Why you won’t find hyperinflation in democracies

By Felix Salmon
September 4, 2012
World Hyperinflations".

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

There are those who believe that the length of a mathematical paper is inversely proportionate to how interesting it is. Something similar can be said about the new paper — short and absolutely first-rate — from Steve Hanke and Nicholas Krus, entitled “World Hyperinflations“. It’s technically 19 pages long, but the first 12 are basically just throat-clearing, and the last two are references. The meat is the five pages in the middle: three pages of tables, and another two of footnotes, detailing every instance of hyperinflation that the world has ever seen.

Hyperinflation, here, has a clear quantitative definition: prices rising by at least 50% per month. (Remember that, the next time some scaremonger starts talking about how US monetary policy risks causing hyperinflation.) And after some three years’ work, Hanke and Krus have managed to come up with an exhaustive list of every hyperinflationary episode in history — 56 in all, or 57 if you include North Korea in early 2010, where the data aren’t solid enough to merit inclusion in the list.

Every entry gets its own footnote, and while there are a lot of relatively easy-to-obtain IMF publications in there, there’s also no shortage of much more obscure source material: Simeun Vilendecic’s Banking in Republika Srpska in the late XX and early XXI century, for instance, or Abram van Heyningen Hartendorp’s 1958 History of Industry and Trade of the Philippines.

The earliest hyperinflation on the list came in France, at the end of the 18th Century, when inflation hit a monthly rate of 304% in mid-August 1796. The famous Weimar hyperinflation in Germany is pegged as taking place between August 1922 and December 1923; it reached a monthly peak of 29,500% in October 1923, with prices increasing at 20.9% per day, and doubling every 3.7 days. And the longest period of hyperinflation comes in Greece, which saw hyperinflation for a whopping 55 months, from May 1941 to December 1945. There’s no particular reason, looking at this list, why Germany should have been particularly scarred by hyperinflation, to the point at which it fiercely attacks even the possibility of relatively modest inflation, where France and Greece (not to mention Hungary or China or Argentina) have been much less deeply affected.

There is, however, a very strong correlation between the length of time that a period of hyperinflation goes on, and the levels that it can reach at its height. If you look at the top six hyperinflations on the list — which include both Germany and that 55-month period in Greece — all but one lasted for longer than a year. Meanwhile, five of the bottom six hyperinflations took place in just a single month, with the sixth lasting just three months.

At their highest, the numbers start to beggar the imagination: in mid-November 2008, for instance, inflation in Zimbabwe reached a monthly rate of 79,600,000,000%. That’s 79 billion percent per month. At that rate, prices pretty much double every day. And Zimbabwe doesn’t even manage to grab the top spot: in July 1946, Hungary saw hyperinflation of 41,900,000,000,000,000%. That’s 42 quadrillion percent in one month, with prices doubling every 15 hours.

The real value of this paper is its exhaustive nature. By looking down the list you can see what isn’t there — and, strikingly, what you don’t see are any instances of central banks gone mad in otherwise-productive economies. As Cullen Roche says, hyperinflation is caused by many things, such as losing a war, or regime collapse, or a massive drop in domestic production. But one thing is clear: it’s not caused by technocrats going mad or bad.

For that matter, there are no hyperinflations at all in North America: the closest we’ve come, geographically speaking, was in Nicaragua, from 1986-91. In fact, if you put to one side the failed states of Zimbabwe and North Korea, there hasn’t been a hyperinflation anywhere in the world since February 1997, more than 15 years ago, despite the enormous number of heterodox central-bank actions in that time.

All of which is to say that hyperinflation, in and of itself, really isn’t anything to worry about. It’s pretty much impossible to predict — and if your country has hyperinflation, it almost certainly has even bigger other problems. In fact, I’d hesitate to categorize hyperinflation as a narrowly economic phenomenon at all, as opposed to simply being a symptom of much bigger failures at the geopolitical level. Those failures are exacerbated by hyperinflation, of course: there’s very much a vicious cycle in these episodes. But you only ever find hyperinflation under extreme conditions, and, with a single exception (Peru), I’m not even sure I can find any genuine democracies on this list.

Update: As many people have helpfully pointed out, Weimar can definitely be considered a genuine democracy, even if it was suffering extraordinary geopolitical burdens.


We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/

Felix, interesting article, but I’m afraid at the end you fell victim to the No True Scotsman move. Weimar Germany was a democracy and had hyperinflation, so apparently it wasn’t a “genuine” democracy.

https://en.wikipedia.org/wiki/No_true_Sc otsman

I buy the conclusion that if you have hyperinflation you have other things to worry about, but that doesn’t mean I can’t spot logical fallacies when I see them.

