Why do certain immigrant groups tend to assimilate, while others don’t? This post argues that immigrant institutions (such as religious communities or foreign language media) have an important role in determining the degree of assimilation.
Moreover, altruistic institutions (those who care about their group’s identity and wealth) and extractive institutions (those who only care about the group’s identity to the extent that it increases their own revenue) may have different effects.
These topics are discussed in a nice working paper by Prummer and Siedlarek (2014). The idea is that an immigrant’s identity is influenced by two main factors: the host society and their own group. Exposure to the host society (e.g. education, working) is beneficial because it provides higher income. On the other hand, the minority’s identification with their group influences how much they invest in these host country-specific skills. A higher identity leads to lower investments.
Thus there is an inherent trade-off between identity and assimilation. Identity itself is shaped by institutions. These could include for instance imams for Muslims, rabbis for Jews, foreign language media for any group, and so on. The paper also refers to these institutions as “leaders”.
But others have an effect on identity as well. Obviously, the host society can influence it (negatively), and other group members (insofar as the person in question is connected to them socially) can have an effect as well.
As mentioned above, members of the minority group may invest in host country-specific skills (e.g. education). The return to these investments is higher if there is lower discrimination or if the person’s ability is high. The (utility) cost of these investments is increasing in the identity of the person.
There are several results derived from this setup. First, if there are no leaders, group members assimilate fully (i.e. their identity goes to zero asymptotically). This rests on the benign assumption that an initial small investment in host country skills is beneficial for everyone. In other words, the corner solution of zero investment is ruled out. If someone does invest nothing in skills, they will still assimilate fully as long as they are connected to another group member with a positive investment.
Second, the situation with a rigid leader is considered. This is a leader whose identity is given exogenously and is non-responsive to any endogenous changes in the model. I.e. they have a fixed unchanging level of identity. In this case, the authors show that there will be no full assimilation, but there will be some. The identity of the group members will lie somewhere between zero (which corresponds to full identification with the host society) and the leader’s identity.
An example of rigid leaders may be Turkish imams in Germany. These leaders are usually brought from Turkey to serve in Germany for 4-5 year periods. They are thus unlikely to be affected by German society, because they’re only there for a relatively short period of time.
Third, the authors finally allow leaders to be strategic. This means that they can adjust their identity in order to reach their objectives. But what are the objectives of leaders? As mentioned in the introduction, two types of leaders are considered: altruistic and extractive. Altruistic leaders care about the identity and the wealth of their groups. Extractive leaders care only about the wealth. They only care about identity insofar as it makes group members spend their wealth with group-specific institutions.
For instance, religious leaders may be thought of as altruistic. They care about the well-being of the community (wealth) because it is needed for instance to maintain a community (build churches, etc.), but they also care about the identity of members. Foreign language media (or any minority-centered business really) can be thought of as extractive. They want group members to be wealthy so that they can spend a higher amount of money at their businesses. Of course, a precondition for spending money at minority-oriented businesses is identification with the group itself. But the main objective here is that they want the group to be wealthy. Identity is only of secondary concern.
Let us consider the altruistic leader first. If the payoff for skills investments is a concave function of identity, then strategic altruistic leaders will choose an identity in-between the zero (i.e. full assimilation) and the maximum possible one (i.e. no assimilation at all). So the most likely outcome of this is that there will be some degree of assimilation for the group. Interestingly, the more discrimination there is against a group, the more it will assimilate.
The fact that the payoff for skills investments is a concave function of identity simply means that if you raise identity by 1 unit, the payoff will decrease less than proportionately (i.e. by less than 1 unit).
If on the other hand, the function is convex (the decrease is more than proportionate), then strategic altruistic leaders will choose an extreme identity: either they will have zero identity (full assimilation) or they will have the maximum possible identity (no assimilation at all). Since the group members will end up with an identity between the host society’s and the leader’s, this means that there can be full assimilation in this case, or potentially imperfect assimilation and an extremist leader.
Contrary to the concave case, the leader is more likely to choose an extremist (as opposed to full assimilation) position if discrimination is higher.
Let us now move on to the extractive leader. The results are quite similar. They’re the same for concave functions. For the other case, the main difference is that an extractive leader never chooses full assimilation. They may be extremist, but never fully assimilated. This is intuitive, as for instance a minority-centered business would have no business at all if all members fully assimilated.
When comparing interior solutions (i.e. the ones with intermediate assimilation), the degree of assimilation may be either lower or higher with altruistic leaders. This depends on the importance altruistic leaders place on the wealth and the identity of their group.
As a final thing, the authors examined how the structure of social connections can affect the results. Extractive leaders seem to prefer highly connected groups. This is because there is a lower dispersion in wealth/identity in this case, as group members influence each other more. If this were not the case, we’d have many rich members with low identity, and poor members with high identity. This is bad for an extractive institution like a business because the former don’t want to buy their products, the latter do but they can’t afford it.
The opposite is true of altruistic institutions: they prefer loosely connected groups. The resulting high dispersion in wealth/identity ensures that the institution has both money and dedicated group members.
An interesting conjecture by the authors regarding this is that Western European immigrants to the United States may have ended up with more altruistic institutions (that the churches they started arguably are) because their societies are more individualistic and thus their group structure was more sparsely connected. At the same time, groups from places like Turkey, Mexico, China or Korea ended up with extractive leaders because their group structure is more connected (owing to their more collectivist societies).
For instance, Turkish imams may be extractive because they tend to go to Germany for a higher salary. Mexican groups are to a very large extent affected by extractive foreign language media. Chinese immigrants had (have?) highly extractive clans in places like San Fransisco (see the paper for examples). Finally, Koreans tend to be very nationalistic and government representatives at their migration destinations (e.g. the consulate general) tend to fulfill the role of extractive institutions.
For me, one of the most interesting implications of this paper is that probably more often than not, more discrimination against minorities helps assimilation. On some level this makes much sense. For instance, immigrants in many Western European countries can get away with using English to communicate with both host country institutions and their colleagues (if they work), which definitely reduces the incentives to assimilate. A similar effect can be seen in the US with Spanish-speaking immigrants, who – due to the large number of Latinos in certain places – have little incentive to learn English.
Another interesting feature is that the results arise without an explicit wish from immigrants to preserve their culture and norms. The only requirement so to speak is that immigrants want to bond with other immigrants from the same country/culture. This creates “business” opportunities for the leaders, who (after establishing their institutions) have an interest in preserving the norms of the minority group.
Last but not least, it must be emphasized that in this model the immigrant group has no effect on the host society. The host society is rigid, unchanging in its culture and norms. This is probably a restrictive assumption for larger groups, but may work well for smaller ones. For instance many groups in the US who are now considered fully assimilated really did influence host country customs. Think of Italians and how their food (amongst others) is now a vital part of American culture. So in a sense, Italians changed the host culture sightly so that they didn’t need to fully assimilate (into the pre-Italian immigration US culture) in order to fully assimilate (into a new post-Italian immigration US culture). Maybe their assimilation would have never been so perfect had the host country been rigid and unwilling to adapt some Italian norms.