ArticlePDF Available

Working women in the city and urban wage growth in the United States: WEINSTEIN

Authors:

Abstract and Figures

Although the female labor force participation rate of women has been steadily rising in the United States, there is substantial variation across cities. Previous cross-county studies find that gender inequality in employment reduces economic efficiency hindering growth. This result is examined in a regional context, across metropolitan areas in the United States. Throughout multiple model formulations including instrumental variables approaches, higher initial female labor force participation rates are positively related to subsequent wage growth in metropolitan areas between 1980 and 2010. Specifically, every 10 percent increase in female labor force participation rates is associated with an increase in real wages of nearly 5 percent.
Content may be subject to copyright.
1
WORKING WOMEN IN THE CITY AND URBAN WAGE GROWTH IN THE U.S.
December 2016
Amanda L. Weinstein
University of Akron, Department of Economics
259 S. Broadway St., Akron, OH 44325
E-mail: aweinstein@uakron.edu
Abstract:
Although the female labor force participation rate of women has been steadily rising in the U.S.,
there is substantial variation across cities. Previous cross-county studies find that gender
inequality in employment reduces economic efficiency hindering growth. This result is examined
in a regional context, across metropolitan areas in the U.S. Throughout multiple model
formulations including instrumental variables approaches, higher initial female labor force
participation rates are positively related to subsequent wage growth in metropolitan areas
between 1980 and 2010. Specifically, every 10 percent increase in female labor force
participation rates is associated with an increase in real wages of nearly 5 percent. (JEL J16,
R11, J24)
Keywords: Female labor force participation rate, Female labor supply, gender equality,
productivity, wage growth
2
Acknowledgements: I am very grateful to Mark Partridge, Bruce Weinberg, two anonymous
referees and the editor for their guidance. I also thank NARSC and SRSA conference
participants for comments. All errors are my own.
3
1. INTRODUCTION
Gender studies in economics are often centered on the impact the economy has on
women e.g., focusing on the gender wage gap and the determinants of women’s increasing
labor force participation. The persistent growth in women’s participation in formal labor markets
is arguably the most significant change in the economy in the last century. Female labor force
participation in the U.S. has nearly doubled from less than 34 percent in 1950 to 58 percent in
2012 (U.S. Bureau of Labor Statistics). Such a significant change surely warrants research into
the driving forces. However, far less research has examined how growing numbers of women in
the labor force are in turn affecting the economy.
There is substantial variation in female labor force participation rates (FLFPR) across
cities in the U.S. (ranging from 36 percent to 68 percent in 1980
1
). Previous research finds that
differences in FLFPR are not due to geographical variations in the characteristics of women, but
that there are fundamental differences in these labor markets (Odland and Ellis, 1998). Peck
(1996) suggests that patterns of labor market segmentation and whether and how women are
incorporated into the economy can be rooted in the local economy itself as a response to the
industry structure and social history of an area. Recent research has focused on the local
determinants of female labor market decisions focusing on how average commute times, local
gender role attitudes and the local business climate, for example, affect women’s decision to
enter the workforce (Black et al, 2014; Patrick et al., 2016). Various fundamental differences in
labor markets affect FLFPR and how women are incorporated into the economy, but may also
impact local economic growth.
1
Calculated using 1980 Census data from IPUMS.
4
Previous cross-country studies find that economic development increases women’s status
in society, raising female labor force participation rates (Psacharopoulos and Tzannatos, 1989;
Goldin, 1995). Goldin (1995) further asserts that not only does economic development promote
gender equality, but that gender equality and the participation of women in the economy also in
turn promotes economic development. Previous cross-county studies confirm that gender
inequality in education, health, and employment reduces economic efficiency hindering
economic growth (Dollar and Gatti, 1999; Klasen, 2000; Knowles et al., 2002; Morrison et al.,
2007; Duflo, 2012). However, cross-country studies may miss important aspects of the complex
interaction between gender and economic development that a regional analysis may better
capture (such as the length of the average commute) or they simply may not hold at lower levels
of geography (especially within a developed country like the U.S.). One study by Esteve-Volart
(2004) narrows the geographic scope by analyzing the relationship across regions in India
finding that a 10 percent increase in the female-to-male ratio of workers increased per capita
state productivity by 8 percent, but it seems no such study exists for the U.S.
This paper contributes to the literature by examining the impact of female labor force
participation rates on wage growth in U.S. metropolitan areas to determine whether previous
cross-country findings hold at lower levels of geography in the U.S. If the entrance of women
into the workforce impacts the local economy only by increasing the quantity of labor i.e. by
responding to a shift in the demand curve or by shifting the labor supply curve, then their
increasing numbers will either have no impact at all or will drive wages down especially if they
are crowded into low-paying female-dominated occupations and industries (either by choice or
discrimination). On the other hand, if the entrance of women into the workforce changes the
qualitative aspects of workers, -i.e. if women are contributing a unique and complementary
5
skillset, then we would expect productivity and efficiency to increase especially as women are
more fairly incorporated into the economy. Thus, estimating the impact of women’s labor force
participation is likely inextricably linked to gender equality and how fairly women are
incorporated into the economy.
Census data from 1980, 1990, and 2000 and American Community Survey data from
2010 are used to estimate the impact of FLFPR (as well as measures of how fairly women are
incorporated into the economy such as industry and occupation segregations and the gender
wage gap) on subsequent MSA median wage growth. Various models are estimated including
instrumental variables approaches to help eliminate the influence of unobserved factors and
determine the direction of causality. Various models suggest that every 10 percent increase in
FLFPR increases real median wage growth by approximately 5 percent, but there is limited
evidence on the impact of other measures of gender equality in how women are incorporated into
the economy.
2. GENDEFR EQUALITY AND ECONOMIC DEVELOPMENT
Previous literature has focused on the determinants of female labor force participation
rather than the effect women may have on the economy. These studies have focused either on the
impact of demand side factors increasing the benefits of female employment i.e. higher wages,
the availability of part-time jobs and jobs in industries and occupations more favorable to women
(Goldin, 2006; Blank, 1989; Smith and Ward, 1985; Weinberg, 2000) or on the impact of supply
side factors lowering the opportunity cost of employment for women through better childcare,
lower fertility, and home appliance advances or factors pushing them into employment such as
divorce or their husband’s labor force status (Waite and Stolzenberg, 1976; Connelly, 1992;
6
Mincer, 1962; Goldin, 2006; Semyonov, 1980; Mincer, 1962). The most dramatic change in
FLFPR stems from the rising labor market participation of married women. Lombard (1999)
finds that the increase in the labor market participation of married women can be attributed not to
demand shifts but to supply shifts, specifically married women’s tastes and attitudes toward work
and higher divorce rates. In general, the research seems to be somewhat divided on whether
demand side or supply side factors play a larger role in the labor force participation trends of
women in the U.S. (Chinhui and Potter, 2006). Regardless of which factors play the largest role
in encouraging women to join the workforce (or because there are both demand shifts and supply
shifts), it is still critically important to estimate the resulting impact of growing numbers of
women in the workforce, if any.
Previous research has shown that higher wages and economic development increase
women’s status in society and their value in the economy, raising their labor force participation
rates (Smith and Ward, 1985; Clark et al., 1991). Specifically, a u-shaped relationship between
economic development and female labor force participation has been found (Psacharopoulos and
Tzannatos, 1989; Mammen and Paxson, 2000; Goldin, 1995). As an economy develops,
women’s labor force participation rates may initially decline but will eventually increase creating
the u-shape. The U.S. and other developed countries now find themselves in the upward sloping
portion of this u-shape as economic development and women’s increased participation in the
formal economy seem to go hand in hand.
Although Goldin (1995) and others show that economic development promotes gender
equality (and women’s participation in the labor force), she also asserts that gender equality
promotes economic development. Morrison et al. (2007) and Duflo (2012) review the literature
linking gender equality to reductions in poverty and gains in productivity across countries.
7
Gender inequality in the investment of education and health care is a market failure that lowers
economic growth (Dollar and Gatti, 1999; Knowles et al., 2002). As gender inequality in
education decreases, human capital levels of spouses will become more equal and women’s time
more valuable. Couples will then substitute the quantity of children for quality reducing fertility
rates while increasing the total human capital level in society through the effect on both women
and on children (Lagerlof, 2003; Klasen, 2002). Gender inequality in employment (and
education) reduces the probability women enter the workforce decreasing the average ability of
the workforce (Klasen, 2000; Klasen and Lamanna, 2009). When women have better access to
labor markets (as well as education and health care), countries become more productive.
However, this line of research has focused on developing countries and cross-country studies
whereas this paper examines this relationship at a smaller geographic scale, namely U.S. cities.
The Impact of Gender across Labor Markets
Initial research examining the impact of women’s increased participation in the labor
market measured the impact of the log ratio of female to male labor supply by state induced by
differences in World War II mobilization rates on individual log wage levels (Acemoglu et al.,
2004). Acemoglu et al. find that female labor market participation during WWII lowered wages.
However, Goldin (1991) finds that the impact of WWII on women’s participation in the
economy was relatively modest and largely temporary. The impact of women during WWII is
also unlikely to be representative of women’s impact on the economy. Lower-educated women
were more likely to be pulled into the labor force and were more likely to work in manufacturing
industries that were not particularly welcoming to women (Goldin and Olivetti, 2013).
