62 comments

[ 0.28 ms ] story [ 113 ms ] thread
It shines a light on the problem with inflationary currencies - one, costs of real goods tend to go down as technology advances, this is the "rising tide that lifts all boats", and is supposed to bring the lower economic classes to enjoy luxuries only available to the riches in eras long ago. Why are we trying to raise the prices on these goods?

Secondly, why are we urging the economy to have exponentially increasing production to maintain nominal value, when we have limited resources on this planet?

There are several parties in the game - the "normal plebians", and the upper eschlons. Guess who benefits the most from inflation? Guess who decides monetary policy that leads to inflation?
I found this remarkably unconvincing: none of the arguments really seem to address the actual concerns with deflationary currencies, but seem to be directed at some (vaguely implied) strawman. I mean, why is the fact that there is a liquid market for bitcoins relevant at all? I don't think that has any bearing on whether a deflationary currency is bad or not. One failure mode for a currency is that it's completely illiquid, but that's not the most likely problem. E.g. there was always a liquid market for gold but it's probably the main example of a deflationary currency in practice.
Agreed. The author is beset by muddled thinking. None of the arguments he brings forward are relevant to the question at hand.

The most important consequence of bitcoin is watching programmers and other tech oriented people play armchair economist. They certainly seem to be having lots of fun but they are playing a game no economist recognizes.

> None of the arguments he brings forward are relevant to the question at hand.

That's an unfair criticism, since the arguments I address are exactly those raised by everyone else who has written on the subject.

I have to agree with the posters above. What is the question at hand? I'd assume from the title you would be explaining why bitcoin is deflationary and why that is okay. (What is okay?)

What you actually seem to be addressing is whether the statement "Bitcoin cannot work as a medium of exchange" is true.

Your third point is good and one that I personally think is the most impactful. At the end of the day, everyone has their "price". You can't take BTC to the grave with you, so at some exchange rate, you are a seller. For some people this amount is $130, or $500, or $5000. But at every new BTCUSD high, it introduces a new group of sellers and eventually all hoarders will get shaken out.

I'm not sold on your second point "someone who sells a Bitcoin for X would have been equally happy to spend it on something they value at X." Using daily volume on Gox as an example, just because $2M changed hands yesterday, it doesn't necessarily imply that these folks would have happily exchange their coins for a good or service. A lot of this volume could represent traders getting in and out of a position, not someone ready to empty their account. However, your general point is that liquidity is a positive reinforcing signal, and I can agree with that. If liquidity is increasing because more merchants need to swap out, this is a direct indication that spenders are willing to transact.

> directed at some (vaguely implied) strawman

The Felix Salmon piece linked at the top, and every other article on the subject. If there's a good one, I haven't seen it.

> I don't think that has any bearing on whether a deflationary currency is bad or not.

Of course it does. Read the "supposedly relevant" link.

I would think a large majority of bitcoins would end up in the hands of people living somewhere with low interest rates - instead of parking their money in banks, they'll park it in bitcoins. The demand for investing - which applies across all currencies - could easily be greater than the demand to spend bitcoins.

Investment demand has caused deflationary problems in other areas too, specifically food prices when a lot of investment was made in certain agricultural futures.

Replying to my own comment, deflation isn't the worst thing that can happen to bitcoin IMO - unstable prices could be a bigger challenge. Since the market for bitcoins is not regulated, it could end up subject to all manor of tricks to create panics or booms. If the price fluctuates rapidly then timing the market becomes a burden of using the currency.
Expect lots and lots of arguments against bitcoin because of its deflationary nature. Nearly the entire universe of modern academic economics rests on the Keynesian and/or Monetarist foundation that increasing money supply (thus inflation) is a good thing.

Most people don't connect the fact that their money is worth less year after year with the fact that it is the government itself devaluing their money. Heck, most people don't even realize their money is losing value. Only savers, especially in our current zero interest rate policy environment, come close to understanding what is going on.

Nothing illustrates better what has happened to our money than the current market value of US coins made before 1965. For instance, a 1964 dime (nominally worth 10 cents) is actually worth $1.64 in silver content. That's somewhere around a 95% devaluation in just the last 50 years!

