A Brief History of Paper Currency and Central Banking

In order to better understand how Bitcoin or some other crypto-currency may evolve to be used as money, I would like to discuss how early paper currencies first developed and came to be used as money, and why central banks were created to regulate them. That is the purpose of this post.

Imagine a world where gold (or maybe silver) is money. That is, pounds of gold or silver are used as the unit in which prices were quoted. (This is actually is how it was. It is why the name of UK money is pound sterling.)

Now, instead of keeping all their gold on hand, people would leave their gold with goldsmiths or deposit their gold at private banks. Private banks would then issue 'bank notes' to their depositors. To get their gold deposits back, people would would simply show up at the bank, hand over their note, and retrieve their deposit. Naturally, instead of going to the trouble of retrieving their gold deposits, people would simply exchange their bank notes when buying/selling goods. Modern banks still use this technology, in fact — we call them "cheques". (Americans call them “checks”.)

Now, of course, people would only accept bank notes as payment if they had faith in the bank that issued them — they had to be sure that the notes really could be redeemed for gold at the bank, no questions asked. This issue of trust is an important concept to remember as we will reference it from time to time. As trust in bank notes grew, more and more they were used as a medium for transaction. (The bank notes are not yet money because prices are still being quoted in gold. See the previous post.)

Except there were problems. Because bank notes were being used for transactions, depositors weren’t likely to redeem their gold too often, so banks found that they needed to keep less gold on hand in their vaults. Instead, the Banks found that they could lend out this extra gold and charge interest. This all works fine until you get an unexpected influx of people requesting to withdraw their gold and you don't have enough to fulfill their requests.

As a result, private banks become susceptible to bank runs: What happens when many people all at once lose faith in a bank and try to redeem their notes? Even if the initial concerns are unfounded, the act of many people withdrawing their deposits really will make a bank that has loaned out its deposits insolvent. You may have heard this referred to as a self-fulfilling prophecy or a self-fulfilling panic. An extension of this is when the loss of credibility in one bank causes people to question the credibility of other banks, even if only one of the banks has been behaving badly. All of a sudden, the loss of faith in just one bank causes many other banks to face huge deposit withdrawals as well. And since the huge withdrawals in fact do cause those once "good" banks to fail, you can see how a cycle would develop. This type of cycle is called financial contagion. Such a contagion happened in 2008 when US regulators allowed the investment bank Lehman Brothers to fail. Right away you can see the important role governments play in maintaining a healthy financial system, which we will return to.

With the widespread adoption of bank notes, very large banks soon realized that they had the ability to put any smaller bank out of business, overnight. How? Well due to their size, large banks accumulated large volumes of bank notes from smaller banks, and they could easily cause any small bank to fail simply by redeeming these notes all at once. Large banks didn't do this, of course, because they were also very well aware of the risk of financial contagion described above. The large banks thus developed a live and let live attitude: A large bank would allow a small bank to remain stable as long as it kept a deposit at the large bank. In the UK, there was just one large bank that played this role and it was the Bank of England. (Eventually the Bank of England became the central bank for the UK  we'll get to that.) In the US, there were several large banks playing this stability role, each dominating a different region. (This is why the US has 12 regional Federal Reserve banks  we'll get to that too.) It is hard to understate the important role these large banks played in maintaining a stable financial system.

Now there was still a problem with these bank issued notes: Not all bank notes had equal value. Notes issued by banks that were seen as less risky traded at a premium, so that even if two notes had the same par value (in terms of gold), they would not be treated as equal value. Merchants would have books full of exchange rates between the notes of issuing banks. These discrepancies made exchanges with bank notes harder than they were originally intended.

