One year of salary analysis – what’s next?

Just under a year after the introduction of the new Equal Opportunities Act, the Competence Centre for Diversity & Inclusion analysed around 120,000 personnel data. The results may seem surprising at first glance, especially when compared to figures published in the media.

Many of the companies we examined (97%) show good to very good results – in other words: either they fall below the tolerance threshold of 5% unexplained pay gap used by the federal government, or they even have no provable unexplained pay gap between men and women at all.

So far, so good.

The results of the 2018 Swiss Earnings Structure Survey (ESS) show a slightly higher difference: in Switzerland, the unexplained pay gap to the disadvantage of women amounts to 8.1% [1]. Where does this high figure come from, when most companies adhere to the tolerance threshold of 5%? (As a side note, the threshold is not a “discrimination tolerance”, but more of an “error tolerance”, since salary analyses are based on models which, firstly, cannot reflect the entire reality and, secondly, must take statistical uncertainties into account).

The simple answer: The two figures are not directly comparable. But what makes the difference? Two major differences make Switzerland’s results (from the ESS) incomparable with individual results from Logib.

Difference 1

In the ESS, all companies are considered collectively, i.e., all persons examined in Switzerland who are classified in lower management, for example, and who otherwise also have similar characteristics, are compared with each other. However, being classified in lower management does not mean the same in all companies. This means that people are grouped who have completely different tasks and therefore also different salaries.

Is that problematic? Not, if women and men are equally affected. Let’s imagine that company A has employed 10 men in lower management, while company B has employed 10 women in lower management. In company A, employees in lower management have responsibility for large teams of up to 50 people, while in company B the team responsibility in lower management is very low. If company A pays an average higher wage to lower management, it looks like there is an unexplained wage difference between women and men. If there were 5 women and 5 men in lower management in both company A and company B, then this wage difference would not be due to gender, on average there would be no wage difference at all between women and men.

Do the national surveys overestimate or underestimate the unexplained wage inequality due to this grouping of companies? The answer cannot be generalized. It always depends on the companies’ employee composition.

From a gender equality perspective, the following considerations are important: If male-dominated industries and functions tend to pay higher wages than female-dominated ones, then in a nationwide consideration based on the ESS, the unexplained wage differences increase – and rightly so.

Difference 2

The complexity of a function is recorded differently in Logib than in the ESS. In the ESS, an international standard (ISCO-08) is used to determine the job’s competence level based on the job title. In Logib, on the other hand, the company has more leeway: The complexity is determined based on the so-called “operational competence level”. Although this operational competence level in Logib has many similarities to the competence level of the ESS, it differs in that the company has more flexibility to rank the complexity of the functions relative to the complexity of the other functions in the company. This allows for a better representation of the operational reality.

Which method is more adequate? From a company perspective, Logib is preferable. Logib is more precise in the way it compares employees within a company. However, if we want to compare wages across Switzerland, the ESS’s standard classification based on job title makes more sense.

The method by which functions are classified based on their complexity can make a big difference in the results (unexplained wage difference), especially since the company’s level of competence is one of the most important characteristics influencing wages, along with occupational status. From an equality perspective, you should, therefore, firstly, be very careful in your company about the respective classifications in terms of operational competence level and professional position and review them from time to time. Secondly, a difference of less than 5% in a Logib analysis does not mean that wage discrimination is not present in your company. Only further in-depth wage analysis can reveal that.

Explanation:

The Earnings Structure Survey of the Federal Statistical Office shows that women in Switzerland earn on average 19% less than men. A large part of this is explainable (and thus not unfounded). A difference of 8.1% remains unexplained (and thus potentially unfounded, i.e. discriminatory)¹.

¹ See https://www.ebg.admin.ch/ebg/en/home/topics/work/equal-pay/background-information/facts-and-figures.html