Greg Mankiew on Modern Monetary Theory

Mankiew just released a https://scholar.harvard.edu/files/mankiw/files/skeptics_guide_to_modern_monetary_theory.pdf he prepared for a talk on Modern Monetary Theory for the AEA. I thought that this might be of interest to many people on this forum since Mankiew's talk is very short and informal. There is not a single equation to be found, but he took the time to sort through a textbook written by proponents of the theory so he could contrast it with mainstream macroeconomics.

The document is short, to the point and overall it is interesting. I think that Mankiew does a good job of capturing some of the intuition behind a very large class of macroeconomic models we call "New Keynesian" models, so the contrast he makes between MMT and NK models probably is meaningful.

I have debated some policy issues with some of its proponents on this forum and I have always felt uneasy about some of the more radical conclusions they have drawn. In particular, I have expressed some of the same concern Mankiew expresses here with respect to the capacity of governments to preclude default through monetary creation. It always felt like these people were creating things out of nowhere, but I now see better what they meant because Mankiew took the pain of translating their idea of exploiting some slack in production capacity into terms I can understand. It sounds a lot less like voodoo, though it still sound like a politically motivated exaggeration.


Quick background on New Keynesian models

For people who are curious about what are New Keynesian models, I can provide some basic context. They are part of what we call dynamic stochastic general equilibrium (DSGE) models. Those models think of business cycle fluctuations are resulting from a flow of random disturbances to which households and firms (and possibly governments and central banks) react (hence the "stochastic" in DSGE). Moreover, those models work in a general equilibrium setting, meaning they treat all markets as interdependent, and they are dynamic meaning everyone involved is forward looking: people try to guess what is going to happen, so anticipations matters. What distinguishes NK models from their "real business cycle" (RBC) counterpart from the 80's and 90's is the presence of nominal rigidities. In the data, a good or service will change price about every 9 months, so we force the models to not allow all prices to adjust every single quarter. Similar things can be said about wages.

..... link?

cpwill said:

..... link?

Click to expand...

I think they're still in the process of reverse repoing it.

I think this is what he's talking about:

https://scholar.harvard.edu/files/mankiw/files/skeptics_guide_to_modern_monetary_theory.pdf

So I took the 15 minutes to read it. The basic premise is that Mankiw still doesnt' get MMT. Let me briefly pick apart his conclusion:

In the end, my study of MMT led me to find some common ground with its proponents without drawing all the radical inferences they do. I agree that the government can always print money to pay its bills. But that fact does not free the government from its intertemporal budget constraint.

Click to expand...

So he get's the general idea that the government can't run out of money but then still can't accept it. He hand-waves about inflation and the fact that congress might decide to default and says MMT doesn't fully address those things. Hint: they do. He completly misses the fact that the real constraints on government spending are the real resources (labor, materials, land, time, etc.) that I would consider core to MMT thought.

I agree that the economy normally operates with excess capacity, in the sense that the economy’s output often falls short of its optimum. But that conclusion does not mean that policymakers only rarely need to worry about inflationary pressures.

Click to expand...

Please Greg, tell me how well the Fed has succeeded at creating inflation since the GFC even though they have been actively trying? How about Japan over the past 20-30 years? I think his inflation boogie-man is a bit harder to lure out than he thinks.

I agree that, in a world of pervasive market power, government price setting might improve private price setting as a matter of economic theory. But that deduction does not imply that actual governments in actual economies can increase welfare by inserting themselves extensively in the price-setting process.

Click to expand...

Another just-so statement with zero evidence.

He also brings up the Quantity Theory of Money, and the view that Reserves Create Loans, both of which have been debunked. In doing so he also conflates bank reserves with money available to be spent on real things.

Etallium said:

So I took the 15 minutes to read it. The basic premise is that Mankiw still doesnt' get MMT. Let me briefly pick apart his conclusion:



So he get's the general idea that the government can't run out of money but then still can't accept it. He hand-waves about inflation and the fact that congress might decide to default and says MMT doesn't fully address those things. Hint: they do. He completly misses the fact that the real constraints on government spending are the real resources (labor, materials, land, time, etc.) that I would consider core to MMT thought.



Please Greg, tell me how well the Fed has succeeded at creating inflation since the GFC even though they have been actively trying? How about Japan over the past 20-30 years? I think his inflation boogie-man is a bit harder to lure out than he thinks.



Another just-so statement with zero evidence.

He also brings up the Quantity Theory of Money, and the view that Reserves Create Loans, both of which have been debunked. In doing so he also conflates bank reserves with money available to be spent on real things.

Click to expand...

Like most mainstream economists, Mankiw can't help but look at everything through his own preferred lens.

