Part of a series debunking bad economics on the internet
Reading time: ~10 minutes.
TL;DR Planet Money breaks a 900 episode streak of good economics by fawning over a pseudoscientific subfield.

(Reddit comments can be found here)

Planet Money is in my opinion the best economics podcast out there[footnote]Stiff competition from Freakonomics. EconTalk is only good when academic guests are on. Conversations with Tyler isn’t specifically about economics[/footnote]. This is why I’ve been disheartened to see them give a full uncritical episode covering MMT.

I’m not a macroeconomist – my training is in empirical methods – so my issue with MMT is not on the substance itself, but the meta-discussion around it. My problem is that MMT acts like a cult, not a scientific endeavor.

13 year old Amy has this question:

I had this idea that because [the government] print money, instead of giving it to the bank and making inflation go up, they could use it just for the public services. And it would be much easier. And it would be it would be, in general, really good because there’s lots of problems with, like, there’s not enough tax to go around all of the schools and hospitals. So I thought maybe this might help. So thank you for listening. Yeah, thanks. Bye.

I don’t expect Amy to know that the central bank is independent of the federal government, especially since the hosts apparently don’t either:

GOLDMARK: Like the first part - printing money. What is stopping the government from printing as much money as it wants?

HELM: Basically nothing.

Fun fact: It’s very important for the central bank to be independent from the government.

If you politicize something, you have to be ready for those decisions to be taken by the worst people you can imagine[footnote]Self-admitted leftists like MMT proponents should be horrified by this at a time where the house, senate, and administration have a majority by people they deeply disagree with. Especially when those politicians are ready to do anything to hold onto power, including rallying behind complete morons.[/footnote].

The central bank has much more power over the economy’s health than the government does. It requires swift decision making at critical times, with potentially dire consequences [footnote][here is]2 a fun game showing the tightrope central banks walk constantly [/footnote]. Political actors on the other hand, spend their time thinking about how to get elected or stay in power. This makes it impossible to take money supply or interest rate decisions in the best interest of the country.

For example, the economic cycle is not synchronized with the election cycle. Central banks act in a “counter-cycle”, boosting the economy in recessions and tampering it in times of wealth. In theory, the government should do the same with fiscal spending – cutting taxes and spending more in downs, and increasing taxes and reducing spending in ups. Of course, the federal government does not do that, because budget decisions are taken for all sorts of decisions that seems positively insane to an impartial observer.

Later in the episode, the hosts note that hyperinflation is a common objection to MMT, which the MMT “expert” handwaved away as a ridiculous objection. But it’s important to know that the historic hyperinflations were only possible in the context of a central bank that wasn’t independent of the government. Sargent (1981) is clear:

“The essential measures in ending hyperinflation in each of Germany, Austria, Hungary, and Poland were, first, the creation of an independent central bank that was legally committed to refuse the government’s demand for additional unsecured credit […]”

Venezuela and Zimbabwe, for reference, are following the same pattern in their hyperinflations.

HELM: OK. Alex, we found the perfect person to answer Amy’s question. She’s an economist, had a similar question about 20 years ago, and has been thinking about it ever since. Her name is Stephanie Kelton. So what do you think? Is Amy onto something?

[17 minutes of gushing over MMT ensues]

Bad planet money! Bad!

Stephanie Kelton is the biggest public proponent of Modern Monetary Theory (MMT). MMT is a fringe economic theory, which has gotten attention ever since its proponents have discovered a fact that marketers, cult leaders and politicians have known forever: if you repeat something loudly and incessantly, some people will start believing it.

You might notice that shouting something repeatedly is not “doing science”. That’s because MMT does not act like a normal scientific subfield. It acts according to the definition of crank science. Let’s take a detour and brush up on how science works:

• Scientific endeavors come up with an idea, then try to disprove it as hard as they can with real world evidence. If that idea survives many rounds of this game, we declare it a good idea.

• Cranks, on the other hand, come up with an idea, then look for evidence to confirm it, and explain away any contrary evidence. Then they make a lot of noise on Twitter either way.