Soon, governments decided to take over the role of issuing notes and stabilizing the financial system, not leaving it in the hands of private banks. In the UK, the Bank of England was nationalized and was given sole authority to issue notes. In the US, the process was more evolutionary. At first, all banks were required to keep a deposit with the US treasury to back any notes that they issued. Moreover, the US treasury took over the responsibility of printing all notes, but outsourced the printing to the largest banks in each region  12 in total. Every bank note that the US Treasury issued looked exactly the same, except for the name of the printing bank which would be printed on the note. The US treasury would redeem any of its notes at the same level, no matter which bank printed it. As a result, all banks honoured each other's notes at par, thus fixing the bank note problem. Eventually, the US treasury created its own regional banks that would oversee the administration of both notes and deposits in each region  12 official regional banks were created with a headquarters in Washington DC, and this became known as the Federal Reserve system.

Now this didn't end the bank runs, unfortunately, because the same mechanics that allowed banks to lend out extra gold deposits also allowed them to lend out extra treasury note deposits. That's why governments had to take the extra step of creating deposit insurance and other regulations. Another important detail is that the Federal Reserve and the Bank of England took on the added responsibility of being the lender of last resort  but we won't get into all that except to say that being the lender of last resort requires access to an unlimited supply of money.

Banks which are granted the monopoly of issuing notes, like the Bank of England in the UK or the Federal Reserve in the US, are called central banks. In different countries, the framework of how these central banks originated are slightly different, but the underlying economics is the same: Governments would guarantee the value of all bank notes and in return private banks could no longer issue their own notes, only the central bank could do that. At first, these notes would be backed by gold, just like the original private bank notes, but that would eventually be removed for reasons soon explained. In America, notes issued by the Federal Reserve are called dollars

So are these dollars money? Not yet. Right now they are only a medium of transaction. Remember, dollars only become money once they are the medium of account  that is, once the price of goods are quoted in dollars (and not gold)  which is of course what happened.

Why were prices suddenly quoted in dollars? Well the reason is very natural: Since there was only one bank now  the central bank  issuing dollars that were convertible into gold, why not just quote prices in dollars instead of gold to make the transaction smoother? And that's indeed what happened. More importantly, though, the government actually made it against the law to quote prices in anything but dollars (i.e. legal tender), and required that all taxes be paid in dollars as well. (You weren’t allowed to pay your taxes in gold.) This created enough network effects to make dollars the medium of account: All prices were quoted in dollars, meaning that dollars were money.

But now notice something: We don't even need the dollars to be redeemable into gold for this system to work. The network effect of people using dollars is enough to give these "fiat" dollars real value. And you are probably already aware that dollars are no longer redeemable in gold. Instead, the dollar gets all of its value from the fact that it is the medium of account  the definition of money  and this is due to the network effects originally imposed by government. In short, dollars are money because dollars are money. And similarly, bitcoins are not money because bitcoins are not money. Yet.

Having the power to create money, central banks soon realized something very important: They realized that by changing the supply of money, they had the power to change interest rates, exchange rates, prices, and even employment, among other things. Having access to unlimited money is also important to the central banks' role of being the lender of last resort, ensuring that they are able to do so. Central banks thus take their responsibility very seriously and they play a crucial role in maintaining stability in modern economies. In fact, many economic crises have been caused by errors made by central banks  the Great Depression, the Great Inflation, Japan's lost decade, the recent Eurozone crisis and the Great Recession quickly come to mind.

The decisions central banks make over the supply of money thus received its own name and is called monetary policy. Through history, central banks have tried to use monetary policy at various times to set interest rates, exchange rates, prices and employment. Today, most modern central banks use monetary policy to keep inflation stable within a pre-defined range. We don't need to get into these details. The point is this: The job of a central bank is very important. We cannot live without effective monetary policy.

So that is the story of how paper currency came to be and how central banks use their monopoly over paper currency to conduct monetary policy. If you really grasp this story, the concepts described, I think you will understand the concept of money better than most economists you meet on a daily basis. The interested reader should also read the story of the Capitol Hill baby-sitting co-op.

Contrary to what Bitcoin proponents say, transacting paper currency is, like Bitcoin, very decentralized. If you and I want to transact using paper currency, we don't need anyone's permission, we simply physically transact. (Transacting through the internet, however, is still very difficult, and this is something Bitcoin is helping to improve.) However, only the central bank has power over the supply of the paper currency. So paper currencies are both decentralized (in transaction) and centralized (in supply). It is likely that crypto-currencies of the future will have this property as well. We will discuss this more in the next post.

