The concept behind separation of powers is to prevent tyranny by not allowing political power to accumulate in the hands of a small number of people. In the U.S. we tend to see it as absolutely critical to liberty.
But is it? For the textbook I’m writing, I did a quick graph comparing separation of powers (none, weak, strong) to Freedom House scores (averaged from 2003-2014) for 23 countries that have been independent and continuously democratic since 1950 (based on the work of Arend Lijphart). Here’s the first graph.
That’s not strongly suggestive of the need for separation of powers to preserve liberty.
But there’s some uncertainty in the data that I have to do some work to clarify. There are a handful of countries I’m not yet sure whether to label as having no separation of powers or weak separation of powers. So in that chart I labeled them all as “weak,” as a conservative approach. But I also made a chart showing the results if I labeled them as not having separation of powers.
Here the trendline is clearly negative, even though the lowest ranked country for freedom is in the non-separation of powers group (India). It’s not strongly negative, though, and of course the small number of cases in the strong separation of powers column indicates we shouldn’t have too much confidence in the result. But it certainly does not support the hypothesis that separation of powers is necessary for freedom, which tends to confirm what I have thought for quite a while now, just from casual observation. After all, it’s pretty hard to look at the Netherlands, Denmark, Iceland and New Zealand and say they’re not very free.
[Notes: The two countries with strong separation of powers are the U.S. and Costa Rica. Indonesia’s current constitution also has strong separation of powers, I believe, but it has not been continuously democratic for very long. Countries where I am at present uncertain just how to label their separation of powers are Austria, Canada, France, and the UK. Most have some characteristics that may make the label choice more artful than scientific. France, for example, apparently has separation of powers only when the president and the parliamentary majority are of different parties. Montesquieu’s model for his theory of separation of powers, the UK, has (it seems to me) steadily shed the elements that embedded separation, and is probably now closer to no separation of powers than a weak separation of powers. And Austria–whose government I know nothing about–is listed in my source as having no separation of powers, but both Wiki and the Austrian government webpage say they do have it. So there’s still a little bit of work to be done. But with any luck I’ll be delivering this data to my students and shocking their minds with the undermining of one of our precious American myths.)