Hacker News Reader

Comment by ucaetano | Hacker News Reader

> A thought experiment: how would an economy run solely on a Bitcoin-like currency behave? And how would it respond to business cycles?

Terribly.

Bitcoin is by design deflationary, which creates an incentive to maintain currency under the mattress. This is accentuated by the fact that bitcoin can't be used in fractional-reserve-banking.

This means that there is little accumulated capital available to borrow and invest.

It also means that any investment would need higher returns than the expected appreciation of a bitcoin, which might set a very high bar for investments, making raising capital nearly impossible.

Today, in most open economies, nobody controls the supply of money, it is increased and decreased dynamically by the market according to supply and demand and to maintain a certain level of risk on the banking system (with some adjustments from the central bank).

View on HN

> This is accentuated by the fact that bitcoin can't be used in fractional-reserve-banking.

I don't know what you mean to say here, but you can absolutely use bitcoin for fractional-reserve banking, just as gold was used through history. Pretty much since the dawn of banking, a bank collects reserves from its depositors, and then loans outs a designated fraction of those reserves. It's not like that process was somehow invented with the Federal Reserve.

What you cannot do is have a lender of last resort with the ability to print money out of thin air, who can then flood the economy with cheap money; But that has nothing to do with fractional-reserve banking.

No, you can't, you need to create a new "currency" to do so. You can't do fractional reserve banking with gold. What banks did was create "gold bonds" that were just a promise to pay.

And then you're just creating a bitcoin-back currency, which at some point will inevitably see the same fate as all currencies did, and being de-linked from bitcoin just as other currencies were de-linked from gold.

People can and did do fractional reserve banking with gold. Notes for deposits was a much later innovation. Nothing stops you from lending out physical specie from your vault as long as you can give depositors their money with high probability when they demand it.

No, they didn't.

They did fractional reserve with gold-denominated currencies.

Sure, you could create a bitcoin-denominated currency (a terrible idea, but let's entertain it) and perform fractional-reserve banking on it. But at that point, bitcoins become irrelevant.

Well, we could go back and forth with "no they didn't" and "yes they did". Let's throw in some citations. You're in a hard place, because proving a negative is tricky. So I think it's incumbent on me to provide some examples.

Let me start by throwing out a little bit of terminology just to make it clear if we're on the same page. Loans have existed since money existed. The only difference between a private loan and fractional reserve banking is that the lender is a) lending someone else's money and b) allowing depositors to withdraw money despite the fact that the sum of all depositor's money is not physically possessed by the bank.

Wikipedia's page [1] is a reasonable place to start. The Babylonians were issuing loans in gold. The Greek and Egyptians were issuing loans in gold. The Romans were issuing loans in gold. None of these (except the Egyptians, with their seed grain short-term currencies) were issuing any sort of derivative currency or notes.

Throughout medieval times, letters of credit were frequently used as a form of correspondence banking [2] for travelers. The lack of high-speed communication meant that the settlement and clearing of loans could take months or years, but the depositors (correspondent banks in foreign countries) were unlikely to make any immediate claims on their deposits except through such letters. The letters could be redeemed, usually, directly for gold. You can argue whether this constitutes fractional reserve banking; it's in some ways closer to what we would now call "corporate paper"; very short-term credit extended to very high trust individuals or institutions.

[1] https://en.wikipedia.org/wiki/History_of_banking#Earliest_fo...

[2] https://en.wikipedia.org/wiki/Merchant_bank

Loaning gold isn't performing FRB on gold.

> Loaning gold isn't performing FRB on gold.

It is, in fact, exactly what fractional-reserve banking means: Loaning a fraction of your reserves.

BTW: The exchanges are creating money by allowing margin: There is already a difference between M0 and M1 in the bitcoin economy.

I think the point they're making is that you cannot literally do fractional reserve banking with gold or bitcoins, because there is a physically or technically limited amount of them.

In your example, they can fractional reserve with letters of credit and trust. At no point can a bank actually loan out more physical gold then it had in its possession.

Exactly.

FRB is essentially a form of "double spending", but under certain conditions to maintain stability.

Bitcoin doesn't allow double spending. Neither does gold.

> FRB is essentially a form of "double spending"

This is nonsense. Loaning money that you own is not "double spending".

