How did European immigration affect the timing or patterns of the Great Migration? Why?

Suppose we want to estimate the causal effect of immigration on an immigrants wages (the return to immigration, essentially). To do so, we collect data on 1000s of men born in Norway around 1860, some of whom stay in Norway and some of whom move to the United States. Why cant we just regress wages in 1900 on immigrant status ? That is, why isnt causal (be specific to the question)?

Suppose we want to estimate the causal effect of African-American migration from the South to the North on wages in the receiving Northern cities. To do so, we collect data on wages in 100s of cities in 1970. Why cant we just regress average city wages in 1970 on the number of Great Migrants arriving in each city from 1910 to 1970 ? That is, why isnt causal (be specific to the question)?