with Israeli State.

For Jewish population, the main effect was larger than the interaction. Together these

coefficients support H3 and H4: under the British Mandate the kibbutzim were founded in

response to opportunities to provide defense, settlement and absorption of immigration to the

Yishuv. Under the State of Israel the kibbutz founding rate was much less responsive to those

factors. Finally, as predicted by H5, development town density has a negative impact on the

founding rate.

24

Given our arguments, it may seem surprising that the main effect of the State of Israel

variable is positive (although only weakly significant). It must be remembered that the creation

of the State of Israel brought major changes in the institutions that governed the Jewish

economy, vastly improving the legal environment surrounding economic exchange. These

changes apparently helped the kibbutzim, as they did many other organizations. This overall

effect of the state on the Israeli economy is distinct from the rivalry between State and kibbutz

on other dimensions of institutional governance.

The other variables in the model yield interesting results. Kibbutz density has the non-

monotonic effect on founding predicted by the theory of density dependence, with founding first

increasing, and then decreasing, with density. Town density was not significant. This provided

greater confidence that the results we find for development-town density are attributable to the

role of those latter settlements in the state’s effort to govern, and are not confounded with some

other influence of non-agricultural settlements on kibbutz founding. Finally, the level of rainfall

had a positive effect on the founding rate, and kibbutzim were significantly more likely to be

founded on land that was good for agriculture.

Models 3 and 4 replicate models 1 and 2 exactly, with the only difference being the

substitution of continuous government size for the dichotomous Israeli State variable. The

substantive results in models 3 and 4 are in all respects comparable to those in models 1 and 2.

The only notable difference is that in the latter models, the negative coefficient of the percentage

of immigrants that came from Asia and Africa obtains significance. This indicates some support

for the idea that the kibbutz founding was reduced by the increasing representation of Sephardim

among Jewish immigrants to Palestine and Israel.

25

These analyses treated all kibbutzim the same. It is possible that kibbutzim had

differential relations with the state as a result of different political affiliations. In supplementary

analysis (available from the authors) we tested that possibility. That analysis shows that there

were some interesting differences in the founding processes of kibbutzim of different political

federations. Generally, however, the processes described above affected all or most kibbutzim.

Implications for Theory: The Interdependence Between States and Organizations

The implications of the kibbutz case for theory concern the interdependence between

organizations and other institutional forms. Although existing literatures examine the

interdependencies between organizations and markets, and organizations and community

governance, the interdependence between organizations and states has received less theoretic

attention. Three questions regarding the relationship between organizations and states emerge

from the experience of the kibbutz. First, why is there sometimes symbiosis between

organizations and states, as with the kibbutzim and the British Mandate? Second, why is there

sometimes competition in that relationship, as with the kibbutzim and the State of Israel? Third,

what explains the specific actions, among the range of alternatives available for both symbiosis

and competition, that the state takes to instigate its relationship with organizations? In this

section we reflect on the kibbutz case, and other relevant research, to suggest answers to the

above questions. Our goal is to develop some grounded theoretical ideas to be used in future

efforts to integrate the state into organizational theory.

To start, we observe that some organizations contribute to the institutional framework

that regulates and thus facilitates the collective pursuit of basic human needs (society). Such

organizational governance can occur if the organization is a federation constituted by other

26

organizations (Ingram and Simons, 2000). It can also occur if the organizations have supplier-

customer relationships to organizations or individuals that pay for governance, as do the major

accounting firms and their clients (Strange, 1996) or the Sicilian mafia and its customers

(Gambetta, 1993). It may also occur when organizations represent specific ideological interests

that they wish to promote, as did Nonconformist Churches of 18th century Wales which provided

insurance against hardship, education, and care for the sick and dying, but also promoted

temperance and the use of the Welsh language (Hechter, 2000). Regardless of their specific

structure and motivation, organizations that seek to provide governance for society are brought

into overlap with the state, which typically survives and thrives depending on its success in

providing that governance (North, 1981).

To date, the common understanding of this overlap has been that it serves the state. This

may occur either because organizations can do things that the state simply cannot (Durkheim,

1951, 1984) or because they can do these things more efficiently than the state (Hechter and

Kanazawa, 1993). Whether by necessity, or for expediency, it is clear that the British Mandate

relied on the kibbutzim. It is also true that kibbutzim were rewarded for their service to the

Yishuv. These rewards included status and material resources from various granting agencies.

