REMITTANCES AND HOUSEHOLD WELFARE IN MIGRANT-SENDING ECONOMIES

Background

International migration and remittances one of most significant sources of external finance for developing economies many cases exceeding FDI and ODA. Nigeria Africa largest recipient absolute terms and Edo State South-South repeatedly documented single largest migrant-sending state with well-established corridor to Europe dating decades. Remittances used variously for household consumption education healthcare housing occasionally productive investment represent potentially significant household-level channel through which international migration affects welfare. Empirical literature while growing faced specific methodological challenge: household decisions to send migrant abroad and resulting remittance flows not randomly distributed. Households successfully sending migrant and receiving substantial remittances plausibly differ in unobserved ways underlying wealth social networks risk tolerance unobserved earning potential differences could independently explain why remittance-receiving appear better off regardless genuine causal effect. Chukwuone Amaechina Eboh Iyoko Okpachu 2008 address concern for Nigeria using IV to assess effect on poverty while Fonta Onyukwu Nwosu 2011 use poverty and Gini decomposition documenting poverty falls substantially across all six zones as result of inflows. Obisie-Nmehielle Ike 2017 using 2009 World Bank Migration Survey find having international migrant and receiving remittances significantly increases welfare using consumer and durable-asset indices analysed OLS and probit though without explicit IV correction. This study builds directly by applying IV 2SLS strategy specifically designed to isolate causal effect using historical community migration-network density and distance to Lagos as instruments following McKenzie Rapoport 2007 Woodruff Zenteno 2007 and migration-cost proxies.

Migration and development economics topics | External: World Bank - Migration and Remittances, McKenzie and Rapoport 2007 - Migration Networks, NiDCOM - Nigerians in Diaspora Commission

Statement of Problem

Existing Nigerian evidence on remittances and household welfare informative but methodologically uneven in treatment of central identification challenge: same unobserved household characteristics entrepreneurial drive social capital pre-existing wealth sufficient to finance costly risky migration attempt that determine whether household successfully sends migrant abroad and receives substantial remittances plausibly also independently determine household welfare regardless of remittances themselves. Where endogeneity not explicitly addressed as in much descriptive and simple-regression-based Nigerian literature estimated remittance effects risk biased most plausibly upward overstating true causal welfare contribution specifically. While Chukwuone et al 2008 apply IV estimation to Nigeria poverty relationship and while international literature offers well-established instrument choices migration-network density McKenzie Rapoport 2007 exchange-rate and rainfall shocks Yang 2008 no identified recent Nigerian study applied community-migration-network instrument specifically within Edo State Nigeria leading migrant-sending state to estimate household-welfare effect net of endogeneity concern nor transparently reported full complement of instrument-validity diagnostics first-stage strength overidentification and formal endogeneity testing that would allow reader assess credibility of identification strategy. Study addresses gap surveys 340 households 71.2% remittance receipt instruments community density 2010 baseline and distance to Lagos.

Aim and Objectives

·         Document incidence and scale remittance receipt among sampled households Edo State

·         Estimate naive uncorrected OLS model relationship remittances and household per capita expenditure as benchmark

·         Estimate 2SLS model same relationship instrumenting remittance receipt with community migration-network density and distance to Lagos

·         Formally test whether remittance receipt in fact endogenous to welfare and assess validity and strength of chosen instruments

·         Compare OLS and 2SLS estimates and draw policy-relevant conclusions on true causal welfare contribution of remittances

Research Questions

·         What proportion sampled households receive remittances and how large transfers relative to expenditure?

·         What is naive uncorrected association between remittance receipt and per capita expenditure?

·         What is IV-corrected causal effect of remittances on per capita expenditure?

·         Is remittance receipt econometrically endogenous to welfare and are chosen instruments valid and sufficiently strong?

Research Hypotheses

·         H01: Remittances received have no significant effect on per capita expenditure

·         H02: Household remittance receipt is not endogenous to welfare i.e., OLS and 2SLS not statistically distinguishable

·         H03: Instruments community network density distance Lagos are not valid relevant predictors of remittance receipt

Tested 5% level. Naive OLS coefficient 0.757 p<0.001 rejects H01; first-stage joint F 83.49 p<0.001 rejects H03 confirming strong correctly signed instruments; 2SLS 0.679 p<0.001; Wu-Hausman 0.922 p0.338 does not provide decisive evidence endogeneity failing to reject H02 in this sample; Sargan 0.308 p0.579 does not reject validity supporting instrument exclusion restriction.

Significance

Significant to Nigerian migration and diaspora-engagement policymakers including Nigerians in Diaspora Commission providing methodologically rigorous causally identified estimate of remittances household-welfare contribution informing diaspora-engagement and remittance-facilitation policy. For Central Bank Nigeria monitoring remittance inflows as component balance of payments offers micro-level evidence complement macro flow statistics. For development partners and NGOs operating in Edo State migration corridor region also subject to significant anti-trafficking and safe-migration programming given history irregular migration to Europe welfare-effect estimates provide evidence relevant to weighing economic benefits of remittance channel against risks associated with irregular routes. Academic extends Chukwuone et al 2008 IV-based Nigerian evidence and Obisie-Nmehielle Ike 2017 welfare evidence with explicit fully diagnosed community-network IV strategy specific to Nigeria leading migrant-sending state. MPC 0.68-0.76 robust contribution recommends continued investment formal low-cost remittance-transfer infrastructure and diaspora-engagement programming recognising substantial welfare contribution.

