MSU ‘ RETURNING MATERIALS: P1ace in book drop to LJBRARJES remove this checkout from w your record. FINES will be charged if book is returned after the date stamped below. r 92.. ." Fri /‘//—- c5357.:- CLASS STRUCTURE, PROPERTY OWNERSHIP, AND INCOME DETERMINATION: A STUDY OF FARM LAND OWNERS IN THE UNITED STATES By Jess Carr Gilbert, Jr. A DISSERTATION Submitted to- Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Sociology 1983 ABSTRACT CLASS STRUCTURE, PROPERTY OWNERSHIP, AND INCOME DETERMINATION: A STUDY OF FARM LAND OWNERS IN THE UNITED STATES BY Jess Carr Gilbert, Jr. This work has two aims: to develop a theoretical model of the class structure of farm land ownership and to use it in an empirical analysis of income determination. I examine property ownership from the viewpoints of institutional economics as well as Weberian social theory. The class structure model elaborated derives mainly from the Marxist tradition, where class is defined in terms of the social relations of production. This is in sharp contrast to the dominant sociological account of social stratification, status attainment, which sees occupation as the chief explanatory variable. Other researchers have compared the class versus occupational perspectives on income determination; what I add to the literature is a focus on land owners. My model includes two basic classes (capitalist farmers and family farmers) and four "contradictory class locations" (pure landlords, farmer- landlords, farmer-tenants, and part-time farmers). Jess Carr Gilbert, Jr. I assess two sets of hypotheses: those at the individual level that compare the statistical explanations of the class structure and status attainment paradigms in predicting income differences, and structural hypotheses that examine how the income determination process varies by class. My data base is the U. S. Department of Agriculture's 1978 Land Ownership Survey. I operationalize six class positions as well as gender, four measures of land size, and the attainment variables of age, education, and occupational status. Using regression analysis, I investigate positive net farm income, negative net farm income, and non-farm household income. The individual-level results provide strong evidence for the class structure approach. Class outperforms occupational status for all types of income. Class structure also compares favorably to the full status attainment model for both net farm incomes and almost equals it for non-farm income. At the structural level, -classes have quite different income determination A . processes. The monetary impacts of the land-size variables and the standard status measures vary significantly by class. However, men and women in the same class still have different "returns" to the independent variables--contrary to my hypothesis. On the whole, though, the class structure model is supported. DEDICATION- To Kathy, David, and Katie ii ACKNOWLEDGEMENTS I want to thank several people for helping me with this work. I am most endebted to Craig Harris, my dissertation adviser. In our years of working together, I have learned much from Craig about sociology in general and research in particular. I have also benefitted greatly from the influence of my other doctoral committee members: Marilyn Aronoff, Rick Hill, and Dick Peterson. Their questions have not only made this a better dissertation, but have led me to become a better sociologist. While at Michigan State, I have made a number of good friends among the other graduate students in sociology. Sharon Powers, especially, has my deep-felt appreciation for years of support. I also thank Milton Steinmueller for serving as the Dean's Representative at my dissertation defense. I take this opportunity to thank the Natural Resource Economics Division of the Economics, Statistics, and Cooperatives Service (now the Economic Research Service), 0.8. Department of Agriculture, and particularly its former director--Melvin Cotner--for making the 1978 Land Ownership Survey data available for my and other researchersl use. . ' iii TABLE OF CONTENTS LiSt Of TableSOOOOOOOO0.0.0.000...OOOOOOOOOOOOOOOOOOOOVi LiSt Of FigureSOOOOOOOOOOOOOOO0.00...OOOOOOOOOOOOOOOOOXi INTRODUCTIONOOOOOO0......OOOOOCOOCDIOOOOOOO...000......1 CHAPTER ONE: I. II. III. IV. CHAPTER I. II. III. CHAPTER I. II. III. CHAPTER II. Part One: Theoretical Development Toward a Sociology of Property-- Land Ownership and Income Rights...........5 Land 5 Property Ownership 8 The Social Relations of Property 12 1. Institutional-Economic and Weberian Views 12 2. A Marxist View 16 Farm Land and Income Rights 20 TWO: The Status Attainment Approach to Income Determination...................26 An Overview 26 Occupation and Occupational Status 30 Critiques #1 THREE: The Class Structure Approach to Income Determination...................47 Class 48 Class Structure and Income Determination 5” Extensions and Criticisms 58 FOUR: The Class Structure and Income Determination Processes of Farm Land Owners in the United States..........65 The Class Structure of Farm Land Ownership 65 Hypotheses Relating the Class Structure of Farm Land Ownership and the Incomes of Farm Land Owners . 85 1. Individual-Level Hypotheses 87 2. Structural-Level Hypotheses 89 iv CHAPTER I. II. III.' CHAPTER I. II. III. IV. Part Two: Empirical Investigation FIVE: Source of Data, Operationalization of Variables, and Analytic Strategies.....92 Data Source 92 Operationalization of the Variables 96 1. Class 96 2. Other Variables 103 Strategies for Data Analysis 105 1. Samples and Sub-Samples 105 2. Statistical Techniques 107 3. Regression Equations and the Hypothses 109 SIX: Research Results......................112 Descriptive Analysis of the Class Structure of Farm Land Owners 112 Individual—Level Results 126 1. Positive Net Farm Income 129 2. Negative Net Farm Income 134 3. Non-Farm Household Income 140 Structural-Level Results: Class Analyses 145 1. Positive Net Farm Income 146 2. Negative Net Farm Income 154 3. Non-Farm Household Income 160 Structural-Level Results: Gender by Class 165 1. Positive Net Farm Income - 165 2. Non-Farm Household Income 172 CHAPTER SEVEN: Summary and Conclusions.............172 Appendix TableSOOOOOOOOO00......0.0.0.00000000000000192 LiSt Of ReferenceSOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO0.0.228 10. 11. 12. 13. LIST OF TABLES Distribution of Farm Land Owners, by Class. Land- Size Measures of Farm Land Owners, by Class (Means). Socio-Economic Characteristics of Farm Land Owners, by Class (Means). Distribution of Land and Income of Farm Land Owners, by Class. Class-Gender Distribution of Farm Land Owners. - Distribution of Occupations of Farm Land Owners Within Classes. Distribution of Classes of Farm Land Owners Within Occupations. Comparison of the Impacts of SEI and Class on Positive Net Farm Income. Comparison of the Relative Changes in Variance Explained of Positive Net Farm Income. Comparison of Models for Positive Net Farm Income. Comparison of the Impacts of SEI and Class on Negative Net Farm Income. Comparison of Relative Changes in Variance Explained of Negative Net Farm Income. Comparison of Models for Negative Net Farm Income.. vi 113 115 118 120 122 124 127 130 _ 131 133 135 138 139 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. ' 27. 28. Comparison of the Impacts of SEI and Class on Non-Farm Household Income. Comparison of the Relative Changes in: Variance Explained of Non-Farm Household Income. Comparison of Models for Non-Farm Household Income. Comparison of Class Processes for Positive Net Farm Income. Total Expected Returns to Land Variables for Positive Net Farm Income, by Class. Comparison of Class Processes for Negative Net Farm Income. Total Expected Returns to Land Variables for Negative Net Farm Income, by Class. Comparison of Class Processes for Non- Parm Household Income. Total Expected Returns to Land Variables for Non-Farm Household Income, by Class. Comparison of Class Processes for Men and Women With Positive Net Farm Income. Total Expected Returns to Land Variables for Men and Women With Positive Net Farm Income, by Class. Positive Net Farm Income of Men and Women, by Class (Means). Comparison of Class Processes for Men and - Women With Non-Farm Household Income. Total Expected Returns to Land Variables for Men and Women With Non-Farm Household Income, by Class. ' Non-Farm Household Income of Men and Women, by Class (Means). vii 141 143 mu 147 153 155 159 161 163 166. 170 171 173 175 177 APPENDIX TABLES A1. A2. A3. Au. A5. A6. A7. A8. A9. A10. A11. A12. A13. Means, Standard Deviations, and Correla- tion Coefficients for All Variables-- Positive Net Farm Income. Means, Standard Deviations, and Correla- tion Coefficients for All Variables-- Negative Net Farm Income. Means, Standard Deviations, and Correla- tion Coefficients for All Variables-- Non-Farm Household Income. Comparison of Models for Positive Net -Farm Income. Comparison of Models for Negative Net Farm Income. Comparison of Models for Non-Farm Household Income. Means, Standard Deviations, and Correla- tion Coefficients for Capitalist Farmers-- Positive Net Farm Income. Means, Standard Deviations, and Correla- tion Coefficients for Landlords--Positive Net Farm Income. Means, Standard Deviations, and Correla- tion Coefficients for Farmer-Landlords-- Positive Net Farm Income. Means, Standard Deviations, and Correla- tion Coefficients for Farmer-Tenants-- Positive Net Farm Income. Means, Standard Deviations, and Correla- tion Coefficients for Part-Time Farmers-- Positive Net Farm Income. Means, Standard Deviations, and Correla- tion Coefficients for Family Farmers-- Positive Net Farm Income. Means, Standard Deviations, and Correla- tion Coefficients for Capitalist Farmers-- Negative Net Farm Income. viii 192 193 194 195 196 197 198 199 200 201 202 203 204 A14. A15. A16. A17. A18. A19. A20. A21. A22. A23. A24. A25. A26. Means, Standard Deviations, and Correla- tion Coefficients for Landlords--Negative Net Farm Income. Means, Standard Deviations, and Correla- tion Coefficients for Farmer-Landlords-- Negative Net Farm Income. Means, Standard Deviations, and Correla- tion Coefficients for Farmer-Tenants-- Negative Net Farm Income. Means, Standard Deviations, and Correla- tion Coefficients for Part-Time Farmers-- Negative Net Farm Income. Means, Standard Deviations, and Correla- tion Coefficients for Family Farmers-- Negative Net Farm Income. Means, Standard Deviations, and Correla- tion Coefficients for Capitalist Farmers-- Non-Farm Household Income. Means Standard Deviations, and Correla- tion Coefficients for Landlords--Non-Farm Household Income. Means, Standard Deviations, and Correla- tion Coefficients for Farmer-Landlords-- Non-Farm Household Income. Means, Standard Deviations, and Correla- tion Coefficients for Farmer-Tenants-- Non-Farm Household Income. Means, Standard Deviations, and Correla- tion Coefficients for Part-Time Farmers-- Non-Farm Household Income. Means, Standard Deviations, and Correlaé tion Coefficients for Family Farmers-- Non-Farm Household Income. ' Means, Standard Deviations, and Correla- tion Coefficients for Capitalist Farmers by Gender--Positive Net Farm Income. Means, Standard Deviations, and Correla- tion Coefficients for Landlords by Gender--Positive Net Farm Income. ix 205 206 207 208 209 210 211 212 213 214 215 216 217 A27. A28. A29. A30. A31. A32. A33. A34. 'A35. A36. Means, Standard Deviations, and Correla- tion Coefficients for Farmer-Landlords by Gender--Positive Net Farm Income. Means, Standard Deviations, and Correla- tion Coefficients for Farmer-Tenants by Gender--Positive Net Farm Income. Means, Standard Deviations, and Correla- tion Coefficients for Part-Time Farmers by Gender--Positive Net Farm Income. Means, Standard Deviations, and Correla— tion Coefficients for Family Farmers by Gender--Positive Net Farm Income. Means, Standard Deviations, and Correla- tion Coefficients for Capitalist Farmers by Gender--Non-Farm Household Income. 'Means, Standard Deviations, and Correla- tion Coefficients for Landlords by Gender--Non-Farm Household Income. Means, Standard Deviations, and Correla- tion Coefficients for Farmer-Landlords by Gender-~Non-Farm Household Income. Means, Standard Deviations, and Correla- tion Coefficients for Farmer-Tenants by Gender--Non-Farm Household Income. Means, Standard Deviations, and Correla- tion Coefficients for Part-Time Farmers by Gender--Non-Farm Household Income. Means, Standard Deviations, and Correla- tion Coefficients for Family Farmers by Gender--Non-Farm Household Income. 218 ' 219 220 221 222 223 224 225 226 227 LIST OF FIGURES FIGURE 1. 10. 11. 12. 13. 1a. 15. 16. Duncan's Socioeconomic Index (SEI) for Major Occupational Categories. The Basic Status Attainment Model of Income Determination. Wright's Model of the Class Structure of Capitalist Society. Wright's Basic Models of Income Determination. Rodefeld's Farm Typology. Buttel's Outline of Agricultural Classes in the Advanced Capitalist Societies. Mooney's Agricultural Class Typology. Possible Combinations of Dimensions of Farm Land Ownership. Dimensions of Farm Land Ownership: Social (Class) Relations. The Class Structure of Farm Land Ownership. Land Ownership Classes and Farm Income. Basic Models for the Two Levels of Hypotheses. The Class Structure of Farm Land Ownership. ~Criteria for Class Positions. Decision Tree for Class Positions. Socioeconomic Index (SEI) Scores for Occupations of Farm Land Owners. xi 38 4O 55 57 67 70 72 77 79 83 84 88 97 98 99 104 INTRODUCTION In this work, I attempt to bring together three different areas of social-scientific inquiry: the class structure theory of capitalist society; the quantitative study of social stratification; and the analysis of private property ownership, farm land in particular. These three domains of investigation occupy distinct places in the social sciences. The first, the class structure view of modern society, has been most fully articulated by students in the Marxist tradition. Sociologists following Max Weber also use class categories. Both Marxists and Weberians are interested primarily in the class structure of capitalism. Statistical studies of income inequality, on the other hand, are of much more recent origin. Over the past fifteen years, "human capital" theorists in economics and status attainment reseachers in sociology have developed dominant paradigms in their disciplines. While divergent in several respects, both highlight individual income differences as one of the important problems in social stratification. I concentrate here on the sociological version. Until lately, proponents of the two social-scientific literatures--class theory and status attainment--rarely interacted. The mid-19703, however, witnessed the emergence of what Bell and Newby (1981:12) call "multivariate Marxism"--statistical research using the concepts of historical materialism. My work may be seen as part of this effort at sub-disciplinary cross- fertilization. In contrast to their attention to certain forms of social inequality, sociologists have largely ignored the role of private property in modern society. Marxists, of course, emphasize the importance of productive property, but their analysis is often at a high level of abstraction. They, too, have downplayed the significance of land ownership per se. The main group of scholars interested in understanding the uniqueness of land ownership, especially farm land, has been institutional economists. In the 19203, they began dealing with problems of property rights, farm tenancy, and land tenure in general. I draw from their analyses of farm land ownership. Thus, I seek to bring a substantive focus on land_into play with quantitative research methods as well as Marxist (and, to a lesser extent, Weberian) theories of - capitalist society.‘ My work can also be characteristized as contributing to the "new," or "critical," sociology of agriculture, also termed the "political economy" of agriculture (see Buttel, 1982). These phrases connote that their referent is generally informed by Marxist (or Weberian) theory. This orientation is a recent development, dating from the late 19703. One substantial source of this growing literature is the work of non-rural sociologists. In fact, much of the best political-economic research is by sociologists with no formal connection to the discipline of rural sociology (e.g., Friedmann, 1978. 1980, 1981; Ehrensaft, 1980; Friedland, 1981; Friedland et al., 1981). They simply study agriculture in the advanced capitalist societies. However, in large measure the new soCiology of agriculture arose from within rural sociology, the American specialty that traditionally studied farming. During the fifties and sixties, though, rural sociologists seemed to loss interest in farmers. The focus re-emerged in the 19703, largely through the work of young rural sociologists such as Richard Rodefeld in the United States and Howard Newby in Great Britain. By the end of the decade, Newby (1980:49) could speak of a "self-conscious attempt to create a new 'sociology of agriculture' which BEGINS from an analysis of landholding and the class structure." And Newby and Buttel (1980:15) could outline the research components of a critical rural sociology: "the structure of agriculture in advanced capitalism, state agricultural policy, agricultural labor, regional inequality, and agricultural ecology" (of. 0033 et al., 1980; Buttel, 1982; Gilbert, 1982). .My work, then, is part of this new rural sociology of agriculture. The objectives of the dissertation are to develop a theoretical model of the class structure of farm land ownership and to use it in an empirical study of income determination. Chapter One first clarifies the meaning of "land ownership," then elaborates several conceptual analyses of property in general and farm land in particular, including its relationship to income. My whole project is based essentially on Wright's CLASS STRUCTURE AND INCOME DETERMINATION (1979), which offers a Marxist challenge to more standard academic views of social stratification. My empirical task is thus to compare the class approach with the status attainment view of income differences. Chapters Two and Three summarize the two positions. Chapter Four reviews previous efforts to delineate the class structure of agriculture and then presents my own theoretical model of farm land owners. It also lays out the hypotheses that I shall assess, including a comparison of the two accounts of stratification as well as some propositions that derive solely from the class structure perspective. Chapter Five details the matters of my data source, operationalization' of the variables, and the different samples, statistical techniques, and regression equations employed in the analysis. Chapter Six presents the research results and Chapter Seven concludes the work. CHAPTER ONE: TOWARD A SOCIOLOGY OF PROPERTY-- LAND OWNERSHIP AND INCOME RIGHTS The principal focus of this research is land ownership. We should be as clear as possible, therefore, about the meaning of the subject matter.' What is land? What is ownership? These two questions will occupy us for the next several pages. To begin with, I rely upon a rather neglected field of economics, institutionalism, which traditionally has been concerned with issues of property rights in land. I later draw from both Weberian and Marxist theory in working toward a sociology of property relations. I. Land It may appear that answering the first query--the nature of land--is quite easy. Land is simply the ground that we stand on, the "surface of the earth" (Bertrand and Corty, 1962:187).. However, such a definition is rather limited for purposes of social-scientific investigation. Students of land, chiefly institutional economists, offer various conceptions of "land." Barlowe (1958:8-9), for instance, begins to clarify the meaning of land by delineating seven different, if overlapping, conceptualizations: 6 1) Space--room and surface for life. 2) Nature--the natural environment of soil, vegetation, and climate. 3) Situationf-location with respect to geographic features, human settlements, and other resources. 4) Consumption good--a site for residential and recreational facilities. 5) Factor of production--the source of raw materials for use in production (along with labor and capital). 6) Property--lega1 rights to an area, with the implication of a sovereign social institution. I 7) Capital--a commodity to be invested (cf. Upton, 1976; Bose, 1980:69). Obviously, this list combines natural, legal, and economic views of land (of. Neufville, 1981:253). Although my interest is ultimately social, let me dwell for a moment on the physical characteristics of land that hold sociological import (Ely and Wehrwein, 1940; Renne, 1947:182-86; Gaffney, 1962:144-47; Breimyer, 1977:10-11): 1) As part of nature, land is not created by human labor. It is a "free gift." As such, land is non- reproducible, relatively limited, and fixed in supply. This aspect of land is reflected in the popular adage: "They're not making any more of it." 2) Land is spatially immobile, both physically extended and locationally situated. It cannot be moved about at will; hence the distinction between real and 7 personal (moveable) property. 3) Land is durable. Indeed, certain physical dimensions of it are relatively permanent and indestructible (e.g., extension and location). 4) Land is not homogeneous but rather quite variable in quality, particularly with regard to fertility and situation. 1 5) Land is necessary for life. People and things must occupy certain spaces and locations. Land is not a single resource but instead constitutes a finite resource base that sustains human and non-human activities (Harvey, 1973:157-59; Andrews, 1979:27-28). However, these characteristic aspects of land are still insufficient for social-scientific purposes. They remain merely physical elements that influence social structures and processes. Consequently, Barlowe (1958:8) defines land as the "sum total of the natural and man-made resources over which possession of the earth's surface gives control" (of. Ely and Wehrwein, 1940:59). This definition emphasizes the economic ("resources") and legal ("possession") senses of land, based on its physical O materiality ("natural," "earth's surface"). Barlowe also points to the social ("control") dimension of property. In fact, this definition is similar to Marx's (1967:615, 634) view of landed property as.the "monopoly by certain persons over definite portions of the globe...," and the "ownership of certain portions of our planet by certain 8 individuals" (of. Murray, 1977:113). This substantial agreement on land itself between institional economics and Marxist political economy, then, leads us to a more detailed consideration of the second word in the term, "land ownership." II. Ownership In common usage, "ownership" has a vague meaning. We generally know, for example, who owns the house we live in, but what exactly does ownership entail? Instead of ownership per se, social scientists often prefer to speak of property or, better yet, different rights in property. Ownership may thus be defined as a claim to certain property rights. Wunderlich and Bierman (1958:288-89) state that "ownership is a rather loose aggregation of human relationships that provides maximum, though limited, use and possession of property objects." Technically, they say, it is more precise not to discuss "ownership" at all, but rather various forms of "estates"--the rights and interests that people hold in property (of. Wunderlich and Chryst, 1958:2953 Wunderlich, 1978:9; Lewis, 1979:11; U.S. Department of Agriculture, 19S9:7; Currie, 1981:169). What, then, is property? In its sociological as well as legal sense, "property" is a right not a thing. It is a socially enforceable claim. Property does not refer to an external object as such but to rights over that object (good or 9 service). Property may be classified into three types according to who, or what, has primary claims to the "thing" in question, and, more importantly, who is excluded by such claims. (1) Common property is the right of those in the society not to be excluded; (2) public property, the right of sovereign governments (the state) to exclude others; and (3) private property, the right of an individual or other "legal person" (e.g., a corporate body) to exclude all others (Macpherson, 1973:124, 1978:4- 5). ' It was with the rise of capitalism that the popular conception of property gradually became narrowed to encompass only one of the basic types, private property. This change in meaning led to the popular confusion of property as a thing rather than a right, as, for example, in the saying, "This land is my property." Feudal property rights in land, on the contrary, clearly meant claims to particular uses of the land. But with the dominance of market exchange in the emerging capitalist societies, land became to be treated like a commodity, something to be bought and sold. Property was transformed into "increasingly saleable absolute rights to things," rather than "in" them, thus blurring the distinction between the right and the object (Macpherson, 1978:8). And the claims to these "things" became predominantly private, no longer either communal or public (of. Andrews, 1979) o. 10 The state, however, as guarantor of all property, reserves for itself certain rights over even private property. In particular, the sovereign holds powers of escheat (reversion to the state when a person dies intestate and without heirs), taxation, eminent domain, and police (laws and regulations). In addition, the state exercises its more general spending power (public expenditures). All of these have significant impacts on private property (Barlowe, 1958). Nevertheless, modern property rights, while not absolute, are practically absolute in comparison with feudal rights. Specifically, unlike earlier forms, modern private prOperty can be alienated (disposed), and it is not necessarily contingent upon the performance of any specific social role or personal obligation (Newby et al., 1978:336; Macpherson, 1973:126, 1978:10). Property taxes, however, may be seen as fulfilling a broad social function (Breimyer, 1977:16). Private property, then, involves the right of persons or legal entities to exclude others from the property- object. But what positive rights inhere in modern private property? A great many; private property in capitalist societies is associated with by far the most extensive set of rights and privileges in world history. That is to say, there are fewer restrictions or limitations on modern private property than on previous widespread forms of property. One classification (Hutter, 1982:139) includes the‘rights of: ' ' ll 1) use or consumption--possession, occupancy, "abandonment, destruction, non-use; 2) free contracts--exchange, sale, rent, lease, granting usufruct, mortgage, credit; 3) transference as a gift; 4) succession; and 5) accumulation (assuming free trade). In elaboration of the above, Harris (1958:280) includes the private property rights to hold, subdivide, consolidate, abuse, waste, exploit, conserve, improve, and bequeath. Macpherson (1973, 1978) simplifies these lengthy lists by presenting the chief rights in modern private property as three: use or control, benefit or reward, and disposal or alienation (of. Rose et al., 1976:703). Until this point, the discussion of property has been rather formalist, legalistic, almost asocial (Newby, 1980:40). Now I consider property ownership from more sociological and political-economic points-of-view. The sociologist Max Weber and institutional economist Frederic Pryor provide a useful transition between the formal definitions above, and the Marxist treatment to follow. 12 III. The Social Relations of Property 1. Institutional-Economic and Weberian Views Barlowe (1958:339) and most other land economists discuss property as a "bundle of rights." But to talk of property in this way is insufficent to capture the broader sociological and economic aspects of property ownership, especially dimensions of power and control. It also confuses two quite different forms of modern private property. These two types may be labeled "property of personal appropriation" (Lafargue, 1901:5), or simply "personal possessions" (Parkin, 1979:48), versus property- as-capital. The crucial distinction between them may be outlined according to the three main characteristics of modern property (Newby et al., 1978:335-38; Rose et al., 1976:705): (1) control: user versus manager; (2) benefit:' use versus profit; and (3) disposal: consumption versus production. The first regards the difference in control between an owner-user and an owner- exproriator; the second, a difference in use between personal and market return; the third, between the functions of subsistence and accumulation. In none of these ways do the two forms hold the same sociological significance. Since almost everyone owns at least some personal possessions (clothes, books, cars, even houses), they are of less relevance to stratification )systems (Rose et al., 1976:710). "To speak of property in 13 the context of class analysis is, then, to speak of capital only, and not possessions" (Parkin, 1979:53). Unlike consumption goods, productive property can be held in quantities much larger than any single individual can use. It entails social relations to the extent that most people are excluded from such ownership, and because it creates the basis for certain unequal relationships (e.g., employer-worker). Property-as-capital depends fundamentally on a class-divided society; it is a class relation (Macpherson, 1977:11). Class relations are discussed at length in Chapters Three and Four. The use versus production distinction is entirely appropriate yet incomplete. What Weberians such as Parkin and Newby miss is precisely what Weber captures by locating classes in the market instead of in production: the intermediate possibility of amassing income from property that is used neither in the production of commodities nor for one's own consumption. The classic case is the rent that a landlord receives for housing property. Another example is working-class homeownership that, especially in inflationary times, amounts to an ‘ accumulation of wealth outside of production (Saunders, 1978, 1979). By differentiating various kinds of "property that is usable for returns," Weber (1946:182) highlights the unique monetary benefits of the rentier as 1 opposed to the entrepreneur. "Positively privileged prOperty classes typically live from property income" 14 (Weber, 1947:425). Thus Weber (1947:241, 267, 322) distinguishes between property as utilities (for consumption), commodities (for market exchange), and capital (for the creation of additional value in the process of production). Weber (1947:424-27) goes on to distinguish property classes from acquisition or commercial classes, those that offer services on the market. Since in a market society the distribution of material property creates specific life-chances, owners are systematically benefitted in the competition: "'Property' and 'lack of property' are therefore the basic categories of all class situations" (Weber, 1946:182). Weber, then, saw property ownership primarily as a means of gaining income, and thus of improving one's chances of living well (Saunders, 1978, 1979). For his part, Pryor (1973:2) defines property as a set of relations between people with regard to some good, service, or "thing." In addition, the relations, or rights to the property-object, must be economically valuable and socially enforceable. This conceptualization has the advantage of emphasizing the social relational aspect of property. Specifically, he stresses that property involves economic power and inequality. Pryor (1973:6-8) focuses on two particularly important types of property rights. The first is income rights--to use a thing in order to obtain money-by means other than self 15 labor. This power derives from either levies (e.g., rent) or use to produce commodities sold on the market for profit. Pryor identifies this special right in property as the essence of "ownership." The second is control rights--to use or dispose of a thing in economic production or exchange. ,Pryor equates this form of property with effective decision-making or economic power. It derives from ownership, or delegation by the owner. "Control" in this sense implies an asymmetric power relationship among people (cf. Vogt, 1979. 