by Randall Wray, 2011

In last week’s post, I responded to Paul Krugman’s critique of Modern Money Theory (MMT), which argues that a sovereign government that issues its own floating exchange rate currency cannot face an affordability constraint—which means it cannot be forced into involuntary default on its own currency debt. His criticisms really boiled down to a misunderstanding over operational details—how banks work, how the Federal Government really spends, and the role played by the Fed in making all these operations work smoothly. I won’t rehash any of that here.
But what we were left with is the argument that if a government operates along MMT lines, then we are on the path to ruinous hyperinflation. Of course Austrians have long argued that all fiat money regimes are subject to these dangers—even ones that don’t follow MMT’s recommendations. MMTers are commonly accused of promoting policy that would recreate the experiences of Zimbabwe or Weimar Republic hyperinflations. These were supposedly caused by governments that resorted to “money printing” to finance burgeoning deficits—increasing the money supply at such a rapid pace that inflation accelerated to truly monumental rates.

No one wants to defend high inflation, much less hyperinflation. In his classic 1956 paper Phillip Cagan defined hyperinflation as an inflation rate of 50% or more per month. Clearly the zeroes would add up quickly, and economic life would be significantly disrupted.
The most popular explanation of hyperinflation is the Monetarist quantity theory of money: government prints up too much, causing prices to rise. However, as prices rise, the velocity of circulation increases—no one wants to hold on to money very long as its value falls rapidly. Wages are demanded daily, so as to spend income each day because tomorrow it will purchase less. What that means is that even though the money supply grows as rapidly as government can print notes, it never keeps up with rising prices. And the faster prices rise, the higher velocity climbs—eventually you demand hourly payment and run to the stores at lunchtime because by dinner prices will be higher.
Essentially, this was Cagan’s explanation for the fact that a simple version of the quantity theory did not fit the data: if prices rise so much faster than the money supply, how can we conclude that the hyperinflation is caused by “too much money chasing too few goods”? To fit the facts of experience, the quantity theory was revised to say that in a high inflation environment, the old quantity theory presumption that velocity is stable (which is necessary to maintain a link between money and prices) no longer holds.
So armed with the revised quantity theory, we can still claim that high and even hyperinflationary inflation result from too much money even though velocity is not stable (it rises as money growth lags behind inflation). And as Monetarists claim that government controls the money supply, hyperinflation must be due to government policy. Add to that the observation that in hyperinflation periods, the supply of government currency (paper notes) rises rapidly (with extra zeroes added). Further, government runs deficits as it finds its tax revenue cannot keep up with its spending, so is said to frantically print money to make up the difference—and that adds to the “too much money chasing too few goods”.
So most of the blame for hyperinflation falls to government money printing to finance deficits. (You can see the parallels to the US, UK, and Japanese situation today—large budget deficits that stuff banks full of reserves that they can supposedly use to pump up the money supply and prices.)
Solution? Tie the hands of government. In the old days, gold could serve as the anchor (and of course Austerian gold bugs want to return to those good old days). Today what we need is discipline, in the form of balanced budget amendments, debt limits, or for deficit doves like Paul Krugman a commitment to “eventually” slash deficit spending once recovery gets underway.
Sorry, that was a long introduction to today’s blog. As usual, I beg to differ with conventional wisdom. I do not accept any part of these arguments. Let’s take a Great Leap Forward to an alternative wisdom based on MMT.

I want to make three points.

1. When MMT says that government spends by “keystrokes”, this is a description, not a prescription. If critics were correct that government spending by “printing money” necessarily leads to hyperinflation, then most developed nations would have hyperinflation all the time. Because we all spend by keystrokes. And all governments that issue their own currency have to spend it before they can collect it in taxes (or bond sales)—no one else can create it. There is no alternative way for these governments to spend—yet they don’t experience hyperinflation (a point to be demonstrated). That we have to look to cases like Weimar or Zimbabwe (or way back in time to American Continentals) tells us a lot about the connection between “printing money” and hyperinflation. The causation cannot be found there.

