25 Apr 2013

Just how much money is there?

I've recently been arguing that there is simply not enough money in the system to pay off the total amount of public and private sector debt. My original calculations were based on just a set of 39 countries that are covered by the BIS Private Sector Debt figures.

For them, I obtained values of $136.89 trillion for combined public and private sector debt, and a figure for total money supply of $68.34 trillion - almost exactly half of the debt level.

It's this massive difference that has got me thinking hard. I suspect that much of this difference could be due to the effects of accumulated interest charges. This is simply based on the idea that if a commercial bank creates $1 million in credit (debt) and charges 5% interest, and if the debt is not paid off, the total amount of debt will have more than doubled in just over 14 years.

I've had a couple of other suggestions for how you might get such a discrepency. Ralph Musgrave suggested that you might get more debt than money if people were delivering goods (he mentioned coconuts) without being paid upfront. This can create effective debt. Quite whether there could be enough such activity to explain the huge gap between the money supply and debt is very unclear to me.

Another suggestion is the possibility that someone could use borrowed money to buy a car, borrow the same money again and buy a second car. I'm not sure how to make this work on a large scale.

So, at least for the time being, my bet is that the big gap is the result of compound interest and the fact that governments have got into the very bad habit of never actually paying off debt. Instead, they just wait until the repayment is due, and then borrow again to pay back the loan plus an extra sum to cover the interest.

Nevertheless, my argument depends on having a clear idea of precisely what the size of the money supply is. There are various measures that are used - M0, M1, M2, M3 and even M4 in some countries like the UK. There don't seem to be well established definitions that are shared. Even worse, the most inclusive measure (namely M3) is not used by many important economies, including the USA.

Using datasets that I was able to download from the remarkable "Trading Economics" website, I have compiled the following table that shows M0, M1, M2 and M3 figures for all the countries where such numbers are available. I converted the raw data into millions of LCU (local currency units) and used the current exchange rate with the dollar to get the distribution of money supplies for all available countries. Here are the results.

There are lots of interesting things to note. First, there's the total for the 99 countries which comes to $71.36 trillion. The major players are clearly China with a money supply of around $16 trillion, the Euroarea with $12.8 trillion, the US with $10.4 trillion, Japan with $8.3 trillion, and the UK with $8.3 trillion.

There are three countries where I was unable to extract an exchange rate (Ecuador, El Salvador and Kuwait), but otherwise, 59 countries have a value for M3, for 34 countries I was forced to use the M2 value, and for 4 countries, the only available number concerned M0. On average, the value for M3 was about 27% higher than M2, which means that it may be possible to get an estimate of what M3 might have been for the 34 countries with only M2 available. This suggests that the number based on M3 would have been around $10.5 trillion higher  - around $82 trillion. The absence of an M3 number is particularly critical for China, where an M3 based value might be over $20 trillion, the US, where an M3 based value could be around $13.2 trillion, and Japan where the M3 based total may be around $10.6 trillion.

One of the take home messages is that clearly we really need a much more methodical approach to measuring the global money supply. Nevertheless, it is clear that even if we use the more generous estimates, there is simply no way that we are going to be able to explain away the huge gap between total money supply and the total amount of debt. My claim that we need someone (the World Bank?) to generate the missing money as debt-free credit stands.

One last point concerns the ratio of M3 or M2 to M0. If M0 is defined as the amount of narrow money (cash and coins) in the economy, it gives an idea of the percentage of the money in the system that is created by banks. In the UK, the Positive Money group often talks about 97% of the money supply being created by commercial banks, and this is visible in the M3/M0 ratio which is  37:1.

If we look across all the countries for which such numbers are available you get an average  value of about 94.4% bank created money using M3, and 91.4% bank created money if you use M2.


  1. Dude, you are spot on with this one. The problem is interest to be paid is never counted as debt, until the money borrowed to pay back the interest is rolled over. If the govt issues $1b in debt, the total bill for it, depending on the interest rate, easily be $2B. But the govt only put 980m into circulation (assuming a 2% discount on the face value of the bond at issuance) And don't forget about those discounts at issuance, that adds up to a fortune too.

  2. Jack, I don't think that it's governments that issue debt - they take on debt by borrowing from the markets. It's the commercial banks that have been creating the vast majority of all the money/debt in the system. And yes, you just have to leave that debt for a bit and the monster gets even bigger.

    The solution? Stop creating money as debt. Take back the money creation powers that have been handed over to commercial banks, and replace them with debt-free money creation by central banks in the public interest.