Seasonally adjusted figures for job vacancies

Published:

The number of job vacancies varies throughout the year, and the pattern in many industries is repeated every year. The statistics on job vacancies are now adjusted for this seasonal pattern, which makes it possible to follow the underlying development from quarter to quarter.

The statistics on job vacancies are based on a quarterly survey of businesses and on jobs reported to the A-ordningen. The survey started in the 1st quarter of 2010, and the figures for the first year were published at the beginning of 2011. The total number of job vacancies and the overall job vacancy rate are published every quarter, and a breakdown is also given for 20 industry groups.

In order to comment on the development in the total number of job vacancies and in the different industries, we have so far only looked at changes from the corresponding quarter in the previous year. This is because several of the industries have variations due to seasonal effects. The series are now seasonally adjusted, which means that the effects of systematic seasonality are removed. With a seasonally adjusted series as a base, we can look at changes from the previous quarter, and can then also follow the short-term development. In addition to the seasonally adjusted figures, the trend-cycle is also given. This series has had both the seasonality and the irregularities eliminated, and gives the long-term development for the series. The trend-cycle and seasonally adjusted figures are available in table 11587 in Statbank

Seasonal adjustment of the total

The series for the total number of job vacancies has a clearly seasonal pattern (See the green line in Figure 1). We see that, for each of the 7 years, there have been most job vacancies in the 2nd quarter and fewest in the 4th quarter. When the effects of seasonality are removed, the 3rd quarter had the most vacancies in 2016, and the 2nd quarter had the fewest. In reality, most vacancies were in the 2nd quarter, but the seasonal adjustment enables us to see that if there were no seasonal variations, the 3rd quarter that year would have had most vacancies (see the yellow line that shows the seasonally adjusted series).

 

Figure 1. Job vacancies

Q1-2010 Q2-2010 Q3-2010 Q4-2010 Q1-2011 Q2-2011 Q3-2011 4. kv. 2011 Q1-2012 Q2-2012 Q3-2012 Q4-2012 Q1-2013 Q2-2013 Q3-2013 Q4-2013 Q1-2014 Q2-2014 Q3-2014 Q4-2014 Q1-2015 Q2-2015 Q3-2015 Q4-2015 Q1-2016 Q2-2016 Q3-2016 Q4-2016 Q1-2017
Non-adjusted 60500 67900 63900 55800 68900 78500 68000 62200 69200 72000 60900 57800 67600 74300 59100 50900 61800 64800 54100 53700 60400 62800 49500 45000 55300 56200 52300 48700 58300
Trend 58900 61200 63000 64000 66500 69400 70100 69100 67300 64700 64100 65200 66100 65700 62200 58400 57800 57900 57700 57900 56600 54500 52600 50900 50500 52100 53900 55000 55300
Seasonally adjusted 58500 60900 66000 63400 66100 70900 70400 70200 66100 64600 63600 66100 64500 67100 61600 57800 58700 58900 56800 59900 57500 56300 52200 50300 52700 51300 54800 54300 55400

The seasonally adjusted numbers fluctuate around the trend-cycle, the purple line in Figure 1. When considering the whole series, the main picture for the trend-cycle is that it was decreasing from the 3rd quarter of 2011 until the 1st quarter of 2016, admittedly with shorter periods of increase. From the 1st quarter of 2016, we observe an increasing trend-cycle. Looking at the latest figures, there were 58 300 vacancies for the 1st quarter of 2017. The seasonally adjusted figure is 55 400 for this quarter, the highest number since the 2nd quarter of 2015.

About the method

Industries with clear seasonal variations

When we examine each of the 20 industry groups, we find the clearest example of seasonal variation in education. In Figure 2, we see a clear peak for the 2nd quarter each year for the non-adjusted series (green line), and clear bottoms for the 3rd and 4th quarters. It is natural to find such a seasonal pattern for this industry, because the teaching institutions seek more new teachers ahead of a new school year than in the middle of it. The trend-cycle has been approximately constant from the start, and seems to be in a weak growth phase now.

Figure 2. Job vacancies in education

Q1-2010 Q2-2010 Q3-2010 Q4-2010 Q1-2011 Q2-2011 Q3-2011 4. kv. 2011 Q1-2012 Q2-2012 Q3-2012 Q4-2012 Q1-2013 Q2-2013 Q3-2013 Q4-2013 Q1-2014 Q2-2014 Q3-2014 Q4-2014 Q1-2015 Q2-2015 Q3-2015 Q4-2015 Q1-2016 Q2-2016 Q3-2016 Q4-2016 Q1-2017
Non-adjusted 3400 6000 2300 2700 2700 5100 2600 2200 4100 6900 2500 2400 3800 6000 2200 2500 2900 5800 2800 2300 3700 7800 2500 2400 3500 5200 2000 2900 4900
Trend-cycle 3500 3600 3600 3400 3200 3100 3300 3700 3900 4000 3900 3700 3600 3600 3500 3400 3400 3400 3400 3600 3900 4200 4100 3700 3300 3200 3500 4100 4400
Seasonally adjusted 3500 3500 3500 3900 2800 3000 4000 3200 4200 4000 3900 3500 3800 3600 3400 3700 2900 3500 4400 3400 3700 4700 4000 3600 3500 3100 3200 4300 4800

In addition to education, we also find clear seasonal variations in manufacture, transportation and storage, and in accommodation and food service activities. For this last industry, there are fewest vacancies in the 4th quarter of each year, while there are most vacancies in the 2nd or 3rd quarter. The higher number of vacancies in spring may be due to higher activity in parts of the industry in the summer, while for instance mountain lodges, where the activity is high during the winter months, may announce vacancies in the late summer and early fall.

Industries without systematic seasonal variations

In some industries, we find no systematic seasonal variations. However, this may change over time when new observations are added. The variations for these series are currently more random. There are four industry groups that are not seasonally adjusted. This applies to human health activities, real estate activities, the industry professional, scientific and technical activities, and to arts, entertainment and recreation. For this last group, it could be envisaged that the industry contains a systematic seasonality in relation to, for instance, amusement parks only open in the summer months. However, we find no systematic seasonal effects, as the industry contains a wide range of businesses. The seasonally adjusted figures published for these industries are the same as the non-adjusted figures.

Contact