Refugee policy is not exempt from evidence-based decision-making.

Just because it’s harder doesn’t mean we shouldn’t do it.

There’s been a lot of noise around the topic of immigration lately, in response to a few EOs from the desk of President Trump.  It’s fair to say that the EOs were met with a lot of emotion from all points on the political spectrum.  Indeed, in many conversations, from those on Facebook to the White House Press Conferences, it was emotion — not evidence — that was driving the dialogue.

Why the lack of evidence?  

First answer: there’s plenty of it, but a lack of will to use it.  It’s not hard to fact-check any of the assertions made in the last few weeks.  The challenge for the next four years and beyond is going to be creating a non-partisan culture of demanding to see the cited facts that support the rhetoric — and then to share those (did I mention cited?) facts respectfully with a purpose of collaboration toward an evidence-based decision.

My second answer addresses not the EOs specifically, but the broader political conversation that the EOs have stirred up: whether or not refugees/immigrants are harmful for the receiving country.  Taylor and colleagues (2016) explain the lack of evidence for this question as having three causes:

  1. The lack of before and after data due to the fact that refugee migration is often unexpected
  2. Impacts of refugees on host countries are varied and complex (and I would like to add: far-reaching)
  3. And, largely as a result of the first two points above, more conventional experimental evaluation approaches are inapplicable

But the authors tackled the challenge anyway, with a method called local economy-wide impact evaluation (LEWIE) and a hint of Monte Carlo simulation.

The authors specifically looked at three Congolese refugee camps in Rwanda.  Researchers conducted surveys of randomly sampled households and businesses around and inside the refugee camps.  They modeled the economic behavior (expenditure, consumption, etc.) of refugees separately from that of the host country, and then compiled it into a single model of the economy which is potentially impacted by refugee resettlement.  From that model, they could predict the impact that adding an additional refugee would have on that economy, given that the refugee received cash aid.

If this doesn’t seem rigorous enough to you yet, you’re right in being skeptical.  At the moment, the results are entirely dependent on the one model, which could be heavily influenced by the population they chose to survey.  To remedy this, researchers used Monte Carlo simulation to re-construct the model 1000 times.  That is, given the characteristics of the refugee and host-country households, they created a random new set of 1000 other possible models.  So then if the “impact” estimate falls within the results of 95% of these models, it is very likely to be true.

In this case, the impact was very positive: “an additional adult refugee receiving cash aid increases annual real income in the local economy by $205 to $253, significantly more than the $120-$126 in aid each refugee receives.”  The authors go on to explain that “most of the difference consists of income spillovers resulting from market interactions between refugees and host-country businesses and households.” These numbers fall within broader 95% confidence intervals which were constructed using Monte Carlo methods.

Of course this study should be applied cautiously (or not at all) to the context of the U.S.  It is much more applicable to the majority of cases in which refugees are resettled in other low-income countries.  But it is a good case-study for how we can adapt our impact evaluation models to fit the needs of the policy question at hand — even expanding the outcome of interest to something beyond questions of economy.

The World Bank also has an interesting report from 2011 which includes case studies of successful programs which mitigated potentially negative impacts of refugees on host-countries.  Again, this report is most applicable to low-income countries (in this case, it focuses on countries neighboring those sending refugees), but it provides a good road map of how we could structure our research and start building evidence for decisions in the U.S. context.

Perhaps next time that current events spark a broader debate on whether refugees are harmful for the U.S. as a host country, we will have a wider pool of evidence from which to draw.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s