Abstract
The sectoral approach has traditionally served as the principal tool for analyzing regional specialization; however, it does not reveal the functional content of employment or the distribution of tasks within economic activities. This article assesses the degree of correspondence between the functional structure of employment and the sectoral structure of output across 17 harmonized regions of Kazakhstan, a resource-dependent post-Soviet economy, over the period 2019–2024. Regional functional profiles are constructed within the task-based framework and comprise four aggregate task categories: non-routine cognitive (NRC), routine cognitive (RC), routine manual (RM), and non-routine manual (NRM). The sectoral structure is represented by the distribution of nominal output across 47 sectors. Inter-regional differences are measured using Jensen–Shannon distance. The empirical analysis is conducted on a dyadic panel of 816 observations, controlling for region size, urbanization, gross regional product per capita, extractive specialization, geographic distance, and time effects. Primary statistical inference relies on the Freedman–Lane permutation procedure within multiple regression quadratic assignment procedure (MRQAP). The results indicate a positive but moderate association between the two structures. The standardized coefficient on functional distance is 0.300 (MRQAP p = 0.001). Including the task variable raises R2 by 5.5 percentage points. The finding remains robust under most alternative distance metrics, different sets of control variables, and sequential exclusion of individual regions. However, the association weakens noticeably when the three largest cities—concentrating a substantial share of non-routine cognitive tasks—are excluded simultaneously. Within-region estimates of correspondence for 2019–2024 are considerably less precise. Pair fixed-effects models and correlations based on net change and cumulative path length do not reveal a systematic contemporaneous relationship between changes in functional and sectoral structures over the observed period. The six-year observation window does not allow the longer-term pattern of their coevolution or possible lagged adjustment to be determined. Overall, functional and sectoral structures are related but not interchangeable. The results provide statistically supported evidence of a positive inter-regional association, although its strength varies by distance metrics and regional subsamples. For regional policy, the findings underscore the value of combining sectoral specialization indicators with information on the characteristics of jobs and accumulated labor competencies.