Additionally, Doepke et al., (2007) find that the female labor supply shock during WWII
8
increased labor market competition for young women which increased both their labor market
exit rates and their fertility. Differences in WWII mobilization rates may not be exogenous if the
way in which women were incorporated into the economy affects the outcomes (wages) of
women and how women may affect men and the economy.
After WWII when women’s role in the economy subsequently entered what Goldin
(2006) calls ‘the revolutionary phase,’ women began to join the workforce more out of choice
than necessity. It is the impact of this choice and this dramatic shift that has yet to be measured
in terms of its impact on regional economies. With access to the pill, women chose to postpone
marriage and childbirth to attend college and make long-term career plans rather than allowing
these factors to dictate their labor force participation decisions (Goldin and Katz, 2002; Bailey,
2006; Waite and Stolzenberg, 1976). Many women now plan to enter the labor force before
marriage, before knowing their husband’s income. The elasticity of married women’s
employment with respect to their own wages is positive and nearly double the elasticity with
respect to their husband’s wage (Mincer, 1962). Heckman and Willis (1977) found married
women’s choice to work followed a u-shaped probability curve. Although the labor force
participation rate for married women was 40 percent, most women’s probability of entering the
workforce was either near 0 or near 100 percent. For those with probabilities near 100 percent,
their labor market decisions more closely resemble that of men’s. Thus, Juhn and Kim (1999)
examine whether more recent (1940-1990) increases in the female labor supply substituted for
male labor lowering male wages. After controlling for the effects of demand side factors, they
find little evidence that women are substitute for men.
With more women now in the labor force than not, social norms no longer work against
the impetus for women to enter the workforce. Women’s steady rise in participation suggests that
9
it is not a singular event like WWII or the pill that is the main factor behind this increase. The
benefit of working for women has steadily increased while the benefit of staying home has
steadily declined. Women’s attitudes toward working especially married women’s tastes have
continued to change. Additionally, local labor markets can differ in important ways that impact
tastes and the decisions of women to participate in the economy such as commute times or
gender role attitudes that may impact how fairly women are incorporated into the economy.
The Gender Wage Gap
When analyzing changes in regional productivity as measured by wages, it is important to
discuss the implications of the gender wage gap and whether it necessarily implies there is a
gender productivity gap or that women may not be fairly incorporated into the labor force. If the
gender wage gap is a result of differences in productivity, then a higher FLFPR will be
associated with lower average levels of productivity in metropolitan areas. Table 1 below shows
that the (unadjusted) wage gap for metropolitan areas since 1980.
TABLE 1: MSA Average Wages by Gender over Time
Year
Women’s Average Wages
Men’s Average Wages
Wage Gap
Nominal
Nominal
Real ($1999)
1980
$5.25
$8.15
$18.39
64.0%
1990
$9.36
$13.40
$17.96
69.8%
2000
$13.92
$18.60
$18.60
74.8%
2010
$18.07
$22.78
$17.98
77.8%
Source: Calculated using respondent information on wage and salary earnings per year, weeks worked per year, and usual hours
worked per week from ACS and Census Data fro
Although it has been narrowing, the gender wage gap persists and its convergence has
been slowing (Blau and Kahn, 2006). Blau and Kahn (1997) find that women earn 72.4 percent
as much as men, but when controlling for factors such as education, experience, and occupation,
that percentage rises to 88.2 percent (the adjusted wage gap). A significant portion of the gender
10
wage gap is due to differences in occupations and industries. Human capital theory suggests that
the wage gap may be evidence of women simply sorting themselves into lower paying
occupations with more menial work in their own self-interests (possibly to incorporate various
non-pecuniary benefits such as a flexible schedule). However, England (1982) and Bergmann
(2005) amongst others find that it is unlikely that human capital theory can explain much of the
observed gender differences in occupation and industry and the resulting gap in wages.
Occupation and industry segregation and the remaining unexplained wage gap may be due to
discrimination, an economic inefficiency that previous studies have shown decreases
productivity. As evidence, Cohen and Huffman (2003) find that high levels of occupational
segregation by gender are associated with all women’s work being devalued.
According to Becker’s (1957) theory of discrimination, competition between firms
should eliminate this discrimination as profit-maximizing firms hire women to lower their labor
costs and wages should converge over time. As evidence, Hellerstein et al. (2002) find that hiring
more women increases profits in plants especially those with market power, but found no
evidence of market forces pressuring firms to eliminate discrimination. It may be women
themselves and their increased participation in the labor forces which is helping to reduce
discrimination and close the gender wage gap. Higher FLFPR have been associated with lower
industry and occupation segregation (Cotter et al., 1998). Women entering male-dominated
industries may also break up an inefficient “old boys’ network” especially as women climb the
ranks to become managers and CEOs. Women’s increased share of high-status managerial
positions can be beneficial for women in lower ranking positions reducing inefficient
discrimination and the gender wage gap (Cohen and Huffman, 2007). Esteve-Volart (2004)
11
found that a 10 percent increase in the ratio of women in managerial positions is associated with
a 2 percent increase in per capita output in India.
Lower occupational segregation benefits all women as low-paying female-dominated
sectors become less crowded (Cotter et al., 1997), but it may also benefit men as efficiency and
productivity increase. As the allocation of jobs and tasks is based more on qualifications than on
gender, the closer correspondence between earnings and skills incentivizes both men and women
to develop their skills increasing productivity. As women are better represented in the
workforce, Hellerstein and Morrill (2011) find that fathers have more incentive to invest in their
daughter’s human capital thereby increasing the likelihood and ability of these daughters to enter
into their father’s occupation. Thus, higher FLFPRs are associated with higher levels of human
capital and lower industry and occupation segregation. Additionally, plant level data from
Hellerstein et al. (2002) also provides evidence that there may be a complementarity between the
labor of men and women. Thus, there may be gendered skills impacting the productivity of men
and women and the gender wage gap may not necessarily imply a gender productivity gap.
Gendered Skills
A shift in the labor supply typically applies downward pressure on wages. However, if
the influx of labor is highly skilled or has specialized skills, the higher labor supply will increase
productivity and wages. Women entering the formal labor market may have specialized skills
and raise human capital levels. Women have surpassed men in obtaining bachelor’s degrees and
may also bring a unique skillset to the table. Genetic differences between the sexes such as
differing hormone levels have been found to result in differing economic and general decision-
making behaviors. For example, lower testosterone levels are associated with higher levels of
12
risk aversion (Sapienza et al., 2009). These gendered skills may provide previously male-
dominated work environments with a variety of complementary skills. Female representation
helps firms better represent their consumers as women make approximately 75 percent of
household decisions (according to recent Mediamark research). As anecdotal evidence, the first
dishwashing machine was invented by a woman, Josephine Cochran, in 1886 at a time when no
man saw the point. The company she founded to manufacture these dishwashers is now
KitchenAid. These same dishwasher reduced the costs of working outside the home for women
further increasing FLFPR.
Women’s presence in the economy may change the skill profiles of cities. Women are
more likely to have altruistic traits (and less likely to have Machiavellian traits) and volunteer in
philanthropic organizations for which they pay a good citizen penalty (Fortin, 2005; Grove et al.,
2011). Thus, women may provide unique value-added skills to the economy that are not
necessarily fully rewarded monetarily. Emerging research on the geographic distribution of skills
in cities finds that productivity effects associated with agglomeration are larger for workers with
cognitive skills and interpersonal skills (Bacolod et al., 2009a, 2009b, 2010) both of which are
traits especially exhibited by women (Bacolod and Blum, 2010; Borghans et al., 2006; Bacolod,
forthcoming). Thus, Phimister (2005) finds a higher urban premium for women than men. Higher
wages for women in cities will certainly increase FLFPR as will other local factors that affect
women’s tastes for working and their supply of labor, but how will these working women in the
city in turn impact their local economies? Despite the higher urban premium for women
(compared to rural women), recent work by Bacolod (forthcoming) suggests that women’s
agglomerative returns to cognitive and social skills is significantly less than men’s.
13
3. THEORETICAL FRAMEWORK
To examine the impact of FLFPR on wage growth in cities, previous papers such as
Glaeser et al. (1992) and Acemoglu et al. (2004) provide a model with which to measure the
growth of cities as a result of changes in labor inputs. We assume labor markets are competitive
and consist of both the quantity of labor () and the quality of labor ( in the aggregate
production function given in equation 1. The quality of labor ( draws upon the standard
Mincer human capital function and accounts for various characteristics of labor inputs that
impact productivity such as educational attainment as well as other worker characteristics such
as gender and race. The aggregate production function also includes technology ().
  (1)
In a competitive market, labor inputs are paid a wage rate () equal to the marginal productivity
of labor shown in equation (2).
  (2)
Assuming, then equation 2 can be rewritten as equation 3. Wages are a
function of the supply of labor hours and human capital. Women’s participation in the labor
force will likely impact both.
   

 (3)
To examine the impact on growth rates, equation 2 is translated into growth rates (equation 3).


 
   
(4)
In order to limit potential endogeneity issues, the initial values of right-hand side
variables are used. A Mincer human capital function is used ( ) where accounts
for levels of schooling and demographic variables such as gender and race. The empirical
estimation of the impact of these variables at the metropolitan level is based on equation 6.
