A couple of weeks ago, just after International Women’s Day, we had a coffee hour in the World Bank’s Ankara office to watch short videos of five women that have recently started up their own business and transformed their lives and that of their families (stay tuned for the release of these films, currently being edited). Their stories were uplifting, but the discussion quickly turned to the dismal field of statistics. Commentators stressed that female labor force participation in Turkey remains at only half of the OECD level, that Turkey loses around 25 percent of its potential GDP because of this, and lamented that social norms and mixed political messages on the role of women in society were preventing greater progress towards gender equality.
So far and so familiar to those working on gender issues in Turkey; and yet, two pieces of evidence have recently caught my attention and left me hopeful that we will see a growing proportion of Turkish women at work in the years ahead.
The first piece of evidence comes from a closer look at Turkey’s own labor statistics. The Turkish economist Insan Tunali has looked at the labor force participation rate of urban women in different cohorts , starting with those born in 1953 and ending with those born 30 years later. The striking finding, captured in Figure 1, is that participation rates have shifted upwards by around 20 percent between the 1963 and 1983 cohorts. As more and more families live in urban areas, the rise in labor force participation rates among young, urban women compensates for the decline of working women in agriculture. As a result, female labor force participation rates have been shifting upward since the mid-2000s by around 6 percentage points and among women in prime working age 25-55 even by 10 percentage points.
The second piece of evidence comes from research by Erik Meyersson published in the January edition of Econometrica . Dr. Meyersson examines the evolution of secondary education enrolment rates since the mid-1990s at the level of Turkish municipalities. Dr. Meyersson’s interest is in the impact of Islamic rule on female school enrolment. The article uses Regression Discontinuity analysis to compare municipalities where candidates from the Islamic Refah party narrowly won in the 1994 municipal election with municipalities where they narrowly lost, and establish a causal link with secondary school enrolment rates for girls. The results establish a robust causal and positive relationship between Islamic rule and female secondary school enrolment (equivalent to a 3 percentage point increase on average), which remains positive even after 17 years. Dr. Meyersson’s interpretation is that Islamic rule reduced barriers to female participation among the poorer and pious segments of society by making them more comfortable about sending their girls to school.
Combining both pieces of evidence, we may start to see a pattern whereby access barriers – whether socioeconomic or social-norm related – to education among girls are being eroded and this in turn leads to growing aspirations among women to enter the labor force. Indeed, among women who make it through higher education, labor force participation rates increase to 62 percent. Net enrolment rates in tertiary education are close to 35 percent for both men and women, up from just 20 percent just 6 years ago. So here, too, the trend is positive.
Does this mean that Turkey will inevitably follow the typical U-shaped pattern of female labor force participation during its transition from middle to high income (e.g. Goldin, 1994 ; Mammen and Paxson, 2000 )? I believe this is quite likely. But that does not mean there is no room for policy to shift the entire U-curve upwards or bend the upward sloping part to make it steeper. Interventions such as improved child care facilities, more flexible work arrangements, including support for part-time work, balanced parental leave provisions (for both men and women) and continued public advocacy are all needed to close the gap that still separates Turkey from its OECD peers.
A particularly interesting question is whether social norms are indeed evolving towards greater gender equality in work life. Figure 2 reports results from several waves of the World Value Survey. The picture is mixed. While the share of respondents who believe men should have priority when jobs are scarce has been declining over the past decade, it is still more than 50 percent (and above levels in the 1990s). This matters, because Turkey’s significant improvement in female labor force participation since the mid-2000s came at a time of record employment growth , with over 4 million jobs created since 2009 alone whereas the late 1990s were a period of economic crisis and poor job market performance. As Turkey enters a period of moderate growth, the key challenge will be to make sure women are not the first ones to suffer from a less buoyant labor market. That may be the real test of whether my optimism in this blog is justified.