The Soccer War was not about soccer. Rather, it was about Salvadorean immigration to Honduras. In 1967, the Honduran government began to expropriate Salvadorean properties and deport Salvadorean immigrants. We do not know how many were deported before the fighting began, but the estimates range between tens of thousands and 100,000. El Salvador invaded to stop the expropriations and deportations. In the peace settlement, Honduras agreed to exactly that.
Problem is, we do not know much about Salvadorean immigration to Honduras before the war, nor what happened to the Salvadoreans afterwards.
Here is what we know. The 1961 census recorded 38,002 Salvadoreans, although the IBRD estimated that the number was actually around 100,000 as early as 1950 ... although that number includes Honduran-born children. It is high compared to the census estimate of 20,285 in 1950, but given birthrates it is not that high: a third of the migrants in 1950 had arrived before 1944; a number around 60,000 in 1950 is quite reasonable.
The war happened in July 1969 and was accompanied by a large flow of Salvadoreans back across the border. El Salvador reported that 81,159 fled during 1969. (See page 59.) That number, however, includes Honduran-born children of the immigrants. In August 1969, the Honduran government reported that a special count of Salvadoreans had uncovered 154,453 people. (See the bottom of page 16 here.)
There is a problem: the 1974 census reported that there were only 100,743 foreigners in the country in 1969. Can that number be reconciled with the count of 154,453 Salvadoreans? The answer is yes.
First, we do not know but it is possible that the larger count includes Honduran-born children. The secondary sources that I have seen imply otherwise, but it is easy to imagine how what we would today call an untrue meme could enter the scholarly literature.
Second, there is a timing issue deriving from how the government produced the 1969 estimate five years later. We do not know the number of foreigners in-country in 1974; the government collected the data but never published the aggregate figures. The March 1974 census schedule, however, asked where you had lived five years earlier. (See question 9 on page 2.) The estimate of the number of foreigners living in-country in 1969 comes from that data. In other words, of the number of foreigners living in Honduras in 1974, 100,743 of them had also been there in 1969. The implication is that even if you do not believe that the 154,453 estimate includes Honduran-born children, it is consistent with the 100,743 number as long as you believe that a lot of Salvadoreans left between August 1969 and March 1974.
Did a lot of Salvadoreans leave during that period? As mentioned, we know around 100,000 left around the war. (The same Salvadorean agency estimated that the ultimate figure came to around 130,000.) We also know that there were only 7,733 Salvadoreans living in El Salvador in 1988. (That figure did not include the 13,325 living in refugee camps, according to page 18 here. I visited one of those camps at the time and am now furious with my younger self for losing the negatives.) Now, it is possible that the decrease was due to natural attrition. But the immigrants would have had to have skewed improbably old in 1969 for most of the them to be dead by1988, so it is not likely.
The decrease also might have been due to naturalization. On page 16, this American document implies that the Honduran census figures recorded country of citizenship, not country of birth. That would be a problem, because Honduras has always had liberal naturalization laws for Central Americans. Article 18 of the Constitution of 1957 allowed Central Americans to obtain Honduran citizenship by a simple administrative procedure after one-year in-county as long as their home country reciprocated. Article 16 of the Constitution of 1965 said the same thing. Article 24 of the current Constitution of 1982 changed the wording again but kept the substance. The Salvadorean Constitution of 1962 granted the necessary reciprocation in Article 12. I have not checked the census volumes, but the good people at CEPAL have and their discussion on page 163 implies that the figures refer to place of birth. (I will check; if you hear nothing more then the CEPAL people are right.)
The upshot is that it appears that a lot of Salvadoreans fled Honduras in the five years after the war, despite the putative victory settlement that allowed them to stay.
But there is still a problem with that conclusion: Salvadoreans might have simply started to lie to the census-takers. The accents are close, the racial mixes near-identical: the nations are at least as similar as the United States and English Canada and closer than England and Scotland.
(Side note: Canadians have a very distinctive accent! Which also happens to be identical to the one spoken in Western New York. Which is a quite distinctive accent compared to the rest of America ... but Rob Ford could credibly claim to be from Buffalo. Should masses of desperate Canadians ever need to flee to the United States, they can pretend to be from Erie County.)
At this point I give up for a blog post.
To recap: El Salvador fought a war to stop the deportation of its nationals from its neighbor. It more-or-less won the resulting war. But nevertheless the Salvadorean population in Honduras disappeared in the next decades anyway despite civil war in El Salvador. What we do not know is if that population decided to hide in plain sight or went back home.
It is a question worth investigating.
UPDATE: I significantly edited this a post on Wednesday afternoon.