Exploiting primogeniture and the degrees of inbreeding as natural experiments, the Ottinger and Voigtländer (2020) paper revisits with unprecedented rigor the question of whether the abilities of leaders matter in the grand scheme of history. The answer is a resounding ‘yes’ for unconstrained rulers.
Can we isolate a leader’s impact?
The Ottinger and Voigtländer (2020) paper joins a large body of literature that investigates whether leaders matter for a country’s performance. This has frequently been a difficult question to answer, as it is often hard to isolate the leader’s impact from other ongoing forces.
However, Ottinger and Voigtländer (2020) shrewdly note that a wealth of data lies just around the corner if exploited appropriately. Medieval and early modern Europe was filled with competing states vying for supremacy, ruled by a small number of clans that frequently intermarried. Next-of-kin marriages caused a considerable degree of inbreeding, leading to grotesques such as Carlos II of Spain.
Figure 1: The inbred Carlos II of Spain, reportedly unable to close his mouth and had one testicle
Source: Wikimedia Commons
In addition, primogeniture, whereby the first-born son is always the dynastic heir, ensures that ruler selection is quasi-random and independent of expected abilities. This opens the door to a successful instrumental variable strategy, a setting creating a natural experiment for econometric estimation.
However, measuring ruler abilities and state performance are two different matters entirely. For those two variables, the authors turned to the meticulous records compiled by the MIT biologist-turned historian Frederick Adams Woods, who scored the ruler abilities on a ten-point scale, and state performance under their reign on a three-point scale, based on meta-analyses of historians’ assessments.
To add more robustness, the authors included another measure of state outcome by considering the number of territories gained or lost during a ruler’s reign, using data from historian Abramson (2017). Thus, Ottinger and Voigtländer (2020) created a dataset of 331 reigns spanning from 990 to 1800 CE.
Findings
Firstly, the paper finds empirical support to the first-stage hypothesis. The degree of inbreeding is negatively correlated with ruler abilities, even when excluding outliers of extreme inbredness such as Carlos II.
Secondly, ruler abilities are found to be strongly correlated with state performance. A one-point rise in ruler abilities is associated with an 11% increase in territories or a whopping 17%, depending on the choice of econometric method used. Measurements using historians’ assessments are even more significant.
Figure 2: Association of Monarch Ability and Country Performance by Country, in coefficient of correlation
Source: Ottinger and Voigtländer (2020)
Notably, it is also found that the relevance of rulers varies from polity to polity depending on the constraints on the rulers. England before and after 1600, for example, shows two rather different pictures of the degree to which ruler abilities matter. The authors use the frequency of parliamentary meetings as a proxy for the degree of constraints, with data compiled by Van Zanden et al (2012). Interacting the degree of constraints with ruler abilities, the authors found rulers’ personal abilities matter much less when constrained by parliaments.
Implications
The implications of the paper are far-reaching. Though drawing data from medieval and early modern Europe, the paper lends its weight in contemporary debates in political science and public discourse over the selection of national leaders, as the historical context is but a tool to identify the underlying causal mechanism.
Even when constrained by parliaments, a more competent leader is still beneficial for the fortunes of the state, though a stronger parliament can help to contain the damages of an incompetent head of state. Countries plagued by incapable populist leaders should pay heed. At the same time, autocracies can reap benefits from an enlightened despot, but also suffer dearly from an incapable one. Fukuyama (2011) famously said that a ‘Bad Emperor’ will be autocratic China’s downfall. Now we can, in principle, make such predictions with statistical precision.
Comments