Myth Busting

Stop Saying MMT Describes Reality – It Doesn’t

I had a good laugh at this Tweet from Jo Michell (who is a very good economist by the way). It seems to be from a UK blog advocating for MMT:

Now, I may not be an expert on sexual intercourse, but I am an expert about money creation and so that qualifies me to tell you that, based on this comment, this guy doesn’t know what sexual intercourse is. The simple fact is that MMT does not, definitively does not, describe the existing reality of the monetary system. Let me explain.

First, MMT assumes that the government is a “sovereign currency issuer” to reinforce the lack of funding constraints.¹ Okay, but this is like assuming that I am really really rich before I tell you that I don’t have a credit constraint. So, this whole narrative around “sovereign currency issuer” reminds me of the old joke about the economist on a deserted island trying to open canned goods and assuming we have a can opener. No, every country isn’t sovereign and MMT’s definition of sovereign is so vague that it’s meaningless. They might as well be saying “really really rich countries have more flexible spending capacity”. This just isn’t a useful description of anything.

Second, the whole basis of the MMT taxing/spending narrative is constructed around the false consolidation of the Fed into the Treasury. Consolidating the Fed and Treasury is fine for communicating general concepts, but they are specifically different entities for specifically different purposes and you cannot claim to be describing reality if your entire money creation narrative relies on consolidating them. In fact, the whole reason they’re separate entities is because the money creation story does not begin and end with the Treasury and Fed.

And that brings us to an important understanding. The whole reason there is a Federal Reserve is because there are private banks and the US government has outsourced money creation to private banks. We don’t have a fully “state money” system as MMT likes to portray. But MMT misconstrues this concept by consolidating the Fed into the Treasury to claim that the total government receives its own liabilities when it taxes. For instance, the US Treasury has an account called the TGA at the Fed. So, if you consolidate the Fed into the Treasury then you can argue that taxes settle by transferring government liabilities to itself. This would logically destroy those liabilities and spending would result in a new reserve credit that creates new money. Hence the MMT narrative that taxes destroy money and spending creates money.

Of course, this isn’t what actually happens. When the Fed settles a payment with the US Treasury they mark down a bank’s reserve account (from existing reserves, reserves that were created specifically by the Fed for the purpose of Monetary Policy and interbank payment clearing) and that debit equals a credit at the Treasury’s TGA. The size of the Fed’s balance sheet doesn’t change. There is no creation of liabilities and no destruction of Fed liabilities. It merely transfers Fed liabilities from a bank to the TGA. And when the Treasury spends the Fed credits a reserve back to a bank. There is no creation and destruction of anything here. There is nothing but transfers. You can even see the credits and debits in real time every single day via the Treasury’s daily statement.²

Importantly, you have to understand why the Fed is even a thing. And the Fed is a thing specifically because we have a monetary system constructed around private banks that the government is specifically subservient to. So you have private banks creating most of the money in the system and they don’t handle interbank clearing well during panics. We saw this originally in the early 1900s and most recently in 2008. So we created a central clearing entity that wouldn’t stop processing payments during panics. That’s what the Fed does and it does this by forcing banks to hold reserve balances.

It’s crucial to understand that the only reason reserves are even a thing is because we have private banks. So when you consolidate the Fed into the Treasury and treat their liabilities as the same thing you logically end up with a de facto nationalization of the banking system. After all, the only way you could argue that the government receives its own liabilities without the Fed being a thing would be by claiming that the Treasury issues all the liabilities. Obviously, if there was no Fed then the consolidated Fed/Treasury would use a regular bank account and they would clear payments by debiting and crediting private bank deposits. If the government wanted to create new money in that world (by running a deficit) their spending would mark up private bank deposit liabilities and taxes would redistribute some of those existing bank liabilities to the government’s spending account. If the government ran a balanced budget their taxes and spending would balance and there would just be redistribution of existing assets/liabilities. Taxes would not destroy liabilities as MMT claims unless the government was the one actually issuing the liabilities as banker to the entire country. Obviously, this isn’t remotely close to our reality.

This is important because we have specific government entities for specific reasons. And we break up their accounting and assets and liabilities to show exactly what the role of those financial instruments are. For example, the Fed holds trillions of dollars of MBS and T-Bonds due to QE. The Fed pays employees and operates their interbank clearing system, in part, by using the revenue from their balance sheet. Do the MMT people really believe this money isn’t there? Do they think it’s all a fiction? That would be outrageously misleading, but in the real world we precisely quantify those assets because they are real assets that have material economic impacts for real people. They aren’t just “destroyed” because some economic theory treats them as arbitrarily consolidated assets.

Consolidating the Fed and Treasury into one sector is about as useful as consolidating the entire world into one sector to show that there are no financial assets and liabilities in the aggregate. Why would that be useful? It wouldn’t, of course, because it obscures the specific details and how specific micro entities exist and perform specific functions.

Third, and related to the comments above, MMT misconstrues the concept of funding in a modern economy. MMT people like to claim that “taxes do not fund anything”. This is a misunderstanding of credit at best and completely wrong at worst.

For instance, let’s say we have two economies with $10T in domestic income. And both of those economies want to spend $2T on healthcare. Economy A funds that spending entirely from domestic income by redistributing money that already exists while economy B creates $2T of new money printed by the government. Now, which economy has more sustainable long-term spending? The obvious answer is economy A because they do not have any risk of creating new inflation because they funded their spending from existing money instead of creating any new money. Now, obviously, a government has a printing press so the government doesn’t HAVE to fund its spending from existing income. And that’s an important concept to understand when we talk about our government being able to “afford” certain things. But it’s also obvious that governments with large taxable income bases have a lot more spending flexibility because they can fund spending from existing money. So yes, taxes absolutely fund spending and the fact that a government can print money does not change that fact. Any theory that says otherwise is not describing reality.

Fourth, and arguably most importantly, MMT is specifically a theory of full employment using a Job Guarantee. The most interesting and experimental aspect of MMT is the idea that the government should give everyone a job and try to maintain price stability through that program. Contrary to popular opinion, MMT is not a theory of large deficits and government spending per se. It is specifically a theory of full employment and the Job Guarantee is how they achieve that. A Job Guarantee has been tried in only a handful of instances in the world and certainly doesn’t exist in any developed economies in the scale that MMT envisions. So no, this hasn’t been tried and MMT’s theory of employment definitely does not reflect reality.

In sum, the assumption of being “sovereign” is vague at best. And their treatment of reserve accounting is misleading at best and wrong at worst. And a Job Guarantee isn’t even a thing. So no, MMT doesn’t describe reality. Not even close.

¹ – Everyone funds their spending, but we won’t get into that MMT misconception since I’ve covered it in detail before. 

² – MMT people invariably respond to this with some hand-wavy generalization about how the government can’t hold its own liabilities. Aside from the fact that the Fed is a hybrid public/private institution, it should be clear that different government entities issue different types of assets for specific reasons. Different government agencies hold various forms of  other intra-government liabilities for specific reasons.  For instance, reserves serve specifically to allow the Fed to operate monetary policy. T-Bonds exist specifically to allow the Treasury to operate fiscal policy. The Fed holds T-Bonds by implementing monetary policy. Are we going to start claiming that the Fed doesn’t really have a $4.5T balance sheet made up of other government liabilities? Are we going to start saying it doesn’t really earn income from those T-Bonds? It doesn’t really fund its operations, in part, from that income?  Look, theories are nice and MMT has some good parts to it, but let’s get real – this theory isn’t reality.