Skip to main content
SpringerLink
Log in

The impact of police violence on migration: evidence from Venezuela

  • Original Paper
  • Published:
Journal of Population Economics Aims and scope Submit manuscript

Abstract

This study unveils the causal effect of authoritative violence on individuals’ likelihood to migrate. Specifically, we examine the migration patterns of Venezuelans during the 2017–2018 political and economic crisis. We draw insights from regional-level data on civilian casualties caused by security forces, along with information extracted from the ENCOVI-2018 survey data that captures migration flows. The estimates rely on travel time from the capital city as an instrumental variable and are robust to the inclusion of several household- and socio-economic regional-level characteristics. The findings strongly suggest that authoritative violence is a significant non-economic push factor for international migration. Moreover, additional evidence indicates that this type of violence influences the skill composition of migrants, especially in the context of South-to-South migration flows.

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

Access this article

Log in via an institution

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

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5

Similar content being viewed by others

Notes

  1. We use the definition of authoritative violence provided by Morrison and May (1994). According to the authors, authoritative violence includes also the actions of state-sponsored actors, such as the so-called death-squad activity, which is often “authorized” by the state, even when the perpetrators are not wearing police or military uniforms or are officially off-duty.

  2. As put by Chiquiar and Hanson (2005), “intra-state” and “inter-state” migration are two different phenomena, which need to be analyzed separately (see also Fernandez-Dominguez (2020)).

  3. We proxy the level of violence through the homicide rate also drawing on UN recommendations. See for more: https://www.unodc.org/unodc/en/data-and-analysis/global-study-on-homicide.html

  4. There are different types of militarization (Flores-Macías and Zarkin 2021; Galavís 2020) The first is the one in which the militarized police rely on military tactics and equipment, maintaining civilian jurisdiction as well as a low-hierarchy structure. The second one is the paramilitary police, operating under military deployment tactics and units, maintaining civilian jurisdiction and a police rationale. The third one is represented by the constabularized militaries, assuming citizen security tasks such as “crime prevention, crime contention, and prison security while reporting to the Ministry of Defence” (Galavís 2020, p. 71)

  5. As shown in Figs. 1 and 2, we dropped the migration data regarding Portuguesa, because we do not have data on violence for that region.

  6. Individual information on migrants’ characteristics and their destinations are reported by the interview respondents, who are the household heads. We report here the questionnaire question translated from Spanish to English: “During the last few years, since June 2013, has anyone who lived with you in this household moved to another country? In what year and month did they migrate?” We restrict the sample of migrants to those who moved between 2017 and September 2018. The reason why we decided to keep the observations of 2018 in the estimation is mainly related to the attempt to grasp the potential time-lagged effect of the authoritative violence “outbreak” of 2017. ENCOVI only accounts for those migrants who have at least a household member left behind. This could imply a loss of representativity of the sample, limiting the validity of our analysis. To overcome such a limitation, we use the Encuesta Dirigida a la Poblaciòn Venezolana que reside en el Pais (see Section 4.1). The survey, performed at the end of 2018, collects information about 9847 Venezuelan migrants residing in Peru, which is the second-largest receiving country.

  7. The variable regarding the education level is a binary one, and takes value 1 if the household head has at least attained a high school diploma.

  8. So far it has been estimated that 5.2 million Venezuelans have moved beyond the border. The most common destination (1.8 million migrants) is Colombia; Peru welcomed 830,000 migrants, Chile 455,000, Ecuador 360,000, and the USA 352,000.

  9. The average age of the migrants in the sample is 29 years old with respect to 41 of the population remaining in Venezuela. Thirty-two percent of the migrants have at least a bachelor’s degree, whereas only 13% of those who have remained at home are college graduates.

  10. It is an institution capillary distributed across the country, which analyzes data on homicides matching governmental sources, journalistic investigation, and international organizations inquiries. In early 2005, the Laboratory of Social Sciences (LACSO) of Venezuela set out to build a Violence Observatory in order to obtain accurate information on the phenomenon of victimization and the perception of insecurity in Venezuela, given the restrictions that at the time existed for journalists and academics in accessing official statistics on “known cases” of violence registered by police (https://observatoriodeviolencia.org.ve/sobre-nosotros/; Uribe et al. (2016)). If until 2010 OVV’s work was mainly relying on statistical predictions, starting from 2016/2017 the observatory has begun applying a more complex methodology of collecting information by media, via victimization surveys, by organizing focus groups and in-depth interviews, and by collecting extra-official information from different institutions (OVV, 2017 in Avila (2018)).

