The economic analysis of politics goes by many names: political economy, rational choice theory, formal political theory, social choice, economics of governance, endogenous policy theory, and public choice. Each of these labels picks out a subtly different intellectual tradition. Each tradition expands our understanding of the world. My favorite, though, remains public choice.
As a GMU professor, you may attribute this to home-team favoritism. Yet before I was a professor at GMU, I was a student at UC Berkeley and Princeton, and neither school fostered the love of public choice… to say the least. The main reason I prefer public choice, rather, is for its normative core. All economists who study politics do cost-benefit analysis, but the public choice approach is wiser. And heretical.
What exactly is this “normative core” of public choice? Simple: After doing standard microeconomic analysis of government policy, public choice adamantly states:
That’s an upper-bound on how well government intervention can work. In the real world, government intervention usually works much more poorly. Before we claim government intervention passes a cost-benefit test, we can, should, and must use past government performance to predict future government performance.
The upshot: Public choice economists end up opposing many government interventions blessed by textbook and policy wonk alike.
Example: Most economists – even economists who study politics – are fans of Pigovian taxation to address externalities problems. What public choice reminds us, though, is that Pigovian taxation is the best that governments can accomplish. In the real world, however, governments are worse in dozens of ways. Before you advocate a regime where government sets Pigovian taxes to address externalities, then, you should estimate what real-world governments will actually do when you give them that kind of power.
Another case: When I was a grad student TA-ing Industrial Organization, I often argued with the junior professor teaching the class. He knew a lot of theory, but almost no economic history, so I told him about quite a few famously anti-competitive antitrust decisions. After a while, I recall a little exchange that went roughly like this:
Junior Professor: Bryan, I don’t care about what government did in the past; I care about what government is going to do in the future.
Me: Shouldn’t we use the past behavior of government to predict the likely future behavior of government?
Junior Professor: By that standard, government should never do anything.
Me: [double-take] Not really, but OK!
For Junior Professor, the normative core of public choice was practically a reductio ad absurdum. But that’s only because he started with a firm pro-government conclusion, and rejected even ironclad premises that undermined it. So when I applied the normative core of public choice, he saw a big bias against government.
This so-called “bias,” however, is simply well-justified pessimism. If actual governments abuse the power to tax, subsidize, and regulate, then it makes cost-benefit sense to put the officials who set tax, subsidy, and regulatory policies in a few chains. Or a lot of chains. Or a solid block of concrete.
Mainstream economists tend to scoff at this mentality. Frankly, that’s because they’re fifty years behind the research frontier. Although textbook demonstrations that well-crafted government policies can make the world better are fun homework problems, they end up being an intellectual smokescreen for demagoguery. The normative core of public choice shows that laissez-faire is undervalued: Even when good government is plainly able to make things better, past experience teaches us to be deeply skeptical that government will do so in practice. Until economists judgmentally study government in action, they have no business recommending that government do much of anything.
READER COMMENTS
BC
Sep 15 2020 at 10:19am
Did Junior Professor recognize the irony that he himself, as a pro-government advocate, was conceding that government’s track record was too dismal to merit granting it intervention power?
Michael Stack
Sep 15 2020 at 10:26am
I remember when I first started learning Econ, and seeing all the wonderful policies that governments can enact to make the world a better place. Then I looked at what governments actually did, and saw little resemblance between the two. It took me a long time to understand why.
Paul
Sep 15 2020 at 11:24am
When I was younger I always thought that “if government would just do X then things would be better” and wondered why this never happened. The solutions were so easy and elegant. After coming across public choice I know why.
KevinDC
Sep 15 2020 at 11:38am
Like many things in life, this reminds me of a Scott Alexander post. He spent a good amount of space accurately describing how textbook economics could enable the government to make healthy food more readily available to poor people. He then goes on to say:
Knut P. Heen
Sep 15 2020 at 11:51am
I refereed a Norwegian language Microeconomics textbook for bachelor students last year. It was full of market failures and suggestions for improvements by the government. In my report, I said straight out that the book was politically biased because there was no chapter on government failure. I requested the authors to put in a chapter on public choice. The new edition came out with a brand new chapter on environmental taxes and no public choice theory. Funny.
When I taught micro I used to show the students that tax revenues typically go down if a monopoly is replaced by a competitive market because competition reduces profits. The government’s incentive is aligned with the monopoly producer. Antitrust action is simply not in the government’s interest. Indeed, various sales taxes and Pigovian taxes may be set such that you hit the monopoly price in every single market in the economy.
Art Carden
Sep 15 2020 at 8:07pm
This reminds me of the first event I attended after coming to Samford. It was a discussion of sports in Birmingham, and if I recall correctly the panelists were in unanimous agreement that Birmingham needed to build a stadium. During Q&A I raised my hand and asked about the standard economist’s objection.
Afterward, someone from the CVB, I think, came up to me and said “I see you have a bias against a stadium.” I replied “if you’ll excuse me, it’s not a bias. It’s a conclusion I’ve drawn on the basis of the best available research.”
Megen de la Mer
Sep 15 2020 at 8:18pm
One of the issues possibly holding back economics is its US centric view. This is unsurprising given that the great majority of advancement in economics is due to the work of US economists. By itself, this is a good thing, but implicitly US economists heavily discount how the rest of the world operates. As such, contempt for – or at least a (correctly) high level of suspicion of – government’s ability has a strong following amongst US academic economists. Yet to this outsider, I see US government at all levels as beset by skewed incentives because heads of agencies are appointed by the political class. That is, these skewed incentives invite nepotism and other sub-optimal outcomes due to being inbuilt into the US systems of government, (e.g.election of minor public officials (police chiefs?) rather than appointment on merit).