2. Hyperinflations are caused by quite specific circumstances, although there are some shared characteristics of countries and monetary regimes that experience hyperinflation. I will not claim to fully understand the causes of hyperinflation—but the Monetarist explanation sheds almost no light on the experience. There is a sensible alternative. We’ll look at three well-known cases from the alternative point of view.

3. There is nothing in the current or prospective future condition of the US (or the UK or Japan) that would lead one to expect high inflation, let alone hyperinflation.
I will address the first point in today’s blog. Next week I will address the second and third points.
Most critics of MMT and of so-called fiat money in general imagine a past in which money was closely tied to a commodity like gold, which constrained the ability of both government and banks to create money “out of thin air”. The best example was the precious metal coin that supposedly gave a “real” value to government money, and forced government to actually get gold in order to spend. A strict gold standard with 100% gold backing against paper notes (issued by government or banks) accomplished the same task.
The reality was always quite different, however.

Obviously I cannot present a history of money here; please see Blog #12 (and look for Blog #13 next week) at the Modern Money Primer (http://neweconomicperspectives.blogspot....rimer.html) for a critique of the commodity coin story. Put it this way, gold and silver coins were the Sovereign’s IOUs that happened to be recorded on metal (rather than on paper or electronic balance sheets). In truth, coins usually circulated far above metallic value, at a nominal value proclaimed by the Sovereign (this is termed “nominalism”—the Sovereign set the nominal value through proclamation, just like today’s pennies that are worth a cent).

And their value was not necessarily stable: governments devalued them by “crying down the coin” (announcing they’d be accepted at half the former value in payments to government). They also “debased” them by reducing metallic content—which did not necessarily change their nominal value at all. To be sure, there are cases of relatively stable coinage and prices for long periods of time, but these are associated with strong and stable governments that adopted strong “nominalism” rather than “metalism” (the principle that a coin would be accepted at a value determined by its metallic content).
Indeed, the most unstable periods for coins coincided with weak Kings that resorted to weighing coins to catch clippers (those who would clip coins to obtain bullion), whilst rejecting light coins. This created “Gresham’s Law” dynamics, forcing everyone to weigh coins, accepting heavy coins in payment but trying to make payments in light coins. A real monetary mess. (See my Blogs 12 and 13 cited above.) This was finally resolved by going all the way to nominalism, coining only base metal and destroying the coin clipper’s business model.
Further, the gold standard did not operate in the manner imagined by today’s gold bugs. First, countries went on and off gold. When a crisis hit, they’d abandon gold. With recovery they’d go back on, until the gold constraints imposed forced them to go off when the economy crashed again. Rather than contributing to monetary stability, the gold standard destabilized the economy.

Second, no one really played by the rules. The temptation was always too great to leverage gold—to issue more IOUs than one could ever convert. (Even Milton Friedman admitted this, which is why he argued that while a gold standard might be ideal in theory, it doesn’t work in practice.)

Third, the periods of relative stability—Bretton Woods post WWII, or Pax Britannica pre WWI—were really dollar and sterling standards, respectively. In each case, the dominant nation agreed to peg the price of gold, and other nations pegged to the dominant currency. It really amounted to a buffer stock program for gold (price ceilings and floors for gold), with international trade actually taking place in pounds and then later in dollars (and the Bretton Wood’s gold safely impounded at Fort Knox).
Conditions for stability were difficult to maintain—which is why neither system lasted long. After WWI the sterling system could not be restored, and indeed set up the conditions for both Weimar (discussed next week) and finally Hitler. The Bretton Woods system collapsed in the early 1970s—lasting barely one generation. In both cases, collapse of these fixed exchange rate systems led to international turmoil.
And that is generally the eventual conclusion of most attempts to tie a domestic currency to some sort of fixed exchange rate standard (whether gold or foreign currency): it works until it collapses.
The gold bug and currency board aficionados are correct that a country that is experiencing high inflation can fairly quickly bring it down by adopting a strict external standard. Argentina did that with its currency board. But that then creates two problems: most countries cannot earn sufficient foreign currency to provide the fiscal policy space needed to keep the economy growing.