Posted by turingfan | Report as abusive

I agree the Hyperinflationists have a lot to answer for, but I also believe it is a case of managing expectations. With many expecting “Hyper” inflation, a small dose, let’s just say 49% per month, could be spun as a victory because it was not the so called “Hyper”.

Posted by MadNumismatist | Report as abusive

A tl;dr version…

1) Inflation happens only in majorly disrupted states
2) Democracies, by definition, cannot be majorly disrupted
3) Therefore, “you won’t find hyperinflation in democracies”

And off to breakfast!

Posted by economicslol | Report as abusive

Germany in 1922 was run by a Representative Parliamentary Democracy during the so called “Weimar Republic” years which succeeded the old Imperial rule in 1919. They were a democracy and they had horrific hyperinflation…

Posted by FifthDecade | Report as abusive

Mr. Salmon — I agree with your general point, that hyperinflation is merely a symptom of a broken political (and social and economic) system. But not to think of the Weimar Republic as a “democracy” is inaccurate. Germany after the Great War had immense problems, to be sure, mostly stemming from the defeat (the search for scapegoats or the mistaken policy of reparations imposed on the country by the victorious Allies), but give them the credit of holding elections, a free press, private enterprise, and other features of a “democracy”.

Posted by mlnberger | Report as abusive

Any discussion of the 1923 hyperinflation of Weimar Germany that does not examine the effects of the reparation demands and the French occupation of the Ruhr is woefully incomplete.

Posted by Ed62 | Report as abusive

Alfonsin’s Argentina was a democracy under hyperinflation. But that’s nitpicking, as ‘extreme conditions’ holds up well enough as a phrase (in the 1922 Weimar as well). There’s an issue is the line to be drawn to class an economy as hyperinflationary as I’ve often seen 20%/month as the benchmark, though agreed shilling about hyperinflation in US (and other industrialized nations) normally comes from those who don’t know what they’re talking about and just think the words sounds good. Anyway, nice catch on the paper, Salmon. Interesting stuff.

Posted by ottorock | Report as abusive

Israel in the 1980s?

Posted by dsquared | Report as abusive

I was also under the impression that Brazil was a democracy at the time of its bout of hyperinflation.

Posted by ryanmburke19 | Report as abusive

“There’s no particular reason, looking at this list, why Germany should have been particularly scarred by hyperinflation”

Germany was not scarred by hyperinflation based upon the list of other nations that experienced hyperinflation. Germany was scarred by hyperinflation based upon the fact that its period of hyperinflation led directly to the collapse of what you now recognize was a democracy, the rise of a totalitarian regime, a war of aggression that nearly destroyed Europe, and a genocide.

Rightly or wrongly, Germany “fiercely attacks even the possibility of modest inflation” because it wants to avert the conditions that led to Hitler, not because it is worried about becoming Argentina.

Posted by AndrewCase | Report as abusive

Generally, those who complain about hyperinflation in the context of modern developed peacetime democracies, are the sorts of folks who consider 5%/year to be “hyperinflation”–and are worried more about devaluation of their investment portfolios than about the collapse of civilization.

Posted by EngineerScotty | Report as abusive

Good find, and from the Cato Institute no less.

I am also going to take issue with your no democracy take. The French inflation of 1796 was sandwiched between the French Revolution and the rise of Napoleon (in the aftermath of the collapse) and from my reading, during this period France has most, if not all, criteria necessary to be called a democracy. There is a well written account of this French attempt to back their currency with lands expropriated from the aristocracy called “Fiat Money Inflation in France” by Andrew Dickson White, President of Cornel University, in 1912.

http://lynncoins.com/fiat-money-france.h tm

Posted by Calinvest7 | Report as abusive

“In a social democracy, all roads lead to inflation.”

Posted by JJButler | Report as abusive

This article is really puzzling. Logically poor to say the least. Brazil was a democracy and had inflation of 82,39% in June 1994. Also, 50% per month is hyperinflation, and if it falls to 30% per month it is not. If one waits 4 months, in one year, the effect will be the same with these 2 inflation rates. The results will be the same, you ll only take longer to get there, but will nevertheless.

Posted by creazyrequest | Report as abusive

IASB defines hyperinflation as about 100% over a three year period.

While 50% per month can, without a doubt, be characterized by hyperinflation I believe a far lower percentage qualifies.

Inflating our troubles away is the only solution sufficiently expedient for cowardly politicians (i.e. ours). It will be characterized as “growing out of our debt” but it will be what we all know it to be.

Posted by EvoShandor | Report as abusive