  11. A death is classified as an intentional homicide following the International Statistical Classification of Diseases and Related Health Problems published by the World Health Organization.

  12. All the violence-related variables are weighted, per 100,000 inhabitants. Data regarding violent deaths refer to the year 2017 and are aggregated at the regional level.

  13. The details of the construction of the dependent variable are outlined in Section 3.1.

  14. In particular, the regional employment rate is calculated using the percentage of employed people aged 19 to 54, and the education level is based on the average number of years of education.

  15. By considering data from 2011, we aim to exclude the heterogeneous effect of the political and economic crisis across regions.

  16. The average access to running water is represented by the percentage of households with at least weekly access to it.

  17. The Encuesta Nacional de Hospitales showed that in November 2018, 33% of the beds in the country’s hospitals were inoperative. Given the inoperability of laboratories, 43% of hospitals in Venezuela do not have the capacity to examine medical tests. In addition, about 70% of hospitals reported experiencing a lack of electrical service and water shortages. Hospitals also experience a shortage of emergency medicines (50% shortage). The ENH is conducted by the “Médicos por la Salud” Observatory and data were collected in the major hospitals in Venezuelan regions during the second week of November 2018.

  18. According to our estimations performed thanks to the data provided by the Encuesta Dirigida a la Población Venezolana que reside en el Pais (see Section 4.1), only the 0,09% of the migrants abandon Venezuela are by sea, the 3,85% by air, and the 1,47% by foot. The high majority of them (94,59%), leave the Country by bus.

  19. The dependent variable used in the test is a dummy variable that takes value 1 if the individual migrated during the period of 2014–2016. Conversely, it takes value 0 if the individual continued to reside in Venezuela until 2016. For individuals who migrated during 2017–2018, the migration variable was set to 0. As a result, these individuals were designated as non-migrants in this particular sample.

  20. Table A1, in the Appendix, reports the coefficients of the Logistic estimation and the related marginal values, performed as a robustness check to support the stability and consistency of the main linear empirical assumptions.

  21. The R-squared greatly increases from 0.001 to 0.074, while the coefficient of authoritative violence ranges from 0.054 to 0.056, also considering the inclusion of the ENPOVE sample as a robustness check.

  22. As suggested by Oster (2019), we choose a Rmax = 1.3R cutoff and we report the values of \(\delta \) for which \(\beta \) = 0.

  23. Since heteroscedasticity, serial correlation, and data clustering can affect instrument strength we also compute F-statistic of Montiel Olea-Pflueger and we report the TSLS critical values (Olea and Pflueger 2013) Again, the F-statistic exceeds the critical TSLS value at 5%, thus confirming the result of the Stock and Yogo under-identification test.

  24. We performed a Durbin-Wu-Hausman test to prove the consistency of both OLS and instrumental variable approach (Baum et al. 2003). The non-significant chi-square statistic (0.60853) suggests that both the estimators are consistent, although the OLS is the more efficient. Despite such evidence, we perform them both to see if their results are comparable.

  25. We perform the falsification test by using both the logarithmic functional forms of travel time and kilometers from the Capital.

  26. The Venezuelan Observatory Monitor de Victimas shows that in the capital city and in the governorate of Miranda between 2017 and 2018, 92% of victims of police violence were male.

  27. In Table 1 we report the descriptive statistics of our migrants’ sample. Forty-four percent of migrants have a high school diploma, and 24% do not have formal education at all.

  28. An in-depth discussion related to these findings is presented in Section 5.

  29. ENPOVE was carried out by the Peruvian National Institute for Statistics (INEI) between November and December 2018. It is “representative by design” of the Venezuelan population residing in Peru. In particular, it was conducted in the five largest cities in the country, where 85% of Venezuelans reside. According to IOM estimates, Peru is the second-largest receiving country for Venezuelan migrants; currently about 1 million out of 5.2 million of them live there. Therefore the ENPOVE sample can be largely representative of Venezuelan households that have entirely migrated.