1981). In summary, both the Weberian tradition in sociology and some brands of institutional economics see "ownership" essentially as a right to, or claim on, property income.) Without employing the term, this is what Newby et al. (1978:335) mean by the modern private property right of benefit or reward: the legal claim to income. Indeed, Macpherson (1973:127-33, 1978:7-8, 206-O7) argues that this view of property is consistent with both pre- capitalist conceptions as well as emerging forms in late capitalist societies (e.g., the right not to be excluded from an income, whether via the means of labor or the welfare state).. The ownership of property, then, is widely interpreted to be fundamentally a claim on income (of. Mandel, 1968:283). Yet Pryor, especially, notes that property entails more. It is not simply, or perhaps even mainly, a legal right to income. For property includes what he calls use 16 or control rights, too. These in turn involve economic power and domination, thus unequal social relationships. Newby and Macpherson also address this profoundly sociological aspect of property, but they fail to develop the point. Even Pryor stops short of laying out a full theoretical treatment of the social relations of property rights. Nonetheless, it seems to me that Pryor's stance contains an implicit criticism of Weber. Weber tends to view property as just another "market capacity," underplaying its role as capital, as the employer of wage- labor. Property is not merely an enhancer of life-chances but rather is a prime basis for social inequality. Recent Marxist discussions make this criticism of Weber more explicit (Westergaard and Resler, 1975:344; Crompton and Gubbay, 1978:145; cf. Saunders, 1978, 1979:66). Other Marxist scholars have gone beyond Pryor in elaborating the constituent aspects of ownership and control. 2. A Marxist View Following Balibar (1970), Bettelheim (1975), and Poulantzas (1975), Vroey (1975), Wright (1976, 1979), and Clegg and Dunkerly (1980) have distinguished three "levels" or "dimensions" of property ownership. These discussions occur-in the context of the class structure of capitalism (see Chapter Three), so all are related in particular to the ownership of capital. The three levels are as follows: (1) Legal ownership--the formal, l7 juridical claim to property income, as discussed above; (2) Real economic ownership, or ownership as a relation of production--the power to assign the means of production to specific uses and to dispose of the product; and (3) Possession--the capacity to put the means of production into operation, or "actual control over the physical operation of production" (Vroey, 1975:3; Wright, 1976:11). The latter two sociological dimensions of property ownership are the ones emphasized in recent Marxist writings, for they concern the process of production rather than merely legal claims. Hazelrigg (1972) speaks of them as relations of factual control over production or "effective private property," in contrast to mere legal title or ownership. They also represent a further analytical refinement of Pryor's (1973) general "control rights." Wright (1976:29-33, 1979:23-37) presents a detailed discussion of economic ownership and possession under capitalism. These two "processes" of ownership actually constitute the substance of (capitalist) social relations of production. .As Pryor (1973) indicates and Wright . stresses, they involve social relations of control. That is, though in the context of a property-object (Pryor's thing), the crucial relationship is not person-to-object but person-to-person--a social relation. "Control" here carries with it a decision—making capacity, the ability to dispose of a.productiveresource. This power is l8 positional or structural, not individual (Wright, 1979:24). "Economic ownership" is ownership as a relation of production. It involves control over the investment and accumulation process, or the "social relations of control over money capital" (Wright, 1979:24). It determines the flow of resources into production, but stops at the; factory (or farm) gate. "Possession," on the other hand, is the ability to put the means of production to work. This is the management function, the direction of production itself, of the immediate labor process (Wright, 1979:29). Wright breaks possession down into two aspects (not "types"): the social relations of control over ‘physical capital, i.e., over the use of the physical means of production; and the social relations of control over labor, or authority in the workplace, involving the supervision and discipline of wage-labor in the process of production. Wright and others hold that, although they are both necessary, of these two basic processes of capitalist production, economic ownership is clearly the prior dimension. It determines, or sets limits upon, the actual production.process. A rentier capitalist, therefore, who is not directly involved in control over physical capital or labor, nevertheless falls within the capitalist class because of the social relations of control over money capital ("real economic ownership" of the means of production) (Wright, l9 1980b:329, 1979:25). The supervision of labor, on the other hand, does not directly affect investment decisions. Together, then, real economic ownership and possession are the very stuff of (capitalist) social relations of production (of. Dahrendorf, 1959:44-47; Vroey, 1975). In the classic capitalist enterprise of the ‘ nineteenth century, all three levels of property ownership were lodged in one class location, the individual capitalist. But capitalist development in the twentieth century has progressively dissociated the latter two dimensions of property, economic ownership and possession. Marx (1967:375, 388) was aware of the analytical difference between the money-capitalist, who owned property used in production, and the industrial capitalist, who was the actual user (employer, manager, operator) of capital in the process of production. This differentiation of functions may be described as capital- as-property and working capital. The first arises prior to production and remains outside the process per se; it yield interest and rent. The second occurs only in production and yields profits, or at least self-earned surplus. Again, the first constitutes real economic ownership and the second, poSsession (Wright, 1979:33-34; Vroey, 1975). In Chapter Three, I shall develop these points with regard to a mOre specific class analysis, which in Chapter 20 Four will be applied to farm land ownership. The remainder of this chapter will be limited to the implications of prOperty rights in farm land itself. IV. Farm Land and Income Rights With respect to property in land, the notion of "tenure" has received considerable attention from . institutional land economists. Tenure means the holding of rights to use land (Wunderlich and Chryst, 1958:295; Miller et al., 1958:562). Others speak of it as the "division of property rights in land" (Renne, 1947:429), or the social relationships established among people that govern their various rights to use, control, and occupy landed property (Barlowe, 1958:373). The two main "tenures," or types of estates (rights and interests) in land, are freehold and leasehold, or ownership and tenancy (Wunderlich and Chryst, 1958:295). Mayer et a1. (1969:3) highlight two attributes of tenure rights in land--claims and access. Claims refer to the demand on income derived from land, a demand arising out of ownership of the resources. Access refers to the decision-making . prerogatives over land, prerogatives arising out of the right to occupy the land and control its use. These two aspects of tenure are the agricultural version of Pryor's more general set of income rights and control rights. Tenure, or property rights in land, breaks down into ownership claims on income and decision-making 21 authority over the land (which itself, of course, can generate income). Hurlburt (1958:176) specifies two sorts of farm land income: the earnings of land as a factor of production and the payment received as a reward for the use of the land in production (contract rent). The two are not necessarily the same; the contribution of the land input may not equal the amount paid for its use. As Hurlburt observes, the monetary rewards to land ownership, or "cost" to the user, often exceeds the earnings ability of the land. How can this be? Because of its unique natural and social characteristics, land constitutes a "specific and peculiar category of the ownership of the means of production in general" (Newby, 1980541, 36; of. Goss et al., 1980:89-92). 'Land represents both wealth as a source of income as well as a necessary input to (agricultural) production (Boyer, 1981:111). That is, land "contains" two different types of property relations: income rights from land ownership itself and control rights over the returns to the use of the land (Murray, 1978:12). How, then, is income derived from land ownership? The ownership of farm land leads to income in two main ways: through control of its actual use in production (possession) and through control of it as an essential element for production (legal-economic ownership).. The first concerns an income surplus, whether earned by one's self (or household) or produced by wage-labor ("profit"); 22 the second refers to rent. The first is generated by a farm operator whereas the second is appropriated by a landlord. Thus, the two types of income are of different origins and deserve separate analysis. The manner in which land leads to income is fairly easy to see in the case of the farm operator. In agriculture, land is a necessary factor of production. The operation, or "possession," of the farm land means that the land is utilized in the process of production. Whether the eventual product is food or fiber or animals, the input of land is essential to farming. Farm operators generally have overall direction and control of this entire process (of. Harris, 1974). That is, they have the decision-making authority over all factors of production: labor, whether household or hired; and other inputs such as machinery, fertilizer, feed, and seeds. Thus, the operator or "possessor" of farm land functions as manager of the farm enterprise. The returns or profits generated in production through the combination-of land, labor, and capital redound to farm operators as a gross income. From this, they must deduct taxes, rent (if any), wages, loans, interest, and other costs of enterprise. The amount left over, gross minus expenses, is net farm income. The central point here is that the farm operator derives income from actual use or possession of the farm land in the proceés of production (of. Vogeler, 1981). 23 The category of rent comes about in a quite different manner. Even though land is a necessary factor in agricultural production, the ownership of farm land itself requires no operation whatsoever. Pure ownership does not imply actual possession of the land (Massey and Catalano, 1978:25). The role of rent is analagous to the place of money-capital that finances a functioning capitalist (Mandel, 1981:64). The role here is one of economic ownership, of ownership as a relation of production. This is capital-as-property (Vroey, 1975; Crompton and Gubbay, 1978:92). This dimension of control is the most basic since it determines investment and resource allocation. It involves the prior flow of resources into production as opposed to their use once in production. The money- capitalist thus has a claim on the product; interest must be paid before industrial profits can be pocketed. So it is with land rent. Landlords my be seen as a relatively passive decision-makers in the farm operation (Harris, 1974)._ Still, they demand a return on their investment in land, which is called rent. Capitalist agriculture in nineteenth century England provides a perfect illustration. The class structure typically consisted of landlord, capitalist tenant farmer, and hired farm workers. While mere ownership gives owners no necessary rights over production, they are entitled by law to a rental payment if their land is used by someone else (a tenant) in production. The rental agreement or 24 contract involves a temporary transfer of property rights in the land from the landlord to the tenant. The former retains ultimate ownership rights while the latter gains most of the rights of use and possession for a prescribed period of time, for which rent is paid (Barlowe, 1972:448; Tribe, 1977:81). Marxist political economists recognize two main types of land rent, differential and absolute. Ricardo argued that rent arises from differential soil fertility (or "favorable market location). Farmers operating highly fertile land will derive an excess or a surplus return beyond that from land of average fertility. However, since the land owner will demand payment for the use of the land, they will siphon off the above-average surplus as differential rent. But even if land is not highly productive, land owners will still require payment for its use. This Marx called absolute rent. It arises from private (exclusionary) prOperty rights in land, a monopoly of control over a necessary input into agriculture (Mandel, 1968; Newby, 1980:42; Kautsky, 1980; Massey and Catalano, 1978:42; Goss et al., 1980). Obviously, rent is the payment that land owners receive for the use of their land. It is thus the economic realization of private property in land (Marx, 1967:618). The classic political economists saw rent as one of the three basic forms of income corresponding to the three great classes of modern society: 'profits accrue to 25 capitalists, wages to workers, and rent to landlords. Ricardo even declared that the purpose of political economy was to explain the varied distribution of income among these social classes (of. Marx, 1967:618, 885). Following the lead of classical political economy in attempting to comprehend capitalist society, I shall not continue elaborating conceptual distinctions of rent. ‘Although there has been much recent theoretical debate over rent, one significant conclusion is that more empirical analysis should be done (see, e.g., Scott, 1976; Tribe, 1977; Murray, 1977, 1978; Ball, 1977, 1980; Fine, 1979, 1980, 1982; Newby, 1980; Hussain and Tribe, 1981a, 1981b). In particular, Massey and Catalano (1978:48, 53) conclude that argument over abstract formal categories such as rent is futile in the absence of concrete investigations of the role of rent in the class structure and its effect on social relationships in capitalist societies (of. Ball, 1977). After reviewing two different theories of income determination, I proceed to such an empirical examination of farm land ownership in the United States. CHAPTER TWO: THE STATUS ATTAINMENT APPROACH TO INCOME DETERMINATION In industrial capitalist societies such as the United States, income derives largely from two sources, the performance of work, or earnings, and the ownership of property, or wealth (Dalton, 1925; Atkinson, 1975). In this and the next chapter, I shall review two of the most prominent sociological models of income determination. The first is based mainly on occupation, or technical work role; the other, on class, or social relationship to the means of production. I. An Overview The tradition of "status attainment" research, the dominant approach to social stratification in the United States, is understandably very attractive to many American sociologists. In less than two decades, it has become widely recognized as one of the few areas of social scientific investigation that is consistently and 'systematically cumulative. The titles of the major books themselves are instructive: THE AMERICAN OCCUPATIONAL STRUCTURE (Blau and Duncan, 1967), SOCIOECONOMIC ‘BACKGROUND AND ACHIEVEMENT (Duncan et al., 1972), INEQUALITY (Jencks et al., 1972), EDUCATION, OCCUPATION, AND EARNINGS‘(Sewell and Hauser, 1975), THE PROCESS or 26 27 STRATIFICATION (Hauser and Featherman, 1977), OPPORTUNITY AND CHANGE (Featherman and Hauser, 1978), and WHO GETS AHEAD? (Jencks et al., 1979). Each builds directly on the ones before. The next-to-last mentioned, for instance, is a precise replication of the first--a rare feat in sociology. Moreover, the statistical techniques employed in these studies of social mobility have become increasingly sophisticated. Status attainment researchers have led the way in advancing the use of quantitative methods-of.data analysis. Indeed, many sociologists, including a leading achievement analyst (Featherman, 1981), acknowledge that their most visible impact on the larger discipline has been methodological. Such features as its cumulative nature and technical advances insure status attainment work a preeminent place in the field of sociology for some time to come (Hurst, 1979:286). Yet, almost as soon as it appeared, a number of sociologists expressed their doubts about the overall paradigm (Colcough and Horan, 1983). While status attainment studies virtually took over the field of stratification in the late 19603 and early 19703, skepticism also grew. By the mid-19703, several serious critiques were leveled at mobility research, and by 1980 sociologists were offering bona fide alternative conceptualizations of social stratification, including the income determination process. I begin this chapter with a review of status 28 attainment research. Extensive treatment is unnecessary here since this literature is one of the best-known in all of sociology and since several recent publications provide excellent overviews of the relevant work (Hurst, 1979:269- 90; Kalleberg and Sorensen, 1979; Matras, 1980; Bielby, 1981; Featherman, 1981; Granovetter, 1981; Colcough and Horan, 1983; and Campbell, 1983). I then explicate the central concept of occupation and its operationalization as occupational status. Lastly, I summarize the major criticisms that have been brought against the status attainment approach. This final point will be expanded in the following chapter, which elaborates one particular alternative approach. The status attainment paradigm did not appear full- blown with Blau and Duncan's publication in 1967. Their research question, however, stood in considerable contrast to earlier American studies of stratification, which usually examined societal rates and trends in the ' intergenerational mobility of broad occupational categories. The new orientation centered on the relative impact of socioeconomic background on an individual's staus attainment (Blau and Duncan, 1967:8-10). Shortly thereafter, Duncan (1968:695) compared the two approaches in this way: In a society where most of the statuses that entitle a person to socially valued and dis- _tributed rewards are, in fact, achieved (rather than ascribed).3tatuses, the crucial question is what factors determine the level of' 29 achievement, not whether achievement involves a person in "mobility" from the status level of his social origins. What exactly is meant by "status," "achievement," and the other operative terms? Heller and Portes (1973:51) provide the best explication. "Statuses are inequalities among social units, such as persons or families, which are more or less institutionalized with the larger social system." The three main status dimensions are wealth, prestige, and power (implicitly following Weber's types of stratification: class, status, and party). Earlier American sociologists, Haller and Portes (1973:52-53) continue, gave descriptive accounts of the overall amount of mobility in the system instead of identifying the factors which "explained" the achievement level of an individual. The language of "mobility," describing changes between status levels, was replaced in the late 19603 by that of "achievement" or of "the process of stratification." However, Heller and Portes (1973:54) argue that the "achievement" label smacks too purely of a pyschological model while "stratification" is overly broad to reflect their approach. They thus propose the phrase, "status-attainment processes" to refer to "those sets of events by which individuals come to occupy their positions in the social hierarchies of wealth, power, and prestige." Although the mobility/achievement terminology is still used, and such researchers often seem to believe that 30 their preferred topics exhaust the field of stratification, most sociologists follow Heller and Portes in characterizing this work as the study of status attainment (cf. Featherman, 1981:97). II. Occupation and Occupational Status Blau and Duncan's "basic model" was designed to account for an individual's occupational status, or, more specifically, to measure the relative impacts of a person's socioeconomic origins on his subsequent occupational attainment. "The prototypical question we ask is how the various ascribed statuses that a man brings to his career affect his achieved status in the occupational structure" (Blau and Duncan, 1967:115). Their empirical answer was equally straightforward: a son's current occupation depends on his father's education, and occupation, his own education, and his first job (Blau and Duncan, 1967:170). This five-variable "causal model" constitutes the underlying explanation of subsequent status attainment research. The method used to make these determinations is a rather sophisticated statistical “technique called path analysis, first introduced to sociology by Duncan (1966). It estimates the relative (direct and indirect) effects of one variable on another in a multivariate model (Hurst, 1979:271-73). In this. manner, Blau and Duncan introduced not only a "rudimentary conceptdal framework-~the socioeconmic life cycle--" but 31 also the methodological procedure to be followed in later status attainment studies (Featherman, 1981:81). But why this conceptual emphasis on occupation, and occupational status in particular? Addressing these questions takes us to the theoretical heart of the status attainment approach, such as it is. Analysts in this tradition actually have little to say about their theoretical dependence on occupation; indeed, they often explicitly disavow any systematic effort at theoretical development (e.g., Blau and Duncan, 1967:1-4; Duncan, Featherman, and Duncan, 1972:17, 255; Featherman and Hauser, 1973:240, 1978:9-16; Hauser and Featherman, 1977:52). Understandably, then, the rationale for their stress upon occupation is not very well developed. Nevertheless, a definite view of modern society, and the central place of occupation within it, emerges from various statements in the status attainment literature. Following the major American approach to stratification (e.g., Warner et al., 1949; see Pease et al., 1970), Blau and Duncan (1967:vii) maintain that occupation is the chief basis of social stratification in contemporary industrial society: ."In the absence of hereditary castes or feudal estates, class differences come to rest primarily on occupational positions and the economic advantages and powers associated with them." . Stratification is best understood, therefore, by studying the processes of social mobility in the occupational 32 structure. This is because occupational position is related to all three of Weber's three dimensions of inequality: economic class, political power, and prestige status. Indeed, occupation is "probably the best single indicator" of overall class position (Blau and Duncan, 1967:1-7). In short, occupation is the concept that best taps into the underlying dimension common to all aspects of stratification. A knowledge of the occupational structure and of the conditions that govern men's chances of achieving economic success by moving up in the occupational hierarchy is, therefore, essential for understanding modern society and, particularly, its stratified character (Blau and Duncan, 1967:vii-viii). Duncan (1968:689) states explicitly: I Indeed it is the assumption that occupation is the source of many of the most important kinds of rewards that allows some sociologists to fix upon. occupation as the key variable in the American stratification system. Other major proponents of the status attainment approach elaborate this position. Haller and Portes (1973:54) assert that in societies where statuses are largely achieved rather than ascribed, occupation is Strategically located in the various status hierarchies; "it appears as the most representative summary measure of a person's general social standing within the context of modern societies." In regard to other major status indicators, for example, education is crucially a determinant, and income a consequence, of occupation 33 (Heller and Portes, 1973:55). Featherman and Hauser (1973:240; Hauser and Featherman, 1977:51-52) add that occupations are highly salient for both the functioning of the society and for the allocation of scarce resources (status such as wealth, esteem, power) to persons....Suffice it to say that sociologists attribute major importance to the characteristics of occupational roles as causal antecedents of economic and social inequalities among persons in industrial societies (of. Davis and Moore, 1945). Elsewhere, they assert the "central place of occupational roles with the structure of industrialized societies, or the linkage of individuals to the society through such roles" (Featherman et al., 1975:330; Hauser and Featherman, 1977:4). This is especially the case, they continue, in capitalist economies, the structure of which explains their emphasis on occupations. They conclude that in such societies, "the MAJOR basis of stratification is occupational socioeconomic status..." (Featherman et al., 1975:358; Hauser and Featherman, 1977:50). They also put the next logical question nicely: "If the structure of capitalist industrial societies justifies the focus on occupations, then what justifies the use of prestige and socioeconomic metrics to scale occupations and to measure social mobility qua occupational stratification and inequality?" (Featherman et al., 1975:330; Hanser and Featherman, 1977:4). However, they 34 do not address this question directly, other than to say that the hierarchy underlying occupational roles is basically socioeconomic. There are at least two possible approaches to the study of occupations. One is to examine the occupational structure in terms of broad divisions such as white-collar and blue-collar (and, some add, farm). This breakdown seems to be along Weberian lines. Parkin (1971:18), for instance, states that such a division is the "backbone of the class structure, and indeed of the entire reward system of modern Western society...." He believes in viewing occupations thus according to the technical division of labor. Hauser and Featherman (1977:4) cite this passage from Parkin to justify their attention to occupation, and Featherman (1981:97) similarly urges a Weberian "class" interpretation of the status attainment tradition (of. Featherman et al., 1975:330). Blau and Duncan (1967) do in fact employ this approach to occupational categories, as do Featherman and Hauser (1978). However, their preponderant use of a second approach to studying occupations would appear_to undercut this theoretical rationale (Parkin, 1971:17-23). The other view of occupations, which status attainment researchers.usually follow, concerns "socioeconomic occupational status" (Blau and Duncan, 1967:5, 117-19; Duncan et al., 1972:3). Here again, there_ are two possible ways of ranking occupations in 35 hierarchical order. One is to classify them according to their social standing or prestige, as evaluated by public opinion surveys. The other is to rank occupations according to a scheme based on educational requirements and monetary returns, to give a socioeconomic index of occupational status. Both of these type of scalings are the subject of some debate, especially regarding exactly what it is that they measure. Briefly, there is disagreement over whether the "prestige" score actually captures that attribute of deference rather than an estimate of "goodness" or "desirability" (Featherman et al., 1975:331; Sorensen, 1979:364). The more important disagreement for our purposes, however, concerns the socioeconomic scales, for this is the approach chosen by Blau and Duncan and used in most subsequent status attainment research. The particular rating they employ, one developed by Duncan (1961), was "designed to give near-optimal reproduction of a set of prestige ratings" (Blau and Duncan, 1967:119). This ambiguity concerning its objective raises the question of what precisely Duncan's Socioeconomic Index (SEI) measures. Hauser and Featherman (1977:27, 50) see a fairly distinct separation between the "objective" socioeconomic index and "subjective" prestige scales; the latter are a "fallible index of occupational status," an "error-prone proxy for socioeconomic status." Because they believe that the "hierarchical structure 36 underlying occupational roles is largely socioeconomic," Featherman et a1. (1975:331) prefer Duncan's SEI over any prestige rating. They thus imply that Duncan's scale is not based on "popular ratings of the social standing of occupations4 (Featherman et al., 1975:335; Hauser and Featherman, 1977:9). Yet the ambiguity of the SEI's original design, quoted above from Blau and Duncan (1967:119), would seem to contradict them on this point. Haug (1977:60) agrees with the founders of the field. Of Duncan's SEI, he remarks: This index, in utilizing occupational incumbents' education and income to determine a work role's score, appears on its face to be a socioeconomic scheme and is so titled. However, the basis for ordering the occupations is a prestige rating, or more precisely, a mathematical surrogate for a prestige rating, which roots the entire scale in the grounds of public opinion. Sorensen (1978:223-24) concurs: "The most commonly used measure of socioeconomic status--Duncan's SEI scores (Duncan, 1961)--in fact is derived from occupational prestige scores." So, it seems that, while the SEI does capture some substantial aspects of socioeconomic status, .it is primarily a "prestige-based measure" (Haug, 1977:74). At least occasionally, Featherman and Hauser (1973:241-42; Hauser and Featherman, 1977:53) appear to admit as much: "Duncan's index of socioeconomic status (SEI), when applied as an attribute of a person, denotes the status accorded to the individual as the incumbent of 37 a particular occupational role." In any case, whatever it measures, the SEI is the preferred index of occupational status among status attainment researchers. One other point about its use deserves attention. Blau and Duncan (1967:123-24) properly remind us of an important assumption: I The use of occupational status scores carries a theoretical implication. We are assuming, in effect, that the occupation structure is more or less continuously graded in regard to status rather than being a set of discrete status classes. This assumption results in the ability to rank occupations quantitatively along a single dimension, "status" (of. Hauser and Featherman, 1977:6). Figure 1 indicates the range and scale of Duncan's' SEI (Featherman and Hauser, 1976:406; Hauser and Featherman, 1977:17, 27). How, though, are these numbers used in empirical studies? Let me address this question by briefly considering the status attainment model of income determination. A After Blau and Duncan's treatment of occupation, researchers extended to model to analyze other types of "acheived status," including income. Their basic model of income determination is the following (Hauser and Featherman, 1977:272; of. Duncan, 1968:693, 1969a:90, 1969b:84; Duncan et al., 1972:39; Sewell and Hauser, 1975:49; Featherman and Hauser, 1978:290): Father's years of schooling and father's SEI determine the son's 38 SEI Professional, technical, and kindred 75 Managers, officials, proprietors 57 Clerical and kindred 45 Sales and kindred 49 Craft and kindred 31 Operatives 18 Service 17 Nonfarm labor 7 Farmers and farm managers 14 Farm laborers . 