Invisible Women: Exposing Data Bias in a World Designed for Men

main aim was to understand how much output could be produced and what consumption needed to be sacrificed to make sure there was enough available to support the war effort.’ To do this they counted everything produced by government and businesses and so ‘what governments do and what businesses do came to be seen as the definition of the economy’. But there was one major aspect of production that was excluded from what came to be the ‘international convention about how you think about and measure the economy’, and that was the contribution of unpaid household work, like cooking, cleaning and childcare. ‘Everyone acknowledges that there is economic value in that work, it’s just not part of ‘the economy’,’ says Coyle.

This was not a mere oversight: it was a deliberate decision, following a fairly vigorous debate. ‘The omission of unpaid services of housewives from national income computation distorts the picture’, wrote economist Paul Studenski in his classic 1958 text The Income of Nations. In principle, he concluded, ‘unpaid work in the home should be included in GDP’. But principles are man-made, and so ‘after a bit of to-ing and fro-ing’, and much debate over how you would measure and value unpaid household services ‘it was decided’, says Coyle, ‘that this would be too big a task in terms of collecting the data’.

Like so many of the decisions to exclude women in the interests of simplicity, from architecture to medical research, this conclusion could only be reached in a culture that conceives of men as the default human and women as a niche aberration. To distort a reality you are supposedly trying to measure makes sense only if you don’t see women as essential. It makes sense only if you see women as an added extra, a complicating factor. It doesn’t make sense if you’re talking about half of the human race. It doesn’t make sense if you care about accurate data.

And excluding women does warp the figures. Coyle points to the post-war period up to about the mid-1970s. This ‘now looks like a kind of golden era of productivity growth’, Coyle says, but this was to some extent a chimera. A large aspect of what was actually happening was that women were going out to work, and the things that they used to do in the home – which weren’t counted – were now being substituted by market goods and services. ‘For example buying pre-prepared food from the supermarket rather than making it from scratch at home. Buying clothes rather than making clothes at home.’ Productivity hadn’t actually gone up. It had just shifted, from the invisibility of the feminised private sphere, to the sphere that counts: the male-dominated public sphere.

The failure to measure unpaid household services is perhaps the greatest gender data gap of all. Estimates suggest that unpaid care work could account for up to 50% of GDP in high-income countries, and as much as 80% of GDP in low-income countries.² If we factor this work into the equation, the UK’s GDP in 2016 was around $3.9 trillion³ (the World Bank’s official figure was $2.6 trillion⁴), and India’s 2016 GDP was around $3.7 trillion⁵ (compared to the World Bank’s figure of $2.3 trillion).

The UN estimates that the total value of unpaid childcare services in the US was $3.2 trillion in 2012, or approximately 20% of GDP (valued at $16.2 trillion that year).⁶ In 2014 nearly 18 billion hours of unpaid care were provided to family members with Alzheimer’s (close to one in nine people aged sixty-five and older in the US are diagnosed with the disease). This work has an estimated value of $218 billion,⁷ or, as an Atlantic article put it, ‘nearly half the net value of Walmart’s 2013 sales’.⁸

In 2015, unpaid care and domestic work in Mexico was valued at 21% – ‘higher than manufacturing, commerce, real estate, mining, construction and transportation and storage’.⁹ And an Australian study found that unpaid childcare should in fact be regarded as Australia’s largest industry generating (in 2011 terms) $345 billion, or ‘almost three times the financial and insurance services industry, the largest industry in the formal economy’.¹⁰ Financial and insurance services didn’t even make second place in this analysis; they were shunted into a lowly third place by ‘other unpaid household services’.

You will notice that these are all estimates. They have to be, because no country is currently systematically collecting the data. And it’s not because there is no way of doing it. The most common way of measuring the amount of unpaid work women do is with time-use surveys. Individuals are asked to keep a time diary of their movements throughout the day – what they are doing, where, and with whom. It is because of this form of data capture, writes prize-winning economist Nancy Folbre, that we now know that ‘in virtually every country, women undertake a disproportionate share of all non-market work, and also tend to work longer hours overall than men do’.

Standard time-use surveys were primarily designed to measure explicit activities such as meal preparation, house-cleaning or feeding a child.¹¹ As a result, they often fail to capture on-call responsibilities, such as having to keep an eye on a sleeping child or be available for an adult with a serious illness while you get on with something else – another data gap. Time-use surveys that explicitly aim to capture such responsibilities demonstrate that the market value of ‘on-call care’, even at a very low replacement wage, is significant,¹² but like with travel data this kind of care work is often lost within personal and leisure data.¹³ Folbre points to studies of home-based care for HIV/AIDS in Botswana which ‘estimated the value of services per caregiver at about $5,000 per year, a number that would substantially increase estimates of total spending on healthcare if it were included’.¹⁴

The good news is that these surveys have been on the increase in many countries. ‘In the first decade of the 21st century, more than 87 such surveys were conducted,