What we need to think about is: Under what scenario would central banks adopt crypto-currencies in the same way they adopted paper currencies? And what innovations would this allow to monetary policy? I keep asking myself this question and in the next post I will try to explore this.

If you liked this article, please share, follow me on twitter or subscribe to RSS.


This is Part 3 of a discussion on the economic consequences of Bitcoin. You can access the remaining parts here:

1.  A Friendly Introduction for Economists to the Bitcoin Protocol
2.  A Monetarist View of Money and Bitcoin
3.  A Brief History of Paper Currency and Central Banking
4.  Will Bitcoins Ever Become Money? A Path to Decentralized Central Banking

A Monetarist View of Money and Bitcoin

I’m market monetarist leaning, so it's not surprising that I mostly agree with Scott Sumner that bitcoins are not money, except for me there is a caveat: it's not money yet, and at least not in its current form. I think, however, that crypto-currencies could become money under the right scenario. More on that later. The purpose of this post is to explain in detail why bitcoins are not money.

But first, what is money, exactly? As Sumner says, the ordinary conversational definition of money can be quite different from what economist call money, especially for market monetarists. For economists, what makes money money is how it relates to and effects the economy as a whole. We'll get to that in a second.

To get straight to the point, we define money as the medium of account: it is the unit in which all prices in the economy are quoted. This is different than a medium of transaction, which can be anything used to facilitate exchange. For example, you might accept apples from me as payment. You might accept a foreign currency. You might also accept bitcoins. That would make apples, foreign currency and bitcoins each a medium of transaction. But if you live in America, you would only accept these payments after converting your original US dollar price. Your true price is measured in dollars because everything else you want to buy in the economy are also priced in dollars. In America, that makes dollars the medium of account  that makes dollars money.

Now, the medium of account is almost always also used as the medium of transaction (that is, I will almost always give you money and not something else), but don’t let that confuse the difference. And don't be confused by the fact that different economies will have different money — for example, euros are money in Europe but not America, and dollars are money in America but not Europe. And then there's the fact that both dollars and euros are used as a medium of transaction pretty much everywhere (most places in Europe will gladly accept dollars, after converting from the euro price of course). Notice that I can say a medium of transaction but will only say the medium of account. Anything can be a medium of transaction, but an economy has only one medium of account.

Let’s try to explore this further with a realistic bitcoin example. Suppose I am selling you something. When you ask me for the price, I will give you a quote in dollars. Dollars are money, after all, and money is my chief concern because everything else in the economy is quoted in money. If you request to pay me in bitcoins, I will simply you give an equivalent bitcoin price by converting my dollar price using the most current exchange rate. Once you send me your bitcoins, I will then immediately convert them into dollars so that I end up with the original dollar price I wanted in the first place (this is literally how it works at every website that says it accepts bitcoins as payment). If the exchange rate between dollars and bitcoins changes, the amount of bitcoins I will require from you will change as well, so that the price I quote you in bitcoins could literally fluctuate by the minute depending on the volatility of the exchange rate. The dollar price, however, will remain unchanged  economists call this phenomenon price stickiness. This is a key feature of money: Prices denoted in the medium of account tend to be sticky, while prices denoted in other units tend to fluctuate. Price stickiness is what makes the medium of account so important.

How exactly did dollars become the medium of account? Patience, we’ll get there.

From here on I will use the words “money” and “medium of account” and “unit of account” to mean the same thing — they are all interchangeable.

So what’s so special about money? Well as Sumner explains, money has a powerful influence over the economy. In particular:

· Changes in the supply/demand of money cause inflation/deflation.
· Changes in the supply/demand of money cause changes in interest rates.
· Changes in the supply/demand of money cause business cycles.

