Dr. Alfred G. Cuzán

With a larger and more diverse set of countries than Cuzán and Heggen (2023a) studied, this article completes a general path of incumbencies that was partially traced by their “cruise and crash” model of the cost of ruling. The missing piece is a bump in the share of the vote in the party’s first reelection. The electoral “bonus” holds more or less steadily until the crash, at which point the vote plunges about 20% below the original win. The analysis suggests that the path of the vote in long-term incumbencies follows a general pattern that is shaped like an inverted asymmetrical letter U or W.

Since the publication of Nannestad and Paldam ([1999]. “The Cost of Ruling. A Foundation Stone for Two Theories,” University of Aarhus, Denmark. Working Paper No. 1999-9. http://www.martin.paldam.dk/Papers/Gamle/Cost-of-ruling.PDF.) on the cost of ruling, their finding that incumbents lose votes in the order of 2 to 3 percentage points per term has gained the status of something like an “inductive law” of politics (Budge [2019]. Politics. A Unified Introduction to How Democracy Works. Abingdon/Oxon/New York: Routledge, 2019). We suggest that this generalization conjures up an inaccurate image, that of a gradual reduction of a ruling party’s vote over the course of an incumbency or “spell.” Our own analyses of national and state or provincial elections in Australia, Canada, Germany, and the United States demonstrate that this interpretation is not correct. Instead, while they remain in office the ruling party cruises along, averaging more or less the same share of the vote election after election, until they fail to win another term, at which point all or nearly all of the cost of ruling, like a balloon payment on a loan, comes due. Moreover, on average the magnitude of the loss is independent of the length of their stay in office. This means that a long incumbency confounds the expectations fostered by the “inductive law” because the overall loss – which we find to be reasonably constant across a wide spectrum of data sets – divided by a spell length exponentially distributed from one to many terms cannot be a constant. Nannestad and Paldam's assertion of 2.25 percentage points per term vote loss is at best a description of a mid-range value.

Abstract In an analysis of 1,825 state or provincial election outcomes in five federal democracies the rate of decay of incumbency ( K ) serves to partition the distribution of the vote for the incumbent party, the party of the head of government, between those who win a subsequent term and those who do not. In conjunction with the mean and standard deviation of the distribution, the weighted mean of the vote in re-election and defeat is identified. The model’s predictions are generally within 2–3 percentage points of the actual outcome.

In “In Laws of Politics and How to Establish Them,” Erik Weber contends that my arguments for the existence of “five laws of politics” are “inconclusive.” Although the “empirical evidence is impressive,” he avers, the “underlying social mechanisms” responsible for the adduced relationships are missing. Without it, he adds, no empirical relationship rises to the special status of a “law” of politics. Helpfully, Weber did not stop there. Using the example of Duverger’s laws, he suggested ways to close the “argumentative gap.” In this article, I aim to do just that.

Exponential functions, widely used in the physical sciences, also have been used to model political phenomena. To our knowledge, however, this tool has not been used to replicate the electoral survival of the government or administration in several democracies. This article reports that an exponential survival model is a good fit for the reelection rate of the party that controls the executive office in states, territories, or provinces in three countries: Australia, Canada, and the United States.

Two decades ago, Peter Nannestad and Martin Paldam (2009) published a paper in which, having analyzed 282 elections held between 1948 and 1997 in 19 developed democracies, they claimed that all incumbent parties on average incur a "cost of ruling" of approximately 2.25% points per term. They called this cost a "robust fact," "an unusually stable constant" that operates across countries, institutions, and time. I evaluate how well N&P's empirical assertions hold up in a much larger set of elections held in a set of well-established democracies similar to the one they studied, as well as in other, more recent electoral democracies outside the OECD region.

In this follow-up to “Five Laws of Politics,” I show that the original estimates of electoral outcomes describing the five “laws” hold up with twice the number of elections and three times as many countries as in the original article. The analysis yields long-run equilibrium values for several electoral parameters that set limits to incumbent support in a democracy. The evidence lends additional support to the notion that elections in democracies appear as if governed by “laws of politics.”