Are you saying there is something wrong with nominal inflation? 2% a year is scary when you talk big numbers like that, but the only time that money spans that large of time spans is retirement, which is invested at well over inflation even if you choose something like government guaranteed bonds.

To my knowledge the only way to lose money to inflation is to sit on cash (stupid for other reasons), have it in short term treasury bonds (which are bought by companies that have nothing to do with their flex cash), or have a bad savings account.

> short term treasury bonds

Hold on. If those are worse than even savings accounts.. why would said companies not put their money in a normal savings account?

I do not run the fiscal policy of a huge company, so I am speculating, but here goes:

Savings accounts are considered perfectly safe due to FDIC. If you have $10 million for six months you can't do anything with but don't want to invest, Treasury Bonds are zero risk and have some kind of return. In contrast there is a minor risk that a bank could go under.

Especially when you consider that you have several accounts with a bank, and you have say $20 million in your accounts total, why add to that?

Well and the FDIC doesn't really apply here since it only covers $250k. You're right on why the treasuries are appealing though, it's basically a bank account for large amounts.
Yes, it is harmful, particularly to savers. Of course, the Holy Keynesian view is that savings is bad and that we all need to be encouraged to spend.

Tell me right now where I can save and earn more than 2%. Maybe 30 year treasuries (with nominal values at 30 year highs, meaning that they have no where to go but down, further eroding savings), but certainly not in savings, a CD, or any other classic savings product. 2% compounding per year is a 22% loss in value over just 10 years. Chasing yield over that 2% per year further distorts savings and markets, and of course that 2% loss per year is good for what? It's just confiscation by the money printers to devalue future debt payments, making it seem all the easier to borrow and spend unsustainably.

But, of course, a steady 2% per year is the pipe dream of central planners and doesn't happen in real life. See, e.g., the 1970s.

Treasury Inflation Protected Securities (TIPS) adjust your principal accordingly with the CPI. You also get the coupon rate.
Small nominal inflation encourages investment, whereas deflation incentivizes people to just hold their money in the form of cash.
Of course, the other option is stability. I'm not saying that bitcoin's deflationary behavior is good, only that it challenges the accepted collective wisdom.
Saving is not harmful nor beneficial. It all depends on what you do with your savings. If you stuff them under the mattress, saving is harmful. If you invest, then it's beneficial.

An inflationary currency promotes investment. A deflationary one promotes mattress stuffing. You may draw the conclusion yourself.

> If you stuff them under the mattress, saving is harmful.

I don't see how. Isn't this equivalent to a distributed loan to everyone who holds dollars?

It is unstable. The currency is deflationary by design and the mattress effect effectively reduces monetary mass, creating more deflationary pressure. Up until the point price drops suddenly. It is the poster example of a bubble: artificial valuation, self feeding cycle, disruptable only by change of public perception.
Milton Friedman wasn't a Keynesian, and yet he advocated monetarism, which is essentially the same as inflation targeting, except instead of keeping inflation constant, you keep the rate of increase of the money supply constant.

In academic economics, even Keynes never advocated permanently distorting the economy towards consumption and away from spending. Of course people who like government spending for other reasons will incorrectly cite Keynes to justify their position.

Real interest rates determine the saving vs consumption decision, not inflation.

See here for the arguments that real economists use to justify a 2% inflation rate

http://econfan.blogspot.com/2011/07/why-target-inflation-at-...

You are always losing money to inflation. First, you always need to have cash because there is no cashless economy and you need cash to pay for everyday expenses, and that is a significant amount of money impacted by inflation.

Second, there is no serious rationale to have inflation (except to make it easy for governments to pay their debt and make everyone else poor), despite what Keynesians bring back on the table every single time without evidence (and citing the current "economist consensus" as a proof, even though they fail to realize economists were thought and raised the believe in Keynes' theory for dozens of years - hardly convincing). As another poster mentioned, even without inflation the money you have now is worth more than the money you have tomorrow, therefore there would still be time-value attached to a currency even if no inflation is there.

Inflation is punishing savers, in turn punishing investment and lowering the economic expectations.

(comment deleted)
> First, you always need to have cash because there is no cashless economy and you need cash to pay for everyday expenses, and that is a significant amount of money impacted by inflation.