Hence I argue that you first need to address and correct the incentives of your institutional arrangements before (quite correctly) pointing out the limitations of government intervention.
For example, would the Westminster system result in better government in the US?
Jon Murphy
Sep 16 2020 at 12:27pm
You are right that outcomes are dependent on the rules and system they emerge from (as the great Public Choice founder James Buchanan said: “order is defined in the process of its emergence”), but the fact that politicians, bureaucrats, and other political appointees respond to incentives is only one part of Bryan’s point. The question is not whether political system X is better (according to who?) than political system Y. Rather, the question is: can political system X solve the supposed market failures? Even if Westminster is better than Congress, it does not imply that Westminster would be better than a market outcome.
Nathanael Snow
Sep 15 2020 at 9:14pm
The Buchanian approach has a potential softer norm.
The role of the political economist is to describe to the public in deliberation the relevant opportunity costs of alternative sets of rules, or policy options.
The public then may deliberate and come to a consensus regarding which option to select.
A liberal will recognize the consensus as the shared conception of the good and respect it.
AJ
Sep 19 2020 at 5:37pm
Unless a GMU economist disagrees with the consensus. Then, it’s bad and is worthy of mocking on a blog post.
Thomas Hutcheson
Sep 16 2020 at 11:30am
The problem I see with public choice theory is that it remains at
a theoretical level. Let’s grant that cost benefit analysis can fail to incorporate the costs in execution. If the solution were to roll up one’s sleeves and, using public choice theory, show how to get a better answer, fine. To often however the analys stops at the level that the analysis is biased upward often with the implicit conclusion that the result is zero and that nothing should be done. The implicit conclusion gets in the way of the critique being taken seriously.
The NPV of a revenue neutral tax on net CO2 emissions as calculated by any given model will probably fail to capture the failings in the way it will be legislated and administered. So should we just do nothing about the increase in the CO2 concentration in the atmosphere, even if we could rule out something a lot worse than a revenue neutral tax on net CO2 emissions coming along?
Jon Murphy
Sep 16 2020 at 12:15pm
Why do you assume the alternative is “do nothing”?
Also, why do you assume “doing something” is better than “doing nothing”?
Econ 101 gives us reason to doubt your first assumption. Public Choice gives us reason to doubt your second assumption.
Jon Murphy
Sep 16 2020 at 12:31pm
Here’s the big takeaway for you from Bryan’s post:
As I am sure you know, a market failure exists only to the extent a government system can improve upon it (Dahlman, 1979). If a Pareto-relevant solution does not exist, there is no market failure. Sure, we may wish costs (transaction, search, whatever) are lower, but that does not imply the market has failed; just the opposite.
As Bryan points out, practical experience tells us to be very suspicious of what governments can and cannot do. So, you cannot even make the claim that CO2 is a problem without a theory of how government operates.
Matthias Görgens
Sep 17 2020 at 11:41am
Now you are making your job too easy.
Obviously you can claim that CO2 is a problem without any reference to government.
What you can’t do is claim that CO2 is involved in a market failure by the definition you have, without involving the government.
Jon Murphy
Sep 17 2020 at 3:27pm
“Problem” is a vague word. Is CO2 a problem because it imposes costs on people? Yes, but so does everything, so “problem” becomes meaningless here.
Is CO2 a problem because it imposes costs on people and those costs can be removed or alleviated? Now we’re getting somewhere, but we need to then ask the question “if there is a solution, why hasn’t it come about?” If the answer to that question is merely “well, the costs of solving it are too high,” then it’s no longer obvious we have a problem. To go back to my luxury car example, it’s a claim that not everyone having a luxury car is a problem because the costs of owning one are high.
Dylan
Sep 18 2020 at 9:31am
Now you’re just getting silly.
Imagine there is a large asteroid on a collision path with earth. Almost uniform consensus that it will likely wipe out most or all life on the planet. If we don’t know the solution to this, does that imply that it isn’t a problem?
Let’s say we have a potential solution, but one that requires unprecedented global government cooperation and spending huge amounts of resources. Does the fact that governments have been incapable of ever pulling anything like this off in the past suggest that we shouldn’t try?
Jon Murphy
Sep 18 2020 at 10:02am
Again, the choice is not “government or do nothing.” That’s just patently absurd. The choice is “who acts?” If governments have shown they are incapable, it is the hight of silliness to expect them to suddenly be competent. Rather, a multiplicity of solutions would be better, whether it be for asteroids or emissions.
Dylan
Sep 18 2020 at 12:03pm
That’s easy enough to say, but I’m not seeing a ton of historical examples of coordinated worldwide action and expenditure without government intervention either.
What does a multiplicity of solutions look like when it comes to an asteroid or GHGs? What is the incentive structure? Who has the power to direct large portions of world GDP?
Jon Murphy
Sep 19 2020 at 8:30am
Look around you. They’re so prevalent it’s like breathing. World poverty and hunger have nearly been eliminated. Trillions of plans are satisfied every day without any governmental coordination.
Matthias Görgens
Sep 17 2020 at 11:42am
Why or how does looking at past performance of government have anything to do with normative questions?
Mark Z
Sep 17 2020 at 7:39pm
I was thinking the same thing. This seems like a ‘theoretical vs. empirical’ issue rather than positive vs. normative.
Jon Murphy
Sep 18 2020 at 10:02am
Because normative informs positive and vice versa.
Comments are closed.