14

       (5)

       (6)
Similar to Glaeser et al. (1992), we assume that technology () has both national and
local components. The national component consists of various nationwide trends in technology
and in prices for specific industries (and occupations).
2
The local component of technology
includes various externalities associated with agglomeration economies (urbanization and
localization economies).
3
4. DATA
U.S. Census Bureau data from 1980, 1990, 2000, and ACS data from 2010 from IPUMS
offer detailed information on labor force statistics.
4
This detailed information is used to estimate
how FLFPRs affect median wage growth in metropolitan areas. Real hourly wages are first
calculated using the wage and salary income (weighted by the CPI99) and the respondents
reported weeks worked in the previous year and usual hours worked each week. Median hourly
real wage by year and metropolitan is then calculated as well as the change in real hourly wages
over time.
5
Figure 1 depicts the change in median wages across the U.S. from 1980 to 2010.
Areas we would expect to see high wage growth are apparent such as New York and the
2
In the next section we describe various measures such as the Bartik type of shift share to account for these
nationwide trends.
3
Total employment and industry diversity measure the impacts of urbanization economies whereas industry
shares measure the impact of localization economies. Additionally, monthly mortgage payments (as a share of
income) and average commute times measure externalities associated with either production or consumption
amenities of a city.
4
The 1980 1% weighted metro sample is used along with the 1990 5% weighted sample, 2000 5% weighted
sample, and 2010 1% weighted sample.
5
Median hourly wages also help eliminated the influence of outliers without relying on somewhat arbitrary cutoff
values.
15
northeastern seaboard along with other large cities such as San Francisco. The dependent
variable in the estimation is the logged wage growth as suggested in equation 6.
FIGURE 1: 1980-2010 Change in Median Wages
Source: Census and ACS Data from IPUMS
IPUMS data is used to calculated total employment ( from equation 6) as well as
various measures of the human capital such as the percent of workers (older than 22) that are
have a college level education.
6
IPUMS data also allows us to calculate various demographic
information about the labor force especially regarding gender and the FLFPR. Figure 2 shows
how FLFPR varies across metropolitan areas in the U.S. where each shade represents a quantile
with darker quantiles representing higher female participation. Table 2 shows the average
FLFPR across MSAs over time. Women made the largest gains in FLPRs in the 1980s (as well
as the largest gains in closing the wage gap- Table 1).
6
We consider someone to be college level education if they reported having 4 or more years of college education.
16
FIGURE 2: 1980 Female Labor Force Participation Rate
Source: Census Data from IPUMS
TABLE 2: FLFPR over Time
Year
Mean
Standard Deviation
Min
Max
1980
50.4%
5.3%
36.1%
67.8%
1990
57.1%
5.3%
41.8%
70.0%
2000
58.2%
5.2%
39.5%
72.3%
2010
59.4%
4.8%
42.4%
75.7%
Source: Census and ACS Data from IPUMS
In estimating the impact of FLFPR, previous research suggests that it may be important
to how fairly women are incorporated into the economy. To account for this, the index of
dissimilarity is used to measure the dissimilarity between the sexes in terms of their occupation
and industry employment (occupational and industry segregation). In equation 7, W (M) is the
total number of employed women (men) and wi (mi) is the total number of women (men)
employed in occupation or industry i. The resulting number is approximately the percent of
women and men that would have to change occupations to yield the same employment
17
distribution. A lower number indicates more similarity (less occupational or industry
segregation).
 (7)
For ease of interpretation, Figure 3 depicts industry segregation where darker colors
represent more similarity and lighter colors represent more dissimilarity. Similarities in the
pattern of industry segregation and FLFPR are noticeable though not exact. There is a significant
correlation between FLFPR and lower levels of industry segregation (-0.304 with a p-value less
than 0.0001) as well as with lower levels of occupation segregation (-0.573 with a p-value less
than 0.0001).
FIGURE 3: 1980 Industry Similarity by Gender
Source: Census and ACS Data from IPUMS
The gender wage gap is also used as a control for how fairly women are incorporated into
the economy (Figure 4). FLFPR is also associated with a lower gender wage gap or higher values
of the female average wage divided by the male average wage (0.369 with a p-value less than
0.0001).
18
FIGURE 4: 1980 Gender Wage Gap
Source: Census and ACS Data from IPUMS
Table 3 provides other descriptive statistics for the data. It also shows that measures of
dissimilarity between the genders in terms of occupation and industry from 1980 to 2010. Again,
women made notable advances in terms of occupational and industry integration in the 1980s,
but not in subsequent decades.
7
Table 3 also shows various other control that are important to
incorporate into an analysis of the relationship between metropolitan wage growth and FLFPR
such as average MSA commute and average MSA monthly mortgage payment. Black et al.
(2014) finds that cities with lower average commuting times have higher female labor force
participation and workers in cities with lower average commute times are willing to give up
wages for this amenity (Timothy and Wheaton, 2001). Johnson (2014) suggests that higher labor
force participation of married women may increase house prices which may also drive wages up.
7
It is important to note that our measures of occupation and industry segregation are based on broad measures
industries and occupations. Given the small share of some occupations and industries, the data cannot provide
reliable estimates at more refined measures.
19
State fixed effects help control for the possibility of any government policies associated with
higher female employment share and wage growth.
8
TABLE 3: Descriptive Statistics
Variable
1980 (N=255)
1990 (N=238)
2000 (N=283)
2010 (N=287)
Mean
Std Dev
Mean
Std Dev
Mean
Std Dev
Mean
Std
Dev
FLFPR
50.6%
5.2%
57.2%
5.3%
58.2%
5.2%
59.4%
4.8%
Female Emp Share
42.1%
3.0%
45.4%
2.5%
46.7%
1.9%
48.1%
2.3%
Gender Wage Gap
60.5%
5.9%
66.9%
4.8%
72.7%
4.5%
76.9%
5.6%
Industry Segregation
0.331
0.063
0.323
0.060
0.319
0.049
0.340
0.056
Occupation Segregation
0.505
0.052
0.452
0.043
0.454
0.037
0.451
0.045
Total Employment
288,026
599,077
350,660
733,863
359,680
715,050
390,571
777,080
Percent College
20.6%
5.9%
24.7%
6.6%
28.2%
7.7%
31.8%
8.4%
Percent Black
9.0%
8.8%
9.8%
9.1%
10.5%
10.1%
11.5%
10.7%
Average Commute
9.13
1.44
18.8
2.79
21.48
3.30
21.09
3.12
Monthly Mortgage
$614.92
$130.94
$786.52
$228.91
$883.07
$225.87
$940.20
$276.70
Poverty Rate
14.0%
4.8%
15.2%
5.4%
14.4%
4.7%
17.3%
4.8%
Average Wage (Real)
$15.07
$1.72
$14.59
$1.92
$15.35
$2.16
$14.99
$2.18
Median Wage (Real)
$11.37
$1.59
$10.99
$1.69
$11.40
$1.68
$11.72
$1.72
ΔMedian Wages
-$0.35
$1.22
$0.53
$0.57
$0.32
$0.65
N/A
N/A
Source: Census and ACS Data from IPUMS
An initial look at the relationship between FLFPR and metropolitan area median wage
growth without all of these various controls is provided in Figure 5. Although this does not yet
include various controls, there is nonetheless a positive and significant relationship. The scatter
plot shows for example, that well-educated cities like Boston, MA and Stamford, CT clearly
experience higher wage growth than expected based on FLFPRs alone. It also shows that areas
like Flint, MI and Youngstown, OH do particularly poorly even more than their lower FLFPR
would suggest.
The difference between the wage growth of cities like Boston and San Jose and cities like
Flint and Youngstown and the differences in their FLPR also highlights the role of underlying
economic trends specifically regarding industries. The manufacturing industry which tends to be
8
For MSAs bordering state lines, the total employment in the city by state is calculated and the city is assigned a
state fixed effect based on which state holds a larger share of total employment.
20
heavily male dominated has also been experiencing decline over the last several decades (due to
globalization, skill-biased technological change, and another of factors unrelated to its female
employment share).
FIGURE 5: Metropolitan Area Real Wage Growth
Source: Census and ACS Data from IPUMS
Thus, various estimations include both industry and occupation shares of total
employment. The effect of industry diversity (measured as 1 minus the Herfindahl index) and
total employment are incorporated as Costa and Kahn (2000) find that power couples are more
likely to be located in large cities with more diverse industries (both of which have been shown
to increase MSA growth). Bartik (1991) style industry (and occupation) mix predicted wage
growth is used to help root out the interaction between industries, occupations, wage growth, and
Anchorage, AK
Reno, NV
Madison, WI
Washington DC
San Jose, CA
Charlottesville, VA
Elkhart, IN
Boston, MA
Stamford, CT
Cedar Falls, IA
San Diego, CA
Flint, MI
Youngstown, OH
Sharon, PA
McAllen, TX
Fort Myers, FL
Sarasota, FL
Johnstown, PA
y = 8.869x - 4.058
R² = 0.084
-$8
-$6
-$4
-$2
$0
$2
$4
$6
30% 35% 40% 45% 50% 55% 60% 65% 70%
1980-2010 Change in Median Wages
1980 Female Labor Force Participation Rate
21
FLFPR. For each industry (i), the Bartik style industry mix predicted wage growth (equation 8)
uses the metropolitan area industry employment share and national median wage growth of
industry (i) measured as the natural log of to estimate how wages would have grown
in an MSA based solely on its industry (or occupation) structure. Thus, with low wage growth in
the manufacturing industry, for example, the Bartik measure would predict low wage growth in
an MSA that is heavily based on the manufacturing industry (regardless of its FLFPR).

