Talk:Hyperinflation

From Wikipedia, the free encyclopedia

[edit] Talk moved from article

"Do there exist any examples anywhere of scientific notation being used on banknotes?"

I suppose 4.19 quintillion means 4.19×10¹⁸ (US notation) and not 4.19×10³⁰ (International notation). Maybe I'm wrong. In any case, could anyone please make it clear? Thank you. Sabbut 22:49, 26 Jan 2004 (UTC)

There is a problem with this figure, I think. If that was the inflation rate for a year, the inflation rate each SECOND would be about 13 Billion percent. Which means a doubling in price every 100,000th of a second. Every 15 hours would be a monthy rate of 3571.2% or a yearly rate of 42000% (ish) —The preceding unsigned comment was added by 68.124.129.219 (talk • contribs) 04:06, 5 October 2006 (UTC)

Please see my calculation below (#Contradiction on Hungarian pengő). If the monthly inflation rate was 4.19×10¹⁶ (41.9 quadrillion) percent (or 4.19×10¹⁴ times) per month, prices would double every 15 hours or so. If prices doubles every 15 hours exactly, then prices would quadruple every 30 hours, multiply by 8 times every 45 hours, and multiply by 2¹⁰ = 1024 times every 150 hours (6 days 6 hours). By the same token, prices would multiply 2.81475×10¹⁴ for 30 days and 6.3317×10¹⁷⁵ in 365 days.

See compound interest and time value of money for the correct formulae. --ChoChoPK (球球PK) (talk | contrib) 21:05, 5 October 2006 (UTC)

This line removed from article:

Nations such as Ghana in North Western Africa continue to this day to have inflation in the order of 30% per annum.

This is not an example of hyperinflation, merely of high inflation. --Geordieandy 13:14, 7 Mar 2004 (UTC)

Monthly rates of inflation in Brazil in the period 1978 - 1987 varied between 30% to 60%, i.e. several thousand percent per annum. Though I wasn't in Brazil in the early 1990's I believe inflation by then had decreased to much lower levels. One other significant (and well-studied) fact about Brazilian inflation is the existence of monetary correction for some financial assets. I would check Instituto Brasilerio de Economia (part of Fundacao Getulio Vargas -- perhaps too far to the right for my taste, that may have changed) or Instituto Brasileiro de Georgrafia e Estatistica. (IBGE) CSTAR 04:17, 13 Jul 2004 (UTC)

Yes I correct myself, inflation in the early 90's was also at 3000+ percent p/annum rate. (although high levels did also exist in the late 70's early 80's. That's one reason why I left!) CSTAR 04:41, 13 Jul 2004 (UTC)

Isn't moving the decimal place correctly referred to as revaluation? I've always understood devaluation to refer to a deliberate governmental inflation in order to take wealth from the general public and/or induce foreign investment. This modification would also involve adjusting the devaluation entry too. Chris Rodgers 18:57, 14 Jul 2004 (UTC)

This is debatable....

The widespread use of fiat money created the increased possibility of hyperinflation. (see seigniorage)

One could argue that fiat money is a consequence of hyperinflation and not a cause of it.

Can you name one example of hyperinflation in progress where the monetary authority switched to a fiat currency basis? There are examples of inflation pressuring the switch, but not to my knowledge hyperinflation. Stirling Newberry 00:24, 16 Jul 2004 (UTC)

Impossible to find an example of hyperinflation without fiat money/govt. printing presses. Spanish empire gold inflation and american colonies tobacco inflation never reached hyperinflation levels.--82.211.83.9 15:29, 8 February 2007 (UTC)

The caption for the 500,000,000,000 dinar note reads "the largest nominal value ever officially printed", yet lower down the article says "The largest denomination banknote ever officially issued for circulation was in 1946 by the Hungarian National Bank for the amount of 100 Thousand Million Billion Pengo (100,000,000,000,000,000,000)." Should the caption be corrected, or am I not understanding "nominal value"? --Earle Martin 11:53, 23 Jul 2004 (UTC)

The caption actually reads "the largest nominal value ever officially printed in Serbia" (emphasis added). Brianjd

what country's currency is the ake? Jdevine 15:43, 3 Sep 2004 (UTC)

Ottoman empire - see "The Great Debasement" Sevket Pamuk, A Monetary History of the Ottoman Empire. Yale 2000

why not put that info in the entry? Jdevine 19:43, 3 Sep 2004 (UTC)

[edit] Mississippi Scheme

An example of this is the "South Sea Bubble" under John Law.

