The Northern Triangle is one of the poorest parts of Central America, comprising of Guatemala, Honduras, and El Salvador. The people living here face many challenges, ranging from a high homicide rate to high levels of poverty. What is the most effective way for the United States to help them?
There are two ways. The first is providing aid and the second is offering refugee status to immigrants. The United States has a fixed amount of money to devote to this problem. Thus, it is instructive to ask — which way is more cost effective? We analyze the cost effectiveness of both options below.
Offering Aid
Cumulatively, the U.S. Agency for International Development (USAID) has given 4.4 billion to El Salvador [1], and 3 billion to Honduras [2]. I could not find a number for Guatemala but I assume it is similar. USAID programs generally focus on reducing crime, improving education, strengthening political governance, and they appear to be effective. For instance, a $40 million dollar Crime and Violence Prevention program enacted in El Salvador focused on 64 municipalities, established 168 youth centers, and provided tutoring and vocational skills to 40,000 children [3]. A Crime and Violence Prevention program that spanned the 3 countries (I could not find the exact cost) led to a 51% drop in murders, extortion, and blackmail for the treated municipalities, according to a 2014 randomized control trial [4]. This is quite a strong effect.
Offering Refugee Status
There are 1.79 million legal immigrants in the United States from Central America that do not have a high school degree [5]. For context, the average education level in Guatemala, El Salvador, and Honduras is 5.5 years, 7.6 years, and 6.5 years respectively [6]. This is substantially worse than Mexico, where 45% of individuals finish high school. In addition, individuals that seek refugee status tend to be less educated members of the population. The net present value cost of supporting an immigrant without a high school level of education ranges from 117,000 to 295,000, depending on whether or not welfare benefits are reduced in the future and whether immigrants bear costs for public goods.
This is indicated by Table 8-12 of the National Academies report on The Economic and Fiscal Consequences of Immigration [7]. This is a 500 page report conducted by the National Academies of Sciences, and is the only peer-reviewed report that estimates the net present value cost of immigration by education level. Note that the cost of supporting a native who has not completed high school is 100K greater, as the native is entitled to more benefits. The substantial cost of supporting individuals without a high school degree reflects the nature of the U.S. economy; low-skill jobs simply do not pay enough for one to support oneself.
Using the range specified above, the net present value cost of supporting Central American immigrants who lack a high school education is $209 billion to $528 billion. Note that this is several orders of magnitude larger than the amount of aid given to Central America. There are 700K immigrants in this category from El Salvador, meaning that the United States has spent 82 - 207 billion dollars to to support these immigrants. By contrast, the population of El Salvador is currently 6.5 million. Thus, the lower range is equivalent to giving each person in El Salvador $12,600 to $31,600 dollars. The lower end of this estimate is 4 times the median income of $3,600. [8]
What about unauthorized immigrants? The costs are reduced here, as the only federal benefit these immigrants receive is the Child Tax Reconciliation Credit (some states choose to provide additional benefits). Through this credit, unauthorized immigrants received $11.5 billion last year but also paid $9 billion in taxes [9,10]. All unauthorized children are entitled to receive a public school education, which amounts to $15,000 per pupil per year, on average [11]. The National Academies report estimates that there is a significant net present value cost to paying for the descendants of those without a high school education. This cost ranges from 8,000 to 116,000, depending on the scenarios described above (Table 8-12). This is because children of immigrants have not completed high school have a 17% chance of not completing high school themselves, compared to a 0.1% chance of not completing high school for children of immigrants have a bachelor’s degree (Table 8-8). Children of immigrants actually have substantially better educational outcomes compared to children of natives with the same level of education. These percentages reflect the reality of the current education system, where it is difficult for children to thrive if their parents are not well-educated.
My point is quite simple. Supporting individuals without a high school education, whether they are natives, legal immigrants, or unauthorized immigrants, is quite expensive. This is not the individuals’ fault. It is simply the nature of the economy and the education system.
