Women in trades over time

Track the number and proportion of women in trade sectors over time

Women in trades over time bannerWomen in trades over time banner


We know that the proportion of women in trade occupations is very low, but how low is it? Additionally, how can we track changes in this proportion over time? In this article we look at a few ways we can measure the number of women in trades using data available in the Integrated Data Infrastructure (IDI). We use data from the 2013 census, tax data from Inland Revenue Department (IRD) records, and data from industry training organisations (ITOs). Census data will give us the most accurate snapshot of a workforce at a point in time, while IRD and ITO data will allow us to see whether the number of women in a particular industry has increased over time, and let us track changes in the future.

Census snapshot

The census, held every five years, asks everyone in New Zealand about their occupation and the industry of the business they work for. We can use this information to define a workforce precisely. We exclude those who work in the trade sector with occupations such as office managers or general clerks in order to concentrate specifically on tradespeople. In the chart below we show the proportion of women in each trade sector. The chart can be filtered using the checkboxes on the right.

In the chart, we can see that Design and Drafting (Kitchen) is the major outlier with about equal numbers of men and women in this sector. Among the more traditional trades, Painting and Decorating has the largest proportion of women at 7%. Most sectors have between 1% and 4% of women in them, with the lowest proportions in Brick and Block, Plumbing, and Rigging (Rigging has fewer than six women so has been suppressed).

Proportion of women over time

The main limitation of the census data is that it is only available for a single point in time in 2013. Although data from the 2018 census will eventually become available, we also need a way to track the number of women on an ongoing basis. We can do this using tax records from the IRD.

Each employee and employer is linked to a business in a particular industry, so we can determine how many people are working in a particular industry each quarter. We do not have information on peoples' occupation, however, so we cannot refine our analysis to only tradespeople. The IRD data will include workers of all occupations in the workforce for each sector. Since a large proportion of administration staff are female, the proportion of women is greater than in the census data.

In general, we see an increase in the size of the workforce over time, with decreases in the years following the recession. Looking at employees only, the trend in the proportion of women varies between sectors. For example, Painting and Decorating has had a strong increase in the proportion of women, while the proportion of women in Cement and Concrete has been flat.

If we look at employers and self-employed workers, an important point to note is that the data for the 2017 tax year are incomplete, so we can only consider the data up to the first quarter of 2016. The trend in the proportion of women among employers is mostly negative, even in sectors where the proportion among employees is increasing. What might be the explanation for this? The proportion of employers who are female is higher than employees, whereas in the census data these proportions are similar. Many of these additional female employers are likely to be partners and spouses of tradespeople receiving income from a jointly owned business. The decrease in the proportion of female employers may result from changes in the way businesses are structured or even changes in marriage rates. For this reason, it is more meaningful to look only at employees when tracking changes in the proportion of women.

When you hover over the chart, arrow buttons will appear in the top-left. You can use these buttons to switch between yearly and quarterly views of the data.

Recently trained workers

The main disadvantage of tracking a workforce using IRD data is that we cannot filter by occupation. This means that we will pick up administration staff when we may only be interested in tradespeople. This can be especially problematic when we are determining the proportion of female tradespeople, as the number of female administration staff in an industry can vastly outnumber the female tradespeople.

To manage this limitation, we combine IRD records with enrolment records from ITOs. We define a recently trained worker as one who is working in an industry in a given year, and has trained with the relevant ITO in the preceding five years. This definition better captures the population of tradespeople in each sector and also represents the population most able to be affected by ITO initiatives.

In the chart below, we show the number of recently trained workers. Some sectors are too small for this definition to be useful and have been omitted. The proportion of women in each sector matches more closely to the census-defined workforce than the IRD-defined workforce. Painting and Decorating in particular shows a large increase in the proportion of women, as well as an increase in the number of recently trained workers overall. Other sectors such as Carpentry and Plumbing have more modest increases in the proportion of women, while the proportion of women in Electrotechnology has been essentially unchanging.


There are a number of ways to define a workforce in each trade sector, and each definition comes with its own advantages and limitations. The census-defined workforce is the most precise and complete definition, but can only give a snapshot in time. The IRD-defined workforce can be tracked over time but cannot be narrowed down by occupation. The combination of IRD and ITO records gives a workforce that can be tracked over time, which consists mainly of tradespeople but represents only a fraction of the total workforce. Each definition contributes to our understanding of the workforce and their combination gives us a clearer picture of the number of women in trades, what roles they have, and how these numbers are changing over time.

Overall, the number of women in trades appears to be increasing, though from a low base. Some sectors, such as Painting and Decorating, are attracting more women at a rapid pace, while in others the proportion is unchanging. Sectors that are not attracting more women should perhaps look to those that are for ideas about how to increase the participation of women.

Related Sweet Analytics articles