I'm not sure what "literally [doing] fractional reserve banking" means. Can you give me an alternative definition to the one I gave above?

Fractional reserve banking means you accept deposits, and you make loans from those deposits, and the difference between the amount you've lent out and the amount you have in deposit is called the "reserve". If you try to maintain your reserve as a "fraction" of your deposits, you're doing "fractional" "reserve" banking. What's the "literal" aspect?

You can't spend dollars that you have in the bank, not in the sense of handing someone cash. You can do money transfers to other accounts or other banks, but when you get down to how a money transfer happens (i.e. deposits with a central bank or correspondence accounts between institutions) there's no reason that you can't do that with any substance or commodity.

There's no double spend involved, because you don't have specific Bitcoins on deposit. The Bitcoins you've deposited are long gone; munged into some central wallet and lent out to depositors, with only a fraction held by the bank. When you ask to withdraw, they take that fraction and give you the amount of your withdrawal out of it, and reduce the paper balance of your account.

I'm living in a community where gold is the currency of choice, and I convince a bunch of people in my community to deposit gold pieces with me. I have a huge pile of gold.

Whenever somebody asks me to give them back their gold, I reach into the pile and hand them pieces equal to the amount they deposited, but not necessarily the same pieces.

But this happens infrequently enough that most of the time, the pile just sits there. So I decide to only keep half the pile sitting there -- more than enough to cover the rate at which my customers make withdrawals -- and loan out the rest to borrowers I deem credible. I charge interest to make up for the losses I take on bad borrowers and actually make enough profit on this interest to keep some and pay some out to account-holders.

I would say I am 1. conducting fractional reserve banking, with a 50% reserve ratio, and 2. not creating a new currency (everybody still transacts in gold pieces).

Are you saying that this is NOT fractional reserve banking, or that it IS creating a new currency?

The loan is the currency. People accept the loan as a medium of exchange. Everybody transacts in loans (paper with promises of value), which then becomes identical to money. The loan is only loosely coupled to the gold backing, because of everybody cashed in the loan (a bank run), you could not pay it out because the gold does not exist for everybody. That very fact demonstrates that the loan is not identical by reference to gold. So yes you have created new currency.

> No, you can't, you need to create a new "currency" to do so.

You are confusing "fraction-reserve banking" with something else, probably "quantitative easing".

There is no requirement to create currency with fractional-reserve banking: A bank simply loans out a fraction of its reserves, which has the effect of creating money, but only in the sense of increasing M1. M0 remains constant.

In different perspectives, you are both correct. Particularly, there is a sense in which bank account balances are a separate currency redeemable in the currency in which they are denominated, so that, yes, you have to create a separate currency to do fractional reserve banking, but it's the kind of currency that various entities are going to be creating as soon as you want to support any kind of useful ecosystem around a cryptocurrency, including exchanges.

Exactly, and the difference of trying to do FRB with bitcoin or gold is that the new currency being created isn't the original currency.

In the case of USD, for example, a USD created by a bank is indistinguishable from one you deposited on the bank.

You could say that being able to double spend in certain conditions is a feature of modern currencies, and largely necessary for a stable economy, which is impossible under bitcoin.

> "Particularly, there is a sense in which bank account balances are a separate currency"

You can define "credits" and "debits" in a currency to be a separate currency, but since that has been in the lexicon for millenia, you'd have a very long road ahead for acceptance of terminology like that.

There is a reason why we define M1 as different from M0. Trying to say that bitcoin is some newfangled currency that demands M0 and M1 be equal? Good luck with that.

Doesn't it depend on whether you deposit your bitcoin into a bank? If everyone's always only keeps their money in personal wallets then you'd be correct, even if banks were just "a wallet but we maintain it" then you'd still be correct.

But if banks continued to work like they do now, where you deposit your money into the bank, and then they show a number when you login, but it's just what they "owe" you. In that case they could continue to do fractional reserve banking just as they do now.

So basically this debate is not whether fractional reserve banking is possible with bitcoin, but whether banking with bitcoin would continue to resemble banking now.

> Doesn't it depend on whether you deposit your bitcoin into a bank?

Why would you? You can get better returns by simply holding it (since it is deflationary) than by lending it out or investing in productive assets.

