How much money can banks create? [Banking 101 Part 4]

You’ve probably seen the standard multiplier explanation of fractional reserve banking that we discussed in an earlier video. In this model, the banks have to keep a percentage of their customers’ money in a ‘reserve’. The reserve ratio given is usually 10%, which means that for every £100 paid into a bank by customers, the bank must keep £10 in a reserve somewhere. This means that the banks can only expand the money supply up to 10 times the amount of real, government created money. We said that this model of banking is completely inaccurate, at least in the UK. For a start, the required reserve ratio in the UK isn’t 10% – it’s zero. But more fundamentally, the reserve ratio would only actually limit the amount of money that banks can create if the ‘reserve’ money was actually taken out of circulation and put into a safe deposit box, or an electronic equivalent. If the Bank of England actually required banks to hold £10 of cash or central bank reserves for every £100 that they typed into their customers’ bank accounts, then that would limit the money supply to around 10 times the amount of base money (the cash and central bank reserves). The pyramid model would then actually apply.

Liquidity ratio?

But this is almost never what happens. When there was a reserve ratio in the UK, it was what’s called a liquidity ratio. A liquidity ratio is deceptively similar to a reserve ratio, but fundamentally different. A liquidity ratio requires banks to hold liquid assets equal to a percentage of their deposits. So if a liquidity ratio was set at 10%, then a bank with £100 in a customers’ account would need to hold £10 of liquid assets.

Now you’re probably thinking, what’s the difference between this and the normal reserve ratio? Well the key point is the term ‘liquid assets’. Liquid assets include cash and central bank reserves, but they also include other things, in particular government bonds. While the reserve ratio used in the textbook model of banking requires banks to hold cash and central bank reserves in proportion to the total balances of their customers’ bank accounts, a liquidity ratio actually allows the banks to use that cash and central bank reserves to buy bonds. The bonds also count towards the liquidity ratio, meaning that the bank could not hold any cash or central bank reserves and still meet the ratio. But the key detail here is that when a bank uses central bank reserves to buy bonds, the central bank reserves then belong to another bank. In other words, they’re not removed from circulation – they’re still circulating through the system. This means that a liquidity ratio, as opposed to a proper cash-and-central-bank-reserves ratio, has no limiting effect on the total amount of money that the banking sector as a whole can create. So a liquidity reserve ratio will not limit the banking sector’s total ability to create money.

We did used to have liquidity reserve ratios in the UK. In fact, from the mid-19th century banks tended to keep an average of sixty percent of liquid assets as a proportion of their total liabilities. This was actually a self-imposed reserve requirement – it’s what the banks knew they needed to keep back in order to avoid the risk fo a run on the bank. In 1866 there was a banking crisis, and the Bank of England then took on the role of ‘lender of last resort’, committing to lend to banks if they ran out of money to make their payments. Once this safety net was in place, banks reduced their liquid reserves to around 30%. In 1947, when the Bank of England was nationalised, they imposed a formal liquidity reserve ratio of 32%. This reserve ratio required banks to hold £32 of cash, central bank reserves and government bonds for every £100 balance in customers’ accounts. Of course, because government bonds would earn the bank some interest, unlike reserves and cash, the banks would try to hold as much of this 32% as possible in the form of bonds. In 1963 this liquidity ratio was dropped to 28%. Then, in the words of the Bank of England, “Before 1971, the clearing banks had been required to hold liquid assets equivalent to 28% of deposits. From 1971, this was relaxed and extended, requiring all banks to hold reserve assets equivalent to 12.5% of eligible liabilities…. This combination of regulatory and economic factors coincided with one of the most rapid periods of credit growth in the 20th century (Chart 10). It also contributed to an ongoing decline in banks’ liquidity holdings, ultimately to below 5% of total assets by the end of the 1970’s.” In this phrase, ‘credit growth’ really means a massive expansion in the amount of bank-created money, and consequently a massive rise in debt. Finally, in 1981, the liquidity reserve ratios were abolished all together.

Natural limit

So if the Bank of England no longer sets a liquidity reserve ratio, is there a natural requirement for banks to keep liquid reserves in proportion to their total customer accounts? In other words, is the system naturally limited? Well let’s look at the central bank clearing system again. Remember that there are 46 banks with reserve accounts at the Bank of England. At the end of the day when all payments are cancelled out against each other, these banks have to ‘settle’ between themselves by transferring money between these reserve accounts. Now the important thing is that this system of central bank reserve accounts is a closed loop. It’s technically impossible for any central bank reserves to leave the loop, because central bank reserves are by definition numbers in accounts at the central bank, and only the Bank of England is able to actually create or destroy central bank reserves. So, when all the payments are cleared at the end of the day and the banks find out how much they actually need to transfer to settle up, some banks will end up having to pay money to other banks ,and other banks will end up receiving money from other banks. What happens if one bank doesn’t have enough central bank reserves at the end of the day to make it’s payments to other banks? Well because it’s a closed loop system, it’s mathematically certain that one of the other banks will have more money than it needs to make it’s payments. What happens then is that the bank that has more central bank reserves than it needs lends some of them to the bank that doesn’t have enough. This lending of central bank reserves between commercial banks is called the inter-bank lending market. And as long as the banks that end up with more reserves than they need are happy to lend it to banks that have less reserves than they need, then all banks will be able to make their payments, and there’s nothing to worry about. So a bank can actually make a loan, creating new money in the hands of the public, even if it doesn’t have the reserves, because it knows that at the end of the day, when all payments are netted out against each other, another bank will be there willing to lend it some reserves to settle its own payments.

