Saturday, August 16, 2014

Positive Feedback Theory of Hyperinflation

I contend that Hyperinflation is a set of Positive Feedback loops.   There are many different ways to describe these loops.  One is that after a government has a large debt and a large deficit that it can get into a loop where the more people stop rolling over their bonds the more the central bank has to make new money and buy bonds, so the government has cash to operate.  But the more new money it makes the more inflation there is and the less anyone wants to hold bonds.    So the phenomenon can feed on itself and grow.  There are more loops, but this gives you one way of explaining one part of hyperinflation.

In comments on the previous post Tom said this was circular reasoning.  A good example of circular reasoning is:

"The Bible is the Word of God because God tells us it is... in the Bible."

In the above the reasoning is circular.  If the Bible is the word of God, then in the Bible God tells you so.   But if it was not the word of God, then God did not say it was.  It is not a good logical argument.

In a positive feedback phenomenon there is a feedback loop made up of real things.   It is not just circular reasoning.   You can watch and measure what is going on in the real world.  You can see how one part pushes another, and that pushes so that it comes back around magnifying the phenomenon.

I think it will help readers to understand my examples of positive feedback loops below, and their common characteristics, before looking at hyperinflation:

Avalanche

A bunch of snow can be sitting on a steep mountain so that it is almost ready to slide.  If something, say a little rabbit, then causes some of it to move, then that can cause more, which causes more, etc.  A small trigger can set off a chain reaction that causes a huge amount of snow to slide down the mountain.

Forest Fire

If there is a high enough density of burnable material in the forest, then a single ignition can set off a chain reaction where a whole forest burns down.

Earthquake

There are two plates of earth touching each other.  But they are slowly pushing past each other.  For many years the ground just slightly flexes under the strain.  However, at some point part of the ground rips.  After the first part rips the strain is too much for the next part and it also rips. This chain reaction continues down the fault line often for miles.

Volcano

Water is mixed in the lava but under such pressure that it does not turn into steam even though it is very hot.  However, if some of the lava on top breaks through the earth dam around the lava and spills out then there is less pressure on the lava below.   If there is enough less pressure that the water turns to steam it can throw off the top of the lava.  This then reduces the pressure on the lava further below which then also releases steam.  This chain reaction can continue till huge quantities of lava are thrown into the air as volcanic ash. From space you can see the ash and water vapor after an eruption.

Hurricane

If the ocean is over 82 degrees F and the ocean is hotter than the air then it can warm up the air as wind blows over he ocean.  The hotter air rises and makes more wind.  The faster the wind goes the more heat is transferred from the ocean to the air.
The more heat there is in the air, the faster the wind blows.  So you get a positive feedback loop leading to really strong winds.

Domino Avalanche

Set up dominoes so that the first one will knock over 2 others, and each of those will knock over 2 more, etc.   So with each time increment there will be twice as many dominoes falling.   One little push can then cause many dominoes to fall over.

Landslide

This is much like an avalanche but earth instead of snow.   There is a bunch of earth on a slope and once some of it starts to break more and more follows till there is lots of earth sliding down.

Earth Dam Failure

If a little bit of water goes over or through an earth dam it can erode a bit of the dam.  This makes for a bigger path for water and so water flows faster.  This then removes earth faster and so water flows faster and faster.   With this positive feedback loop, an earth dam can be quickly destroyed.

Snowball

With the right hill and snow you can start a small snowball rolling down the hill and it will grow and grow and grow till it gets to the bottom of the hill.

Ping Pong Balls and Mousetraps

If you put a bunch of ping pong balls on set mousetraps and trigger one, you can get a chain reaction.   It only takes a small amount of energy to trigger a mousetrap and then the ball flies off with lots of energy to bounce around and hit others.   This is not part of nature, but the video is fun and does illustrate the idea of a chain reaction.

Common Characteristics for these and Hyperinflation

In each case there is some set up condition.  Something that has to be in place for a growing chain reaction to be possible.   Under the right conditions, a small thing can set off a large chain reaction.  Once the chain reaction starts it is very hard to stop it.  Eventually it will burn itself out and stop on its own.   Under the "proper conditions", we say there is a risk of this chain reaction.  Even if the reaction has not yet happened, an expert can still see that there is a risk of it happening.