For example, for the years 1940 through 1945 almost 50% of the Jewish Agency’s total

investments for settlement were directed to kibbutzim (Rozenman, 1997: 291). The rewards to

the kibbutzim also included the ideological satisfaction of implementing their utopian model for

Jewish society in Palestine.

The competitive relationship between the kibbutzim and the Israeli State is much more

difficult for existing theory to explain. The historical explanation we cited, that state leaders

perceived the kibbutzim as a political threat, suggests that there are limitations to the state’s

27

reliance on subordinate organizations to provide governance. That reliance may curtail the

autonomy that is required for the state to manage the institutional framework effectively

(Rueschemeyer and Evans, 1985; Migdal, 1988; Grinberg, 1993). The state may become

captured by specific organized interests, and unable to sufficiently service community-wide or

societal interests. As the kibbutz case illustrates, organizations may also be rivals to the state,

threatening to usurp its role and displace the individuals that populate it.

The Israel-kibbutz relationship is not the only instance of state-organization competition.

Tilly (1985: 173) observed of the response of state-building kings in sixteenth- and seventeenth-

century Europe that “disarming the great stood high on the agenda of every would-be state

maker.” The “great” were Lords, subordinate to the King but still in control of vast

organizational resources. Strange (1996) presents other examples, including the suppression of

Knights of Templar in France in 1307 and the dissolution of Catholic monasteries by Henry VIII

in England of the 1530s.

Among contemporary examples, Strange’s (1996) account of the Italian mafia stands out.

This case parallels the kibbutz-case in that it includes periods of both symbiosis and competition

between the mafia and the state. In the period of symbiosis, “the state in effect delegated to the

mafiosi the functions of social intermediation and arbitration, protection of property and persons

and the preservation of order (p. 115).” Starting in the 1970s, however, the delegation of

authority to the mafia stopped, politicians began covering up links to the mafia, and the state

attacked the mafia through the legal system. What caused this radical shift in state-mafia

relations? According to Strange, the problem was that the mafia gained power relative to the

state as a function of a quantum leap in the mafia’s financial resources that resulted from the

internationalization of their criminal activities. In terms of the argument we are developing, the

28

mafia came to threaten the autonomy of the Italian State, which in response discarded its

symbiotic relationship to the mafia and initiated a competitive one.

Other contemporary examples come from the same geographic area that we have studied.

The Palestinian Authority currently struggles with Hamas, a religious organization that provides

education, social services, and engages in military action. In recent years, Israel has had

increasing conflict with a religious organization called Shas. Shas operates a school system that

parallels Israel’s public system and provides a wide range of social services. Recent

commentary has echoed the rhetoric of the kibbutz-State conflict by referring to Shas as an anti-

mamlachtiyut organization.

The above arguments suggest a way for organizational theory to endogenize the state, by

characterizing the state as an actor that pursues its interests through systematic interaction with

organizations. The conceptualization of the state as an actor has already proven useful as a

building block for better understanding of the state’s participation in a number of realms (e.g.

North, 1981; Skocpol, 1985; Evans, 1995). The idea that state-organization relations depend on

the latter’s threat to the former could be used to predict symbiosis or competition between the

state and a number of organizational populations, including political parties, banks, social

movement organizations, trade associations, churches, universities, and newspapers. At the same

time, we caution that the full influence of the state cannot be captured by negative or positive

coefficients in models of organizational founding and failure. We see utility in the treatment of

the state as an actor in a system of interdependent organizations, but this simplification must not

be an excuse for ignoring the complexity of the state’s influence on organizations.

The state’s repertoire of influence is broad, and contains unique elements. Most

definitions of the state identify its monopoly over the legitimate use of violence, and indeed,

29

violence is brought to bear on organizations that threaten states. As we write, the Palestinian

Authority and Hamas are engaged in a back-and-forth that has resulted in a number of deaths in

the last week. The leader of Israel’s contemporary organizational rival Shas was recently

imprisoned by the State. His followers claim the charges were trumped up.

We think that analysts and organizational participants should be conscious that the threat

of violence always lurks in the background of state-organization relations. However, recent

work has rightly emphasized another tool for state influence on organizations that seems more

common, the control over legitimacy. An emerging literature identifies the state’s efforts to

implement its policies not by force but by encouraging organizations with more direct control

over participants to act according to the state’s preferences. States such as the United States,

which seem secure in their autonomy, may be weak in terms of their capacity to implement

policy (Dobbin and Sutton, 1998). Even the rule of law, which is traditionally linked to the

state’s potential for violence, may in practice be more reliant on its ability to employ normative

forms of influence (Edelman, Uggen and Erlanger, 1999).

