From office to construction site: Differences in work practices across occupations, industries and regions

2026-06-04T10:02:01+03:00

Session room:

Numerical and adaptive problem-solving skills of professionals in Finland and Baltic countries: analysis of the PIAAC Cycle 2 data
Mareks Niklass, University of Latvia, Latvia

This presentation examines the numeracy and adaptive problem-solving (APS) skills of professionals (an occupational group defined at the 1-digit level of the International Standard Classification of Occupations (ISCO)) in Finland and the Baltic countries. Professionals are typically required to hold at least a college degree to enter their occupation, and access to these professions is often strictly regulated by law and professional associations (e.g., for doctors, lawyers, and teachers). Once employed, professionals are generally expected to participate in further training, and for many, lifelong learning has become an integral part of their professional careers and ethos. The presentation aims to identify factors that account for cross-country differences in numeracy and adaptive problem-solving scores using data from PIAAC Cycle 2. Professionals in Finland achieve significantly higher scores in both numeracy (M = 328.21, SE = 1.62) and APS (M = 370.77, SE = 1.45) tests. Corresponding results in Estonia are somewhat lower (numeracy: M = 312.47, SE = 1.29; APS: M = 288.02, SE = 1.08). Scores in Latvia (numeracy: M = 289.34, SE = 2.38; APS: M = 266.64, SE = 2.04) and Lithuania (numeracy: M = 268.41, SE = 2.16; APS: M = 244.35, SE = 1.73) are clearly more modest. Across all countries examined, more than 80% of professionals hold at least a college degree. More than 90% were employed during the week preceding the interview, and close to 90% had an indefinite employment contract. In other words, professionals in Finland and the Baltic countries are predominantly employed in stable white-collar jobs. What factors account for these differences in skill levels? While older respondents and women tend to score lower in numeracy and APS tests, these effects are relatively weak compared to work-related factors captured by questionnaire items, such as how often respondents take measurements, read or create charts, or use specialized software or programming languages at work. For example, 36% of Lithuanian professionals reported never using specialized software at work, compared to only 14% of Finnish professionals. In conclusion, the results suggest that what matters most is whether—and how often—numeracy and ICT-related skills are used in the workplace.

Presentation to be updated

The division of labor and routinization of work
Peter Njekwa Ryberg, Jönköping International Business School, Sweden

This paper examines how the division of labor affects the routinization of work at the level of individual workers. Building upon the theory that the division of labor is limited by the extent of the market, I examine how the local density of economic activity impacts task specialization and in turn routine intensity. To measure task specialization and routine intensity of individual workers, I utilize the Swedish 2011/2012 and 2022/2023 cycles of the Programme for International Assessment of Adult Competencies (PIAAC) Survey of Adult Skills, which encompass detailed accounts on work-related tasks and skills. I link workers to geocoded establishments using longitudinal linked employer-employee microdata, to measure the density of economic activity in the area where they work. To test the proposed hypothesis, I use a mediation analysis regression framework where average treatment effects are obtained using an instrumental variable correlated random coefficients model. Historical data on railway passenger traffic, linked to the geography of the workers in my sample, instruments contemporary density and task specialization. The results indicate that density has a positive direct average treatment effect on task specialization, which fully mediates the relationship between density and routine intensity. Increasing density-induced task specialization by one standard deviation leads to an average increase of routine intensity by 7 percent. The average treatment effect is heterogenous by codifiability of tasks, occupation, education, and industry, and grows in size at increasing levels of density. The findings show how a finer division of labor yields a greater routine intensity in carrying out work-related tasks. Insights into how the organization of work influences its quality, drawn from the usage of PIAAC data in combination with microdata, can deepen our understanding of worker autonomy, monopsony power, and well-being.

Parallel session G

From office to construction site: Differences in work practices across occupations, industries and regions