Changes in the supply or demand of bitcoins do none of these things and that alone means they are not deserving of the name “money”. The reason money has these effects is because of the price stickiness I described above and a similar phenomenon called money illusion. If you want to understand these effects, you should read Scott Sumner’s blog or read this paper. And since central banks control the supply of money, the decisions they make greatly impact the economy — we call decisions concerning the supply of money monetary policy. As an example, in Canada, the Bank of Canada has decided that it will control the supply of money in such a way so that medium-term inflation is 2% — and they have been very successful. The ability to conduct monetary policy is a crucial feature for Bitcoin or other crypto-currencies to have if they are ever to become money.

The question I would like to explore is: Under what scenario would bitcoins become money? And what would monetary policy look like if they did?

Understanding the difference between money and something that is simply a medium of transaction will be important to understanding why a central bank might adopt a crypto-currency like Bitcoin as money and how monetary policy would be conducted in such a world. Before that, we need to discuss how paper currencies originated, why central banks were created to regulate them and how they became to be money. That will be next — stay tuned.

If you liked this article, please share, follow me on twitter or subscribe to RSS.


This is Part 2 of a discussion on the economic consequences of Bitcoin. You can access the remaining parts here:

1.  A Friendly Introduction for Economists to the Bitcoin Protocol
2.  A Monetarist View of Money and Bitcoin
3.  A Brief History of Paper Currency and Central Banking
4.  Will Bitcoins Ever Become Money? A Path to Decentralized Central Banking

A Friendly Introduction for Economists to the Bitcoin Protocol

Bitcoin has become an intense area of interest of mine as it overlaps several of my main interests: mathematics, monetary economics, technology, innovation. The more I learn about Bitcoin, the more I see how wrong the popular press presents it. To discuss and explore Bitcoin from a monetary economics perspective, we first need to understand what Bitcoin is because so few economists seem to know. The purpose of this post is to explain the Bitcoin protocol in a less technical format so that ordinary economists can understand how it applies to economic theory.

Bitcoin is merely a protocol. A set of rules. The protocol tells us how to assign ownership of bits of information without duplication — these bits are called bitcoins. As in other sources, I refer to the protocol as Bitcoin (with a capital B) and the units of currency as bitcoins (lowercase).

The Bitcoin protocol consists of two main components: a ledger and a network. The ledger, also called the blockchain, records all transactions that use the protocol. Copies of this ledger are stored on computers that comprise the network. Any computer can be part of the Bitcoin network, it just has to be set it up for such. If two people want to enter a transaction of bitcoins, they simply have to record their transaction on the ledger. To do this, they announce the transaction to the network, and each computer within the network records the transaction on their respective copy of the ledger. The ledger will identify the two people using their Bitcoin addresses, which you can think of as a Bitcoin account number. It is very easy and costless to create a Bitcoin address, and you don’t need anyone’s permission.

But before recording the transaction, the computers in the network will verify that the payer actually owns the bitcoins being transferred. To verify ownership, the payer must provide a digital signature that is associated with their Bitcoin address and the bitcoins being transferred. The computers use this signature to solve a (hard) mathematical equation which, if solved, proves that the bitcoins being transferred really do belong to the payer. After verifying ownership, the transaction is grouped with other yet-to-be-processed transactions to form a block, which still needs to be processed. To process the block, the computer solves another (very hard) mathematical equation — this process is called proof of workOnce the block is processed, the block is part of the blockchain, and the transaction is complete.


The proof of work for each transaction is dependent on the proof of work of the preceding transaction, and therefore every other preceding transaction. This means that the ledger is backward dependent, and it is thus impossible to go back and rewrite a transaction without affecting every transaction that follows. It is for this reason that the public ledger is referred to as a blockchain and each transaction is referred to as a block. The backward dependence of the blockchain makes it (nearly) impossible to rewrite the transaction history, as changing even one transaction would require the computational power to change every following transaction as well.


