In the first couple of posts on government spending, I’ve inevitably written about the money creation process – i.e. where money comes from, how it enters the economy. I’ve spoken explicitly about how the government creates fiat money; but without talking about how commercial banks also create money, or more correctly credit money, we’re missing half of the story.

Professor Steve Keen gives a technical treatment of this on his blog

Credit money is different to fiat money. Fiat money can be thought of as a final means of payment. In the economy as a whole, it is money that can extinguish all debts. This government created money does not need to be paid back. Remember fiat money is created by the government when they credit bank accounts held by the private sector.

Credit money on the other hand comes with a corresponding debt. It needs to be paid back at some point down the line. It may be used to pay off a debt, but one debt is merely replaced by another. So when banks create credit (make a loan), by definition it must be paid back. This is the fundamental difference between state/fiat money and credit money.

Banking; the conventional story

Now when we think about the role of banks, there’s a popular story that gets bandied about. This story says banks are simply financial intermediaries. That they simply act as middlemen between savers and borrowers, suggesting that they take in deposits, and they make profits by lending a portion of those deposits. This theory of how banking works is called the loanable funds approach. In this approach, banks create (credit) money by lending out the deposits of the private sector (households and firms). Where do the deposits come from in the first place? From the government creation of fiat money, referred to as the money base.

Money base is created when the government credits the bank accounts of the private sector. So the conventional theory says that only when this occurs, i.e. a deposit is made with the bank, will the bank then lend out the deposit to a borrower, and keep a small fraction as reserves for prudential reasons. Remember in the last post I spoke about how reserves are held to ensure payments clear on a daily basis. From that first loan extended by a bank, call it bank A, the borrower may then go and spend that money on a house. The seller may receive the payment in the form of a deposit in another bank B. Bank B can then loan out that deposit keeping some small fraction as reserve, and the process plays out again. The next borrower takes the loan from bank B, and spends it at a retailer. The retailer receives the payment in the form of a deposit in bank C. The process plays out again…. In Professor Keen’s example, an initial injection of $1000 of fiat money with banks holding onto 10% as reserves and lending the rest creates an additional $9000 of credit money ($9000 of corresponding debt). After an initial injection of fiat money or money base, the banking system ends up multiplying this initial deposit by several magnitudes. This is supposedly how banks create money.

This loanable funds approach provides some of the theoretical grounding for ideas like government spending causing inflation. If government creates lots of deposits in the banking system, the banks will end up making huge amounts of loans to the private sector multiplying the supply of money/credit, and the amount of credit will outpace the supply of goods and services causing inflation.

This approach also provides some of the theoretical grounding for ideas such as giving money to the banks to stimulate the economy in times of recession. The idea ofcourse that when money is given to the banks, they lend it out to households and firms who will use the loans to increase consumption and investment and create employment.

If you’ve tuned into what has been happening globally, central banks around the world have engaged in what is known as quantitative easing. Remember in the last post I discussed how central banks, when they try to maintain the rate of interest, they sell the excess reserves created by the government as bonds to the private sector, draining reserves from the system and maintaining their target interest rate. With quantitative easing, the central bank instead tries to create some quantity of excess reserves, and it will do this by purchasing those bonds from the private sector. When the central bank buys government bonds – the commercial bank recieves payment in the form of increased reserve balances. The central banks demand for government bonds increases the price for government bonds, which encourages banks to lend out the excess reserves for other profit making opportunities rather than simply buying bonds. This will supposedly stimulate consumption, investment, and employment as banks lend out the reserves.

The Evidence?

So how do we know this theory is wrong? In the aftermath of the global financial crisis, in 2008 the US Federal Reserve tripled the money base over 2 years. That is, it created nearly $1.8 trillion of fiat money out of thin air (after approval from the government/treasury), and ‘purchased’ government bonds from commercial banks, adding reserves to the system, in the hope that banks would lend it out and stimulate the economy. What did the banks do? They simply sat on the reserves.


Source: US Federal Reserve

The graph above shows the money base tracked since 1959, as well as M2. M2 is the money aggregate used by the Federal Reserve (as well as the RBA in Australia) that measures the amount of money in the private sector including credit money. This includes money in people’s savings accounts, term deposits, superannuation funds, cheque and savings deposits held by businesses and firms. It is essentially a broad measure for the amount of money circulating in the system. You’ll see that there is substantially more credit money circulating in the system than fiat or base money. In the graph, there isn’t a big problem with the loanable funds story until about 2008 where you see the Federal Reserve start quantitative easing, and the money base triples in size over 2 years. Yet M2, or the amount of credit money the banking system would supposedly create does not multiply correspondingly.

The graph below shows the ratio of M2 to the money base. Remember that if banks simply lend out deposits created by the government as per the loanable funds approach, then the Federal reserve creation of money base should have corresponded to a multiplied increase in credit money. It did not. It had next to no effect, in fact the amount of credit money with respect to the amount of money base is at its lowest level in 50 years. Quantitative easing, or creating money base/fiat money and giving it to the banks has had no effect on banks’ credit creation. Something is amiss.