The kibbutz case indicates the competitive punch of the state’s control over legitimacy.

The state’s displacement of the kibbutzim, from political status if not from the land, was non-

violent. And although some kibbutz members complained about the state’s absorption of their

institutional apparatus, there was very little transgressive resistance. Even the demilitarization of

the kibbutzim, which kibbutz members understood as a deep strike against their political

position, ultimately occurred without violence (Eilat, 2000). Would the kibbutzim have acceded

so quietly to another organizational form that tried to establish governance over the Yishuv? Or

to a state that they did not perceive as legitimate? We say no.

30

The Israeli State’s legitimacy-based attacks on the kibbutzim are also notable for the

persistence of their negative effects. Anti-kibbutz rhetoric and activity has never been as intense

as in the immediate years after the formation of the State. Indeed, in the subsequent fifty years

many acts of the State appear to have helped the kibbutzim. An obvious example is State-

brokered restructuring of kibbutz debt in the mid-1980s, which may have saved many kibbutzim

from failure. But make no mistake, the organizations that the State saved in the 1980s are

substantially different than the ones it attacked in the 1950s. The kibbutzim, and their ability to

threaten the State, were reduced in that earlier interaction, and they have never recovered. As the

secretary of the Kibbutz Artzi federation recently commented, “We were subjected to

demonization. No matter how hard we try, we can’t shake it off (Levy-Barzilai, 2000: 9).” The

pattern of the state’s actions towards the kibbutzim reinforces the emerging idea that for

organizational analysis, the state’s control over legitimacy may be as important, or more, than its

monopoly over violence.

The above discussion links structural factors (specifically the state’s autonomy relative to

organizations and its capacity to impose its will on those organizations) to the pattern of state-

organization relations. We close this section with a caution against over-emphasizing the

structural determinants of state-organization relations. As with organizational competition,

political contention is influenced by structural factors, but also relies heavily on dynamic

interaction between the parties (McAdam, Tarrow and Tilly, 2001). For example, we believe

that the interaction between the Israeli State and the kibbutzim could have turned violent if

different perceptions, interpretations and characterizations of the state were made at a given

moment, in social settings that allowed them to spread. Our case allows us insight into the

influence of structural factors on state-organization relationships because we have variance in

31

those factors with the transition from the British Mandate to the State of Israel. But, we have

only one set of dynamics. We encourage comparison of cases that are similar in terms of the

structure of state autonomy and capacity vis-à-vis organizations to distill insight into the role of

competitive and contentious dynamics on the relations between organizations and states.

Conclusion on the Kibbutz

In this final section, we reflect on the implications of our results for the perception of the

kibbutz as a failed experiment in utopia. Our results indicate support for some of the most

popular explanations for the arrested ascension of the kibbutzim. The kibbutz did indeed suffer

competition from the growth of the capitalist economy, as indicated by the competitive impact of

corporation density. And the competitive impact of the moshavim suggests that many potential

kibbutz participants were dissuaded by the kibbutz’s control over consumption and family life.

One other favorite explanation which we doubted, that the kibbutz was harmed by the increased

representation of African and Asian Jews among immigrants to Israel, received limited support

from our analysis.

Our results also support the idea that the kibbutz suffered from the emergence of the State

of Israel. The tensions between the State and kibbutzim in the period immediately following the

formation of Israel are well known. In subsequent generations, however, the impact of that early

conflict has been de-emphasized and many Israelis feel that the kibbutzim have received

preferential treatment from the state (e.g. Yiftachel and Tzefadia, 1999). While the state has

acted in ways to sustain some kibbutzim our results indicate that kibbutz founding was

substantially reduced when the Israeli State assumed key activities from the kibbutzim and

promoted the development towns as alternatives for settlement. Whatever advantages or

32

disadvantages the kibbutzim may now have relative to other settlement and organizational forms,

the balance would have been more in the kibbutz favor had not the Israeli State acted to

permanently change the trajectory of kibbutz evolution.

Recall that our founding analysis does not capture changes within existing organizations.

However, we do know some things about how life on the kibbutz has changed, and they are

consistent with our premise that the kibbutz has been reduced in the post-state period. In the last

fifty years, kibbutzim have moved away from once-sacred organizing principles, because of

external pressure from the state or capitalist organizations (Rayman, 1981; Simons and Ingram,

1997). Over the same period the number of kibbutz members in the Knesset has steadily fallen.

