Content warning: Frequentist statistics
While economics is often derided as the dismal science, I believe that economists have done much to improve policymaking in the world. Their models are never perfect, but they are often useful. I find that Political Science as a field has been less successful at producing utility, as evidenced by our more tenuous relationship with policymaking.
Firstly, political scientists (poli sci's) have a more difficult subject matter: case numbers are usually small, strategies are constantly changing, and the room where it happens is often intentionally barred to researchers. Secondly, Poli Sci's have a more diverse set of normative missions, and no consensus on revealed preference. Thirdly, usefulness often depends on predicting highly uncertain events, but calibration and decision theory are omitted from the traditional training.
The purpose of this blog series is to recognize poli sci's who's work has high consequentialist value. I admire these writers for not just satisfying academic incentives but writing articles with real potential to improve the lives of their fellow man. They are my role models.
Are Parliamentary Systems Better? by Gerring, Thacker and Moreno
The largest differences between the democracies of the world are the unity of legistlature and the executive, the number of veto positions, the form of election, and federal vs. unitary states. When countries form or democratize constitution writers must make difficult, uncertain choices on these axis. Gerring et al. compare the outcomes caused by presidential systems and parliamentary systems. They skip labyrinthine debates about the risks and advantages of each design by looking directly at the outcomes through a massive regression analysis.
Under presidentialism, legislature and executive are elected separately. Each have a mandate from the people and they often disagree about policy, resulting in gridlock. Americans will be familiar with this problem. But the multiple elected parties can also "check and balance" one another, which might result in more stable economic policy. Under parliamentary systems the legislature chooses the prime minister directly. A majority of members of parliament are free to replace the prime minister at any time by a no confidence vote. The government cannot be split and there are fewer veto points. Broadly speaking, presidentialism is common in Latin American, the US and Africa. Parliamentalism is common in Europe, South Asia, East Asia and British colonies.
The great innovation of Gerring and Tacker is to measure outcomes with not one but 14 separate measure of government outcomes. The challenge to the regression is that so many different factors (culture, political traditions, starting gdp, malaria, democracy, legal origin, resource curse) affect outcomes that identifying the parliamentalism effect is fraught. They solve this with 14 outcomes including GDPPC, telephone mainlines, trade openness, investment ratings, corruption indices, bureaucratic quality indices, and illiteracy.
The results find that parliamentary government leaders to better outcomes. They find statistically distinguishable positive effect on 7 of 14 governance outcomes and no negative effects. The most plausible mechanism is that unitary governments are better at solving collective action problems. I also suspect that parliaments are easier for voters to understand and thus punish bad governance (split systems allow the party holding each branch to blame the other, confusing voters). These affects seem stronger than the hypothesised benefits of presidentialism (see federalist papers, etc.)
Gerring et al. certainly don't have the last word on these issues, future regressions have argued the real effect is smaller. But the decision to use many dependent variables greatly improves their robustness. Furthermore no subsequent article (afaik) has found evidence supporting presidentialism. So if one is writing a constitution, choosing parliamentalism is smart unless you have a massive prior for presidentialism.
It hardly needs saying that the longterm value of a better constitution is massive.
Good questions Thomas. The point of the blog series is to highlight papers that ask the right questions and use the right methods to have consequentialist value. I am not arguing that the Gerring paper is the last word. I'll answer a few of your questions, though.
We know they aren't p-hacking in the selection of dependent variables because there are very few such variables that cover every country-year of interest. How many organizations measured the governance quality of Liberia, Columbia and Denmark in 1953. I'm working on introducing a new one using weather station quality.
I didn't want to dive into the regression table in my blog post. All models used adjust for continent. They also adjust for distance to financial center. I would also point out that if the continents with lots of presidential regimes have less cross-border trade, this is evidence against quality of governance of presidentialism.
There is a later study with an expanded dataset that supported the null on GDP, but I didn't include it because it ignored the 13 other governance indicators. This isn't my main research area so I won't do a full literature review for this blog post. In municipalities the same result is robustly observed.
If robustly establishing causation means "adjusting for every factor which could possibly affect governance outcomes at the country level", then the question is clearly unanswerable. There are hundreds of such factors and RCT's are impossible. But as consequentialists our goal isn't to achieve some arbitrary degree of confidence in our beliefs. The goal is to make better decisions. Since your prior on pres v. parl should be near .5, this evidence compellingly moves us toward the parl side, maybe to .7 . For a constitutional designer, that's a hugely valuable update. There remains a 30% chance of making the wrong decision, but that's way better than a 50% chance of making the wrong decision. Therefore if even one constitutional designer reads this paper, the QALY's that Gerring et al. have made is huge.