Professor Keen explains additionally that if the loanable funds approach were true, not only should we see multiplied increases of credit money in response to base money, but by that very definition there should be some time lag between the two. That is, if the banks require deposits to make loans, we should see base money increase in the system first, and only then credit money increase some time after. This is the clincher. Professor Keen cites research by two leading mainstream economists that shows that M2, or credit money, consistently grows before money base. It has taken on average 9 months for money base to start growing in response to credit money growth. Notice that the causation is completely reversed.

Not only did the tripling of fiat/base money in the US cause no inflation (inflation is less than 2%), giving it to the banks created next to no employment. In January 2008 the unemployment rate was 5%, today it is 7.8%. This does not include the portion of the unemployed who have stopped looking for work, but say they want to work. If these people are included the unemployment rate actually climbs to 14.7%. This means there are still 22 million people in the US who are unemployed who want a job!

Source: Bureau of Labor Statistics

How can this be so?

Banks Do Not Require Deposits To Make Loans

Banks make loans first, and create deposits in the process. They do not loan out deposits as convenient a picture that may seem. How do they do this? Through double entry bookkeeping. Say you go to the bank for a $1000 loan. If you are credit-worthy the loan gets approved. The bank then credits your account with $1000. The bank also correspondingly adds to their assets this $1000 loan (a loan is an asset to a bank because it makes profit through the interest on the loan). The bank gives you $1000 and then essentially says that you owe the bank $1000. The money needn’t have come from anywhere. And the loan creates the deposit in the borrower’s bank account. Credit creation out of thin air again. And what does this credit creation depend on? Demand from credit worthy borrowers, not on reserves. If there is demand for credit and borrowers are credit worthy, the bank sees a profit opportunity and will make the loan. If the bank is short of reserves, it will borrow this at a later date from the central bank in the overnight market. Loans do not depend on reserves, they depend on demand for loans. This is why credit money grows in the economy first. With credit money there is a corresponding debt. This means whenever banks create credit money, they create debt.

So when you have a system that is heavily indebted (the ratio Private debt to GDP is 270% in the US and 160% of GDP in Australia) trying to stimulate employment by stimulating banks’ credit creation simply adds more and more debt to the system. This is an important point that Professor Steve Keen has devoted his blog to. If people have too much debt (their income cannot service any more debt), then there is no demand for banks’ loans, and no amount of additional reserves or quantitative easing will persuade the private sector to take on more debt.

Fiat money is different in this regard. There is no corresponding debt in the private sector as discussed. The government simply spends fiat money into existence, and the money does not need to be paid back. So the government is the only party that can create net financial assets in the economy. It can create money that can extinguish the private sector’s aggregate level of debt. If the private sector is heavily indebted, the government creation of fiat money can stimulate consumption, investment, and employment, because it does not need to be paid back – the private sector does not need to be persuaded to accept fiat money the way it needs to be persuaded to take on debt.

Now debt is not a bad thing per se. In fact debt is necessary. If debt is used to create investment, create goods and services, and produce more than enough income to consume those goods and services and service that debt… well this is the debt that fuels technical progress, and rising standards of living. However the inevitable outcome of banks’ ability to create credit out of thin air and their profit motivation to do so means that the banking system wants to create as much private debt as possible. To do this their lending standards over time fall, especially in good times. The fall in lending standards means banks end up financing activity that will NOT produce enough income to service the debt created. So there inevitably comes a point where creation of debt starts to outstrip the amount of income being generated. When the private sector starts to feel this debt burden, this is when a crisis starts and you see a downturn in the economy. There is a fall in demand for goods and services as people start paying off their debts, employers start laying off staff, prices of assets like houses and shares get re-evaluated downwards amplifying into a recession. I will save the detail for a later post.

Reality Check

So when you hear about the boom and bust cycle, what this business cycle actually refers to is a financial (banking) sector cycle. A debt cycle. Booms are driven by private sector credit creation, and busts occur when private sector debt exceeds the debt servicing capacity of the economy (when the income generated cannot keep up with debt service costs). Notice how small a role government spending and fiat money creation actually plays in this process. This boom and bust cycle, this debt dynamic, occurs independently of government, it is how capitalism works because it is how banking works. In fact this kind of debt cycle is not unique to capitalism, and traces back thousands of years to antiquity. It follows that there much room for fiat money creation to reduce volatility in the debt cycle.

One of the greatest failings of modern economics is how it has obscured, rather than aided our understanding of money.  You would think economists would be all over it, but their models do not include money and debt – what they essentially model is a barter economy where everyone simply exchanges one set of goods and services for another set of goods and services. When we start to understand money, more specifically fiat money and credit money, then we start to see solutions to things like unemployment, how we can fund public initiatives and even the transition to a low carbon economy.

If we can regulate banking such that lending standards remain strong over time, and encourage banking credit to finance investment and some consumption rather than speculation, we’ll avoid large asset price bubbles, and reduce volatility in demand and employment. If we use government money to create employment for those who want to work but cannot find it in the private sector, we can have true full employment and utilise idle labour for productive initiatives without inflation problems. The full story is that we live in a system of Credit and Fiat Money. The banking system and government sector can create credit money and fiat money at will, let’s use it!