9 Figure l. Duncan's Socioeconomic Index (SEI) For Major Occupational Categories. Source: Hauser and Featherman, 1977:17. 39 education, which in turn affects his own SEI, which finally influences his income. Obviously, this is but a simple extension of the occupational status model, with the addition of income as a dependent variable. It may be presented in a "path diagram" as in Figure 2. The rationale for such a model is best stated by Sewell and Hauser, 1975:2-3: We believe that earnings, here defined as direct monetary reward for occupational performance, are only one of a number of rewards that may accrue to a person because of his performance of social roles--in this case an occupational role. For this reason we think the distribution of earnings should be interpreted in the context of the allocation of occupational roles and their related perquisites. That is, earnings are a status achievement, just as is educational attainment or occupational achievement. Moreover, earnings are to some extent dependent on occupational attainment, just as occupational attainment is to some extent dependent on educational attainment (of. Duncan, 1969a:89-91; Featherman, 1971; Wang and Sewell, 1980). The above relatively coherent statement of the empirical relationships in the model is in contrast to the following admission: We lack a compelling theory of the relationship btween earnings and occupational status. The Duncan SEI scales occupations with regard to their educational requisites and economic rewards; therefore, the relationship between occupation and earnings indexes the degree of correspondence between personal economic reward and position in the socioeconomic structure via occupational incumbency (Featherman and Hauser, 1978:290). 40 Father's Education \ . Education \ Income / occupation Father's Occupation Figure 2. The Basic Status Attainment Model of Income Determination. Source: Hauser and Featherman, 1977:272. 41 This, then, is the fruit of "two decades of cumulative social science" research in the status attainment tradition (Featherman, 1981). III. Critiques No version of social stratification goes unchallenged. As we might expect, the status attainment approach, including its analysis of income inequality, has come under severe theoretical and empirical attack. Before presenting an alternative conceptualization for comprehending income inequality, let me summarize the major criticisms of the still-dominant status attainment approach. Other than the "review symposium" of Blau and Duncan (1967) in the AMERICAN SOCIOLOGICAL REVIEW (April 1968), there were few sustained critiques of status attainment research before 1974. Parkin (1971:17) did criticize in passing the tendency of American sociologists to reduce stratification to a multidemensional set of statistical rankings such as the SEI, which "obscures the systematic nature of inequality." Since the mid-19703, however, at least four critical themes have been.evident (Colcough and Horan, 1983). First is a challenge to the assumption that prestige or status scores, as currently operationalized, represent the most significant aspect of occupational differentiation. LAlternative conceptualizations of 42 occupation include authority, power, or autonomy in the workplace (Dahrendorf, 1959; Baron and Bielby, 1980; Kalleberg and Griffin, 1980; Kalleberg et al., 1981; Horan and Tolbert, 1982). Bottomore (1968) digs even deeper by inquiring into the relationship between occupation and class and raising the important issue--addressed later in this work--of the role of property ownership in social stratification (of. Wesolowski et al., 1979:9). Second is the relative inattention devoted to race and gender. Status attainment models do not explain black-white or male-female differences very well. In fact, they did not seek to do so initially. Still today, most mobility work deals only with white males--another tradition started by Blau and Duncan (1967). Critics charge that ignoring blacks and women, often the most oppressed groups, hardly gives an accurate portrayal of the stratification system (Crowder, 1974:38; Coser, 1975; Burawoy, 1977; Hurst, 1979:281-85). As Gintis (1980:13) ,comments in a related context: "A more flagrant case of 'touting your wares' would be hard to find." Third is the relative inability to explain economic inequality, especially income determination. While status attainment models can predict occupational prestige and educational achievement moderately well, they generally fail to account for-much variation in income (Jencks et al., 1971:226; Crowder, 1974:25; Wilson, 1978, 1981; ° Hurst, 1979:279-87; Colcough andeoran, 1983; see also 43 Haller, 1977, 1982; Featherman and Hauser, 1978:289). This shortcoming is of course central to the focus of my work; I later dwell on it at length. Fourth is the theoretical underpinning of status attainment research. In his presidental address to the American Sociological Association, Coser (1975) accused path analysts, and stratification researchers in particular, of "mindless empiricism" (cf. Coser, 1976:37; Featherman, 1976; Treiman, 1976). Burawoy (1977) reiterated Coser's charge that status attainment work is atheoretical and un-sociological since it ignores the impact of social structure. He sees this at least in part as a consequence of the use of "linear statistics," including regression analysis (of. Treiman, 1977). This last criticism, at the level of theory, deserves greater treatment. I briefly review the more significant theoretical critiques of status attainment research. Sorensen (1976:73, 80) argues that the work devotes too little effort at specifying the causal mechanisms that it claims to have identified. That is, status attainment analysts have no Etheory about how observed outcomes are produced." Although they interpret their empirical findings as being an inequality of results determined by the distribution of individual characteristics, the correlations may instead reflect the inequality of opportunity founded on the social structure as a set of "empty places," not the attributes of individuals. At L111 base, then, Sorensen (1976:76) maintains, status attainment research is theoretically ambigious. Bielby (1981) defends this approach to some extent by pointing out it is essentially concerned with the sorting of individuals within a given social system. Indeed, many sociologists note that status attainment work focuses. mainly at the individual level and tends to overlook social structure or "positional inequality" (Bielby and Kalleberg, 1981; Kalleberg and Griffin, 1980). Granovetter (1981:15) observes that even the term "attainment" itself reflects an individualistic bias. Halliman and Olneck (1983:1) state that status attainment analysis is "inherently incapable of explaining the institutional and interactional processes which generate the relationships the paradigm so ably documents." Campbell (1983:59) agrees: "One cannot examine the impact of social structure by holding it constant." And Hurst (1979:287) concludes that status attainment "occurs WITHIN a system of inequality" (see also Pawson, 1978:624). Crowder (1974) develops all of these criticisms in a particularly detailed and convincing manner. 'He offers a thoroughgoing critique of the leading exponent of status attainment research in the United States, 0. D. Duncan. "Duncan's major conceptual error, the ignoring of systemic constraint upon reward acquisition, leads directly to labelling a model which represents only a segment of the stratification system as THE model of the systemn 45 (Crowder, 1974:40). Duncan's models do not contain structural features of the society that exist prior to and independently of any person's entry into it. Although conceptually aware of such a system (Duncan, 1968:690), his empirical work on "the process of stratification" makes no such allowance. Crowder argues that an individual's characteristics, whether ascriptive (family origins) or achieved (education), simply are not the major factors that determine income. As evidence for this claim, Crowder (1974:38) points to the fact that status attainment models rarely explain, in a statistical sense, more than 25 percent of the variance in individual income. Therefore, if the aim is to understand the system of stratification, then Blau and Duncan's original, and oft- repeated, question--the relative impacts of familial ascription and individual achievement in status attainment-~13 misdirected. It fails to raise, much less to answer, the central underlying causes of social inequality. The other detailed critique of the literature is Horan's (1978). Unlike Coser (1975) and Burawoy (1977), he maintains that status attainment work is not I atheoretical at all but quite "theory-laden"--and that the theory is inappropriate. Horan (1978:535) asks, "what would our basic conception of stratification have to be in order for the status attainment model to be appropriate?" He attacks the use of Duncan's SEI as an indicator of 46 occupation; numerical prestige scores imply a "continuum of presumably consensual popular evaluation in which differences between occupations can only be matters of degree." Occupational differentiation, and social structure in general, is thus reduced to "shared values among individuals"--and the sociological theory that supports such a view is Parsonian functionalism, which Horan views as essentially flawed (cf. Beck et al., 1978, '1980; Hauser, 1980). So, even if status attainment work is not atheoretical, it is non-structural. The conventional models are not only individualistic, but they posit the open, competitive, and homogeneous "market process" (cf. Horan, et al., 1980). Each person is justly rewarded according to their value to society, exactly as neoclassical economic theory ordains. Why, though, is status attainment research so individualistic? Both Coser and Burawoy say that it is due to the methodological procedures of the research itself such as the use of "linear statistics." Horan (1978:538) disagrees, attributing that individualism (e.g., market homogeniety) to functionalist/neoclassical theory rather than to any- particular method of analysis (of. Featherman, 1981). The next chapter offers an alternative to the status attainment approach that seeks to overcome many of the criticisms expressed above. CHAPTER THREE: THE CLASS STRUCTURE APPROACH TO INCOME DETERMINATION As we have seen, Otis Dudley Duncan pioneered the status attainment approach to social stratification. With considerable justification, we may say that the "Duncan" of the class structure perspective on income inequality is Erik Olin Wright. A major difference, of course, is that Wright works from an explicit theoretical perspective \(Marxism), and a particular variant within that. Wright is one of the most influential young Marxist sociologists in the United States. His work represents a unique blend of two ostensibly disparate fields, French structuralist Marxism of a highly theoretical sort (Althusser's philosophy, Poulantzas' sociology) and quantitative empirical analysis of the kind begun by status attainment researchers. His contributions lie in both Marxist class theory and studies of income determination. This chapter reviews both these areas, the first of which establishes the framework for the second. The class structure approach to income determination is offered as an alternative formulation to the status attainment tradition. I first introduce this perspective through a discussion of the central concept, "class," then present its analysis of income inequality. I conclude with a survey of its most recent applications and 47 48 extensions as well as the major criticisms that have been levelled against it. I. Class Wright defines class in terms of the social relations of production. This, of course, anyone claiming to be in the Marxist tradition would do. But what do these terms signify? What is distinctive about the Marxist approach to class? Marx himself was not exactly clear on what constituted a class (see, e.g., Marx, 1967:885-86; Ollman, .1968). As Wright (1980b:327-28) points out, his particular answers to these questions represent only one stance within a larger debate among Marxist scholars. Wright (1979:4, 1980b:325) begins by contrasting other sociological definitions of class with each term in the phrase, "social relations of production." First, he understands class as being essentially relational rather than gradational. American stratificationists, following Parsons (1940, 1970) and Warner et al. (1949), typically adopt a gradational view. They see class hierarchically, as levels or strata that rank higher or lower along quantitative dimensions such as income or occupational status. From this angle, class is simply a descriptive, statistical aggregatione-a variable on a continuum. From the relational standpoint, however, classes are categorical; they represent qualitative divisions in society.£ One class is not merely relative to, but in- 49 relation with, another. The focus is not on more or less of any given attribute but rather on distinctly different positions within a system of social relations. As against many contemporary stratification theorists, both Marx and Weber take the relational view of the concept of class. Secondly, Wright (1980a:177, 1980b:325) sees classes as fundamentally social instead of technical relations. Other sociologists identify classes as relational, but locate them in the technical organization of economic relations. Dahrendorf (1959) and Lenski (1966), for example, emphasize the authority relations derived from the technical organization of production. Alternatively, Davis and Moore (1945) and Parkin (1971) rely on the technical division of labor (e.g., white-collar and blue- collar) to ground their conceptualizations of. (occupational) classes. Wright (1980b:325), on the other hand, prefers to view class relations as "irreducibly social, and thus the analysis of those relations requires a systematic analysis of the forms of social organization of economic relations." This is the task of Marx's concept of "mode of production," as elaborated below. Finally, Wright (1979:10, 1980b:326) locates classes in the sphere of production rather than that of the market. The latter position is the focus of Weberians. Weber's (1946:181) analysis defines classes as social groups that have common causal factors determining their life-chances on commodity or labor markets. Weber thus 50 identifies classes only with exchange economies. Other forms of stratification--social status and political power--were historically more important, but under modern "rationalized" capitalism, the economic element of class is generally the chief basis of inequality (Weber, 1946:183-90; cf. Giddens, 1973:105-07; Parkin, 1979:46- 58). In contrast, Wright (1979:10, 1980b:326) sees classes as being basically structured by the social relations within the production process itself. Exchange relations remain important, but secondary, in a Marxist analysis. A particular concept of exploitation is crucial here. It refers to a relation in which people in one social position can appropriate the surplus labor, or the social product of that surplus labor, of people in another, subordinate position. Class analysis in this sense seeks first to determine precisely how surplus labor is extracted. This is the purpose of the theory of mode of production, which specifies the predominant manner of appropriation--e.g., rent in the feudal mode, wage-labor in the capitalist. Classes, then, are broadly conceived according to their location within such relations of exploitation (Wright, 1979:15-17, 19803:179—80). After thus differentiating his Marxist conceptualization from gradational, technical, and market- exchange definitions, Wright (1979:17, 1980b:326; Wright and Perrone, 1977:33) describes classes as "common 51 structural positions within the social organization of production." This stance follows Lenin's (1975:479) elaboration: Classes are large groups of people which differ from each other by the place they occupy in a historically determined system of social production, by their relation (in most cases fixed and formulated in law) to the means of production, by their role in the social organization of labour an , consequently, by the dimensions and method of acquiring the share of social wealth of which they dispose. Classes are groups of people one of which can apprOpriate the labour of another owing to the different places they occupy in a definite system of social economy. It should be noted that this definition is an economic, structural approach that downplays political and cultural features necessary for a complete view of class (Ollman, 1968). The assumption here is that an analysis of class structure is prior to a full grasp of these other aspects of class (Wright, 1979:20-33, 1980b:365-70, 1980c:1438). While not an ideal solution to the problem of studying class, this is one frequently followed by social scientists (e.g., Hill, 1973:234; Braverman, 1974:27). I return to this problem at the end of the chapter. What classes, then, can be identified in the capitalist mode of production? Traditionally, Marxists analyze the class structure of capitalism in terms of (1) ownership of the means of production, (2) the buying of labor-power, and (3)-the selling of one's own labor-power. These criteria result in the delineation of two basic 52 classes: capitalists (1 and 2 above) and wage-laborers (3 above). Capital and labor are thus social categories, defined only in relation to one another. They compose the capitalist mode of production, based on a relatively few owners of productive property that employ the mass of - workers who are "free" to enter into the social relations of production with them (Marx, 1976; Wright and Perrone, 1976:33). This mode of production level of analysis aims to describe the fundamental, abstract determinants of social conflict and change. To study concrete, historical social formations, however, requires the identification of other classes; no mode of production exists in pure form. Capitalist society also contains, for example, a petty bourgeoisie, those who own and operate their means of production but neither hire nor sell labor-power. This class, while not strictly part of the capitalist mode of production, always appears in capitalist societies (Marx, 1976:927). It makes up the independent (petty or simple) form of production. In addition, for the investigation of contemporary capitalism, Wright (1976, 1978a, 1979, 1980a, 1980b) has developed the concept of "contradictory class locations." These positions fall between the above three basic classes according to the social relations of control over production. The social relations of production under capitalism consist of three interdependent processes or dimensions: 53 social relations of control over money capital (flow of investments and the accumulation process); social relations of control over physical capital (use of the Aactual means of production); and social relations of control over labor (authority in the work process). "Control" here does not mean primarily a relationship of people to things (tools, buildings, land) but of people to 'other people. In the process of production, capitalists have the power of control, workers lack it. The social relations of production, then, are not the attribute of an individual so much as of a position within the class structure (Wright, 1979:24). If someone were to change from being a capitalist to a worker, for instance, s/he 'would thereby lose contrel over the means of production. The three social processes of control, outlined above, are not always clear cut, giving rise to contradictory class locations that partake of some but not all of these dimensions. -These structurally ambiguous class positions are in contrast to basic classes--those that exist within modes of production per se (Wright and Singelmann, 1982:8180-81; Wright et al., 1982:710). The most important contradictory classes are the following: (1) Managers and supervisors, who usually have some control over physical capital and labor, occupy the "space" between capitalists and workers. (2) Semi- autonomous workers (e.g., employed engineers, professors) maintain a significant degree of control over their own 54 labor process, so are located between the petty bourgeoisie and workers. (3) Small employers control investment and the means of production but hire only minimal wage-labor; they are between the petty bourgeoisie and capitalists (Wright, 1979:39-49). A first approximation of the class structure of advanced. capitalist societies thus includes capitalists, wage- laborers, and the petty bourgeoisie plus managers, semi- autonomous workers, and small employers. Wright presents these class positions in a visual display reproduced here as Figure 3. II. Class Structure and Income Determination This, then, is the conceptual scheme that Wright (1978b, 1979, Wright and Perrone, 1977) applies to the problem cf income inequality. He is not the first to use historical materialism in an anlysis of income.(see, e.g., Hill, 1973; Bowles and Gintis, 1976), but he makes significant advances. Wright (1979:58-96) first "reconstructs" a Marxist theory of income determination, using social structures, classes, and individuals as units of analysis. Since he is concerned primarily with one certain type of social structure, capitalist society, he focuses on classes and individuals. In Wright's theory, income is determined basically by location in the class structure rather than by a cluster of individual traits, as in status attainment research. 55 Capitalist Mode Simple Commodity of Production Production CAPITALISTS Small Employers Top Managers Middle Managers Technocrats -PETTY ‘BOURGEOISIE Bottom Managers Foremen Line Supervisors Semiautonomous Employees WAGE—LABORERS UPPER CASE = Basic Class Positions lower case = Contradictory Class-Positions Figure 3. Wright's Model of The Class Structure of Capitalist Society. - Source: Wright, 1979:42.' 56 Classes acquire money in fundamentally different ways. Workers, for instance, earn wages by selling their labor power to capitalists; capitalists, in turn, exploit workers to derive profits. Both are in contrast to the self-earned income of the petty bourgeoisie. And landlords receive money through the payment of rent. In addition, classes establish limits within which exchange and technical relations influence income. For example, regardless of the educational attainment of a wage- laborer, s/he cannot hope to get the monetary rewards of a large capitalist--a situation that nicely illustrates a class-level analysis as opposed to the individual level. This does not mean, however, that an individual's education and occupation are irrelevant to personal income determination, only that their effects are structured by class relations. In other words, the class structure mediates a person's income attainment. Figure 4 illustrates these conceptual models of income determination. Based upon these suppositions, Wright develops three sets of hypotheses, which he proceeds to test empirically. First, he shows that class, understood as a common structural position in the social relations of production, influences income independently of occupation, and, in fact, offers a better statistical explanation of income level than does occupational status (Duncan's Socioeconomic Index). Second, he illustrates the claim 57 A. Individual-Level Model for Comparing Class and Status. //7 Location\ Social ——-> Education 4 Income Background ~~~M‘\\\\\\\\‘~\*Occupational Status B. Basic Model for Investigating Class Mediations. Class Location Location Within Exchange Relations 5 Income (Primarily Education) Figure 4. Wright's Basic Models of Income Determination. Source: Wright, 1979:98-99. 58 that class mediates income by showing that monetary "returns" to education vary significantly across classes. Specifically, education affects managers' incomes much more positively than it does for either workers (who can't fully benefit from it) or capitalists (who don't need it). ~Third, Wright demonstrates that class also mediates the effects of race and gender. In particular, whites and blacks, and men and women, in the same class location have incomes more similar than if class is not taken into account. This is not to deny the negative effects of racism or sexism on income, but to suggest, again, that class structures those impacts. One of the conclusions that Wright (1979:223-24) draws from his detailed analysis is that property ownership remains extremely consequential in advanced capitalist society. Despite decades of "post- industrialism," including Featherman and Hauser (1978), and the "managerial revolution" theorists, the ownership of the means of production is still decisive. The empirical part of the present work is centrally concerned with this point. III. Extensions and Criticisms Wright's innovative approach has already spawned a good deal of empirical work in sociology. Let me briefly .mention the various pieces of research. Robinson and‘- .Kelly (1979) use Wright's class structure (as well as 59 Dahrendorf's authority) approach and report that the class models perform quite well relative to the status attainment paradigm in predicting income (of. Attewell and Fitzgerald, 1980; Robinson and Kelly, 1980). Along similar lines, Kalleberg and Griffin (1980), following Wright, conceptually distinguish class from occupation and employ the concepts to analyze income inequality among individuals. Both operationalizations appear to exert independent effects on the outcome variable (see also Kalleberg et al., 1981). Robinson and Kelly (1979) and Kalleberg and Griffin (1980) suggest that there are two stratification systems in advanced capitalism, one occupational (based on the technical division of labor), the other class-based (the social relations of production). This empirical evidence supports Bottomore's (1968) original criticism of the Blau-Duncan (1967) model, that it ignores the relationship between occupation and class. Several scholars have applied Wright's approach to nations other than the United States.. Koo and Hong (1980) follow Wright's lead in an analysis of income stratification in Korea. They find that class and occupational status are equally determinate of individual income. They also modify Wright's framework by incorporating a Weberian white-collar--blue-collar distinction that captures the social division of labor; this considerably improves the statistical explanatory 60 power of their model. Winn (1982) conducts a similar analysis of Sweden and obtains similar results. Heller and Pastore (1982) report that the category of capitalist, or "self-employed employer," significantly outperforms all other variables in predicting income differences in Brazil. _ ' Several other students of stratification have employed Wright's scheme in modified fashion. Lord and Falk (1980) compare Wright's "structural" approach, along with a segmented labor market analysis (Beck et al., 1978), with the conventional "individualistic" account of income differences. Again, both theories contribute to the explanation, with the former working better for men and the latter, for women (cf. Morrissey, 1981; Lord and Falk, 1981). In a study of capitalists' income, Aldrich and Weiss (1981) transform Wright's model by assuming a continuous rather than categorical conceptualization of class divisions. By focusing particularly on workforce size as differentiating the "owning" classes, they Obliterate the distinctions between capitalist, smaller employers, and petty bougeoise. This continuous indicator of firm size predicts property-owners' incomes much better than do competing interpretations based on age, education, or job tenure.- Even though Wright's class structure approach is only a few years old, it faces a considerable number of 6l critiques. It is useful to divide the objections into those of conventional sociologists and those of more radical social critics, including Marxists. Within mainstream sociology, several criticisms are levelled. One is that Wright's operationalization is base on authority relations and, thus, derives more from Dahrendorf than from Marx (Robinson and Kelly, 1979:40; Koo and Hong, 1980:617-18; Parkin, 1979:21-23; cf. Plotke, 1980:99-103). Aldrich and Weiss (1981:281) further criticize him for not differentiating the capitalist class (of. Singer, 1982:106). Baron and Bielby (1980) lump Wright together with others (the "new structuralists") who look at the wrong level of analysis. In particular, they argue, Wright and the others are guilty of trying to explain micro processes (e.g., income determination) with macro mechanisms (class structure). Baron and Bielby suggest that the crucial missing link is firm-level analysis. Two other critics of Wright are more harsh. In his review of Wright (1979), Keyfitz (1981a) charges that the empirical results interpreted according to the thesis of exploitation face an alternative explanation based on notions of efficiency and productivity. In response to Wright's (1981) and Burawoy's (1981) rejoinders, Keyfitz (1981b) further criticizes Wright from a neo-classical economic point-of-view. He adds that Wright's data cannot .address his claimS. ,This latter point and other 62 methodological concerns are the subject of the most developed critique by a non-Marxist. Clogg (1982) observes that while Wright insists on the theoretical distinction between class and occupation, their empirical overlap is considerable (of. Wright, 1980a). He suggests that Wright's classes are simply regroupings of occupations. He also is critical of the specific regressions compared, and says that Wright fails to entertain alternative explanations of his findings. Giddens (1981:302-05) addresses Wright's work on the problem of the "new" or white-collar middle class. This pertains to the concept of contradictory class locations. He views Wright generally as tending toward a Marxist functionalism and objectivism (cf. Connell, 1979; APpelbaum, 1979). Giddens continues that contradictory class locations fail on three counts (besides being misnamed): it downplays the importance of (1) the labor market, (2) the manual versus non-manual distinction, and (3) women in lower white-collar jobs (cf. Connell, 1979:311). In line with Giddens' general evaluation, many critics point to‘a crucial shortcoming of Wright's stance. It tends to bypass the subjective interests and activities of people. Whether calling his approach formal, static, abstract, ahistorical, economistic, undialectical, or "structural schematism," all are critical of his emphasis 63 on structure over process (Greenberg and Mayer, 1978:169- 76; Ehrenreich and Ehrenreich, 1979:325; Eyerman, 1979; Plotke, 1980:99-103; Wolff and Resnick, 1982:16). Carchedi (1981:344) summarizes the danger as that of "sliding from a structural to a structuralist approach." The most detailed critique of Wright has to do with the three processes of the social relations of production and the resulting class structure, including the contradictory classes. Greenberg and Mayer (1978:174), for example, pick out his analysis of the petty bourgeoisie, which controls money-capital and the physical means of production but not the labor-power of others. Formally, this combination of criteria would seem to locate the petty bourgeoisie between the capitalists (who control all three dimensions) and the wage-laborers (who control none). Yet, Wright says that the petty bourgeoisie emerges from a different mode (or form) of production altogether, and thus is not part of the capitalist mode. Greenberg and Mayer (1978:174) view this argument as unconvincing "since all class locations are interrelated at the level of the social formation." Are‘ Wright's three criteria sufficient to identify classes, or not? Falling back upon a "mode of production" analysis at this point seems, at least to some, to smack of theorizing by fiat (cf. Crompton and Gubbay, 1978:182-90; Connell, 1979; Singer, 1982:106). 