And even if you deposited, you can't "double spend" the bitcoin, so you can't do FRB with it.

> So basically this debate is not whether fractional reserve banking is possible with bitcoin, but whether banking with bitcoin would continue to resemble banking now.

It is both. You need FRB to dynamically adjust the supply of money and keep a balance with demand.

And the fact that nobody will invest, which kills economic growth.

No new startups, no new factories, no new bridges, no new houses, etc...

> Why would you? You can get better returns by simply holding it (since it is deflationary) than by lending it out or investing in productive assets.

There's obviously a balance in this scenario, as slower investment pace, makes for slower expectation of BTC value increase.

Only "investments" of very low expected return are being prevented by lack of monetary inflation.

If BTC-based monetary system produces relatively slow .5% global growth, that means only investments with expected return lower than .5% are not going to happen. Which is a feature, not a bug. It makes growth cycle much more stable and self-balancing. Too high growth slows down investing cycle, low growth speeds it up. No need for central bankers.

And it's not that much different than current monetary system anyway, where people just throw they currency into speculative, unproductive assets to escape inflation. Nowadays things like housing serves the role of Bitcoin.

Sure there is a balance.

Just like in emerging markets with high interest rates: if your government pays 10% per year (real terms) interest on its bonds, would you invest in a business that would only return 9%?

No, nobody would ever invest in it. But there's a reason the government is paying 10%: to reward the risk of lending money to it.

In the case of bitcoin, the only reason there is this deflationary "tax" is because someone arbitrarily decided that the supply of bitcoins should remain constant.

So you're voluntarily fucking up your economy.

Sure, "we could relax that and make the supply increase over time", but that doesn't address fluctuations on the demand for money, resulting in harder oscillations.

And then "but then we could have the supply adjust dynamically based on some guidance from economists plus the balance of supply and demand for money".

Well, congrats, you just recreated the modern monetary system.

Just the deflationary nature of a currency isn't enough to destabilize cycles. Cycles run on credit-worthiness, or how likely is someone to return x% return in the future on an investment made today.

Compared to the Fed, which makes arbitrary (non-cyclical) adjustments to the "cost of money", Bitcoin's more predictable deflationary attributes might lead to more predictability in financial cycles.

This is of course, assuming the hundreds of other problems with Bitcoin as a currency, or credit-instrument have been solved.

> Just the deflationary nature of a currency isn't enough to destabilize cycles.

No need for cycles: a deflationary currency means that you're better off holding it than investing it.

This is disastrous for the economy.

> Compared to the Fed, which makes arbitrary (non-cyclical) adjustments to the "cost of money", Bitcoin's more predictable deflationary attributes might lead to more predictability in financial cycles.

Quite the opposite. The money supply wouldn't adapt dynamically to the demand for money, leading to harsher crises.

>This is disastrous for the economy.

This is FUD. The US experienced deflation through much of the 19th century, yet saw higher growth rates then than in the 20th century. For instance, from https://www.investopedia.com/ask/answers/040715/were-there-a... (don't have time to find a more formal source): "The period between 1873 and 1879 saw prices drop by nearly 3% per year, yet real national product growth was almost 7% during the same time." And overall "the price level (the average of current prices across the entire spectrum of goods and services produced in the economy) was actually 50% higher in 1800 than it was in 1900." So clearly it's possible to have strong economic growth during deflationary times.

You're confusing deflation with deflationary currency.

> a deflationary currency means that you're better off holding it than investing it.

It is ridiculous to assume that any artificially designed deflationary digital asset will produce long term returns greater than an index like SPY, which is intrinsically tied to human productivity in the economy.

An average, semi-rational investor will always have the incentive to lend to a credit worthy person/business/asset, regardless of how much artificial deflation you can induce in your asset.

It isn't returns. It is deflation.

You're literally creating a broken system by design. Don't complain when it doesn't work.

So in some segment of the 1800s economic growth was larger than deflationary pressures which might have been linked to increasing economic efficiency... Seems pretty narrow.

Unfortunately I don't see how bitcoin would be likely to maintain a deflationary rate anywhere near balanced with economic growth if it were used as a primary currency considering its hard capped at 21 million BTC, and an unknown number of coins are pretty much permanently lost. I think it'd have an effect of making primary interest rates have to exceed the rate of Bitcoin's deflation which sounds like a death spiral to me.