So as long as all banks are increasing their lending at roughly the same rate, the money supply can keep increasing without the need for additional reserves. So banks don’t really depend on having reserves before they can create money. They can make the loan first and find the reserves to settle the payment by borrowing them from another bank. And collectively, banks can increase the money supply almost indefinitely without being restrained by the amount of central bank reserves. In fact, before the financial crisis the ratio between the bank-created money in the hands of the public, and the central bank reserves, was 80:1. Of course, this only works if the banks are willing to lend to each other. If they think that the other banks might not repay them, then they’ll refuse to lend. If some of the banks decide to sit on their reserves and refuse to engage in the inter-bank lending market, it becomes a mathematical certainty that one of the other banks will struggle to make its payments. If this happens then the entire payment system could very quickly fall apart. This is what happened during the financial crisis. The only way to avoid this is for the central bank to pump in such a huge quantity of reserves that every single bank has more reserves than they need. This would mean that they no longer need to lend to each other. This is effectively what the Quantitative Easing scheme did, by pumping reserves into the banks and making it unnecessary for them to lend to each other.

Capital adequacy ratios or Basel accords?

So far we’ve seen that there is no liquidity reserve ratio, and that banks don’t really need to have central bank reserves in order to lend. But what about the capital adequacy ratios or Basel accords that everyone is talking about? Well the capital adequacy ratios relate to something quite different, but to understand why we need to look at the balance sheets again. Remember that the assets side of the balance sheets shows everything that the bank owns, including all its loans and mortgages, and the liabilities side shows everything that the bank owes to other people or companies. There’s a third part of the balance sheet, which is something called shareholder equity. Shareholder equity is very simply what’s left for the owners of the company when all the assets are sold and all the liabilities are paid off. To avoid going bankrupt, a bank needs to make sure that its assets are greater than its liabilities. When somebody defaults on a loan and stops making repayments, then the bank has to repossess the house and sell it off, usually at an auction. The bank will usually get less at the auction than the original value of the loan. That means that it loses money by repossessing the house. The mortgages on the balance sheet that was originally quarter of a million has turned into a house that will be sold for less than a quarter of a million. So when loans and mortgages go bad, it reduces the assets of the bank. Now if only a small percentage of the loans go bad,there’s no problem. The bank already expects at least one or two out of every 100 mortgages to go bad – that’s just part of the risk, and besides the interest payments from the loans that don’t go bad should cover these losses. But if everyone starts defaulting at the same time, then the banks’ assets can start shrinking rapidly. If the assets shrink so much that the bank’s assets are less than their liabilities, then the bank is insolvent and should be liquidated and shut down. The problem is that while the bank is being liquidated, most customers will be unable to access their money. This can cause big problems in the economy, and could even trigger a panic that leads to people trying to get money out of their other accounts and causing those banks to have difficulties too. So to try and prevent this from happening, there is something called the Basel accords, or capital adequacy ratios.

Capital adequacy ratios basically require the banks to keep a buffer that’s big enough to absorb any losses by the banks. The bigger the buffer, the more of a bank’s loans can go bad before it becomes insolvent. We won’t go into too much detail on how this scheme works here, but the key thing you need to know is this. When the bank makes a profit on its loans, then this profit increases the size of the capital buffer. If it’s capital buffer is bigger then it can afford to make more loans. So when the economy is improving, the ability of banks to lend will also increase. This leads to them lending more, making more profits, and further increasing their ability to lend. In other words, capital adequacy requirements don’t limit the ability of banks to create money when the economy is doing well. However, they do limit the ability of banks to create money when the economy is doing badly. And as we’ve seen, the money supply of the nation is dependent on the lending of banks, which means that the capital buffers make the instability in the money supply even worse. But the important thing is that capital adequacy reserves are not, and never have been, intended to limit how much money the banks can create, or how much reckless lending they can do. It’s simply about trying to ensure that when things do go wrong and loans start going bad, the banks have enough of a buffer to avoid going bankrupt.

So what does actually limit the ability of banks to increase the money supply?

We’ve seen that the type of reserve ratio that’s discussed in the textbooks has never even existed in the UK. We’ve seen that the liquidity ratios that did exist have been reduced and eventually abolished, and that even when they did exist, they only limited the speed that the money supply could increase, but put no limit on the total size that it could grow to. We’ve also seen that the Capital Adequacy Ratios and Basel accords are about preventing banks from going bust when loans go bad, rather than limiting their dangerous lending or limiting how much money they create through lending. And although the capital adequacy requirements can restrain lending after a banking crisis, it doesn’t do anything to restrain lending in a boom. We’ve also seen that there is no natural limit on how quickly the banks can create money. They know that even if they don’t have the actual central bank reserves to make payments, they’ll be able to borrow those reserves from other banks, or even the central bank.

All this comes together to imply that the only thing that truly limits the creation of money, is the willingness of banks to lend. And their willingness to lend depends on their confidence. In other words, the money supply of the nation depends on the mood swings of banks and the senior bankers that run them. This is surely an insane way to run an economy.