John Law was responsible for the Mississippi Scheme.

I'd like to clarify my concerns with the current version of the article, following the reversion of some of my edits by Stirling Newberry.

My general concern is that it doesn't accurately describe the debate about various causes of hyperinflation. The article seems to me to be describing a stereotyped view of each side of the debate, which exagerrates the amount of controversy and obscures the amount of common ground. "Monetary" and "Crisis of confidence" explanations of hyperinflation are describes as if they are two completely different competing theories, held by different schools of thought ("monetarists" and "neo-liberals").

I think that the overwhelming majority of economists would see both monetary and confidence factors as being important as the initial 'root cause' of hyperinflation, with the importance varying from case to case. A lot of the economics is very uncontroversial. For example:

It's totally uncontroversial that a massive increase in the money supply will cause hyperinflation. Virtually any economist will agree that a big increase in the money supply is a possible trigger for hyperinflation, even if they don't believe it's the only one. To say imply that neo-liberal economists don't believe that some hyperinflations are initially triggered by monetary expansion is to caricature their position.
It's totally uncontroversial that loss of confidence in the monetary system has a big role in hyperinflation. To imply that monetarists don't see a role for reestablishing confidence in the currency - that just 'stopping the printing presses' is enough - is just misleading. Monetarists talk about ways of establishing confidence all the time.

I would be interested to hear from Stirling what specific inaccuracies were in my edits, and why he deleted the statement "The two models are closely related, and often economists may explain hyperinflation as being due to a combination of both effects."

Enchanter 20:44, Sep 30, 2004 (UTC)

[edit] Characteristics of hyperinflation

International Accounting Standard 29 describes four signs that an economy may be in hyperinflation:

But there are five items under that sentence. Brianjd

[edit] Hyperinflation under commodity money

Governments in a commodity or specie money system can

Confiscate assets.
Invade another country and "liberate" the money stock of the other country - also known as "pillage".
Change the basis of the unit of account or "peg".
Emit notes of credit - which while not legal tender, can be required to do business with the government.
Borrow.
Debase.
Shave.
Grant patents and sell stock.

All of the above have caused inflation in the past, but never hyperinflation. For a simple reason: the quantity which a government can inflate this way is limited. You can easily print 1000€ for each 1€ in existance. Confiscating enough gold/land/tobacco to x1000 the amount already in the market is almost unthinkable.--Flix2000 18:59, 8 February 2007 (UTC)

[edit] Removed Cleanup tag

A naked cleanup tag without explanation is not particularly helpful. Please do not add naked clean up tags - particularly with no edit summary - without listing in the talk section why the article needs clean up. Stirling Newberry 04:39, 11 Jan 2005 (UTC)

[edit] Contradictions

In the Hungary section its says: Previously, between 1922 and 1924 inflation in Poland reached 98%.

And in the Polish one: Previously between 1922 and 1924, Polish inflation reached 275%. Sitenl 22:44, 7 December 2005 (UTC)

[edit] removed UK

Intro says hyperinflation is 50% per month, this section says UK inflation was 30% per year. Doesn't seem to fit. What's the lowest rate of hyperinflation among other included examples?