Members’ attachment to their kibbutzim has fallen, and as a consequence, their likelihood to

leave has increased (Rosner, et al., 1990). In the mid-1980s many kibbutzim experienced a

crisis of indebtedness from which they emerged more beholden to the state and to banks. Crime

rates on the kibbutz have increased, while members’ confidence that the kibbutz can deal with

crime internally has decreased (Ben-Rafael, 1997). So, while we do not claim that the founding

rate is the only way to measure the fate of the kibbutz population, we believe that comparable

analyses of other measures would lead to similar conclusions.

Given that all three of the institutional alternatives of market, community and state seem

to have impinged on the kibbutz, it is natural to wonder about the relative magnitudes of their

affects. The negative-multinomial model is multiplicative, so the magnitude of effects must be

understood as the effect of a given variable on the founding rate as determined by other

variables. This measure is called a multiplier of the rate. Figure 2 shows multipliers of the

kibbutz-founding rate for the variables associated with the market and state institutional

alternatives in the post-1950 period. Those multipliers are graphed on a logarithmic scale since

33

they become very small. (The impact of moshav density is not shown because it is much smaller

than that of the other variables. In the post-state period, the kibbutz founding rate fell about 18%

due to increasing competition from the moshavim). The figure shows that the negative influence

of the increase in the number of corporations is substantial, reducing the founding rate to a small

fraction of what it would have otherwise been. But for most of the post-state period, the negative

influence of state competition is even larger.

Figure 2 casts doubt on the common interpretation that the outcomes of the kibbutz

population indicate the inferiority of utopian socialism to capitalism. It shows that the contest

was not only the kibbutz vs. capitalism, but also the kibbutz vs. the State. How would the

kibbutz have faired if the Israeli State had not competed with it, directly and indirectly? Our

models allow some informed speculation on this question. It is possible to project the growth of

the kibbutz population as it might have been had the State pursued a different strategy, or if a

different type of state had come to power in 1948. We conducted a simulation of the growth of

the kibbutz population absent the competitive effects of the State (the negative interactions

between the state and the population and political violence variables, and the negative influence

of the number of development towns). The simulation takes all other variables at their historical

values, and estimates their impact on the founding rate using the coefficients from our best fitting

model, model 2. Figure 3 presents the mean kibbutz density, and the ninety-five percent

confidence interval of that estimate, from 100 iterations of the simulation.

The simulation indicates that the kibbutz population would have been substantially larger

by 1997 were it not for State competition (616 kibbutzim, compared to the 267 that actually

existed). Of course, if the State had been different, and if the kibbutz population had been larger,

other influences on kibbutz founding would have been different also. For example, had the

34

kibbutz retained control over the education of immigrants, and the resulting ideological

hegemony, the resistance of the Sephardim to the kibbutz may have been deflated or reversed.

As for the capitalist organizations, the arguments supporting hypothesis 1 can be reversed to

suggest that more kibbutzim would have reduced the number of corporations. We have

conducted other simulations that indicate that even a small decrease in the number of

corporations could result in a larger increase in the number of kibbutzim (simulations using a

post-State growth rate of the corporation population of six percent, instead of the historic six-

and-two-thirds, produce, on average, 1572 kibbutzim by 1997).

The point of introducing these simulations is simply to show that absent competition

from the State, the kibbutz population might have been much larger than it is now. The

possibility of an alternative outcome of the kibbutz “experiment” should serve as a caution, not

just for the evaluation of utopianism, but also for broader comparisons of institutional

alternatives. Since the fall of state socialism, social scientific analysis has shifted away from the

comparison of capitalism to its alternatives, and towards the comparison of alternate forms of

capitalism (e.g., Stark and Bruszt, 1998). Our analysis suggests that this shift may be premature

by showing that capitalism alone did not defeat the kibbutz. In Israel, as elsewhere, the rise of

capitalism rests not only the advantages of capitalist organization and markets, but also on a

symbiosis with a very specific type of state (Miliband, 1969). Just as we think that there are

realistic institutional configurations under which the kibbutz could have faired much better, we

recognize the possibility that there are as yet untried combinations of institutional forces that

may produce systems that can compete with capitalism.

