New Economist

More about menstruation and the gender earnings gap

After my recent post, Does menstruation explain the gender pay gap?, some more research on a similar theme. Andrea Ichino from the European University Institute and Enrico Moretti at the University  of California, Berkeley have a new NBER working paper, Biological Gender Differences, Absenteeism and the Earning Gap, which argues that women's higher rate of absenteeism due to menstruation reduces their earnings and accounts for almost 12% of the gender earnings gap:

In most Western countries illness-related absenteeism is higher among female workers than among male workers. Using the personnel dataset of a large Italian bank, we show that the probability of an absence due to illness increases for females, relative to males, approximately 28 days after a previous illness. This difference disappears for workers age 45 or older. We interpret this as evidence that the menstrual cycle raises female absenteeism. Absences with a 28-day cycle explain a significant fraction of the male-female absenteeism gap.

To investigate the effect of absenteeism on earnings, we use a simple signaling model in which employers cannot directly observe workers' productivity, and therefore use observable characteristics – including absenteeism – to set wages. Since men are absent from work because of health and shirking reasons, while women face an additional exogenous source of health shocks due to menstruation, the signal extraction based on absenteeism is more informative about shirking for males than for females. Consistent with the predictions of the model, we find that the relationship between earnings and absenteeism is more negative for males than for females. Furthermore, this difference declines with seniority, as employers learn more about their workers' true productivity. Finally, we calculate the earnings cost for women associated with menstruation. We find that higher absenteeism induced by the 28-day cycle explains 11.8 percent of the earnings gender differential.

For those without access to NBER publications, you can find a version here (PDF).

It's an interesting thesis and a careful piece of analysis. But clearly, like the earlier paper I cited, it's only one part of the gender earnings gap picture. I am not a biological reductionist, and very much doubt that menstruation can explain most of the gender gap. On my somewhat cursory reading of the quite vast literature, the two main drivers of the gender divide in most countries are:

• The effect of time off work from maternity leave and career breaks, with the gender gap being greater for women who take longer off work, or who have more than one child and hence multiple breaks.
• The effect of mothers not returning to the same employer and same job (in many cases, due to switching from full-time to part-time employment)  - in many countries, effectively trading away career prospects and training for greater working time convenience and flexibility (though not, apparently, in the Netherlands).

These two factors are likely to be be more important than absenteeism or bargaining/risk aversion.

Hat tip: Adam O'Neill at The Lowest Deep

UPDATE: Tyler Cowen at Marginal Revolution links to this post, and in a subsequent addendum he also links to the paper by Yan Chen, et al, Why Can't a Woman Bid More Like a Man? (PDF), which was the subject of my first post on this topic. For my thoughts on it, please refer to that post.

Does menstruation explain the gender pay gap?

Yan Chen, Peter Katuscak and Emre Ozdenoren ask an interesting question: Why Can't a Woman Bid More Like a Man? (PDF). No, this paper is not about My Fair Lady (or Pygmalion, for that matter). It's about menstruation and the gender pay gap.

There is a small but growing body of evidence that women in the workforce are, on average, more risk averse and less likely to bargain (or to ask for pay rises) than men. See for example the book Women don't ask: Negotiation and the gender divide by Linda Babcock and Sara Laschaver, and the paper by Deborah Small, et al: Who goes to the bargaining table? Understanding gender variation in the initiation of negotiations (PDF). Might there be a biological explanation for this? The paper by Chen, etal, suggests there is. Here is the abstract:

We find systematic evidence that demographic characteristics, especially gender, race and the number of siblings, education backgrounds, as well as menstrual cycle, significantly affect bidder behavior in the first and second-price sealed-bid auction in the laboratory. In particular, we find that women bid significantly higher than men in the first-price auction, while the likelihood of dominant strategy play in the secondprice auction is not different between men and women.

This finding provides support for the hypothesis that risk attitude rather than cognitive ability is the main driving force for the gender gap in competitive environments. At a biological level, we find that, in the first-price auction, during menstruation, when levels of estrogen and progesterone are the lowest, women do not bid differently from men. The gender difference in the first-price auction is driven by women during other phases of the menstrual cycle with higher levels of estrogen and progesterone.

