Parliamentarism passes another test
Parliamentarism passes another test
I wrote a recent blog post on how parliamentarism passes the Pritchett test. The Pritchett test is one of the most stringent tests a policy can pass. To recapitulate, in Romer's words, the policy must fulfill the following requirements:
In a cross-sectional comparison of levels, do countries that are more developed have more X?
In cross-sectional comparison of growth rates, do countries that have rapid growth in X also tend to experience a rapid increase in standards of living?
When we look at the few countries for which we have long historical records, do the ones that become much more developed also acquire much more X?
If we look for countries that switch from a regime of slow economic development to a regime of rapid development, do we see a parallel shift in the rate of growth of change in X?
Hard indeed. What I realized is that parliamentarism also passes another test, not as stringent, but still pretty hard. Not nearly as famous, but still meaningful to me, because it is a test I devised myself five years ago, well before I ever thought about writing on parliamentarism. Most of the criticism I got at the time was that it was too stringent, and I argued that it was not. Still, I re-read it with some satisfaction that I still agree with the test (with one slight modification I'll mention below), and that parliamentarism passes it, even though I had forgotten I had ever written this.
This is the test, in a flowchart:
Following the flowchart you will see that there are many conditions which make a policy not recommended. That is on purpose. Policies are an intervention on the freedom of people, at the very least because they cost taxpayer's money to be implemented. So the default position should be to not carry out with any policy. This does not mean, of course, that no policies are ever justified. To see that parliamentarism passes it, let's follow the flowchart. First, I should note that I was - inappropriately - conflating "evidence" with "experimental evidence".
Step 1 - Is there plenty of [experimental] evidence for it? - No
2 - Is there evidence against? - No
3 - Is there evidence of absence of evidence? - No
4 - Can we get evidence with small RCT? - We could, but not a realistic proposal
5 - Can we get better evidence with full scale implementation? Experimental evidence, no. Better evidence, yes. So we can choose eithe answer. Let us say we went with "no".
6 - Are the models and observational data extremely eliable, and are there reasons to believe net benefits of running the project would be huge? Yes.
he main change I would make to this chart (other than clarifying that I mean "experimental evidence" on the fist step) is to change the color of the orange "recommend it" panel. I wrote this when I assigned a greater probability that experiments would become ubiquitous in public policies. If that was the case, there could definitely be some expectation of small experiments (in cities, or other types of smaller organizations) to try to assess the relative benefits of parliamentary vs presidential arrangements. But I don't think that such a revolution will come anytime soon, and I believe that we must act with the best information available at any given time. So, if I were to redraw this chart, I would have a bright green color for both "recommend it" panels.