35

References

Aharoni, Yair. 1991. The Political Economy of Israel. Tel Aviv: Am Oved (Hebrew) Baratz, Joseph A. 1954. A village by the Jordan: Story of Dagania. London: Hawill. Barkai, Haim. 1977. Growth Patterns of the Kibbutz Economy. Amsterdam: North Holland. Barnett, William P., and Glenn R. Carroll. 1993. “How institutional constraints affected the organization of early American telephony.@ Journal of Law, Economics, and Organizations, 9: 98-126. Barron, David N., Elizabeth West, and Michael T. Hannan. 1994. “A time to grow and a time to die: Growth and mortality of credit unions in New York, 1914-1990.” American Journal of Sociology, 100: 381-421. Baum, Joel A.C. and Christine Oliver. 1992. “Institutional embededness and the dynamics of organizational populations.” American Sociological Review, 57: 540-559. Ben Chorin, Tzvi. 1983. Kibbutzim and Development Towns: From Paternalism to Partnership Haifa: Haifa University (Hebrew). Ben-Rafael, Eliezer. 1988. Status, Power and Conflict in the Kibbutz. Aldershot, UK: Aubury. Ben-Rafael, Eliezer. 1996. A Non-Total Revolution. Ramat Efal: Yad Tabenkin (Hebrew). Ben-Rafael, Eliezer. 1997. Crisis and transformation: The kibbutz at Century’s End. Albany: SUNY Press. Ben-Zadok, Efraim. 1985. “National Planning – The Critical Neglected Link: One Hundred Years of Jewish Settlement in Israel. International Journal of Middle East Studies, 17: 329-345. Biger, Gideon. 1994. An Empire in the Holy Land: Historical Geography of the British Administration in Palestine, 1917-1929. New York: St. Mark’s Press. Bloomfield-Ramagem, Sonia M. 1993. The Moral Economy of a Kibbutz in a Time of Crisis. PhD Dissertation, The Catholic University of America. Carroll, Glenn R., Jerry Goodstein, Antal Gyenes. 1988. “Organizations and the State: Effects of the Institutional Environment on Agricultural Cooperatives in Hungary.” Administrative Science Quarterly, 33: 233-256. Carroll, Glenn R. and Michael T. Hannan, 2000. The Demography of Corporations and Industries. Princeton, NJ: Princeton University Press.

36

Chafets, Ze’ev. 1998. “On the kibbutz, dirty hands at last.” New York Times Magazine, September 27: 40-44. Cohen, Mitchell. 1987. Zion and State. New York: Basil Blackwell. DiMaggio, Paul J. and Walter W. Powell, 1983. “The iron cage revisited: Institutional isomorphism and collective rationality in organizational fields.” American Sociological Review, 48: 147-160. Cohen-Almagor Raphael. 1995. “Cultural Pluralism and the Israeli Nation-Building Ideology.” International Journal of Middle eastern Studies, 27: 461-484. Dobbin, Frank and Timothy J. Dowd 1997 “How policy shapes competition: Early railroad foundings in Massachusetts.” Administrative Science Quarterly, 42: 501-529. Dobbin, Frank, and John R. Sutton. 1998. “The strength of a weak state: The rights revolution and the rise of human resource management divisions.” American Journal of Sociology, 104: 441-476. Dobrev, Stanislav D. 1999. "The Dynamics of the Bulgarian Newspaper Industry in a Period of Transition: Organizational Adaptation, Structural Inertia and Political Change." Industrial and Corporate Change 8-3: 573-605. Don, Yehuda 1988. Industrialization of a Rural Collective. England: Avebury. Durkheim, Emile. 1951. Suicide. New York: Free Press. Durkheim, Emile. 1984. The Division of Labor in Society. New York: Free Press. Edelman, Lauren B., Christopher Uggen, and Howard S. Erlanger. 1999. “The Endogeneity of Legal Regulation: Grievance Procedures as Rational Myth.” American Journal of Sociology, 105: 406-454. Eilat, Gidi. 2000. "A security party - form and image" pp 91-174 In: Eli Tzur (ed.) Can't Do Without Us: Mapam's Views on Security and Foreign Policy 1948-1956. Tel Aviv: Hakibbutz Hameuchad Publishing (Hebrew). Einat, Jeheskel. 1997. “Normative Rotation in Kibbutz Factories: The Effects of executive Succession, Successor Type and Structural Changes on the Organizational Performance.” MA thesis, Department of Labor Studies, Tel Aviv University (Hebrew). Etzioni, Amitai. 1966. "Change of the system: The decline of Neo-Feudalism in Israel." In: Etzioni Amitai (ed.), Studies in Social Change. New York: Holt, Rinehart and Winston.