One can expect howls of protest from some quarters when these findings become more widely known. But even if women's risk aversion and bargaining account for some part of the gender earnings gap, no one would argue that they account for all of it.

I expect cross-national surveys to find significant variation between countries on women's propensity to bargain, suggesting an important role for culture and education as well as for biology. There is nothing stopping women from being encouraged and educated to bargain harder. Indeed, that may be easier to achieve than overcoming direct sex discrimination and prejudice.

Explaining the gender pay gap

Men earn more money than women, and a lot of research has been done over the years to explain this gender pay gap. Two recent papers by Andrés Erosa, Luisa Fuster and Diego Restuccia present an innovative approach to the problem:

A General Equilibrium Analysis of Parental Leave Policies, Federal Reserve Bank of Richmond Working Paper no. 05-08, August 2005.

A Quantitative Theory of the Gender Gap in Wages, Federal Reserve Bank of Richmond Working Paper no. 05-09, September 2005.

The first paper develops and calibrates a general-equilibrium model of fertility and labor-market decisions, building on the Mortensen and Pissarides (1994) labor-market framework. The second paper, using panel data from the National Longitudinal Survey of Youth, finds that "a large portion of gender differences in labor market attachment can be traced to the impact of children on the labor supply of women".

The quantitative life-cycle model of fertility, labor supply, and human capital accumulation decisions developed in these two papers offer, to my mind, a more sophisticated approach than earlier work. The latest Richmond Fed Region Focus magazine has a good summary of these two papers by Doug Campbell: The Pay Divide (PDF). It explains the authors' approach well. Here's the nub of it:

...the findings mostly suggest that corporate discrimination is not as important as once thought. ..Restuccia sees a greater role for home-based discrimination than market-based. Societal norms pull more women into the home than market norms.

There is more research to come, with some interesting policy implications. According to Campbell:

The extent to which availability of flexible and part-time jobs, as well as child care services, can close the gender wage gap is next up on their to-study list. “Maybe policies should be implemented to allow women to balance the time they have to put into home with their kids but still be in the labor market,” Restuccia says. “As long as they’re able to supply labor time to the market that is close to that of males, then the incentives are going to work out that women acquire as much human capital as men.”

Why fertility rates and female employment rates are positively correlated

Not so long ago I posted a chart from The Economist which showed a positive correlation between women's fertility rate and their labour force participation in OECD economies. It was a somewhat counter-intuitive result, and the post generated quite a few comments. Now I've come across an explanation, by Luisa Fuster and Jose M. Da Rocha.

The paper is entitled Why Are Fertility Rates and Female Employment Ratios Positively Correlated Across OECD Countries? (PDF). Forthcoming in the International Economic Review, their paper finds that higher unemployment reduces fertility rates:

We develop a quantitative theory of fertility and labor market participation decisions in order to investigate the role of labor market frictions in generating the observed positive association between fertility and employment among OECD countries. We find that unemployment induces females to postpone and space births which, in turn, reduces the total fertility rate. ...We also find that labor market frictions (low probability of finding a job) can generate a positive association between female employment ratios and fertility rates across economies.

This is not necessarily a fixed relationship, however. The authors note that:

Our findings suggest that the sign of the correlation between employment ratios and fertility rates across countries could change as the job-finding rate of females increases.

Working women and bigger government

Chris Dillow at Stumbling and Mumbling blog has an intriguing post about a cost of women workers: bigger government.

There's a trade-off between gender equality and limited government. That's the message of this paper (pdf) from the CEPR:

Higher participation by women in the labor market is indeed positively associated with larger governments...We find strong evidence of a causal link between participation in the labor market and government size: a 10 percent rise in participation in the labor market leads to a 7 to 8 percent rise in government size. This effect is robust to the country sample, time period, and a set of controls.

This is because government spending is partly on childcare and education, which make it easier for women to go out to work rather than raise children. So working women vote for bigger government.