Now why would anyone donate their computers towards the network to solve these equations for someone else’s transaction? Because the first computer to process a block is given a block reward of new bitcoins — this reward is built right into the Bitcoin protocol and is called mining. Every bitcoin in existence is created by mining — there is no way to remove it. The number of bitcoins rewarded for processing a transaction reduces over time and converges to (but never reaches) zero. By this construction, there will never be more than 21 million bitcoins ever in existence, making bitcoins scarce. This rate of reduction is arbitrary, however, and the block reward can actually be adjusted to any value that the network collectively agrees upon. This fact will be very important if/when central banks choose to issue their own crypto-currencies. (Although many proponents of Bitcoin cite the built-in scarcity as a key strength of it as a currency, it is actually a fatal obstacle to it ever becoming real money.)

So how can you acquire bitcoins? Like I said, you have to mine them by donating your computer to the network to process blocks. The more computing power you donate to the network, the more bitcoins you are likely to be rewarded. So if you want a lot of bitcoins, you have to buy a lot of computing hardware, run them using a lot of electricity, and connect them to the internet with a lot of bandwidth. Of course, you can just buy bitcoins from people who have already mined some. At the time of this posting, you can buy 1 bitcoin for 550 US dollars in the open market.

You might be thinking: Why is this such a big deal? Remember, one of the key powers of computer technology is that bits and digital information can be easily copied and duplicated, which many media companies know all too well. But forget media companies, duplication is a fatal problem for financial assets to have. Credit card companies and financial institutions spend huge resources on security and fraud prevention, making digital transactions very slow and very costly. With Bitcoin, we now have an easy and low-cost way to create bits that cannot be duplicated. And as I've alluded to, Bitcoin is just one implementation of a crypto-currency protocol: We can easily create other crypto-currencies similar to Bitcoin by simply changing some of the underlying rules of the protocol (and people have, such as Litecoin and Dogecoin). So when I talk about Bitcoin, I am really talking about crypto-currencies, and more generally the blockchain, as a technology where "Bitcoin" is just one implementation. It is like talking about paper currency as a technology, as opposed to Dollars or Yen.

Now to be clear, Bitcoin doesn’t actually stop the duplication of bits, it simply tells us who owns the bits (remember that it is impossible to rewrite the transaction history of the ledger). But for the sake of economic analysis, this distinction is not important: Economists can simply think of the protocol as telling us how to create bits that cannot be counterfeitedBitcoin, or the blockchain in general, facilitates many applications, not just as a currency. For example, you could imagine using a Bitcoin-like protocol to trade shares in the stock market: you simply assign one share for one crypto-unit. Trading stocks then becomes very easy and without the need to have a broker (or pay their high fees). The creation and trading of options and other derivative contracts also then become seamless, transparent and very low-cost. The blockchain in general can be used to store bits of information where the ownership of those bits is well-defined, but that's an application that monetary economists need not worry about. I hope you truly grasp what a powerful technology Bitcoin and the blockchain are. It is also important to notice that the protocol doesn't tell us how to apply it or what to use it for. It especially doesn't tell us what the value of bitcoins are. Bitcoin is simply a tool — who knows what other applications are out there.

Also a note on anonymity: Bitcoin is touted for its anonymity, but this is not accurate. Every single Bitcoin transaction is recorded on the public ledger. If someone can find a way to associate you with your Bitcoin address, they will know every single transaction you have ever conducted. In a sense, Bitcoin is the most transparent financial instrument ever. Of course, you could create multiple Bitcoin addresses…

So there we have it, that's what Bitcoin is. It is a way of transacting digital bits which cannot be counterfeited. That is a very important quality for money to have, although not the only one. I will soon explain.

If you liked this article, please share, follow me on twitter or subscribe to RSS.


This is Part 1 of a discussion on the economic consequences of Bitcoin. You can access the remaining parts here:

1.  A Friendly Introduction for Economists to the Bitcoin Protocol
2.  A Monetarist View of Money and Bitcoin
3.  A Brief History of Paper Currency and Central Banking
4.  Will Bitcoins Ever Become Money? A Path to Decentralized Central Banking

Has the Bank of Canada developed a credibility problem?

Andrew Coyne says the Bank of Canada has abandoned its policy of targeting inflation, allowing prices to increase in order to accommodate a lower Canadian dollar. Coyne says that this new policy objective is the result of direct interference from the Federal Finance Minister. I think that's ridiculous. Let's discuss.