64 These criticisms are not to be taken lightly. Some of them strike at the heart of Wright's approach. They raise questions, for instance, about what constitutes class analysis. Does it primarily, or even initially, deal with the economic structure of society? The founder of the tradition, Marx, would probably say no. Class analysis must treat the political and cultural elements of social life as well; the point is to grasp how people can make their own history. Unfortunately, the approach I have taken and the data that I use in this work do not allow for such a full class analysis. While my research thus cannot address many of the above criticisms, I shall return to discuss some of the issues in-the concluding chapter. For now, however, I proceed with the research at hand. In the next chapter I modify Wright's perspective to make it applicable to farm land ownership. Then I compare this class structure approach with the status attainment model in an empirical analysis of the income determination processes of farm land owners. CHAPTER FOUR: THE CLASS STRUCTURE AND INCOME DETERMINATION PROCESSES OF FARM LAND OWNERS IN THE UNITED STATES I.’ The Class Structure of Farm Land Ownership At first glance, it may appear that all owners of farm land in capitalist societies are in the same social class, the "landlords" of classical political economy. However, in the United States agriculture, land owners have never constituted a single class, nor do they anymore even in England (Roweis and Scott, 1981:151). Classes, as we have defined them above, are groups that occupy the same relative positions within the social organization of production. In particular, a clas is composed of those who have a similar relationship to the means of production (understood as the social relations of production, i.e., person-to-person rather than person-to-thing relationships). In agriculture, the primary means of production is land. Over three-fourths of the farm sector assets is in real estate (Cochrane, 1979:325). As elaborated in Chapter One, "land" connotes the social relationships of property rights in land. I thus delineate the class structure of farm land ownership on the basis of the various relationshipss of groups to land as a means of production. 65 66 Any discussion of the class structure of United States agriculture, or of land ownership as a part thereof, must confront the notion of the "family farm." There is no general agreement on the meaning of the term. This situation has led several students of capitalist agriculture to attempt to specify the content of "family farm" (e.g., Freidmann, 1978, 1980; Mooney, 1979. 1982; Goss et al., 1980; Buttel, 1980, 1982; Singer, 1979. 1982; Vogeler, 1981; Brewster, 1979, 1980). Rodefeld undertook the initial effort to re- conceptualize the family farm. His full classification of farm types considers four factors of production (land, labor, capital, and management) and five associated status-roles (ownership of the first three of the factors, plus operational and organizational management (Goss et al., 1980:113; Rodefeld, 1976, 1979; cf. Lancelle and Rodefeld, 1980; Harris, 1974; Vogeler, 1981:21). Rodefeld's (1976:101, 1978:159, 1979:385) simpler typology, displayed in Figure 5, assumes that land ownership, capital ownership, and operational management reside largely in the same unit. It differentiates farm operators on two bases: land ownership and labor provision. The family farmer owns most of the land operated and provides most of the labor utilized. The three other types (tenant, industrial, and larger-than- family) deviate from this basic structure. As seen in the labels, Rodefeld preferred a somewhat less connotative \ 67 .mmfinmsmfi .ofioeooom noonoom .zwoaoqze Emma n.6Hmmmcom .m mmswam .mme to mom u 30H mmom swap once u swfima mamplamfimumsccH Q mamuuucmcoe m vowmcme mezzo Accmav :oz ”sag camelmafismm Icmnplmowpmq o cumulhafismm < cmwmcme mezzo mama ”swam UmNHmImHHEmMIcmnulpmwwmq omNHmImHHEmm 30A Hpowmcmz spam an swam coamfi>omm momma mo Ho>mq Hmowmcmz each an Qanmpoczo mama mo Hm>oq 68 term than "family farmer." Buttel followed him in this preference, and with a higher degree of theoretical specificity. Buttel (1980) makes the case that the concept of family farm is too imprecise and ahistorical for fruitful social-scientific inquiry. He recasts the general notion in terms of "independent commodity production," a more historically-specific concept. Independent (or petty or simple) commodity production is a form of production in which the direct producers and their household own the means of production, provide the labor themselves, produce commodities and enter into market exchange for money that supports the household (Kelly, 1979). Applied to agriculture, independent commodity production implies that the farm family (1) owns and controls the land and other "inputs" to production, (2) provides the labor used on the farm, (3) sells agricultural products on the commercial market, and (4) derives sufficient income from farming to meet its basic needs. . Buttel (1980) argues that thus reconceptualizing the ”family farm" is essential to an understanding of contemporary United States agriculture. First of all, these attributes capture the ideal of Jefferson's independent yeomanry-éthe mainstay of American agrarianism. Secondly, they identify the type of agriculture that arose in Wes tern Europe and many of its colonies during the transition between feudalism and 69 capitalism. Finally, these four defining characteristics of independent commodity production may serve as "crucial benchmarks for gauging structural change in agriculture" (Buttel, 1980:11; see also Flinn and Buttel, 1980; Harris, 1974; Brewster, 1979). This last point is most important: As any of these aspects of "family labor farming" undergo transformation, the independent commodity producers move toward other class locations in the wider, more differentiated class structure of agriculture. What are the other class positions in United States ~farming today? In the capitalist mode of production, as described by Marx, there are two essential classes, the owners of productive property and wage-laborers. Along with independent commodity producers, these constitute the central class locations in capitalist society. Buttel (1980:19) lays out a class typology of American agriculture in Figure 6. This listing is a starting point but really does not get at many of the underlying social relationships that define class positions. Buttel (1980:19) himself notes its major faults: it has no place for tenants, sharecroppers, or contract farmers; it is a. rather static formulation; and it ignores the larger economy. Mooney, however, has attempted a fuller analysis. Mooney (1979, 1982) outlines the three basic classes of capitalist farmer, farm worker, and family labor farmer. These are "basic" in the sense discussed in Chapter Three: 70 A. Agrarian Wage—Laborers l. 2. Agricultural working class (propertyless seasonal or permanent laborers) "Semi-proletarianized" or part-time farmers B. Agrarian Petty Bourgeoisie l. 2. Petty or household commodity producers (i.e., family labor farmers or petty bourgeoisie who employ relatively little labor) (Unpaid) family laborers C. Agrarian Capitalists l. 4. Figure Source: Petty capitalist (or "larger-than-family") farmers who employ considerable hired labor Competitive sector capitalist farmers (farm operations owned by small nonfarm corporations) Monopoly sector capitalist farmers (farm operations owned by large nonfarm corporations such as Tenneco) ‘ Absentee landlords 6. Buttel's Outline of Agricultural Classes in the Advanced Capitalist Societies. Buttel, 1980:19. 71 they are completely polarized within the social relations of production. Moreover, he identifies six social relationships in agriculture that further differentiate and may eventually transform this basic class structure: tenancy, indebtedness, contract production, off-farm work, hiring wage-labor, and renting land out to others to farm. Any of these additional social conditions leads one out of simple commodity production and into the capitalist mode of production per se. In particular, the first four, Mooney claims, entail a move toward the working class, whereas the last two, toward the capitalist class. Mooney follows Wright in conceptualizing these potential social locations as "contradictory class positions" because they mix the pure criteria for class identification (the three dimensions of control over money-capital, physical capital, and others' labor-power). Among the contradictory locations that Mooney specifies are those of the farmer-landlord, who combines owner-operatorship with the appropriation of rent; the farmer-tenant, or part- owner, who may combine the renting in of land with any of the other criteria; and the part-time farmer, who actually combines two different claSs situations (farm and non- farm). Mooney's (1979:55) class positions are schematized in Figure 7. Although Mooney draws extensively from Wright, he departs from him in two important ways. Wright (1979:39) states explicitly that his main concern is with structural 723 .mea .zmcooz "momzom .mwoaoaze mmmao ampsuasoamw< w.>mcooz .w mpswfim mco.~muo_ mac—u u . . >couu.vmuucou u «conducts mmMMWuHWMM nununnana - WIQHMHHHHmn" co.uu:v0ua eo>o .oLucou ocu .neooseoem .v mo>ohc¢o " 1111 Logos cone. to o.on ecu r .W uuaeueou m . nmw thh12\111 c..os_o> us.aesn ousvotau rr-------.rmmmmuwa-i h< ¢uu0m u nee—u um._oa.aau u co..o:u0ea Deceased can- . —11 36: 11“ mo necn_>uoasn “ acuE>oa “mucous. can: n .nuouce-e no.eo.om " acct can: u " v.0:omsog uuuuuunnauu_uaunnaunnuuu c_zu.3 os_c> m:.¢c:m c_cuouu w.w.¢uo¢=oo >hhwc fill lo I III amnwmm_mew-ia---- ---- "mmumpmmm,-. . neoguo be taxed Loan—nu :o_uuauoLa do acquit: so.u-.sisuuo\ueoluno>e_uu 33025.3 525 .9556: 2o.»u=ooxa no use: >C.oozzoo moax.m oncogene oozes: .esooes-- Loan. ‘05.:un «motoue.un accent «couscoCa “use.t sot» 03—0) as_aesa duo-eaotcco magma; Hooacoo .codumOoHHm monsomoh cam unusumm>:« Lw>o donucoo .Amucwfip oeoocav aupoaona ou cacao ammo; .aOLucoo o: .HosuCOo Hqu .w msswam HNM 77 + l I + -t 4- + + «r 4- + -p + + «r + «e + + -+ -+ + -+ + -r + AmanmnOLQEHo nmwmcm: ooh“: cuaz noczo o>ummmm AcfiawnongHv cLoHocmA m>ummmm AmanmnOLQEHo possum pomhucoo LoEme sowsscoo smaamssamo Lospmm pomgucoo \LoonQEm HHmEm cLoHocmq o>auo< AoHanoLQEHv howmcm: pond: cud: poczo o>auo< AoaanOLQEHV AonnwpoLaEHv Lesson hmasmm .LoEme unfiamauomo LuonQEm HHmEm .oa .mH .za .MH .NH .HH .oa Looscocm somnfio mcmcuo uo coduonoosm no Lozomlponmq memo: Hmofinzcm codewonnom nuznhoczo oasocoom m . m coauoscOLm ho acoaumamm Hmaoom undamanomo )[ H oucnpoczo ammo; 78 absentee owners with little day-to-day interest in the' farm (say, a New Yorker who inherited a ranch in Wyoming). The other major categories are contract farmers. These are owner-operators who enter into legal agreements with. processing firms before planting time. They are thus assured of a market for their produce, but they often relinquish most of the control of the operation. The firm occasionally even provides capital inputs (other than land) and demands adherence to their, rather than the farmer's own, time-table. In such cases, the agribusiness corporation holds effective ownership of the farm (Pfeffer, 1982). Let me now turn to the major class locations. Figure 9 has the same dimensions as the previous one. It identifies and situates the six different class positions that are primarily engaged in farm land ownership. As mentioned above, the basic classes in capitalism are capitalist and wage-laborer, or, in agriculture, the capitalist farmer and farm laborer. In addition, outside the capitalist mode of production as such is the agricultural version of the independent commodity 'producer, the family farmer.l These three classes, again, ‘are "basic" in that they have unambiguous social relations of production; it is clear whether or not they own and possess. Capitalist farmers alone exercise control over the labor-power of others. This, of course, is central to the .mcouamamm Awmmdov Hmdoom “aficmhmczo mama spam no acoamcmsdo .m ossmam .Aaafisocuzm momaoxcoz ocm =Hmuaomo Hmoumznocv :ofiuozcogn uo cofiumpmoo Hashes Lo>o magmas Hosucoom .coaumooaflm monsomma ccm ucosumm>ca po>o Honucoom H .HOLacoo oson n +\I “Hosp:Oo o: u I .HOhucoo scams u I\+ «Hopscoo dash n + .Amunwua csoocdv zanooogq o» manna ammo; 79 + I + + + possum osuaIasmm + I I\+ +\I +\I acmcoeILmEpmm I\+ I I\+ I\+ + ccoaccquLoEme I I +\- I\+ + egoflecaa “Anncczo ocmav uneaumooa mmmao mucuouomuucoo + I + + + scenes aaasmm + I I I I Apononmq Shaky I + + + I\+ Losses smaqwsaamo "momnmao owwmm wnocuo ho :ofiucscOLm no Lozomlsonmq memo: Hmodm»:m Loosoosm pompfin mcodmmmmmom moannuocxo oqucoom Hauzmnoczo Hmmmq :ofiuozcopm ho acoaumamm Hmaoom undamuaamo 80 very definition of "capitalist." However, I shall not discuss this column further since all the other classes ' lack such control. Two other points about capitalists: First, while they do not necessarily have legal title to all of the farm land that they operate, they do have "real economic ownership" over the property. This illustrates the important Marxist distinction between formal juridical rights and substantive control over productive resources (Wright, 1979:33). Secondly, Figure 9 indicates that capitalist farmers are not direct producers. This is appropriate here since I am emphasizing a theoretical analysis of capitalist agriculture. But it is an oversimplification. Many large farmers, including those who hire wage-labor, are also involved as workers in the actual process of production; in this they are similar to family farmers. Still, here I am highlighting the theoretical difference between these two basic classes, which concerns the employment of wage-labor. Capitalists as capitalists do not work directly in the process of production. The other essential class in capitalist agriculture, obviously, is the farm-laborer. Farm workers are not land owners at all. .I include them here to complete the basic .class typology and, especially, to I'complement" the capitalist farmer. “These two classes are shown to be mirror images of one another in every respect. The remainder of the-class positions in Figure 9 all assume 81 some degree of farm land ownership. Other than controlling labor-power, the family farmer is the only class position that is positively engaged at all dimensions of property ownership, being both an owner and a worker. This, again, is to reinforce Mooney's and Buttel's portrayal of the "family labor farmer" as fundamentally outside of the capitalist mode of production. Davis (1980) characterizes this form of independent commodity production as "propertied labor"--an apt description in line with Figure 9. Landlords occupy the only class location (other than capitalists) that is clearly a non-producer. Yet landlords--along with the other contradictory class positions--are ambiguous on the two major dimensions of economic ownership and possession of the physical means of production. As relatively passive owners, pure landlords maintain control over only part of the investment and accumulation processes while they relinquish almost all of the control rights over the property (for the time stipulated in the rental agreement). A non-owning full tenant farmer would exhibit Just the opposite relations of (ownership and control. Farmer-landlords and farmer-tenants are even more contradictory than pure landlords. -Both directly operate some of their own land and so have "majority control" over the physical means of production. Farmer-landlords also rent part of their land out to tenants,‘thus having a 82 measure of economic ownership similar to full landlords, and an ambiguous place in the "direct producer" column.‘ Farmer-tenants, on the other land, have only partial ownership (both legal and economic) since they rent in most of the land they operate, but they are clearly direct producers. Part-time farmers are in a class by themselves. They actually are located in two class systems simultaneously-- one on the farm, the other in their "off-farm" job. While formally like the family farmer in being fully engaged in all aspects of property ownership without hiring outside labor, part-time farmers are actually only partially in agriculture. This situation, for them, is what constitutes their "contradictory class location" (of. Mooney, 1979:52; Singer, 1982:105; Buttel and Larson, 1982; Buttel et al., 1982; Coughenour and Swanson, 1983). Figure 10 presents the class locations of contemporary American agriculture and farm land ownership in schematic form. This summarizes the end-point of my theoretical development; I shall shortly turn to empirical .investigation. One thing that the classificatory schemes in Figures 9 and 10 accomplish is that they suggest the _different bases for income among the various classes. Figure 11 makes this explicit. Each pure class has a unique income determination process. Capitalists acquire money in the form of profit, derived from the exploitation of wage-labor. Landlords receive rent from their tenants. 83 mmzm Income Class lLand Variables, Gender, Age, and Education. B. Model of Class Mediations For Structural-Level Hypotheses. Class Gender -j:::::::::i:::::::::i:Income Status Attainment2 2 Age, Education, and SEI. Land Variables Figure 12. Basic Models For the Two Levels of Hypotheses. 89 attainment tradition maintains that occupation is "the key variable in the American stratification system" (Duncan,’ 1968:689). Marxists, on the other hand, hold that class is a more fundamental determinant of social stratification. Hypothesis 2 seeks to get at this debate by comparing the relative effects of class and occupational status on income. HYPOTHESIS 3: Class alone will give a better explanation of income differences than will the status attainment approach. This constitutes a comparison of paradigms-~the "strongest" test of Marxist class theory at the individual level. _There is a certain amount of overlap between Hypotheses 2 and 3. If 3 is confirmed, then 2 must also be. However, it is possible that 2 will receive support while 3 does not. Hypothesis 3 thus culminates my examination of competing theoretical accounts of income inequality among farm land owners. 2. Structural-Level Hypotheses Analysis at the structural level is obviously central to the class structure approach to social stratification. At this level, classes themselves, and not individuals, are the proper unit of analysis. As discussed earlier, each class is assumed to have a fundamentally different manner of acquiring income. This is another way of saying that classes are structural mediations of the income determination process. The following hypotheses, then, 90 are formulated at the level of class. They are further specifications of the claim that classes have essentially dissimilar income determination processes. HYPOTHESIS u: Various measures of farm land (amount owned and operated, rented in and operated, rented out, market value) will have differential impacts on income depending on class. This proposition is portrayed in Figure 4 above: the six classes use farm land to acquire money in different ways. They should therefore "cash in" on their land in differential fashion. HYPOTHESIS 5: The standard determinants of status attainment (age, education, occupation) will likewise differentially affect income by class. If classes are real entities, then they should influence the entire income determination process, not just the land components. This hypothesis, then, extends Hypothesis u to propose that class "structures" the monetary returns to various measures of status. HYPOTHESIS 6: The returns to both the land measures and the status indicators will be similar for men and women within the same class. If the above two hypotheses are true for both men and women, then they will be so for each alone. Hypothesis 6 goes further in positing that class mediates the specific effects of gender on income. If class rather than gender per se accounts for much of the observed sexual diScrimination, then class may be said to structure gender relations just as it does the returns 91 to land and status. HYPOTHESIS 7: The difference between the average incomes of men and women in each class will be smaller than the overall income difference between men and women. Some of the income difference between them is due to the class-gender distribution rather than to a global "sexual discrimination." If this is true, we would expect a smaller income difference within a particular class than if class were not considered at all. These, then, are the seven hypotheses that I shall test in this study. While they do not exhaust the questions that could be raised about the relationship between class structure and income determination (much less those about class theory in general), I believe that they constitute a meaningful series of propositions that are important to the understanding of how class operates in the agricultural sector of capitalist societies such as the United States. I now proceed to the empirical- analytic half of the work. CHAPTER FIVE: SOURCE OF DATA, OPERATIONALIZATION 0F VARIABLES, AND ANALYTIC STRATEGIES I. Data Source The data used in this study are from the 1978 Land Ownership Survey conducted by the Natural Resource Economics Division of the Economics, Statistics, and Cooperative Service (now the Economic Research Service) of .the United States Department of Agriculture. The survey was based on a 2 percent stratified random sample of the nation's land area. One-third of this sample, which included every county in the United States (excluding Alaska), was randomly selected for the Land Ownership Survey. The sampling frame consisted generally of 160-. acre parcels of land, called "primary sampling units," selected from all land units of 7,680 acres. A common probability of selection was 1/1uu (: 1/3 x 160/7680). Within the sampling units, a single point was randomly picked, and its owner identified. In this manner, the Department of Agriculture attempted to survey u5,898 private land owners. Mail questionnaires, telephone interviews, and personal contacts ultimately resulted in usable data on over 37,000 land owners, for a response rate of 81 percent._ Adjustments for non-response were made following a sample of non-respondents (D. Lewis, 1980:3; J. Lewis, 1980:33-37). 92 93 Because of the random selection of points of land, not all owners had equal chances of being surveyed. Owners of large parcels were more likely to receive a questionnaire. Therefore, in order to generalize from the sample to the population of land owners, respondents owning small acreages had to be weighted relatively more heavily than the owners of large parcels. The Department of Agriculture accomplished this weighting through differential expansion factors, equal to the inverse of the probality of being included in the sample. That probability is equal to the probability that one's primary sampling unit was selected times the probability that one's parcelreporting unit, or R.U.) was chosen within the sampling unit. The expansion factor for 160-acre sampling units thus equals 1/(1/3 x 160/7680 x R.U./160) = 23,040/R.U. Each sample land owner has an expansion factor, or weight, computed in this way by the Department of Agriculture; each respondent, then, represents a number of land owners equal to the size of the expansion factor (D. Lewis, 1980:2-3; J. Lewis,-1980:34-35). The Land Ownership Survey was a 70-item questionnaire that provided the following kinds of information about the owner and his or her land in the specified county (J., Lewis, 1980): 9n 1) Type of owner--individual, family, partnership, corporation, etc. Size of holding in acres. Decade of acquisition of land. A) Manner of acquisition--gift, purchase, inheritance. Estimated market value of land. Type of land use--three catagories: (a) farm, (b) residential, commercial, industrial, and similar "urban" uses, and (c) other, including forest, waste, and idle land. 7) Tenure--whether the owner is a farm operator, and whether the owner (of all types of land) rents to or from others. 8) Acres of land owned in other counties or states. 9) Residence-~whether in same county or state as land. WM vv O\U'| vv The following information was asked only of individual- type owners: 10) Occupation: "What was your principal occupation during 1977?" 11) If "farmer": "Did you work at an off-farm job 100 days or more during 1977?" 12) Age. 13) Sex. 1“) Race. 15) Years of schooling. 16) Net farm income (categorized up to $50,000). 17) Non-farm household income (categorized up to $50,000). 18) Number of dependents in household, and number contributing non-farm income. 19) National citizenship. The Land Ownership Survey presents some rather serious limitations. As Castells (1980:27) observes, data gathered by government agencies in capitalist sicieties can hardly be expected to fit Marxist categories. Class, understood as position with the social relations of production, is a prime case in point. The questionnaire contains no information on whether the owner employs wage- labor--a crucial criterion for separating simple commodity producers from proper capitalists. Therefore, I.cannot 95 differentiate these two classes on that basis. Secondly, only three land uses are identified, and only one, farm land, may reasonably be said to be in production. The other two types mix very different uses of the land such as residential with commercial and forest with waste land. A third shortcoming of the survey is that it gives no income data for corporate or institutional owners. Many of the larger farms and other capitalist enterprises are certainly incorporated, but they must be excluded from my empirical analysis of income determination. Yet there are some strong points to the Land Ownership Survey. For one thing, it is the only large national study of land ownership per se ever done in the United States. In other words, until 1978, there was no good information on Just who owned the land. The only comparable study, conducted in 1946, was limited to farm land ownership (see Gilbert and Harris, 1981, and forthcoming). Another benefit of the survey is that it offers considerable data on some quite wealthy Americans-- information that is very difficult to come by. In fact, Wright (1979:22fl) and others often claim that the capitalist class cannot be studied by the survey method. On this point, the United States Department of Agriculture has shown them to be wrong. It is as if the Department of Commerce or the Treasury had taken a "sample" of stocks and bonds, traced them to their owners, and submitted them to a social survey. The resulting information would be 96 invaluable, as is the Land Ownership Survey. That is not to say that it is without drawbacks. It is, however, the best that we have to work with. II. Operationalization of the Variables 1. Class Figure 13, taken from the culmination of the theoretical discussion in Chapter Four, lays out the class structure of farm land ownership. As indicated above, operationalizing a Marxist concept of class with the Land Ownership Survey data is problematic. However, a reasonable approximation to delineating the class structure of farm land ownership can be made. Figure 14 shows how the following variables are cdmbined for specific classes: tenure, occupation, number of acres owned, and number of acres operated. The operationalization is hierarchical; the "decision tree" in Figure 15 gives the ordering. The two "land-size" criteria are used to designate only the capitalist class; I begin with them. Since there are no data on the hiring of wage-labor, I have resorted to the use of a proxy to indicate such employment: number of farm acres operated (not '. necessarily owned). The 1978 Census of Agriculture (U. S. Bureau of the Census, 1981:77) presents information on hired farm labor by size of farm. In 1978, 76 percent of all farms of 2,000 acres and over (almost 63,000) hired 97 .QHSmpmczo pcmcmelpmapmm mm2mo :« anon :« econ / x /\ peso zcm puma euso hem pack 7 X BmHacaHm Hobos .< .< asses mooeom .e 60:30 re sans .mmmao an .mamazo mama Eamm mo mEooaH mam mama mo aoHusaHapmam .a cache 121 acquire their "expected" share; part-timers do not. Based on proportion of total acres operated, however, capitalists do very poorly: using 37 percent of the land, they get only 3 percent of the net farm income. This is in part a reflection of the low value per acre of their land (see Table 2). From this standpoint, farmer-tenants do somewhat better, and family farmers much better; the latter class operates 15 percent of the land and receives 38 percent of the farm income. The distribution is reversed for non-farm income, with the two landlord classes and the capitalists obtaining roughly proportionate percentages (based on sample size). Part-time farmers receive over half of all non-farm income. Farmer-tenants and family farmers, who together constitute 27 percent of farm land owners, get only 11 percent of the non-farm income. These proportions should be kept in mind as we analyze the income determination processes for the various classes. I conclude this descriptive section with two separate cross-tabulations, class by gender and occupation. Panel A of Table 5 reveals, first of all, that women constitute 16 percent of all farm-land owners. Their distribution across the classes, though, is highly variable, ranging from 4 percent of farmer-tenants to 35 percent of landlords. Men dominate every class, making up about 90 percent of all except the two landlord ones. Panel B indicates that a majority (60 percent) of all women are 122 Table 5. Class-Gender Distribution of Farm Land Owners. A. Gender by Class Men Women Total N (%) (%) (7) (000) 'Capitalist Farmer 94.6 5.4 100 48 Landlord 65.0 35.0 100 . 1,127 Farmer-Landlord 77.8 22.2 100 . 253 Farmer-Tenant 96.4 3.6 100 447 Part-Time Farmer 90.9 9.1 I 100 1,639 Family Farmer 94.0 6.0 100 682 All 84.3 15.7 100 4,196 B. Class by Gender Men Women -All (%) (Z) (Z) Capitalist Farmer 1.3 0.4 ' 1.2 Landlord 20.7 59.9 26.9 Farmer-Landlord 5.6 8.5 6.0 Farmer-Tenant 12.2 2.5 10.6 Part-Time Farmer 42.2 22.5 . 