Russ Roberts predicts my policy positions


Earlier this year, there was an interesting debate between Russ Roberts of EconTalk and Adam Ozimek of Forbes about ideology and economics. Basically, Roberts (mostly on Twitter) took the cynical position that ideology governs much of people's stances on policy positions, that this is inevitable, and that we should just accept it. Ozimek said no, economists aren't as ideological as Roberts thinks, even commentators in the public sphere. He also said that if you find that your own positions are driven by ideology, it's a sign that maybe you should rethink how you form your positions.

More recently, the argument flared up again. Roberts declared the following:


I then challenged Russ to predict my positions on various policies. Initially I suggested that I would tell him three of my positions, and then name three other issues and ask him to predict the second three. He counter-suggested that he would merely pick a bunch of issues and predict my positions on all of them. I agreed, despite the fact that this was not as good a test of Russ' thesis.

Anyway, though, Russ did come through with the promised predictions, and posted them on Twitter. Here they are (sorry for the weird embedding of reply-tweets):









OK, here is a scorecard. Russ named 13 policy issues and predicted my positions on all of them. I will give 1 point for a correct prediction, 0 points for an incorrect one, and 0.5 points if I don't really have a position or am not really sure. I will also include brief explanations of why I hold the various positions - not because I love hearing myself dispense opinions, but so I can prove that I'm being honest about what I think. Here goes:

1. ACA (Obamacare): I'm not sure. I don't really know much about health care policy. My instinct says that the most important health care problem in America is high excess cost of care, not the number of the uninsured. Obamacare does do some things to try to address the cost problem, and costs do seem to have decelerated somewhat in the last few years, but I just don't know how big of a factor Obamacare has been, or whether the cost slowdown is a permanent thing.

Score: 0.5 pts.

2. Minimum Wage Increase: I favor local experiments with $12 or $15 minimum wages but not a national minimum wage hike - at least, not until the results of the experiments have come in, which will take years. Even then, national minimum wages are generally not great because they don't take local cost differences into account. Also I think EITC dominates minimum wage in most respects, and if paid as a wage subsidy instead of a tax credit would dominate in all respects.

Score: 0 pts.

3. 2009 Stimulus: Yes, probably a good idea. The temporary tax credits didn't do a whole lot, but support for state spending probably did do substantial good. And the stimulus included a substantial amount of money to shore up our badly underfunded infrastructure. Moreover, the deficit (and spending as a percent of GDP) has since come down to normal levels, quieting people's worries that the stimulus was a cover for permanent increases in spending and/or deficits.

Score: 1 pt.

4. Bernanke: Yes, Bernanke did a good job. Monetary policy was probably a factor in averting a Second Great Depression.

Score: 1 pt.

5. Bailouts: Probably something like that was necessary. But I think bailouts were handled badly in the heat of the moment; should have been much harder on management of big banks, to avoid future moral hazard. Still, long-term net costs to the government of most of the bailouts (with the exceptions of AIG and GM) were zero. So I'll give Russ the point.

Score: 1 pt.

6. Higher Taxes on the Rich: I don't have a moral problem with raising taxes on the rich, and I doubt the efficiency cost would be that high (rich people aren't really working to consume). However, I am very pessimistic about our chances of actually getting the money from the rich people, who are very good at avoiding taxes. Taxes on the rich used to be much higher, but the share of tax revenue as a percent of GDP was about the same. I do, however, think a higher inheritance tax would be a great idea. Not only would it tax unproductive "trust fund babies", but it would also probably raise the happiness of many rich people themselves in the long run. I think most rich people - or at least, most "self-made" rich people - don't realize how much their kids will be spoiled by large inheritances. So inheritance taxes can help save rich people from their own mistakes!

Score: 0.5 pts.

7. CFPB: Seems like it has been doing good so far, though too early to tell whether it will remain effective in the long run.

Score: 1 pt.

