I forget how, but some conversation recently got me thinking about coattails, i.e. the phenomenon wherein someone popular at the top of the ticket will draw in voters for downballot races, and presumably an unpopular candidate will also hurt the other candidates.
It is, of course, difficult to measure, but there have been elections in which one party did overwhelmingly well in every area, like 1980, when Ronald Reagan brought 13 new Republican Senators with him along with his landslide victory over Jimmy Carter. This swung control of the chamber to his party until 1986; similar gains occurred in the House, where Republicans gained 34 seats and mustered the support of many conservative Democrats to boot, but remained under 200 members.
However, we also have situations like 1972, when Richard Nixon's 24-point defeat of George McGovern could not prevent the net loss of two seats in the Senate, and the gain of 12 in the House was just under a third of what they needed to take the majority. Oddly enough, House Republicans held 192 seats after each of the elections I've mentioned.
The first step should really be to define this as strictly as possible: what are coattails, exactly? The idea is that someone with high name recognition and race notability will tend to affect the votes of candidates running for less prestigious offices on the same ballot, whether positively or negatively.
As previously discussed with the initial election of Reagan and reelection of Nixon, this doesn't always happen, but that simply turns the task into delineating the differences and ascertaining what the requirements are for the coattails phenomenon to come into play in a certain election. Part of this will be simple reasoning, which could well be wrong, and part will be examination of data, which is limited to what I can get my hands on for the time being (since I don't have access to a university library).
For the first part, it stands to reason that the presidential candidate should be viewed as the leader of the party, with nobody else vying for the position or being identified as the leader of one wing; this would lead voters to lump every other member of the party in with that person, and assume, to some extent, that they will have the same agenda. Again, this could easily be completely off-base, but it seems reasonable for now.
This would also apply to both sides—if one party has an unpopular presidential candidate, but the party is not unified behind him, it would make sense for that unpopularity to have relatively little effect downballot. I honestly just thought of this, and it makes a lot of sense for the difference between Nixon in 1972 and Reagan in 1980; Republicans were united in both cases, and Democrats coalesced behind Carter after Kennedy conceded (how much difference that campaign made is the subject for another column), but a lot of Democrats were pissed at McGovern.
See, he was the head of the commission designed to prevent the anger and infighting of 1968 from happening again, and one of the changes he and his team made was giving primaries a lot more power to decide the nominee. That power came from party leaders, and they were pissed. They were mad when the report was released in 1969, and they didn't cool down when McGovern won a bunch of primaries despite their lack of support in 1972. Plenty of Democrats didn't like McGovern, and either voted for Nixon or didn't pick either, but they still sent Democrats everywhere else.
