Uncovering the Hidden Work of Women in Family Businesses

A History of Census Undernumeration

Lisa Geib-Gunderson

• Publisher: Taylor & Francis
• Year: 2016
• Print ISBN: 9781138244368
• eBook ISBN: 9781351996679
• Edition: 1
• Topic: Business
• Subtopic: Business General
• Index: Business

Data from the United States Census of Population indicate that there has been a dramatic increase in the labor force participation of married women over the twentieth century. This book, first published in 1998, takes issue with this well-known stylized fact. Whereas the labor force literature comments extensively on men's transition from home production to market work, the effect on women's employment has gone more or less unnoticed. The objective of this book is to uncover the work usually omitted from descriptions of wage work and housework – that is, work done in the household for market use – and to examine the various implications of this omission for analysing married women's participation in GNP-producing work over the course of the past century.

CHAPTER I

An Historical Description of Married Women’s Labor

THE U-SHAPED PATTERN OF MARRIED WOMEN’S LABOR SUPPLY

Data from the United States Census of Population indicate that there has been a dramatic increase in the labor force participation of married women over the twentieth century. This book takes issue with this well-known stylized fact.

Census figures indicate that in 1900 the labor force participation rate of married women was 5.3 percent and that by 1990 this rate had climbed to 57.7 percent (see Table 1.1).¹ Virtually all studies of early female labor force participation in the U.S. are based upon the census, which first collected data on women’s occupations in 1860, first collected this data by marital status in 1880, and first published aggregate statistics of marital status and occupation in 1890.² Extrapolation to earlier years is based upon our knowledge of production practices in the colonial period of the United States. Married women’s participation is thought to have followed a pronounced U-shaped pattern, first falling and then rising with industrialization.³

Claudia Goldin (1994) explains that the basis for this pattern rests in our understanding that in the colonial period, household production was the dominant form of market activity with married women playing key roles in household production. Therefore, during these early years, women’s participation rates were very high. The left-hand, or declining phase of the U is said to have resulted from the separation of home from work in the early stages of industrialization (between the colonial period and the early twentieth century). In this phase, men are said to have followed “work” out of the home while their wives remained in the home, specializing in the production of goods and services for the consumption of their families. Since this household production is not generally defined as “work”, married women’s employment rates fell. The right-hand, or increasing phase of the U is said to represent a new era in which educated married women, after World War II, increasingly found employment opportunities in while collar occupations outside the home.

Whereas the labor force literature comments extensively on men’s transition from home production to market work, the effect on women’s employment has gone more or less unnoticed. We know from labor force statistics that this structural shift was accompanied by a decline in the proportion of men who were self-employed and an increase in the number who work for wages or salaries. However, the effect of such structural shifts on women’s economic activity is not completely clear since, for the most part, the United States labor force statistics do not report their participation as unpaid family workers.⁴ As women shifted from unpaid family labor to more easily recognizable wage work, the labor force statistics reported an increase in the general rate of participation. It appears, therefore, women’s increased labor force participation was brought about by industrialization.

There have been numerous feminist critiques of this conventional belief which have as their premise a challenge to the categorization of the unpaid household production of married women as something other than “work”.⁵ According to this view, the amount of work performed by married women did not change over the course of industrialization but, rather, the location of their work activity changed—especially relative to that of men.

The starting point of this book is fundamentally different. I accept, for example, the standard definition of “work” as involvement in GNP-producing activity. Instead, I focus on the timing and extent of the decline in the family economy (or the unity of home and work) that facilitated the productive employment of married women prior to industrialization. Census data indicate that self-employment was chosen by families in the American economy well into the twentieth century. This institution, therefore, remained a viable one for women’s participation in GNP-producing activities far longer than is generally recognized.⁶ The objective of this book is to uncover the work usually omitted from descriptions of wage work and housework—that is, work done in the household for market use—and to examine the various implications of this omission for analyzing married women’s participation in GNP-producing work over the course of the past century. In so doing, I find levels and trends in participation that are significantly different from the standard view.

FROM HOUSEHOLD PRODUCTION TO LABOR MARKET PARTICIPATION

Work performed by both men and women can be categorized into two basic types: work involved in the production of goods and services to be sold in the market (GNP type) and work involved in the production of goods and services to be consumed directly (non-GNP type). Both types of work are necessary in any economy and both can be performed at home or outside the home. Below I reproduce a chart devised by Penelope Ciancanelli (1983) which categorizes this labor by the location of production (home/outside the home) and the destination of outputs (market/own use).⁷

	For Market	For Own Use
At Home	Farm wives harvesting cash crop Taking in boarders and lodgers Piece work for local factory	Repairing and improving home Cooking and cleaning Child care
Outside the Home	Wage work in office/factory Peddling, selling Running a business outside home	Subsistence farming

Figure 1.1: The Locational and Destination Dimensions of Women’s Work

These locational and destination dimensions of work are of particular interest in the study of women’s economic activity during the early decades of this century. While in the current period in the United States most home production is for own use and most “outside the home” production is for market, this was not always the case. In the years around the turn of the century and before, many women produced market goods at home. However, this does not mean that their work was entirely devoted to the production of use values for the consumption of their families.

The “social relationship” of labor is analyzed by Ciancanelli using Census-related terms describing work relations. This analysis is characterized in Chart 2 below.⁸ The point of this chart is that, although both unpaid family labor (production of goods and services within the home for sale in the market) and wage work are market work, they are fundamentally different in terms of institutional and societal ideology. The implications of these differences for counting women’s work are discussed in the next section.

	Social Relations
	Unpaid Family Worker	Self-Employed	Employee	Employer
Own Use	Housework	Subsistence farming		Supervising servants
Market	Farm wives Small business	Farmer, Artisan Boarding services Taking in laundry	Wage worker	Running a small business with hired help

Figure 1.2: The Social Relations of Production

EARLY CENSUS BIASES

In this book, I argue that married women were, in fact, participating in the labor market through their activities in family businesses much later in the twentieth century than is generally recognized. A close examination of Census enumerator instructions reveals a number of factors that led to the exclusion of many married women from official employment statistics in the earliest years of data collection (1880 through 1930).⁹ The census office itself recognized this problem and the possibility of enumerator error was openly acknowledged:

Women and children are omitted in large numbers … either through failure of the enumerator to ask the questions related to occupation concerning such persons, assuming that they have no avocation outside their homes, or from the indisposition of the persons themselves or the heads of families to speak of them as in employment.¹⁰

There is reason to doubt the standard U-shaped story and, in particular, the low rates of participation recorded for married women (the bottom of the “U”) around the turn of the century. Understanding the factors that led to this omission makes this fact clear; these are discussed below.

Changes in definition: The recorded level of married women’s participation may have been too low in the earliest years of the census due to the manner in which a person’s employment status was recorded by enumerators. The technical definition of employment has changed since 1880, becoming more inclusive of women’s work since 1940. Prior to 1940, a person was recorded as “gainfully occupied” if he or she (but usually he) reported having an occupation, whether or not the person was currently working, retired, or working only seasonally.¹¹ The socia...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Original Title Page
• Original Copyright Page
• Dedication
• Table of Contents
• Tables and Figures
• Acknowledgments
• Chapter I: An Historical Description of Married Women’s Labor
• Chapter II: The Apparent Undercount of Productive Women in U.S. Censuses
• Chapter III: A History of Wives’ Participation in Family Businesses
• Chapter IV: Explaining the Racial Gap in Married Women’s Labor Force Participation
• Chapter V: Conclusions, Implications, and Extensions
• Bibliography
• Index