[edit] add back uk

Smallbones, you can not arbitrarily remove a whole paragraph like that without discussion. You may be right. maybe 30%p.a. is not hyperinflation. But just because someone says it starts at 50% a month does not mean he is right. Accounting standards say Hyperinflation exists when price increases approach 100% in three years. That makes around 30% a year. Photomaniac 19:49, 13 December 2005 (UTC)

[edit] discussion: how much inflation is hyperinflation

Discussion is good. What are these "accounting standards"? It would be good for the article to have internal consistency; I don't have enough knowledge to comment on the actual definition, but this is so far short of 50%/month it the article seems out of step with itself. Can you elaborate on these other definitions? If they belong here then perhaps so does the example, though I've never heard of that period referred to as hyperinflation (yes, I live in the UK). Can someone provide a citation for the 50%/month maybe too? Notinasnaid 20:03, 13 December 2005 (UTC)

It's not clear that the UK reached the levels SUGGESTED in the IAS standards. It would take 30% annual inflation for a full three years. Did UK inflation get that high for that long? In any case the standard calls for judgement, and other indicators as well and this level seems so far removed from the other cases here that "hyperinflation" seems to be a gross exageration. A google search on "hyperinflation in the UK" and "hyperinflation in Britain" (with quotes) comes up with only 2 irrelevant articles (something about literature in the Edwardian period). I'd be inclined to keep the UK in here only if it doesn't have by far the lowest inflation level of the examples cited. I'll check.

Smallbones 21:31, 13 December 2005 (UTC)

Other "low" inflation levels: Dominican Republic 55% per year, Zimbabwe and Austria about 135% per year, there are a couple in the 200% range. Any cut-off is arbitrary, but I'll suggest 100% per year as being the absolute low, just to be consistent with all the other examples. BTW 50% per month is something like 13,000% per year. I'll remove UK and Dominican Republic, but am open to having them returned with some sort of reasonable explanation.Smallbones 21:50, 13 December 2005 (UTC)

According to official UK price level stats [[1]] (funky site look for CDKO series)price levels doubled (i.e. cumulative 100% inflation) between June 1974 and Sept 1978 (over 4 years) and didn't double again until April 1987 (over 8 years). If you want a different period March 76 - August 81 (over 5 years) almost doubles. Jan 75 - Dec 79 (almost 5 years) almost doubles. So it seems that the UK doesn't fit the IAS standard's suggestion (under 3 year doubling period). Smallbones 22:47, 13 December 2005 (UTC)

[edit] GAPP and International Accounting Standards definition

IAS Standard 29, paragraph 3: The Standard does not establish an absolute rate at which hyperinflation is deemed to arise. It is a matter of judgement when restatement of financial statements..... becomes necessary. Hyperinflation is indicated by charachterististics of the economic environment of a country which include, but are not limited to the following:

(a) the general population prefers to keep its wealth in non-monetary assets or in foreign currency....
(b)....
(c)Sales and purchases on credit take place at prices that compensate for the expected loss of purchasing power....
(d) interest rates, wages and prices are linked to a price index
(e) the cumulative inflation rate over three years is approaching or exceeding 100%

Trickyt 20:43, 13 December 2005 (UTC)

In the section "Hyperinflation around the world", both Poland and Romania experienced hyperinflation after WWI and WWII, respectively. Giving them both 2 distinct periods of monetary inflation. Does anyone plan on revising these sections to reflect that? Or are you just going for the highest inflation rate? The article has improved much since I put my 2 cents in some years back - so I really don't want to disrupt your hard work. Also, the quote "One of the firms printing these notes submitted an invoice for the work to the Reichsbank for 32,776,899,763,734,490,417.05 (3.28×1019, or 33 quintillion) Marks." is wrong, I believe. I will have to research it, but when I created my website (www.milliondollarbabies.com) I distinctly remember my original source referring it to be a number from the Reichsbank's Annual Report for 1923. I will dig it out and correct the entry if necessary. Any comments would be appreciated - you seem to have it all well in hand. Bobtheplanet 00:58, 28 January 2006 (UTC)

[edit] Wikicities link removed

I couldn't quite figure out what the Wikicities link to "money is free" was doing here. Wikicities looked a bit like Wikipedia with ads. The "money is free" article looked like a 6th grade version of this article.