  30. While ENCOVI is representative of the whole Venezuelan population (25 to 28 million people estimated), ENPOVE is representative of approximately 550,000 Venezuelan migrants. The merge has been performed by applying the appropriate sample weights.

  31. We considered only those Venezuelans who declared not to have any left-behind member of their family.

References

  • Ades AF, Glaeser EL (1995) Trade and circuses: explaining urban giants. Q J Econ 110(1):195–227

    Article  Google Scholar 

  • Altonji JG, Elder TE, Taber CR (2005) Selection on observed and unobserved variables: assessing the effectiveness of catholic schools. J Polit Econ 113(1):151–184

    Article  Google Scholar 

  • Anatol M, Kangalee QM (2021) Crime in Trinidad and Tobago: the possible impacts of increased crime due to migration from Venezuela. Migration and Development 10(2):260–272

    Article  Google Scholar 

  • Anderton CH, Brauer J (2016) Economic aspects of genocides, other mass atrocities, and their prevention. Oxford University Press

    Book  Google Scholar 

  • Arceo-Gómez EO (2012) Drug-related violence, forced migration and the changing face of Mexican immigrants in the United States

  • Ascani A, Bettarelli L, Resmini L, Balland PA (2020) Global networks, local specialisation and regional patterns of innovation. Res Policy 49(8)

  • Avila K (2018) cómo analizar las actuales cifras delictivas en Venezuela. Amnistía Internacional. Venezuela. Disponible en: https://issuu.com/amnistia/docs/-cifras_delictivas_de_venezuela_an (Consulta del 21-2-2019)

  • Bahar D, Ibáñez AM, Rozo SV (2021) Give me your tired and your poor: impact of a large-scale amnesty program for undocumented refugees. J Dev Econ 151:102652

    Article  Google Scholar 

  • Baum CF, Schaffer ME, Stillman S (2003) Instrumental variables and gmm: estimation and testing. Stand Genomic Sci 3(1):1–31

    Google Scholar 

  • Becker GS (2011) Cep discussion paper no 1079 september 2011 fear and the response to terrorism: an economic analysis gary s. becker and yona rubinstein

  • Becker GS, Rubinstein Y et al (2004) Fear and the response to terrorism: an economic analysis. University of Chicago mimeo 93(1)

  • Bohra-Mishra P, Massey DS (2011) Individual decisions to migrate during civil conflict. Demography 48(2):401–424

    Article  PubMed  Google Scholar 

  • Boulding KE, Singh P (1962) The role of the price structure in economic development. Am Econ Rev 52(2):28–38

    Google Scholar 

  • Briceño-León R, Perdomo G (2019) Violence against indigenous children and adolescents in Venezuela. Cad Saude Publica 35:e00084718

    Article  Google Scholar 

  • Brown LA, England KV, Goetz AR (1989) Location, social categories, and individual labor market experiences in developing economies: the Venezuelan case. Int Reg Sci Rev 12(1):1–28

    Article  PubMed  Google Scholar 

  • Brown LA, Goetz AR (1987) Development-related contextual effects and individual attributes in third world migration processes: a Venezuelan example. Demography 24(4):497–516

    Article  PubMed  Google Scholar 

  • Bull B, Rosales A (2020) The crisis in Venezuela. European Review of Latin American and Caribbean Studies/Revista Europea de Estudios Latinoamericanos y del Caribe

  • Cederman LE, Buhaug H, Rød JK (2009) Ethno-nationalist dyads and civil war: a gis-based analysis. J Conflict Resolut 53(4):496–525

    Article  Google Scholar 

  • Cerna CM (2019) Counter-terrorism resolutions and initiatives by regional institutions: The organization of American States and the inter-American commission on human rights. International Human Rights and Counter-Terrorism 75–107

  • Chiquiar D, Hanson GH (2005) International migration, self-selection, and the distribution of wages: evidence from Mexico and the United States. J Polit Econ 113(2):239–281

    Article  Google Scholar 

  • Clemens MA (2017) Violence, development, and migration waves: evidence from central American child migrant apprehensions. Center for Global Development Working Paper

  • Clemens MA, Mendola M (2020) Migration from developing countries: selection, income elasticity, and simpson’s paradox