Paying interest is a policy choice, and it doesn't equate to actual debt. Unemployment doesn't need to be solved by private sector expansion. There is no "natural" level of output. There are so many examples of Mankiw setting the parameters by which he measures MMT that it's probably easier to dismiss everything he says and simply explain MMT from a clean slate.

JohnfrmClevelan said:

Like most mainstream economists, Mankiw can't help but look at everything through his own preferred lens.

Click to expand...

I do not think it is a completely fair criticism of Mankiw, in large part because this doesn't apply specifically to either Mankiw or economists in general. Everyone involved in a discussion arrives with background information, as well as personal preferences and motivations. In other words, we all have some tendencies with regards to how we want to frame the discussion.

Besides, I think he actually made an honest effort. I don't get the impression that he is trying to bury anything behind methodological preferences, as much as he is trying to informally translate intuitions across two very different points of view. He might be wrong (I don't know how well he captures MMT), but he's not exactly trying to do something very simple here.

JohnfrmClevelan said:

Paying interest is a policy choice, and it doesn't equate to actual debt.

Click to expand...

Interest payments on deposits held by central banks are part of a policy choice. Specifically, monetary policy is (ordinarily) conducted by playing around with interest payments on borrowing and lending. If you commit to paying positive interest rates, part of the bank's liabilities include the payments it owes (as well as the deposits, obviously). You can set that rate to zero (or even to a negative interest rate), so you wouldn't owe interest payments. On the other hand, historically, a null interest rate would be a very expansionary policy. Both go hand in hand, so to speak.

Am I getting you right, or you meant something else than a central bank paying no interests on deposits?

Interest payments on deposits held by central banks are part of a policy choice. Specifically, monetary policy is (ordinarily) conducted by playing around with interest payments on borrowing and lending. If you commit to paying positive interest rates, part of the bank's liabilities include the payments it owes (as well as the deposits, obviously). You can set that rate to zero (or even to a negative interest rate), so you wouldn't owe interest payments. On the other hand, historically, a null interest rate would be a very expansionary policy. Both go hand in hand, so to speak.

Click to expand...

So this right here would be a key difference between traditional economics and MMT. MMT points out that interest on reserves is expansionary monetary policy. All those reserves that would normally sit around not earning interest are now earning interest increasing the assets of banks. Since all the QE that took place during the GFC, most banks are flush with reserves and instead of having to borrow them to meet requirements they have plenty of excess. So the interest paid on reserves is now an income channel for the banks which increases their profits.

Think about how a personal interest-bearing checking account would impact your behavior. On one hand, it would encourage you to keep more money in it than otherwise as you earn more interest. However, those interest payments increase your income which means you have more money to spend. Orthodox economics almost completely ignores that second point that interest is income and just assumes that extra returns cause people to save more--to infinity. In reality there is some interplay between the increased savings desire and the increase in income. Where those two functions overlap is not quite certain.

However, unlike people bank lending is not constrained by reserves (another MMT insight: the Loanable Funds Theory is false). So paying interest on reserves is purely adding income without increasing savings desires. You can read this paper from the Bank of England here if you don't believe me.

Don't even get me starting on the idiocy of negative interest rates on personal deposits. Anyone with any sense knew that would be a disaster and only silly economists with flawed models thought otherwise. Actually, if you think negative interests rates were a good idea, I'd love to have that argument with you as it will easily show you how flawed orthodox models are.

TheEconomist said:

I do not think it is a completely fair criticism of Mankiw, in large part because this doesn't apply specifically to either Mankiw or economists in general. Everyone involved in a discussion arrives with background information, as well as personal preferences and motivations. In other words, we all have some tendencies with regards to how we want to frame the discussion.

Besides, I think he actually made an honest effort. I don't get the impression that he is trying to bury anything behind methodological preferences, as much as he is trying to informally translate intuitions across two very different points of view. He might be wrong (I don't know how well he captures MMT), but he's not exactly trying to do something very simple here.



Interest payments on deposits held by central banks are part of a policy choice. Specifically, monetary policy is (ordinarily) conducted by playing around with interest payments on borrowing and lending. If you commit to paying positive interest rates, part of the bank's liabilities include the payments it owes (as well as the deposits, obviously). You can set that rate to zero (or even to a negative interest rate), so you wouldn't owe interest payments. On the other hand, historically, a null interest rate would be a very expansionary policy. Both go hand in hand, so to speak.

Am I getting you right, or you meant something else than a central bank paying no interests on deposits?

Click to expand...

