Analysing the evolution of income from employment in the first years after obtaining a qualification through the Swiss education system is key to assessing how well the education and training system caters to the labour market.
Some very interesting sources for this are the individual accounts of the Central Compensation Office (CCO) and the FSO's Structural Survey (SS).
The methodological report documents the methodology used by the LABB programme (longitudinal analyses in the field of education) to determine the standardised income for full-time workers, and people's hourly and gross incomes. It also provides, using data from 2016, a detailed descriptive analysis of the differences between the constructed income variable and the standardised wage from the Swiss Earnings Structure Survey (ESS), and provides information on the limits of interpretation.
The CCO individual accounts provide a measure that should, in principle, come very close to that of the total gross monthly income from employment received by a person, while the SS measures the number of hours worked per week for all paid work as reported by the respondent.
The method of calculating the “actual” standardised income for a full-time position considers the “actual” income that the person would have if they worked a “usual 100%” in a given occupation. This usual 100% is calculated on the basis of full-time references defined based on responses from SS respondents. This means that the calculation also takes account of any overtime (paid or unpaid) worked in general by people in the occupation. This income therefore differs conceptually from the ESS wage and it is not possible to formally validate this “actual” standardised income with the ESS.
The methodological report documents in detail – for all dimensions and indicators that LABB plans to examine – to what extent this income differs from the ESS wage. This also provides a better understanding of the significance and the utility of this “actual” standardised income for education research.
Two complementary approaches were adopted.
- The first approach compares the results from the ESS with those from the CCO/SS within universes that are as comparable as possible. The advantage of this approach is that the comparisons allow conclusions to be drawn for a large proportion of workers resident in Switzerland. The disadvantage is that the differences observed may be due to measurement effects between the two sources.
- The second approach is based on the possibilities offered by AHVN13 to link the samples from the CCO/SS and from the ESS and to examine, at individual level, the differences observed between the two sources. For this linked database, important selections are carried out in order for the comparisons to make sense. The advantage of this approach is that it limits measurement effects. The disadvantage is that it covers a more restricted portion of the population. This second approach also requires a reweighting to allow analyses and direct comparisons of incomes to be made at the level of the reduced population. The report detailing the reweighting carried out is also available below.
The comparisons are based on two types of indicator: the first measures the distribution of differences in income between the ESS and the CCO/SS on the basis of a comparison between each individual, while the second makes aggregated comparisons.
This document should allow the reader to better interpret and utilise the results of future LABB publications based on the use of the CCO/SS constructed incomes.
Concerning the separate databases, the 2016 ESS standardised wage is CHF 64 (median value) higher than the income constructed via the CCO/SS 2016 (CHF 6 680 for the ESS and CHF 6 616 for the CCO/SS), although the difference is not statistically significant.
Concerning the linked database, the median of the differences between the “actual” CCO/SS standardised income and that of the ESS is CHF 150 (or 2.2% of the ESS median wage). This result reflects the fact that the number of hours worked is higher on the whole in the SS than in the ESS.
Despite these limited deviations, it is important to remember that the constructed standardised CCO/SS income by definition measures something different from the ESS standardised wage. It measures an income for a full-time position which is calculated on an “actual” basis, whereas the ESS wage reflects the “contractual” aspect. The difference is intrinsically important for people who do a significant amount of paid overtime not compensated by leave or variable working hours. We conclude that this CCO/SS income calculated on an “actual” basis can be an advantage for analyses in social and economic sciences because it focuses on the employee's actual work situation.