> which creates an incentive to maintain currency under the mattress.

Continuing the thought experiment, consider what goods and services people would continue to spend money on, in spite of the expected increase in the value of money.

> Today, in most open economies, nobody controls the supply of money, ...

With all due respect, this is demonstrably false. The Federal Reserve and its member banks control the supply of money.

"Continuing the thought experiment, consider what goods and services people would continue to spend money on, in spite of the expected increase in the value of money."

I love this comment. Everyone just blindly says "People will put money in their mattress. Economy will go down" and then it's case closed.. but then what happens next? They just won't buy food? water? won't go to the mall? They just stay at home and hunker down? They won't even be watching entertainment since they won't be paying for a Netflix subscription too, apparently.

It's not like it never happened before. Just read up on the great depression and you'll get an idea.

Deflation and deflationary episodes aren't the same as a deflationary currency.

read up on the gold standard then.

Sight...seriously?

Gold wasn't a deflationary currency, not only that, the supply of gold increases over time, and some times suddenly such as when the Spanish empire started extracting gold and silver from the new world resulting in the Spanish Price Revolution.

Seriously, please go take some economics classes...

The proposition is not that people won't spend money. The proposition is that they won't invest the money that they're not spending. Keeping money under the mattress is not competing against groceries and Netflix, it's competing against government bonds and/or interest-bearing bank accounts.

If people sudddenly hoard money, they're essentially reducing the supply of money, which means the value of money will go up (as there's less of it), effectively acting as a transfer of purchasing power to the people who are actually spending it at that time. Imagine for instance I do something that generates a lot of value (money), then burn that money (or hide it under my mattress 'til I die of old age). This essentially means I gave "value" to society for free, as I created "value" but did not consume any "value" in return, leaving that "value" I could have consumed with my money to be consumed by others.

Nope, the problem isn't people hoarding money, the problem is doing so outside of a banking system, where such money is put to work in investments.

If the economy is growing in the long term and the supply of money is constant, the value of money (vs. goods) will always increase, and everyone has an incentive to hold currency instead of investing in the economy.

The result is that you can't raise capital for a startup or for an infrastructure project.

This is not responding to GP's argument: the point is that "hoarders" implicitly provide capital for others by reducing the circulating supply.

Driving deflation isn't providing value.

The claim isn't even wrong, it doesn't make sense.

To earn money, somebody has to have created value (people are paid for creating value). Spending money is consuming value. Imagine there is $100 worth of total value in existence. Somebody creates $5 of value, they're paid $5, and total value existing is now $105. They could immediately consume or invest this $5 of value, leaving society with $100 of value, and if they invested some productively then they'd have increased the rate at which society's able to generate value in future. Or, they could do nothing with that $5, leaving society with $105 of value. How does society access this extra $5 of value? The person who choose to hoard the $5 effectively removes that money from circulation, which decreases the supply of money and hence increases its value. This means the purchasing power of other currency holders is increased. Those who then decide to spend money can consume or invest the $5 of value that was left unused.

I'm not talking about spending, I'm talking about savings, capital accumulation and re-investment. Your entire point (and your parent's) isn't even wrong.

And no, the federal reserve doesn't control the supply of money by any means.

It influences it, sure, but it doesn't control.

The deflationary argument is about a difference in degree, not in kind. A deflationary monetary regime does not stop all commerce, it just creates incentives to defer or forgo discretionary spending and investment. That, by traditional economic measures, is Bad.

> Continuing the thought experiment, consider what goods and services people would continue to spend money on, in spite of the expected increase in the value of money.

Sorry, but you didn't understand how capital accumulation works.

People work and so on, get paid and so on. Then they spend part of it on their living expenses and save part of it (that's the domestic savings rate).

The money saved doesn't just sit under a mattress, it becomes the stock of capital in a country (plus and minus capital imports and exports and so on), which is then invested in projects (startups, infrastructure, etc.) that need capital.

Banks are the intermediaries in this market, taking capital from savers and providing investments and loans to projects that need capital. That's why so many emerging countries are desperate to get people into the banking system and to move their savings to the banks (the US did it by Executive Order 6102 for example).