Both IOR and interest on bonds. (We lump the Fed in there with Treasury as functionally part of the government.) Neither one is operationally necessary. If you start with the fact that a sovereign government is perfectly capable of printing up and spending their own currency, then you see bonds as merely another form of government liability, but with some handy features - but you shouldn't see bonds as "debt" in the normal sense of the word. The normal definition of "debt" being (to me) something that sets you back in real resources when it comes time to repay; I can't print money to meet my obligations, so I have to work, or sell something, or spend less. That's not true of the government.

This is one thing I took issue with - Mankiw saying that because the government paid interest, then it must be debt. Then he assumed limits on the government's ability to create and spend and the resulting inflation risks that should also apply to mainstream economic thinking, but are seldom brought up. How often do you hear the word "hyperinflation" tossed around when mainstream economics is being discussed? But when MMT is the subject, it's rarely not mentioned.

So in the MMT cost/benefit analysis of government spending, the cost is "whatever effects (if any) more dollars/bonds in existence have on the value of the dollar." But not "real debt," in the sense that today's debt makes it harder to spend tomorrow.

The MMT analysis of the benefits is also different, because we reject NAIRU reasoning. There is no point at which unemployment can be "too low."

As far as interest rates go, we reject any notion of a "natural" rate of interest. There is also no cost to the government creating and spending money, so bond interest and IOR is considered an unnecessary giveaway. Commercial banks should charge interest based on risk and time, not any underlying "cost" of reserves (not that there is any cost post-QE). We pretty much toss monetary policy in the trash, and would prefer that the Fed left interest rates at or near zero. (That's not a universal position among MMTers, though.)

There really is a lot of learned economics that one needs to dispose of to really get into MMT.

JohnfrmClevelan said:

As far as interest rates go, we reject any notion of a "natural" rate of interest.

Click to expand...

As Warren Mosler says, "The 'natural' rate of interest is zero."

I just came across this today from a post at Naked Capitalism.
CONGRESSIONAL TESTIMONY: Reexamining the Economic Costs of Debt

This is from Randy Wray (the guy who wrote the textbook from the Makiw piece). It's about 30 pages--so slightly longer than Mankiw--but has a lot of charts and tables which make for a quick read.

Let me just quote form the introduction here as a rebuttal to Mankiw:

Supposedly, [MMT] calls on central banks to just print money to pay for ramped-up spending. It is purported to claim that deficits don’t matter. It is said to ignore the inflationary consequences of spending without limit, and even to invite hyperinflation....

None of these claims is true. MMT is based on sound economic theory. Most of it is not even new. Rather it represents an integration of a number of long-standing traditions that heretofore had not been linked. It does reach some surprising conclusions, but these conclusions are more consistent with real world outcomes that mainstream theory has trouble explaining.

Click to expand...

Looking forward to hearing more from TheEconomist

JohnfrmClevelan said:

Both IOR and interest on bonds. (We lump the Fed in there with Treasury as functionally part of the government.) Neither one is operationally necessary. If you start with the fact that a sovereign government is perfectly capable of printing up and spending their own currency, then you see bonds as merely another form of government liability, but with some handy features - but you shouldn't see bonds as "debt" in the normal sense of the word. The normal definition of "debt" being (to me) something that sets you back in real resources when it comes time to repay; I can't print money to meet my obligations, so I have to work, or sell something, or spend less. That's not true of the government.

Click to expand...

The normal sense of the word "debt" is something that is owed or due to someone else. It would be less confusing if you added an asterisk about how loose is the capacity of the government to pay back what everyone understands to be debt.

JohnfrmClevelan said:

This is one thing I took issue with - Mankiw saying that because the government paid interest, then it must be debt. Then he assumed limits on the government's ability to create and spend and the resulting inflation risks that should also apply to mainstream economic thinking, but are seldom brought up. How often do you hear the word "hyperinflation" tossed around when mainstream economics is being discussed? But when MMT is the subject, it's rarely not mentioned.

Click to expand...

In mainstream macroeconomics, we impose solvency restrictions and restrictions on the behavior of central banks. So, the problems you mentionned are being accounted for in what is possibly the most conservative way possible.

You can think about it this way. You suspect that mainstream macroeconomics imposes too strict a restriction on government's capacity to pay its debt and to issue money. The idea being that if we were to not listen, it would take a much bigger violation than anticipated to create hyperinflation. Then, this shouldn't be a surprise at all. We don't fear hyperinflation under the stricter standards.

JohnfrmClevelan said:

There really is a lot of learned economics that one needs to dispose of to really get into MMT.

Click to expand...

That is fairly possible.

In order to steal successfully, capitalist only need to have agility and greed. Greed is especially necessary, because for a small theft, you can get sued.