39.1 Family Farmer 18.1 6.2 16.2 Total 100 100 100 N(000) 3,536 660 4,196 123 non-operator landlords. More men (42 percent), in contrast, are in the part-timer class than in any other. This is the second most common position for women. If we add to landlords the proportion in part-timers (22 percent), it accounts for over 80 percent of all land- owning females. These two are in fact the largest classes (see Table 1). Men are also concentrated in them, but not so tightly (65 percent). Table 6 presents the class-by-occupation breakdown. Almost all of the capitalist class (93 percent) identify themselves primarily as farmers, as do most farmer-tenants (80 percent). Nearly half of all landlords are retired, and 10 percent of them are housewives. Together, these two occupational categories constitute over 36 percent of the farmer-landlords, followed by occupational farmers (25 percent). Part-timers are the most varied occupationally: 27 percent are white-collar; 40 percent blue-collar; 19 percent retired; and 3 percent housewives. The family farming class can only be farmers. Overall, the farm land owning occupational structure is 18 percent white-collar, 22 percent blue-collar, 33 percent farmers, 22 percent retired, and 4 percent housewives. . Table 7 breaks down class by occupation. Over half of the following occupational categories are non-operator landlords: physicians and dentists, attorneys and judges, farm laborers, housewives, and retirees. Most of the remaining occupations are primarily part-time farmers, the - 124 mmmxaoz 0.0 II 0.ma 0.m 0.0 m.m m.0 mm>apmamao mamxaoz 0.0 II m.0a 0.0 0.0 0.0 0.0 pumao mamxaoz 0.0 II w.m m.a 0.m 0.m II amoaamao mamxaoz 0.m II m-m a.m m.m 0.m m.0 mmamm amapo mamxokm no mpamw< 0.0 II 0.0 a.0 a.a m.a II mumumm Hmmm maoumapmaaaema 0.0 II m.ma 0.0 m.» 0.0 0.a 0 naommcsz maoxaoz Hmoaaaooe a a.m II 0.0 0.0 m.m 0.0 a.a mamaoammmmoam amauo mmwm30 0.0 II m.0 II 0.0 0.0 0.0 0 hooszca mpmapamm m.a II a.a m.0 0.a 0.m 0.0 a ncmaoansam ARV 100 A00 A00 A00 A00 A00 amsamm_ amEamm pamama maoamama maoa amspmm Ham maaemm mEHBIuamm Iamsamm Iamsamm Imama pmaamuadmo .mmmmmao aaapaz mamazo mama Ehmm mo maoapmasooo ho aoausaaapmam .0 magma 125 . OcHQN H II H00.: 000 m0m.a ms: 000 000.a 00 A000vz 00a 00a 00a 000 000 00a 000 ampoe 0.: II 0.0 0.0 0.0 0.0 0.0 mm>azmmsom 0.a0 II H.0a m.a H.00 0.0: 0.a mmmaaumm mamwmamz Eamm 0.00 ooa 0.0 0.00 0.00 0.0 H.00 mam mameamm amanom mam a.0 II 0.0 II a.0 0.0 0.0 mamaoama Eamm Amaoammzom mum>aam wafimzaoaHv 0.a II 0.0 0.0 0.a a.0 II mamxaoz moa>amm Asamm pamoxmv 0.0 II 0.0 H.H 0.0 0.0 0.0 mamaoama A00 A00 A00 A00 A00 A00 A00 amaamm ameamm pamame maoamama mLOH ameamm 00¢ 0005mm oEHBIpamm amepmm Iamshmm Imama umaamuaamo .A0.ceoov 0 manna 126 highest percentages being among blue-collar and service workers. In fact, every occupation has at least a third, and often two-thirds, of its members in the part-timer position--with the exceptions of housewives (29 percent) and farmers (10 percent). The majority of farmers (56 percent) are in the family farmer class. Over 26 percent of all farmers are in the farmer-tenant class ldcation. There are more farmers than any other occupation in the capitalist class, but this is only 3.4 percent of all farmers. Again, this distribution of occupationseby-class should be kept in mind as we compare occupational status and land-owning classes below. Having covered in some detail the descriptive characteristics of the classes, I now turn to the regression alalysis of income determination among and between them. II. Individual-Level Results This section presents the results of the individual- level regressions. Hypotheses 1-3 offer progressively more stringent tests of the class structure model compared to the status attainment paradigm. In addition to the means and standard deviations of all variables, the correlation coefficients used in the regression analyses are in the Appendix (Tables A1-A3). 127 mamxaoz H00 00a II 0.00 0.0 0.0 0.0a II mm>apma000 mamxao3 000 00a II a.a0 0.0 0.0 H.0a a.0 c0na0. . mamxaoz 00 000 II 0.00 0.0 0.0 0.0: II amoaaoao mamxnoz 00a 000 II 0.00 0.ma 0.0 0.00 0.0 nmamm 00000 mamxoam Lo mpamw< 00 00a II 0.a0 a.0 0.0 0.00 a.0 000000 ammm maOpmapmaaHEma mam ooa II 0.00 m.m 0.0 .:.00 0.0 0 0000000: mmmxaoz amOfiaaomE a 000 000 II 0.00 0.0 0.0 0.00 0.0 nanaoanmmaoam 00000 . momm30 ma 000 II 000m II 0.0a 0.00 a.0 0 npo0sca mpmauamn 00 ooa II 0.00 0.0 0.0 0.00 0.0 0 hcmacan000 100 A00 A00 A00 A00 000 000 amEamm amsamm pamame maoamama maoH mosamm Aooovz HGDOE hHHEmm mEHBlphmm 1.09:me IeamEramm IUCGA DmHHMuHQmU .ma0a0m03000 aaauaz mamazo mama Eamm mo mmmmmao mo aoausnahumaa .0 mHQmB 128 .OVHGN " ll a00.0 0.0a 0.00 0.0a 0.0 0.00 0.H HH< 000 00a II 0.00 0.0 0.0 0.00 H.0 mm>azmmsom 000 00a II 0.00 0.0 0.0 0.00 a.0 mmmafiumm mammmamz.eamm 000.0 00a 0.00 0.0 0.00 0.0 0.0 0.m 0:0 ntmaama amEmaom mam 0 00a II 0.0: II 0.0 :.a0 0.0 mamaoama Eamm Amaoammsom mpm>HLm waamSHOaHv 00 000 II 0.00 0.0 0.0 0.00 II mamxaoz moa>amm Neamm ummoxmv 0aa 00a II 0.00 0.: 0.0 0.a0 II mamaoama A00 A00 A00 000 A00 A00 ARV ameamm ameamm pamame maoamama maoH amsamm 000002 Hmuoe zHaEmm mEaBIuamm Iamsamm Iamsamm Imama umaamuaamo .a0.seccv 0 oaaaa 129 1. Positive Net Farm Income Hypothesis 1 holds that class will exert an influence on income that is independent of occupational status. Table 8 displays the results of four non-hierarchical regressions of positive net farm income on various combinations of the Land variables, Gender, Age, Education, SEI, and the five Class dummy variables. Column 1 presents the standardized coefficients (betas) when all variables are included in the regression equation. The SEI coefficient is only -.11, in contrast to three single Class dummy variables (part-timers, landlords, and farmer-tenants) that have larger betas. This indicates immediately that Class influences net farm income independently of occupational status. A stronger assertion is the concern of Hypothesis 2, which can now be investigated in more detail. As evident in Table 8, all variables in the equation explain 18.3 percent of the variance in positive net farm income (column 1). The baseline model of Land, Gender, Age, and Education accounts for only 3.8 percent. SEI increases this percentage to 9.0, but (in the absence of SEI) Class raises it to 17.6 percent. Table 9 summarizes this comparison of SEI and Class. In terms of percentage change in variance explained as each enters the baseline equation, the main status attainment variable has a net effect of less than half that of Class (5.2 percent and 13.8 percent).' For total net effect, there is no real 130 Table 8. Comparison of the Impacts of SEI and Class on Positive Net Farm Income. (Standardized Regression Coefficients) (1) (2) (3) (49 All Land, Gender, Land, Gender, Land, ender, Age,Education. .Age, Educa- Age, Educa- tion, SEI tion, Class A. Owned & .08 .15 .13 .08 Operated A. Rented -.02 .03 .01 -.02 In 8: Operated A. Rented , .06 .04 .05 .06 Out Value of .05 .07 .06 .05 A. Owned Gender -.02 -.05 -.02 -.02 (Female=l) Age -.05 —.04 -.03 -.05 Education .06 -.O4 .09 .02 SEI -.11 -.26 Capitalist .04 .04 Farmer Land- -.15 -.19 lord Farmer- -.08 -.09 Landlord . Farmer— .12 .12 Tenant Part-Time -.29 -.34 Farmer ' 2 R . .183 .038 . .090 .176 131 Table 9. Comparison of the Relative Changes in Variance Explained of Positive Net Farm Income. Equations Interpretations of Change in Variance Compared Comparison Explained (%) (From Table 8) 3—2 SEI net of Land, ’ 5.2 Gender, Age, Edu- cation - 4-2 Class net of Land, 13.8 Gender, Age, Edu- cation 1-4 SEI net of Land, 0.7 Age, Gender, Edu- cation, Class 1-3 Class net of Land, 9.3 Age, Gender, Edu- cation, SEI 132 contest. SEI alone explains merely 0.7, and Class 9.3, percent of the variance. On this point the results appear quite clear: Class is a far superior predictor of positive net farm income. Hypothesis 2 is strongly confirmed. Let us now consider Hypothesis 3, regarding the full Status Attainment model (Age, Education, and SEI) versus the Class Structure approach. Here we are not so much concerned with comparing individual variables within the same equation as with estimating the overall effect of theoretically divergent equations, with Gender included as a control in each case. Table 10 repeats the R squared of .183 for the full equation. (Appendix Table A4 presents the unstandardized coefficients.) The four Land-Size variables alone account for only 3.7 percent (column 2), while the Status Attainment model nearly doubles that amount, to 6.5 percent (column 3). The Class Structure model, however, doubles again the explanatory power attributable to the Status Attainment variables; it contributes fully 15.9 of the total 18.3 percent variance explained (column 4).. Again, the interpretation seems straightforward. .For positive net farm income, the Class Structure model vastly outperforms that of Status Attainment. 133 Table 10. Comparison of Models for Positive Net Farm Income. (Standardized Regression Coefficients) (l) (2) (3) (4) All Land-Size Status Class Attainment Structure A. owned & .08 .15 Operated A. Rented In -.02 .03 & Operated A. Rented .06 .03 Out Value of .05 .07 A. Owned Gender -.02 —.05 -.O2 -.03 (Female=1) Age -.05 -.01 Education .06 .10 SEI -.11 -.28 Capitalist .04 .08 Farmer Land- -.15 -.18 lord Farmer- -.08 -.O9 Landlord Farmer- .12 .12 Tenant Part-Time -.29 —.34 Farmer 32 .037 .065 .159 .183 134 2. Negative Net Farm Income Tables 11-13 compare the effects of SEI/Status Attainment and Class Structure on negative net farm income. Like the results for positive net farm income, these findings confirm Hypotheses 1-3. My discussion of the negative income regressions will thus be brief. Some more general observations, though, are in order. The first thing to notice is that the signs on almost all the coefficients are opposite those for positive net farm income (of. Tables 8-10). This suggests that farm income losers have a different determination process than farm income gainers. For example, operating more of one's own farm land loses money for the former while it increases the income of the latter. This finding may reflect the phenomenon of farmers aggressively expanding production but still not "breaking even." In particular, throughout the later 19703, many young farmers bought land at extreme prices and interest rates. These "costs of production" could be paid only with very high yields as 'well as product prices. Yet after 1976, farm prices dropped off sharply as interest rates continued to climb. aAnd'1977--the year that the Land Ownership Survey asked about income--was."probably the worst year for farmers since the depression" (General Accounting Office, 1980:26). One possible interpretation of these negative Vnet farm income recipients, then, is that they overexpanded and thus could not meet their ever-rising 135 Table 11. Comparison of the Impacts of SEI and Class on Negative Net Farm Income. (Standardized Regression Coefficients) (l) (2) (3) (4) All Land, Genden Land, Gender, Land, Gender, Age,Education. .Age, Educa— Age, Educa- tion, SEI tion, Class A. Owned & -.O6 -.O9 -.08 -.06 Operated ' A. Rented -.19 -.23 -.21 -.20 In 8: Operated A. Rented -.O3 -.O4 -.O4 -.03 Out Value of -.O7 -.08 -.O8 -.07 A. Owned Gender .07 ' .07 .07 .07 (Female=1) Age .03 .01 ... .03 Education -.07 -.O2 -.O9 -.03 SEI .09 .13 Capitalist -.05 -.05 Farmer Land- -.05 -.03 lord Farmer- .05 .07 Landlord ‘ ' Farmer- .02 .02 Tenant Part-Time .11 ' .15 Farmer R2 .110 . .080 .093 .105 . = beta less than .01. 136 debts--even though they kept buying and operating more land. A similar interpretation could apply as well to the high negative betas for Acres Rented In and Operated. In fact, the negative land betas overall indicate that no matter what these farm land owners tried to do with their land, they lost money. I There are other important differences between negative and positive net farm income recipients. Women benefit from being among the farm losers but are discriminated against among farm gainers. This may be explained by the fact that the largest "loser" variable, Acres Rented In and Operated, has a mean of 736 for men and only 32 for women. That is, the leading proximate reason that this subsample loses money is that they rent too much land in, and women do very little of this. Therefore, women as a whole do better than men in this net income category. In addition, both age and SEI tend to increase the income of the losers whereas it decreases that of the gainers, and education has precisely the opposite impacts. Actually, the only independent variables that act in the same direction for both samples are the class dummy variables for landlords (negative) and farmer-tenants (positive); Even here, the betas for the ‘ negative net income sample are near zero in each case. These results, then, provide strong evidence that the income determination processes for negative and positivei net farm income recipients aré substantially different. 137 Another remarkable difference between the two samples is the greatly increased explanatory power of the land variables for the net farm income losers. Although all have negative effects, the four together with gender explain 8 percent of the variance. All the other independent variables account for only 3 percent more. The implication is that land per se is much more important (i.e., costly) in determining the income of net farm losers than of gainers. Again, this is probably related to the overexpansion in terms of land among young aggressive farmers. In the late seventies, as total farm income began to drop, they tried to compensate by increasing the size of their operations. They overexpanded by buying too much high-priced land and by renting in even more (Penn, 1979). Both strategies often failed; their debt was too large to overcome by any means. Besides being interesting in itself, this outcome on the importance of land also affects the tests of the individual-level hypotheses. Since the baseline model (Column 2 in Table 11) takes up so much of the total R squared, it leaves relatively little for either SEI or Class to explain. Nevertheless, Table 11 reveals that the Class variables are indeed independent of SEI (Hypothesis 1). Further, Table 12 summarizes the comparison of the two predictors, showing that Class has a greater impact on income than SEI (Hypothesis 2). Finally, the last two . columns of Table 13 indicate that the Class Structure 138 Table 12. Comparison of Relative Changes in Variance Explained of Negative Net Farm Income. Equations Interpretations of Change in Variance Compared Comparison Explained (%) (From Table 11) 3-2 SEI net of Land, 1.3 Gender, Age, Education 4-2 Class net of Land, 2.5 Gender, Age, Education 1-4 SEI net of Land, Age, 0.5 Gender, Education, Class 1-3 Class net of Land, 1.7 Age, Gender, Edu- cation, SEI 139 Table 13. Comparison of Models For Negative Net Farm Income. (Standardized Regression Coefficients) (l) (2) (3) (4) A11 Land-Size Status Class Attainment Structure A. Owned & -.O6 -.O9 Operated A. Rented In -.19 -.23 & Operated A. Rented -.03 -.04 Out Value of -.O7 -.08 A. Owned Gender ' .07 .07 .08 .07 (Female=1) Age .03 -.02 Education -.07 -.14 SEI .09 .18 Capitalist -.05 -.15 Farmer Land- -.05 -.03 lord Farmer- .05 .06 Landlord Farmer- .02 -.01 Tenant Part-Time .11 . 15 Farmer R2 .110 .079 .033 .060 140 model is nearly twice as powerful as the overall Status Attainment paradigm (Hypothesis 3). (Appendix Table A5 presents the b's.) These findings, while not as overwhelming as with net farm income gainers, also point to the superiority of Class over SEI/Status Attainment in predicting negative net farm income. To this point, I have tested the first three hypotheses on two different samples, positive and negative net farm income recipients. 'In each of the six cases, Class has clearly outperformed the Status Attainment model. Now I turn to the third and final type of income to be examined. 3. Non-farm Household Income Table 14 compares the impacts of SEI and Class. Note that the R squared is twice as large as for positive net farm income. The independent variables explain over one- third of the total variation in non-farm income. Although some of the Class betas are increased, this dramatic rise in variance explained is due largely to the improved predictive power of Education and SEI--variables that are traditionally associated with non-farm income anyway. As before, Hypothesis 1 is supported: Class clearly exerts independent effects on the dependent variable. Hypothesis 2, that Class will be a better predictor than SEI, is also confirmed, but just barely. The difference between columns 3 and 4 is only .016. In addition, as- 141 Table 14. Comparison of the Impacts of SEI and Class on Non-Farm Household Income. (Standardized Regression Coefficients) ( 1) ( 2) ( 3) All Land, Gender, Land, Gender, Age,Education. .Age, Educa- (4) Land, Gender, Age, Educa- R .358- -l45 .293 tion, SEI tion, Class A. Owned & .01 -.05 -.Ol .01 Operated A. Rented -.03 -.O8 -.04 -.04 In 8: Operated A. Rented .02 .05 .04 .02 Out Value of .... .01 A. Owned Gender -.14 -.08 -.13 -.12 . (Female=1) Age .04 .03 .01 .06 Education .20 .37 .17 .31 SEI .29 .44 Capitalist .02 .02 Farmer Land- .22 .33 lord Farmer- .10 .14 Landlord Farmer- .05 .07 Tenant Part-Time .33 .45 Farmer 2 .309 = beta less than .01. 142 Table 15 shows, Class outperforms SEI by 1.6 percentage points. While this does support Hypothesis 2, it bodes ill for Hypothesis 3, which pits the full Status Attainment model against Class. As we would expect given the above results, SEI plus Age, Gender, and Education account for more of the variance explained in non-farm income than does Class Structure and Gender. Table 16 reports an R squared of .289 for the Status Attainment variables and .223 for Class. (Appendix Table A6 presents the b's.) Hypothesis 3 is thus not supported for non-farm income. In retrospect, this is not really surprising. The Class variables are defined on the basis of farm land ownership, yét here we are trying to predict non-farm income. The more surprising result, in this light, is that the Class Structure approach competes as well as it does with the Status Attainment model, and even outperforms SEI alone. This finding may be due, at least in part, to the class-occupation interactions. Recall from Tables 3 and 7 that the higher status occupations tend to be grouped in the landlord class--implying that class is to some extent a proxy for SEI. An alternative interpretation, of course, is that occupation is a proxy for class. In either view, the point is that there is considerable overlap between the two measures, and it . becomes especially notable when the two "compete" closely with one another, as in predicting non0farm income. 143 Table 15. Comparison of the Relative Changes in Variance Explained of Non-Farm Household Income. Equations Interpretations of Change in Variance Compared Comparison Explained (%) (From Table 14) 3—2 SEI net of Land, 14.8 Gender, Age, Edu- cation 4-2 Class net of Land, 16.4 Gender, Age, Edu- cation 1-4 SEI net of Land, 4.9 .Age, Gender, Edu- cation, Class 1-3 Class net of Land, 6.5 Age, Gender, Edu: cation, SEI 144 Table 16. Comparison of Models for Non-Farm Household Income. (Standardized Regression Coefficients) (1) (2) (3) (4) All Land-Size Status Class Attainment Structure A. Owned & ' .01 -.04 Operated A. Rented In -.03 -.O6 & Operated A. Rented .02 .07 Out Value of ... -.01 A. Owned Gender -.14 -.O7 -.13 -.12 (Female=1) Age .04 .01 Education .20 .16 SEI .29 .44 Capitalist .02 .03 Farmer Land- .22 .40 lord Farmer- .10 _ , .15 Landlord Farmer- . 05 ' . 07 Tenant Part-Time .33 ' .49 Farmer . 32 .358 .016 .289 ' .223 = beta less than .01. 145 In summary, the three individual-level hypotheses are substantially supported. Hypotheses 1 and 2, that Class is independent of, and superior to, SEI, receives confirmation with all three samples: both positive and negative net farm incomes as well as non-farm income. Hypothesis 3, that the Class Structure variables will predict income better than will the entire Status Attainment model (Age, Education, and SEI), is confirmed for the two net farm incomes but not for non-farm income, as discussed above. This single instance of a superior performance by Status Attainment is where we might expect it to offer a fuller statistical explanation than would the Class Structure of farm land ownership, concerning non-farm income. Overall, the Class Structure approach has performed quite well at the individual level. I now investigate the classes themselves as units of analysis. III. Structural-Level Results: Class Analyses This section addresses Hypotheses 4 and 5, which concern the income determination processes of classes instead of individuals. The regressians were run for each class. The means, standard deviations, and correlation matrices by class are in Appendix Tables A7-A24. 146 1. Positive Net Farm Income Table 17 presents both standardized and unstandardized regression coefficients for positive net farm income by class. The first (beta), as in the previous section, indicates the relative weight of the variable within a regression equation. The second (b), on the other hand, is the estimate of the monetary return to the variable. For example, on the Acres Rented Out variable, both landlords and farmer-landlords receive around $3.50 for every acre they lease out. However, their respective betas differ substantially. For landlords, Acres Rented Out is the most important variable predicting income (beta = .19). For farmer-landlords, two other variables are more important (beta = .08 compared to .32 and .14). So, although unstandardized coefficients may be rather similar, the relative-importance of the variables can be quite different across the various classes. Both comparisons are germane to the structural hypotheses. I deal first with the betas, then with the b's. They both tend to support Hypotheses 4 and 5. Looking at the betas in Table 17, we see that the relative importance of the independent variables varies greatly by class. The major variable predicting positive net farm income for capitalists is Acres Rented and Operated, but it has a negative impact (beta = -.17). Acres Owned and Operated and the value of that land itself are the main positive contributors to capitalists' income.‘ 147 000. 000. 000. 000.. 000. 000. 00 . Same 00.000.0 00.000 00 000.00 00.000.0 00.000.0 00.000.00 scmpnaoo 00. 00. .00.I . :0. 00. 00.000 00.00 00.000I * 00.00 00.000 000000000 00.I 00. 00.I 00.I 00.I 00.000I 00.0 00.000I 00.00I 00.0I 0 000 00.I 00.I 00.I 00.I 00.I. 00I0005000 * 00.000I 00.000.0I 00.000I 00.000I 00.000.0I 000000 00. 00. 00. 00. 00. 00. 00020 .0 0000. 000. 00. 000. 000. 000. 0c ms0m> 00. 00. 0:0 II II II 00.0 00.0 0 000000 .0 00. 00. 00. 00.I 00000000 0 00.00 00.00 00.: II II 00.I :0 000000 .0 00. 00. 00.I 00. 00. 00000000 00.0 00.00 00.0I 00.00 II 00. 0 00:30 .0 ameamm ameamm pamamB maoamama maoH 0080mm 0006mm mEaeIpamm Iamsamm Iamsamm Imama anHmufiqmo pamHOHMMmoo aofimmmawmm mmwfimammampm pamaofimmmoo aoammmawmm mmuamammampmaa .mEooaH 50mm 002 m>HuHmom pom mmmmmooam mmmao no acmaamQEoo .00 maame 1H8 Legend for Table 17. * = not significant at .01 level. -- = not applicable. 149 Landlords' income, not surprisingly, is most positively affected by Acres Rented Out as well as the value of owned land. Farmer-landlords have a somewhat similar determination process, with the major difference that they benefit very much from operating their own land (beta 0 .32). No class approaches such a high relative return to this land variable except for part-timers (beta = .25). And the second most important determinant for their income is renting in and operating land, which farmer-landlords do not do at all. Farmer-tenants, on the other hand, are the only class position to benefit most from the value of owned land (beta 0 .22). These relative measures of the impact of the independent variables for each class provide tentative support for the claim that the classes have fundamentally different income determination processes. The fairly wide range of variances explained (from 5.5 percent for capitalists to 17.8 percent for farmer- landlords) is also evidence for such an assertion. However, in order to compare the estimated monetary effects of the variables across class, we need to examine the unstandardized coefficients. The returns to the same predictor variable are duite' different depending on class. Focusing on the land V variables for the three classes most centrally involved in commercial agriculture: capitalists "cash in" very little from operating their own acres (b = .53), farmer-tenants lose considerably (b : -6.7fl), and family farmers gain 150 significantly (b = 7.78). Why such large differences? Recall that these unstandardized coefficients estimate the monetary return per unit change in the independent variable (e.g., for every owned acre operated, capitalists get $0.53). Capitalist farmers have the least valuable land (see Table 2), which indicates lower productivity, and this leads to less income per acre fror the class. (Other factors, though, are necessary to explain why farmer-tenants lose money by operating their own land. As discussed above, such loss probably reflects the overexpansion of land ownership on the part of younger, aggressive, commercial farmers--who typically rent in most of the land they operate (i.e., they are in the farmer- tenant class location). Yet the land they bought in the 19703 was so expensive, and the interest rates so high, that they could not recover the costs of production from their land. Family farmers, on the other hand, are not nearly as large as farmer-tenants (see Table 2). Apparently, they are at the right level of ownership in order to capture high returns (nearly $8.00 per acre owned and operated). _ The income of capitalists is likewise little affected by renting in and operating land (b = -.28), whereas farmer-tenants gain a good bit (b = n.53) and family farmers gain a lot (b = 10.99). Similarly, farmer-tenants receive relatively high returns to the market value of their owned land (b = .03), capitalists significantly less 151 (b = .001), and family farmers practically nothing (b = .0002). I conclude from the wide differences in the above coefficients that the classes do indeed use their .productive farm land differently to generate income. These results thus support Hypothesis u. There are also considerable differences in the effects of the other, non-land variables according to class. It "costs" women a large amount to be either capitalists (b = -67u0) or farmer-tenants (b = -7092). These two classes rent in most of the land they operate and are by far the largest operators. Women who farm very large acreages apparently do much worse in terms of income than their male counterparts. In the other classes, women are discriminated against to a lesser extent. Age, in contrast, has a large negative effect in the incomes of farmer-tenants and family farmers, both of whom operate significant amounts of land. Age does not matter to capitalists, who presumably do not work as hard physically as the other two, non-employer commercial producers. Finally, Education increases the incomes of capitalists and family farmers the most (both over $500 per year of. schooling), makes no difference to farmer-landlords, and penalizes farmer-tenants (b = -156). The differential effects of these variables by class, then, are substantial. Hypothesis 5 asserts such differences and is thus confirmed for positive net farm income. The above discussion has centered on the regression 152 coefficients, interpreted as the amount of change in income associated with a unit change in the independent variable. Another way to approach the class differences in income determination is to estimate not the effect per unit (e.g., acres) but the total effect. This measure of the overall impact for the land variables can be approximated by multiplying the unstandardized coefficient by the mean value for that variable. For example, the b for capitalists on Acres Owned and Operated is .53, and they own and operate an average of 2,608 acres (see Appendix Table A7). To get the total expected impact of this variable on capitalist income, multiply the two numbers, which gives $1,382. Table 18 presents these estimates for the land variables, three of which are measured in acres and one (market value) in dollars. The table reveals that both capitalists and farmer-landlords receive about $1,uoo from operating their own land. Farmily farmers have somewhat higher returns, nearly $1,800. In contrast, farmer-tenants lose $6u7 from operating owned acres. However, by renting in farm land, they make almost $2,300, while capitalists lose over $1,500. Only the two landlord classes obtain any money from rentingland out, pure landlords getting twice as much as farmer-landlords., All class positions benefit slightly from the total market value of their owned land, except farmer-tenants, who benefit a great deal ($2,664). .These calculated total'returns provide further evidence Table 18. 153 Total Expected Returns to Land Variables For Positive Net Farm Income, by Class. (Unstandardized Regression Coefficient x Mean) A. Owned A. Rented In A. Rented Value of & Operated & Operated Out A. Owned (:15) ($) (15) ($) Capitalist 1,382 -l,562 * 690 Farmer Land- -- -- 507 391 lord Farmer- 1,393 . -- 212 351 Landlord Farmer- -6H7 2,270 -- 2,664 Tenant Part-Time 520 54 -- 87 Farmer Family 1,766 363 -- 35. Farmer -- = not applicable. * = b not significant at .01 level. 15“ that the classes use their farm land to acquire income in quite different ways. 2. Negative Net Farm Income As with the individual-level analysis, the structural or class analysis of negative net farm income has the same theoretical conclusion as with positive net farm income: Hypotheses u and 5, positing different impacts by class of the land variables and the status attainment variables, are substantially confirmed. The discussion here will therefore be brief. Table 19 presents the coefficients resulting from the regression of negative net farm income on the independent variables. The most striking finding is that every one of the land coefficients is negative. In other words, no class position obtains positive returns for any of the four land variables. Granted, we are dealing here with negative net farm income recipients, yet it seems remarkable that all classes lose money by using land in any way whatsoever. A possible reason for this result is that land is considerably more influential on negative net farm income than on positive, as discussed in the previous section. This being the case, given that these recipients lose money in agriculture, then they most probably do so via their land. Thus, all the land measures have negative effects. An indication of the different processes governing 155 000. 000. 000. 000. 000. 000. _ 00 5009 00.000.0I 00.000.0I 00.000.0I 00.000.0I 00.000.0I 00.000.0I 00000000. 00.I 00.I 00. 00.I 00.I 00.I 00.00I 00.000I 00.000 00.000I 00.000I 00.00I .000000000 00. 00. 00. 00. 00.00 00.0 00.00 00.0 0 0 000 00. 00. 00. 00. 00. 00u000000v 00.000 00.000.0 00.000.0 00.000.0 00.000.0 a 000000 00.I 00.I 00.I 00.I 00.I 00030 .0 000.I 0000.I 00.I 000.I 0 000.I 0o 0000> No.1 00.I 050 II II II 00.I 00.I 0 000000 .0 00.I 00.I 00.I 00000000 0 0 00.0I 00.I II II 00.I 00 000000 .0 00.I 00.I 00.I 00.I 00000000 00.0I 00.0I 00.0I 00.0I II a 0 00030 .0 005000 005000 020:09 00000500 0000 005005 000500 050910000 1005000 1005000 Iccmq 0000000000 00000000000 COHmmmpwmm 0000©Lwocmpm 05000000000 £00mmmpwmm cmNacnmocmmeD .mEooCH 5000 002 0>0pwwwz 00% 000000000 00000 mo 0000000500 .00 00209 156 Legend for Table 19. * = not significant at .01 level. -- = not applicable. 