Robertas Zubrickas is a Senior Research Associate at the University of Zurich.
Over the past decade, there has been an almost exponential rise in international remittances. We know from recent research that remittances are critical for the well-being of individual households in developing countries – helping them to emerge from poverty, send their children to school, and invest in small enterprises, health, education and housing. Yet not much is known about determinants of remittance flows within transnational households (those with one or more members working abroad), an increasingly important topic for policy makers with the sums involved.
Given that the geographical separation might not allow the wife to be fully aware of her migrant husband’s earnings abroad, and given that a discrepancy in information matters for economic outcomes, the question arises: Does awareness matter for remittances? In particular, will the husband remit less if his wife cannot observe his earnings? Our study “Asymmetric Information about Migrant Earnings and Remittance Flows” (done jointly with Ganesh Seshan), which focuses on Qatar and India, shows that the answer is “absolutely”, with higher earners benefiting more from the lack of awareness than lower earners.
From Qatar to Kerala
Migration into Qatar and other oil-producing countries of the Arabian Gulf took off in the 1970s with rising oil prices that subsequently fueled a large construction boom. In fact, in 2011 there were an estimated 17 million contract workers in the Arabian Gulf countries alone, mostly from developing countries. For our study, we decided to focus on Qatar, where about 90% of the 1.7 million population age 15 or older are foreign born, rendering it the nation with the highest share of immigrants in the world. And we picked the Indian state of Kerala, which until recently accounted for more than half of the Indian migrants to the Gulf.
To assess differences in information about overseas earnings, we collected migrants’ reports about their earnings and contrasted these reports with the reports about their earnings collected from the remittance recipients. To our knowledge, this is the first study using a matched household dataset that collects such cross-reports on remittance flows. Our key findings are twofold:
Observation 1: More earnings, more discrepancy in information. Wives report on average only about 79% of their husbands' earnings, which signifies a substantial discrepancy in information. Moreover, underreporting is not uniform across households – it is more prevalent in households with higher earning migrants. In particular, we observe the gap between wives’ and husbands’ reports to widen in husbands’ earnings (see figure below).
Observation 2: More discrepancy in information, less remittance. Underreporting is associated with lower remittances. In the sample, a wife who understates her husband’s earnings by the average amount receives 15 percent less in annual remittances than a wife who has perfect information about her husband’s earnings. Hypothetically, closing this information gap about foreign earnings would be associated in India with an increase in annual remittances of $432 – or nearly two months worth of monthly household expenses. Furthermore, data show that husbands remit on average 58 cents from every dollar of below-the-median income, but only 17 cents from every dollar above the median income.
Work Abroad as a Family Investment Project
What explains our observations? First, we are able to rule out the explanation that differences in information arise from subjective biases in reporting behavior. A more probable and natural explanation for differences in information – as also suggested by the evidence on remittance behavior of Tongan migrants to New Zealand – is that husbands tend to hide overseas earnings from their wives to reduce pressure to remit. But harder questions are why husbands are more truthful about their income when it is low and why they remit a larger share of a lower income. To answer these questions, we approach migration as a family investment project. A family sends its member abroad for work and expects the migrant to remit a share of earnings, which, however, are observable by the family only if it tries to verify the report (for example, by making inquiries in a migrant network).
When verification costs are not trivial, the most efficient “expectation” for remittances that the family can impose on the migrant takes the form of a simple remittance threshold – which is what occurs in the most efficient loan contract if a costly audit is involved. Namely, if the migrant remits less than the remittance threshold on his account of low income, then the family can take the effort to verify his reported account of income (and punish him if he is found lying). On the other hand, no verification is undertaken if the migrant remits at least the threshold. Thus, it implies that the migrant, once capable of meeting the minimum expectation for remittances, may choose to be silent about any additional income earned and send no additional remittance thereof. As this implication is strongly supported by our empirical observations, it suggests that the same economic arguments that apply to investment contracting also apply to forming expectations for remittances and actual remittance behavior.
Our work signifies the role of information for remittance flows by demonstrating a direct connection between how much families receive in remittances and how much they know about foreign earnings. As more awareness benefits remittance recipients and, accordingly, their countries, a desirable policy is to improve it. Particular measures at the disposal of policy makers in remittance-receiving countries would be improving financial literacy and means of communication.