8. Unemployment Insurance Extension: No. We're way past the recession. Unemployment insurance is an automatic stabilizer, but it does discourage work. And the more work gets discouraged, the more people's skills and resumes and work ethic decays, and the more danger they are in of falling into the ranks of the long-term unemployed.

Score: 0 pts.

9. School Vouchers: My God, what a terrible idea. Privatized education just fails, fails, fails whenever it's tried. History is clear: It does not work. Vouchers are also a form of faux-privatization, where the government pays the bills but doesn't administer the service. That is a clear and unequivocal recipe for ineptitude, waste, fraud, corruption, and inefficiency. Russ is totally right about my position on this one.

Score: 1 pt.

10. More Govt. Funding of Education: Not sure. Don't know how effective this is. It would probably depend crucially on how the money was spent, though I don't know enough to know what the "good" way is.

Score: 0.5 pts.

11. Fannie & Freddie: Bad. These agencies seem like yet another example of faux-privatization. Government provides the funding (in this case, in the form of a guarantee), but doesn't do the administration, leading to bad incentives. Also, I think we incentivize homeownership way too much in this country.

Score: 0 pts.

12. Stricter Gun Control: Probably not. I grew up in small(ish)-town Texas, where tons of people had guns and there weren't any shootings that I ever heard of (though probably some accidents). Canada has relatively high gun ownership and very little crime, including few mass shootings. Brazil has a small fraction of the gun ownership we have, and much higher crime. Meanwhile, we've had a huge drop in crime in the last two decades with no real increase in gun control. Let's try to replicate that success before we start disarming the populace. I will admit that my stance on this has wavered recently, in light of the rash of mass shootings, but I still don't think gun control is likely to have a huge effect. A much, much more important policy for reducing gun death would be to end the War on Drugs.

Score: 0 pts.

13. Trade and Immigration: Russ is right. I'm basically pro-trade and immigration, but not for open borders. I'm definitely more skeptical of free trade than the average economist - I think industrial policy can be useful for developing countries, and I think that trade with countries that manipulate their currencies can sometimes be self-defeating. I also think that international capital flows can be destabilizing and can reduce productivity in some countries. But I will still give Russ the full point.

Score: 1 pt.


Total score: 7.5 out of 13


That is slightly better than random chance. But remember, Russ follows me on Twitter, which gives him an advantage. Also, he was able to choose the issues, which gives him a number of additional advantages - he can choose positions on which a substantial majority of Americans agree, he can select several issues correlated to those on which he is most sure about my position, etc. Given this, I don't think Russ was able to predict my positions that well.

The only obvious cluster of predictive success was on policies relating to the financial crisis. Russ correctly predicted my positions on Bernanke, the bailouts, and the CFPB.

I think this exercise shows a number of different "failure modes" of attempting to model people's policy positions based on an assessment of their ideology. For example:

1. You may fail to assess people's ideology correctly. Russ probably didn't expect that I'd intrinsically value the personal liberty of owning guns. He also probably thought I would be more eager to tax the rich simply in order to reduce inequality (regardless of its efficacy).

2. Your model of ideologies may have errors. Russ probably assumed I'd have ideological reasons to support Fannie and Freddie, because he thinks I'm a liberal, and liberals support Fannie and Freddie. But I'm not sure this is true - I've never seen liberals defend those agencies. Maybe some have, but it doesn't seem to be a pillar of liberal ideology.

3. No model of ideology is perfect. "Liberal" is a name given to a cluster of ideas, but few people precisely fit that cluster. Most people have at least one or two positions that don't toe the ideological or party line. Each individual's ideology is complex, and the standard clusters, like "liberal" and "libertarian", are only approximate models. Russ probably didn't guess that my values would be "liberal" on the CFPB, "libertarian" on gun control, etc. Even I don't know what to call myself, ideologically.