157 net farm income losers as compared to gainers is that almost all of the coefficients have opposite signs between the two samples (cf. Table 17). Only Acres Owned and Operated for farmer-tenants, Acres Rented In and Operated for capitalists, and Age for landlords and part-timers work in the same direction in their influence on net losers. The most remarkable overall shift is probably the Gender variable. In direct contrast with net gainers, women who lose money in agriculture are benefitted by their gender in relation to men; all the Gender coefficients are positive, indicating that women have higher net farm income returns than do men. I have no ready explanation for this dramatic shift in the effect of being female. The main point about negative net farm income recipients, though, is that it still matters which class one is in. The betas, which estimate the relative importance of the variables for a given class, differ by class. The largest betas are even more widely distributed than for positive net farm income. Only two classes, farmer-landlords and family farmers, have the same variable making the greatest impact on their income (Value of Acres Owned). The least important variables for each class are also generally different, as are the variances explained, ranging from u percent to over 30 percent. This, again, suggests that the classes have different income determination processes: 158 Although all the land variables have negative impacts (they all contribute to the loss of farm income), the classes still vary quite a bit in monetary returns. The b's for the Acres Owned and Operated variable range from statistical insignificance to -7.00; Acres Rented In and Operated, from insignificance to -3.15; Acres Rented Out, from -.36 to -.76; and Value of Acres Owned, from -.0003 to -.02. Age is somewhat more varied by class, from farmer-landlords (b = 6.78) to farmer-tenants (b = 86). The Education coefficients are all negative except for farmer-tenants, and the range is smaller (-31 to -157). Yet these differences are substantial enough to support Hypotheses u and 5, if not as strongly as with positive net farm income recipients. Table 20 gives the total expected returns to the land variables. As noted above, all the impacts are negative. However, the losses due to land are not that great. Family farmers, for instance, lose over $800 by operating their own land, and capitalists lose slightly more by renting in land. Farmer tenants and family farmers lose nearly $1,000 from the value of their land; capitalists and farmer-landlords, about half as much. The total effects, then, generally differ by class. The main thrust of this table, though, is that the land variables per se have relatively little overall impact on the income of. negative net farm income recipients. Table 20. 159 Total Expected Returns to Land Variables For Negative Net Farm Income, by Class. (Unstandardized Regression Coefficient x Mean) A. Owned A. Rented In A. Rented Value of & Operated & Operated Out A. Owned ($) ($) ($) ($) Capitalist * -873 * -53A Farmer Land- —- -_ -109 * lord Farmer- -279 -- -31 -671 Landlord Farmer— -472 -565 -- -958 Tenant Part-Time -398 -22 -- -17 Farmer Family -821 * -- -927 Farmer * = b not significant at .01 level. -- = not applicable. 160 3. Non-Farm Household Income Hypotheses A and 5 for non-farm income are somewhat similar to negative net farm income: they are supported by the data, but not as vigorously as for positive net farm income. Table 21 presents the results of the regressions. Looking first at the betas, we note that the highest coefficients are all for either Education or SEI, while most of the lowest ones are Acres Rented Out or Value of Owned Acres. The unstandardized coefficients give a different portrait of the class differences for non-farm income. Acres Owned and Operated ranges from - 6.37 (farmer-tenants) to 5.68 (part-timers), with both capitalists and family farmers being about zero. 0f the four classes that rent in land to operate, three have negative returns, but the range is wide (-.08 to -13.32). Family farmers appear to benefit slightly from renting in. land (b = .52). However, Table 22 shows that the impact of the land variables on non-farm income is rather small, with the partial exception of the capitalist class. The most significant effects are that capitalists obtain nearly $1,000, and farmer-tenants losé half that much, from the Acres Owned and Operated variable" Capitalist farmers also'lose over $500 by renting in land. All the class positions except farmer-tenants gain a slight amount from the market value of their'land. Yet, overall, the impact of land on non-farm income is minimal} There is little 161 om. mm. o:. Hm. oo. nu Am.oMH em.omo Hm.omm mo.HmH om.o~m Hmm ma. om. ma. am. no. . om.ao: HH.HmN.H * Ho.ooo mo.ooo.a om.omo cooomooom HH. mo. oo.u mo. oH. oo.om om.oa * mm.omu Afi.om om.moH two mo. ma.u oa.u Am.s mo.n AauonEmmo No.5mm mo.ooo.ou * oo.moo.ou oe.oom.mau oo.:mfl.ou amocmo Ho. Ho. mo.u oo. Ho. mo. omczo .< moooo. mooo. Hoo.u moo. mooo. moood co moam> Ho. mo. mo. . poo I- n- us No. . 3H.H om. omocom .<. Ho. mo.u oo.u oo.u omomnmoo o mm. mm.ma- oH.u u- u- oo.-. :H omecmm .< Ho.n mo. no.: so. .oH. omompmoo :H.u oo.m om.on oH.m .1 mm. o omezo..< hospwm pmeumm pcmcme whoaccmq whoa hmEAmm mafismm mefielphmm ILmEmmm Immenmm Icsmq pmfiampfiamo ucmHOHmmmoo scammmpwmm cmufivpmpsmum ucmfiofimmmoo coammmnwmm UmmfiopmocmmeD .mEooCH caonmmsom Ehmmucoz pom mmmwmoopm mmmao no comHLmQEoo .Hm manna 162 .maanaaaqw no: n In .Hm>ma Ho. pm acmOHMchHm no: u * :mo. Nod. oom. mom. oam. omm. mm . same *Hm.ooo.mu ofi.aoo.ou *o:.om oo.mom.ou mm.o~o.aau om.omm.oaa ocoumcoo smegma LmEme panama whoaccmq whoa pmemmm zafismm mEHBIppmm lamenmm InmEpmm Iccmq umfiamufiamo .Ao.ocooo Hm magma 163 Table 22. Total Expected Returns to Land Variables For Non- Farm Household Income, by Class. (Unstandardized Regression Coefficient x Mean) (A. Owned A. Rented In A. Rented Value of & Operated & Operated Out A. Owned <$> <$> _ ($) ($) Capitalist 959 -577 36 190 Farmer Land- -- -- 152 21 lord Farmer— 350 -- 71 276 Landlord Farmer— -529 -l63 -- -70 Tenant Part-Time 375 -53 -- 16 Farmer Family 33 18 -- 3 Farmer -- = not applicable. 16H reason to expect otherwise. To continue with Table 21: The variable that exhibits the most difference by class is Gender. Women are disadvantaged in four of the classes, but, again, this ranges from over $A,000 (capitalists) to over $12,000 (landlords). Gender makes no difference in the farmer- tenant class, whereas women in the family farming class are benefitted in non-farm income. Age is not as differentially rewarded. Only farmer-landlords receive negative returns to increased age, although it has no effect on farmer-tenants. Every class except farmer- tenants benefit from more education, with highest rewards going to landlords (b = 1910). Similarly, all classes acquire more money from higher occupational statuses, capitalists getting the most return and part-timers the least. Finally, we should note the different amounts of variance explained for each class, ranging from a little over 3 percent for family farmers to nearly 40 percent for farmer-landlords. One reason_that the R squared for family farmers is so low is that the major explanatory variable, SEI, is constant in that class (farmers = 1A). These results indicate that Hypotheses A and 5 receive moderate support with regard to non-farm income. Class continues to make a difference in the variable effects, but not nearly so much as it did with net farm income. 165 IV. Structural-Level Results: Gender by Class Hypotheses 6 and 7 deal with gender differences within class as compared to the differences if class is not considered. Hypothesis 6 follows all the above hypotheses in that it contrasts regression coefficients. Hypothesis 7, however, is unique in that it simply compares the mean incomes of the genders by class. This section is also different since it treats only two types of income: positive net farm and non-farm household. There are not enough women in several classes of negative net farm income recipients to make the comparisons needed for this set of hypotheses. The means and standard deviations, as well as the correlation matrices used to evaluate Hypothesis 6, are in Appendix Tables A25-A36. 1. Positive Net Farm Income Table 23 presents the unstandardized regression coefficients by gender within each class. Hypothesis 6 holds that the returns for the land and the status variables will be similar for both men and women within each class. One immediate indication that this may not be the case is the large differences in variance explained of incomes for men and women in almost every class. The smallest differential is over 6 percent (farmer-landlords) and the biggest is 53 percent (farmer-tenants)! (This latter difference may be attributable to the small sample of women in the farmer-tenant class who were net farm 166 .~o>mm .mommo am can» mama u + .oanmoaagqm no: n In AD. as acmoauacmam uo: u - omm. mom. one. mom. Hom. omm. :msoz omo. mam. omo. ooH. moo. ooo. cm: mm mmz.am Hom.m oom.o~ mam.m moo.m- omo.m= case: same mom.o mosi oom.mm ooo.= m=H.: mmm.mm cm: scmomcoo oo.omo.m- mm.oomu mo.m=on . mo.mmm . case: oo.o:o mm.mo Hm.=omu . Hm.mm oo.omm co: comooooom om.=mm- oo.o~- mm.oom- oo.mm- oo.mm om.Homu case: HH.Ho- so.mm oo.omH- Hm.mmu oo.mmu m :62. om< mo. mo. mo. mo. mo. . case: ooczo .< mooo. moo. mo. moo. moo. moo. co: co moam> .. u- 1: om.m m:.H + nose: ago 1. u- .. o=.m oo.m . so: ooocmm .< + + mo.o u- u: . + :meo: cmomaoao a om.o oo.mm om.z u- u- om.u cm: :H ooscmm .< mm.m- mo.m m~.m mo.m- us mm.~ case: casebooo o =o.o mm.oH oo.ou oo.Hm n. m:. cm: emczo .< smegma LmEme ucwcme ULoHccmq choa Loahmm hamemm oefielucmm ILoEme Inoemmm Imama umuamuaamo Amocomouummoo codnmoawom conacpmocmumcsv .chocH Spam uwz m>auamom new: :mEo: ccm :0: no: mmmnmoopm mmmao no :Omfismosoo .mm magma 167 income gainers, only 25.) In any case, a close analysis of Table 23 fails to offer strong support for Hypothesis 6: men and women within the same class appear to "cash in" somewhat differently on both the land and the status variables. Let us examine the more intriguing differences, some of which are contrary to my theoretical predictions. Within the capitalist class, women gain three times as much per acre of land owned and operated as do men (b's = 1.32 and .HS). Hen receive significant returns to both the market value of the land they own and to Education whereas women do not. Age, on the other hand, disadvantages women while it has no effect on the income of male capitalists. In the landlord class, Age works against increased income for men and in the opposite direction for women. The other three variables which affect landlord income are of the same sign for both genders, but they exert significantly divergent impacts. Men's returns to Acres Rented Out--which defines landlords--is over twice as large as women's. Women's returns to Education are over five times as great as men's. The differential effects between men and women landlords who also operate part of their land are not as great, yet they are still important. For example, farmer- landlord males have large, positive returns to Acres Owned and Operated (b = 21.60), but females in that class have moderate, negative returns (b = -5.05). The Education‘ 168 variable, however, does have similar impacts on both of them (b's = -25 and -20), and Age is fairly similar (b's = -37 and -15). Farmer-tenants are the most alike by gender. Only two of the relevant variables have substantially different effects, Acres Owned and Operated (which have opposite signs) and Education (with women having negative impacts nearly three times as great). Male and female part- timers, on the other hand, are different on every measure. Age and Education both affect them in opposite directions; men gain thrice more by operating their own acres, but women benefit mUch more from the value of the land they own. Family farmers, finally, are also quite different by gender. Hen benefit from Acres Owned and Operated while women lose; Agenegatively influences the incomes of both, but women much more so; and Education greatly helps men (b = 6H7) while hurting women (b = -1079).' The conclusion here is equivocal: some of the class positions significantly mediate the effects of gender while others do not. For instance, for farmer-landlords and farmer-tenants, the returns to all of the land variables (except Acres Owned and Operated) as well as to‘ Age and Education appear to be quite similar for men and women. The other classes, however, seem to have different variable impacts according to gender. Hypothesis 6 thus does not receive enough support to be fully confirmed.' Wright also reaches this conclusion. He found that 169 significant differences exist between the income determination processes of men and women even when controlling for class. Interestingly, this gender differential is much greater than for race; sexism may be less "class-bound" than is racism (Wright and Perrone, 1977:50; Wright, 1979:219). In any case, classes themselves do not appear to mediate most of the effects of gender on positive net farm income. This point is reinforced by Table 24, which presents the total impacts of the land variables for each class by gender. As in the above analysis, men and women in the farmer-tenant class are the most alike. They also receive the largest total returns to land, $u,346 for males and .$3,337 for females. Both genders in the capitalist class also obtain significant "pay-offs," especially for operating their own land; women get over $3,800 and men over $1,100. On the other variables, though, capitalists differ substantially by gender, as do the other class positions on practically every measure. Therefore, the overall expected returns to land for positive net farm income recipients fail to support Hypothesis 6. Hypothesis 7 takes another approach to the class- gender relation. It posits that the average mean incomes of men and women within a particular class will be closer to each other than will the overall gender difference in income (i.e., not taking class into account). Table 25 presents the relevant data. It appears that this 1 170 Table 2“. Total Expected Returns to Land Variables for Men and Women With Positive Net Farm Income, by Class. (Unstandardized Regression Coefficient x Mean) A. Owned A. Rented In A. Rented Value of & Operated & Operated Out A. Owned ($) - ($) ($) ($) Capitalist Farmers . Men 1,16H -1,683 * 670 Women 3,839 + + * Landlords Men -- -- 538 312 Women -- -- 190 1,709 Farmer- Landlords . Men 1,858 , -- 188 252 Women -232 -- “93 1,739 Farmer— Tenants Men -685 2,330- -- 2,691 Women 3 202 1,001 -- 2,13A Part-Time Farmers . Men 536 75 -- 86 Women 172 + —- 926 Family Farmers Men . 1,875 328 -- 36 Women -919 + -- 2,694 * b not significant at .01 level. + II II sample size too small (less than 30) to estimate b. -— = not applicable. 172 Table 25. Positive Net Farm Income of Men and Women, by Class (Means). ' Men Women Women/Men Ratio ($) ($) (%) A11 5,8u9 3,527 60.3 Capitalist Farmer 18,922 1N,h89 76.6 Landlord D M,OM7 3,351 82.8 Farmer—Landlord 4,657 3,256 69.9 Farmer-Tenant 14,77u u,785 32.u Part-Time Farmer _2,189 1,770 80.9 Family Farmer 10,999 10,510 95.5 172 hypothesis is largely confirmed. Average women, regardless of class, have 60 percent of the positive net farm income that average men do. Analyzing men and women by class, however, raises that ratio considerably, from 70 percent (farmer-landlords) to 95 percent (family farmers). There is one exception: female farmer-tenants make only 32 percent as much as males in that class. The main point, though, is that the positive net farm income differentials by gender are significantly reduced in every class except one, compared to the overall gender differential. Hypothesis 7 is largely confirmed. 2. Non-farm Household Income Table 26 shows that the gender differences within classes for non-farm household income are not as great as for positive net farm income. (One indication of this is the general similarity of the variance explained for men and women in each class position--much closer than in the above section. Hypothesis 6 thus receives somewhat more support with non-farm income. ,Yet, it still falls short of solid confimation. . _ ' Capitalists are largely alike on the land variables. They differ by gender mainly in the returns to Age (men are'"positive," women "negative") and SEI (men are highly benefitted, women are not affected). Landlord returns are in the same direction for men and women with regard to their definitive feature, Acres Rented Out, although the 173 .uouao cm.:mzu one“ I + .ounsouuqqa ac: I I- .Ho>o~ ac. am assouuncmun yo: I - omo. wmm. mam. mmm. mmm. cam. nose: smo. mas. mam. mom. mom. mom. co: m: .omm.ma- ~o=.mm- om~.~m .mmo.m .om..- om..o- ease: snow mmm.m- omm.m- ocm- omm.o- omm.~m- .mo.od- so: announce - om.~m~ om.oc~ om.oom .m.mm . case: - as.~=~ mm.mma oo..am .m.oo~ ma.oom as: How oo.mom =H.mmm.H mo.mmm oa.ooo mo.mmm.m o=.oom.m cone: mo.ooa am.s=m.~ - oo.mmo.m o..moo.m ...moc so: comsaosem =o.mm m=.oom oo.mom- mm.mm- sm.~o- .m.mmm- coco: os.oo o~.om om.o ms.ma- m~.mom mw.o- co: om< moo.- . Ho.- moo. moo.- . case: gonzo .< moooo. . moo.- moo. mooo. mooo. no: mo ooma> - - - oo.m- om.m + case: use - - - am.H oo. mm. :o: eoscom .< o=.mm + so.m - - .m.- cone: .eosaaoao . . Hm.m- o~.- - - oo.- no: no case»: .< oo.m mo.m- ao.mm- mm.m - mm. case: eosanoqo a mm.- mm.m -.w- s:.m - am. no: macro .< possum possum access escaccaa Oped Luanda aqusmm mama-usmm. unoEme shoanom 10:09 . uuumaauaao Anacououuuooo coaun0hmuz condonoccmaocav .oEoo:H uuozousoz Esau-co: can) solo: ecu :0: non nounOoOLL mango ho coouhmnsoo . mm ca 99—. 17u coefficient for females is over three times the size. Landlords differ substantially by gender on land value and Age; on the other variables they are similar. As with net farm income, farmer-landlords and farmer-tenants are the most alike by gender. Men and women in the former class are quite close on every variable except Acres Rented Out, which is, however, a definitive characteristic of this class. Likewise, farmer-tenants differ by gender on the crucial variable, Acres Rented In and Operated, yet have the same sign on most of the other measures; women generally have larger coefficients. Both part-timers and family farmers have comparable returns to the status attainment variables but not to land. As in the previous analysis, then, Hypothesis 6 is only partially supported for non-farm income: some class positions have similar income determination processes for both men and women whereas others do not. Table 27 presents the total effects of the land variables by class and gender. Capitalist women average nearly $2,500 by operating their own land while men bring in $800. Landlords generally benefit from renting land out, if only slightly. The largest returns are to female farmer-tenants, who lose over $1,200 by operating their own land but earn nearly as much by renting in land. I cannot explain this dramatic turnabouto-unless women in this class are an exaggerated version of the aggressive commercial farmers who overexpanded by buying too much 175 Table 27. Total Expected Returns to Land Variables For Men and Women With Non-Farm Household Income, by Class. (Unstandardized Regression Coefficient x Mean) A. Owned A. Rented In A. Rented Value of & Operated & Operated Out A. Owned ($) ($) ($) ($) Capitalist Farmers Men 813 -596 36 308 Women 2,ugu -727 + ' x Landlords ‘ Men -- -- 136 22 Women -- -- 357 -75 Farmer- Landlords Men 399 -- 107 293 Women 202 '-- -307 162 Farmerb Tenants Men -504 -l67 -- -70 Women -l,23A 1,209 -- -647 Part-Time Farmers Men 399 -63 -- * Women —205 + —- * Family Farmers Men . -79 '* -- 5 Women 557 356 -- -207 * = b not significant at .01 level. + = sample size too small (less than 30) to estimate b. -- = not applicable. 176 land at very high prices and now compensate by leasing land for increased net returns. If this is the case, however, then why don't men to it, too? I do not know. The upshot of Table 26 backs up the above negative evaluation of Hypothesis 6: For most of the class locations, men and women appear to have different income determination processes. Unlike with net farm income, however, Hypothesis 7 is also rejected for non-farm income. Table 28 reports the overall mean non-farm incomes for men and women, the ratio being 70 percent. The class breakdowns reveal only two classes for which the gender differential is lessened. Farmer-tenant women make 82 percent of the men's income in that class. Most remarkably, females in the family farmer class have higher non-farm incomes than do men, the ratio being 129 percent. Other than these two, however, the ratios are smaller than for men and women in general. Landlords are the lowest (“1 percent); capitalists, farmer-landlords, and part-timers all have female/male income ratios around 65 percent--lower than the overall figure of 70. Thus, Hypothesis 7 is not supported for non-farm income. To conclude: Of the four tests of the gender4class hypotheses, only one was convincingly confirmed. For ' positive net farm income recipients, class does appear to mediate absolute levels of gender differences in income.‘ 177 Table 28. Non-Farm Household Income of Men and Women, by Class (Means). Men ‘ Women. 'Women/Men Ratio ($) ($) (%) A11 13,375 9,418 70.4 Capitalist Farmer 7,620 5,074 66.6 Landlord 22,315 9,265 u1.5 Farmer-Landlord 14,770 9,458 64.0 Farmer-Tenant 7,236 5,935 82.0 Part-Time Farmer 19,594 12,952 66.1 Family Farmer 3,113 4,022 129.2 178 This is not the case with non-farm income. Furthermore, the income returns to the predictor variables are usually rather different by gender within the same class. Thus, we cannot claim that class effectively accounts for the different income determination processes of men and women. CHAPTER SEVEN: SUMMARY AND CONCLUSIONS This study is basically an extension of Wright's (1979) class structure perspective on income determination. However, it is not simply an extension because of the subject matter--1and ownership. This leads to some divergence from Wright. He analyzed a random sample of the 0.8. working population, and found, for example, that individuals' incomes rise as their SEI scores increase. In contrast, my analysis indicates that as SEI scores go up, net farm incomes decline. This otherwise inexplicable finding is easily understood once we realize that my sample consists of farm land owners and so is not representative of the 0.5. population as a whole. That is, many farm land owners are farmers, who have a low SEI measure (14), but who also tend to have high net farm incomes. Occupational status, then, plays a somewhat different role in the income determination process of farm land owners than it does for other people. In this sense, many of my results do not parallel those of .Wright. Moreover, whereas Wright relied mainly upon authority relations to designate classes, I focus upon the social relations of property ownership itself. Carchedi (1980), for one, criticizes Wright for overemphasizing workplace 179 180 domination while excluding ownership. At the end of his study, Wright (1979:224) admits the need for examining the propertied. Some time ago, Stinchcombe (1961) urged the priority of property relations over occupation, at least for the study of rural social stratification. Newby (1978, 1980) recently repeated this suggestion. ,My investigation of farm land ownership took place within this context. As laid out in Chapter One, land has several characteristics that make it unique among the means of production. The relative fixedness of land--in quantity, quality, and location, for example--in conjunction with the social institution of private property leads to a distinctive class structure of farm land ownership, and therefore of agriculture, in capitalist societies (Newby, 1978, 1980). My theoretical model of this class structure is an initial effort at specifying some of the more important economic dimensions of owning farm land. It is certainly not an ideal specification. I think, though, that the class structure model of farm land ownership presented here makes a contribution to the "new" sociology of agriculture. Next I employed this model in an analysis of income determination. First, I compared it to the standard sociological explanation of stratification, the status attainment paradigm. Wright (1979) found that class compares favorably with occupational status in predicting 181 individual income. I wanted to see if this would hold for farm land owners, a very different population than Wright studied. Actually, I used a much more stringent "test" of the class structure model. He pitted it against only occupational status while my comparison included the additional attainment variables of age and education. Since farm land owners usually receive income from two sources, farm and non-farm, I applied the competing approaches to both types. For non-farm income as well as net farm income, class outpredicted the status attainment variables. These results furnish strong empirical support for the class structure view. I then sought to discover whether class itself makes a difference in income determination processes. These were my structural-level hypotheses that the effects of both land size measures and conventional determinants of status would vary significantly by class. The findings here also provided evidence in favor of class analysis, although considerablymore so for net farm than for non- net farm income. That is, the regression coefficients differ more by class with net farm income. Finally, I investigated class mediations of the "returns" to gender. Do men and women within the same class have similar income determination processes compared to those across class? Wright (1979) found that this is generally the case. But most of my results for farm land owners indicated that class is not a very important mediator of the proposed 182 gender-income relation. Men and women appear to have substantially different processes for acquiring money, regardless of class location. This suggests that by and large gender is not effectively "structured" by class. This set of results was my main negative findings. For the most part, however, the structural-level analyses reinforce the class approach to income determination. What does this imply for our understanding of social stratification? Most especially, does it mean that occupation and education are unimportant in determining an individual's income? No; in only a few cases were the effects of these variables insignificant (cf. Wright, 1979; Crompton and Gubbay, 1978:149—65). Throughout this work, I have discussed the two approaches as though they were mutually exclusive. Several sociologists maintain that this is not necessarily so, for two different reasons. First, status attainment researchers deal almost exclusively with the individual level of analysis. They determine how people are sorted into various positions-- assuming, often implicitly, the given social structure of "empty places." This line of argument holds that there is nothing intrinsically incompatible with a class structure that-is filled with people according to their individual characteristics such as educational attainment (Kalleberg and Griffin, 1980; Kalleberg, 1982; Bielby, 1981; Bielby and Kalleberg, 1981). Wright (1979) concludes, for 183 example, that educational impacts are structured by class position. This is somewhat akin to seeing stratification as comprised of resources and rewards (Coleman and Rainwater, 1978; Parkin, 1971:33). Productive property or wealth is an instance of the former while prestige or income exemplifies the latter (of. Wright, 1980c). The dichotomy can be extended to include Marx's emphasis on production and Weber's on distribution (Crompton and Gubbay, 1978:16, 153; Saunders, 1978:68; Giddens, 1973:75- 96; Parkin, 1979:3, 11). Second, some sociologists identify two distinct structures in the economic realm, class and occupation (e.g., Bottomore, 1968:295; Hill, 1973:175; of. Wilson, 1978, 1981). Following Wright (1979), Kalleberg and Griffin (1980:735) see classes in terms of the social relations of production and occupations as similar "technical activities within a division of labor." They do not, though, restrict occupation to occupational status and in fact extend the concept to include other measures of occupational differentiation such as complexity of involvement, relative cognitive development, and training time (of. Kalleberg et al., 1981; Bielby and Kalleberg, 1981). Thus, there may well be two somewhat overlapping but theoretically distinct social structures that govern the productive organization of society, one occupational, , the other class-based. If this is correct, it still is reasonable to inquire 184 into the relative impacts of class and occupation. Which is the better predictor of income? My research addresses this issue and the answer is fairly straightforward. First, at the individual level, the class structure of farm land owners is evidently more important than even the combination of age, education, and occupation. Secondly, it is also apparent that class mediates the impacts of the land-size measures as well as the status attainment variables (but not of gender). Therefore, I conclude that class substantially influences both individual-level characteristics and the occupational structure itself. Class is evidently more important than occupation in structuring the income determination process. Let me now summarize some of the major limitations of my work. First of all, it is a very incomplete form of class analysis. As mentioned at the end of Chapter Three, the Marxist concept of class includes not only economics but also politics and culture. This research dealt only with the economic structure of farm land ownership. One obviously missing element is the political determinants of farm income. Throughout 0.8. history, and especially since the New Deal, the national state has played a vital role in agricultural development and in income policy in particular. Yet I had no data on this impact--a serious shortcoming indeed since I was trying to explain the income determination process. Therefore, I do nbt claim 185 to have offered a full class analysis of farm land owners. A related limitation is that my data are for only one point in time. This leads to several problems. One is the volatility of farm income. The year that income was reported in my data (1977), was probably the worst year for farm income since the 19303 (General Accounting Office, 1980:26). This situation could give a highly biased view of the incomes of farm land owners. Moreover, my work lacks an historical perspective on 0.8. agriculture. How did the six class positions arise? Which are just now emerging, and which are declining? My cross-sectional data cannot begin to answer such questions. Finally, I am unable to account for most of the gender differences that my results point to. Men and women within the same class appear to have quite different income determination processes--contrary to my hypothesis. I had reasoned that much of the income differences between males and females would disappear if their class location were held constant. By and large, this did not occur, and I cannot adequately explain why. Such an inability is often the outgrowth of an approach to sociology that Mills (1959) castigated as "abstracted empiricism." Without a detailed, concrete knowledge of one's subject matter, the numbers tend to dominate the research. Real social relationships give way to abstractions. What suggestions, then, could I offer for further O 186 research in this area? At the empirical level, I would ask for better measures of class relations. The most glaring deficiency in this work is the lack of information on the employment of wage-labor. Data on what kind of farm is owned (e.g., fruit, wheat, livestock) would also be extremely useful since type of enterprise varies so much, especially at the national level. More information on the jobs of the non-farmers would allow us to locate them in the larger class structure of the 0.8. Questions on the reasons that they own farm land would also help us to understand their motivations. What are they hoping to get out of the land? This would permit a comparison of the subjective attitudes of the different classes. In light of the above self-criticisms, a number of other recommendations can be made. Historical examination of the various class positions could shed light on the sociology of property relations. For instance, a regional study comparing the rise of part-owner-operators and the decline of tenant farmers in the South would be instructive. Such an investigatibn should take note of the role of the New Deal in transforming the old plantation system. How were those public policies made, and who benefitted from them? Furthermore, what place did the sharecropper unions of the thirties have in hastening their own decline? Did they realize the eventual outcome of their protests?. These queries hint at the sort of historical class analysis that could be performed. 