The analysis of unemployment in Siberia at an initial stage of COVID-19

Labor market processes reflect the economic situation both at regional and national levels. The large-scale economic shock caused by the COVID-19 pandemic showed the need for conceptual changes both in terms of analysis and forecasting of labor market. In particular, the demand for research aimed at finding higher-frequency proxy indicators for modeling a short-term situation on the labor market increased sharply. Scientific literature accumulated certain experience in using it for research of different economic segments, including the labor market, at national and international levels. However, there is lack of knowledge in the mesoeconomic context. In this paper, we incorporated search data into econometric models to analyze the regional labor market in Siberia during the COVID-19 pandemic in 2020. We selected such keywords as «job» and “employment service” from Yandex as the main proxy indicators. In this research, we construct a set of models based on panel data: pooled regression model, model with random effects, model with fixed effects, dynamic panel data model. The findings show that the search data can be used as a significant factor for modeling regional unemployment. The use of dynamic models improved the accuracy by including the lags of dependent variable — unemployment rate. The applied logic can be utilized to analyze the impact of a wider range of shocks. 

1. Federal State Statistics Service (2020). Regions of Russia. Socio-economic indicators. Retrieved April 15, 2022, from https://gks.ru/bgd/regl/b20_14p/Main.htm

2. Kurovskiy, G. S. (2019). Using textual information to predict in macroeconomics. Moscow University Economic Bulletin, 6, 39–57. https://doi.org/10.38050/01300105201965

3. Podvoisky, G. L. (2021). Russian labor market under COVID-19 conditions: analysis, evaluation, prospects. Economic Sciences, 8, 67-84, DOI: 10.14451/1.201.67.

4. Samarukha, V. I., Krasnova, T. G., & Plotnikova, T. N. (2018). Migration movement of the population of Siberian Regions. Bulletin of Baikal State University, 8, 56–62, DOI: 10.17150/2500-2759.2018.28(1).56-62

5. Shcherbakov V. S., Kharlamova M. S., & Gartvich R. E. (2022). Methods and models of nowcasting of economic indicators using search queries. Materialy mezhregional’noj nauchnoj onlajn-konferencii «Razvitie ekonomiki regionov: prostranstvennaya transformaciya, global’nye vyzovy i perspektivy ekonomicheskogo rosta», 117–127.

6. Ulyankin,F. (2020). Forecasting Russian macroeconomic indicators based on information from news and search queries. Russian journal of money and finance, 4, 75–97. DOI: 10.31477/rjmf.202004.75