Because a small input can move things to another state, the initial state can be viewed as unstable.   We also tend to see positive feedback loops as causing destruction.

The positive feedback loops above have some form of stored energy which is used up in powering the chain reaction.  The stored energies above are gravitational potential energy, chemical energy,  heat energy, or elastic potential energy.  I think the equivalent in hyperinflation is the total value of all the currency and bonds in that currency being used up.

The best prediction you can give for most of these is "a high risk" and not an exact date.   Earthquakes, volcanoes, forest fires, avalanches, etc. are things where an expert can tell when there is a danger, but still not be able to say  exactly when the trigger will be.    I think it is the same with hyperinflation.

It would make as much sense to accuse each of these real world phenomenon of circular reasoning as it does to claim this of my explanation of hyperinflation.

Now on to Hyperinflation

Once you understand the above examples of positive feedback loops you should then read the different ways of explaining the positive feedback loops in hyperinflation

1. I concede that postulating a positive feedback loop is not circular reasoning. But any explanation which is not independently testable flirts with it. That was my problem. Replace your "feedback loops" with X and say that X causes hyperinflation, except when X isn't strong enough, and the only way to know if it is strong enough is if hyperinflation is present. In this case X, no matter what X is, is part of a circular reasoning loop. Maybe that's not the best way to describe it, but in any case it's logically irrelevant, even if it's logically consistent (internally) and consistent will all observed data. Occam asks what value the X hypothesis adds? Why not just cut it out? The story is completely different if you can independently measure X: then the X hypothesis can be tested. What I asked you to do was to provide a probability (one means of measuring X) of hyperinflation per year, over the next several years, for a host of nations. If not that a formula, or at least a strategy for devising such a formula. That would constitute a means of testing the X hypothesis (at least eventually). If your feedback hypothesis can be used as a basis of a model to compute those probabilities, that would be about as good as we can hope for.

I can propose an alternative testable hypothesis right now: the probability of hyperinflation in any one economy in any one year is independent year to year and is equal to the ratio of the population of people who are male, left-handed, and have type B+ blood to the full population in that economy.

This is a simple, quantifiable and testable rule. I don't have high hopes for it. I won't be gambling any money on it. It's not a particularly clever or insightful or meaningful rule, but it has one big advantage over what I've seen here: testability. We have a conceivable way to now decide if my hypothesis is correct or not... even if we can't get our hands on all the data we need to make the calculations for every economy. Perhaps we can devise ways to approximate it in some cases. This opens some doors.

But what I hear on your blog over and over is that we are in terrible grave danger... but I never see the numbers. How much danger? Moon sized asteroid hitting the Earth in the next year danger? Or I'm going to take the last drop of coffee at work so I have to make a new pot danger? How will we decide if you're likely wrong or not going forward?

Sorry, I made the above comment w/o reading your full post (I stopped when I saw my name mentioned)... so now I'll go back and hopefully you'll address my concerns (and I'll apologize!) :D

1. Seeing hyperinflation repeatedly *not* happen, year after year, is NOT evidence that the feedback loops operate like an avalanche and are putting many countries in grave danger. This evidence is instead more consistent with many other hypotheses, including an alternative feedback-loop avalanche hypothesis which is exactly the same as yours, but ranks the current risk for this happening as "very low" (perhaps < 1%) for the 1st world countries you're concerned about.

2. I never claimed that hyperinflation not happening was evidence of my positive feedback theory of hyperinflation.

The put options on the Yen seems consistent with the market believing there is less than a 1% chance of hyperinflation in Japan. I think it is much more than that. I don't know the number but I feel like it is more than 50% in the next 2 years. If I am right and the market is off by a factor of 50, I will do very well. It does not matter to me so much if the real odds ought to be 40% or 60%, I think the market is way off on this because it does not understand the positive feedback nature of the problem.