Without this stock of capital, it is impossible to start a business, or to fund large infrastructure projects. It is a key and necessary component of a healthy economy.

And bitcoin kills it completely.

>The money saved doesn't just sit under a mattress, it becomes the stock of capital in a country (plus and minus capital imports and exports and so on), which is then invested in projects (startups, infrastructure, etc.) that need capital.

Money doesn't become the stock of capital; money isn't capital. Money is a proxy for value. Capital is created by consuming less value than is produced, so the non-consumed value can be invested, which savings can achieve even if the money itself is not invested. This is because saving/hoarding essentially reduces the money supply, thus increasing the value of money - transferring purchasing power from the saver to people currently spending money. If I create x dollars of value, then burn the money I recieved in compensation, I've essentially gifted the value I produced to other currency holders, which they can consume or invest.

The US experienced deflation through much of the 19th century, yet saw higher growth rates then than in the 20th century. For instance, from https://www.investopedia.com/ask/answers/040715/were-there-a... (don't have time to find a more formal source): "The period between 1873 and 1879 saw prices drop by nearly 3% per year, yet real national product growth was almost 7% during the same time." And overall "the price level (the average of current prices across the entire spectrum of goods and services produced in the economy) was actually 50% higher in 1800 than it was in 1900." So clearly it's possible to have strong economic growth during deflationary times.

Money is a proxy for capital when used for productive ends. So, yes, savings generate stock of capital.

It's true that Bitcoin can't be used legally in fractional-reserve banking, but given the behaviour of exchanges to continue trading despite a proportion of their holdings having been hacked or embezzled, there was an element of de-facto fractional reserve going on.

That's not fractional reserve, that's just fraud.

Exchanges are not banks.

I don't know what it means to say that it can't be used legally in fractional-reserve banking.

Fractional-reserve banking is just a pooled-risk lending instrument. If you can lend something, then you can fractionally reserve bank with it.

To put it another way, when you have dollars deposited in a bank, you exchange your dollars for a contract saying that the bank will remit so many dollars to you on demand. You value that contract at pretty much the face value of the dollar amount of the deposit because the risk of default on that contract is considered extremely low.

Fractional reserve banking in Bitcoin would work the same way -- the bank would give out loans while promising to make their depositors (creditors) whole should they require their deposits. They cannot satisfy that contract if more than a certain number of depositors come knocking, but that's true for regular banks too.

With a liquid credit market, even runs on the bank, as long as they are slow, are fairly tractable; you can just sell your debts for their face value (discounted, most likely, for the fact that the counterparty to the sale may not use the same risk metrics) and thereby make your depositors whole without having to call in any outstanding loans.

> fractional-reserve-banking

You are confusing this with something else. If there was no fractional-reserve-banking under the gold standard, then why did bank panics happen? If the bank had $1 million loaned out, and also had $1 million of gold in its vaults, then why would anyone panic? Yet bank panics were common. And that's because the banks all engaged in fractional-reserve-banking -- if they had $1 million in gold in their vaults, they would make loans worth $5 million.

Because you literally can't loan $5M in physical gold if you only have $1M in physical gold in your vaults.

You can't just make gold out of thin air. Same with a bitcoin.

You can, however, issue gold-denominated notes in $5M even if you only have $1M in the bank. But you're not lending gold, you're lending paper currency.

I can't tell if you are talking about real life, or if you are talking about a hypothetical scenario. You are aware that bank panics were common when the gold standard was in force? Banks routinely made $5 million in loans when they had $1 million of gold in their vaults?

Loaning out a gold-backed currency isn't loaning out gold.

You seem to be confusing "1 pound of gold" with "a piece of paper saying that bank X will trade this for 1 pound of gold".

The two are not the same.

> Bitcoin is by design deflationary

I keep hearing this. Saying that Bitcoin is deflationary is exactly equivalent to saying that the value of Bitcoin will always go up. In a Bitcoin economy, the value of Bitcoin would vary over time in response to various factors; technological innovation causing actually cheaper goods, and credit markets that allow people to trade on the future vs. the present value of money.

If the supply of money is constant but the demand for it (due to economic growth) is growing, the value of money will always increase, and the prices of things will always decrease.

This is trivial.

And the consequence is that everyone is better off holding currency than investing it. Which is disastrous for the economy.