Coyne's evidence of the shift in policy:

  • Finance Minister Jim Flaherty said earlier this month that a lower dollar would be good for exports and the economy, and that interest rates would rise later this year.
  • A resurgent manufacturing sector in southern Ontario would be good for Tory election prospects.
  • Flaherty is rumoured to have been heavily involved in the selection process of appointing Governor Poloz.
  • Poloz used to work at Export Development Canada.
  • The Canadian dollar fell after the latest policy rate announcement. 

All these points are true. The first one is especially irresponsible for a Finance Minister. But why do they imply that the Bank of Canada has abandoned targeting inflation for the exchange rate? We have to check if the Bank of Canada really has abandoned its inflation target (keep in mind that this target is legislated — the Bank is obligated by law to pursue it).

To start, nowhere in the Bank's recent Monetary Policy Report or in Poloz's public statements is there any mention of a desire to see a lower dollar. Indeed, the day after Flaherty's irresponsible comments, the governor went out of his way in an interview with CBC to say that the Bank does not care about the exchange rate. And many, many times, in every publication and news release, the Bank has said over and over again that it only cares about inflation. That has been true for years and it has not changed under Poloz. Poloz is in fact the first governor to describe the inflation target as "sacrosanct." All monetary policy at the Bank of Canada is viewed through the lens of inflation. If you don't understand this, you will not understand the Bank of Canada.

But those commitments are just words. What has actually happened — is inflation staying on target? Coyne says that inflation is currently not that far below target and that the dovish language in this week's policy announcement was thus unwarranted to maintain the Bank's target. But that's not true. To see, you just need to graph it:


Since inflation targeting began, inflation has been following a consistent trend. But somewhere in 2011, inflation slowly began to diverge below trend. That is, inflation has actually been too low. It's very slight, but the Bank of Canada does not wait for huge divergences to act — it tries to steer inflation in real time. As you can see, this divergence is the biggest since inflation targeting began. This is what Poloz means when he says that inflation has been "too low for too long."

 (The Bank says that it targets the inflation level and not the path of CPI above, but never mind that. If you don't believe me, think of the Bank as targeting the slope of the graph above and the argument still holds.)

The point is, to stay on target, the Bank clearly needs to loosen policy, which it has done this week. And looser policy happens to also depreciate the dollar. So what? This doesn't mean the Bank is now targeting the exchange rate instead of inflation. It is not. The Bank cares about inflation — and it only cares about the exchange rate insofar as it affects inflation.

That's what the Bank says and the evidence is consistent with that. All the evidence suggests that the Bank of Canada is simply following its inflation mandate. There is no conspiracy.

But Coyne's tirade does raise an important issue: Does the Bank of Canada have a communication problem? Is Coyne some crank pundit trying to sell a headline or should his complaints be taken seriously? I don't know. (Coyne's father was governor of the Bank of Canada many decades ago and was instrumental in establishing central bank independence, but I don't think intellectual authority can be inherited.) Though Coyne emphasizes interference by the Federal Finance minister, several other news agencies also reported this week that the Bank of Canada is attempting to devalue the Canadian dollar — it's just not true.

Many people think that the Bank of Canada works by adjusting interest rates, but that's not actually true. In fact, in the policy rate announcement this week that Coyne has taken issue with, there was actually no change in the policy rate whatsoever. The dirty secret of monetary policy is that the real tool is in communication, for reasons we won't get into right now. If the Bank is failing to communicate its objective, if its credibility is in jeopardy, the Bank of Canada has a serious problem.

How to judge a central bank?

The Bank of Canada is often credited for having fared better than the US Fed in handling the Great Recession. In fact, Mark Carney was handed the Bank of England governorship mainly because of the Bank of Canada's success.

But you can't see that success when you look at nominal GDP (cc Scott Sumner):

On the other hand, Australia has done very well:

Australia hasn't had a recession in 20 years. 20 years! But like Canada, Australia says it targets inflation, and certainly not nominal GDP. Yet Australia is faring much better.

Very peculiar. Maybe we're overlooking something.