Monetary Sovereignty and the Velocity of Money

I had another interesting discussion with MMT supporters on Twitter. This time, it was about monetary sovereignty and the velocity of money. There are many common misconceptions about these topics, so I figured it’s best to write a brief post.

It’s been a long-standing critique of MMT that it does not apply to all countries. For example, you don’t hear talk of MMT in Brazil, Argentina, Turkey, Russia, etc. The reason is high inflation. Any attempt at fiscal stimulus during recessions gives rise to high inflation. Such countries, despite issuing their own currencies, are nominally-constrained, which contradicts a basic MMT premise. In other words, printing money results in higher prices as opposed to improved output and employment.

MMT’s response is to say that such countries (e.g., emerging and developing economies) lack monetary sovereignty. While they meet two of the conditions for monetary sovereignty, namely issue their own currency and have a common fiscal authority, they fail the other two tests: their governments borrow in foreign capital markets and they run large trade deficits. For a good overview of this position, please refer to a recent podcast by Fadhel Kaboub, a prominent MMT economist.

I have to say the MMT message to emerging and developing economies is somewhat depressing. If only you could have a diverse, vibrant economy that is largely self-sufficient, or even better, have a strong export sector, you will have monetary sovereignty. In other words, the path to monetary sovereignty is on the real side of the economy. The problem is that real constraints are notoriously difficult to overcome. The MMT offer to emerging markets is a kind of Catch 22: overcome real constraints to gain monetary sovereignty, and then you will be in position to use your new monetary sovereignty to improve the real economy.

The good news is that monetary sovereignty is not about real constraints. It is all about nominal and institutional factors, namely independent central bank and commitment to price stability. The example, I give, is the US in the 1970s. The US met all 4 criteria for monetary sovereignty as laid-out by MMT[1]. Inflation, nonetheless, was still a binding constraint. In other words, the US could not have printed its way out of the stagflation of the 1970s.

To understand what’s going on we need to talk about money demand. By now, readers of my blog should be familiar with the components of money demand. Transaction money demand represents the demand for money by agents who do not use credit to facilitate purchases but prefer to spend out of prior-period incomes. To illustrate, entrepreneurs borrow from banks to finance the production of goods and services. Banks respond by creating money. The newly-created money flows from the entrepreneurs to the workers in the form of wages. Workers now have the money to buy the very same goods and services they produced. Entrepreneurs receive the money back as revenue and use it to pay off their credit-line debt incurred at the beginning of the period, upon which the money is destroyed.

If this process of money creation and destruction took place within the same period, the money balances at the end of each period would be exactly zero. The velocity of money would be infinite, which is a non-sensical result. The reason we observe velocity is because this process extends beyond the current period. Inventory takes time to produce, and workers want to accumulate pre-cautionary savings to prepare for the eventuality that they may lose their jobs. It is because of this demand for money (aka, transaction money demand) that money persists at the end of the period, allowing us to observe the velocity of money. In other words, in an endogenous money world, where agents can create and destroy money at will, velocity is not a measure of how many times money changes hands. Instead, velocity is a measure of money demand — how much money people want to hold in proportion to their incomes (or NGDP, in the aggregate).

The second component of money demand is what I refer to as asset money demand (Keynes’ called it speculative demand for money). This is where things start to get interesting. Asset money demand represents savings held in the form of money as a long-term investment (not for short-term spending). To visualize asset money demand, think of a person who believes her stock portfolio will decline in value. Accordingly, she sells some of her stocks, allocating a larger portion of her long-term savings into money. That demand for money is asset money demand.

Monetary sovereign countries are countries where domestic agents demand domestic currency to satisfy their asset money demand. In the example above, if that person were an American, she will sell her US stock portfolio for US dollars and gladly continue to hold such dollars either in a bank deposit or money-market mutual fund. However, if she were Brazilian, she would sell her Brazilian stocks for reals but immediately exchange such reals for US dollars. That’s the difference. The US dollar is considered a store of wealth, a reserve capable of satisfying asset money demand, while the Brazilian real is not.

