Pennsylvania Politics -- The Eight-Year Cycle

Real Clear Politics features an article by Terry Madonna and Michael Young on Ed Rendell's problems in the gubenatorial election this fall. Their hook is the "eight year cycle" they discern in Pennsylvania voting. They write:

Under Pennsylvania's celebrated "eight-year cycle," Rendell, as a first-term incumbent, should be entering the re-election period as a prohibitive favorite destined to win a second term just as incumbent governors or incumbent parties have for more than half a century.

But this time:

[T]he governor's job is in jeopardy. In the vaunted "horse race" between him and Lynn Swann, he is barely ahead of a candidate who has never sought government office before. Worse perhaps--in the polls' fine print--the percentage of voters who have a favorable opinion of him are just above 40 percent, his job approval hovers around 45 percent, and his "re-elect" scores (the proportion of voters who thinks he deserves re-election) are about 40 percent.

They then go on to explain the "unexpected" situation in terms of voter dissatisfaction [a tautology if ever I read one]; regional differences in the electorate; and Rendell's ineptitude as governor; and Swann's superstar appeal.

Read it here.

This is the sort of nonsense that passes for serious analysis among political consultants, journalists, and sports fans. They take a limited and statistically insignificant series of data points, claim to discern a coherent pattern in them, and then articulate a "law of politics" based on totally inadequate data. In this case they posit:

an almost clockwork-like pattern established during the past half century of elections in which Pennsylvania voters have ousted the incumbent party from the governor's mansion every eight years. The cycle began in 1954, and since then, we have consistently had a pattern of eight years of Democrats followed by eight years of Republicans.

Fifty years, six changes of party, twelve elections; and they admit that the pattern was almost overturned in 1982. Not much on which to base a "venerable" and widely accepted "law of politics."

Each election has to be considered in its own terms. An election in the 1950's is hardly comparable to one in the seventies or with one today. Milton Shapp is not Bob Casey is not Tom Ridge is not Ed Rendell. None of them were subject to any implacable, deterministic forces that guaranteed the outcome of their efforts. The pattern is essentially meaningless. All it indicates is that neither major party has a lock on Pennsylvania.

That, of course, doesn't mean that journalists and political consultants, like Madonna and Young, will not claim to find meaning in the pattern. It serves as a good "hook" for a story, and when pitched to unsophisticated clients or readers can produce the illusion of understanding. And that's really what counts -- producing easily digested arguments that seem to explain a complex reality. It really doesn't matter if they do or don't. The illusion of wisdom will suffice.

I seriously doubt that either Ed Rendell or Lynn Swann, when considering their chances in the upcoming election, paid any attention at all to the "venerable" eight year cycle. They're too smart to swallow that sort of sillyness.