Second, there is no easy way off the currency board once it is recognized that fiscal policy space has evaporated making it possible to deal with a burgeoning problem of no domestic growth and rising unemployment. Argentina experienced a speculative attack on its dollar reserves (even though fiscal policy was quite tight, unemployment was high, and growth had evaporated—causing markets to doubt its ability to convert on demand) and it took a crisis to get off the dollar. Once it did so, it fairly quickly restored economic growth with the fiscal space provided by return to its own currency.
And that brings us to “how governments really spend”. Any government that issues its own currency spends by “keystrokes”—crediting the account of the recipient and simultaneously crediting reserves to the recipient’s bank. (It could print currency and make payments that way—but the effect will be the same because recipients would make deposits in banks, which would receive credits to their reserves.) I repeat, this is not a proposal. It is reality. There’s no other way. You cannot print up Dollars in your basement. Government has to keystroke them into existence before you can pay your taxes or buy Treasuries.

There can be somewhat complex operational procedures that involve special private banks as well as intermediate steps involving the central bank and treasury bank accounts, but the end result is always the same: bank accounts get credited and banks get reserves. Usually the treasury then sells bonds to let banks earn higher interest than they receive on reserves. (See the discussion last week.)
On a floating exchange rate that is the end of the story. Banks can use their reserves to buy treasuries; depositors can demand cash (in which case the central bank ships it to the banks while debiting the banks’ reserves). But no one can return government IOUs to demand gold or foreign currency at a fixed exchange rate. There is no affordability constraint; there is no foreign currency or gold constraint. Government can meet all demands to convert to cash; and can pay all interest as it comes due through additional keystrokes.

However on a fixed exchange rate, or currency board, central bank and treasury IOUs have to be converted to foreign currency (or gold). And for that reason, a prudent government must limit its keystrokes. It can run out of foreign currency or gold. It can be forced to default on its promise to convert. That of course counts as a default on debt. Its “affordability” is called into question by markets when they doubt ability to do the conversions at the promised exchange rate. Imprudence is deadly. History is of course filled with imprudent governments—those that issue too many IOUs relative to the reserves promised for conversion on demand.
The floating rate provides policy space that can be used by prudent governments to pursue domestic policy goals with a greater degree of freedom. History is of course filled with imprudent governments there, too. There is no substitute for good governance. Still, it is curious that except for the losers of WWI (plus Poland and Russia, which were on the winning side but lost the war—so to speak–anyway), there are no cases of nominally democratic Western capitalist countries that have experienced hyperinflation in the past century. And if we limit our data set to those with floating currencies, there aren’t any with exchange rate crises, either.

Quite curious, no? Only countries with fixed exchange rates or other promises to deliver foreign currency or gold (such as debts in foreign currencies) seem to have hyperinflations and currency crises. And that always seems to come down to imprudent expansion of these IOUs relative to ability to actually deliver the foreign currency or gold.
While it appears that the fixed exchange rate guarantees prudence that is a foolish notion. The fixed exchange rate introduces exchange rate crises plus involuntary default as possibilities, in the pious hope that government will be prudent. Unfortunately, governments on fixed exchange standards more often adopt the prayer of St. Augustine: “Lord please make me prudent, but not just yet.”
Far from ensuring prudence and protection from high inflation, when sovereign government promises to deliver foreign currency it actually exposes the nation to Weimar Republic hyperinflationary risks—as we will see next week.
To conclude. It is not true that fixed exchange rates eliminate risks of exchange rate crises and hyperinflations. Because sovereign governments are not necessarily prudent.
And even if they are, their banks are not necessarily prudent. (Think Ireland! While the government was the paragon of financial prudence, its banks lent in foreign currency until the cows came home. When borrowers defaulted, the Irish government took on all the foreign currency debt—quite imprudent!)
Further, it is not true that floating rate standards invariably lead to hyperinflations. If that were true, we’d have hyperinflation all the time.
Finally, it is not true that ability to “print money” through keystrokes necessarily leads to hyperinflations. All sovereign governments that issue their own currency spend by keystrokes. Even if they promise to convert at a fixed exchange rate, they still spend by keystrokes. If keystrokes invariably cause hyperinflation, we’d have hyperinflation all the time.
We don’t. Hyperinflations are unusual outcomes. Next week we’ll take a look at some. And we’ll take a look at the case made by hyperventilators for prospective hyperinflation in the US.