“What policies are correct” is a fairly under-talked about question in political discourse. This is because discourse usually takes place in echo chambers or high-profile clashes of bad-faith people with drastically different opinions trying to gin up controversy. It’s surprising that in honest discussion, people rarely attempt to make the case that a given policy is good or bad. Maybe they accept their position unthinkingly on faith, or maybe deep down we understand that these questions are very complicated - so we leave it to the experts.
Recently I argued that if we view politics through a cultural and economic axis, with cultural and economic issues being the driving factor behind the differences between factions, that economic questions are inherently more important than cultural ones. For cultural questions, people have a lot of experience with them in their own lives. For a throwback to the 2010s, consider gay marriage. Most people are heterosexual and some people are homosexual. You could make high-brow arguments about western culture that says that gay marriage is actually bad, but the average anti-gay marriage advocate is a heterosexual who thinks that gay sex is disgusting. They have the requisite information to have that opinion, because it basically boils down to if you think it’s wrong or not, which is subjective. Their intuition is telling them that it’s wrong, because in a biological sense it is, at least to them.
But economic questions aren’t always like that. On some level we have experience with the market: we look for jobs, we compare offers, we buy this pair of headphones over that pair and so on. But the most important questions about economics involve trillions of dollars in market activity, and it’s not clear that humans innately have a sense of what a 1 or 2 percent tax increase on a given industry actually does in reality. Intuition isn’t a reliable guide here. Our inability to mentally comprehend The Economy at large doesn’t mean it doesn’t exist, it just means that during the time we spent in trees and caves we didn’t interact with a global market economy based on mutual exchange. To showcase this point, consider that even macroeconomists can’t fully explain the Great Depression. That over 100 years of data and breakthroughs in computing and economic experimentation doesn’t give a clear answer on the biggest economic disaster in modern history demonstrates that this stuff is really hard, even for experts.
Then again, just because we don’t have answers to the biggest questions doesn’t mean we shouldn’t listen to experts. Doctors haven’t cured aging, but you would trust them to stitch your tendon in the case of a knife accident. Physicists haven’t yet explained the Big Bang, but we can still launch rockets into space. Economists can’t explain the Great Depression, but rent control has failed everywhere it’s been tried. Many questions do in fact have correct answers, even if we have not answered literally every question.
II. When experts disagree, what do you do?
Then again, experts don’t always agree.
When two people disagree about something and you are trying to decide who is correct, what do you do? I asked ChatGPT this same question:
There are a few good points here, but there is a running theme with the ones that aren’t great. Evaluating credentials to some degree requires you to be an expert. If one professional clown went to clown community college, and the other clown was the president of the American Clowning Association then it would be easy to guess who knows more about Clowning. But if both had similarly situated credentials and experience, who is correct then?
ChatGPT’s next 2 points suffer the same fate. Checking if there is bias requires you to verify evidence of bias and evaluating data requires you to do the same. If you’re able to do that then good for you, but for the sake of the thought experiment we are assuming that the data is good because otherwise why would either expert think they are right? It’s totally normal to run experiments and get different results and the experiment being conducted properly in both cases. This isn’t physics, these are the Human Sciences, and people are different from each other. You can definitely get different results each time you run an experiment. Answers 5, 6, and 7 also fall victim to this same issue. The human sciences just don’t work like regular science.
But there are 2 answers here that are intriguing: Consensus in the Field and Independent Verification. For now I’ll focus on Consensus, because Verification goes hand in hand with their being a consensus in the first place. Consensus is a great answer - imagine a distribution of opinions on a scientific question, with a clear median sticking out. We can reason that if scientific errors are random, then the experiments on the tails should cancel out in favor of a result that keeps appearing over time. Take the following professional graphic.
A completely hypothetical illustration.
In theory, if people and places and contexts randomly vary in time, then given enough experiments conducted in those places and on those people, the results will look something like a bell curve. There should be tons of results with many on one side or the other, but there should be a result that keeps repeating, because it is true. If random variation occupies the tails of the distribution, then the median outcome can be understood as the effect that the policy actually has.
Imagine you then poll experts about their opinion on something. To some extent you are just measuring whatever experimentation that very expert has done, but to another extent you are measuring what they heard about from someone else or what they themselves have researched. Maybe you’re measuring their vibe that day. You’re measuring a lot of things. At the end of this poll you’ll still end up with some type of distribution. If you asked Physicists why the Earth revolves around the sun you’d basically get one answer. Economics isn’t physics, so you won’t get literally the same answer for an economic question - again, this isn’t a deathblow to the exercise, this is just the reality of the profession.
So what does the distribution of expert answers look like for the rent control question?
For whatever reason these pollsters don’t straightforwardly ask “is it good or not?” That’s annoying, but in my view these are pretty bad answers for rent control in any case. The answers themselves do appear to be normal distributions which is what you would expect given that we aren’t talking about basic physics questions.
Another way to think about this is to consider what the impact of expertise actually has on your opinion, that is: what if we could see what non-experts think and compare that to what educated people think, and then compare both of those results to what domain-specific experts think? If we knew the answers all 3 groups have to the same question, then we can predict that expertise is not only changing people’s opinion, but that it does so in a predictable direction.
