Effect of Remittances on Mexican Financial Inclusion: Analytical Essay

Introduction

International remittances have become a particularly prominent source of development finance since the turn of the century, dwarfing official international foreign aid figures in comparison. Key literature on this issue highlights the debate surrounding its impact on financial inclusion, which itself concerns the provision of a full range of quality financial services universally that is effectively utilised by its consumers (Copestake, 2017). Wage differentials between receiving and sending countries is one of the most conspicuous contributors to the migration and remittance mechanism, consequently giving rise to the Mexico-United States corridor. Of the estimated $440.5bn remittances received by developing countries in 2015, $25.2bn of this was via this corridor (World Bank, 2016). While statistics on this topic are readily available, several unquantifiable elements mean it has proved more challenging to thoroughly analyse the direct effect on Mexican financial inclusion and how this filters through to its social and economic development.

This paper scrutinises existing studies and aims to evaluate whether remittances received by Mexico significantly affect its population’s financial inclusion, relating specifically to the saving, borrowing and investment opportunities of receiving households. Likewise, it is relevant to cross-examine the geographical, educational and income differences of recipients. This is a crucial issue to understand due to the significant benefits of having an inclusive financial system; subsequently the lack of this can contribute to persistent income inequality and slower growth over time. It also highlights the importance of implementing appropriate monetary policy in order to maximize the benefits of the recipient. As previously mentioned there have proved to be multiple unquantifiable elements, thus it becomes necessary for studying to incorporate both quantitative and qualitative aspects when conducting analysis.

Context

While the overall consensus of the topic’s literature upholds the belief that remittances are beneficial to the receiving country, many development economists maintain the view that notable drawbacks hinder the progress of emerging economies. One such issue is the ‘Brain Drain’, which refers to the migration of skilled workers from developing countries into MEDCs, usually to reap the economic benefits in the developed world by taking advantage of income opportunities. One line of argument is that the brain drain from poorer countries robs them of the personnel they require to escape poverty (The Economist, 2018). Having said that, only 13% of Mexican migrants received a tertiary education (Carrington and Detragiache, 1998). This number pales in comparison to figures of migrants from the Eastern Hemisphere, suggesting this should be a more pressing issue elsewhere. Other potential negative effects of remittances include the Dutch Disease and moral hazard, which can contribute to weak labour force participation (Hudson, 2014). Despite this, a variety of empirical studies and papers have highlighted the contributions of remittances to developing countries on a macroeconomic level through their impact on poverty reduction and economic growth (e.g. Adams, 2011; Cohen, 2011).

There is a well-documented analysis of the link between remittances and economic growth through extensive worldwide case studies, however less research has been conducted on the effects on financial inclusion of individual households. Therefore, this case study focuses on evaluating the traditional theoretical framework of migration and remittances and uncovering the relationship between remittances and financial inclusion of households receiving these remittances from abroad, relating specifically to their saving, borrowing and investment opportunities. Statistics show that Mexico is a particularly relevant country for examining this relationship through both the prominence of remittances yet simultaneously the lack of financial access for the average household. At 2.3% of its GDP (World Bank, 2016), remittances make up the second largest source of its external financing (trailing only to oil exports), a figure that has been mostly increasing since 1979. The same journal conducted a study in 2015 which revealed only 44% of Mexican adults owned a bank account. This indicates the saving opportunities of Mexican households are limited simply to storing cash, in turn preventing record keeping (e.g. credit scores) which could be important for securing future loans for investment opportunities. Instead, the relative absence of links to the financial sector restrict any investment to fixed assets such and land and cattle. The introduction of Mexico’s National Financial Inclusion Strategy in 2016 shows the relevance of the issue today and provides an interesting outlook as to how policymakers can influence financial inclusion. Successfully linking remittances with financial services can have a positive effect on both remittance-receiving households and countries, which is why policy surrounding this has become such a high priority (Orozco, 2004; Terry and Wilson, 2005).

Key Literature

Ambrosius (2016) develops a household level study with the aim of uncovering the indirect effects of remittances on receiving economies through links to the financial sector. In this context, remittance receiving households are defined as those in which at least one member received a transfer from a USA residing family member during the last year.

The paper draws evidence from the Mexican Family Life Survey (MxFLS). This combines data on migration history, monetary transfers and financial services access & usage among others, with a large enough sample size to be nationally representative and account for various differences. Using a survey showing that from 2002-2005 7.8% of households receive remittances in rural communities (defined as having less than 2000 habitants) compared to just 4.8% in urban communities ensures standardization is a necessity.