4. People disagree on the facts, not just on values. In general, people with heterogeneous priors about the state of the world will fail to reach agreement even after seeing all of the same evidence. And when people form their policy positions, they consider efficacy of policies, not just whether the intended effect would be a good thing. Russ probably didn't bet that I would be pessimistic about the efficacy of taxing the rich, the usefulness of the ACA's tax credits, or the effectiveness of gun control. He also probably underestimated my uncertainty about the effect of Obamacare on health costs, the usefulness of education spending, and the employment effects of minimum wage hikes.

So anyway, this was a fun exercise, and thanks to Russ for taking the time to do it. I'm not sure how it really relates to the original Roberts-Ozimek debate, but it was interesting nonetheless.


Updates

A few people have suggested I was too hard on Russ in my scoring. I gave him 0 points on minimum wage, despite the fact that I like the idea of conducting natural experiments with local minimum wages. And I gave him only 0.5 points on taxing the rich, despite the fact that I favor a higher inheritance tax.

Well, maybe; it's hard to score precisely. Some things are closer to 0.7 or 0.3 than 0.5. I did the best I could. On minimum wages, my opposition to a national minimum wage, and my belief that EITC is just better, made me give it 0 instead of 0.5. In reality it's more like 0.2, rounded down. On taxes-on-the-rich, I gave 0.5, because it really depends on what we use the revenue for. If it's for inefficient subsidies or boondoggles like the F-35, I say *cut* overall taxes on the rich, while shifting from income to inheritance taxes. I don't want to raise taxes on the rich just to "soak the rich" and stamp out inequality, as many do. So I don't think this deserves more than a 0.5. 0.6 at most.

Then there were a couple where I rounded *toward* Russ' prediction. For example, on bailouts, I gave him 1 point, even though I think the bailouts were not executed well and created moral hazard. So that's really more like 0.8, rounded up. And on the stimulus, I think about half of it went to utterly useless (but not particularly harmful) tax credits. So that's more like 0.8 or 0.9, rounded up. My skepticism about the pro-trade consensus probably makes that a 0.9 as well, rounded up.

So overall I think I was fair, and my inevitable roundings came out a wash. Plus, this isn't a scientific test, so changing to 7/13 or 8/13 wouldn't change the basic message, which is about how "ideology cluster" models can fail.

Comments

Post a Comment

Popular posts from this blog

Econ theory as signaling?

Image
I don't expend much effort dissing macroeconomics these days, but every once in a while it's good to give people a reminder. I wrote a Bloomberg post about how academic macro (or more accurately, mainstream macro theory) has not really helped out the finance industry, the Fed, or coffee house discussions. The reason, as Justin Wolfers recently pointed out , is basically that DSGE models don't work. Brad DeLong then wrote a post riffing on mine , which is excellent and which you should read. A super-fun Twitter discussion then followed, part of which Brad storified for posterity . But that leaves the question: Assuming Wolfers and DeLong and I aren't just blowing smoke out of our rear ends, and DSGE models really don't work, why do so many macroeconomists spend so much time on them? One obvious hypothesis is that a huge percent of their human capital is already invested in knowing how to do this technique, so they just keep doing what they know how to do, and teachi...
Read more

Robert Lucas in biology class

Image
Back in August, a bunch of people were talking about Paul Romer and Bob Lucas and history of macro and stuff like that. Somehow I missed this post , where Brad DeLong dug up a Bob Lucas memoir and made fun of Lucas' college biology class exploits. For reference, here's a longer version of Lucas' story: The only science course I took in college was Natural Sciences II - a biology course. We read a modern anatomy text, and also selections from Darwin, Mendel, and others...  [T]here was nothing spooky about Mendel’s genetic theories. They were clear, they made some kind of sense (though there was nothing molecular in our Nat Sci II readings), you could work out predictions that would surprise you, and these predictions matched interesting facts. We did a classroom experiment with fruit flies, focused on eyes, and pooled the results. Our assignment was to write up the results in a lab report and compare them to predictions from a Mendelian model. I had not enjoyed the actual l...
Read more