3. The average currency is dead in about 25 years. If we simplify to say the hyperinflation happens in one year, then there is about a 4% chance of hyperinflation in any given year. Really hyperinflation might take more than one of those 25 years, so it is higher than 4%. I agree that advance western democracies are averaging much longer than 25 years. But even if we give them 100 years there is still about a 1% chance of hyperinflation on an average year. After the government debt is over 200% of GNP, spending is twice taxes, and the central bank doubles the money supply in 2 years, you don't have an average year's odds. It is much higher. So I am sure that the market has priced the risk of Yen hyperinflation too low. So I am sure it makes for a good gamble, whatever the exact odds are.

4. Thanks Vincent. That's helpful. Care to outline a possible technique for calculating how much higher the probability gets as a function of government debt, spending relative to taxes, and central bank base money expansion? I'm not asking for the world here, but only for a description of a strategy for how you would theoretically go about establishing such a calculation? Are you proposing instead of a theoretical approach, an approach based solely on fitting a curve of some kind (e.g. polynomial), with no particular reason behind the functional form other than convenience, of perhaps an approach which establishes a functional form (based on theory) with some parameters to be fit to past data to make it applicable to any one particular economy (akin to Jason's approach)?

BTW, the following article is not about hyperinflation, but you might enjoy it anyway:

http://informationtransfereconomics.blogspot.com/2014/08/rationality-and-entropic-forces.html#comment-form

I think it speaks to your estimation that the market has it wrong. What I find intriguing about his idea is that it addresses critics of rational expectations, and validates their complaints to some extent, but yet says that in the end their complaints perhaps don't matter too much in a large market: it will tend to be efficient and appear "rational" anyway.

Elsewhere Jason has stated that perhaps 90% of the movement of big macro variables such as the average price level, might be due to these entropic forces which lie beyond the realm of analysis about particular human motivations (i.e. trying to reason out how "the market" will "rationally" react in various circumstances), and perhaps 10% of the variation we see is human motivated in a way not explainable by these entropic forces.

In trying to explain this idea to a third party I likened it to waves on the ocean, and how they contribute to the mean sea level at any one point: each individual human contributes a wave of a particular volume, speed, direction and timing, but overall, his contributions don't matter most of the time aside from the differentially small volume of water he contributes... although there are occasionally "Rogue Waves" in which a larger than expected number of human decisions (waves) line up and create a crest or trough significantly different than the mean. I realized that one problem with the rogue wave analogy is that it does not take account of any positive feedback effects. I ran that analogy past Jason to see if I was capturing his point, and he was more or less OK with it. Commenting on my positive feedback point, he agreed and said that sometimes human behavior causes what should have been a 2 sigma event to be a 4 sigma event (e.g. a "panic").

I don't think Jason is prepared to mount a strong defense of his 90%/10% breakdown of entropic to human forces: he's stated that's a guesstimate based on what he's seen since looking at macro.

Here's Jason's 2-sigma vs 4-sigma comment (it's short) in response to me:

http://informationtransfereconomics.blogspot.com/2014/08/zen-koan-inflation-targeting.html?showComment=1407737014420#c1628467965234869331

5. ... regarding a potential approach for modeling probabilities: perhaps you can start with your existing simulation model, and add some elements which tend to retard or inhibit the runaway feedback loops: but which suddenly give way in certain circumstances when a tipping point is reached. Then one could potentially use this model in a Monte-Carlo analysis to estimate probabilities of hyperinflation.

2. You might add the "hunter's drawn bow" to your list of positive feedback loops. The hunter slowly pulls back the bow with arrow in place, and then releases. The arrow flies away with all the "energy" stored during the pull-back process.

It appears to me that QE is a storage of potential GDP. I do not know when the stored GDP will be released.

I think that government bonds also represent stored GDP. Bonds lock up GDP for a time period making bonds a measurable friction on the speed of release of additional GDP.