Are you a science? The game edition

Here’s a game you can play in any normal subfield of a science: Give me your theory as equations and a way to test it statistically with real world data. Note that we need to represent ideas in mathematical form to make them unambiguous. But the idea and equation can be extremely simple, as long as they’re accurate.

Big theories (at least the core we agree on as fact) pass this test in scientific subfields. Some examples:

• The Solow macroeconomic growth model can be tested by replicating Mankiw, Romer, Weil (1992). You can easily replicate this paper with freely accessible macroeconomic data in a weekend.

• The Spence Signalling model can be tested by replicating Bedard (2001) or Murnane, Tyler and Willet (1999). It’s a bit harder to find good data to replicate those studies.

• Most of the classical and quantum physics theories you’ve heard of can be tested in a lab. Hell, most of you probably tested a few classical mechanics equations in your high school lab. That is, you “did science”.

Well written scientific courses follow that pattern: high level idea in words, a mathematical representation, and empirical tests of the idea [footnote]David Autor’s lecture notes are a great example 1, 2, 3[/footnote].

It’s important that the empirical check explicitly test the theory (eg. it shouldn’t pass because something else is true) and they have to be replicable in multiple independent settings.

For instance, later in the Planet Money episode, the hosts note that the recent low inflation with low unemployment are real world evidence of MMT. It’s not an empirical test because those facts are consistent with a ton of theories, MMT being one of them. This is classic crank behavior: looking in smoke clouds and choosing the patterns you like as evidence of what you want to believe.

On the other hand, when you read about the replication crisis in social sciences, or all the brouhaha in modern macro, you should be happy about those subfields’ intellectual rigor. The fields have a self-defense mechanism against crankery, though a slow one [footnote]Some fields have inherent problems with this. String theory for example is untestable, so it’s tenuous to even call it science at this point.[/footnote].

Sidebar: Why Nassim Taleb is a crank but Picketty isn’t

We’re now armed with enough definitions to call Nassim Taleb is crank [footnote]I mean, the content of his twitter should make that clear, but apparently lots of very intelligent people don’t get that [/footnote]. Nassim has ideas, which he rarely if ever writes them down as models, and never, ever explicitly tries to falsify with tests. He then yells happily at any moment where something could be interpreted as a boost to his ideas, and goes on outraged rants whenever he feels challenged.

On the other side of the spectrum, we have Thomas Picketty. Thomas had an idea, which he immediately expressed in math: “If r > g, then wealth will concentrate over time.” He then spent years collecting data and wrote a book on how his idea fit to the data. His idea was met with empirical criticisms (here and here) which in turn sparked discussions to the merits of all sides.

Picketty’s big idea may very well be wrong, but he’s doing good science either way. We’ll know who’s correct eventually, and everyone involved in this process was doing good work regardless of who “won”. Nassim Taleb on the other hand might be speaking Truth Itself as handed to him from God, but we couldn’t know because he’s not doing science, he’s looking to bolster the Cult of Nassim Taleb.

How does MMT stack up?

The conclusion to this game is asking an MMT proponent the same questions, at which point you are inevitably met with backflips and rationalizations. This is not because data is impossible to get – it’s very easy to get data in macroeconomics. It’s easy to make tests invalid by over-identifying macroeconomic models to the number of variables, but MMT theories generally aren’t 20 equation monsters like the ones in modern macroeconomic models.

Note that I’m not engaging the ideas of MMT on their own merit! They could be correct for all I know. But we don’t get to learn that because its proponents are not honestly engaging in the pursuit of knowledge. It strange to state this confidently without sources, but notice that the problem here is that there aren’t any. Empirical work in MMT amounts to looking at historical events and subjectively interpreting them.

The reason is that MMT’ers are not doing science, they are proselytizing a cult. They don’t care very much about falsification or empirics, unless it’s to write Medium articles gushing over how right they are [footnote]Notice Stephanie Kelton’s site is geared towards public appearances – it’s very hard to find her vita or even publication list online. Because the research is not the point of her public persona[/footnote].

If they start having discussions on the empirical merits of their theories, instead of discussing how constellations lining up totally confirm their beliefs this time for real, then we should actually start listening.