If anybody can explain this to me ... please do. Smallbones 12:21, 26 February 2006 (UTC)

[edit] Reparations Just a Scapegoat

I changed part of the Weimar Hyperinflation description. It is a myth that the reparations imposed on Germany after WWI necessitated the government's resort to hyperinflation. Here is more information: http://www.libertyhaven.com/countriesandregions/germany/greatgermaninf.html

And if you just want the data: http://www.libertyhaven.com/countriesandregions/germany/ydefrep.gif --Isaac 19:05, 17 March 2006 (UTC)

This is a good source, but the article needs changing I think. It is right that this view should be represented, but as a view. The alternative view, that reparations did cause it, seems to be widely held and should also be reported as a view. Wikipedia requires that a neutral point of view present the different sources, even if you disagree with them, without showing a bias towards them. However, I'm not in a position to produce a citation for the opposite view, which would be required for balance. Notinasnaid 19:15, 17 March 2006 (UTC)

[edit] Nepers per second??!

The Nepers per second idea is interesting and from a purely scientific viewpoint it may be a more precise method of describing things. But I think it is out of place here because 99% of the readers will not have any idea what is being said. It is also out of place in its precision - it seems to describe a process that has some sort of control or pattern to it. Really though hyperinflation is something that's completely out-of-control. The digits are really just added as fast as the government can print them - so what's the use of describing it scientificaly?

Exponential notation has some similar drawbacks, e.g. it's never been used during a real hyperinflation espisode. But I think many people will at least understand it.

I'm certainly open to feedback on this. Smallbones 09:32, 20 March 2006 (UTC)

I added the nepers per second figures. The big advantage of them is that it's a rate that one can calculate with. It's non-trivial to see how "125% per month" compares with "50% per month", but you can immediately see that "308 nNp/s" is twice the rate of "154 nNp/s": it's doubling prices twice as fast. When you get to the region of billions of percent per month I think the percentages become quite meaningless. There's also the advantage that plain nepers can be used to measure the amount of inflation in a particular time, and it's then meaningful to add together the numbers of nepers for consecutive bits of inflation. It's a logarithmic phenomenon, so is best measured using a logarithmic unit, which is what the neper is.

As for the precision, I don't think we're giving unwarranted precision. Being out of control doesn't mean that it can't or shouldn't be measured precisely and scientifically. A substantial part of the material in this article is about the extent and rate of various episodes of hyperinflation, and these are stated using various numerical statistics. I certainly didn't change that aspect of the article's flavour, so if you think that giving numbers is inappropriate then you have a more general issue.

I think your point that most readers won't understand it is correct, at least initially. It's not common practice to describe inflation, interest rates, stock market movements, and so on, using nepers. (I wish it were, for the reasons that I described two paragraphs ago: they're logarithmic phenomena.) I have in fact never seen the neper used in discussion of inflation. So if we are to get the benefit to comprehensibility that comes from using nepers then an explanatory paragraph is in order. In any case, note that it is not necessary for every reader to understand nepers for it to be useful. I'm proposing adding to, not replacing, the traditional form of measurement. 81.168.80.170 23:18, 20 March 2006 (UTC)

There is a key point here: is this a practice used in published research on hyperinflation? If not, it sounds like original research, which doesn't belong here no matter how useful... Sorry, I see you already confirmed this. See Wikipedia:No original research. Notinasnaid 23:23, 20 March 2006 (UTC)

I don't think this counts as original research. It's not a new interpretation or explanation of hyperinflation, nor is it new numerical data. It's a unit conversion. It's analogous to converting between different scales for expressing magnitudes of earthquakes. There is some novelty in using this particular unit to express measurements of this phenomenon, but this novelty is only in the expression, not in the information itself. The unit is not novel, and it is an appropriate one for the phenomenon. Having read Wikipedia:No original research I conclude that that policy doesn't apply here. 81.168.80.170 00:05, 21 March 2006 (UTC)

I've got a great deal of sympathy for the nanoNepers, but am still not convinced. It's not quite "Original Research" - more like "Original expression in this context." It also won't communicate well to readers. My biggest objection however is the false precision it suggests. The article now is rife with conflicting measurements. Official inflation figures, as noted in the article, are essentially meaningless, but are mentioned anyway. The other main measurement seems to be how many zeros are on the currency - silly in its own way, but is there a better real measure?? I think the chaos in measurement reflects the chaos of the phenomenon - it's quite close to being non measurable.