187 My work holds some implications for agricultural policy. For example, the issue of the separation of the ownership and control of farm land has recently been raised within the U.S. Department of Agriculture (1981; cf. Raup, 1980). Should non-farming owners of farm land be allowed to gain control of the agricultural sector, or should operating farmers increase their own power? Does nominal ownership imply actual direction of farm land use, or are landlords often dominated by their large, commercial farming tenants? The structure of agriculture has long been moving away from a family labor farming system and toward more capitalistic arrangements. Which farm land owning classes are in ascendency? Which should receive public support? The national state plays a significant role in farm incomes. My results would suggest that it matters a great deal whether government payments are tied to land or to enterprise--and that different classes are differentially affected by income policies. I shall conclude with some further theoretical reflection on property relations. Of my two basic aims in this work--to develop a theoretical model of the class structure of farm land ownership and to employ it in an empirical analysis of income differences among farm land owners--the results from the latter are encouraging. However, even though the model did perform rather well 188 empirically, I am not completely satisfied with its conceptual basis. It is not so much the delineation of the class structure of farm land owners that is problematic, but rather the several theoretical foundations upon which it is based. Specifically, there is an unfortunate lack of interaction among the three perspectives on property: Marxism, Weberianism, and institutional economics. This issue is not directly tied to income determination although it holds implications for such. Everyone agrees that property ownership contains the legal claim to income. For instance, a leading Marxist economist (Mandel, 1968:283) approvingly quotes the following institutional-economic view of land ownership: "The buyer of land is actually buying the right to receive a series of annual incomes..." (Renne, 1947:215). Let me elaborate upon the broader conceptual problem of property ownership with respect to the further integration of the three theoretical standpoints. As pointed out in Chapter One, Pryor's (1973) analysis of property highlights economic inequality and domination. This institutional-economic focus on the power relations of property strikes a sociologist-as being straight out of Weber. Yet the relationship between the two is unclear. Institutional economists occasionally claim Weber as one of their own, but my interest here is not mainly intellectual history. Rather, what are the similarities and differences between-Weber's sdcial 189 economics and that of the institutionalists? True, both view property ownership as including a claim on income. This point, however, does not exhaust the amount of overlap. As regards farm land, both institutional agricultural economists and Weberian rural sociologists focus on the concept of tenure. Land tenure analysis can be a formal, arid exercise, as, for example, in the simplistic "bundle of rights" concept of property. The problem with this notion is two-fold. On its own terms, it usually fails to differentiate any hierarchy of rights; they all seem equally important. Secondly, from the broader standpoint discussed above, property is not just "rights." As the best practioners of both subdisciplines stress, property is not essentially limited to income rights. They focus instead on actual control of farm land (e.g., Harris, 1958, 1974; Breimyer, 1977; Raup, 1980) and, more generally, on the role of power in agricultural property relations (Newby et al., 1978; Newby, 1980). Sometimes they also broach the subject of the "social relations of production." 9 ‘ This phrase is hat a perfectly clear one. The version I have employed is Wright's (1979, 1980b), but there are competing renditions. Bowles and Gintis (1974:33; cf. 1976:55-61) state: By'the "social relations of production" we mean the system of rights and responsibilites, duties and rewards, that 190 governs the interaction of all individuals' involved in organized productive activity. Giddens (1977:206) distinguishes four such social relationships: paratechnical, authority, market, and consumption (of. Giddens, 1973:86, 1981:299). Therborn (1976:376-86) discusses three aspects of the relations of production: distribution of the means of production and means of subsistence, the objective of production, and the structured social relations between workers and appropriators. Cutler et a1. (1977:243-62) state simply that the effective possession of, or separation from, the means of production constitutes the basis class relationship. Many of these definitions overlap to a considerable degree, but they are far from being identical. Wright and others, as outlined in Chapters One and Four, break the capitalist social relations of production down into three dimensions of control over property, real economic ownership (resource investment decisions) and possession (of both the physical means of production and the labor-power of others). These concepts relate closely to the institutional economists' sophisticated formulations of property (e.g., Pryor's (1973) control rights) and land tenure (Mayer et al.'s (1969) access--the "decision-making prerogatives over land.") Do these labels point toward different--or the same--aspects of the ownership and control of production? Writing from an 191 ostensibly Marxist point-of-view, Goss et a1. (1980:96) argue that tenure categories tend to obscure instead of advance class analysis. My work suggests that this may not in fact be the case. The conclusion to be drawn from both the theoretical and empirical analysis is that tenure is a useful indicator of economic class position, of one's location within the social relations of production. Finally, what of Marx and Weber themselves? Saunders (1978, 1979:66-102) presents concise summaries of the issues. Weber studied tenure and property ownership chiefly as exchange situations that could lead to income via the market. This was central to his analysis of property classes. Marx, in contrast, was relatively unconcerned with tenure and property as such. Only as these implied the exploitation of labor did his attention turn to them. While Marx (1973) commented extensively on the changing place of property in history, his primary interest was in class relations--the social relations of production (Hunt, 1979). This concept still seems to me to be the best one available for investigating class structures and their social-economic implications. . Many of these issues, however, remain unresOlved. 'We are still debating "what Marx (and Weber) really said." But, more importantly, we need to move forward to deal with substantive matters, whatever they are called: powsr, control, ownership, or exploitation. Then we can develop a more satisfactory sociology of property, APPENDIX .1592 .so. can» duos . . omols m:. o:. mm. mm. mm. as. mm m mm mm. Amwm. 1 mom doom oom .o.m oomo mm. om. AH. mo. om. mo. om mm om co. Amuwv .mm mmm mmm u ... am. om.- mm. mo.- oo.- ms. mm.- mo.- mo.- oo.- oo. mo. mo. as. oaoocn sans so: .mao .. m=.- os.- ms.- am.- oo.- om.- mm.- om. .o.- mo. mo.- .o.- oa. gonads sud-us logo ... om.- oa.- om.- oo.- om. ma. mm.- mo.- mo.- -.- mo.- o~.- gonads onus-suns Ammo ... oo.- om.- so.- .oH.- mo.- as.- so.- . mo.- oo. mo.- accede-sounds .mmo ... oH.- mo.- Ho.- ao.- mm. .o. mo. no. mo.- Ho.- caduceus-posses Ammo ... oo.- om. ms. oo. om. . mm. .o.- om.- oaomecus .omo ... so.- so. no. mo.- oo. mo. on. on. possum snodauadao .o. ... on. .H.- an. no.- oo. oo.- o~.- . Hum Aoo ... om.- mo. so. mo. . . comaaoseu A». ... so. no. so. mo.- mo. om<.AoV ... so.- no. mo.- mo.- AH - «masons soecoo Am. ... oo. mo. ma. eocso .< co o=Ha> lav ... . on. use cacao: .< and ... mu. cousaono . cm guano: .< Am.- ... couscous . cacao .< .m. solo lass Amao Ammo A.Ho loss Aoo Aoo as. Amy Amy Amy Am. .mo .av .osoo:H sass no: o>uuuuom-no~nmuss> -< sou nucououuuooo sodasausaoo can .ucouusu>on cascade» .ncao: .~< ounce 193 .mo. can» anus . . Aoooo Sow om. o... m... mm. mm. 3. mm m 3 mm. 5.... H: on: 23 5.» slow. oi. o3. mm. mo. oo. mo. om ms o. mo. .mww. om moo mos u ..of :. S.- ..o. eof :.- 3. .3.- - ea. on... so... m~..- 3f once—5 spas so: o>msamuz Anus ... om.- mm.- as.- as.- so.- mm.- .~.- om. mo.- no. mo.- oo.- .o. gonads usages ..Ho ... ==.- oH.- am.- m~.- om. oo. m~.- . mo.- oo.- m~.- oo.- sounds onus-anus .mav ... m_.- oH.- oo.- mm.- oo.- mm.- mo.- mo.- oo.- as. oo.- scones-gossas Am“. ... mo.- mo.- mm. so. mo. no. mo. oo. no.- mo.- caduceus-possum as". ... so.- om. No. oa. oo. .o. mm. .o.- mo.- eaoaeeas Aodv ... oo.- so. so. so.- as. mo. om. ma. possum summaoqauu “a. . .. cm. 2..- S... no; no. 2..- 8.- Hum 8v ... mm.-H . mo. .o. co. mo. cousuoseu Am. ... mo. mo. .o. .o. no. one Ac. . ... so.- mo. mo.- . AH . cannon. dengue Am. ... mo. . o“. eocso .< co and.» ..V ... so.-. so. ago cacao: < Am. ... mo. condense . o :H guano: .¢ .m. ... eoaasodo . cease .< .mv logo A=Ho .mlo Amso “Mao Rosy Aoo Aoo A». Amy Am. 5.. and .m. Ago .mEoocH Ewa— umz U>dadMOZIIDOADGdLa> -< hOL Due—Odouhhooo :OdadHDLQ—OO 0:6 ..flOdadd>00 260288” .0200: .N< 0Hnub 1£94 .Ho. can» doom - . zmmms m:. o:. mm. mm. om. mm. mm m mm hm. Ann“. pom oaam moo .o.m oaomH mm. mm. mm. mo. mm. so., om mm on oo. Amway om com mm" x ... mm.- Hm. oH.- mo. as. so.- om. om. oo.- mo.- do.w mo. mo.- ao.- . osooen ascended: sacs-co: Ammo ... ~=.- mm.- ms.- mm.- so.- mm.- Hm.- mm. no.- mo. so. mo.- oo. sounds.»m«sas ..mo ... mm.- os.- mm.- oH.- hm. oo. o~.- no.- mo.- om. oo.- oo.- possum came-sacs Ammo ... oH.- ms.- mo.- o~.- mo.- .~.- no.- mo.- mo. as. mo.- oedema-posses Ammo ... o~.- mo.- mo. mo.- oH. so. so. no. mo.- mo.- odomecos-auaans Ammo ... mo.- om. ms. oo. 5H. . mm. no.- oo.- saddened .omo ... mo.- no. mo. mo.- oo. mo. mm. on. posses suomusmdoo Ao. ... as. oo.- mu. mo.- oo. mo.- oo.- .Hum Ao. ... am.- so. mo. no. so. so. zomoooseu Am. ... no. no. me. s cc. om< any ... Ho.- no. mo.- mo.- .H . cannons sconce Ans. ... no. mo. as. cease .< do osma> “so ... . mo. ago eased: .< Any ... co. causaoao . cm ease»: .¢ Am. ... eouasoao a cacao .¢ Amy lm_o Also lmso lmlo Ass. loss lav lo. lea lo. .m. ..o .m. .mo ls. .msoo:H cmozmnsoz sham-:oz-Iuounsasa> ~H< not oucououuuooo couusuousoo use .oco«»a«>oa cascade» .ucso: .m< amass Table A4. (Unstandardized Regression Coefficients) l95 Comparison of Models for Positive Net Farm Income. (l) (2) (3) (4) A11 Land-Size Status Class ~ Attainment Structure A. Owned & 11.73 3.08 Operated A. Rented In -.17 .19 & Operated A. Rented 3.18 1.88 Out Value of .0007 .001. A. Owned Gender -690.94 -2239.12 -757.52 -1225.74 (Female=1) Age -44.62 O-l4.27 Education 238.07 403.01 SEI -58.05 -149.18 Capitalist 4867.93 8559.95 Farmer Land- -4576.56 -5741.95 lord Farmer- -4110.33 -4936.76 Landlord Farmer- 4443.49 4392.28 Tenant Part-Time -7051.9l -806l.76 Farmer ' Constant 10424.52 6135.98 6766.75 10347.36 Term R2 ,183 .037 .065 .159 196 .079 . Table A5. Comparison of Models For Negative Net Farm Income. (Unstandardized Regression Coefficients) (l) (2) (3) (4) All Land—Size Status Class Attainment Structure A. Owned & -.29 -.44 Operated A. Rented In -.20 -.24 & Operated A. Rented -.83 —1.23 Out Value of -.0006 -.0007 .A. Owned Gender 1669.73 1619.52 1832.83 1687.30 (Female=1) Age 11.72 -8.65 Education -127.94 -224.27 SEI 20.42 39.44 Capitalist 1757.92 -4995.72 Farmer Land- -913.18 -575.63 lord Farmer- 1083.26 1395.99 Landlord Farmer- 234.41' -124.15 Tenant Part—Time 1115.56 1509.63 Farmer . Constant -4859.12 -4755.65' .2831.48 5590.3u Term ‘ . R2 .110 .033 .060 Table A6. Comparison of Models for Non-Farm Household I ncome . 197 (Unstandardized Regression Coefficients) .(l) (2) (3) (4) A11 Land-Size Status Class Attainment Structure A. Owned & .30 -.95 Operated A. Rented In —.17 -.35 & Operated A. Rented 1.58 5.43 Out Value of .0001 -.0002 A. Owned Gender -8164.77 -4227.46 -747l.33 -7133.33 (Female=1) Age 55.26 12.36 Education 1025.43 859.27 SEI 201.43 312.56 Capitalist 1986.43 3496.98 Farmer Land- 9529.48 16939.25 lord Farmer- 6693.98 10679.29 Landlord Farmer-I 2233.24 3108.93 Tenant Part-Time 10359.80 ' 15343.02 Farmer_ Constant -13093.00 -6110.69 4290.64 Term R2 .358 .016 .289 .223 198 .Ho. deco mama fl * Aooov Noamm m mm mm. moom mom seams osoe .o.m Aoooo memom mm mm mo. ooo ma ommm ooom . M ... NO. HO. :O.l mo. 30. ©H.I MH. . .- QEOOEH Sham pmz m>HpHmom Amv ... oa.- oo. so. so. no. om. doaaeeoom Amo ... mo. MO. m0. NO.I m0. . . mw< Amy ... no. Ho. 50.: mo. AH u mHMEmmV panama Amv ... mm. so. om. omczo .< no man>.on ... * mm. poo deadam .a Amv ... wo.- woumamao s 2H oeeaem .a Amo ... oaoenaao a cacao.aa AHV Aoo Aav Aoo Amo Ass Amy Amv Amy .mEOOCH 59mm 962 m>HpHmomIImameamm umfiawpfiamo pom mucmHOHmmmoo coapmHmaaoo cam .mCOHumH>mQ camocmpm .mcmmz .aa manna 199 .manmoaaoam no: u -- Aooov Haas m as ma. omm mom - - .o.m Aooov :Hmm mm mm am. mm 0mm - - w mEOOCH ... ao. mo.- mo.- mm. mm. - . - same oez e>auanom Aoo ... oz.- mo.- oo. oo. - - casemesom Asa o... NH. NO. MO. II II . ®w¢ ADV ... mo.- Ho.- II I- AH u mHmEmmv stucco Amy ... . ON. II In omcso .< Mo msam> may ... - - poo condom .< Amy ... II Umpdhmao w CH Omucmm .< Amv ... oeoeaeao a noose .< Aao 2: E so . E 2: 5 as E .mEoocH Spam umz m>HpHmom IlmoaoHUcmq pom mpcmaoammmoo coaumampaoo can .mcoaumfi>ma vamncmpm .msmmz .m< manna 200 . .Ho. :m:n mmma u * .mHDmOHHQQm no: u I: noooo nomo m en as. me: nom - man .o.m Aoooo meme an no Hm. nan an - we . w ... NO. moeI NO.I 3N. HN. II mm. GEOOCH Sank noz m>anamom Amv .... oz. on.- mo. no. - on. connneoom new ... MH. mool *. II NO.I . Owdw AWV ... mo.- mo.- 1| HH.- AH n mHmEmmv,:mo:ow Amm ... ma. II mm. coczo .< Mo mSHm> sz ... - em. poo omncam .a Amv ... II Umnmnmao . . a an owned: .< Amo ... oeneaeao a cacao .< Ana on Aav on Amo Ago Amo Amo “no .mEoo:H Esau noz m>HnHmom Ilmonoau:mq-:mesmm pom mn:mflofimmmoo :oanmamnnoo c:m .m:0Han>mQ Unmo:mnm .w:mmz ..m: manna 201 .mnnmonnaaw no: u.- noooo memon m mn mm. mmn - mom Ann .o.m noooo . nomnn mn on mo. oo - nom oo w ... no. mn.- mn.- on. - on. oo. . . esoecn sans nez e>nnnnoa non ... . nm.- mn.- mo. - mo. no. donnnasom nnv ... NH. #0. II mfiel NO. ®w< Amy ... mo.- - mn.- mo.- An u annsmmv neodmo omo ... - no. mm. . om:3o .< no msnm> Ago ... - - nso owned: .< nmo ... mo. Umnmnmao a an owned: .< nmo ... oenmaaao n cacao .< nnv non Any noo Amy new Amy Amy nno .oEoo:H Snob nmz m>nnnmom I-mn:m:mB-nmsnmm :om mn:mnonmnooo :Onanopnoo c:m .m:onnmn>ma Unmo:mnm .m:mmz .on: wanna 202 .no. :m:n mmmn n a .mnannnaqm no: u -- nooov. mom: m MH mm. omH II om HHH .Q.m noooo mzam NH om mo. m: II N om M ... no. mo. mo.- on. - om. nm. esoecn . spam nmz m>nnnmom Amv ... N:.l NO. NO. ll HO.I HO. . COHHMOSUM ANV ... no. No. II * mo. mw< Amy ... a II No.1 mo. AH u mHmEmmv Loc:mo Amv ... II on. am. Um:3o .< no manm> sz ... - - nso own:mm .< Amv ... nm. omnmnmao a an oencem .a nmv ... oennaeao a oenzo .< nno 2: E a: E 2: E as E .mEoo:H Enmm nmz o>HnnmomII m:mE:mm mane-nnmm no: mn:mnonmmmoo :onnmnmnnoo v:m .m:onnmn>ma cnmc:mnm .m:mm2 .HH< manme 203 .Ho. :mnn mmmn u x .manmonnoam no: H II noooo mnnnn m mn am. oomn - no oom . .o.m noooo omoon nn am mo. own II mm omm M ... on. oH.I no.I :o. II ma. om. mEOo:H sans nez a>nnnaom noo ... om.I mo. mo. II an. ma. :onnmosom on ... no. * II MH.I mo.l mw< Amv ... no.I II wo.I oo.I AH u mHmEmmo pmo:mo Amv ... II :0. mo. nmczo .¢ ho o3am> Aav ... - - ego oancam .< nmo ... oz. omnmnoao a :H omn:mm .¢ Amv ... omnmnooo a 06:30 .¢ Adv nwo ANV Aoo Amo Ago Amy nmo Any .mEoo:H Spam nmz m>nnnmom IImnmEnmm mnnEmm mom mn:mnonmnmoo :onnmnmnnoo o:m .m:0nn@n>mo onmo:mnm .m:mmz .mn: mnnt 204 .no. can» mean nooov omsm m nn om. omo onm onoon moom .o.m noooo mmmonI mn mm :o. :mm o: mono mmmm M ... on.- mo. no. mn.- mo.- om.- mo.- esoecn Enmm nmz m>Hnmwoz Amy ... :m. * mo. HO. m:. :0. COHQMOSUM ANV ... HO. mool mo. 00.I NO. mwfi va ... no. mo. mo.I oo. An u mnmsmmo :oo:mo Amv ... mo. mH.I mm. Um:zo .< ho 03Hm> “no ... so.- no._ noo condom .< Amo. ... mn.I omnmnmao a :H omn:mm .< Amv ... owneneao a cause .< nno Amv on Aoo nmo non nmo nmo Any nmnnMnnowo pom mn:ononmmmoo :onnmnmnpoo o:m .m:onnmn>oo onmo:mnm .m:mmz .mEoo:H Emma nmz m>nnmwm2IImmepmm .ma< oHDmB 205 .onnmonnqqm no: . noooo momm m mn ma. mzmm a»: II II .o.m noooo :oon mn no mm. mun :an II II M ... NO.| NO. mH. NO.I NO.I II II QEOOCH . Spam nmz m>nnnwoz nwo .... mmol Noel NO. MO. II II COfiPmOZUm ANV ... om. no. mo.I II II . mw< nov ... mo.I no. II II An n mnmEmmV nmo:oo nmv ... mn. II II oo:so .< no osnm> nio ... - - noo oeacam .a nmo ... II omnmnmao a an oeonam .a nmv ... empenaao a oeczo .n nno nmo Any noo nmo Ann nmo nmo nno .mEoo:n Spam nmz m>nnmwmz IInULOno:mn pom mn:mnonmnmoo :onnmnmsnoo o:m .m:0nnmn>mo unmo:mnm .m:mw: .:n¢ mnnme 206 .mnnMonnoqm no: H II noooo mnon m on mm. :mm onm - omm .o.m noooo mowm- mn 2m mn. omn om -- :mn m ... wHol 00. 3H.. 03.] MH.I II mmol @EOOCH Epmm nmz m>nnmwoz nwv ... mm.I :o.I mn. mo. II mo. :onnMosom on ... MH. NO. NO. II W0. ®w¢ AQV ... mo.- no.- - oo.- An I onosomo aoonoo nmo ... mm. - em. cacao .< no oono> neo ... II mm. nso omn:om .¢ nmv ... II omnmnooo s on ooonom .a nmo ... ooooaoao.n oonzo .n nno Aoo nno noo nmo noo nmo nmo nno .mEoo:n Spam nmz o>nnmwm2Ilmononc:mn Inmenmm pom mn:mnonmomoo :onnmnmnnoo o:m .m:onnmn>mo ohmo:mnm .m:mmz .mn< mnnme 207 .no. :mcn nmon u a .mnnmonnoam no: u II noooo mmmm m on mn. om II owno mm .o.m noooo nwmmI mn :: mo. w: II mmmn no N ... H0. 00. HH. 5m.l ll mm.l mN.l mEOOCH Spam nmz.m>nnmwmz va ... * NO. mo. ll mo. HO. . COHmeSUm Abv ... :0. HO. ll MN. OH. mw< Amv ... mo.I II :o.I mo.I An u mnmEmmv nmo:mo nmv ... - nn. oo. ooazo .e no osno> nno ... - - noo condom .a Amy ... mo.I omnmpooo n on condom .< nmo ... ooonnooo a oonso .n nno noo Awo noo nmo. Ago Amy . nmo nnv .mEoo:H Epmm nmz m>nnmmmZIImn:m:mB Ipme:mm pom mn:mnonmmmoo :Onnmnmnnoo c:m .m:onnmn>mo onmo:mnm .m:mmz .on« snows .no. :m:n mmmn u * .mnnmonnoom no: u II 208 noooo moo: m nn om. nom - om non .o.m noooo oonm- mn on on. om - . n no N ... :n.- mo. mn. no.- - oo.- on.- oaoocn Enmm noz o>nnmwoz nmv ... .om.- mm.- no. - no. no. cononooom nno ... NH. 8 ll No.1 . OH. . ®w< va ... no.- - mo.- oo.- An u onnsomo aoocoo nmo ... - an. mn. ooczo .< no oonn> nno ... I- - nso condom .< nmo ... nm. oonnnodo o 2n oondom .a Amv ... oonoaooo n oodzo .< Ann Amo Aoo nov nmo nno Amy nmo nno .mEoo:n Sham nmz m>nnmmm2IImnoEme mEnBInnmm pom mn:mnonmmmoo :Onnmnmnnoo o:m .m:onnmn>mo chmo:mnm .m:mmz .~n< mnnme .no. :m:n mmmn u * .mnnmonnoom no: u II 209 noooo onnm m on nm. nom .- oon oom . .o.m noooo oozm- mn om mo.. mmn - Fm mum M ... MH.I MN. m0. mmol II MHel Hmol @EOOCH Spam noz o>nnmwoz va ... om.- mo. on. - mo.- mo. connoooom Any ... NO. wfioI II mfiol NH.I ®w< AWV ... oo.- - mo.- oo. An u.onosomo noocoo nmv ... II on. mo. omszo .< no osnm> Azo ... II - .nso noncom .onmo ... om. omnmnooo a :H omn:mm .< Amv ... oonmnmoo a 60:30 .< nnv nmv Aoo Aoo nmv nzo Amy nmo Ano .mEoo:n spam nmz m>nnmmeIImnmEme annEmm pom mn:ononmnmoo :Onnmnmnnoo o:m .m:onnmn>ma o:mo:mnm .m:moz. .mn< onnme 210 .no. coco anon noooo onzon mn m on on. own ooo mmnmn non: .o.m noooo oomn on on mm no. :mo so nnmn omnm n ... m3. HN. GO. mcel mo. 00. OH.I mH. GEOOCH Unonmmzom Ehmmlcoz Amy ... om. mo. on.- a no. oo.- so. now noo ... nm.- mo. mo. mo. nm. no. connnosom nno ... mo. mo. mo. no.- oo. two Aoo ... mo. * no.I mo. An n onMEmmv Loo:oo Amv. ... mn. no.- on. oodzo .a no oono> nno ... no.- on. one condom .< nmo ... nn.l Umnmhoao n on condom .< nmo ... oonoaoao a oodzo .n nno noo Aoo . nno noo nmo non nmo nmo nno pom mn:ononmnmoo :onnmnm:noo o:m .m:onnmn>mo o:mo:mnm .m:mmz .mEoo:n ono:mmsom Spam-:OZIIanEme nmnnManmo .mn¢ onQMB 211 .mnnmonnoom no: .no. coco moon I * mmoon om. m nn om. nmmmo on: o o . .o.m swoon mo. mn mm on. Amwmo mmn o o w ... mm. on. mo.I mm.I no. mo. II II msoo:n. cno:mmsom EnmmI:oz on ... mo. oo.- mo. no.- a - - now noo ... om.- oo.- oo. oo. - - conooooom.nno ... on. mo. oo. - - own noo ... mo.- mo.- - - An I onosooo noocoo Amy ... mn. II II Uo:3o .< no osnmb sz ... - - nso noncom.J< nmo II II omnmnoao n on ooncom .< Amo - oooaaoao a ooczo .< nno Aoo Aoo nno nov nmo nno nmo Amo nnv pom mn:ononnnmoo :onnmnonnoo o:m .n:onnmn>mo onmo:mnm .m:mmz .mEoo:n ononmmsom EnmmI:OZIImoLono:mn .om< onnt .mnnmonnoam no: u II 212 noooo omnon mm m mn mm. mom nmm o non . .o.m mmozn nm mn mm mn. Ammwv mm o. onn M ... om. om. mm.- on.- no. mo. - mo. esoocn ono:mmso: EnmmI:oz on ... om. oa.- mo. mo.- mo.- -- on.- . . nan Aoo ... nm.I mo.I on. mo. II mo. :onnmozom any ... so. no. no. -- oo. one loo ... :o.I mo.I II mo.I An I onmEmmv noocoo Amv ... on. II mm. om:zo .< no mznm> sz ... - nm. noo oooaom .a nmo II II . omnmnmoo w :H concom .< Amy ... ooooaooo a oonzo .< nno nov va Ano noo Amy non Amy Amy nnv .meoo:n Uno:mmsom Sham-:OZIImGLOno:mn ILoEme :om mn:mnonmomoo :onnmnmnnoo o:m .m:0nnmn>mo onwo:mnm .m:mmz .nm¢ onnme .mnnmonnqom no: 213 I IN mEoo:n ono:mm:om EnmmI:oz nmm :onnmdsom mw< An n onmEmmo nmo:mo om:30 .< no msnm> . .nso oon:om .< omnmnooo w :n omn:om .< omnmhmao a om:3o .< 5 :3. E so so 2: 5 Amy nno .mEoo:n ono:mm:om EnmmI:02IImn:m:mB Insanmm pom mn:mnonnnmoo :onnmnmnnoo o:m .m:onnmn>ma onwo:mnm .m:mm2 .mm< onnme 214 .no. can» mean u * .mnnmonnoam no: I II III III noooo ommmn mm m . on mm. mnm I- mm omn I .o.m noooo nonon om mn n: no. so - z oo M ... om. .mm. no.- nn.- no. - mo. no. osoocn ononmmzom EpmmI:oz on ... om. mo.I mn. mo. II mo.I no.I Hmm nwv ... mm.I mo. no. II * mo. . :Onnmosom Any ... mOoI NO. II HO. @0- mw< va ... a II mo.I mo.I An n onwEmmo :mo:mo nmv ... II NH. mH. UWCKO .¢ MO ®3Hm> sz ... II II n50 omn:mm .< Amv ... mm. ooooaoao a an condom .a nmo ... omnmnmoo a om:3o .< Any on Amy nno nov Amo non Amy Amo nno . .msoo:n ono:mmsom EammI:OZIIwLoEpmm. mEnBInpmm pom mn:mnonnmooo :onnmnmnnoo o:m .m:onnmn>oo ohmG:Mnm .m:mmz .mm< mnnme .no. :m:n mmon n a .mnnmonnaqm no: u II 215 . noooo nmmw o . m mn mm. nmzn II ow owm .Q.m noooo . nwnm 2n nn mm mo. mun II mm nmm M ... II mn. mo. mo. no. II * mo. mEoo:n onozmmsom EnmmI:oz nov - ... - - I- - - - - nmm noo ... mm.- mo. no. - oo. . mn. . :onnoosom nnv ... mo. * II on.- mo.- ow< noo ... no.I II mo.I mo.I An u mnmEomv Lm©:mo Amv ... - so. on. oonzo .< no oono> nno ... II II nso own:om .< an ... oz. nonmnmoo a :H concom .¢ Amv ... oonmnmoo w 66:30 .< Anv on nwo nno non Amo nov nmv nmv nnv .mEoo:n ono:om:om EnmmI:oz IImLmEpmm annEmm mom mn:ononmnmoo :onnmnmpnoo o:m .:onnmn>oo onmo:mnm .m:mmz .:m< mnnme 216 ..no. anon moon I * noooo nnoon m nn oonm nnm nznm onnm dose: noooo ononm m on nonn moo mmnmn zoo: do: .o.m noooo oooon mn oo noow mnn .onmn ooom nose: nooo mmoon mn mm ono nn moom nomm do: n ... no. mo. mn. no. nn.- mn. oeoocn stem noz o>nonnoo nnv oo. ... on.- no. no. no. on. connoozom Aoo om.- oo.- ... mo. mo. no.- oo. own nmo * no. mn. ... nn. oo. mm. ooczo .< no oono> nzo mo. oo.- oo. mo.- ... * om. noo ooncom .n nmo on. oo. oo.- mo.- mo. ... oo.- oooonooo s an condom .a Amy on. om. on. no. oo.- mm. ... ooooaooo n ooczo .< nno nno non nmv Ann nmo nmo nno Anm:owmno m:n zonmn :me o:m m>onm :mEozv .mEoo:H Spam nmz m>nnnmomIInmo:mo hp mmeme nmnnmnnomo pom mn:mnonnmmoo :Onnmnmppoo o:m .m:onnmn>mo opmo:mnm .m:mmz .mm< onnme 217 .mnpmonnqqm no: u II nooov omnm m nn omn nom - - cosoz noooo mnon m an onm mom - - no: .o.m noooo nmmm nn mo mo mmn II II :mEoz noooo . nnon mn no non omn - - . do: n ... m0. ScoI @H. WW. II II QEOOCH anon ooz o>nnnmon nnv on. ... on.- no. no. - - eonoaooom noo. no. mm.- ... mo. mo. - - own Aoo no. nn. mo. ... on. - - ooczo .< no mono> nno om. on. mo. om. ... - - ooo oooeom .< Amo II II II II II ... II Umnmnmoo a on ooonom .< nmo II II II II II II ... omnmnmoo w 66:30 .< Anv nnv Aoo Amy nzv Amv Amo Any nnm:owmno m:n zonmn :mE o:m m>onm :mEo3v .mEoo:n Enmm nmz m>nnnmomIInmo:mo an monono:mn pom mn:mnonnnmoo :onnmnonnoo o:m .m:0nndn>mo onwo:mnm .m:mmz .omo mnnwe 218 .no. anon moon I a .onnmonnoom no: I II II III nooov nnno m on mmn nmn - nnn . canon noooo momo n nn non onm - non so: .o.m noooo oomm on oo no no - on . cosoz noooo noon nn oo omn on - oo do: n ... oo. no.- mm. om. - mn. osoocn . . anon noz o>nonaon nno oo. ... oo.- oo. no. - oo. conooooom non mo.- nn.- ... mo.- * - mo.- omn Amy on. mn. mo.- ... on. - om. oodzo .a no mono> nno mm. om. no. on. ... - nm. poo noncom .n nmo II II II II II ... II oonmnmoo n on ooonom .< nmo oo. nn. oo.- nm. om. - ... oooonoao o ooczo -n nno Ano Aoo noo nno nmo nmo nno nnm:owmno o:n zonmn :os o:m o>onm :oEozv . .mEoo:n Enmm nmz m>nnnmomIInmo:oo an monono:mnlnmsnmm nom mn:mnonnnmoo :onnmnmnnoo o:m .m:onnmn>ma onmo:mnm .m:moz .nm< mnnme 219 .no. :m:n moon .mnnmonnoom no: noooo mono o mn onn - oom oo cosoz nooov omoon m on omn - ooo onn . no: .o.m nooov oonn on oo nn - oon no nosoz noooo nnnnn mn on oo - omo oo do: n ... mo. nn.- on. - on. oo. osoocn anon noz o>nonaon nno om. ... on.- no. - mo. mo.- connoooom noo om.- nn.- ... oo. - nn.- oo. own noo oo. mn. on.- ... - * oo. oodzo .< no oone> nno - - - - ... - - one ooocom .< Amo no. mm. nn.I Mn. II ... no. omnmnmoo n on ooonom .n nmo no. om. mm.- oo. - on. ... oooonoao n ooczo .n nno nno noo noo nno Amy nmo nno Anm:owmno m:n zonon :me o:m o>onm :mEoBV mn:w:melnmsnmm nom mn:mnonnnooo :OnnMnmnnoo o:m .m:onnmn>mo onwo:mnm .m:mmz .mEoo:n Enmm nmz m>nnnmomIInmo:oo an .mm¢ onnme 220 .no..:o:n moon u * .mnnmonnoom no: u II noooo nonn m on oon - nn nnn dose: . noooo mnnn m on non - om onn . no: .o.m . noooo onnn mn mo on - n oo nose: noooo oonm mn oo on - m oo co: m ... no. mo. on. - om. no. osoodn ._ anon ooz o>nonmon nno * ... mn.- no. . - no. no. cononooom non e . no.- ... mo. - n no. one Aoo oo. on. mo. ... - on. mm. ooczo .n no oono> Ano - - - - ... . - - ooo condom .n Amo om. no. no.- nn. - . ... mm. ooonnoao . n on ooocom .n nmo mo. e on. on. - on. ... oooonooo n ooczo .n nno I II III II II I nnv noo noo nno non nmo nno Anm:ommno m:n zonmn :os o:o m>oom :oEozv .oEoo:H Enmm noz m>nnnmomIInoo:mo on mnoenmm mEnBInnmm nom mn:ononnnooo :onnmnonnoo o:o .m:0nnon>oo Unmo:mnm .m:omz .om< onnme 221 .no. anon anon u.r .onnmonnoom no: I II noooo onoon m on mnn. - on nnm nose: noooo ‘ . ooomn m on .nnon - oo nom no: .o.m noooo . onoon mn oo oo - nn oon noEo: noooo oooon nn no mon - om nom do: n ... on. nn.- no. - on. nm. onoocn anon ooz o>nonnon nno mo. ... no.- no. - mn. nn. connoooom noo om.- nn.- n.. n - mn.- oo.- own noo nm. oo. oo. ... - no. oo. oonzo .n no oono> nno - - - - ... - .- . poo ooonom .< nmv om. mo. nm.- on. - ... nn. eooonodo n on ooonom .< nmo no. oo. on. on. - om. ... oononoao n ooczo .< Ano nno noo noo nno nmo nmo nno nnm:ommno onn zonmn :me o:o o>ono :mEozv nnnEom_nom mn:ononnnooo :onnononnoo o:m .m:onnmn>oo onoo:mnm .n:ooz .om< .oEoo:H Enmm nmz m>nnnmomIInoo:oo an mnmEnmm onnt 222 nno. coco moon I e noooo oommn nn m mn Aoomw ooo nomm nmmn cosoz ooo mmoon mn m on non oon nooon noon do: .o.m nooon nnoo on on no moow oo onmn nonm nooon nooo omon on on mo nno no nonn oonm co: m ... on. nm. on. no. oo. on.- nn. . osoocn . nno:omsom EnomI:oz Amv no. ... om. mo. e no. oo.- oo. nnm nno no. oo. ... nm.- no. no. nm. no. conomoooo noo nn.- no. om.- ... a mo. no.- oo. omn noo mo. * no. nn. ... on. no. nm. oonzo .n no oonn> nno mo. om.- nn.- nn. mo.- ... no.- om. ooo noncom..< nmo no. oo.- om. on.- no.- no.- ... mn.- . A onoonooo n on ooncom .< nmo on. e on. oo. e mo. on. ... onoonooo n ooczo .n nno noo nno Aoo Aoo nno nmo nmo nno Ano:ommno m:n zonon :mE o:m o>onm :oEoBV nonnonnomo nom mn:ononnnmoo :onnonmnnoo o:m .m:onnMn>oQ onoo:mnm .m:oos .mEoo:H Uno:mmzom EnmmI:02IInmo:mo an mnmanom .nm¢ onnme 223 .no. :onn nnon u * .onnoonnooo no: I II nooon ommw mm m on nnn omm II II :oEoz nooon onnmn mm m nn now on: II II . :oz .o.m . noooo . memo m: mn om mn mnn II II . :mEoz nooon mnmmm m: mn mm onn nmn II II :o: M ... NJ. N:- moel MO. mo. II II GEOOCH . ono:mm:om EnmmI:oz nmv on. ... no.. mn.- . . -- -- nan nno om. mm. ... mm.I mo. mo. II II :onnmosom nov no.- mo. no. ... oo. oo. - - own nov mn. no.I mm. mo. ... mn. II II oo:3o .< no mono> nno mn. mo.I mn. no. mm. ... II II . .nso oon:om .< Amv II II II II II II ... II oonmnooo - a :n omn:om .< nmv II II II II II II II ... oonmnmoo a om:3o .< nno noo nno non noo nno nmv nmo nno nnm:ommno mnn 30nmn :oE o:m o>onm :mEo3v .mEOO:H ononomsom EnmmI:02IInoo:mo on mon0no:mn nom mn:ononnnmoo :onnononnoo o:m .m:0nnmn>mo onmo:mnm .m:moz .mm< onnoe 224 .no. :m:n nnon u a .onnmonnoom no: I II noooo momnn om m on oon non - onn nooon nooon . . oomnn mm m on nno omm - oon . do: .o.m noooo mono mm nn no no mo -- oo :osoz nooon onnnn no mn oo nnn oo - onn co: m ... mm. m:. Hm.l 00. MO. II NO. ®EOOCH ono:omsom Enmml:oz va mn. ... oo. om.- oo.- oo.- - nn.- nmm nno oo. on. ... oo.- oo. oo. - oo. donooooom non nn.- on.- no.- ... mo. oo. - no. one noo mo. no.- on. * ... on. - mm. ooezo .n no oonn> nno .mo.1 mo.- on. oo. on. ... - om. . ooo condom .n noo II II II , II II II ... II omnmnooo n on oonnom .n nmo mo. mo.- no. oo. om. oo.. - ... oononooo n oodzo .n nno Aoo nno nov Aoo nno Amo Amo nno nnm:owmno m:n zonmn :oE ocm m>onm :mEozv I I I II .oEoo:n ono:mmsom EnomI:02IInoo:mo on monono:onInoEnom nom oncononnnooo :Onnononnoo o:m .m:onnon>oo onoo:onm .m:oo2 .mm¢ onnoe 225 .onnmonnqoo no: u nooon oomo o m mn nnn - oom oo nose; nooon . mnoo mn m. mn onn - onnn onn no: .o.o noooo oooo on nn no oo - nom oo nooon noooo oomn on mn on on - onon no do: n ... oo. on. oo. oo.- - on.- nn.- osoocn ono:mm:om EnmmI:oz Amo no. ... nm. oo. oo.- - oo.- mn.- nno nno nm. on. ... no.- no. - oo. no.- connoosoo noo no.- oo.- oo.- ... oo. - on. on. ooo noo on.- on. nm. on. ... - oo. oo. ooczo .< no oono> nno - - - - - ... - - ooo condom .< noo mn. no. nn. no.- no. - ... oo.- onoonooo n on ooncoo .o nmo on.- mo. on. mo. on. - oo. ... onoonooo n oonzo .o nno noo nno noo noo nno non nmo nno Ano:owono o:n zonon :oE o:o o>oom :oEozv .oEoo:n ononomzom EnomI:02IInoo:mo on mn:o:oelno5nom nom mn:mnonnnooo :Onnwnmnnoo o:o .m:onnMn>mQ onnn:Mnm .m:omz .:m¢ mnnt .no. :m:n nnon u.* .onnmonnoom no: u II 226 . noooo moonn om o mn no - on nnn nooon noooo . onnon mm o on ooo - . om oon no: .o.o nooon . . mmmmn on mn m: m: II N mm :mEoz noooo nooon oo mn nn no - o no co: m ... no. oo. mo.- no. - mo.- no. onoocn ono:omsom EnmmI:oz on nn. ... on. mo.- mo. - oo.- no.- . . nno nno oo. on. ... om.- no. - o oo. cononooom non oo. om.- nn.- ... mo. - no. oo. ‘ ooo non mo. on.- on. mn. ... - mn. on. ooczo .< no oono> nno - - - - - ... . - - ooo ooocoo .o noo oo.- no.- no. no.- no. - ... oo. oononooo _ n on condom .o nmo no. oo. on. on. oo. - on. ... onoonooo n ooezo .n nno noo nno non noo nno noo nmo nno nnm:owmno m:n zonon :mE o:m o>onm :mEozv .mEoocn onononsom anon-cozwInoo:oo no mnoEnom oEnBInnom nom on:ononnnooo :onnmnmnnoo o:o .mconnmn>oo onwo:onm .m:moz .mm< onooe 227 .no._:o:n mmon u * .onnoonnooo no: u II . noooo nonon o m mn onn - nn onm nose: noooo oono o o on oonn - no onm no: .o.o noooo mmon nn nn oo oon - nn oom . nooon nooon onno nn nn no nnn - oo oom co: m ooo II mHo mo. HO. II * HO. QEOUCH ono:om:om EnmmI:oz Amv - ... - - - - . - - . . nno nno on. - ... oo.- no. - oo. nn. connooooo noo no. - nnu- ... n - nn.- oo.- oon Aoo no. - om. oo.- ... - no. on. ooczo .< no osno> nno - - - - - ... - - ooo ooocoo .o noo mn. - oo. om.