1. Roger, regarding the QE portion of your analogy, let me take a stab at arguing why I think it's benign. Imagine that the CB is committed to stopping any hint of hyperinflation once it gets started and that they're prepared to use all their legal powers to do so. Then here's my go-to example of one strategy they might use: Raise the reserve requirements (RR). So for example, say that reserves were running at an industry wide average of 170% of deposits. Conceivably the CB could then instantaneously raise the RR to 170%. This would cost the CB nothing and wouldn't require any massive changes in the CB's balance sheet nor any OMOs at all. Now of course some banks would be running lower than 170% and some higher, but the ones with a deficit of reserves could borrow from the ones with an excess. The CB will subsidize this borrowing because they are paying IOR now anyway. The lending banks get the same return either way, so it shouldn't be an issue for them. But I put forth that this would prevent all those reserves from "leaking out" into the economy like Vincent is worried about instantaneously. Now he complains that even though the banks with reserves in excess of 170% could conceivably lend to banks with reserves less than 170%, that such a move might bankrupt some banks anyway, and I'll concede that jumping overnight to 170% (or whatever the figure is on paper) might be a bad plan because the paper calculation might not reflect the stock of reserves (in excess of requirements) that banks desire to hold: thus I propose a slight modification of the plan: instead of jumping to the theoretical maximum (the RR level that would make all legally excess reserves vanish), why not creep up on it over a period of time: 10% a week, or per day. Perhaps the CB will find that raising the RR to 120% does the trick, or maybe is a bit far and they need to back off to 115%.

Now you might argue that RR is a terrible way to do what adjusting the IOR level can also do, but better. I won't argue that: that might be correct. I propose the RR approach because it serves as an emergency brake of sorts and it avoids the objections people like Vincent have to competing techniques: it does not require any OMOs like adjusting the CB's BS would, and it does not "cost" the CB anything different that what they're already paying, like playing with the IOR level would. His sole complaint is that it's a danger to the banks, but I think I've effectively addressed that by proposing an incremental approach.

My conclusion: QE is nothing to worry about: it's effects can be quickly and easily neutralized if need be.

2. Tom, My worry about QE comes from a different direction. Before explaining, I will try to repeat your argument in my words, hoping that I understand your thoughts.

The Central Bank can control inflation by controlling borrowing from banks.. Control over bank borrowing is done by control of the Reserve Rate, Interest on Reserves, Reserve Requirements and other methods. The increased money supply caused by QE can be offset by reduced bank lending if the Central Bank found that inflation had increased to a level that required remedy.

I think that Paul Volcker demonstrated in the 1980's that this sort of control can be accomplished.

My concern comes from a different direction. To me, QE is a method of financing the Federal Government with newly created money. It is important to me that the amount of QE is roughly the same as the amount of new Federal Debt over the course of the program. All of this new debt is in the hands of the public and represents money which can be spent as soon as the holders decide to spend. There is no need to invoke new loans to make this spending occur as the money has already been placed into public hands.

I think the reason that inflation has been very mild (relative to some expectations) is that the excess money has come to rest in the hands of non-spenders. Think here of wealthy people who save and buy government bonds, and not-so-wealthy-people who have a long term pension plans which buy government bonds. If it is correct that most of the new money flows into few hands, hands that are not spenders, then it should not be a surprise that inflation (and hyperinflation) has not occurred.

If in fact, the new money is correctly seen as already existing, when might the money return with a rush and cause hyperinflation? First, I think that the Central Bank can delay spending by increasing interest rates and transferring the funding load for government deficits to the private sector. If the Central Bank is unwilling to do that, then what rate of annual increase in money supply is needed. We have seen that a rate of increase of about 10% annually is not enough to cause inflation (in the American Economy) so it is logical that an even larger annual increase is required but how much larger?

The problem that developed economies seem to be running into is not inflation but an increasing divide between "haves" and "have-not's". I think this is a problem inherent in economies funded extensively by borrowed money (which increases the money supply).

3. Japan is spending twice what they get in taxes. I don't think it is possible to fix this by "transferring the funding load for government deficits to the private sector. " The private sector is so strained they have not been having kids for few decades. Now there are lots of retirees and not as many workers. No way you can double the taxes without destroying the economy.

