Both Australia and Canada practise a Skills-based immigration policies. In Australia in paticular, the children of immigrants do better at school than native children. Teriary educated immigrantion applicants are far more likely to be accepted, and seems to have a large effect on PISA scores.
http://1.bp.blogspot.com/_-EMpadQx4hM/TRKZ5kB-49I/AAAAAAAAAY...
If the immigrants were exactly like the natives, their school results would be exactly the same as well. Or, if the school was perfectly adapted to the immigrants, they would score the best in the world.
In reality, you can look more into how to keep "undesirable" people out of your country (and deport the disadvantaged people who lower the average scores?), or you can look more into how to develop the school system so that even the disadvantaged people get a good education out of it. Finland chooses the latter, more constructive, more humane, more equal, and as far as we believe the PISA and similar studies, more efficient stance. Ethics aside, you're free to choose otherwise.
I've never made the claim that immigrants cause non-immigrants to perform badly, nor have I heard anyone else make this claim. What is the proposed mechanism? Bad study habits of non-immigrants rub off on native children or something? (I've heard of some yuppies in NYC trying to do the opposite, and make their kids socialize with Asian kids, but I've never seen data on this.)
As to immigrants causing non-immigrants to perform badly, that's how I understand the arguments that concentrate on the immigration differences while that's hardly the only difference between the two countries (as if the poor in the US didn't lower the average as well, and as if the scores of the Caucasian students weren't below Finland as well).
As to the proposed mechanism, I could imagine school funds and teacher resources being directed towards integrating the immigrants in the school. At least in Finland it is a concern, as the schools have less funds than the US ones.
Wouldn't Finland's flatter income distribution be a potential confounding variable in this case? When the difference between the 25th and 75th percentile in income means less in absolute terms, there are probably a whole lot of things that are associated with income in the US that show less of an association in Finland. For example, high-IQ types might be less likely to strive for high-income occupations in Finland than in the US, lessening the correlation between income and IQ. With child IQ being strongly correlated to parental IQ, it's easy to see how this could lessen the correlation between school performance and economic background.
Of course, you may have been comparing Finland to other countries with similarly flat income distributions, which would make my above point moot. Even then though, it's not clear that Finland's education system is necessarily "more equal" than other countries -- it could just be more effective. If the school system in, say, Norway is less effective than in Finland, higher-income students are likely to see less of a drop in scores because they have more educational resources outside of the system.
First, I know little about this gap in Finland or what's causing it.
However, I can hypothesise that it has to do with the immigrant children having attended school in Finland for less time. Further, the Finnish school system has been developed for Finnish citizens and is less prepared to attend to the needs of different ethnicities than to different socio-economic backgrounds. For starters, the immigrants need to learn the non-Indo-European language.