Well, that poll has already been done. From The Myth of The Rational Voter,
In 1996, the Washington Post, Kaiser Family Foundation, and Harvard University Survey Project collaborated to create the Survey of Americans and Economists on the Economy (henceforth, SAEE).9 Based on interviews with 1,510 randomly selected members of the American public and 250 economics Ph.D.’s, the SAEE is ideally structured to test for systematic lay-expert belief differences.10 It also features remarkably diverse questions, which lets us explore belief differences in depth. A further advantage of the SAEE is its rich set of respondent characteristics. One can use this information to test theories about the origin of lay-expert belief gaps. The rest of this book draws heavily on the SAEE, so it is worth exploring at length. Its 37 questions break down into four categories.11 Questions in the first two categories ask whether various factors are a “major reason,” “minor reason,” or “not a reason at all” why “the economy is not doing better than it is.” There are 18 questions of this form. Questions in the third category ask whether something is good, indifferent, or bad for the economy. There are seven questions with this structure. The last category is a grab bag of a dozen miscellaneous questions.
…
Each of these questions has three summary statistics:
• First, the “raw” average belief for the general public.
• Second, the “raw” average belief for Ph.D. economists.
• Last, the estimated belief of the Enlightened Public.
…
And so on. The basic gist of this survey is that the public thinks one thing, economists think another, and informed citizens think something else that’s closer to the economists than the public. This indicates that information reliably leads you away from what the general public thinks and towards what economists think.
Simply taking the argument at face value though proves too much, because it could be the case that instead of the public’s belief being based in ignorance, economists’ beliefs are based in a type of self-serving bias - they only believe something because it serves them, and the public has the actual true belief that would serve society.1
I think that’s a weird way to think about it though. We don’t treat any other field’s experts this way, even though by all accounts economics as a profession does pretty well in not being completely P-hacked:
Also, we can come up with plenty of self-serving narratives that the public holds that leads them to disagree with economists. Maybe they are more racist so they dislike immigration more? Maybe they have a lower time preference so they don’t think debt is a big deal? Maybe as a group they are lower IQ so they are worse judges of information? There are lots of narratives to throw around in either direction - if you want to take the side of the public on complex questions like these, be my guest, but you should probably lower your faith in experts in general across the board - not just in the case of economics specifically. The least you could do is actually argue your case if you’re so confident the experts are wrong!
The unwashed masses that supposedly know more than experts.
Another response is well maybe economists are simply misled? Sure, they aren’t scheming and plotting based out of class interest, but they could be unintentionally making mistakes. I think this is a decent response, but the rate of science failures is pretty low. I wouldn’t hold it against you if you limited your opinions to only the ones that have a lot of consensus, and remained agnostic towards ones that don’t, because that would be an improvement over what the average person thinks.
The best rebuttal I can think of is to only have opinions that would benefit you, personally. “Economists may be right in the abstract, but I am a current renter in a city and rent control would benefit me.” I have no problem with this, because it’s not in the voter’s best interest to vote for socially-optimal policies, it’s in the voter’s best interest to do what would benefit them even if it comes at a greater cost to the rest of us. This isn’t a scientific problem though, it’s a problem with Democracy. Our leaders should be able to see above the fray and do what will benefit the most people with the expectation that they will be frequently asked to do things for one group at the expense of another.
But I don’t care about that as much because that’s boring. Of course people will benefit themselves, I just think it’s more interesting to think about what is correct in the abstract, and in the abstract - you should probably agree with the economic consensus.2
In short, when you are ignorant you should try to become educated. When you’re ignorant and being educated is too much work, copy expert’s homework.
And Democrats are the party of academia and higher-ed funding. That’s at least one reason to suggest that the answers economists give are even more to the left of what they actually should be.
All voters looking out for themselves probably end up worse off than a society that had some forward-looking leadership. Rent control is a pretty good example: If everyone acted in self interest and rent control was passed everywhere, we would all be worse off. My guess is that this is generally the case with most policy. A few people can free ride off the market, but if we all do that then we become poorer than if we didn’t. Policy-at-large may be a collective action problem. Politicians should be able to see this coming and adapt the specifics of their policy implementations to be less destructive on the margin.
But what if the vibes, and the podcast I listen to regularly tells me otherwise? My uncle last thanksgiving made a very compelling case that the Fed is a tool of Soros and Schwab, thus I know exactly what macroeconomic policy they should pursue (abolish central banks, return to the gold standard and bitcoin). These “economists” who say otherwise are clearly being paid off/compromised by the global elite and/or billionaires and/or coastal liberals to inflate away my earnings anyways.
But what if the vibes, and the podcast I listen to regularly tells me otherwise? My uncle last thanksgiving made a very compelling case that the Fed is a tool of Soros and Schwab, thus I know exactly what macroeconomic policy they should pursue (abolish central banks, return to the gold standard and bitcoin). These “economists” who say otherwise are clearly being paid off/compromised by the global elite and/or billionaires and/or coastal liberals to inflate away my earnings anyways.