The chasm between what the man and woman on the street knows about monetary policy, and the jargon-filled gibberish of economists and financial analysts, is probably deeper than for any other economic topic.

For example, how do governments "create" money? That is the very relevant question a Sun reader asked me by email a couple of weeks ago after reading one of my columns. Although the financial press is filled with news about "QE3" (a possible third round of "quantitative easing," the euphemism for money creation, by the Federal Reserve), it is indeed not obvious how this is being done.

Many people will recall seeing images on TV of printing presses churning out large sheets of dollar notes. But physical cash is only a tiny portion of existing money. Most of it nowadays simply exists as digits in computers. The main way central banks put more dollars into circulation is not by printing it but by buying government obligations and crediting the seller's account with previously nonexistent money.

You might ask: Why buy obligations? Why not directly hand over the money to the government?

Central banks superficially function like normal commercial banks, with assets and liabilities. They may be able to create money at will, but when they "spend" that money, they need to have some new asset as a counterpart so that their books balance.

This has several advantages. It makes it easier to manage the money supply. Central banks usually buy government obligations not directly from the government as they are issued, but from financial institutions that hold them. They can thus influence how much money commercial banks will have to lend.

It also allows them to reduce the money supply if the need arises, usually in order to tame price inflation. They do so by selling back the obligations they bought. They then "destroy" the money they get as payment, just like they created it out of thin air in the first place.

These operations take place in a context where central banks are part of the government apparatus, but autonomous to some extent. Current mainstream thinking in economics, very much influenced by Milton Friedman, is that when governments can simply order their central banks to print more money, it usually leads to inflationary disasters, as in Germany in the 1920s, Argentina in the 1980s and Zimbabwe a couple of years ago. Politicians have no incentive to destroy money: they usually want to spend more! It's safer to keep some distance between them and the money machine.

There are economists who nonetheless criticize the current system as still too inflationary and prone to creating boom-bust cycles. Friedrich Hayek for example defended the idea that money should be "denationalized" and managed as a private good to avoid the pitfalls of government meddling. This is not an idea that is as revolutionary as one might think. Indeed, the Bank of Canada was only created in 1934 and the U.S. Federal Reserve in 1913.

Central banks have constantly expanded the money supply in recent decades and thus debased the currency's value. This explains why stuff you could buy for a dollar in 1970 will cost you $6 today. They've simply been doing it at an extraordinarily faster pace since 2008. And they've been buying not only government obligations as was traditionally the case, but also other types of securities, including so-called "toxic assets" related to the subprime crisis. That could threaten the value of their balance sheets and make it more difficult to reduce the money supply in the future if nobody wants to buy these worthless assets.

Granted, monetary economics is one of the most boring and technical topics in the field of economics. But given what is at stake in this risky experiment, we all have an interest in better understanding what is going on.

Follow Michel Kelly-Gagnon on Twitter: www.twitter.com/iedm_montreal