Japan has no acceptable choice but to have the central bank fund the deficit. When it gets to that point you are clear for hyperinflation takeoff.

4. Vincent, you write:

"The private sector is so strained they have not been having kids for few decades."

I watched a 2010 BBC documentary called "No Sex Please, We're Japanese" which looks at the low birth rate in Japan. I think you can find it on youtube. Economics was probably mentioned as both a cause and consequence, but it was far from the primary one they focused on. Now perhaps the BBC got it wrong in their focus, but having seen the myriad plausible sounding potential causes they presented I'm far from prepared to accept at face value your statement here. (It's actually a pretty interesting documentary: I recommend it).

5. Kids are expensive and child labor is no longer legal in the West. So advanced countries, not just Japan, do to tend to have less children when times are hard.

3. Roger, I read your reply, and I think I get your main point. Before addressing your point, just to be sure we're on the same page, I drew out on simple balance sheets the consequences of two kinds of QE:

http://brown-blog-5.blogspot.com/2013/08/banking-example-41-quantitative-easing.html

In the first case the net seller of the Tsy bond is the banking sector. In the second case the net seller is the non-bank sector. Both result in the same dollar increase in reserves in the banking sector (assuming the non-banks don't withdraw any additional cash). Neither result in an increase in equity for anybody (I'm not including any profits or fees made on the sale of the Tsy bonds, so this is a 0th order approximation).

Vincent has expressed concern that the bank reserves which pile up as a direct consequence of QE can be converted into cash, and distributed to the public, thus we need to be very worried about it. I contend that this is not the case, because nobodies equity has changed, and in any case these reserves can be instantaneously neutralized with no "unwinding" of the Fed balance sheet required, and with no additional "cost" to the Fed in terms of increased IOR payments to the banks by simply raising the reserve requirements (RRs) (past 100% if necessary) to a level which eliminates all excess reserves from bank balance sheets. Excess here means "in excess of desired amounts" and I'll further assume that desired amounts are always greater than or equal to the legally required amount.

If we look at a simplified economy, with no foreign trade, and just four players: Government (Tsy), Central Bank (CB), Banks, and Non-Banks, then we have:

http://brown-blog-5.blogspot.com/2013/08/banking-example-11-all-possible-balance.html

Which shows that the equity held by the public in money and Tsy bond financial assets (not including private ownership of houses, cars, businesses, etc) is always equal to the government debt, regardless of the division of Tsy bonds between the CB and the banking sector. Look at the interactive spreadsheet, and try it out, or look at case 1 and set Ut=D=0 for simplicity, which is done for you in the "Public (Simplified)" balance sheet towards the bottom).

The public's total deposits = L + B + F - C
The public's total cash = C
The public's total Tsy debt holdings = T - B - F

L = total loans on bank balance sheets to non-banks (BSs)
B = bank held Tsy debt
F = CB held Tsy debt
C = cash held by public (public = non-bank private sector)
T = Government debt

Simplified further:

public's money = L + B + F
public held Tsy debt = T - B - F
public's total assets = L + T

public's total liabilities = L

public's equity = T

I'd claim that the equity held by the public is not affected (to 0th order) by QE, and how wealthy the public is, is partly a function of this equity (it should also include their ownership of houses, cars, businesses, etc).

4. Roger, I notice from the title of your blog that you're interested in the "mechanics" of the economy. Also you write above:

"The problem that developed economies seem to be running into is not inflation but an increasing divide between "haves" and "have-not's". I think this is a problem inherent in economies funded extensively by borrowed money (which increases the money supply)."

As I laid out above, to 0th order, I don't see what you're getting at with QE anyway. For one thing, all money is borrowed in modern systems: whether it's privately borrowed or borrowed by the government... there's always an equal and offsetting liability somewhere on somebody's balance sheet to any monetary asset.

I can see an argument about government debt, as that contributes directly to non-bank private sector equity, but QE only changes the composition of the assets which contribute to that equity.

