How to model Russia’s interregional migration flows on the base of PADS method // Edinak, Potapenko

Ekaterina Edinak, Vadim Potapenko — IEF RAS, Russia

Summary: To develop high-quality demographic forecast by region it is necessary to forecast the future scale of interregional migration flows. In the study, incoming migration flows are modeled by region on the base of PADS methodology, which is traditionally used for modeling changes in personal consumption. In migration context, PADS is considered as a system of nonlinear regressions. The difference in per capita incomes between regions is explanatory variable of the regressions. It is assumed that if the growth of per capita income in a region i is faster than in other regions, then the incoming migration flow to this region is going to increase. For most of Russian regions, this hypothesis has been confirmed: income per capita elasticities of the incoming migration flow
by region are positive. However, for some regions (about 30), in addition to income, other migration factors, such as infrastructural, climatic, social, etc., are also important. As a whole, the PADS method demonstrates adequate results: the fitting values of the incoming migration are very close to the actual ones in the retrospective period.

For reference: E. Edinak, V. Potapenko. Forecast of Russian Interregional Migration Flows. 5 September 2019. Russia, Sochi, 27th Inforum Conference, URL: https:// ecfor.ru/publication/forecast-of-russian-interregional-migration-flows/

Link: https://ecfor.ru/publication/forecast-of-russian-interregional-migration-flows/

——————————

This presentation happened 5 September 2019, Russia, Sochi at 27th Inforum Conference.

The conference & materials: https://ecfor.ru/27th-inforum-world-conference/

Playlist with the other videos from the conference: https://www.youtube.com/playlist?list=PLJBJFs8UgQgpmWWFzqw87jdwOBQXQmLfu
Inforum’s site: http://www.inforum.umd.edu/

Inforum, or Interindustry Forecasting at the University of Maryland, was founded nearly 50 years ago by Dr. Clopper Almon, now Professor Emeritus of the University. It is dedicated to improving business planning, government policy analysis, and the general understanding of the economic environment.

The host side was the Institute of Economic Forecasting of the Russian Academy of Sciences. IEF RAS specializes in fundamental, applied and exploratory scientific research in the field of analysis and forecasting of the socio-economic prospects of Russia and its regions, and the development of recommendations and proposals with a goal to improve the quality of economic policy in Russia. The site is https://ecfor.ru.

——————————

Notes:

Why do we need the forecast of migration flows between regions?

1) Migration affects the demographic situation in the regions;
2) migration influences the parameters of regional labor markets;
— in particular, migration contributes to the growth of regional inequality in terms of population and labor
3) migration affects regional infrastructure load;
4) migration redistributes population incomes from one region to another;
5) and much more…

Some difficulties of modeling

1) Changing the methodology of accounting for internal migrants, breaking the time series.
2) The variety of types of internal migration: educational, labor, etc.
3) In this regard, there are many factors affecting the structure of migration: social, demographic, infrastructural, climatic, geographical.

Database for modeling

1) The dynamic series of the arrived population by region (xi), 2011-2017.
2) The dynamic series of income per capita by region (yi), 2007-2017. 3) Estimates of income per capita over the past 5 years have been explanatory variables.
The number of regions is 82.

Conclusions

1) The shown settlement system on the base of “PADS” was developed primarily for forecasting internal migration.
2) The number of regions with negative income elasticity of migration less than 50%.
3) For regions with the negative income elasticities of migration, only the additive part of the equation can be used to forecast the incoming flow

Further research need

— Forecast of the incoming migration and the outgoing migration by region.
— Embed this forecast in a regional demographic forecasting model.
— Modeling separate flows of labor migrants.
— Assessment of the scale of redistribution of population income due to internal migration.