Skip to main content

Advertisement

Log in

The effect of environmental change on out-migration in the Brazilian Amazon rainforest

  • Original Paper
  • Published:
Population and Environment Aims and scope Submit manuscript

Abstract

We study how environmental change affects out-migration in the Brazilian Amazon using census data on municipal-level migration flows between 2005 and 2010. We characterize environmental change in terms of increases in municipal deforested area and gradual changes in temperature and precipitation as well as extreme weather events. Our empirical analysis is based on gravity models of migration, which consider simultaneously characteristics of origins and destinations as determinants of migration flows, treating for potential sources of endogeneity. We find evidence that out-migration from the Amazon region of Brazil is mainly associated with prior levels of deforestation, which is partially explained by the life cycle channel: a younger generation leaving the old frontier of colonization to an aging population. Furthermore, the links between environmental change and migration are more evident in the context of rural-urban and intra-regional migration, contributing to the process of urbanization in the Amazon.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Map 1
Fig. 1
Fig. 2
Fig. 3
Fig. 4

Similar content being viewed by others

Notes

  1. The Brazilian Law (Lei n. 11.326, July 24th 2006) recognizes as family farm the rural households that carry out agricultural and extraction activities (forest products and fishing, for example) for subsistence and/or cash income. This way, this group does not only comprise the small landholders settled by the government but also indigenous peoples, quilombolas (communities of Afro-Brazilians created by escaped slaves before the Abolition in 1888), caboclos (mixed white and indigenous or black and indigenous ancestry), seringueiros (rubber gatherers), and riverine families. According to the 2006 Rural census, the Amazon had nearly 413 thousand family farms, which represent 87% of the total number of farms in the region (IBGE 2018).

  2. Although the census may underestimate migration flows to inhospitable areas of the Amazon, this may not be a main concern in our estimates if we assume that measurement errors in the dependent variable are unrelated to the random errors in our models (Wooldridge 2003, sec. 9.3).

  3. We used the administrative division of municipalities of the Demographic Census 2000 because some of our explanatory variables use information of both the censuses 2000 and 2010 (to be explained below).

  4. In log-linear models, the percentage variation in Y given changes in the categories of a binary variable is equal (eβ − 1) × 100 (Halvorsen and Palmquist 1980).

  5. Although we cannot empirically prove this assumption, we did not have strong reasons to believe that deforestation and migration are a result of similar spatial processes, mainly because these variables present different patterns of spatial dependence: the coefficient of spatial correlation (I Moran) is 0.78 for the deforestation rate and nearly zero for the log of out-migrants.

References

Download references

Acknowledgements

This work was supported by CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (grant number 99999.000041/2017-03) and FAPESP – Fundação de Ampara à Pesquisa do Estado de São Paulo (grant number 2015/09312-0).

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Alexandre Gori Maia.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendix

Appendix

Table 5 Estimates of the first stage of the control function approach for the dependent variable deforestation in the municipality of origin, Brazilian Amazon
Table 6 Estimates of the control variables of the ZINB model (Eq. 3) for all types of migration flows between 2005 and 2010 (clustered robust standard errors between parentheses), Brazilian Amazon
Table 7 Estimates of the ZINB model (Eq. 3) using control function for all types of migration flows between 2005 and 2010 (clustered robust standard errors between parentheses), Brazilian Amazon
Table 8 Estimates of the ZINB model (Eq. 3) for rural-rural migration flows (cluster-robust standard errors between parentheses) between 2005 and 2010, Brazilian Amazon
Table 9 Estimates of the ZINB model (Eq. 3) for rural-urban migration flows (cluster-robust standard errors between parentheses) between 2005 and 2010, Brazilian Amazon
Table 10 Estimates of the ZINB model (Eq. 3) for urban-rural migration flows (cluster-robust standard errors between parentheses) between 2005 and 2010, Brazilian Amazon
Table 11 Estimates of the ZINB model (Eq. 3) for urban-urban migration flows (cluster-robust standard errors between parentheses) between 2005 and 2010, Brazilian Amazon
Table 12 Estimates of the ZINB model (Eq. 3) for intra-regional migration flows (cluster-robust standard errors between parentheses) between 2005 and 2010, Brazilian Amazon
Table 13 Estimates of the ZINB model (Eq. 3) for inter-regional migration flows (cluster-robust standard errors between parentheses) between 2005 and 2010, Brazilian Amazon

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Gori Maia, A., Schons, S. The effect of environmental change on out-migration in the Brazilian Amazon rainforest. Popul Environ 42, 183–218 (2020). https://doi.org/10.1007/s11111-020-00358-2

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11111-020-00358-2

Keywords

Navigation