Scope and Limitations

Delimited to households selected communities Edo State encompassing both remittance-receiving and non-receiving households data collected household expenditure remittance receipt demographic characteristics alongside community migration-network data and distance to Lagos. Examines per capita monthly expenditure as primary welfare outcome in relation to per capita monthly remittances received. Limitations: while IV strategy designed address endogeneity arising unobserved characteristics its validity rests on exclusion restriction that community migration-network density and distance to Lagos affect welfare only through effect on remittance receipt not other channel; while standard defensible assumption in international literature McKenzie Rapoport 2007 cannot be definitively proven assessed only indirectly through overidentification test Sargan 0.308 p0.579. Cross-sectional does not capture dynamic cumulative effects sustained receipt over multiple years. Self-reported remittance and expenditure subject recall social-desirability bias particularly given portion Edo historical corridor involves irregular trafficking-adjacent routes respondents may be reluctant discuss; survey instrument Appendix deliberately worded focus remittance receipt and use rather than migration circumstances to mitigate. Geographically concentrated Edo State may not generalise to states different migration histories.

Operational Definitions

Remittances: Money or in-kind transfers sent by international migrant to family remaining origin measured per capita monthly Naira value received - 71.2% of 340 households report receipt.

Household Welfare: Economic wellbeing measured through per capita monthly expenditure consistent standard practice Nigerian and international literature.

Migration Network: Share households in respondent community that had international migrant as of historical 2010 baseline used as instrument for current remittance receipt grounds established networks lower cost risk future migration and remittance transfer for other households same community without directly affecting those other households current welfare - McKenzie Rapoport 2007 instrument design.

Endogeneity: Situation explanatory variable interest here remittances correlated with error term outcome equation due unobserved confounding reverse causality measurement error biasing OLS - naive OLS 0.757 potentially upward biased.

Instrumental Variable: Variable correlated with endogenous explanatory variable relevance but uncorrelated with outcome equation error except through effect on endogenous variable exclusion restriction used to obtain consistent estimate via 2SLS - community density and distance to Lagos joint F 83.49 p0.001 strong.

Two-Stage Least Squares: Technique endogenous regressor first regressed on instruments and exogenous controls first stage outcome then regressed on fitted values second stage yielding consistent estimate under valid instruments - 2SLS 0.679 p0.001 somewhat below OLS expected direction Wu-Hausman 0.922 p0.338 Sargan 0.308 p0.579 valid; MPC 0.68-0.76.

Conclusion

Naive OLS benchmark coefficient 0.757 p0.001 per capita remittances in per capita expenditure equation. First-stage confirms both instruments strong correctly signed joint F 83.49 p0.001. 2SLS estimate 0.679 p0.001 somewhat below OLS in theoretically expected direction though Wu-Hausman statistic 0.922 p0.338 does not provide statistically decisive evidence of endogeneity in this sample and Sargan overidentification statistic 0.308 p0.579 does not reject instrument validity. Concludes remittances make statistically and economically significant robust contribution to household welfare Edo State estimated marginal propensity to consume out of remittances approximately 0.68-0.76 and recommends continued investment in formal low-cost remittance-transfer infrastructure and diaspora-engagement programming recognising substantial welfare contribution.

FAQs

What is incidence remittance receipt Edo State sample?

340 households 71.2% report receipt leading migrant-sending state Nigeria well-established corridor to Europe.

What is naive OLS association remittances welfare?

Coefficient 0.757 p<0.001 on per capita remittances in per capita expenditure equation uncorrected benchmark potentially upward biased due endogeneity.

What instruments address endogeneity?

Community migration-network density share households community with international migrant as of 2010 baseline and household distance to Lagos proxy migration logistics cost following McKenzie and Rapoport 2007 network instrument design.

Are instruments strong and valid?

First-stage joint F 83.49 p<0.001 confirms strong correctly signed; Sargan overidentification statistic 0.308 p0.579 does not reject validity supporting exclusion restriction that network density and distance affect welfare only through remittances.

What is causal 2SLS effect?

2SLS estimate 0.679 p<0.001 somewhat below OLS 0.757 in theoretically expected direction; MPC out of remittances approx 0.68-0.76 robust significant welfare contribution.

Is remittance receipt endogenous?

Wu-Hausman statistic 0.922 p0.338 does not provide statistically decisive evidence of endogeneity in this sample failing to reject that OLS and 2SLS not distinguishable though theory suggests potential endogeneity.

What is New Economics of Labour Migration framework?

Framework viewing migration household decision not individual to diversify income and overcome market failures remittances part of household strategy - grounds Edo study.

What are policy implications?

Continued investment formal low-cost remittance-transfer infrastructure and diaspora-engagement programming recognising substantial welfare contribution for NiDCOM CBN development partners anti-trafficking safe-migration programming weighing economic benefits vs irregular route risks.

What are limitations?

Cross-sectional not dynamic cumulative effects; self-reported recall social-desirability bias particularly irregular trafficking-adjacent routes mitigated by focusing questionnaire on receipt and use not circumstances; IV exclusion restriction cannot be definitively proven only indirectly via overidentification test; geographically Edo only may not generalise.

How does study extend prior Nigerian literature?

Extends Chukwuone et al 2008 IV-based poverty evidence and Obisie-Nmehielle Ike 2017 welfare evidence OLS probit without explicit IV correction with fully diagnosed community-network IV strategy specific to leading migrant-sending state transparently reporting first-stage strength overidentification and endogeneity tests.