Nearly two decades ago researchers pointed out the sharp decline in marginal districts in elections for the US House of Representatives. That observation led to an outpouring of research describing the electoral changes, explaining their bases and speculating about their consequences for the larger political system. Recently Gary Jacobson has offered a major corrective to that line of research, arguing that ‘House incumbents are no safer now than they were in the 1950s; the marginals, properly defined, have not vanished; the swing ratio has diminished little, if at all; and competition for House seats held by incumbents has not declined’. While Jacobson advances an extremely provocative argument, there are complicating patterns in his evidence that support additional and/or different interpretations. We argue that the marginals, ‘properly defined’ have diminished, the swing ratio has declined, and party competition for House seats held by incumbents has lessened. While fears that the vanishing marginals phenomenon would lead to lower responsiveness on the part of ‘safe’ House incumbents have proved groundless, the collective composition of Congress does appear to be less responsive to changes in popular sentiments. Thus, the vanishing marginals have contributed to the occurrence of divided government in the United States and in all likelihood do have the effects on congressional leadership and policy-making that many analysts have claimed.
Policy making in divided government: A pivotal actors model with party discipline