If we want something that is more precise, comparable, and easy to understand, I'll suggest doubling period. But, if others (plural) think Nepers is understandable, I certainly won't stand in the way Smallbones 08:37, 21 March 2006 (UTC)

[edit] Face Vs. Scrap Value

In the article, it says that in coins, the scrap value often ends up enourmasly exceeding the face value, but can this happen with notes? I assume it does, but just thought I should ask here first. Can it? - ^R_H^e _o^d_t 13:26, 2 April 2006 (UTC)

Think toilet paper. I'm not kidding, if a roll of TP costs 1,000,000 coupons and the wad of coupons is bigger than the roll of TP, who is going to buy the TP? This is not a hypothetical example - it's from Ukraine, 1993.

Oh. Ugh! I can't imagine it's very comfortable. - ^Red _Hot 16:59, 12 April 2006 (UTC)

[edit] Repetition

There is a lot of redundant repetition in this article. 134.174.157.216 13:37, 26 July 2006 (UTC)

I agree. --Chochopk 16:21, 26 July 2006 (UTC)

[edit] Contradiction on Hungarian pengő

Section "Characteristics"

The most severe known incident of inflation was in Hungary after the end of World War II at 4.19×10¹⁶ percent per month (prices double every 15 hours) (The math is correct)

Section "Hyperinflation and the currency"

The Post-WWII hyperinflation of Hungary holds the record for the most extreme monthly inflation rate ever — 41,900,000,000,000,000% (4.19×10¹⁶%) for July, 1946, amounting to prices doubling every fifteen hours.

Section "Hyperinflation around the world"

The rate of inflation was 4.19 quintillion percent. (that is 4.19×10¹⁸%)

So I search the internet and found about equal number of results of either hypothesis =( . Then I gathered these numbers from Hungarian pengő:

The largest pengő denomination issued was 10⁸ b.-pengő (10²⁰ pengő), which was worth about US$ 0.2 on the day of issuance (11 July 1946). That means 1 US$ was approximately equal to 5×10²⁰ pengő.
Forint was introduced on August 1, 1946. At that time,

1 forint = 4×10²⁹ pengő

1 United States dollar = 11.74 forints = 4.696×10³⁰ pengő

Hypothesis 1

In July 1946, the inflation rate was 4.19×10¹⁶ (41.9 quadrillion) percent (or 4.19×10¹⁴ times) per month. Using this number, and assuming constant exponential inflation, we can derive the following

	Time required to multiply prices			Price multiplier for this much time
	2x	10x	100x	1 hour	1 day	1 week	2 weeks	3 weeks
If 1 month is 31 days	15.32 hours	2.12 days	4.24 days	1.0463	2.96	2004	4.01×10⁶	8.04×10⁹
If 1 month is 30 days	14.82 hours	2.05 days	4.1 days	1.0479	3.07	2581	6.66×10⁶	1.72×10¹⁰

4.696×10³⁰ / 5×10²⁰ = 9.39×10⁹, which supports this hypothesis.

Hypothesis 2

In July 1946, the inflation rate was 4.19×10¹⁸ (4.19 quintillion) percent (or 4.19×10¹⁶ times) per month. Using this number, and assuming constant exponential inflation, we can derive the following

	Time required to multiply prices			Price multiplier for this much time
	2x	10x	100x	1 hour	1 day	1 week	2 weeks	3 weeks
If 1 month is 31 days	13.47 hours	1.86 days	3.73 days	1.0528	3.44	5668	3.21×10⁷	1.82×10¹¹
If 1 month is 30 days	13.04 hours	1.8 days	3.61 days	1.0546	3.58	7560	5.72×10⁷	4.32×10¹¹

