I was deciding between (B) and (E) and finally picked correct answer choice (E), because (B) said that "since 1947, it has had little effect." We can infer from paragraph 1 that while union membership may not have decreased after 1947, there is an effect of wage depression through 2 mechanisms:
1) reduction in number and spread of union members; and possibly
2) collusion of manufacturers and suppliers in right to work states (Carroll).
Paragraph 2 says that if there is a reduction in power of unions to negotiate wages (which we know could occur through 1) and 2) above) then black workers would suffer disproportionate relative loss. The question is only concerned with the wages and not economic positions, so the rest of paragraph 2 (lines 61 to 66) are not of concern. In other words... If there is such a reduction in unionship as to cause a reduction in the "economic power" of unions to raise wages or to stop collusion, right to work actually has an effect of causing a disproportional decline in relative wage positions of black workers.
But we don't know that this is true... So (B) is jumping the gun. Is my reasoning right?