female_inventors.knit

A Gender Gap to More Innovation in Switzerland

May 2021 by Matthias Niggli and Christian Rutzer

It was only 50 years ago when Switzerland finally gave women the legal right to vote. Since then, women have continued fighting for gender equality and equal representation in politics, civil society, and the economy. Progress has been made but has generally taken a long time. Women remain underrepresented in many fields and this is especially the case in some domains of the labor market.

According to the NZZ, only 4 of the 100 largest Swiss employers had a female CEO in 2018, and a Business Monitor survey found that in 2019 the female share in management positions of Swiss companies was only 23.9 percent. Another domain where women still tend to be underrepresented is in companies’ innovative activities. For example, the United States Patent and Trademark Office (USPTO) recently reported that in 2019 the share of patents with at least one female inventor was 21.9 percent. Although this share has increased steadily since the 1980s, women clearly continue to be underrepresented in R&D.

This is obviously problematic regarding equal gender representation as such. But can it also be of concern beyond this consideration? As innovations and technical inventions are the main drivers of growth and prosperity in the long run, there might be several economic reasons why the gender bias may be harmful. First, if there are more women who, in principle, could be scientists and engineers than those currently involved in innovative activities, the economy is most probably not able to utilize the skills of its domestic workforce in an optimal way. Stated differently, if more women could be attracted to R&D careers, this could result in more innovation and growth (Bedee et al., 2011). Second, mixed teams simply tend to perform better. This has been shown to be case for more diverse management (Dezsoe & Gaddis Ross, 2012) and research teams (Wikhamn & Wikhamn, 2020) alike. Hence, the economic advantages of increasing the share of female inventors are potentially quite straightforward: Having more female inventors may not only increase the quantity of innovations, but also improve their overall quality.

Interestingly, some countries have been more successful than others in achieving higher levels of female participation in R&D. This is shown in Figure 1. The dynamic graph is based on patent data from the United States Patent and Trademark Office (USPTO) and allows countries to be compared over a long period of time with regard to their share of female patent inventors.

Figure 1: Proportion of women among all inventors over time

Source: CIEB`s own estimations based on data from the USPTO. The data used for this plot is available on GitHub.

At the beginning of the 1980s, the rate of female inventors was very low in almost all countries. But it was especially low for Switzerland, where less than two out of 100 inventors of Swiss patents published at the USPTO were women. Female representation in Switzerland has improved somewhat over the last 40 years. But in 2018 still only 9 out of 100 Swiss patent inventors are women.

Only 9 out of 100 Swiss patent inventors are women

This puts Switzerland somewhere in the lower midfield in a country-comparison. Over time, the female shares of inventors have increased steadily in most countries, but in no country have they ever come close to equal representation of men and women. At the same time, there are a lot of differences among countries: For example, in 2018 still fewer than 7 out of 100 German inventors were women. In turn, Asian countries and territories, such as Hong-Kong, South Korea and Taiwan, had rather high rates, with around 20 women per 100 inventors.

What could be possible reasons for these low female shares? One reason is simply that not so many women graduate in fields like Science, Technology, Engineering or Mathematics (so-called STEM fields), which is generally a precondition to becoming an inventor. Accordingly, the overall supply of potential female inventors is rather low. Another point could be that among the female graduates in STEM fields, not so many go on to become inventors. This, in turn, would imply that inventor careers are less attractive for women compared to men. Figure 2 allows us to investigate both of these channels. We use the same USPTO data as before to calculate the proportions of female inventors (stated on the vertical y-axis) and relate it to data from the OECD on female STEM graduate shares (stated on the horizontal x-axis). Due to data availability, we use information for the period 2010-2015 and aggregate the data of this period to smooth out yearly fluctuations. Every point in Figure 2 represents a country and we plotted the point for Switzerland in red for better visibility.

Figure 2: Female inventor shares and female graduate shares in STEM

Source: CIEB`s own estimations based on data from the USPTO and the OECD. We use information for the period 2010-2015 and aggregate the data of this 5-year period to smooth out yearly fluctuations. The data used for this plot is available on GitHub.