Now you might have a point regarding how that equity is divided (regardless of its composition) between the haves and have-nots. But let me present an alternative view anyway: what if ~90% of inflation, etc, is explainable outside of any considerations of human motivations (i.e. to explain 90% of the big macro variables, like average price level (P), say we can ignore ideas like "expectations" "rationality" "incentives" etc), and instead explainable by entropic forces which are simply a consequence of the information carrying function of money. Here's an introduction to the mechanism:

http://informationtransfereconomics.blogspot.com/2014/08/rationality-and-entropic-forces.html

Some basic concepts:

http://informationtransfereconomics.blogspot.com/2014/03/how-money-transfers-information.html

How well the resulting mathematical models fit the data for a number of global economies:

http://informationtransfereconomics.blogspot.com/2014/06/output-and-price-level-behavior-across.html

Personally I'm very intrigued by the idea that perhaps 90% or so of some key macro trends have nothing at all to with particular human motivations, expectations, incentives, etc, but are rather mechanical emergent phenomena beyond any human scale considerations. I find the concept intuitively appealing and a refreshing alternative to what seem like endless "just-so" stories attempting to explain these same trends. Now whether it's true or not is another matter, but luckily the author is not interested in adding "epicycles" to his theories to patch them up (should they fail), and he has a number of testable predictions on record for a host of different economies around the globe. So if he's totally off base, we'll know soon enough. Most econ bloggers like to protect their precious hypothesis from the facts. In my view they should be going out of their way to expose them to falsification as Jason has.

One thing I'll say for Vincent, is that he's also put himself out their with at least one prediction for Japan: hyperinflation by the start of 2016!

1. Tom, I am not familiar with these concepts except I am familiar with information technology in a very beginning-student way. I have not taken time to understand the concepts presented in your links.

My mechanical approach to economics has focused on understanding that all the parts of economics must function smoothly together as in a functioning machine.

Toward this end, I have concluded that money is NOT debt. Yes, the quantity of money issued can be measured by adding up all the debt but that fact does not turn money into debt. Instead, money needs to be considered simply as PROPERTY, identical to all the other varieties of physical property.

Defense of this position is probably not an appropriate topic for the comment section of Vincent's blog.

If we think of money as PROPERTY, then it is easier to see my concern about QE (and perhaps Vincent's concern). The QE program changes debt into PROPERTY which becomes an increase in the amount of physical property present in the economic world.

The Japan example is one we should follow more closely. Japan has a "value added economy" in the sense that they have no resources so all their products are the result of adding labor to resources purchased abroad. All of Japan's products are produced with payment of YEN to labor. Then, most of Japan's products are sold overseas in NOT-YEN currencies. The imbalance between purchased resources and sale price has resulted in Japan holding large amounts of internal debt and foreign debt. In a macro-sense, someone has been working for BOTH money and debt with imbalances building for many years. I would like to better understand the mechanical relationships happening here.

2. Roger, I'm not sure I fully understand your last paragraph, but you write:

"I have concluded that money is NOT debt."

I agree so far as I'd say that money is not debt to all parties. To some it's an asset (not debt), and to other parties it represents an IOU that they've issued (looks like a debt to them, at least in an accounting sense).

I'll take another stab at your last paragraph.

5. Tom, Thanks for visiting my blog. Thanks also for the detailed (and complicated) comments. I will take a careful look and try to follow all your logic.

My latest post may apply to this issue of QE.

http://mechanicalmoney.blogspot.com/2014/05/the-banking-car-money-analogy.html

In one sense, clearly QE does not matter. In the second perspective, QE is an enabler that allows continuation (or correction) of an underlying repeating distortion. The substitution of cars for "money" allows this distortion to be more easily recognized.

More later after I do my homework.

6. I added "Landslide" and "Earth Dam Failure".

1. Also updated the common characteristics a bit.

7. Vincent, you might be interested to know that Jason makes a similar analogy wrt recessions. Here's the last sentence of his latest post:

I guess human behavior could trigger these avalanches or earthquakes.