Comparison of Cost-Effectiveness
Offering aid is far more cost effective than supporting additional immigrants. For only $40 million dollars, one can support 341 people without a high school education or reach 40,000 children through a USAID program. Aid represents an effective, scalable way to improve the outcomes for people in the Northern Triangle. It leads to a more equitable distribution of money compared to the status quo, where large amounts of money are spent on the select few who make it into the United States. Indeed, USAID has already reduced violence by 50% in treated municipalities. Additionally, the population of countries in the Northern Triangle is rapidly growing. For instance, the population of Guatemala increased by 17% in the last 10 years. Investing in aid can slow this population growth as well as benefit future generations, making it additionally cost effective.
Taking in large numbers of immigrants also has two unintended side effects.
Firstly, migration is driven by “push” effects (violence in native country) as well as “pull” effects (family and peer networks in the United States). An increase in “pull” effects can lead to an increase in migration even when violence reduces. For example, some of the regions in El Salvador with the highest amounts of unaccompanied children (e.g. Sierra Madre de Chiapas mountains of northern El Salvador) do not have the highest levels of violence [12].
Secondly, the promise of entry to the United States incentivizes individuals to put themselves through dangerous circumstances. Migrating is a long and dangerous process, and can sometimes be fatal. Individuals pay around 6,000 to 10,000 dollars to human “smugglers” to have a chance of entry [13]. Individuals detained at the border suffer from poor living conditions, and the scale of the problem has led to large numbers of missing children [14] Migrants are simply trying to ensure a better life for themselves and their children, and many of us would behave similarly if we were in their circumstance. The fault is with the system for providing perverse incentives.
The status quo is tragic for all parties. Americans are spending large amounts of money on solutions that are not effective. Central American countries continue to be plagued by violence and instability, and those who migrate face a difficult, expensive journey as well as poor living conditions during detainment.
The United States has a fixed amount of money that can be used to address this problem. Thus, we should seek the most cost-effective option. The solution is to dramatically increase aid to Central American countries and also shut down the US-Mexico border. This is less extreme than it sounds; the border was closed for all nonessential travel during Covid-19. This solution helps all parties. Firstly, living conditions in Central American can improve as a result of additional targeted aid. Secondly, increasing aid levels is still several orders of magnitude cheaper than supporting individuals without a high school education. Thirdly, individuals aren’t incentivized to risk their safety and face poor living conditions in detainment.
The main negative impact would be to the Mexican economy, especially border towns. This can be offset by more aid—the United States has spent $209 to $528 billion assisting immigrants but has spent ~$3 billion cumulatively for each country in Central America. Shutting down the border is so cost effective that we can afford to dramatically increase aid to Mexico and Central American countries. Additionally, important cargo could be transmitted via air or sea.
My main point is this. We should focus on aid as the main pillar of international development, as opposed to taking in immigrants with limited education. This is vastly more effective, and avoids unintended side effects. Currently, however, the opposite is happening.
[1 ]https://www.usaid.gov/el-salvador/frequently-asked-questions
[2] https://www.usaid.gov/honduras/history
[3] https://www.usaid.gov/sites/default/files/documents/1862/Fact_Sheet_-_USAID_Crime_and_Violence_Prevention_Project.pdf
[4] https://pdf.usaid.gov/pdf_docs/pbaab431.pdf
[5] https://www.migrationpolicy.org/article/central-american-immigrants-united-states
[6] https://drupalwebsitepardee.s3-us-west-2.amazonaws.com/pardee/public/The+Future+of+Guatemalan+Education.pdf
[7] https://icfdn.org/educational-snapshot-mexico-infographic/
[8] https://www.statista.com/statistics/1070121/gross-national-income-per-capita-el-salvador/
[9] https://cis.org/Camarota/Illegal-Immigrants-Get-105-Billion-Child-Tax-Credit-Reconciliation-More-Previously
[10] https://bipartisanpolicy.org/blog/how-do-undocumented-immigrants-pay-federal-taxes-an-explainer/
[11] https://nces.ed.gov/fastfacts/display.asp?id=66
[12] Violence, development, and migration waves: Evidence from Central American child migrant apprehensions
[13] https://qz.com/1632508/this-is-how-much-it-costs-to-cross-the-us-mexico-border-illegally/
[14] https://cis.org/Arthur/Biden-Administration-Lost-Yes-Lost-Nearly-20000-Migrant-Children