37

Evans, Peter. 1995. Embedded Autonomy: States and Industrial Transformations. Princeton, NJ: Princeton University Press. Gambetta, Diego. 1993. The Sicilian Mafia: The Business of Private Protection. Cambridge, MA: Harvard University Press. Gavron, Daniel. 2000. The Kibbutz: Awakening from Utopia. Maryland: Rowan & Littlefield. Gordon, Aaron David. 1938. Selected Essays. New York: League for Labor in Palestine. Granovetter, Mark S. 1985. “Economic action and social structure: the problem of embeddedness.” American Journal of Sociology. 91: 481-510. Grinberg, Luis Lev. 1993. "The crisis of statehood: A weak state and strong political institutions in Israel." Journal of Theoretical Politics, 5: 89-107. Guo, Guang. 1996. “Negative multinomial regression models for clustered event counts.” Sociological Methodology, 26: 113-132. Hagana, 1954. The Hagana Book. Tel Aviv: Maarachot (Hebrew). Hakibbutz Hameuchad. 1967. The Report to the 19th Convention of HaKibbutz HaMeuchad. (Tel Aviv: HaKibbutz HaMeuchad. Hannan, Michael T. and John Freeman, 1989. Organizational Ecology. Cambridge, MA: Harvard University Press. Hausman, Jerry, Bronwyn H. Hall, and Zvi Griliches. 1984. “Econometric models for count data with an application to the patents-R&D relationship.” Econometrica, 52: 909-937. Hechter, Michael. 2000. Containing Nationalism. New York: Oxford University Press. Hertz, Rosanna, and Wayne Baker. 1983. “Women and men’s work in an Israeli kibbutz: Gender and allocation of labor.” In Michal Palgi, Joseph R. Blasi, Menachem Rosner and Marilyn Safir (eds.), Sexual equality: The Israeli kibbutz tests the theories, pp. 154-173. Norwood, PA: Norwood Editions. Hechter, Michael and Satoshi Kanazawa. 1993. “Group solidarity and social order in Japan.” Journal of Theoretical Politics, 5: 455-493. Horowitz, Dan, and Moshe Lissak. 1978. Origins of the Israeli Polity: Palestine Under the Mandate. Chicago: University of Chicago Press. Ingram, Paul and Tal Simons. 2000. "State Formation, Ideological Competition, and the Ecology of Israeli Workers' Cooperatives, 1920-1992." Administrative Science Quarterly, 45: 25-53.

38

Katz, Yossi. 1995. “The religious kibbutz movement and its credo, 1935-1948.” Middle Eastern Studies, 31: 253-280. King, Gary. 1988. “Statistical models for political science event counts: Bias in conventional procedures and evidence for the exponential Poisson regression model.” American Journal of Political Science, 32: 838-863. Lapidot, Avi Levia Applebaum and Mira Yehudai. 2000. From Protection to Competition. Ramat Efal: Yad Tabenkin (Hebrew). Leviatan, Uri. 1985. “Interpretations of sex differences in work centrality among kibbutz members.” Sex Roles, 13: 287-310. Lieblich, Amia. 1981. Kibbutz Makom. New York: Pantheon Books. Lipshitz, Gabriel. 1998. Country on the move: Migration to and within Israel, 1948-1995. Boston: Kluwer. McAdam, Doug, Sidney Tarrow and Charles Tilly. 2001. Dynamics of Contention. New York: Cambridge University Press. McTague, John J. 1983. British Policy in Palestine, 1917-1922. Lanham, MD: University Press of America. Migdal, Joel. 1988. Weak states and strong societies. Princeton: Princeton University Press. Migdal, Joel. 2001. Through the lens of Israel: Exploration in state and society. New York: SUNY Press. Miliband, Ralph. 1969. The State in Capitalist Society. New York: Basic Books. Mintz, Beth, and Michael Schwartz. 1985. The Power Structure of American Business. Chicago: University of Chicago Press. Mizruchi, Mark S., and Linda Brewster Stearns. 1994. “Money, Banking and financial Markets.” Pp. 313-341, in: Smelser, J. Neil and Richard Swedberg (eds.) Handbook of Economic Sociology. Princeton, NJ: Princeton University Press. Near, Henry. 1992. The Kibbutz Movement Volume 1. Oxford: Oxford University Press. Near, Henry. 1997. The Kibbutz Movement Volume 2. London: Littman. North, Douglass C. 1981. Structure and Change in Economic History. New York: W.W. Norton. Orren, Elhannan. 1978. Settlement amid struggles: The Pre-state Strategy of Settlement 1936-1947. Jerusalem: Yad Ben-Zvi Publications (Hebrew).