 (8)
The FLFPR could be spuriously related to productivity growth (or wage growth) if
women are sorting into certain sectors (e.g. the services sector) with different wage growth rates
or different demand shocks. Juhn and Kim (1999) stress the importance of accounting for
demand side factors when estimating the impact of the increasing labor supply of women. Bartik
style industry (and occupation) measures are used to predict employment growth in each MSA
(by replacing U.S. wage growth with U.S. employment growth by industry in equation 8).
Women’s initial median wage is also used to account for the demand for women and the
incentive for women to enter the labor force. By accounting for various demand side factors and
other controls, we can better estimate the impact of the FLFPR.
5. EMPIRICAL ESTIMATION
Table 4 provides the results of the OLS estimation of equation 6 with various controls
(listed in Table 3). FLFPRs are consistently associated with higher wage growth. The preferred
model 1 suggests that every 10 percent increase in FLFPR is associated with a 5 percent increase
22
in metropolitan area wages.
9
The coefficients of the control variables are not surprising. Larger
labor markets are continuing to experience wage growth. Higher levels of human capital are
associated with higher wage growth rates (though not significant in many models). Higher
average commute times are associated with higher wage growth as workers must be
compensated for dealing with a long commute. To account for both higher home prices in an
area and higher natural amenities, we use the average total monthly mortgage payment as a share
of worker’s (and spouse’s) income. Workers are willing to give up wages and experience lower
wage growth in order to live in nicer areas.
10
The negative association between the initial median
wage rate and wage growth rates suggests that cities may be converging or that workers are
migrating to cities with high wages, applying downward pressure on wages.
TABLE 4: OLS RESULTS
ln(Median Wage Growth)
VARIABLES
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
FLFPR
0.476***
0.560***
0.641***
0.373***
0.460***
0.460***
(0.082)
(0.081)
(0.113)
(0.075)
(0.080)
(0.081)
MLFPR
-0.269**
(0.108)
Female Labor Share
0.975***
(0.233)
Female Emp Share
0.783***
(0.225)
ln(Med Wage)
-0.729***
(0.040)
ln(Male Med Wage)
-0.321***
-0.276***
-0.304***
-0.308***
-0.314***
-0.304***
-0.298***
(0.042)
(0.042)
(0.043)
(0.044)
(0.044)
(0.045)
(0.047)
ln(Female Med Wage)
-0.409***
-0.360***
-0.419***
-0.380***
-0.386***
-0.426***
-0.426***
(0.050)
(0.047)
(0.051)
(0.051)
(0.051)
(0.053)
(0.054)
ln(Total Emp)
0.006*
0.012***
0.008**
0.006*
0.006
0.008**
0.006
0.006
(0.004)
(0.003)
(0.004)
(0.004)
(0.004)
(0.004)
(0.004)
(0.004)
Percent College
0.0159
0.296***
0.0488
0.204
0.221*
0.0814
-0.00165
0.000246
(0.127)
(0.0778)
(0.126)
(0.124)
(0.124)
(0.117)
(0.130)
(0.130)
Poverty Rate
-0.529***
-0.452***
-0.563***
-0.577***
-0.592***
-0.594***
-0.534***
-0.529***
(0.111)
(0.0940)
(0.111)
(0.109)
(0.109)
(0.105)
(0.111)
(0.111)
9
The FLFPR is measured as the percent of the female population age 16 or older that are in the labor force as used
by the BLS and others.
10
Models with the logged real monthly mortgage payments were also conducted with no significant changes to
other variables.
23
Commute
0.005***
0.006***
0.006***
0.005***
0.005***
0.006***
0.006***
0.005***
(0.001)
(0.001)
(0.001)
(0.001)
(0.001)
(0.001)
(0.001)
(0.001)
Home/Inc
-0.003
-0.004**
-0.003
-0.004**
-0.004**
-0.003
-0.003
-0.003
(0.002)
(0.002)
(0.002)
(0.002)
(0.002)
(0.002)
(0.002)
(0.002)
Gender Wage Gap
-0.150**
(0.069)
Occ Seg
-0.093
(0.099)
Ind Seg
-0.086
(0.093)
Occ/Ind Shares
Y
N
Y
Y
Y
Y
Y
Y
Occ/Ind Pred Growth
Y
Y
Y
Y
Y
Y
Y
Y
State/Year FE
Y
Y
Y
Y
Y
Y
Y
Y
Observations
752
752
752
752
752
752
752
752
R-squared
0.665
0.575
0.669
0.658
0.654
0.701
0.666
0.666
Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1. Additional variables not shown include industry diversity
and percent black.
Measures of how fairly women are incorporated into the economy are explicitly
controlled for in models 6-8. Although the coefficient on both occupation and industry
segregation is negative, they are both insignificant (possibly due to broad industry and
occupation categories used due to data limitations). Some portion of the impact of higher
FLFPRs could simply be due to the narrowing of the gender wage gap. Indeed we find that
metropolitan areas with larger gender wage gaps, experiences higher wage growth and the
impact of FLFPR is smaller.
As a robustness check, we also examine the impact of male labor force participation rates
(MLFPR) on wage growth (model 3). As expected, the coefficient on MLFPR is very different
than for FLFPR with higher MLFPR being associated with lower wage growth and an even
larger positive impact for FLFPR on wage growth. The impact of female share of the labor force
and the female share of employment were also estimated with similar though larger impact on
wage growth. We find that a 10 percent increase in the female employment share is associated
with an 8 percent increase in real wage growth (similar to Esteve-Volart’s paper that found a 10
24
percent increase in the female to male ratio of workers in India was associated with an 8 percent
increase in state per capita productivity).
11
The Impact by Decade
The previous regressions control for year (or decade) fixed effects, but breaking out the
estimation by decade provides additional insight. Because women made large gains in
employment and in the gender wage gap in the 1980s but progress subsequently slowed, the
impact of the FLFPR may also vary by decade. Table 5 shows that the impact of FLFPR does
vary by decade with the largest impact occurring in the 1980s and decreasing each decade after.
Over a 30 year period from 1980 to 2010, a 10 percent increase in the female employment share
is associated with an increase in the wage growth by 6 percent, larger than the shorter 10 year
impact.
TABLE 5: Breakout by Decade
ln(Median Wage Growth)
VARIABLES
1980
1990
2000
1980-2010
FLFPR
0.824***
0.184
0.213*
0.642***
(0.149)
(0.120)
(0.112)
(0.187)
Occ/Ind Pred Growth
Y
Y
Y
Y
Occ/Ind Shares
N
N
N
N
State FE
Y
Y
Y
Y
Year FE
N
N
N
N
R-Sq
0.876
0.654
0.665
0.752
Observations
233
237
282
246
Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1. This model is similar to Model 2 in Table 4.
The Impact by Skill Level and Gender
11
Other models not shown include a regression on average wages instead of median wages which suggested that a
10% increase in FLFPR was associated with a 2% increase in real average wage growth at the 0.6% significance
level. Another model removed the state and year fixed effects with MSA fixed effects with the results suggesting a
10% increase in FLFPR was associated with a 4% increase in median wage growth at the 0.2% significance level.
25
It appears that FLFPRs may be, at least in part, capturing the impact of the narrowing
gender wage gap. Some portion of the closing gender wage gap may be due to increasing
FLFPRs (for example, previous research shows that as FLFPRs increase the higher prevalence of
females in management positions will help close the gender wage gap through various
managerial decisions). Previous research also shows that the gender wage gap has been closing
due to global economic trends such as SBTC and the growing importance of people skills, trends
unrelated to FLFPRs (Borghans et al., 2006). Despite our Bartik measures used to account for the
impact of demand side factors, there may still be some factors that are unobserved. First, we
focus the analysis of the impact of higher FLFPRs to all men and men by education level (the
breakout for women is also provided for comparison). In the next section, we will use 2SLS to
address these concerns. Table 6 shows that a 10 percent increase in the FLFPR is associated with
higher median wages for men by approximately 2.4 percent (about half the overall impact).
TABLE 6: Impact Across Skill Levels
Ln(Median Wage Growth)
VARIABLES
All Men
College
Men
HS Men
No HS
Men
All
Women
College
Women
HS
Women
No HS
Women
FLFPR
0.243***
-0.007
0.321***
0.424**
0.463***
-0.057
0.501***
0.493***
(0.093)
(0.102)
(0.101)
(0.188)
(0.074)
(0.088)
(0.078)
(0.167)
College FLFPR
0.050
0.022
0.063
0.087
-0.005
0.013
0.024
-0.007
(0.060)
(0.071)
(0.065)
(0.127)
(0.049)
(0.058)
(0.053)
(0.109)
High School FLFPR
0.194**
0.102
0.215**
0.362**
0.318***
0.067
0.382***
0.411***
(0.086)
(0.093)
(0.095)
(0.171)
(0.070)
(0.081)
(0.074)
(0.156)
No High School FLFPR
0.095
0.077
0.144**
0.489***
0.278***
0.012
0.323***
0.465***
(0.065)
(0.072)
(0.068)
(0.141)
(0.051)
(0.062)
(0.053)
(0.139)
Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1. Controls not shown are the same as those listed in Table 4
Model 1. Each of the cells in Table 6 represents a separate regression.