- no. - ... on. onoonooo n on ooocoo .4 nmo on. - om. oo. nn. - om. ... onoonoao o ooczo .n nno Aoo nno on Aoo nno nmo Amy nno nnm:ommno m:n zonmn :mE o:m o>onm :oEozv .mEoo:H ono:om:om EnmmI:OZIInmo:oo no mnoenom onnsmm non mncononnnooo :Onnononnoo o:m .m:onnmn>mo onoo:mnm nm:mmz .om< onooe LIST OF REFERENCES LIST OF REFERENCES Aldrich, Howard, and Jane Weiss 1981 "Differentiation within the United States capitalist class." AMERICAN SOCIOLOGICAL REVIEW 46 (June):279-90. Allan, G. J. Boris 1971 "Simplicity in path analysis." SOCIOLOGY 8 (May):197-212. Anderson, Charles H. 1974 THE POLITICAL ECONOMY OF SOCIAL CLASS. Engelwood Cliffs, N.J.: Prentice-Hall, 1974. Andrews, Richard N.L. LAND IN AMERICA. Lexington, Ma.: D.C. Heath, 1979. Appelbaum, Richard 1979 "Born-again functionalism?" INSURGENT SOCIOLOGIST 9 (Summer):18-33. ’ Atkinson, A.B. 1975 THE ECONOMICS OF INEQUALITY. Oxford: Clarendon. Balibar, Etienne 1977 "The basic concepts of historical materialism." Pp. 199-308 in L. Althusser and E. Balibar, READING CAPITAL. London: New Left Books. Ball, Michael 1977 "Differential rent and the role of landed property." INTERNATIONAL JOURNAL OF URBAN AND REGIONAL RESEARCH 1 (October):387-403. 1980 "On Marx's theory of agricultural rent." ECONOMY AND SOCIETY 9 (August):304-26. Barlowe, Raleigh 1958 LAND RESOURCE ECONOMICS. Englewood Cliffs, N.J.: Prentice-Hall. . 228 229 Banks, Vera, and Judith Z. Kalbacher 1981 "Farm income recipients and their families." Washington, D. C.:' Economic Research Service, U. S. Department of Agriculture. Rural Develop- ment Research Report No. 30. Baron, James N., and William T. Bielby 1980 "Bringing the firms back in." AMERICAN SOCIOLOGICAL REVIEW 45 (October):737-65. Beck, E.M., Patrick M. Horan, and Charles M. Tolbert II 1978 "Stratification in a dual economy." AMERICAN SOCIOLOGICAL REVIEW 43 (October): 704- 20. 1980 "Social stratification in industrial society." AMERICAN SOCIOLOGICAL REVIEW 45 (August):712-18. Bell Colin, and Howard Newby 1981 "Narcissism or reflexivity in modern sociology." POLISH SOCIOLOGICAL BULLETIN 53 (1):5-19. Bell, Daniel 1973 THE COMING OF POST-INDUSTRIAL SOCIETY. New York: Basic. Bertrand, Alvin L., and Floyd L. Corty 1962 RURAL LAND TENURE IN THE UNITED STATES. Baton Rouge: Louisiana State University. Bettelheim, Charles 1975 ECONOMIC CALCULATION AND FORMS OF PROPERTY. New York: Monthly Review. Bielby, William T. 1981 "Models of status attainment." Pp. 3-26 in D.J. Treiman and R.V. Robinson (eds.), RESEARCH IN SOCIAL STRATIFICATION AND MOBILITY, Vol. 1. Greenwich, Cn.: JAI. Bielby, William T., and Arne L. Kalleberg - 1981 "The structure of occupational inequality." QUALITY AND QUANTITY 15:125-50. Blalock, Hubert N., Jr. 1972 SOCIAL STATISTICS. New York: McGraw-Hill. Blau Peter M. and Otis Dudley Duncan 1967 THE AMERICAN OCCUPATIONAL STRUCTURE. New York: Wiley. Bose, Arun 1980 MARX ON EXPLOITATION AND INEQUALITY. Delhi: Oxford University. 230 Bottomore, T.B. 1968 "Review symposium."‘ AMERICAN SOCIOLOGICAL REVIEW 33 (April):294-96. Bowles, Samuel, and Herbert Gintis 1974 "IQ in the United States class structure." Pp. 7-84 in A. Gartner C. Greer, and F. Reissman (eds.), THE new ASSAULT ON EQUALITY. New York: Harper & Row. 1976 SCHOOLING IN CAPITALIST AMERICA. New York: 'Basic. Boxley, Robert F. 1979 “Ownership and land use policy." Pp. 161-67 in STRUCTURE ISSUES OF AMERICAN AGRICULTURE. Agricultural Economic Report 438. Washington, D.C.: Economics, Statistics, and Cooperatives Service, U.S. Department of Agriculture. ----- , and Larry Walker 1979 "Impact of rising land values on agricultural structure." Pp. 88-96 in STRUCTURE ISSUES OF AMERICAN AGRICULTURE. Agricultural Economic Report 438. Washington, D.C.: Economics, Statistics, and Cooperatives Service, U.S. Department of Agriculture. Boyer, M. Christine 1981 "National land use policy." Pp. 109-25 in J.I. de Neufville (ed.), THE LAND USE POLICY DEBATE IN THE UNITED STATES. New York: Plenum. Braverman, Harry 1974 LABOR AND MONOPOLY CAPITAL. New York: Monthly Review. Brewster, David E. 1979 "The family farm." Pp. 74-79 in STRUCTURE ISSUES OF AMERICAN AGRICULTURE. Agricultural Economic Report 438. Washington, D.C.: Economics, Statistics, and Cooperatives Service, U.S. Department of Agriculture. 1980 "Changes in the family farm concept." Pp. 18-23 in FARM STRUCTURE. Committee on Agriculture, Nutrition, and Forestry, U.S. Senate. Washington, D.C.: Government Printing Office. ' Breimyer, Harold.F. ' . . 1977 FARM POLICY. Ames: Iowa State University. Burawoy, Michael - 1977 "Social structure, homogenization, and 'The process of status attainment in the United States and Great Britain.'" AMERICAN JOURNAL OF SOCIOLOGY 82 (March):1031-1042. 1981 ‘"Comment on 'Explaining the saleries of 5 . 231 managers.'" CONTEMPORARY SOCIOLOGY 1O (September):612-13. Buttel, Frederick H. 1980 "W(h)ither the family farm?" CORNELL JOURNAL OF SOCIAL RELATIONS 15 (Summer):10-37. 1982 "The political economy of agriculture in advanced industrial societies." Forthcoming in 8.0. McNall and G.N. Howe (eds.), CURRENT PERSPECTIVES IN SOCIAL THEORY. Greenwich, Cn.: JAI. ----- , and Oscar W. Larson III 1982 "Political implications of multiple jobholding in U.S. agriculture." RURAL SOCIOLOGY 47 (Summer):272-94. ----- , Oscar W. Larson III, Craig K. Harris, and Sharon Powers 1982 "Social class and agrarian politicl ideology." SOCIAL FORCES 61 (September):277-83. Campbell, Richard T. ' 1983 "Status Attainment Research." SOCIOLOGY OF EDUCATION 56 (January):47-62. Carchedi, G. 1981 Review Of Wright, 1979. SCIENCE AND SOCIETY 45 Carlin, Thomas A., and Linda M. Ghelfi 1979 "Off-farm employment and the farm structure." Pp. 270-73 in STRUCTURE ISSUES OF AMERICAN AGRICULTURE. Agricultural Economic Report 438.~ Washington, D.C.: Economics, Statistics, and Cooperatives Service, U.S. Department of Agriculture. Carr A. Barry 1980 "A profile of the commercial agricul‘tural sector." Pp. 24- 35 in FARM STRUCTURE. Committee on Agricul- ture, Nutrition, and Forestry, U.S. Senate. Washington, D.C.: Government Printing Office. Castells, Manuel 1980 THE ECONOMIC CRISIS AND AMERICAN SOCIETY. Princeton, N.J.: Princeton University. Clegg, Stewart, and David Dunkerly 1980 ORGANIZATION, CLASS AND CONTROL. London: Routledge & Kegan Paul. Clog , Clifford C. 19 . "Some methodological problems in Wright's CLASS STRUCTURE AND INCOME DETERMINATION. " Presentation at the annual meeting of the American Sociological Association, San Francisco, Ca. 232 Cochrane, Willard W. 1979 THE DEVELOPMENT OF AMERICAN AGRICULTURE. Minneapolis: University of Minnesota. Colclough, Glenna, and Patrick M. Horan 1983 "The status attainment paradigm." SOCIOLOGICAL QUARTERLY 24 (Winter):25-42. Coleman, Richard P. ., and Lee Rainwater 1978 SOCIAL STANDING IN AMERICA. New York: Basic. Congressional Budget Office 1978 PUBLIC POLICY AND THE CHANGING STRUCTURE OF AGRICULTURE. Background Paper, U.S. Congress. Washington, D.C.: Government Printing Office. Congressional Research Service 1979 FARM INCOME AND FARM STRUCTURE IN THE UNITED STATES. Report No. 79-188 S. Washington, D.C.: -Library of Congress. Connell, R.W. 1979 "A critique of the Althusserian approach to class." THEORY AND SOCIETY 8 (November):303-46. Coser, Lewis A. 1975 "Two methods in search of a substance." AMERICAN SOCIOLOGICAL REVIEW 40 (December):691-700. 1976 "Reply to my critics." THE AMERICAN SOCIOLOGIST 11 (February):33-38. Coughenour, C. Milton, and James A. Christenson 1982 "Farm structure, social class, and farmers' policy perspectives." Forthcoming in D. Brewster, W.D. Rasmussen, and G. Youngblood (eds.), FARMS IN TRANSITION. Ames: Iowa State University. Coughenour, C. Milton, and Louis Swanson 1983 "Work statuses and occupations of men and women in farm families and the structure of farms." RURAL SOCIOLOGY 48 (Spring):23-43. ‘ Crecink, John C. 1979 FAMILIES WITH FARM INCOME. Washington, D.C.: Economics, Statistics, and Cooperative Service, U.S. Department of Agriculture. Crompton, Rosemary, and Jon Gubbay 1978 ECONOMY AND CLASS STRUCTURE. New York: St. Martin's. Crowder, N. David 1974 "A critique of Duncan's stratification nesearch." 233 SOCIOLOGY 8 (January):19-45. Currie, J. M. 1981 THE ECONOMIC THEORY OF AGRICULTURAL LAND TENURE. Cambridge: Cambridge University. Cutler, Anthony, Barry Hindess, Paul Hirst, and Athar Hussain. 1977 MARX'S "CAPITAL" AND CAPITALISM TODAY. London: Routledge A Kegan Paul. Dahrendorf, Ralf 1959 CLASS AND CLASS CONFLICT IN INDUSTRIAL SOCIETY. Palo Alto, Ca.: Stanford University. Dalton, Hugh 1925 THE INEQUALITY 0F INCOMES.. London: Routledge. Davis, John Emmeus 1980 ”Capitalist agricultural development and the exploitation of the propertied laborer." Pp. 133- 53 in F.H. Buttel and H. Newby (eds.), THE RURAL SOCIOLOGY OF THE ADVANCED SOCIETIES. Montclair, N.J.: Allanheld, Osmun. Davis, Kingsley, and Wilbert E. Moore 1945 "Some principles of stratification." AMERICAN SOCIOLOGICAL REVIEW 10 (April):242-49. ' Duncan, Otis Dudley 1961 "A socioeconomic index for all occupations." Pp. 109-38 in A. J. Reiss (ed. ), OCCUPATIONS AND SOCIAL STATUS. New York: Free. 1966 "Path analysis." AMERICAN JOURNAL OF SOCIOLOGY 72 (JUlY) 51-160 1968 "Social stratification and mobility." Pp. 675-719 in E.B. Sheldon and W.E. Moore (eds.), INDICATORS OF SOCIAL CHANGE. New York: Russell Sage Foundation. 1969a "Inheritance of poverty or inheritance of race?" Pp. 85- 110 in D. P. Moynihan (ed. ), ON UNDERSTANDING POVERTY. New York: Basic. - 1969b "Contingencies in constructing causal models." Pp. 74-112 in E. F. Borgatta and G. W. Bohrnstedt (eds. ), SOCIOLOGICAL METHODOLOGY 1969. San Francisco: Jossey-Bass. ----- , David L. Featherman, and Beverly Duncan 1972 SOCIOECONOMIC BACKGROUND AND ACHIEVEMENT. New York: Seminar. Ehrenreich, Barbara and John 1979 ”Rejoinder." Pp. 313-34 in Pat Walker (ed.), BETWEEN LABOR AND CAPITAL. Boston: South End. 234 Ehrensaft, Philip 1980 "Long waves in the transformation of North American agriculture." CORNELL JOURNAL OF SOCIAL RELATIONS 15 (Summer):65-83. Ely, Richard T., and George S. Wehrwein 1940 LAND ECONOMICS. New York: Macmillan. Eyerman, Ron 1979 Review of Wright, 1978a. THEORY AND SOCIETY 8 (November): 431- 35. ' Featherman, David L. ' 1971 "Residential background and socioeconomic achievement in metropolitan stratification systems." RURAL SOCIOLOGY 36 (June):107-24. 1976 "Coser's ... 'In search of substance.'" THE AMERICAN SOCIOLOGIST 11 (February):21-27. 1981 "Stratification and social mobility." Pp. 79-100 in J.F. Short (ed.), THE STATE OF SOCIOLOGY. Beverly Hills: Sage. ----- , and Robert M. Hauser A 1973 "On the measurement of occupation in social surveys." SOCIOLOGICAL METHODS A RESEARCH 2 (November): 239-51. 1976 "Prestige or socioeconomic scales in the study of occupational achievement?" SOCIOLOGICAL METHODS A RESEARCH 4 (MayL 403-22. 1978 OPPORTUNITY AND CHANGE. New York: Academic. ----- , F. Lancaster Jones, and Robert M. Hauser 1975 "Assumptions of social mobility research in the U.S." SOCIAL SCIENCE RESEARCH 4:329-60. Fine, Ben ‘ 1979 ”On Marx's theory of agricultural rent." ECONOMY AND SOCIETY 8 (August):241-78. 1980 "On Marx's theory of agricultural rent." ECONOMY AND SOCIETY 9 (August):327-31. ' 1982 THEORIES OF THE CAPITALIST ECONOMY. London: Arnold. Finke, Marvin, and Kim Schopmeyer 1980 "The dynamics of class formation and the problem of the middle class." Presentation at the annual con- ference on the current state of Marxist theory, Louisville, Ky. Flinn, William L., and Frederick H. Buttel 1980 "Sociological aspects of farm size." AMERICAN JOURNAL OF AGRICULTURAL ECONOMICS 62 (December): 946- 53. Friedland, William H. 235 1981 "Prospectus for a sociology of agriculture." Presentation at the annual meeting of the American Sociological Association, Toronto, Can. ----- , Amy Barton, and Robert J. Thomas 1981 MANUFACTURING GREEN GOLD. New York: Cambridge University. Friedmann, Harriet 1978 "World market, state, and family farm." COMPARATIVE STUDIES -IN SOCIETY AND HISTORY 20 (4): 545- 86. 1980 I'Household production and the national economyJ JOURNAL OF PEASANT STUDIES 7 (2):158-84. 1981 nThe family farm in advanced capitalism." Presenta- tion at the annual meeting of the American Socio- logical Association, Toronto, Can. Gaffney, M. Mason 1962 "Land and rent in welfare economics." Pp. 141-58 in J. Ackerman, M. Clawson, and M. Harris (eds.), LAND ECONOMICS RESEARCH. Washington, D.C.: Resources for the Future. Geisler, Charles 0., Nelson L. Bills, Jack R. Kloppenburg, Jr., and William F. Waters 1982 "The structure of agricultural landownership in the United States, 1946_and 1978." Forthcoming, Agricultural Experiment Station Bulletin, Cornell University. General Accounting Office 1978‘ CHANGING CHARACTER AND STRUCTURE OF AMERICAN AGRICULTURE. CED-78-178. 'Washington, D.C. 1980 AN ASSESSMENT OF PARITY AS A TOOL FOR FORMULATING AND EVALUATING AGRICULTURAL POLICY. CED-81-11.' Washington, D.C. Giddens, Anthony 1973 THE CLASS STRUCTURE OF THE ADVANCED SOCIETIES. New York: Harper A Row. - 1977 STUDIES IN SOCIAL AND POLITICAL THEORY. New York: Basic. 1979 CENTRAI. PROBLEMS IN SOCIAL THEORY. Berkeley: University of California. 1981 '"Postscript (1979)." In A. Giddens, THE CLASS STRUCTURE OF THE ADVANCED SOCIETIES, 2nd edn. London: Hutchinson. Gilbert, Jess 1982 "Rural theory."' RURAL SOCIOLOGY 47 (Winter):609-33. ----- , and Craig K. Harris ' 1979 I'The changing structure of farm land ownership in 236 the United States, 1946-1978." Presentation at the annual meeting of the Rural Sociological Society, Guelph, Can. Forthcoming in H. Schwarzweller (ed. ), RESEARCH IN RURAL SOCIOLOGY AND DEVELOPMENT, Vol. 1. Greenwich, Cn.: JAI. Gintis, Herbert 1980 "The American occupational structure eleven years later.“ CONTEMPORARY SOCIOLOGY 9 (January):12-16. Goss Kevin, Richard D. Rodefeld, and Frederick H. Buttel 1980 "The political economy of class structure in U.S. agriculture." Pp. 83-132 in F.H. Buttel and H. Newby (eds.), THE RURAL SOCIOLOGY OF THE ADVANCED SOCIETIES. Montclair, N.J,: Allanheld, Osmun. Granovetter, Mark 1981 "Toward a sociological theory of income differences." Pp. 11-47 in I. Berg (ed.), SOCIOLOGICAL PERSPECTIVES ON LABOR MARKETS. New York: Academic. Greenberg, Edward S. and Thomas F. Mayer ' 1978 Review of Wright, 1978a. KAPITALISTATE 7:167-86. Gujarati, Damodar 1978 BASIC ECONOMETRICS. New York: McGraw-Hill. Haller, Archibald 0. 1982 "Reflections on the social psychology of status attainment. " Pp. 3- in R. M. Hauser, D. Mechanic, A. O. Haller, and T. S. Hauser (eds. ), SOCIAL STRUCTURE AND BEHAVIOR. New York: Academic. ----- and Alejandro Portes 1973 "Status attainment processes." SOCIOLOGY OF EDUCATION 46 (Winter):51-91. ----- , and Kenneth I. Spanner 1977 "Occupational income differentiation in status attainment." RURAL SOCIOLOGY 42 (Winter):517-35. ----- , and Jose Pastore . 1982 "Labor market segmentation, sex, and income in Brazil." Manuscript in progress. Halliman, Maureen T., and Michael R. Olneck 1983 "Letter from the editor." SOCIOLOGY OF EDUCATION 56 (January):1-2. Harris, Marshall 1974 ENTREPRENEURSHIP IN AGRICULTURE. Agricultural Law Center Monograph No. 12. Iowa City: University of Iowa. - ° . 237 Haug, Marie R. 1977 Harvey, 1973 Hauser, 1980 "MeaSurement in social stratification." Pp. 51-77 in ANNUAL REVIEW OF SOCIOLOGY, Vol. 3. Palo Alto, Ca.:- Annual Reviews. David SOCIAL JUSTICE AND THE CITY. Baltimore: Johns Hopkins University. Robert M. "On 'Stratification in a dual economy. '" AMERICAN SOCIOLOGICAL REVIEW 45 (August): 702- 12. ----- , and David L. Featherman 1977 THE PROCESS OF STRATIFICATION. New York: Academic. Hazelrigg, Lawrence E. 1972 "Class, property, and authority." SOCIAL FORCES 50 (June): 473- 87. Hill, Richard Child 1973 "Urban income inequality." Ph.D. dissertation. University of Wisconsin. Horan, Patrick M. 1978 Hottel, I979 nIs status attainment atheoretical?" AMERICAN SOCIOLOGICAL REVIEW 43 (August):534-41. E.M. Beck, and Charles M. Tolbert II "The market homogeneity assumption." SOCIAL SCIENCE QUARTERLY 61 (September):278-92. Bruce, and David H. Harrington "Tenure and equity influences on the incomes of farmers." Pp. 97-107 in STRUCTURE ISSUES OF AMERICAN AGRICULTURE. Agricultural Economic Report 438. Washington, D.C.: Economics, Statistics, and Cooperatives Service, U.S. Department of Agriculture. Hunt, E.K. 1979 "Marx' s theory of property and alienation." Pp. 283- 315 in A. Parel and T. Flanagan (eds.), THEORIES OF PROPERTY. Waterloo, Can.: Wilfrid Laurier University. ' Hurlburt, Virgil L. 1958 "Distribution of income from farmland. " Pp. 176-82 in LAND. The Yearbook of Agriculture, 1958, U.S. Department of Agriculture. Washington, D.C.: Government Printing Office. 238 Hurst, Charles E. 1979 THE ANATOMY OF SOCIAL INEQUALITY. St. Louis: Mosby. Hussain, Athar, and Keith Tribe 1981 MARXISM AND THE AGRARIAN QUESTION, 2 Vols. Atlantic Highlands, N. J.: Humanities. Hutter, Michael 1982 "Early contributions to law and economics." JOURNAL OF ECONOMIC ISSUES 16 (March):131-47. Jencks, Christopher, and 7 others. 1972 INEQUALITY. New York: Basic. ----- and 11 others. 1979 WHO GETS AHEAD? New York: Basic. Jensen, Harold R., Thomas C. Hatch, and David H. Harrington 1981 ECONOMIC WELL-BEING 0F FARMS. Agricultural Economic Report 469. Washington, D.C.: Economics, Statistics, and Cooperatives Service, U.S. Department of Agriculture. Kalleberg, Arne L. 1982 Review of Wright, 1979. AMERICAN JOURNAL OF SOCIOLOGY 87 (JanuaryL 970-72. ----- , and Larry J, Griffin 1980 "Class, Occupation, and Inequality in Job Rewards." AMERICAN JOURNAL OF SOCIOLOGY 85 (January):731-68. ----- , and Aage B. Sorensen 1977 "The Sociology of Labor Markets." Pp. 351-79 in ANNUAL REVIEW OF SOCIOLOGY, Vol. 5. Palo Atlo, Ca.: Annual Reviews. ----- Michael Wallace, and Robert P. Althauser 1981 "Economic Segmentation, Worker Power, and Income Inequality." AMERICAN JOURNAL OF SOCIOLOGY 87. (November):651-83. Kautsky, Karl 1980 ,"Summary of selected parts of THE AGRARIAN QUESTION" (ed. Jarius Banaji). Pp. 39-82 in F. H. Buttel and H. Newby (ed8.), THE RURAL SOCIOLOGY OF THE ADVANCED SOCIETIES. Montclair, N.J.: Allanheld, Osmun. Kelly, Kevin D. 1979 "The independent mode of production." REVIEW OF RADICAL POLITICAL ECONOMICS 11 (Spring):38-48. . Kenny, David A. ’ 1979 CORRELATION AND CAUSALITY New York: Wiley. 239 Keyfitz, Nathan 1981a "Explaining the saleries of managers." CONTEMPORARY SOCIOLOGY 10 (January):85-87. 1981b "Reply to Wright and Burawoy." CONTEMPORARY SOCIOLOGY 10 (September):613-15. Koo, Hagen, and Doo-Seung Hong 1980 "Class and income inequality in Korea." AMERICAN SOCIOLOGICAL REVIEW 45 (August):610—26. Lafargue,_Paul 1901 THE EVOLUTION OF PROPERTY. London: Sonnenschein. Lancelle, Mark, and Richard D. Rodefeld 1980 "The influence of social origins on the ability to attain ownership of large farms." RURAL SOCIOLOGY 45 (Fall):381-95. Larson, Donald K. 1974 "Wage and salary income." JOURNAL OF NORTHEASTERN AGRICULTURAL ECONOMICS COUNCIL 3 (May):64-75." 1975 "Economic class as a measure of farmers' welfare." AMERICAN JOURNAL OF AGRICULTURAL ECONOMICS 57 (November):658-64. ----- , and Thomas A. Carlin 1974 "Income and economic status of people with farm earnings." SOUTHERN JOURNAL OF AGRICULTURAL ECONOMICS (December):73-79. Larson, Oscar, III, Gilbert W. Gillespie, Jr., and Frederick H. Buttel 1983 "Sources of class indentification among farmers." RURAL SOCIOLOGY 48 (Spring):82-103. Lenin, V.I. 1975 "A great beginning." Pp. 477-88 in R.C. Tucker (ed.), THE LENIN ANTHOLOGY.‘ New York: Norton. Lenski, Gerhard 1966 POWER AND PRIVILEGE. New York: McGraw-Hill. Lewis, Douglas G. 1980 "Users guide to the 1978 ESCS landownership survey." Washington, D.C.: Natural Resource Economics Division; Economics, Statistics, and Cooperatives Service; U.S. Department of Agriculture. Lewis, James A. - 1979 "Land ownership and the 1978 resource economics survey." Presentation at the Southern Association of Agricultural Scientists, New Orleans, La. 1980 240 LANDOWNERSHIP IN THE UNITED STATES, 1978. Agriculture Information Bulletin 435. Washington, D.C.: Economics, Statistics, and Cooperatives Service, U.S. Department of Agriculture. Lewis-Beck, Michael S. 1980 1980 Lord 1982 APPLIED REGRESSION. Beverly Hills, Ca.: Sage. George F. III, and William W. Falk "An exploratory analysis of individualist versus structuralist explanations of income." SOCIAL FORCES 59 (December):376-91. "Dual economy, dual labor, and dogmatic Marxism." SOCIAL FORCES 60 (March):891-97. Lyson, Thomas, and Patrick M. Horan n.d. "Notes on the social and economic stratification of farmers." Manuscript in progress. Macpherson, C.B. 1973 1977 1978 Mandel, 1968 1981 DEMOCRATIC THEORY. Oxford: Clarendon. THE LIFE AND TIMES OF LIBERAL DEMOCRACY. Oxford: Oxford University. "The meaning of property." Pp. 1-13 in C.B. Macpherson (ed. ), PROPERTY. Toronto: University of Toronto. Ernest MARXIST ECONOMIC THEORY. New York: Monthly Review. "Introduction." Pp. 9-90 in K. Marx, CAPITAL, Vol. 3. New York: Vintage. Marx, Karl 1967 1973 1976 Massey, 1977 1978 Matras, 1980 Miller, 1958 CAPITAL, Vol. 3. New York: International. GRUNDRISSE. New York: Vintage. CAPITAL, Vol. 1. New York: Vintage. Doreen "The analysis of capitalist landownership." INTER- NATIONAL JOURNAL OF URBAN AND REGIONAL RESEARCH 1 (0ctober):404-24. and Alejandrina Catalano CAPITAL AND LAND. London: Arnold. Judah "Comparative social mobility." ANNUAL REVIEW OF SOCIOLOGY, Vol. 6. Palo Alto, Ca.:' Annual Reviews. Walter G., Max M. Tharp, and Lawrence A. Jones "Tenure and the use of farm resources." Pp. 562-67 in LAND. The Yearbook of Agriculture, 1958, U.S. Department of Agriculture. Washington, D.C.: 2Ul Government Printing Office. Mills, C. Wright 1959 THE SOCIOLOGICAL IMAGINATION. New York: Oxford University. Mooney, Patrick H. 1979 "Class relations and class structure in the Midwest." Manuscript in progress. 1981 "Toward a class analysis of Midwestern agriculture." Presentation at the annual meeting of the Rural Sociological Society, San Francisco, Ca. Morrissey, Marietta 1981 "The dual economy and labor market segmentation." SOCIAL FORCES 6O (March):883-90. Moyer, D. David, Marshall Harris, and Marie B. Harmon 1969 LAND TENURE IN THE UNITED STATES. Agriculture Info- mation Bulletin 338. Washington, D.C.: Economic Research Service, U.S. Department of Agriculture. Murray, Robin 1977 I'Value and rent theory, part 1." CAPITAL & CLASS 3 (Winter):100-22. ‘ ' 1978 "Value and rent theory, part 2." CAPITAL & CLASS A (Spring):11-33. ‘ Neufville, Judith Innes de 1981 "Conclusion.“ Pp. 251-55 in J. I. de Neufville (ed.), THE LAND USE POLICY DEBATE IN THE UNITED STATES. New York: Plenum. Newby Howard 1976 "The rural sociology of advanced capitalist societies." Pp. 3- 30 in H. Newby (ed. ), INTERNATIONAL PERSPECTIVES IN RURAL SOCIOLOGY. London: Wiley. ----- Colin Bell, David Rose, and Peter Saunders 1976 PROPERTY, PATERNALISH, AND POWER. London: Hutchinson. ----- , and Frederick H. Buttel . 1980 "Toward a critical rural sociology." Pp. 1-35 in F.H. Buttel and H. Newby (ed3.), THE RURAL SOCIOLOGY OF THE ADVANCED SOCIETIES“ Montclair, N.J.: Allanheld, Osmun. Ollman, Bertell 1968 ”Marx' 3 use of 'class.'" AMERICAN JOURNAL OF SOCIOLOGY 73 (March): 573- 80. Parkin,rFrank 1971 CLASS INEQUALITY AND POLITICAL ORDER. New York: Praeger. ' 242 1979 MARXISM AND CLASS THEORY. New York: Columbia University. Parsons, Talcott 1940 "An analytic approach to the theory of social stratifiction." AMERICAN JOURNAL OF SOCIOLOGY AS (May):8fl1-62. 1970 "Equality and inequality in modern society." Pp. 13-72 in E.O. Laumann (ed.), SOCIAL STRATIFICATION. Indianapolis: Bobbs-Merrill. Pawson, Ray , 1978 "Empiricist explanatory strategies." SOCIOLOGICAL REVIEW 26 (August):613-45. Pease, John, William H. Form, and Joan Huber Rytina 1970 "Ideological currents in American stratification literature." THE AMERICAN SOCIOLOGIST 5 (May):127-37. Penn, J.B. 1979 "The structure of agriculture." Pp. 2-23 in STRUCTURE ISSUES OF AMERICAN AGRICULTURE. Agriculture Economic Report #38. Washington, D.C.: Economics, Statistics, and Cooperatives Service, U.S. Department of Agriculture. Perelman, Michael 1975 "Natural resources and agriculture under capitalism." AMERICAN JOURNAL OF AGRICULTURAL ECONOMICS 57 (November):701-OA. Pfeffer, Max John 1982 "The labor process and capitalist development of agriculture." THE RURAL SOCIOLOGIST 2 (March):72-80. Plotke, David 1980 "The United States in transition." SOCIALIST REVIEW 10 (November-December):71-123. ‘ Poulantzas, Nicos 1975 CLASSES IN CONTEMPORARY CAPITALISM. London: New Left. Pryor Frederick L. 1978 PROPERTY AND INDUSTRIAL ORGANIZATION IN COMMUNIST AND CAPITALIST NATIONS. Bloomington: Indiana University. Raup Philip M. ' 1980 "Some issues in land tenure, ownership and control in dispersed vs. concentrated agriculture." Pp. 153-59 in INCREASING UNDERSTANDING OF PUBLIC PROBLEMS AND POLICIES. Oak Brook, Il.: Farm Foundation. ' ' 243 Renne, Roland R.- - 4 19A? LAND ECONOMICS. New York: Harper & Brothers. Robinson, Robert V., and Jonathan Kelly 1979 "Class_as conceived by Marx and Dahrendorf." AMERICAN SOCIOLOGICAL REVIEW 44 (February):38-57. 1980 "Synthesis and comparison of stratification theories." AMERICAN SOCIOLOGICAL REVIEW 45 (April):328-34. Rodefeld, Richard D. . - 1976 "The assessment of farm (operating unit) and farm operator characteristics by the U.S. Census of Agriculture." Pp. 82-102 in AGRICULTURAL CENSUS. Hearings before the Subcommittee on Census and Popu- lation, Committee on Post Office and Civil Service, U.S. House of Representatives, 94th Congress, 2nd Session. Washington, D.C.: Government Printing Office. 1978 "Trends in U.S. farm organizational structure and type." Pp. 158-77 in R.D. Rodefeld et a1.-(eds.), CHANGE IN RURAL AMERICA. St. Louis: Mosby. 1979 "The family—type farm and structural differentiation." Pp. 379-AA8 in FAMILY FARM ANTITRUST ACT OF 1979. Hearings before the Subcommittee on Antitrust, Mono— poly and Business Rights, Committee on the Judiciary, U.S. Senate, 96th Congress, 1st Session. Washington: Government Printing Office. - Rose, David, Peter Saunders, Howard Newby, and Colin Bell 1976 "Ideologies of properties." SOCIOLOGICAL REVIEW 24 (November):699-730. Roweis, Shoukry T., and Allen J. Scott 1981 "The urban land question." Pp. 123-57 in M. Dear and A.J. Scott (ed8.), URBANIZATION AND URBAN PLANNING IN CAPITALIST SOCIETY. New York: Methuen. Saunders Peter 1978 ‘Domestic property and social class." INTERNATIONAL JOURNAL OF URBAN AND REGIONAL RESEARCH 2 (June):233-51. 1979' URBAN POLITICS. Eondon: Hutchinson. Schertz, Lyle P., and others. 1979 ANOTHER REVOLUTION IN U.S. FARMING? Agricultural Economic Report uu1. Washington, D.C.: Economics, Statistics, and Cooperatives Service, U.S. Department of Agriculture. Scott, Allen J. 1976 "Land and land rent." Pp. 101-H5 in PROGRESS IN GEOGRAPHY, Vol. 9. London: Arnold. 2AA Sewell, William H., and Robert M. Hauser . 1975 EDUCATION, OCCUPATION, AND EARNINGS. New York: Academic. Singer, Edward G. 1982 "Class relations and political attitudes among Ohio farmers." Ph.D. dissertation. Ohio State University. Sorensen, Aage B. 1976 "Models and strategies in research on attainment an? opportunity." SOCIAL SCIENCE INFORMATION 15 1 671-910 . 1978 "Education, the process of attainment and the structure of inequality." Pp. 221-42 in R. Girod (ed.), DEMOCRATISATION DES ETUDES ET PROGRES SOCIAL. Vevey: Delta. Reprint 327, Institute for Research on Poverty, University of Wisconsin, Madison. 1979 "A model and a metric for the analysis of intra- generational status attainment process." AMERICAN JOURNAL OF SOCIOLOGY 85 (2):361-84. ‘ Stinchcombe, Arthur L. 1961 "Agricultural enterprise and rural class relations." AMERICAN JOURNAL OF SOCIOLOGY 67 (September):165-76. Treiman, Donald J. 1976 "A comment on professor Lewis Coser's presidential address." THE AMERICAN SOCIOLOGIST 11 (February):27-33. Therborn, Goran 1976 SCIENCE, CLASS, AND SOCIETY. London: New Left. Tribe, Keith 1977 "Economic property and the theorisation of ground rent." ECONOMY AND SOCIETY 6 (February):69-88. U.S. Bureau of the Census 1978 CENSUS OF AGRICULTURE, 197A. Vol. 2, Part 2. Washington, D.C.: Government Printing Office. 1981 CENSUS OF AGRICULTURE, 1978. Vol. 1, Part 51. Washington, D.C.: Government Printing Office. U.S. Department of Agriculture. 1959 LAND OWNERSHIP IN THE GREAT PLAINS. ARS #3-93. Washington, D.C.: Agricultural Research Service, U.S. Department of Agriculture. 1981 A TIME TO CHOOSE. Washington, D.C.: Government Printing Office. Upton, Martin ‘ . , 1976 AGRICULTURAL PRODUCTION ECONOMICS AND RESOURCE-USE. 245 London: Oxford University. Vogeler, Ingolf 1981 THE MYTH OF THE FAMILY FARM. Boulder, Co.: Westview. Vogt Roy 981 "Property rights and employee decision making in West Germany." JOURNAL OF ECONOMIC ISSUES 15 (June): 377-860 Vroey, Michael de 1975 "The separation of ownership and control in large corporations." REVIEW OF RADICAL POLITICAL ECONOMICS 7 (Summer):1-10. Wang Charlotte Shiany Yun, and William H. Sewell 980 "Residence, migration, and earnings." RURAL SOCIOLOGY 45 (Summer):185-206. * Warner, W. Lloyd, Marchia Meeker, and Kenneth Eels - 1949 SOCIAL CLASS IN AMERICA. Chicago: Science Research. Weber, Max 1946 FROM MAX WEBER. Ed. H.H. Gerth and C.W. Mills. New York: Oxford University. 1947 THE THEORY OF SOCIAL AND ECONOMIC ORGANIZATION. Ed. T. Parsons. New York: Free. Wesolowski, Wlodzimierz, and Kazimierz M. Slomczynski, and Bogdan W. Mach 1979 "Trends in social mobility studies and Marxist theory of class structure." POLISH SOCIOLOGICAL REVIEW (1):5-18. Westergaard, John, and Henrietta Resler 1975 CLASS IN A CAPITALIST SOCIETY. New York: Basic. Wilson, Kenneth L. 1978 "Toward an improved explanation of income attainment." AMERICAN JOURNAL OF SOCIOLOGY 84 (November):684-97. 1981 "Reoalibrating occupation." SOCIOLOGY AND SOCIAL RESEARCH 65 (April). ' ' Winn Stephen VanZandt 1982 "Social class and income returns to education in Sweden." Presentation at the annual meeting of the Southern Sociological Society, Memphis, Tn. Wolff, Richard, and Steven Resnick 1982 "Classes in Marxian Theory." REVIEW OF RADICAL POLITICAL ECONOMICS 13 (Winter):1-18. 246 Wright, Erik Olin 1976 "Class boundaries in advanced capitalist societies." NEW LEFT REVIEW No. 98 (July-August):3-41. 1978a CLASS, CRISIS, AND THE STATE. London: New Left. 1978b "Race, class, and income inequality." AMERICAN JOURNAL OF SOCIOLOGY 83 (May):1368—97. 1979 CLASS STRUCTURE AND INCOME DETERMINATION. New York: Academic. ' 1980a "Class and occupation." THEORY AND SOCIETY 9 (January):177-214. ‘ 1980b "Varieties of Marxist conceptions of class structure." POLITICS & SOCIETY 9 (3):323-70. 1980c Review of Coleman and Rainwater, 1978. AMERICAN JOURNAL OF SOCIOLOGY 85 (6):1433-39. 1981 "Reply to 'Explaining the salaries of managers.'" CONTEMPORARY SOCIOLOGY 10 (September):610-12. ----- , and Luca Perrone 1977 "Marxist class categories and income inequality." AMERICAN SOCIOLOGICAL REVIEW 42 (January):32-55. ----- ,Cynthia Costello, David Hachen, and Joey Sprague 1982 ”The American class structure." AMERICAN SOCIOLOGICAL REVIEW 47 (December):709-26. ----- , and Joachim Singelmann 1982 "Proletarianization in the changing American class structure." MARXIST INQUIRIES (AMERICAN JOURNAL OF SOCIOLOGY 88, Supplement 1982):S176-S209. ' Wunderlich, Gene 1978 FACTS ABOUT U.S. LANDOWNERSHIP. Agriculture Info- mation Bulletin 422. Washington, D.C.: Economics, Statistics, and Cooperatives Service, U.S. Department of Agriculture. ----- and Russel W. Bierman 1958 "What do we mean by 'ownership'?" Pp. 287-94 in LAND. The Yearbook of Agriculture, 1958, U.S. Department of Agriculture. Washington, D.C.: Government Printing Office. ----- and Walter E. Chryst 1958 "Farm tenure and the use of land." Pp. 295-301 in LAND. The Yearbook of Agriculture, 1958, U.S. Department of Agriculture. Washington, D.C.: Government Printing Office. -