Differences in the price of labor are sociological and political in nature, not a matter of personal preference, so that, e.g., native, unionized workers, who enjoy full political rights will demand higher wages and be more likely to resist employer prerogatives than undocumented immigrant, non-union workers from poorer countries.
According to Bonacich, likely outcomes of a split labor market include not only antagonism but, depending on the political power of higher-priced workers, a caste-like system where lower-priced workers are restricted to specific occupations, or total exclusion of the lower-priced group from the labor market.
The split labor market theory attributes events to social structure rather than to individual preferences.
[1] Split labor market theory traces the roots of racial/ethnic stratification to social and political differences that predate inter-group contact in the labor market, but the specific outcomes (caste system, exclusion, or something else) result mainly from the actions of the higher paid segment of the working class and their power relative to that of capital.
[3] In one of the best-known applications of split labor market theory, William Julius Wilson argued in The Declining Significance of Race that a split labor market provided much of the racial antagonism between blacks and whites during the earlier years of what he called the period of industrial race relations.