And while we know that at least one of the two numbers is erroneous, we might as well validate the inflation rates on Germany, Yugoslavia, Greece, etc. --Chochopk 09:43, 31 July 2006 (UTC)

[edit] Chile

The section on Chile was both inaccurate (140% is almost never considered hyperinflation, particularly when the average rate of inflation in the economy before had been 35% per year) and heavily dependent on Pinochet statements about the economy, which are both self-serving and incorrect. There was a coup before a plebescite, and the coup did not come on the back of a popular uprising, but was primarily backed by outside elements. Everyone admits that Allende's experiment ran away from itself between 1971-72, but only Pinochet apologists call the episdoe "hyperinflation". The central bank of Chile, for example, does not label the period one of hyperinflation, nor do the documents in the US library of Congress.

Stirling Newberry 08:50, 18 September 2006 (UTC)

Milton Friedman was involved in Chile for a few months. I'm not exactly sure of the dates... Carbonate 15:52, 19 September 2006 (UTC)

[edit] Mexico

The statement "Hyperinflation in Mexico eventually forced prices so high that in 1993 Carlos Salinas de Gortari had to replace the peso ($) with the nuevo peso (N$). The parity was N$1 for $1000 – in short, he stripped three zeroes from the peso." has nothing to do in the article. Highest annual rates in Mexico were ~100% in 1982–1983 and ~130%, which is far from hyperinflation.

Régis Lachaume (talk) 19:27, 24 October 2006 (UTC)

[edit] Hungarian banknote images

The largest denomination banknote ever officially issued for circulation was in 1946 by the Hungarian National Bank for the amount of 100 quintillion pengő (100,000,000,000,000,000,000, or 10²⁰). image (There was even a banknote worth 10 times more, i.e. 10²¹ pengő, printed, but not issued image.)

The banknotes in the images say Százmillió (100 million) and Egymilliárd (1 milliard, or 1 billion for you buffalo-eaters). How do these represent the far greater figures mentioned in the article? JIP | Talk 09:42, 2 January 2007 (UTC)

The "b" in b-pengő stands for billion in long scale (10¹²). See Hungarian pengő paper money#Postwar inflation series (1945-1946). --ChoChoPK (球球PK) (talk | contrib) 10:52, 2 January 2007 (UTC)

Oh. Now I understand. I thought it was just some symbol identifying the currency's name. Thanks. JIP | Talk 10:58, 2 January 2007 (UTC)

[edit] Root causes

Has there been one single instance of HYPERinflation in which huge amounts of money printing (paper or electronic) was not directly involved? Not just in the real world, I'll accept models, videogames, novels, dreams.... really has it ever happened?

—The preceding unsigned comment was added by Flix2000 (talk • contribs) 20:23, 14 March 2007 (UTC).

Hyperinflation may begin as rapid increases in checking account money, but ultimately ends with inflation of paper money because people who receive the money want to spend it as quickly as possible. Neither buyers nor sellers want to wait for checks to clear. Greensburger 15:52, 15 March 2007 (UTC)

Yes in an electronic money or book-entry money system there can be rapid inflation without actual physical printing.. but without massive money creation at the source (central banks in most cases), which can then be multiplied by fractional reserve banks, it is not possible and has never happened.--Flix2000 15:27, 29 March 2007 (UTC)

[edit] Mistake in the "Hyperinflation around the world" section?

The "Hyperinflation around the world" section mentions both the "Republic of China" and Taiwan which are basically the same.

Shouldn't "Republic of China" be "People's Republic of China" instead? Sinkiang is in mainland China, not Taiwan I believe. —The preceding unsigned comment was added by 155.198.116.81 (talk) 18:10, 22 March 2007 (UTC).

There is no mistake. The People's Republic of China was founded in 1949. China was in a state of civil war between 1945-49. At this time, the official title was "Republic of China". The communist won, the nationalist remained control of Taiwan only. --ChoChoPK (球球PK) (talk | contrib) 04:39, 23 March 2007 (UTC)

Retrieved from "http://en.wikipedia.org../../../h/y/p/Talk%7EHyperinflation_e7b1.html"