How long do you hold onto your cash? I mean an actual bill. Remember inflation only impacts cash while you have a particular bit of cash.

If I hold onto $200 cash as an emergency fund, that $200 is impacted by inflation. If my wallet goes from $200 to $10 every week, then only $10 is impacted by inflation.

> Second, there is no serious rationale to have inflation

You are going to have to try much harder to debunk inflation than you are doing here. Most of the modern economic theory calls for a nominal amount of inflation (1-2%) as good for the economy. I was correlating time-value to inflation, as they are similar impactors.

> Inflation is punishing savers, in turn punishing investment and lowering the economic expectations.

Except every system I have ever seen that isn't your mattress is designed to give you inflation coverage + an ROI to compensate you for potential loss. How is that hurting savers?

I apologize ahead of time for going off topic but I have to argue against the bashing of Keynesianism that is getting so popular lately.

Keynesian economics works, and it is responsible for the longest and largest sustaining economic growth ever seen on this planet. It works, and moreover it is shown to be the only thing that works over a long period of time.

The fact that the competing theory is simpler does not mean that the competing theory works. Simplicity does not mean accuracy. Also expansion of the money supply does not necessarily mean inflation. Remember that the amount of goods and services produced in the world is also expanding.

There's a difference between "works" and "has worked". And it hasn't existed for "a long period of time", just long when compared against the lifetime of modern economics.

Which is not to say it's wrong. Merely that the world hasn't settled on Keynesianism as the One True Path™, and there exists no logical proof that it is successful, much less the only successful scheme, because you can't construct irrefutable proofs when you have to include human behavior.

The bashing may be popular now, but Keynesianism has been popular for quite a while, during a period of great wealth increase (which you cannot attribute solely to it, for the same reasons you cannot prove it works), and I suspect it has blinded people to other possibilities. At least this gets people talking about it, and it has generally been civil as far as I have seen.

"Keynesian economics works"

well, yes, if you call a increasing gap between the wealthy and the poor "work"ing.

"Nearly the entire universe of modern academic economics rests on the Keynesian and/or Monetarist foundation that increasing money supply (thus inflation) is a good thing."

Imagine my shock that user jstalin is a monatery crank.

If your currency is worth X today and more than X tomorrow you will see less economic activity than otherwise. That simple fact that is true whether or not you agree with the rest of Keynesian or Monetarist economics.

> If your currency is worth X today and more than X tomorrow you will see less economic activity than otherwise.

This is something that's confused me for a while. If the value of your currency is going up, doesn't that mean your purchasing power is going up too? Which means you feel richer because the relative prices of goods are going down.

So your argument seems to be saying: falling prices depress economic activity. Which is true in some cases, but isn't true as a general rule. I think it's more of a threshold effect.

No one ever says "I'm never going to buy an iPhone because the price keeps going down." They say "Once the price drops to $X, I'll buy one."

"Once the price drops to $X, I'll buy one."

This is a naive view of how supply works. If fewer people are buying today this lowers the demand curve. This decreases the amount of the good that will be produced in the first place. You don't get a chance to buy those lost phones later because they never got made.

Yes, but the standard supply and demand model makes assumptions about the nature of money markets that I don't think apply in this case, which is why I'm confused.

Bitcoin has some unique features (e.g. non-repudiable "teleporting") that aren't normally considered -- Econ 101 assumptions prove too little. However, this doesn't take a deflationary death spiral off the table. It could easily be Bitcoin's downfall, I just think other things are more likely.

...And this whole time I thought supply-side conservative economics was all about propping up and subsidizing the holders of capital.
> That's somewhere around a 95% devaluation in just the last 50 years!

And a good thing! Everything was shit in 1964. Cars, computers, retail logistics, cancer treatments, surgical techniques, metallurgy, airplanes, the telephone network, movies, animation, TV, air conditioning, the relative strength of the USSR: all shit.

It all needed to be burned to the ground to get the wonders we live with today. If Grandma Retard's pass-book account had to go with it, so be it. If the deflationists had their way, we'd still be using stable, value-preserved stone knives and bear skins.