How financial access is characterised can be interpreted differently depending on access to credit options, saving options and commercial banks among other factors. Ambrosius measures access to financial services using just two key indicators; whether or at least one household member has a bank account and the availability of credit from financial institution. Notice that only the availability of borrowing options is considered rather than their actual use, which rigidly distinguishes financial access from the demand of financial services. Using this, financial access is found to be particularly restricted in rural areas, with only 9% of rural households owning a savings account and 22% with borrowing options, compared to 22% and 36% in urban areas respectively. One problem faced was accounting for several non-financial differences between households regarding number of children, housing conditions, employment situation and education levels. Combatting this, Schreiner’s Index (2011) was used to aggregate information into a ‘poverty score’ from 0 (lowest probability of falling below poverty line) to 100 to allow for the controlling of these characteristics. Following this, spending per capita per month was used as an indicator of households’ income and included in the regression. Binary dummy variables are then used for the two measures of financial access, with a value of ‘1’ given to households who fulfil the criteria in a given time period. Ambrosius then uses a panel logit (fixed effects) model followed by a random effects model in order to analyse the effect of received remittances on financial access.

Two key findings emerge from the regression results. There is a large and statistically significant coefficient for remittances with respect to households owning savings accounts, with a larger coefficient produced by the fixed effects model compared to the random effects model. More specifically, sizeable differences between various geographical and socio-economic groups are exposed when observing the correlation between remittances and ownership of household savings accounts. Firstly, the probability of owning a savings account is increased to a far larger degree for poor households in comparison to richer households. Similarly, dividing into rural and urban population sees again a large coefficient for remittances on households for those living in rural areas, while statistical significance on urban households only occurs at the 10% level. Despite this, rural households generally have ownership of savings accounts in the microfinance sector rather than commercial banks, raising the topic of the potential for a government-led initiative to encourage financial institutions to further branch out to rural communities. Moving on to access to borrowing options, the relationship becomes slightly more ambiguous. This is to be expected as commercial banks are unlikely to accept informal income such as remittances as a means to facilitate access to credit. Therefore, the only statistically significant result is found for borrowing from microfinance institutions as opposed to banks, which is again higher for rural households. In summary, a supportive view is provided for the hypothesis that there is a positive correlation between remittances and financial inclusion.

When assessing the impact of remittances on financial access, there is an issue of reverse causality between variables (Posso, 2015). Ambrosius provides two example of how this could occur in the context of this study. Migration costs could theoretically be financed through loans from financial institutions, which would appear in the results as being correlated with remittances. An argument against this alternative interpretation would be that institutions are unlikely to be willing to lend in a situation where repayment could be difficult to enforce. Secondly, it could be argued that access to financial services simply makes receiving remittances easier, however previous literature suggests this only causes the adaptation of sending behaviour as opposed to the prevention of remittances being sent.

Discussion

Ambrosius’ journal’s main strength lies in its relevance to this case study, as it is useful in showing the impact of remittances on financial inclusion by providing empirical evidence from the models. However, the reviewed journal is limited in its scope regarding this case study. Despite its usefulness in empirically assessing correlations, the paper provides less insight into how these are reflected in the wider economy. This case study aims to assess the reasons behind the importance of the correlation as well as the ways in which it can be exploited by policymakers for the benefit of households. Ambrosius (2016) is limited in delivering any qualitative analysis and theory backed up by further research used for policy recommendations.

The results of this study supplement the theory that a demand for savings accounts emerges from the need to store cash when remittances are received. Several theories have been put forward by other development economists that could explain the inconclusive results on borrowing opportunities. One of these is that the interest payments on loans from microfinance institutions tends to be considerably high (higher than commercial banks). Recalling that Ambrosius finds access to credit from microfinance to be far more abundant than commercial banks in rural areas, income from remittances may be used as a substitute in order to avoid these interest payments (Hudson, 2014). It is worth noting that Ambrosius’ finding concerning the distinct lack of commercial banking sector presence in rural areas could stem from their higher poverty rates, thus increasing the perceived risks surrounding loan delinquency. From these results rise an opportunity for Mexican policymakers to impose and adapt policy to achieve three main objectives. The first is to incentivise beneficiaries of remittances to take advantage of the saving and borrowing access they currently have. The second is to probe the commercial banking sector to venture further into more rural, poverty-stricken areas. There is empirical and qualitative evidence to suggest causality between financial inclusion and inclusive growth at the individual, micro level (Demirgüç-Kunt, 2017). Linkage between financial inclusion and growth on a macroeconomic scale is less researched, however certain papers have documented empirical evidence of a close positive relationship in financially developing countries (e.g. Kim, Yu and Hassan, 2018). Lastly, studies have shown a remittances have promoted growth in developing countries by providing an alternative way to finance investment (e.g. Giuliano and Ruiz-Arranz, 2009), so setting policy around increasing the overall level of remittances should be a priority.

Airola (2007) discovered that a high proportion of remittances sent to Mexico are being spent on healthcare, housing and durable goods. While investment in durable goods is a step in the right direction, money spent on healthcare and housing could suggest that remittances are sent home to family members as a form of aid when desperately required. Theoretically, if the Mexican government were able to find the funds to allocate towards public housing and healthcare, a higher proportion of remittances could be saved or invested into durable goods. Doing so would reduce the need to spend, but further policies could be introduced to raise the opportunity cost of immediate consumption by increasing average return on investment.