This is not trivial. Even a pure monetarist admits monetary velocity into this equation. If we lived in a universe where the price of Bitcoin would always increase in a reliable fashion, then it is straightforward to integrate that information into the current price of Bitcoin based on prevailing interest rates.

There is simply no such thing as an asset that will always increase in value -- _that_ is trivial.

Oh, since you want to bring velocity into the game: a currency with infinite velocity has 0 value, and is therefore irrelevant.

That's why it is trivial, the velocity of money doesn't change the deal, unless it is varying wildly, and in that case, you're screwed anyway...

Why do we have to jump to infinity? Why can't the velocity of money vary based on market forces, including economic growth? Credit markets are the closest thing we have to visibility into monetary velocity.

This is all assuming that we're talking about modern Chicago-school monetary theory. In Keynsian (i.e. most mainstream economics) there's not even a universally accepted connection between money supply and price inflation/deflation. Most Keynsian theorists (and practitioners) agree that directly manipulating credit markets through artificial price floors on future money is a much more sane way to affect price inflation and deflation, while also having the secondary effect of manipulating the employment market.

The idea of an asset that always increases in value (or a "free lunch", let's call it) is explicitly disowned by every theory of economics, from the Austrians & monetarists to the Keynsians & the Marxists.

> Why do we have to jump to infinity? Why can't the velocity of money vary based on market forces, including economic growth?

Because the velocity of money isn't determined by market forces, it is determined by how easy it is for money to flow.

If bitcoin was the dream it aspires to be (instant, easy, transparent money transfers anywhere in the world), velocity of money tends to infinity, and the value of bitcoin goes to zero.

> The idea of an asset that always increases in value (or a "free lunch", let's call it) is explicitly disowned by every theory of economics, from the Austrians & monetarists to the Keynsians & the Marxists.

I'm not talking about an asset that always increases in value. I'm talking about a fixed supply of money, which would on average always represent more value (since the economy is growing on average) year after year.

As the supply of bitcoin will never increase, but there demand will (on average) always increase, a single bitcoin will represent a larger amount of value each year after year.

Therefore, it is a deflationary currency.

> Because the velocity of money isn't determined by market forces, it is determined by how easy it is for money to flow.

I mean... that's not what that means [1]. Monetarists use the "Equation of Exchange" [2] to determine/define inflation.

> I'm not talking about an asset that always increases in value. I'm talking about a fixed supply of money, which would on average always represent more value (since the economy is growing on average) year after year.

Bitcoin is an asset -- "always represent more value" is another way of saying "increase in value". What's the thesis here -- Bitcoin will decrease in value because it will increase in value? I think you'd do well to take a moment to digest what your argument is.

You're far from alone here; many people are under the impression that Bitcoin is a deflationary currency. This is a very common trope, made more annoying by the fact that Bitcoin proponents have a lean towards the Austrian school of economics, which defines "deflation" and "inflation" to mean a decrease (and increase) in the money supply, independent of prices. But that's not the mainstream definition of inflation; mainstream economics uses "debasement" to refer to the direct increase in the money supply, but it is used rarely since most economic schools feel that the money supply is only very weakly coupled to price inflation.

[1] https://en.wikipedia.org/wiki/Velocity_of_money

[2] https://en.wikipedia.org/wiki/Equation_of_exchange

> Bitcoin is an asset -- "always represent more value" is another way of saying "increase in value". What's the thesis here -- Bitcoin will decrease in value because it will increase in value? I think you'd do well to take a moment to digest what your argument is.

Where am I saying that bitcoin will decrease in value? Bitcoin is a deflationary currency, not an inflationary one.

If bitcoin was the only game in town, it would always (disregarding short-term cycles) increase in value, because the economy is growing.

>And the consequence is that everyone is better off holding currency than investing it. Which is disastrous for the economy.

Don't copy-paste your reply :)

But again, you are confusing deflation with deflationary currency.

perversely hyper* deflationary

We were on the gold standard in the late 19th century and the economy grew very fast. A fixed money supply puts more economic power in the hands of the industrialists and their retained earnings instead of the banks.

Gold standard != using gold as currency.

The gilded age isn't a joyful standard to follow, however

For the late 19th century, compared to the rest of the world, it was pretty good.

Reply on news.ycombinator.com