  • Contreras VR (2014). The role of drug-related violence and extortion in promoting Mexican migration: unexpected consequences of a drug war. Latin American Research Review 199–217

  • Corrales J (2020) Democratic backsliding through electoral irregularities. European Review of Latin American and Caribbean Studies/Revista Europea de Estudios Latinoamericanos y del Caribe 41–65

  • Engel S, Ibáñez AM (2007) Displacement due to violence in Colombia: a household-level analysis. Econ Dev Cult Change 55(2):335–365

    Article  Google Scholar 

  • Fernandez-Dominguez AO (2020) Effect of actual and perceived violence on internal migration: evidence from Mexico’s drug war. IZA Journal of Development and Migration 11(1)

  • Flores-Macías GA, Zarkin J (2021) The militarization of law enforcement: evidence from Latin America. Perspect Polit 19(2):519–538

    Article  Google Scholar 

  • Galavís NG (2020). Rule of law crisis, militarization of citizen security, and effects on human rights in Venezuela. European Review of Latin American and Caribbean Studies/Revista Europea de Estudios Latinoamericanos y del Caribe 67–86

  • García Marco D (2016) Una pena de muerte disimulada: la polémica operación de liberación del pueblo, la mano dura del gobierno de Venezuela contra el crimen. Retrieved from https://www.bbc.com/mundo/noticias-america-latina_38126651

  • Goldman RK (2009) History and action: the inter-American human rights system and the role of the inter-American commission on human rights. Hum. Rts. Q 31:856

    Article  Google Scholar 

  • Ibáñez AM, Vélez CE (2008) Civil conflict and forced migration: the micro determinants and welfare losses of displacement in Colombia. World Dev 36(4):659–676

    Article  Google Scholar 

  • Ibáñez Londoño AM, Vélez CE et al (2005) Civil conflict and forced migration: the micro determinants and the welfare losses of displacement in Colombia

  • Ingram MC, da Costa MM (2019) Political geography of violence: municipal politics and homicide in Brazil. World Dev 124:104592

    Article  Google Scholar 

  • Kleibergen F, Paap R (2006) Generalized reduced rank tests using the singular value decomposition. Journal of Econometrics 133(1):97–126

    Article  MathSciNet  Google Scholar 

  • Knight BG, Tribin A (2020) Immigration and violent crime: evidence from the Colombia-Venezuela border (Tech. Rep.). National Bureau of Economic Research

  • Legler T (2020) A story within a story. European Review of Latin American and Caribbean Studies/Revista Europea de Estudios Latinoamericanos y del Caribe 135–156

  • Marsteintredet L (2020) With the cards stacked against you. European Review of Latin American and Caribbean Studies/Revista Europea de Estudios Latinoamericanos y del Caribe 87–106

  • Martínez Herrera LA (2020) Crimen organizado y violencia homicida en ciudades intermedias

  • Massey DS, Axinn WG, Ghimire DJ (2010) Environmental change and out-migration: evidence from Nepal. Population and environment

  • Maya ML (2014) Venezuela: the political crisis of post-chavismo. Social Justice 40(4 (134):68–87

  • McDoom OS (2014) Predicting violence within genocide: a model of elite competition and ethnic segregation from rwanda. Polit Geogr 42:34–45

    Article  Google Scholar 

  • Moore WH, Shellman SM (2006) Refugee or internally displaced person? to where should one flee? Comp Pol Stud 39(5):599–622

    Article  Google Scholar 

  • Morrison AR (1993) Violence or economics: what drives internal migration in Guatemala? Econ Dev Cult Change 41(4):817–831

    Article  Google Scholar 

  • Morrison AR, May RA (1994) Escape from terror: violence and migration in post-revolutionary Guatemala. Lat Am Res Rev 29(2):111–132

    Article  ADS  Google Scholar 

  • Mummolo J (2018) Militarization fails to enhance police safety or reduce crime but may harm police reputation. Proc Natl Acad Sci 115(37):9181–9186

    Article  ADS  PubMed  PubMed Central  Google Scholar 

  • Neumayer E (2003) Good policy can lower violent crime: evidence from a cross-national panel of homicide rates, 1980–97. J Peace Res 40(6):619–640