When we break out the FLFPR by the education levels of women, it appears that this
relationship is driven primarily by the impact of higher FLFPRs of less educated women on less
educated men. This may be indicative of the importance of innate gendered skills, skills that are
not necessarily gained through higher levels of formal education. Census and ACS data suggest
26
that real wages have increased since 1980 only for college educated men (and declined for
others).
12
Thus, women have slowed the decline of real wages for men with a high school level
of education and below (this aligns with the conclusions of Juhn and Kim, 1999). Women also
seem to benefit when a higher percentage of women join them in the labor force. Higher FLFPRs
are associated with lower occupational segregation which leaves the secondary labor market
(likely mainly comprised of less educated workers both high school graduates and high school
dropouts) less crowded which aligns with the findings of Cotter et al. (1997). Women may also
be replacing less productive men, which would explain the higher wages for men that remain in
the market as well as lower participation rates of men over time (note by Juhn and Potter, 2006).
It may also explain the negative sign on the coefficient for MLFPR (Table 4); cities with higher
MLFPR may be cities where women are having a more difficult time replacing less productive
men. Finally, these results suggest that there may be diminishing marginal returns to increases in
FLFPRs and that the relationship between FLFPR and economic development (wage growth) in
the U.S. may be concave and not the convex U-shape previously suggested. The average FLFPR
for college graduates (in MSAs across 1980, 1990, 2000, and 2010) is over 71 percent whereas
the FLFPR for high school dropouts is only 33 percent.
13
This may also explain why the largest
impact of higher FLFPRs occurred in 1980 (Table 5). To test this hypothesis, FLFPR squared is
incorporated into Model 1 in Table 4. The results confirm that there is a concave relationship
between FLFPRs and wage growth.
14
Thus, low skill women and men receive the largest benefits
from the FLFPR of low skill women.
12
Based on calculations of real wages over time for men using the IPUMS Census and ACS data.
13
The average FLFPR for high school graduates is approximately 60%. Calculated using IPUMS Census and ACS
data.
14
The coefficient of FLFPR becomes 1.854*** (0.579) and the coefficient of FLFPR2 is -1.295** (0.533).
27
Considering Causality
Aside from demand side factors and larger economic trends contributing to the complex
relationship between economic development and women’s participation in the economy, there
may be other unobserved factors confounding these results. For example, forward looking
women may choose to enter the labor force in cities where they expect wages to grow based on
past wage growth (and not just the current wage level). Thus, women may be more likely to enter
the labor force in places like Boston and not places like Flint. Conversely, previous research has
shown that households smooth consumption in economic shocks through female labor force
participation (Fallon and Lucas, 2002). Thus, women may instead be more likely to enter the
labor force in areas like Flint where their husbands face layoffs and less likely in places like
Boston. To directly address this concern, the real wage growth from the previous decade is
incorporated into model 1 (Table 4). The impact of the FLFPR remains positive and significant
(at less than 1 percent significance level) with a 10 percent increase in the FLFPR associated
with an increase in wages of approximately 4 percent.
There may also be other unobserved factors specific to cities that are biasing results.
Within the metropolitan analysis, state fixed effects control for government policies (including
affirmative action policies and other progressive policies) associated with wage growth and
female employment share. State and year fixed effects can also be replaced with MSA fixed
effects. In this model, the impact of the FLFPR is still positive and significant (at the 0.2 percent
significance level) and suggests that a 10 percent increase in FLFPR is associated with a 4
percent increase in median wage growth. Finally, instrumental variables that affect women’s
attitudes and tastes for work and does not directly impact wage growth (or the demand for
working women).
28
Previous literature on supply (and demand side) factors that increase FLFPR provides a
host of potential instruments that are related to female employment share, but few are unrelated
to wage growth (other than through their relationship to FLFPR). Cities with higher divorce rates
will likely have a higher FLFPR (as shown in the first stage results in table 5), but we would not
expect divorce rates to be associated with higher wage growth nor should this increase the
demand for women. However, divorced women may not pay a marriage wage penalty. Similarly,
lower fertility levels are strongly associated with higher FLFPR, but are also directly related to
wage growth. The wage penalty for motherhood has been well-documented in the literature
(Anderson et al., 2003). Thus we rely on lagged and deeply lagged values of various factors
related to FLFPR over time but unrelated to current wage growth. Fogli and Veldkamp (2011)
argue that FLFPR is dependent on the information women learn about the impact of maternal
employment on children. Thus, cities with a higher initial FLFPR and a higher share of working
mothers will experience faster increases in FLFPRs.
The 10 year lagged value of FLFPR is first used as an instrumental variable and also the
1950 FLFPR. We expect that the 1950 level of FLFPR will impact future FLFPR through a role
model effect for young women as well as establishing a work environment earlier than other
areas that is welcoming to women. Additionally, Hellerstein and Morrill’s research suggests that
higher FLFPR in 1950 will incentivize father’s to encourage their daughters to enter the same
occupation. Lower fertility levels in 1950 increase the FLFPR in 1950 and increases the
likelihood that the daughters (and daughters-in-law) of these women will also plan to enter the
workforce (Fernández et al., 2004). The most dramatic change in FLFPR can be found in the
labor market decisions of married women. The 1950 share of married women that are working is
used to capture this cultural shift and instrument for recent FLFPR.
29
All of our 2SLS estimation suggest that the impact of FLFPR are positive and significant.
They also suggest that OLS estimates may be biased downward. The 2SLS estimates using all of
our deeply lagged instruments that impact women’s current tastes for work suggest that a 10
percnet increase in the FLFPR is associated with an increase in wage growth of over 8 percent.
TABLE 5: Instrumental Variables Estimation
Instrument
Instrument
Lag(FLFPR)
1950 FLFPR
1950 Fertility
OLS
0.425***
0.734***
0.734***
(0.103)
(0.142)
(0.142)
2SLS
0.777***
0.838***
1.275*
(0.253)
(0.360)
(0.695)
First Stage
0.281***
0.298***
-0.022***
(0.033)
(0.050)
(0.007)
1st Stage F-Stat
73
36
9
Hausman
0.123
0.755
0.410
Instrument
1950 Percent Married
Women Working
Divorce15
1950 FLFPR, Fertility, Married
Women Working
OLS
0.734***
0.476***
0.734***
(0.142)
(0.082)
(0.142)
2SLS
0.821***
0.516*
0.843***
(0.283)
(0.269)
(0.279)
First Stage
0.316***
0.648***
(0.038)
(0.094)
1st Stage F-Stat
69
48
24
Hausman
0.724
0.876
0.576
Overid
1950 Percent Married
Women Working
0.800
Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1.
6. CONCLUSION
Utilizing regional differences in FLFPR across metropolitan areas, the impact of women
on the local economy (rather than how various factors affect the labor market decisions of
women, as in previous literature) can be estimated. This research provides evidence similar to the
conclusions of cross-country studies; women’s participation in the economy increases economic
15
The divorce instrument was also used in a model with MSA fixed effects instead of state fixed effects. The first
stage estimates are positive and significant (with an F-stat of 20) and the 2SLS estimate for the impact of the FLFPR
in this model is 1.197** (0.485) compared to 0.382*** (0.121).
30
growth. From 1980 to 2010, a 10 percent increase in FLFPR is associated with a 5 percent
increase in wages (or every 1 percent increase in FLFPR is associated with 0.5 percent increase
in wages).
Table 6 (and 7) show the top 10 (and bottom 10) MSAs in terms of the 1980 FLFPR
(among the 100 most populous MSAs). Metropolitan areas with the highest FLFPR experienced
higher median wage growth from 1980 to 2010 than those with the lowest. The top 10 MSAs in
terms of female employment share also had lower industry and occupation segregation. Women
are choosing to work where their opportunities are more equal and their skills are more highly
valued, evidenced by the higher median wages for women and lower industry and occupation
segregation.