But what causes women's participation in the labour force? Here's something that's overlooked - the prices of home appliances such as dishwashers and microwaves. Falls in these make it cheaper to buy ways of saving on labour in the home, which in turn make it easier for women to go out to work. Another paper by the same authors, Tiago Cavalcanti and Jose Tavares, says:

A decrease in the relative price of appliances - the ratio of the price of appliances to the consumer price index - leads to a substantial and statistically significant increase in female labor force participation. In the United Kingdom, for instance, the decline in the relative price of home appliances accounts alone for about 10 to 15 percent of the increase in female labor force participation from 1975 to 1999.

Put these two papers together and we have the message - technical progress has contributed to bigger government. There are very few wholly good things.

One of the comments on his post alleges 'spurious regression'. But my recent post pointing out the positive correlation between fertility rates and women's labour force participation lends support to the Cavalcanti and Tavares thesis. As I noted in a comment on my post:

The two main mechanisms at work here, I think, are that as the proportion of women of child bearing years in the workforce grows:

1) So too does the supply of formal childcare places, either by government or the private sector; and

2) The more likely governments are to respond to working women's calls for assistance with childcare costs, maternity leave and pay, flexible working arrangements, child tax credits and the like. These policies act to lower the opportunity cost of having children and to make it more feasible to combine work and being a parent.

It seems reasonable to expect that additional government assistance could, if sufficiently generous, substantially boost total government spending.

UPDATE: Andrew Norton at Cattalaxy blog has also posted on the subject: Are women to blame for big government?

The rise of the older worker

Two recent reports highlight the growing importance of older workers in the labour market. Andrew Taylor writes in today's Financial Times that older workers take most new jobs:

More than half the [UK] jobs created in the past year were filled by people above the state pension age, according to official statistics. The rise in the number of working older people reflected increasing financial pressures created by pensions shortfalls and a growing willingness of employers to take on older staff, employers’ organisations said.

The employment rate for “pensionable” workers, men aged 65-plus and women aged 60-plus, ranged between 7.5 per cent and 8 per cent for most of the 1990s but rose to 10.4 per cent during the three months to the end of January, according to the Office for National Statistics.

...Sam Mercer, director of the Employers’ Forum on Age, said that employers faced with skills shortages were also more prepared to hire older people who were physically and mentally fitter than previous generations. Large supermarket chains, such as Tesco and J Sainsbury, in particular, had made an effort to employ older staff to reflect the balance of society at large and an ageing population that provided many of their customers, she said.

The number of people of pensionable age in employment rose to 1.13m during the three months to the end of February; up 85,000 on the corresponding period 12 months ago. That accounted for more than half the total rise of 147,000 in the number of people in employment. There was also a strong increase in employment among people aged between 50 and pension age, which rose by 57,000. ...By comparison, employment figures for younger people fell by 13 per cent for 16- and 17-year-olds and by 0.7 per cent for the 25-34 age group.

Meanwhile Elizabeth Olsen from the New York Times reported Tuesday on the growth of employment agencies aimed at seniors: In Demand: Employment Agencies for 50-Plus:

The demand is so great that employment services specializing in the over-50 age group are cropping up, and large agencies like  Manpower Inc. and  Kelly Services are finding that more older adults —  born between 1946 and 1964 — are seeking  temporary or permanent jobs.

According to the United States Bureau of Labor Statistics, 16.2 percent of workers — more than 40 million people — were 55 or older in 2005, and the bureau expects that number to rise to 20 percent or more by 2015.

"Interest is growing because people are contacting me every week, even from abroad, to ask about setting up an agency like ours," said Susan M. Allan, executive director of Senior Employment Resources.

Recognizing that more retirees are looking for work, AARP has partnered with Manpower and Kelly Services to help place workers over 50. The association has a "featured employers" section on its Web site, aarp.org, to connect job seekers with national corporations.

Melanie Holmes, Manpower's vice president for corporate affairs, said that her company began focusing on older workers five years ago with its Reinvent Retirement program. "About 10 percent of our work force is 50 and over now, and we expect that number to keep growing," she said.

A decade ago, Kelly Services started a program for older adults. Jocelyn Lincoln, the director of product management for the company, said the agency places about 67,000 older workers yearly.

