Dodging the glass ceiling? Networks and the new wave of women entrepreneurs.(Statistical Data Included)

Stephan Weiler [*]

Abstract

Openings of women-owned businesses have radically accelerated recently. This paper explores the causes and results of this phenomenon. Noting the predictive weaknesses of the canonical neoclassical perspective, an extended institutional framework incorporating the impact of male-dominated networks seems to better explain women's situations in both the traditional and entrepreneurial labor markets. Theoretical and empirical evidence points to the paradoxically obstructive role of information networks as the source of women's market difficulties. In the light of this, the paper considers possible motivations for the continued influx of women to entrepreneurship and its potential implications for women's economic status. [C] 2001 Elsevier Science Inc. All rights reserved.

1. Introduction

The relative surge of women entrepreneurs in the last decade has been nothing short of revolutionary. Census data (Tables 1 and 2) provide clear evidence of the rapid increase in entrepreneurship among women, as both the creation of firms as well as sales per firm grew more rapidly for women than men entrepreneurs. According to the National Association of Women Business Owners (NFWBO) in 1995 there were 7.7 million women-owned businesses in the US employing 15.5 million people and generating approximately $1.4 trillion in sales (including the sales of full corporations). From 1990 to 1994 the number of women-owned businesses grew 43% (NEWBO, 1995, p. 2). Current estimates indicate that women own approximately 30% of the businesses in the US, a share which is likely to rise to 50% by the year 2000 (Rios, 1997). The rising tide of women entrepreneurs is unmistakable.

Little research has been done on the causes and consequences of this rapid growth in entrepreneurship among women. One of the main impediments to such work is the lack of data on women entrepreneurs (Moore & Buttner, 1997, p. 13-14). This paper attempts to review the existing literature on this topic, analyze the existing data, and provide an analytical framework for better understanding women's entrepreneurship. Neoclassical economics provides both the tools and the perspective to begin to understand this economic phenomenon. Yet, given the influence of social norms and established patterns of behavior, institutional detail is crucial in framing constraints and interactions. This paper thus provides a neoclassical and institutional synthesis of theory and evidence to better understand the new wave of women entrepreneurs. The principal finding is that established networks, which normally are assumed to increase economic efficiency by reducing transaction costs, may paradoxically represent significant impedim ents to women's success in both traditional and entrepreneurial labor markets.

2. Framing the issues

Increased entrepreneurship by women could be a reaction to continuing discrimination in the formal labor market. Despite mounting legal support and slow erosion of wage and opportunity gaps, women are finding themselves stalled short of equality in the labor market. Neoclassical economic theory suggests that efficient markets will tend to pay workers according to their productivity. While it is true that women are often mired in lower-productivity occupations, which accounts for some of the female/male wage gap, women are also paid less for comparable work (Strober & Arnold, 1987). These facts suggest that, in addition to occupational segregation, wage discrimination for equally placed and productive workers continues.

Yet such long-term discriminatory gaps seem to defy the most basic economic logic. Given the profit motives of entrepreneurs, there seems to be considerable incentive to hire women at wages between their current market value and their productivity. Such enterprises should then have a distinct cost advantage over discriminatory firms, and have greater success in the marketplace. Economic theory suggests that the surge of women-initiated enterprises may indicate that women themselves are hoping to take advantage of the unprofitable logic of discrimination.

This explanation for the growth 'in women's entrepreneurship provided by neoclassical theory thus provides a convenient starting point for our analysis. Two unavoidable issues complicate the neoclassical view however, and require clarification from institutional details. First, discrimination has persisted, despite the legal barriers and profit vulnerability of discriminators. While effective discrimination against women may be easing, the longer-term effects of established male networks and their impact on job access may in effect cause hysteresis in hiring patterns. Thus, positive discrimination towards males may be responsible for prolonging the difficulties faced by women in the traditional labor market. Second, women-owned businesses appear to be performing more poorly than their male brethren despite their supposed competitive advantage from nondiscriminatory hiring. Networks, which generally are touted as providing valuable information conduits to efficient markets, may once again be at the source of women's difficulties, as their firms may find themselves struggling against an established male-dominated system of customers, suppliers, and creditors.

The paper proceeds as follows. Section III reviews the current understanding of women's position in the labor market, by formally presenting and extending Arrow/Becker's classic model to include institutional factors in an otherwise neoclassical model. The implications of this model are then compared with recent statistics on the creation and failure rates of firms headed by women and by men. Section IV further extends the formal institutional facets of the model, explicitly incorporating the impacts of discriminatory customer/supplier networks, and surveys further rationales for female entrepreneurship. Conclusions are offered in the final section.

3. Battling networks and "glass ceilings"

3.1. Economic logic and illogic

Even when all possible factors influencing an individual's productivity, such as experience and education, are taken into account, women's pay continues to lag about 20% behind that of men. When one further considers the apparently significant effect of occupation type on wage rates, the gap narrows to about 10% (Devine, 1994a). While both of these differences have declined over time, the disparities nevertheless remain significant and overall convergence has in fact slowed. Both in terms of access to higher-wage occupations as well as equal pay for comparable work, women are apparently still subject to discrimination.

Yet as both Becker (1971) and, more explicitly, Arrow (1972) indicated in their studies of black/white discrimination, such practices are inherently contradictory to the survival-of-the-most-profitable axiom of the free market. In a purely competitive market, less prejudiced owners should hire workers facing discriminatory wages (i.e., earning less than their productivity contribution to the firm), pay them a wage between their market wage and their true productivity, and effectively drive discriminatory firms out of business. While imperfect competition and resultant product and capital market rents may allow a discriminatory "firm to remain in business and indulge its distaste ... for hiring Blacks" (Arrow, 1972, p. 91), nondiscriminatory firms should still be more successful. In any case, wage differences should fall, which in fact they slowly have.

Yet, to this point, there has been relatively scarce direct evidence of such labor market arbitrage. "Sherlock Holmes, a man much concerned with the formulation of hypotheses for the explanation of empirical behavior, once asked about the barking of a dog at night. The local police inspector, mystified as usual, noted that the dog had not barked at night. Holmes dryly noted that his silence was precisely the problem" (Arrow, 1972, p. 91). So far, it seems that the dog has indeed not barked.

However, the popular press has recently highlighted two interestingly concurrent phenomena. Women's frustration at discriminatory behavior, especially the corporate "glass-ceiling," appears to be growing. In particular, access to the highest echelon of jobs is still a major problem. The result is that talented productive women are finding that they cannot gain access to the highest levels, and highest pay scales, of the economy, effectively halting their previously steady progress to equality. Twenty-seven percent of women-initiated enterprises are in fact motivated either by barriers to advancement or lack of challenges in current careers (Jackson, 1998).

Devine (1994a) found that the relative returns to skill for women were higher in self-employment than within the wage-and-salary sector. Consistent with the glass-ceiling hypothesis, which …