1. He mentions stock crashes as also in this category, positive feedback loops. I agree. I think we are about due.

8. Tom, I made it through your post(s). Very well done and a lot of work to put them together. Thanks!

Your post come to the conclusion that treasury debt is private equity (excepting debt owned by Central Banks and Banks).

Next, when we consider QE, you come to the conclusion that QE does not change the equity of the private sector. I would agree with that position.

So what does QE change that makes it worth the effort of undertaking the program? It seems to me that QE changes the tension on the hunters's bow string (to continue the analogy I earlier advanced). Most economist do not consider bonds to be money so it follows that when government sells bonds, the money supply in the hands of the public is reduced. (Then, when government pays out the money it just borrowed, the supply of money in public hands again increases.) QE increases the rate of re-introduction of money back into public hands by NOT waiting until government spends, or (more directly), reverses the damping effect on spending from previous borrowing.

It is this increase in tension in the hunter's bow string that concerns me (and I think concerns Vincent).

Will the present world-wide expansion of the money supply(s) lead to hyperinflation? I think the controls on money value reside in taxation and in the willingness of government to reduce the amount of money in public hands by borrowing. Both mechanisms are consciously controlled by government and can prevent hyperinflation if vigorously utilized.

I conclude that hyperinflation is a choice, not a accidental outcome, but accidents can not be completely ruled out of the realm of possible outcomes.

1. I contend that out of all the hyperinflations there is not one where the central bank or government made a deliberate choice ahead of time to have hyperinflation. Hyperinflation is the market response to bad choices by government after they have gone too far. But the politicians neither understand hyperinflation nor worry about the long term enough to take the hard choices to avoid it. They need to cut government spending to save the long run, but for the next election it seems better to borrow/print/spend more. The politicians are just focused on the next election. So while it is theoretically true that "government can prevent hyperinflation if they vigorously utilised the mechanisms under their control", I think there is no chance that Japan will do so and not much that the USA will.

2. "I contend that out of all the hyperinflations there is not one where the central bank or government made a deliberate choice ahead of time to have hyperinflation."

You may be correct, but how about Hungary after WWII? That's perhaps the worst hyperinflation in recorded history, is it not? I don't know the history very well, but it occurred to me that this hyperinflation could have been engineered as part of a communist plot to destroy the upper and middle class resources and solidify the communist party's resources.

And how about Khmer Rouge in Democratic Kampuchea from 1975 to 1979: there's a government that effectively stamped out any possibility of hyperinflation by banning money altogether. Their grateful citizens sing the praises of the Khmer Rouge to this day for protecting them from the evils of hyperinflation. :D

3. Vincent, I will play devil's advocate writing a competing set of complaints about short sighted cynical politicians (which I don't necessarily believe, but which sound as plausible to me as yours do):

----------------------------------

But the politicians neither understand recessions caused by tight money nor worry about the long term consequences of them enough to take the appropriate choices to avoid them. They need to print money like mad to avoid recessions and depressions, but for the next election it seems better to play to their electorate who's been sold on simple Austrian platitudes like "Printing money is bad! Gold is good! Inflation is always bad and is the same as communism!" The politicians then become obsessed with cutting spending and tightening the money supply. This plays well to their populist base voters AND simultaneously to the wealthy elite which fund their campaigns. How? To the lower middle class populist voter base because the politician cynically claims government spending ends up in the hands of immigrants and people of the "wrong" race (they don't say "wrong race" but they imply it by saying "welfare queens" and "inner city"). These same politicians simultaneously don't say a word about medicare or social security spending, and imply that their older white lower middle class constituents deserve those payments because it's their money (even if they long ago withdrew all the funds they ever put into the system) and imply that's nothing at all like money going into the "inner city" or to "foreign aid" and "programs to help immigrants." These politicians cynically lead their base to believe that money going to "the inner city" and to "immigrants" and foreign aid makes up WAY more of the budget that is actually the case, thus the "man on the street" of their base will estimate that perhaps 20% of Federal government spending goes to undeserving "foreigners" and "inner city" people ... but at the same time this base will hold up signs in protest saying "Keep your dirty Federal government hands off of my social security and medicare!!!" forgetting completely where social security and medicare come from.