TABLE 6: Top 10 MSAs by Female Employment Share
Rank
Metropolitan Area
FLFPR
Change in
Median Wages
Men’s
Median
Wage
Women’s
Median
Wage
Industry
Segregation
Occupation
Segregation
1
Madison, WI
62.5%
$2.30
$14.72
$9.86
0.278
0.470
2
Washington, DC
61.5%
$3.93
$17.89
$12.14
0.197
0.482
3
Des Moines, IA
60.3%
$0.39
$15.89
$9.16
0.343
0.504
4
San Jose, CA
60.1%
$4.19
$18.41
$11.40
0.221
0.427
5
Denver, CO
59.7%
$1.12
$16.56
$9.94
0.256
0.469
6
Minneapolis, MN
59.5%
$1.65
$17.66
$9.95
0.290
0.481
7
Las Vegas, NV
59.4%
$0.21
$14.66
$9.60
0.246
0.464
8
Raleigh, NC
59.3%
$2.55
$13.20
$9.47
0.283
0.428
9
Columbia, SC
58.8%
$0.93
$11.09
$9.05
0.249
0.419
10
Charlotte, NC
58.8%
$2.20
$12.69
$8.69
0.239
0.435
Average
60.0%
$1.95
$15.28
$9.93
0.260
0.458
TABLE 7: Bottom10 MSAs by Female Employment Share
Rank
Metropolitan Area
FLFPR
Change in
Median Wages
Men’s
Median
Wage
Women's
Median
Wage
Industry
Segregation
Occupation
Segregation
91
Bakersfield, CA
45.4%
$0.11
$13.42
$8.55
0.367
0.580
92
Johnson City, TN
45.0%
-$0.45
$12.88
$8.40
0.339
0.449
93
Youngstown, OH
44.6%
-$3.70
$17.93
$9.32
0.382
0.508
94
Fort Lauderdale, FL
44.6%
$1.37
$13.25
$8.85
0.279
0.498
31
95
Mobile, AL
44.4%
$0.53
$13.25
$8.13
0.412
0.552
96
Scranton, PA
44.3%
$0.53
$13.24
$8.39
0.235
0.433
97
Beaumont, TX
44.0%
-$3.11
$19.06
$8.28
0.534
0.625
98
Tampa, FL
43.1%
$1.81
$11.58
$8.25
0.309
0.488
99
West Palm Beach, FL
42.3%
$2.22
$12.60
$8.36
0.322
0.499
100
Pittsburgh, PA
42.2%
-$1.49
$17.90
$9.38
0.386
0.548
Average
44.0%
-$0.22
$14.51
$8.59
0.357
0.518
When considered in a Roback (1982) context, the higher wages (shown here) and higher
housing prices (suggested in Johnson, 2014) associated with women’s involvement in the
workforce may further suggest that working women may be a type of productive amenity in their
cities. Women’s impact on the economy in the long run has been positive, but a breakout by
decade shows that the impact has decreased over the last three decades along with progress
slowing progress in decreasing the gender wage gap (and occupational and industry segregation).
This may be due to diminishing marginal returns from increase in FLFPRs. The diminishing
marginal returns may also be a result of slowing progress in closing the gender wage gap and
FLFPR growth. A recent paper by Bacolod (forthcoming) suggests that difference in the gender
wage gap across cities can be explained by difference in the returns to agglomeration by gender.
Although women seem to be increasing wages in cities, they benefit less from having skills that
are usually rewarded in agglomeration. Thus, it may be just as important today for women and
society as a whole to continue to reduce discrimination and fairly incorporate women into the
labor force. The U.S. may need to consider policies that help women (and men) freely enter the
labor market specifically related to managing both a career and family, e.g. affordable childcare,
healthcare, paid maternity (and paternity) leave, etc. The United States is the only industrialized
country without a paid maternity leave policy (Human Rights Watch, 2011) and only 5 states
provide income replacement for maternity leave (Sandberg, 2013). Women looking to some type
of childcare services to fill this gap, families may be sorely disappointed. Childcare costs rose
32
twice as fast as the median income for families with children from 2000 to 2010 (NACCRRA,
2010). Women, especially mothers, will find it difficult to continue to make advances if the costs
of working outside the home increase. If we replace FLFPR in Model 1 with the share of married
women working or the share of mothers working, the results are similarly positive and
significant with a 10 percent increase associated with a 4 percent and 3 percent increase in wage
growth, respectively. This paper provides evidence that gender equality in the opportunity to
participate in labor markets may increase economic growth as measured by growth in median
real wages. It should also be noted that policies that allow women to freely (and fairly) enter the
labor market may also be valuable in their own right regardless of the impact on economic
development.
REFERENCES
Acemoglu, Daron, David Card and David Lyle. 2004. Women, War, and Wages: The Effect of
Female Labor Supply on the Wage Structure at Midcentury.” Journal of Political
Economy, 112(3), 497-551.
Anderson, Deborah, Melissa Binder, and Kate Krause. 2003. The Motherhood Wage Penalty
Revisited: Experience, Heterogeneity, Work Effort, and Work-Schedule Flexibility.
Industrial and Labor Relations Review, 56(2), 273-294.
Bacolod, Marigee. Forthcoming. “Skills, the Gender Wage Gap, and Cities.” Journal of Regional
Science.
Bacolod, Marigee and Bernardo Blum. 2010. Two Sides of the Same Coin: U.S. Residual
Inequality and the Gender Gap. Journal of Human Resources, 45(1), 197-242.
Bacolod, Marigee, Bernardo Blum, and William Strange, 2009a. Skills in the City. Journal of
Urban Economics, 65(2), 136-153.
-2009b. Urban Interactions: Soft Skills versus Specialization. Journal of Economic
Geography, 9(2), 227-262.
- 2010. Elements of Skill: Traits, Intelligences, Education, and Agglomeration. Journal
of Regional Science, 50(1), 245-280.
Bartik, Timothy. 1991. “Who Benefits from State and Local Economic Development Policies?”,
W. E. Upjohn Institute for Employment Research, Kalamazoo, MI.
Borghans, Lex, Bas ter Weel, and Bruce Weinberg. 2006. People People: Social Capital and the
Labor-Market Outcomes of Underrepresented Groups. NBER Working Paper.
33
Bailey, Martha. 2006. More Power to the Pill: The Impact of Contraceptive Freedom on
Women’s Lifecycle Labor Supply. Quarterly Journal of Economics. 121, 289-320.
Becker, Gary. 1957. The Economics of Discrimination. University of Chicago Press, Chicago,
IL.
Bergmann, Barabara. 2005. The Economic Emergence of Women. 2nd Ed. Palgrave
Macmillan, New York.
Black, Dan, Natalia Kolesnikova, and Lowell Taylor. 2014. Why Do So Few Women Work in
New York (and so many in Minneapolis)? Labor Supply of Married Women across U.S.
Cities. Journal of Urban Economics.79, 59-71.
Blank, Rebecca. 1989. “The Role of Part-Time Work in Women’s Labor Market Choices over
Time. The American Economic Review. 79(2), 295-299.
Blau, Francine and Lawrence Kahn. 1997. Swimming Upstream: Trends in the Gender Wage
Differential in the 1980s. Journal of Labor Economics, 15(1), 1-42.
- 2006. The U.S. Gender Pay Gap in the 1990s: Slowing Convergence. Industrial
Labor Relations Review, 60(1), 45-66.
Clark, Roger, Thomas Ramsbey, and Emily Stier Adler. 1991. Culture, Gender, and Labor
Force Participation: A Cross-National Study. Gender & Society, 5(1), 47-66.
Cohen, Philip and Matt Huffman. 2003. Occupational Segregation and the Devaluation of
Women’s Work across U.S. Labor Markets. Social Forces. 81(3), 881-907.
-2007. Working for the Woman? Female Managers and the Gender Wage Gap.
American Sociological Review. 72(5), 681-704.
Connelly, Rachel. 1992. “The Effect of Child Care Costs on Married Women’s Labor Force
Participation.” The Review of Economics and Statistics 74(1), 83-90.
Costa, Dora and Matthew Kahn. 2000. Power Couples: Changes in the Locational Choice of the
College Educated 1940-1990. Quarterly Journal of Economics, 115(4), 1287-1315.
Cotter, David, JoAnn DeFiore, Joan Hermsen, Brenda Marsteller Kowalewski, and Reeve
Vanneman. 1997. All Women Benefit: The Macro-level Effect of Occupational
Integration on Gender Earning Inequality. American Sociological Review, 62, 714-734
- 1998. The Demand for Female Labor. American Journal of Sociology, 103(5), 1673-
1712.
Doepke, Matthias, Moshe Hazan, and Yishay Maoz. 2007. The Baby Boom and World War II:
A Macroeconomic Analysis.” NBER Working Paper No. 13707.
Dollar, David and Roberta Gatti. 1999. Gender Inequality, Income, and Growth: Are Good
Times Good for Women? The World Bank.
Duflo, Esther. 2012. “Women Empowerment and Economic Development.” Journal of
Economic Literature, 50(4), 1059-1071.
England, Paula. 1982. The Failure of Human Capital Theory to Explain Occupational Sex
Segregation. The Journal of Human Resources, 17(3), 358-370.
Esteve-Volart, Berta. 2004. Gender Discrimination and Growth: Theory and Evidence from
India. London School of Economics and Political Science. DEDPS Working Paper No.
42.
Fallon, Peter and R.obert Lucas. 2002. The Impact of Financial Crises on Labor Markets,
Household Incomes, and Poverty: A Review of Evidence. The World Bank Research
Observer. 17(1), 21-45.
34
Fernández, Raquel, Alessandra Fogli, and Claudia Olivetti. 2004. Mothers and sons: Preference
formation and female labor force dynamics. Quarterly Journal of Economics. 119,
1249-1299.
Fogli, Alessandara and Laura Veldkamp. 2011. “Nature or Nurture? Learning and the Geography
of Female Labor Force Participation.” Econometrica, 79(4), 1103-1138.
Fortin, Nicole. 2005. Greed, Altruism, and the Gender Wage Gap. Paper presented at the 2005
SOLE/EALE Meetings, San Francisco, CA.
Glaeser, Edward, Hedi Kallal, Jose Scheinkman, and Andrei Scleifer. 1992. Growth in Cities.