39

Parag, Yael. 1999. "Organizational Death? The Failure of Kibbutzim," MA thesis, Department of Labor Studies, Tel Aviv University (Hebrew). Putnam, Robert D. 2000. Bowling Alone. New York: Simon & Schuster. Rayman, Paula. 1981. The kibbutz community and nation building. Princeton, NJ: Princeton University Press. Rosner, Menacham, Itzhak Ben David, Alexander Avnat, Neni Cohen and Uri Leviatan. 1990. The second generation: Continuity and change in the kibbutz. New York: Greenwood Press. Rosolio, Daniel. 1999a. Crises, Adjustments and Changes in the Kibbutz Movement. Tel Aviv: Am Oved (Hebrew). Rosolio, Daniel. 1999b. “The rise and fall of the Labor State in Israel and its two pillars: The Histadruth and the Kibbutz.” No. 86, Haifa: University of Haifa, The Institute for Study and Research of the Kibbutz and Cooperative Idea. Rothschild, Joyce and J. Allen Whitt. 1986. The Cooperative Workplace: Potentials and Dilemmas of Organizational Democracy and Participation. New York: Cambridge University Press. Rozenman, Abraham. 1997. Settling the Country: The Story of the Settlement Department Before the Establishment of Israel. Jerusalem: Mosad Bialik (Hebrew). Rueschemeyer, Dietrich and Peter B. Evans. 1985. “The state and economic transformation: Toward an analysis of the conditions underlying effective intervention.” In Peter B. Evans, Dietrich Rueschemeyer and Theda Skocpol (eds.), Bringing the State Back In: 44-77. Cambridge: Cambridge University Press. Schlesinger Yaffa. 1977. “Sex Roles and Social Change in the Kibbutz“ Journal of Marriage and the Family, 39: 771-779. Segev Tom 2001. One Palestine Complete: Jews and Arabs Under the British Mandate. NY: Owl Books. Shafir, Gershon. 1996. Land, Labor and the Origins of the Israeli-Palestinian Conflict, 1882-1914. Berkeley: University of California Press. Shalev, Michael. 1992. Labour and the political economy in Israel. Oxford: Oxford University Press. Silverman, Brian S., Jackson A. Nickerson and John Freeman. 1997. Profitability, transactional alignment, and organizational mortality in the U.S. trucking industry. Strategic Management Journal, 18 (summer special issue): 31-52.

40

Simons, Tal, and Paul Ingram. 1997. “Organization and ideology: Kibbutzim and hired labor, 1951-1965.” Administrative Science Quarterly, 42: 784-813. Skocpol, Theda. 1985. “Bringing the state back in: Strategies of analysis in current research.” In Peter B. Evans, Dietrich Rueschemeyer and Theda Skocpol (eds.), Bringing the State Back In: 1-43. Cambridge: Cambridge University Press. Spilerman, Seymour, and Jack Habib. 1976. “Development towns in Israel: The role of community in creating ethnic disparities in labor force characteristics.” American Journal of Sociology, 81: 781-812. Stark, David and László Bruszt.1998. Postsocialist Pathways. New York: Cambridge Univ. Press Strange, Susan. 1996. The Retreat of the State. Cambridge: Cambridge University Press. Streeck, Wolfgang, and Pillippe C. Schmitter. 1985. “Community, market, state…and associations? The prospective contribution of interest governance to social order.” In W. Streeck and P. C. Schmitter (eds.), Private Interest Government: Beyond Market and State: 1-29. London: Sage. Swaminathan, Anand. 1995. “The proliferation of specialist organizations in the American Wine Industry, 1941-1990.” Administrative Science Quarterly, 40: 653-680. Talmon, Yonina. 1972. Family and Community in the Kibbutz. Cambridge: Harvard University Press. Tilly, Charles. 1985. “War making and state making as organized crime.” In Peter B. Evans, Dietrich Rueschemeyer and Theda Skocpol (eds.) Bringing the State Back In. New York: Cambridge University Press: 169-191. Tzur, Zeev. 1984.The Kibbutz Meuchad in the settlement of Eretz-Israel Vol. III 1949-1960. Ramat Efal: Yad Tabenkin (Hebrew). Uzzi, Brian. 1996. The sources and consequences of embeddedness for the economic performance of organizations: the network effect. American Sociological Review. 61:674-98 Wade, James B., Anand Swaminathan and Michael-Scott Saxon.1998. "Normative and Resource Flow Consequences of Local Regulations in the American Brewing Industry, 1845-1918" Administrative Science Quarterly, 43: 905-935, 1998. Warhurst, Christopher, 1998. “Recognizing the possible: The organization and control of a socialist labor process.” Administrative Science Quarterly, 43: 470-497. Weintraub, D., M. Lissak, and Y. Azmon. 1969. Moshava, Kibbutz, and Moshav. Ithaca, NY: Cornell University Press.