The average share of female STEM graduates across all countries being considered is 36.7 percent, which confirms the well-known underrepresentation of women in these fields. However, it is interesting to note that the shares vary considerably across countries. For example, Eastern and Southern European countries tend to have more female graduates (e.g. Poland, Estonia, Italy, or Greece). Therefore, it might very well be possible for other countries to increase their female graduate shares in STEM. Even more striking is that every country in our sample is below the dashed diagonal line (see the green dots in the graph). This diagonal line shows the female inventor shares that could be achieved if countries employed the exact same proportion of women in R&D as they train through their education systems. Take Switzerland, for example: 28.8 percent of all students who graduated in STEM fields between 2010 and 2015 were women. However, Switzerland’s female inventor share for the same period was only 9.2 percent. Hence, only a very disproportional share of Swiss female STEM graduates pursues inventor careers.

Figure 3: Conversion Rate between female STEM graduate and inventor shares

Source: CIEB`s own estimations based on data from the USPTO and the OECD. We use information for the period 2010-2015 and aggregate the data of this period to smooth out yearly fluctuations. The dashed vertical line states the median ratio among these 20 countries. The data used for this plot is available on GitHub.

Figure 3 allows us to examine this more directly for the 20 countries with the most inventors between 2010 and 2015. The plot highlights the ratio between female STEM graduate shares and female inventor shares, which can be interpreted as the fraction of female STEM graduates who become inventors. In other words, Figure 3 depicts how far away from the diagonal line a country is located in Figure 2. A ratio of 100 percent would imply that the country is located exactly at the diagonal line. Noticeably, South Korea has clearly the highest ratio, suggesting that it is relatively more successful in motivating female STEM graduates to become inventors. Switzerland, in turn, shows a ratio of only 32 percent. Since this is slightly above the median, this might seem acceptable at first sight. However, this is the case only because almost all countries perform relatively badly. Since all countries have ratios below 100 percent in Figure 3 and are located below the diagonal line in Figure 2, the share of female patent inventors is lower than the share of female STEM graduates in every country. In other words, not only are there relatively few female STEM graduates in many countries, but those women who chose to study a STEM field are less likely to realize an R&D career, compared to their male peers in general. And the farther away a country is positioned from the diagonal line, the more pronounced is this imbalance.

Female shares differ substantially among countries

That said, there are notable differences across countries. Figure 4 zooms in to examine these differences more closely. The zoomed plot shows the same relationship as Figure 2, but additionally depicts the average female inventor share (the dashed horizontal line) and the average female STEM graduate share (the dashed vertical line) across countries.

Figure 4: Female inventor shares and female graduate shares in STEM

Sources and notes

CIEB`s own estimations based on data from the USPTO and the OECD. We use information for the period 2010-2015 and aggregate the data of this period to smooth out yearly fluctuations. The dashed lines indicate the respective median among these 39 countries. The data used for this plot is available on GitHub.
The selectable technology groups are based on Schmoch et al. (2008) and contain the following fields:
Overall: All technology fields.
Chemistry: Biotechnology, Chemistry, Pharmaceuticals.
Electrical engineering: Audio-visual technology, Computer technology, Digital communication, Electrical machinery, Semiconductors, Telecommunications.
Instruments: Control, Measurement, Medical technology, Optics.
Mechanical engineering: Engines, Handling, Machine tools, Mechanical elements, Textile and paper machines, Transport.
Other fields: Furniture, Games and Civil engineering.

Recall from Figure 3 that South Korea seemed to perform a lot better than other countries. However, there are some caveats to this argument. As you can see, countries like South Korea or Belgium start off with rather low female inventor shares. This could mean that there is just stronger self-selection. That is to say, among their lower numbers of female graduates, there are more who are determined to become inventors than in other countries. Another argument might be that the advantage we see of these countries could be driven by country-specific industry specialization. Think of it this way: If there are some technology areas where women inventors are more adequately represented (for example, Chemistry compared to Mechanical Engineering), a country that specializes in such technology areas might have a higher overall female share. But this higher overall share would not be due to some successful policy framework in this country, but rather to industry specialization.