Iosif, I appreciate your strong opinions about inflation. I'm sure Lenin would be proud [1]. However, you're too into the politics of it. We in HN should try to do some rational armchair economics, try to see the inflation/deflation debate as about tradeoffs, not principles of morality. "Government steals through insidious inflation!". "Rent seekers fatten when deflation happens!". Yeah, very useful.

First, the Keynesian/Monetarist Block does emphatically not claim that inflation is good, since last time I checked ethics was not a sub-field of economics. Inflation is a tool.

Second, deflation really is a big deal. Unless you're ready to argue that debt-deflation does not exist. Please spare us from such "deep" arguments such as found on the Bitcoin Wiki [2].

Third, the main problem of Bitcoin is not that it is deflationary, because it's not. Well, it is now, but not necessarily forever. The problem is that it imposes, through its fixed supply, a senseless corset on economic growth. It's neither central planning, which would have at least the possibility of being a somewhat enlightened planning, nor a rigorous market mechanism, that responds to shocks on the economy. It's just fixed because Satoshi said so. This lack of rigor in economic thought is what's really problematic in Bitcoin.

[1] https://en.wikipedia.org/wiki/Inflationism#Political_debate

[2] https://en.bitcoin.it/wiki/Deflationary_spiral . A sample from the well of deepness: "The key difference is that people don't foresee a fixed cost (unit amount) that they must pay with Bitcoin". Yes, they do, or will when lending in Bitcoins begins. It's called the principal. You should laugh now.

The article hits on an entire class of silly arguments I've seen against bitcoin from the start: people claim that bitcoins will be worthless because they will be too valuable to trade.

The language used for "too valuable" may vary (scarcity! illiquid! deflation!) But, as the author notes, that's not at all what matters -- all that's necessary is that they're tradeable and people have different values for a bitcoin at a particular time.

Hm, HN traffic has grown. The server will be back up in a couple minutes. Edit: All good, I think.

This post has been in the draft folder for a while, but I decided to just run it. A better title would be "Common Flaws in the 'Bitcoin is Deflationary' Argument". There might actually be a convincing "Bitcoin is Deflationary" argument to be made, it's just that no one has made it yet.

A lot of these arguments against the future of Bitcoin rest on the ill-founded assumption that the Bitcoin protocol is static and unchanging. This isn't necessarily true; there's already reactionary plans should SHA-2 or secp256k1 become comprised.

It seems fair to reason that a community consensus would alter the protocol should deflation ever really become a problem.

Replacing the underlying crypto is 1000x simpler than replacing the coin production mechanism, for both political and practical reasons.
I think miners who have invested in farms of (very application specific) ASICs would disagree.

Then again, in theory it's the "economic majority", not that miners, that matter: https://en.bitcoin.it/wiki/Economic_majority

You can replace RIPE or ECDSA or even SHA (in the address-generation mechanism) without impacting the miners at all.
Since the entire premise of Bitcoin is a fanatical devotion to the idea that no evil government will ever be able to produce more of the currency, it seems highly, highly unlikely that the deflationary "bug" in the Bitcoin protocol is one that will ever be changed by Bitcoin participants. Any other bug might be fixed. But not that one.
Falling prices and strong growth tend to be mutually exclusive in the real economy. Suppose you’re an investor and an entrepreneur comes to you with a proposal to build a factory, and the business plan shows next year the factory will cost 5% less to build, and the products will sell 5% cheaper. Why build a factory today if you can get 5% a year increase in purchasing power just by hoarding currency? You need much bigger margins than are currently typical outside tech and zero marginal cost products. Incur debt, and even at the zero interest rate bound, you need 5% more real income next year to pay it back. In a modern economy, especially where there's a lot of debt finance, price deflation is death.

Anyway, since Bitcoin isn't a currency in the sense of a transaction medium, a stable store of value, and a unit of account, but more akin to a digital pet rock or Beanie Baby, deflation doesn't really matter very much.

Why build a factory today if you can buy your country's bonds and enjoy a risk-free x% a year?

Why buy a smart phone today if you can wait 1 year and get a much better one for the same price?

(comment deleted)
The short answer is, you don't. When real interest rates are at 5%, that is not normal and those are not usually good times.