Also emerging are online services like Retireecareers.com and Seniorjobbank.org. As far as training, there are a limited number of programs. Experience Works, a nonprofit group, and the National Council on Aging, in conjunction with AARP, run job placement centers nationwide, and provide office skills training for low-income older workers.

But many older people looking for work do not need training, experts say. As the first wave of post-World War II babies turns 60 this year, the demand for jobs will only grow. And the majority of these people will have up-to-date skills because they have worked for all or most of their lives and never expected to retire.

A 2005 study by the John J. Heldrich Center for Workforce Development at Rutgers University said that most baby boomers think traditional retirement is obsolete. The survey found that more than two-thirds of workers in that generation expected to work after the traditional retirement age; 15 percent anticipated starting their own business after retirement; and 13 percent expected to stop working entirely. The study reported that the number of workers saying they would never retire rose to 12 percent in 2005 from 7 percent in 2000.

These trends are certainly encouraging. But they don't mean older workers have an easy time of it. The quality of jobs available to seniors is far from ideal. As the FT story reported:

Patrick Grattan, chief executive of the Third Age Employment Network, said that there were still problems “with the poor quality and low rates of pay of some part-time and full-time jobs filled by older workers”.

And as last Sunday's BBC TV Panorama programme, Must have own teeth, clearly showed, age discrimination is still pervasive in Britain. the website has several accompanying pieces, including two on ageism research.

The Economist: A guide to womenomics

An Economist article on women and the world economy argues that "the future of the world economy lies increasingly in female hands". The piece, A guide to womenomics, claims that:

In rich countries, girls now do better at school than boys, more women are getting university degrees than men are and females are filling most new jobs. Arguably, women are now the most powerful engine of global growth.

The article gives particular attention to the growing role of women at work. The most interesting chart highlights the positive correlation between women's fertility rate and their labour force participation:

It is sometimes argued that it is shortsighted to get more women into paid employment. The more women go out to work, it is said, the fewer children there will be and the lower growth will be in the long run. Yet the facts suggest otherwise. Chart 3 shows that countries with high female labour participation rates, such as Sweden, tend to have higher fertility rates than Germany, Italy and Japan, where fewer women work. Indeed, the decline in fertility has been greatest in several countries where female employment is low.

It seems that if higher female labour participation is supported by the right policies, it need not reduce fertility. To make full use of their national pools of female talent, governments need to remove obstacles that make it hard for women to combine work with having children. This may mean offering parental leave and child care, allowing more flexible working hours, and reforming tax and social-security systems that create disincentives for women to work.

Countries in which more women have stayed at home, namely Germany, Japan and Italy, offer less support for working mothers. This means that fewer women take or look for jobs; but it also means lower birth rates because women postpone childbearing. Japan, for example, offers little support for working mothers: only 13% of children under three attend day-care centres, compared with 54% in America and 34% in Britain.

The whole article is free online.

The end of sisterhood?

For the first time in history, women in developed societies can take up any occupation or career they please. This has brought enormous benefits. But it has also had some less positive consequences—the death of sisterhood, a decline in female altruism and growing disincentives to bear children. So argues Professor Alison Wolf in her cover story for the April 2006 issue of Prospect magazine. Here's an extract from her article, Working girls:

Women used to enter the elite as daughters, mothers and wives. Now they do so as individuals.

This marks a rupture in human history. It is one that has brought enormous benefits to many people, and to many women in particular. But its repercussions are not all positive, either for society as a whole, or for all women. We are no more likely to return to the old patterns than we are to subsistence agriculture, so we need to understand what the new female labour market means for all our lives.

Three consequences get far less attention than they deserve. The first is the death of sisterhood: an end to the millennia during which women of all classes shared the same major life experiences to a far greater degree than did their men.

The second is the erosion of "female altruism," the service ethos which has been profoundly important to modern industrial societies - particularly in the education of their young, and the care of their old and sick.

The third is the impact of employment change on childbearing. We are familiar with the prospect of demographic decline, yet we ignore, sometimes wilfully, the extent to which educated women face disincentives to bear children.

India's job reservation policy: extend, or rethink?