41

Williamson, Oliver E. 1985. The Economic Institutions of Capitalism. New York: Free Press. Wholey, Douglas R., Jon B. Christianson, and Susan M. Sanchez. 1992. “Organizational size and failure among health maintenance organizations.” American Sociological Review, 57: 829-842. Yanai, Nathan. 1982. Political Crises in Israel. Jerusalem: Keter Publishing House (Hebrew). Yiftachel, Oren and Erez Tzefadia. 1999. Policy and Identity in the Development Towns: The Case of North-African Immigrants 1952-1998. Working Paper Ben Gurion University (Hebrew).

Table 1 Descriptive Statistics

Mean Standard Deviation

(2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14)

(1) Kibbutz Foundings 0.01 0.01 -.05 .05 -.01 -.06 -.06 .0 .04 .01 .03 .11 .08 -.05 .00 (2) Jewish population /

100,000 20.03 14.04 .03 .13 .89 .96 .16 .10 .17 .14 -.11 -.14 .95 .17

(3) Political Violence 0.38 0.48 .01 -.03 .04 .05 .02 .02 .04 .02 .03 -.00 .10 (4) Development-town

density 0.05 0.25 .14 .13 .29 .22 .44 .28 .23 .20 .11 .11

(5) State of Israel 0.66 0.47 .80 .16 .10 .17 .08 -.14 -.18 .90 .50 (6) Corporation density 23.40 20.41 .17 .10 .18 .17 -.07 -.09 .87 .04 (7) Kibbutz density

(logged) -1.84 1.11 .61 .42 .32 .50 .42 .11 .15

(8) (Kibbutz density)2 0.13 0.62 .21 .22 .30 .22 .07 .09 (9) Moshav density 0.53 1.84 .43 .44 .34 .13 .12 (10) Town density 0.25 0.84 .38 .39 .07 .01 (11) Land quality 0.23 0.33 .81 -.17 -.04 (12) Rainfall 2.33 2.38 -.21 -.05 (13) Government Size 0.48 0.31 .18 (14) % Immigrants from

Asia and Africa 0.22 0.18

Table 2 Negative-Multinomial Models of Kibbutz Founding

Variable Model 1

Model 2 Model 3 Model 4

Intercept

-6.838** (.362)

-7.469** (.407)

-6.695** (.360)

-7.159** (.405)

Jewish population

0.395** (.029)

0.399** (.046)

0.427** (.044)

0.439** (.054)

Jewish population * State of Israel

-.234** (.082)

Political Violence

0.503** (.168)

0.717** (.223)

0.805** (.135)

-1.328** (.259)

Political Violence * State of Israel

-.762* (.435)

Development-town density

-1.195** (.387)

-1.158** (.390)

-1.192** (.389)

-1.175** (.392)

State of Israel

-1.279** (.417)

1.762+ (1.01)

Corporation density

-.299** (.023)

-.165** (.036)

-.306** (.022)

-.256** (.025)

Kibbutz density

0.277** (.079)

0.182* (.085)

0.274** (.079)

0.235** (.082)

(Kibbutz density)2

-.346** (.071)

-.365** (.073)

-.350** (.070)

-.355** (.072)

Moshav density

-.144** (.036)

-.152** (.037)

-.153** (.034)

-.157** (.035)

Town density

0.055 (.111)

0.083 (.117)

0.061 (.109)

0.068 (.114)

Land quality

2.984** (.400)

3.223** (.442)

3.001** (.407)

3.105** (.423)

Rainfall

0.096+ (.056)

0.101+ (.062)

0.095+ (.056)

0.098+ (.059)

% of Immigrants from Asia and Africa

-.322 (.620)

-1.039 (.966)

-1.852** (.474)

-2.174* (.576)

Government Size -2.551+ (1.39)

-1.489 (1.82)

Jewish population * Government Size

-.091+ (.071)

Political Violence * Government Size

-1.832** (.680)

Log-likelihood -1279.73 -1269.00 -1282.96 -1276.31

** p < .01, * p < .05, + p < .10

44

Figures

45