First things first: The question is about how many Eritrean citizens will apply for asylum, not how many will be granted asylum. Obviously the latter will depend heavily on the former. According to the Council on Foreign Relations, “EU countries grant asylum to about 90 percent of Eritrean asylum seekers,” which was higher than I expected. Switzerland (7,475), Germany (5,500), and Sweden (4,645) took the lion’s share in 2015[1].

Now, a little recent history: The number of Eritrean asylum applicants went up significantly in 2014, from 14,485 to 36,925[2]. That’s an increase of 154.9%. Why? Violence in South Sudan increased the total number of asylum seekers in east Africa[3], but Eritrea doesn’t share a border with South Sudan. No, the main problem seems to be internal to Eritrea – in particular, the massive and ongoing violation of human rights within the country, according to the United Nations Committee on Human Rights[4]. Here’s the Council of Foreign Relations’ take on it:

The UN commission of inquiry found that national service often entails “arbitrary detention, torture, sexual torture, forced labor, absence of leave, and the [sic] ludicrous pay,” calling it “an institution where slavery-like practices are routine.” For many, leaving national service is only possible by deserting the military and fleeing the country, the commission found[5].

At any rate, the number of applicants leveled off in 2015, though it remained much higher than in 2013. I’ve seen this number estimated as 37,000[6].

So how many asylum seekers should we expect in the first quarter of 2016? Nothing much has changed in Eritrea, so we might expect 2016 to be a lot like 2015. One could just divide this number by 4 in order to get a (very rough!) estimate for Q1 of 2016: 9,250. However, it’s worth noting that there seems to be a seasonal component to the number of asylum seekers from Eritrea. The number of applicants during Q1 of 2015 was much lower than you’d expect if you simply divided the total number of asylum seekers for the year by the number of quarters. In fact, asylum seekers in the first quarter of 2015 numbered a mere 3,155. In contrast, the number of asylum seekers skyrocketed around Q3 in both 2014 and 2015. I suspect that’s because the weather during this time of year makes it easier to travel to Europe. That’s just speculation on my part, but I note that a very large number of asylum seekers land in Sicily, and a sea crossing during the summer months is probably a better bet. Indeed, the number of asylum seekers – regardless of nation of origin – in Europe is lowest during Q1[7].

Anyway, here’s a thought: It might be better to average these two data points in order to get an expected number of asylum seekers for Q1 of 2016:

- ((37,000)/4) + (3,155))/2 = 6203 (rounding up to the nearest whole number)

Given the paucity of data, that’s *not* *crazy*. And *not crazy* might be good enough. The standard deviation for this (tiny!) data set is 4,310 (rounding up to the nearest whole number). That means that there’s about a 74% chance that the number of asylum seekers will be between 0 and 9,999, and a 17% chance that the number will be between 10,000 and 15,000. That gives us a picture of Q1 that looks roughly like this:

- 74% Less than 10,000
- 17% Between 10,000 and 15,000, inclusive
- 4% More than 15,000

However, there is a little more data available. As of January 31, Europe had received 6,029 refugees by sea. Recently, 25% of refugees of this sort have been Eritrean[ibid.]. Assuming that all of them are asylum seekers, that would mean that one third of the way through Q1 there have been about 1,500 Eritrean asylum seekers. That’s very close to the number for January 2015: 1,295. So I’ll use a weighted average, instead of an unweighted average, giving more weight to Q1 of 2015 and less to the average of the quarters of 2015:

- .33x(37,000)/4) + .67x(3,155) = 5167 (rounding up to the nearest whole number)

The standard deviation for this data set is 3,519 (rounding up to the nearest whole number). That means that there’s about a 84% chance that the number of asylum seekers will be between 0 and 9,999, and a 8% chance that the number will be between 10,000 and 15,000. Hence,

- 88% Less than 10,000
- 9% Between 10,000 and 15,000, inclusive
- 3% More than 15,000

If the numbers continue to trend this way over the next few weeks, then I’ll modify my forecast, but that’s all I have for now.

Finally, let me just say that a lot of other very smart people did much of the leg work gathering the data I’m talking about here[8]. If I get close to the truth, it’s mostly because of there hard work, not mine.

[Post updated for typos and a earlier failure to take into consideration the impossibility that the number of asylum seekers will be less than zero!]

[1] http://www.cfr.org/eritrea/authoritarianism-eritrea-migrant-crisis/p37239

[2] http://ec.europa.eu/eurostat/statistics-explained/images/d/d5/Countries_of_origin_of_%28non-EU%29_asylum_seekers_in_the_EU-28_Member_States%2C_2013_and_2014_YB15_III.png. (See also http://www.unhcr.org/546606286.html)

[3] http://www.unhcr.org/pages/49e4838e6.html

[4] http://www.ohchr.org/EN/HRBodies/HRC/CoIEritrea/Pages/ReportCoIEritrea.aspx

[5] http://www.cfr.org/eritrea/authoritarianism-eritrea-migrant-crisis/p37239

[6] http://ec.europa.eu/eurostat/statistics-explained/images/f/f1/First_time_asylum_applicants_in_the_EU-28_by_citizenship%2C_Q3_2014_%E2%80%93_Q3_2015.png

[7] http://data.unhcr.org/mediterranean/country.php?id=105

[8] https://www.gjopen.com/questions/110-how-many-eritrean-citizens-will-apply-for-asylum-in-europe-for-the-first-time-in-the-first-quarter-of-2016