Now why does the message and policies of these cynical short sighted politicians appeal to the tiny but extremely wealthy donor class which funds their campaigns of disinformation? Because by being hyper-vigilant about any hint of inflation, they of course help creditors at the expense of debtors. This message is lost on the voting lower middle class base, who are debtors, and who are thus lead to believe that inflation is bad (not realizing that inflation includes wage inflation, and thus this inflation would help with repayment of their debts to the creditor class). Thus the cynical short sighted politician directly helps the financial status of this tiny population of already rich donors, while simultaneously getting his voting base to support him even though his policies are not in their direct interests (by cynically playing to the racial anxieties of this same voting base).

These short sighted cynical politicians don't consider for a moment the dangers of creating fascist extremism (think neo-Nazis in Hungary, Golden Dawn in Greece, or neo-fascists in Italy right now) by quashing any possibility of economic growth by keeping money too tight, but they do serve their donor class masters very nicely, since every incremental fall in the inflation rate below expectations adds \$billions to their real (not nominal) profits.

--------------------------------------------------------

4. Steve Roth makes this point about lowered inflation transferring real (not nominal) wealth from the debtor class to the creditor class in this comment to an excellent Simon Wren-Lewis post:

http://mainlymacro.blogspot.com/2014/08/the-symmetry-test.html?showComment=1408548772285#c6543981246972817937

9. http://robertnielsen21.wordpress.com/2013/04/06/why-do-we-print-money/

What are the bottlenecks currently in the economy that will lead to inflation? Or what bottlenecks will occur once people start spending money?

1. As a country gets close to hyperinflation it seems that necessities go up in price first. People reduce purchases of optional things, so they don't go up so much. But people need food.

What do you think of my positive feedback theory of hyperinflation?

2. Here is another way to look at it. If food is going up 5% per year and the bank pays 0.5% Interest then a family is better off buying extra cans of tuna than putting their money in the bank. If people start buying extra canned food then the supply of canned food becomes a bottleneck. If canned food is going up 15% per year then it is far better to buy that then put money in the bank. And so it goes...

3. 1. I only eat canned beans. At 4 different competitor stores the price has been going DOWN over the past year. As usual your analogies leave much to be desired. (Notice I too, picked ONE item. It totally DESTROYS your theory of inevitable hyperinflation. I suggest you short sell those Japanese bonds you hold. LOL.)
2. By the way its been 3 years and we are STILL WAITING Vincent (last name withheld).http://mythfighter.com/2010/09/13/debt-hawk-predicts-hyperinflation-in-2011/

4. Steve, from that link it says, " The debt hawks are to economics as the creationists are to biology.". Countries with too much debt run into trouble all the time. Look at Venezuela, Argentina, etc. It is just not true that a country that can print its own currency does not have to worry about debt. Can you argue against my positive feedback theory of hyperinflation with something stronger than the price of canned beans going down?

10. ["The Bible is the Word of God because God tells us it is... in the Bible."]

Terrible example and not what Christians or Jews believe. The Bible is believed to be the word of God due to prophetic fulfillment and miracles validating the authenticity of its writers in the past. Current generations accept the testimony of past generations to these events as well as Biblical reliability on other matters of history and human relations. You may disagree with these conclusions, but what you wrote is simply false.

1. I did not say any particular person or religion believed that. It is just an example of circular reasoning.

11. Needing to know exactly when a positive feedback loop will start is like putting an open can of gasoline next to a fire and demanding to know at exactly what time a spark will set it off. Run away and bet on an explosion.

12. Wow, Vincent, just reading the comments section here, evidently no good blog post goes unpunished! I'm no economist or anything close, but I think your point is pretty clear. The issue is not about what the precise moment will be when a positive feedback loop kicks off, but rather just the fact that given the right set of conditions, it's a virtual certainty.

-- Jim

13. eToro is the ultimate forex broker for new and full-time traders.

Looking for polite debate on ideas. Never attack a person. Be nice.