    Article  Google Scholar 

  • Niedomysl T (2011) How migration motives change over migration distance: evidence on variation across socio-economic and demographic groups. Reg Stud 45(6):843–855

    Article  Google Scholar 

  • Olea JLM, Pflueger C (2013) A robust test for weak instruments. Journal of Business & Economic Statistics 31(3):358–369

    Article  MathSciNet  Google Scholar 

  • Osse A (2006) Understanding policing. Amnesty International Netherlands, A Resource for Human Rights Activists (Netherlands, p 2012

  • Oster E (2019) Unobservable selection and coefficient stability: theory and evidence. Journal of Business & Economic Statistics 37(2):187–204

    Article  MathSciNet  Google Scholar 

  • Pareja FJA (2020) Archipiélagos politicos bajo la tormenta en Venezuela. European Review of Latin American and Caribbean Studies/Revista Europea de Estudios Latinoamericanos y del Caribe 21–40

  • Rivera M (2016) The sources of social violence in Latin America: an empirical analysis of homicide rates, 1980–2010. journal of peace research 53(1):84–99

  • Rodriguez C, Villa E (2012) Kidnap risks and migration: evidence from Colombia. J Popul Econ 25(3):1139–1164

    Article  Google Scholar 

  • Sassen S (2013) The global city: New York. Princeton University Press, Retrieved from, London, Tokyo. https://doi.org/10.1515/9781400847488

    Article  Google Scholar 

  • Schultz TP (1971) Rural-urban migration in Colombia. The Review of Economics and Statistics 157–163

  • Schutte S, Donnay K (2014) Matched wake analysis: finding causal relationships in spatiotemporal event data. Polit Geogr 41:1–10

    Article  Google Scholar 

  • Sprout HH, Sprout M (2015) Ecological perspective on human affairs. Princeton University Press

    Google Scholar 

  • Starr H (1978) “Opportunity" and “willingness’ as ordering concepts in the study of war. International interactions 4(4):363–387

    Article  Google Scholar 

  • Stock J, Yogo M (2005) Asymptotic distributions of instrumental variables statistics with many instruments (vol. 6). Chapter

  • Uribe PM, Mazuera-Arias R, Murillo E, Albornoz-Arias N (2016) 6. homicidio en la frontera: Análisis de la violencia homicida en el estado táchira, venezuela49. Crimen organizado y violencia homicida en ciudades intermedias 242

Download references

Acknowledgements

First of all, the authors wish to thank the Catholic University Andrés Bello de Caracas for the provision of the survey data for research purposes, and the Observatorio Venezolano de Violencia for the provision of data on violence. We also thank the Editor-in-Chief Klaus F. Zimmermann and the three anonymous reviewers of the Journal of Population Economics for the relevant contributions and challenging comments thanks to which this paper has reached its final form, along with the Managing Editor Madeline Zavodny. Finally, the authors also thank Giacomo De Luca, Alessandra Faggian, Alexander Moradi, Steven Stillman, Claudio Detotto, Andrea Ascani, Daria Denti, and all the faculty members of the Authors’ affiliations who intervened during the internal seminars, for the constructive comments and the useful suggestions provided.

Author information

Author notes

  1. Carlo Caporali contributed equally to this work.

Authors and Affiliations

  1. Department of Economics and Management, Free University of Bozen, Piazza Universitá, 1, Bolzano, 39100, Italy

    Federico Maggio

  2. Department of Economics, University of Bologna, Piazza Scaravilli, 2, Bologna, 40126, Italy

    Federico Maggio

  3. Area of Social Sciences, Gran Sasso Science Institute, V.le L. Rendina, 28, L’Aquila, 67100, Italy

    Carlo Caporali

Authors
  1. Federico Maggio
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Carlo Caporali
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Carlo Caporali.

Ethics declarations

Conflict of interest

The authors declare no competing interests.

Additional information

Responsible editor: Klaus F. Zimmermann

Publisher's Note

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

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary file 1 (pdf 122 KB)

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Maggio, F., Caporali, C. The impact of police violence on migration: evidence from Venezuela. J Popul Econ 37, 24 (2024). https://doi.org/10.1007/s00148-024-00997-x

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1007/s00148-024-00997-x

Keywords

JEL Classification

Search

Navigation