This website uses cookies.

By clicking the "Accept" button or continuing to browse our site, you agree to first-party and session-only cookies being stored on your device to enhance site navigation and analyze site performance and traffic. For more information on our use of cookies, please see our Privacy Policy.

  1. Home
  2. About the AEA
  3. AEA Committees
  4. Committee on the Status of Women in the Economics Profession
  5. CSWEP 2015 Statistical Report on the Status of Women in the Economics Profession

CSWEP 2015 Statistical Report on the Status of Women in the Economics Profession

By Margaret Levenstein, Associate Chair and Director of the Survey

A. Women’s Status in the Economics Profession

In 1971 the AEA established CSWEP as a standing committee to monitor the status and promote the advancement of women in the economics profession. In 1972 CSWEP undertook a broad survey of economics departments and found that women represented 7.6% of new PhDs, 8.8% of assistant, 3.7% of associate and 2.4% of full professors. Much has changed. At doctoral institutions, women have more than quadrupled their representation amongst new PhDs to 34.7%, tripled their representation amongst assistant professors to 28.2%, increased their representation at the associate level more than six fold to 23.5% and increased their representation at the full professor level five-fold to 12.2%. This report presents the results of the 2015 survey, with emphasis on changes over the last two decades, including entry of women into PhD programs and the progress of cohorts of new PhDs as they progressed through the academic ranks.[i]

1. The CSWEP Annual Surveys, 1972-2015

In fall 2015 CSWEP surveyed 124 doctoral departments and 126 non-doctoral departments.[ii] Of these, all 124 doctoral and 117 non-doctoral departments responded, yielding response rates of 100% and 87%, respectively. This report includes harvested faculty data from the Web for the non-responders. The non-doctoral sample is based on the listing of “Baccalaureate Colleges – Liberal Arts” from the Carnegie Classification of Institutions of Higher Learning (2000 Edition). Starting in 2006 the survey was augmented to include six departments in research universities that offer a Master’s degree but not a PhD degree in economics. We are in the process of harmonizing and documenting the departmental-level data from the 1970s to the current period to improve our analysis of long-run trends in the profession.

 

2. 2015 Results

This overview begins with an oft-neglected group, teaching faculty outside of the tenure track. These faculty typically hold multiyear rolling contracts and carry titles such as adjunct, instructor, lecturer, visitor or professor of the practice. As seen in Table 1, in doctoral departments, the representation of women in these positions runs high, currently standing at 36.8%, exceeding that not just of assistant professors but even that of new PhDs. In 2015 the share of non-tenure track women was almost twice their share of all tenure track positions combined (19%), and this disparity is greater still in the top 20 departments (Table 2).

 

With regard to doctoral departments, the representation of women at each level of the academic hierarchy has increased since the 1970s. However, progress has slowed during the last two decades. Since 1997, there has been only a very small increase in the proportion of assistant professors who are women (28.2% in 2015 versus 26.0% in 1997). The representation of women amongst first-year PhD students has not increased at all, standing at 31.6% in 2015 versus 31.3% in 1997. During the last decade the share of first-year students who are women averaged 32.2%, a slight decline from the previous decade’s 33.7%. This was the case despite an increase in the share of baccalaureate degrees going to women. The increased entry of women into the profession during the late 20th century led to increasing representation of women in higher ranks, with women now making up almost one-quarter of tenured associate professors and just over 12% of full professors.

At every level of the academic hierarchy, from entering PhD student to full professor, women have been and remain a minority. Moreover, within the tenure track, from new PhD to full professor, the higher the rank, the lower the representation of women (Figure 1). In 2015 new doctorates were 34.7% female, falling to 28.2% for assistant professors, to 23.5% for tenured associate professors and to 12.2% for full professors. This pattern has been characterized as the “leaky pipeline.” Our reliance on this leaky pipeline for gradual progress in women’s representation in the profession depends on continued growth in entry, which no longer appears to be forthcoming.

Because the growth in women’s representation has differed across ranks, the gaps in representation between adjacent ranks have changed. Thus, following some convergence of women’s representation at the associate level to that at the assistant level around the turn of the century, convergence seems to have ceased. The gap between women’s representation at the full and associate levels is much higher than it was in the 1990s. It is worth noting that the latter is not necessarily an unwanted development. It is the result of relatively good growth in women’s representation at the associate level as compared with the full level, where women’s representation changes only slowly as the stock of full professors at any given time reflects something like a 25-year history of promotions from associate to full.

Turning to a comparison of non-doctoral with doctoral departments, at every level in the tenure track, women’s representation in non-doctoral departments runs higher – over 10 percentage points higher – than in doctoral departments (compare Tables 5 and 6). Similar to the trend in doctoral departments, women’s representation has mildly trended up at the assistant professor level and somewhat more so at the full level. Deserving of attention, the non-doctoral departments do not share the strong upward trend at the associate level exhibited by doctoral departments. Among non-doctoral departments the trend in women’s representation at the associate level seems fairly flat over the past 12 years at a little over one-third (Figure 2).