The Times of India reported yesterday that the UPA government plans job quotas in the private sector:

In a significant departure from its oft-repeated stance that implemention of quotas in private sector should be "voluntary," the UPA government is considering enacting a law to extend job reservations to the non-government sector. Sources say government, which had so far maintained that it would leave it to the private sector to consider introducing job reservations, is now bringing both an amendment to the Constitution, as well as enacting a new law for the purpose. Law ministry has already brushed aside the view of attorney-general Milon Banerjee that there’s no room in the Constitution for getting the private sector to implement quotas.

Is this such is a good idea? Amit Varma ceratinly doesn't think so. And a new paper by economists Vani K. Borooah, Amaresh Dubey and Sriya Iyer also raises some serious concerns. Has Job Reservation Been Effective? Caste, Religion, and Economic Status in India (PDF) looks at the Indian policy of reserving a certain proportion of government jobs for the so-called “Scheduled Castes” (SC) and “Scheduled Tribes” (ST). They investigate, using data from the 55th round (1999-2000) of the National Sample Survey, the extent to which job reservation for the SC and the ST have benefited these groups. The authors found that positive discrimination did boost employment:

Our conclusion was, that in the context of regular salaried and wage employment in the country, job reservation succeeded in raising the representation of persons from the SC and Tribes by about 5 percentage points.

However the job reservation policy has three important flaws. First, it has a "discriminatory bias against Muslims who do not benefit from such policies". Second, it emphasises caste or tribe rather then income or wealth:

The goal of reservation in India has been to bring about an improvement in the welfare who, historically, have been economically and socially depressed. But, in arriving at this judgement about who should be eligible for reservation, the criterion has been a person’s caste rather than his income or wealth. Consequently, groups belonging to what Article 115 of the Indian Constitution calls “socially and educationally backward classes” have benefited from reservation even though, in practice, many of these groups could not be regarded as “backward”. This has meant that many of the benefits of reservation have been captured by well-off groups from the depressed classes (for example, chamars from the SC) while poorer groups from the depressed (for example, bhangis from the SC) have failed to benefit.

Finally, the job reservation policy does not emphasise education:

...job reservation policies placed little emphasis on improving the job-related attributes of persons from the SCT. Our simulations showed that even modest improvements in the education levels of SCT persons could deliver significant employment gains.

...We conclude that an alternative, and more effective, way of raising the proportion of prime-age men from the SC/ST groups in regular salaried or wage employment would be to improve their employment-related attributes.

Rather than extend the policy, it may be time for a fundamental rethink.

Roemer on equality of opportunity

John E. Roemer has written a short contribution on Equality of opportunity (PDF) for the New Palgrave Dictionary:

Equality of opportunity is to be contrasted with equality of outcome. While advocacy of the latter has been traditionally associated with a left-wing political philosophy, the former has been championed by conservative political philosophy. Equality of outcome fails to hold individuals responsible for imprudent actions that may, absent redress, reduce the values of the outcomes they enjoy, or for wise actions that would raise the value of the outcomes above the levels of others’. Equality of opportunity, in contrast, ‘levels the playing field,’ so that all have the potential to achieve the same outcomes; whether or not, in the event, they do, depends upon individual choice.

A brief, clear introduction to the topic. But for those who want to delve deeper, I recommend his recent paper (with Gary Fields, V. Dardononi and L. Sanchez) How demanding should equality of opportunity be, and how much have we achieved? (PDF). It applies his ideas to several datasets, to compute what kinds of equal opportunity have been achieved in the US and the UK:

This paper proposes tests of various notions of equality of opportunity and applies them to intergenerational income data for the United States and Britain. Equality of opportunity holds if the chances that individuals have to succeed depend only on their own efforts and not upon extraneous circumstances that may inhibit or expand those chances. Exactly what “effort” and “circumstances” should include is a contentious issue.

We discuss four channels through which parents affect the income earning capacities of their children: provision of social connections, formation of beliefs and skills in children through family culture and investment, genetic transmission of native ability, and instillation of preferences and aspirations in children. We test for equality of opportunity using ever more stringent definitions and find only the weakest of support in the United States and no support at all in Britain.