Journal of Political Economy. 100(6), 1126-1152.
Goldin, Claudia. 1995. The U-Shaped Labor Force Participation Function in Economic
Development and Economic History. T.Paul Schultz ed. Investment in Women’s Human
Capital Investment. Chicago, IL: University of Chicago Press, 61-90.
-1991. The Role of World War II in the Rise of Women’s Work.” American Economic
Review, 81(4), 741-756.
-2006. The Quiet Revolution That Transformed Women's Employment, Education, and
Family.” 2006 Ely Lecture, American Economic Association Meetings, Boston MA (Jan.
2006), published in American Economic Review, Papers and Proceedings 96 (May
2006), 1-21.
Goldin, Claudia and Lawrence Katz. 2002. The Power of the Pill: Oral Contraceptives and
Women’s Career and Marriage Decisions. Journal of Political Economy.110(4), 730-
770.
Goldin, Claudia and Claudia Olivetti. 2013. Shocking Labor Supply: A Reassessment of the
Role of World War II on Women’s Labor Supply.” American Economic Review: Papers
and Proceedings, 103(3), 257-262.
Grove, Wayne, Andrew Hussey, and Michael Jetter. 2011. The Gender Pay Gap Beyond
Human Capital: Heterogeneity in Noncognitive Skills and in Labor Market Tastes.
Journal of Human Resources, 46(4).
Heckman, James and Robert Willis. 1977. A Beta-logistic Model for the Analysis of Sequential
Labor Force Participation by Married Women. Journal of Political Economy. 85(1), 27-
58.
Hellerstein, Judith, David Neumark, and Kenneth Troske. 2002. “Market Forces and Sex
Discrimination. Journal of Human Resources, 37(2), 353-380.
Hellerstein, Judith. and Melinda Sandler Morrill. 2011. Dads and Daughters: The Changing
Impact of Fathers on Women’s Occupational Choices.” Journal of Human Resources,
46(2), 333-372.
Human Rights Watch. 2011. Failing its Families: Lack of Paid Leave and Work-Family Supports
in the U.S. http://www.hrw.org/sites/default/files/reports/us0211webwcover.pdf
Institute for Women’s Policy Research. 2007. Women and Paid Sick Days: Crucial for Family
Well-Being. http://www.iwpr.org/publications/pubs/women-and-paid-sick-days-crucial-
for-family-well-being/at_download/file
Johnson, William. 2014. “House Prices and Female Labor Force Participation. Journal of
Urban Economics, 82, 1-11.
Juhn, Chinhui and Dae Il Kim. 1999. The Effects of Rising Female Labor Supply on Male
Wages. Journal of Labor Economics, 17(1), 23-48.
Juhn, Chinhui and Simon Potter. 2006. “Changes in Labor Force Participation in the United
States Journal of Economic Perspectives, 20(3), 27-46.
35
Klasen, Stephan, 2000. Does Gender Inequality Reduce Growth and Development? Evidence
from Cross-Country Regressions. The World Bank, Policy Research Report on Gender
and Development Working Paper Series, No. 2.
-2002. Low Schooling for Girls, Slower Growth for All? Cross-Country Evidence on the
Effect of Gender Inequality in Education on Economic Development.” The World Bank
Economic Review, 16(3), 345-373.
Klasen, Stephan and Francesca Lamanna. 2009. The Impact of Gender Inequality in Education
and Employment on Economic Growth: New Evidence for a Panel of Countries.
Feminist Economics, 15, 91-132.
Knowles, Stephen, Paula Lorgelly, and P. Dorian Owen. 2002. Are Educational Gender Gaps a
Brake on Economic Development? Some Cross-country Empirical Evidence. Oxford
Economic Papers.
Lagerlof, Nils-Peter. 2003. Gender Equality and Long-Run Growth. Journal of Economic
Growth, 8, 403-426.
Lombard, Karen V. 1999. “Women’s Rising Market Opportunities and Increased Labor Force
Participation. Economic Inquiry, 37(2), 195-212.
Mammen, Kristin and Christina Paxson. 2000. “Women’s Work and Economic Development.
Journal of Economic Perspectives. 14(4), 141-164.
Mincer, Jacob. 1962. “Labor Force Participation of Married Women: A Study of Labor Supply.”
Aspects of Labor of Economics. National Bureau of Economic Research, 63-106.
Morrison, Andrew, Dhushyanth Raju, and Nistha Sincha. 2007. “Gender Equality, Poverty, and
Economic Growth.” The World Bank. Gender and Development Group and Poverty
Reduction and Economic Management Network. Policy Working Paper 4349.
National Association of Child Care Resource & Referral Agencies (NACCRRA), 2010. “Parents
and the High Cost of Child Care.” http://eyeonkids.ca/docs/files/cost_report_073010-
final.pdf
Odland, John and Mark Ellis. 1998. “Variations in the Labour Force Experience of Women
Across Large Metropolitan Areas in the United States.” Regional Studies, 32(4), 333-347.
Patrick, Carlianne, Heather Stephens, and Amanda Weinstein. 2016. “Where are all the Self-
Employed Women? Push and Pull Factors Influencing Female Labor Market Decisions.”
Small Business Economics, 46, 365-390.
Peck, Jamie. 1996. Work-Place: The Social Regulation of Labour Markets. Guilford Press,
London.
Phimister, Euan. 2005. Urban Effects on Participation and Wages: Are There Gender
Differences? Journal of Urban Economics, 58, 513-536.
Psacharopoulos, George. and Zafiris Tzannatos. 1989. Female Labor Force Participation: An
International Perspective. The World Bank Observer, 4(2), 187-201.
Roback, Jennifer. 1982. Wages, Rents and the Quality of Life. Journal of Political Economy
90(6),1257-1278.
Ruggles, Steven, Katie Genadek, Ronald Goeken, and Matthew Sobek. 2010. Integrated Public
Use Microdata Series: Version 5.0 [Machine-readable database]. Minneapolis: University
of Minnesota.
Sandberg, Sheryl with Nell Scovell. 2013. Lean In: Women, Work, and the Will to Lead. Alfred
A. Knopf.: New York.
36
Sapienza, Paola, Luigi Zingales, and Dario Maestripieri. 2009. Gender Differences in Financial
Risk Aversion and Career Choices are affected by Testosterone.Proceedings of the
National Academy of Sciences of the United States of America, 106(36), 15268-15273.
Semyonov, Moshe. 1980. “The Social Context of Women’s Labor Force Participation: A
Comparative Analysis. American Journal of Sociology, 86(3), 534-550.
Smith, James and Michael Ward. 1985. Time-Series Growth in the Female Labor Force.
Journal of Labor Economics. 3(1), 59-90.
Timothy, Darren and William Wheaton. 2001. Intra-Urban Wage Variation, Employment
Location, and Commuting Times. Journal of Urban Economics, 50(2), 338-366.
U.S. Bureau of Labor Statistics, 2014. Women in the Labor Force: A Databook. BLS Reports.
http://www.bls.gov/cps/wlf-databook-2013.pdf
Waite, Linda and Ross Stolzenberg. 1976. Intended Childbearing and Labor Force Participation
of Young Women: Insights from Non- Recursive Models. American Sociological
Review, 41, 235-252.
Weinberg, Bruce. 2000. Computer Use and the Demand for Female Workers. Industrial and
Labor Relations Review, 53(2), 290-308.
... These findings imply a lesser impact of increasing labor force participation on the gender wage gap. They and the related literature could be concealing the total effect of increased labor force participation of women who bring complementary skill sets to the labor market, increasing overall productivity [Weinstein (2017)]. We believe this issue in the Mexican labor market deserves further inquiry. ...
... t and a m t , or other determinants of wages, such as bargaining power. This could not be true if women bring complementary skill sets to the labor market, increasing overall productivity, and if their greater labor force participation changes their bargaining power[Weinstein (2017)]. ...
Article
This article analyzes the effect of international migration on the wage gap between women and men who remain in Mexico. We use historical distance to the U.S. border over early twentieth-century railroad networks as an exogenous factor causing changes in the relative supply of men and women, due to predominately male migration. A 10% decrease in the relative labor supply of men tends to increase the wage gap between women and men by approximately 1.1 percentage points, suggesting that they are not perfect substitutes. However, the results imply a greater elasticity of substitution between men and women than that suggested by previous studies.
... Third, differences in labor force participation rates are not strongly correlated with differences in quality of life (0.10) and the direction of the correlation suggests that women prefer cities with higher labor force participation rates. That is important since research has shown that female labor force participation is correlated with higher female wages (Weinstein, 2017), which holding housing costs constant would tend to lower estimated quality of life. So in this case, higher female labor force participation is associated with higher female wage effects and even larger positive housing effects. ...
Article
The quality of life literature revealed the role that natural resources and local consumption amenities play in household preferences for cities and their growing importance in determining the success of cities. As women's control over household resources has increased, a growing body of consumer behavior literature has been documenting gender differences in preferences for consumer goods. Yet, the quality of life literature does not seem to make any distinction along gender lines in determining preferences for local consumption amenities. We estimate the quality of life across U.S. cities separately for unmarried men and women, and using a rich set of location‐specific amenities, examine which amenities cause preferences to differ between men and women. Our results suggest that men and women's valuations of cities are correlated, suggesting commonality in preferences for many location‐based amenities, most notably natural resources, but there are also important deviations in valuations especially for local public goods and gender role attitudes. This article is protected by copyright. All rights reserved.