In countries where domestic agents prefer foreign currency for their asset money demand, the domestic currency is held primarily for the purpose of facilitating near-term spending. In other words, the primary component of money demand is transaction money demand. The price level is determined by the foreign-currency exchange rate, and economic conditions are dictated by foreign capital flows, giving rise to a positively-sloped Phillips Curve. Any fiscal stimulus in the form of money-financed fiscal deficits simply boosts inflation because domestic agents attempt to exchange the newly-created money into FX, driving down the value of the currency and pushing-up domestic prices. It is precisely for this reason that such economies have very little policy space for countercyclical policies. As a result, they face the difficult choice of imposing capital controls, that stifle much needed foreign investments. Another option is to borrow from foreign-capital markets because only by supplying FX can such governments affect domestic market conditions. In other words, foreign government debt is often not the cause but the symptom of monetary sovereignty loss.

The same thing was happening in the 1970s in the US, the only difference being that the reserve of choice was not FX but gold. Despite the gold exit by the Nixon Administration, domestic agents continued to prefer gold for their asset money demand. The capital flight to gold during recessions is what gave rise to inflation as the price of gold determined the domestic price level as illustrated on Chart 1. For all practical purposes, the US lacked monetary sovereignty in the 1970s because it lacked the fiscal and monetary policy space to implement full-employment policies.

Chart 1 (Source: Fred)

With respect to velocity, the observed historical record can be explained by expressing the velocity of money in terms of the velocity of transaction money and adjusting for asset money demand (Chart 2).

Chart 2 (Velocity of Money)

The velocity of transaction money should be stable. That’s because transaction money demand is a stable proportion of NGDP as production schedules and desired pre-cautionary savings are not subject to large fluctuations. This means that observed fluctuations in velocity are largely driven by asset money demand. When asset money demand is high relative to the total supply of money, velocity declines and vice versa.

The velocity of money in countries without monetary sovereignty should be subject to small fluctuations within a narrow band. That’s because asset money demand is a small component of domestic money supply since agents prefer to hold FX instead. These are the precise conditions for the quantity-of-money equation to work. Any injection of new money causes prices to rise, hence there is very little fiscal policy space. Furthermore, peaks in velocity should coincide with recessions as agents flee the domestic currency, and the little asset money demand there is declines even further.

The opposite pattern is observed in countries with monetary sovereignty. Since domestic agents are willing to hold domestic currency for their asset money demand, velocity is subject to larger fluctuations. The quantity-of-money equation breaks down because money demand is unstable as evidenced by large fluctuation in velocity. Furthermore, velocity now peaks in expansions. That’s because asset money demand declines during economic booms (going back to the example above, investors reduce cash allocations if they expect equity prices to rise). During recessions, the opposite occurs. Asset money demand rises as people expect declining incomes, causing velocity to decline as well. This is perfectly borne in the US data as illustrated in Chart 3.

Chart 3 (Source: Fred)

During the 1970s, the US lacked monetary sovereignty as people preferred gold for their asset money demand. Therefore, velocity exhibited small fluctuations with peaks coinciding with recessions. In the 1980s, things started to change. The Volcker shock established the US dollar as the global reserve currency capable of satisfying asset money demand. Since then the velocity pattern has reversed — fluctuations are much larger with peaks coinciding with booms.

I prefer to call this the Volcker dividend. Volcker re-established the US dollar as a store of wealth. As a result, the US regained its monetary sovereignty, which was lost in the aftermath of the gold exit. This created the policy space for active monetary and fiscal policy ever since, giving us the possibility to consider and implement full-employment policies. As exemplified by Volcker, the key to monetary sovereignty does not lie in real factors, but in commitment to price stability and central bank independence.

[1] With respect to trade deficits, the US ran a trade surplus in the first half of the decade and a small trade deficit toward the end of the decade, that was puny in comparison to the deficits that followed in the 1980s.

[2] Someone could argue that in aggregate, individuals cannot get rid of the money by buying stocks (or create more money by selling stocks). That’s true, but what is happening is that the price of stocks adjusts such that people are okay holding the money that does exist. By the way, this is the connection between monetary policy in countries with monetary sovereignty and asset booms and busts, but that’s a topic for another day.