Bonds in the US have earned a real 2.3% since 1926. From 1937 to 1981 bonds lost a real -0.72% per year.

You don't buy bonds to earn the rate of growth of real GDP or better. If you want to do that, you have to take risk.

The last 30 years of declining inflation have been the anomaly. People built businesses instead of buying bonds because bonds looked pretty unattractive, but when inflation kept surprising to the downside, bonds did better than expected.

re: "since Bitcoin isn't a currency in the sense of a transaction medium"

$2M per month of transactions on SilkRoad beg to differ.

A minor airline's frequent flier mile transactions are 10 times that. An average Target store does more than that. What convertible currency doesn't have a giant multiple of that? And Silk Road isn't even legit, enforceable transactions.
>Why build a factory today if you can get 5% a year increase in purchasing power just by hoarding currency?

You don't; but the reality is that if the need for something becomes great enough, the potential profits from investing in production of that thing surpass the potential profits of saving. And if no one is investing in production, prices increase, combatting deflation. As usual, the market will solve this problem given a little time.

>In a modern economy, especially where there's a lot of debt finance, price deflation is death.

Then I guess businesses can continue to use deflationary currency while everyone who wants to save can use Bitcoin or some other non-deflationary holding.

>Anyway, since Bitcoin isn't a currency in the sense of a transaction medium, a stable store of value, and a unit of account, but more akin to a digital pet rock or Beanie Baby, deflation doesn't really matter very much.

A deflationary currency is certainly not a stable store of value either. Bitcoin is, of course, a transaction medium and a unit of account. I guess if you wanted, you could conduct transaction in beanie babies, but it would be much less convenient.

You might want to build that factory now to be the first on the market, and reap all the benefits from that. The guy who waits until next year will have to face more competition.
The comparison to housing seems a little off to me. For example with the property boom, people didn't want to sell their houses because they were going up in value. Therefor they go up in value some more. Therefor people can't afford to buy them, so we fix that with sub prime mortgages..
Right, whereas, if we didn't create subprime mortgages, if people couldn't afford to buy them, eventually the price would have had to come down, or else you have idle stock. Instead, we inflated the price bubble by enabling risky individuals to make purchases they shouldn't be able to, and so when the prices actually came crashing down, not only did they take down the housing sector, but the defaulted loans exerted magnified, collateral damage across the entire economy.
I think it is not ok, and I think that the deflationary nature of Bitcoin is a wonderful practical example of the problem with deflation.

To the best of my knowledge bitcoin does not seem to be used as currency that much. It's explosion in popularity is not matched in explosion in currency usage. And why would it? If you, for example, buy a toaster online with bitcoin and it keeps appreciating, this toaster may end up costing you a $1000 in a year.

If bitcoin were to fullfil its promise, a large portion of online transactions should take place with bitcoins. AFAIK, thus far this is only happening for illegal drug transactions, for which there is no other alternative.

Thus far bitcoin is something people buy and hold hoping it appreciates. If they are tired of holding it, they sell it back. It is not a currency, it is an object of investment/speculation.

What stops you from doing business in fractional bitcoins? I'm pretty sure shift is a basic instruction in every architecture.
Dealing in fractional bitcoins doesn't solve deflationary problems in the same way that Zimbabwe printing a $100 trillion dollar banknote didn't solve their inflationary problems.
None of these arguments is at all persuasive.

"First of all, the present value of a perpetually appreciating asset is not infinity"

Straw man. No one is claiming the present value of anything has to be infinity. And I don't see anyone saying real estate is a good medium of exchange.

"Second, there is a liquid market for Bitcoins."

Another straw man. No one is saying there will not be any trade in deflationary currencies. The amount of Bitcoins that are not being circulated speaks to exactly what people don't like about deflationary currencies though.

"Third, the marginal utility of money is diminishing."

A true fact that has no impact on the criticism of Bitcoin because it's deflationary. Red herring.

Deflation may prevent Bitcoin from replacing all government fiat currencies, but I don't think it has much impact on Bitcoin becoming a popular form of payments / transfer / store of wealth.

Some Bitcoin proponents would be disappointed by that prospect, but I think it would still be a big win.

It likely means several orders of magnitude difference in price, but in either case the price would go up significantly.