Article
There is a growing amount of literature in economic geography showing that historical episodes can leave long‐lasting cultural and institutional legacies across space. For credibly identifying such persistent effects the analyses should not pick up trends preceding the respective episodes. Against this background, the paper re‐examines the famous case of the German division and re‐unification. The empirical focus is on the persistent mark‐up of women in work in East relative to West German regions that is often associated with legacy effects of the socialist regime that was in place in East Germany during the country’s four decades of division. In contrast to the conventional wisdom in academia, policy, and the public, the current paper shows that the higher share of working women in East German regions is not due to a legacy of socialism. Female labour force participation was already remarkably higher in the East before the introduction of socialism. The general lesson is that any attempt to explain spatial variation in individual decision‐making by persisting institutional and cultural legacies of certain historical episodes needs to assess regional conditions pre‐dating these episodes. This article is protected by copyright. All rights reserved.
Article
Using tools rooted in welfare economics, this paper explores the social welfare loss that arises from occupational segregation by gender and race in the U.S. at the regional level. After controlling for characteristics, the losses are lower in the Northeast than in the South and West according to a wide range of indicators, including those that take into account the relative size of disadvantaged groups (incidence), the magnitude of their losses (intensity), and the inequality among those groups. The West has the highest (conditional) losses, although the intensity of the phenomenon barely differs from that in the South or Midwest.
Article
Men are more likely than women to commit tax offenses, and previous research has suggested that this difference can be partially explained by status processes. This study investigates whether the relationship between status and tax offenses has changed over time in synchrony with changes in social stratification based on gender, race, and class. We hypothesize that if gender equality has increased within U.S. society (particularly within the formal economy), then gender-based discrepancies in tax offending should have decreased as well. Using administrative data from federal criminal cases sentenced between 1995–2015, we examine how the effects of different indicators of status – including race, class, and gender – on the odds of being convicted for a tax offense have changed over time. Results from a series of logistic regression models indicate that over time the difference in the probability of men and women being convicted of a tax offense versus another white collar crime decreases. Other variables of status including race, age, and education are also analyzed and discussed.
Article
The present study is a general review of the published papers on women and the city with the aim of identifying knowledge gaps in the research. In order to achieve this goal, 275 papers were selected from 43 journals drawn from four data bases: Taylor and Francis; Wiley Online Library; Science Direct (Elsevier); and SAGE Libraries. The results of the paper suggest the United States of America and the United Kingdom have the highest number of papers in relation to women and the city, and in considering the period 1977–2019, the highest number of papers were published in the past 10 years. The knowledge gaps and debates which left behind are presented in five discussion documents which can be used for future research. These five documents include debates on the health and safety of women in cities; economic studies representing labour, migrants and poverty; social justice debates, and the road to the right to the city. The fifth document, at the end of this paper, presents a typology based on three aspects: theories of women in urban studies, ‘the theoretical debate’; the designer's efforts create urban spaces in order to achieve urban inclusivity, ‘the form-of-city debate’; and the inclusion of women in future urban development planning processes together with emerging gender planning theory, ‘the planning process debate’.
Article
Gender inequality is persistent in the labor market despite equality between genders being one of the fundamental principles of the EU law. This is a reason why we have decided to analyze gender inequality in unemployment rates of the 27 European Union countries. Unemployment gender inequality differs across EU countries. There are countries where gender gap is persistent, but there are also countries where gender gap changed during the observed period of 2005Q1–2017Q2. Moreover, in some countries gender inequality decreased, while in others it was steady. Unemployment rates differ across countries similarly as unemployment gender inequality does, therefore, by choosing these countries, we have obtained a portfolio of distinct behaviors throughout the time. We have also obtained the relations between them. We test the hypothesis of the hysteresis to check the relation between the unemployment gender inequality and the unemployment rate levels, gender disadvantaged through unemployment gender inequality and the unemployment rate levels. This hypothesis was tested for both univariate and panel data country series. We could not reject the hypothesis of the hysteresis in univariate country series using a battery of GLS unit root tests proposed by Carrión-i-Silvestre et al. (Econ Theory 25:1754–1792, 2009). However, ILT test of Im et al. (Oxf Bull Econ Stat 67:393–419, 2005) rejected the hypothesis of the hysteresis for panel data of the European Union countries. Finally, the relationship between unemployment gender inequality and disadvantaged gender in inequality depends on the country analyzed. This result suggests that further analysis is needed to identify what causes unemployment gender inequality and what should be done to decrease it.
Article
Full-text available
This paper links gender wage gaps with the urban wage premium. First, the study documents gender wage gaps are narrower in larger U.S. metropolitan areas in 2000 and 2010. Skill agglomerations are then considered to explain this. Specifically, if men and women employ heterogeneous skills, and these skills have differential productivities across city sizes, agglomerative forces may differentially reward men and women. Occupational data show that women are concentrated in jobs relatively more intensive in interactive and cognitive skills, while men are comparatively in physical skill-intensive jobs. Decomposing the gender wage gap shows that explanatory factors (education, skills, and location) predict women would outearn men. Instead, agglomerative skill returns account for majority of the gap. These estimates suggest that even as women employ skills rewarded in agglomeration economies, they benefit less from agglomerations than men, resulting in the observed gap.
Article
Full-text available
Previous research focuses on factors that influence self-employment participation, in part because entrepreneurship has been associated with economic growth. This literature has tended to focus only on men or the comparison of women to men, while ignoring substantial heterogeneity in employment decisions among women. By investigating the impact of individual, household, and local economic and cultural characteristics on the labor market outcomes of different groups of women, we get a more comprehensive picture of their self-employment decision. Recognizing self-employment as one of multiple labor market choices, we use multinomial logit and two confidential, geocoded micro-level datasets to study women`s career choices in urban areas. We find that the effects of various push and pull factors differ between married and unmarried women. In particular, more progressive gender attitudes pull married women into self-employment, while household burdens associated with children push them into self-employment. For unmarried women, the local business climate and individual characteristics have the strongest influence. In both cases, the motivations for women are quite different than men.
Book
This new edition of a classic feminist book explains how one of the great historical revolutions - the ongoing movement toward equality between the sexes - has come about. Its origins are to be found, not in changing ideas, but in the economic developments that have made women's labour too valuable to be spent exclusively in domestic pursuits. The revolution is unfinished; new arrangements are needed to fight still-prevalent discrimination in the workplace, to achieve a more just sharing of housework and childcare between women and men, and, with the weakening of the institution of marriage, to re-erect a firm economic basis for the raising of children.
Article
The release of Enovid in 1960, the first birth control pill, afforded U. S. women unprecedented freedom to plan childbearing and their careers. This paper uses plausibly exogenous variation in state consent laws to evaluate the causal impact of the pill on the timing of first births and extent and intensity of women's labor-force participation. The results suggest that legal access to the pill before age 21 significantly reduced the likelihood of a first birth before age 22, increased the number of women in the paid labor force, and raised the number of annual hours worked.
Article
Is there a causal connection between house prices and labor force participation of married women ? The simple correlation between house prices and married women’s labor force participation across U.S. metro areas is positive. Plausible, informal arguments have been advanced to support causation in either direction: prices raising participation (negative income effects of higher house prices lead more married women to work) or participation raising prices (richer two-earner households bid up the price of scarce housing). I construct an equilibrium model of location, labor supply and real estate (land) prices within a metro area which predicts that 1) metro areas with exogenously less buildable land will have higher house prices and more labor force participation of married women, while 2) metro areas with married women exogenously more prone to work will have higher house prices. Using geographic instruments for housing supply, I find little evidence of a positive effect of house prices on married women’s labor force participation, but a somewhat greater possibility that house prices raise their earnings. Likewise, an instrument for married women’s labor supply reveals no consistent significant causal effect of two earner households on housing prices, although the possibility of a positive effect cannot be ruled out.
Article
This report assesses the impact of culture on women's share of the labor force. Measuring both economic factors and cultural milieu, we found that culture was related not only to levels of women's share of the labor force but, in some instances, to changes in those levels. A secondary finding of the study was that the economic development of a nation had a strong positive association with increases in women's share of the labor force and that one measure of dependency (commodity concentration) had a strong negative association with such change.
Article
Abstract Using two single-cohort longitudinal surveys, the NLS72 and the NELS88, I investigate the impact of greed, ambition, leadership, altruism, gender role attitudes, family values and in- come expectations on the gender wage gap. Appealing to a hedonic wage function framework, I find that in the NLS72, gender dierences in these values and linked behaviors account for a share of the gender wage gap that exceeds (is comparable) in 1979 (in 1986) the share at- tributable to gender dierences in labor market experience and job tenure. In the NELS88, gender dierences in work values have closed substantially. Thus in 2000, gender dierences in income expectations, which are significantly linked to greed, become the most important soft explanatory factor of the gender gap in pay. Methodologically, this paper proposes a correction to the Oaxaca-Blinder-Ransom decomposition that results in a truly decomposable approach compatible with the simple pooled regression that includes a gender dummy. Keywords: Altruism, income expectations, gender wage gap, regression-compatible Oaxaca-