Figure 4 allows us to examine the latter caveat by selecting a technology area in the slider panel. After selecting a given technology area, the graph states female inventor shares of patents exclusively from this technology area across countries. Observing these country-level differences in female inventor shares, we see they are generally less pronounced within the same technology area. For example, countries are more closely positioned to each other along the y-axis in the fields “Mechanical Engineering” or “Electrical Engineering”. Accordingly, technological specialization seems to be one good explanation for differences across countries. However, substantial gaps remain between countries (e.g. in “Chemistry”) and it is noteworthy that South Korea is always most closely located to the left among all countries, that is, towards the diagonal 45-degree line. This suggests that beyond industry characteristics, policies and cultural norms are also important.

Let us now switch back to the female inventor shares in the overall economy. Note that countries located in the upper right of Figure 4 perform the best, as they have both a relatively high share of female STEM graduates and female inventors. This is the case for Latvia, Lithuania, Poland, Colombia, Portugal, and Spain in particular. Countries in the upper left do not have a high share of female STEM graduates. But they at least make good use of their potential for new inventions and innovations. In this regard, South Korea is successful, but France and Belgium are also doing better relative to other countries.

Switzerland and all other German-speaking countries are amongst the most poorly performing countries.

In contrast, countries in the lower left perform the most poorly. Remarkably, Switzerland and all other German-speaking countries are located in this area. There can be several possible reasons for this, ranging from cultural peculiarities to less family-friendly work environments, or more attractive outside options for female STEM graduates (for example, more decent salaries in more family- friendly occupations, such as teaching or public service).

What can be done?

In short, Switzerland does not perform well compared to other countries and there seems to be room for improvement. On the one hand, the female graduate share in STEM fields is low. In this regard, female role models (Porter & Serra, 2020; Antecol et al., 2015; Bottia et al., 2015 ), gender norms (Kahn & Ginther, 2017; Alesina & Nunn, 2013; Cvencek et al., 2011) and the overall education system (Lim & Meer, 2018; Kulturel-Konak et al., 2011; Griffith, 2010) are important factors to consider. However, achieving lasting changes within these frameworks takes time. For example, according to the OECD, the Swiss share of female STEM graduates increased only slightly from 28.1% in 2010 to 29.4% in 2018. Hence, Swiss policy-making, academia and the business world should foster existing projects (for example Swiss TecLadies and Girls Can Code, or initiatives from ICT Switzerland, EPFL or Zukunftstag) and tackle ongoing challenges more intensively.

On the other hand, the conversion rate of female STEM graduates to female inventors in Switzerland – while above average in international comparison – is relatively low in absolute terms. One important aspect to increase it may be the compatibility of family and career (Cech & Blair-Loy, 2019). In R&D-related occupations, this might be particularly challenging today, because long lab work hours are often the case, to give one example. Therefore, investments and more flexible models in child day care could be an important channel to consider. From an industry perspective, a shift towards enabling more part-time work could be a promising way to increase the attractiveness of R&D careers for female STEM graduates (Mavriplis et al., 2010). More generally, adaptions in the Swiss tax system could be important as well (Ecoplan, 2019).

Making good use of the skill potential of its workforce is key for Switzerland’s future prosperity

To sum up, policies in different areas are important channels to raise the participation of women in inventive activities. As the Swiss economy constantly requires new technical innovations, new ideas and improved processes to remain one of the world`s leading innovation hubs, making good use of the skill potential of its workforce is key for Switzerland’s future prosperity. Hence, there is a strong case for pushing policies aimed at increasing the attractiveness of R&D careers for women.

Did you enjoy this article?
Share it with your friends and colleagues on Social Media

This report was written using R Markdown, R shiny and Plotly. Data and code to reproduce the figures presented in this article are available on GitHub. Calculations and data processing were performed at sciCORE scientific computing center at the University of Basel. We thank Conny Wunsch for valuable feedback and Monty Sufrin for excellent proof-reading.

CIEB | University of Basel | Follow us on Social Media