 

A further comparison by rank shows that the representation of women declines as the emphasis on research increases, averaging 39% for (full-time) non-tenure track teaching positions in non-doctoral departments, 36.8% of non-tenure track teaching positions in doctoral departments, 33.5% of all tenure track positions in non-doctoral departments, 19% in all doctoral departments, 14.3% in the top 20 departments and 13.6% in the top 10 departments. This represents a remarkable decline in women’s representation as departmental research intensity increases. The share of new PhDs going to research-intensive (doctoral) departments who are women has increased since the 1990s (Table 3), but women are still over represented in non-academic (especially private sector) placements (Table 4).

With regard to the advance of cohorts of academics through the ranks, this report presents a simple lock-step model of these advances (Figures 3 and 4).

With a maximum of 41 years of data on each rank we can track the gender composition of some relatively young cohorts from entering graduate school though the PhD and of other older cohorts from receipt of the degree though the assistant and associate professor ranks. Unfortunately, these data do not suffice to analyze the advance of cohorts from associate to full professor. Over the last decade, the proportion of women receiving their PhDs has been almost exactly the same as the proportion of women entering PhD programs six years prior. Women are, if anything, more likely to graduate in five years than their male co-matriculates.

 

There is evidence of attrition from academia after graduate school, however, as women’s share of new assistant professors is on average about 5% less than their share of new PhDs (Figure 3). Women’s disproportionate exit from traditional academic jobs has, if anything, increased in the last decade (examining those who entered PhD programs in 1997-2003).

The female share of the entering class of students in PhD programs overall has been steady, at between 31 and 35%, over the last 20 years (Figure 1). The female share in top 20-programs, however, has fluctuated in ways that raise concern (Table 7).

Between 1997 and 2001, the average female share in these programs was about 30%. During the period 2002-2006 this stays roughly constant, suggesting continued integration of women into economics and a flow into the pipeline. During 2007-2011, the average fell to 27.3%. This could easily have reflected small numbers and not a trend. However, the average female share in top-20 programs has remained at or below this level during 2012-2015and fell below 25% in 2015. There is considerable variation in the share of females in the first PhD class across the 21 schools in the top 20 (Table 8). Note that while we are not breaking out the top 10, to protect the confidentiality of individual school data, the pattern is not different across the top 10 and the schools ranked 11-20.

 

3. Conclusions

Past intakes and subsequent advancements of women and men determine the contemporaneous distribution of men and women on the academic economists’ ladder. This report points to two critical junctures: the failure to grow of the representation of women at the intake; and, relative to men, the subsequent poorer chance of advancing from untenured assistant to tenured associate professor. With regard to the first, in the face of the growing representation of women at the baccalaureate level, the stagnation of the share of women in entering PhD classes means that entering PhD students represent a declining fraction of new baccalaureate women. This latter decline is no doubt rooted in the analogous decline in the fraction of women undergraduates who major in economics and may in part stem from the way we teach economics at the undergraduate level, as stressed by Goldin (CSWEP Newsletter, Spring/Summer 2013). This is an issue for both doctoral and non-doctoral departments (see Tables 5 and 6).

With regard to the second juncture, the advancement of women from untenured assistant to tenured associate professor is no doubt intertwined and jointly determined with family-related decisions. Here, the institutional setting (length of the tenure clock, gender-neutral family leave, on-site child care and so forth) can play significant roles. These policies are generally the same across academic disciplines, so they cannot explain the relative lack of progress for women in economics when compared with other disciplines.

Finally, it is worth recognizing the high representation of women in non-tenure track teaching jobs. Fully one-third of the full-time female faculty in top-20 economics departments are in non-tenure track positions.

In closing out this summary, it is worth noting that the 44 years of data on the evolution of faculty composition at the department level are unique in the social sciences and beyond. It is time to steward these data in a way that meets professional standards, to put in place a system for maintenance for future years and to make the descriptive statistics at group levels (e.g., doctoral, non-doctoral and others) available online. We also recommend making departmental-level data available for research purposes in a manner that protects the confidentiality of the responding departments. The new Associate Chair and Director of the Survey has begun the process of identifying and documenting the extant data so that it can be properly archived and shared.

 

[i] Survey respondents include all 124 PhD-granting economics departments in the United States and 117 economics departments without PhD programs. Nine non-PhD programs failed to respond to the survey; information on the composition of the faculty at those institutions (Earlham, Eastern Mennonite, Mills, Nebraska Wesleyan, the New College of Florida, Oglethorpe, Roanoke, Spelman and Westmont) was culled from their websites. No information on student composition is available for those schools.

[ii] The 2015 survey pool for doctoral departments remained the same as in 2013 and 2014. The 126 non-doctoral departments surveyed are the same as those surveyed in 2014 with the addition of a recently identified undergraduate department that had been conflated with a PhD department in the same university (but different college) in previous years. 21 departments composed mainly of business faculty were dropped from the 2014 survey of non-doctoral departments and continue to be excluded in this year’s survey.