The Idea of Banking
The intention is to eventually reflect on the operations of commercial banks based on the principles of banking and ground them in history.
If you try and search for the history of banking, you may get lost in the history of money and its sporadic appearance and disappearance throughout history.[1] You may also become lost in institutions that were bank-like and had their own ups and downs with money. In this post, we will attempt to uncover the structure of banking based on its root idea, which remains mere conjecture.[2] The intention is to eventually reflect on the operations of commercial banks in current times and contrast them with their historical predecessors. Things have changed, yet the basic structure remains intact, and we still haven’t figured out the main role of banks!
Was it safety which forced us to bank?
The standard story is that banks started in Florence, Italy with the Medici family in 1300s. These were merchants who had gold coins from depositors which could be traders or commoners seeking one very simple service – security of their wealth. In fact, it is funny that security would be the key idea which takes people closer to Medici family. This would imply that the Medici’s were more capable of securing the wealth than individuals who themselves possessed wealth sizable enough to command external safety. Further, if wealth was unequal as was bound to be, then optimal size of operations would require that even average amount of wealth needed safe keeping.
However, one cannot certainly be sure of the need for safe keeping of such modest amount of wealth. We might not put a 100$ we earn in a lottery in the bank, we rather keep it with ourselves and spend it on our daily needs. Unless we live in a neighborhood with a lot of thieves which could make the trip to the Medici family worth it. However, it could also imply nonexistence of adequate deposits. In the bid to safeguard wealth, we might use the 100$ and buy essentials which would be required in the immediate future. This would lower the number of deposits further and make the scale economies harder to achieve for the Medici. Thus, there is some such thing as optimal theft to deposit ratio which could extend the argument that we keep deposits because of safekeeping of wealth.
This argument misses one important concern. The optimal deposits are an equilibrium outcome of the supply of deposits by people who want safekeeping and the demand for deposits by the Medici and their likes. What if it was the case that the Medici themselves wanted more and more deposits to attain a scale of operation commensurate with the overheads of maintaining such a strong empire. This would imply that they would need to attract the average depositor with a lure to maintaining deposits rather than keeping cash at hand with themselves. This would require them to be a part of a reasonably and relatively stable macroeconomic environment which maintains the value of money over the lifetime of deposits. If the Medici were in an environment with rapid inflation or fluctuating prices of necessity commodities, then depositors would not keep deposits but would rather consume a greater part of their incomes/wealth. Thus, stable macroeconomic environment was important driver for the origin of banking in the first place.
Story of Two Interest Rates
Another more obvious incentive is to attract depositors with an interest (i) on deposits. This would still require stable macroeconomic environment, but perhaps do more to make a case for deposits. Especially if you have un-used wealth for a period, then putting them into deposits makes sense. This would induce those who without an interest thought to be too costly to go to the Medici deposit facility for safe keeping. Now they have added interest in deposits which overcomes operational cost of maintaining deposits. Thus, they now have interest bearing deposits, which are liquid enough to use on demand and does not cost them to maintain. Thus, banks started accepting deposits.
All of this begs an important question – what are the Medici getting in all of this? We are asking the wrong question here. Various historical records point to the need for credit as the starting point for the demand for deposits. The story develops much earlier, or so can be conjectured. With improvements in agriculture, we had people specialize in other occupations. The specialization brought with it a need to trade with others. This is where money comes into picture. Once money was established, the specialization flourished further. However, with trading came the need to establish line of credit and the gap between purchase, sale, and payment settlement systems. This is where Medici enter the picture.[3]
The need of the hour was a line of credit. Like all trades, this trade entailed risk. We can think of this risk on a variety of fronts. What if the transport of the product was not secure enough and the buyer reneged on their promise of payment? There could be a million reasons which force the creditor to default on their line of credit. Additionally, the Medici also take a liquidity risk – the risk of not having enough money when the depositors ask for a transaction. In order to induce them to take this risk, they would charge a premium on the line of credit that they extend which makes it worth their while to engage in the process of intermediation in the first place.
Getting back to the argument of safekeeping as the starting point of deposit acceptance implies that the Medici were providing a service. This service could in turn could have commanded a fee, which should be paid by the safe keeper. They might even do so if the threat of theft is a credible one. However, to give interest on deposits is an invitation which invites deposits and makes this service fee negative. This then feeds into the idea of credit at a higher interest rate and makes Medici as the quintessential financial intermediary of the older times.
Now there is a matter of causality which is not yet clear, historically. Did the interest on deposits force the Medici to command a higher interest on loans or was it the demand for loans which pushed them to provide interest on deposits.[4] These led to the formation of Modern banks which accepted deposits, gave out loans and lived to tell the tale of profits as the difference between the two interest rates. This is very standard in modern banks too. They accept deposits, provide interest on deposits, give out loans and charge an interest on these loans which is higher than the interest on deposit they pay.
It is important here to understand that the Medici did not gain this power in vacuum and there are other parallel stories of origins of banks which differ entirely from ours. The alternate history is one in which the Medici started giving out these loans to the sovereign, who in turn ensured that banking institutions become entrenched in the law of the land and are legitimized in power. But we leave it here and return to our incentives-based story.
Can we trust the Banks?
In our story, the deposit-loan relationship is essentially an optimal behavior from the Medici to continue and thrive. If we place the deposits as liabilities and loans as assets in their balance sheets, we describe most modern banks which accept deposits. One difference being that the deposits are broken down into various types – but remain similar in spirit. For instance, you can deposit into an account where you earn little interest, but get charged no fee, can freely transact and can even use a debit card for your digital transactions. You can also deposit your money in a much more premium deposit account which has a debit card which can give you rewards for usage – i.e. now you have an interest income and an income in kind, which in some cases can even be encashed! Similarly, loans can be structured in a variety of ways.
This is where depositors start getting suspicious. Why is someone giving you more money than what you give them for safekeeping? This winds a lot of depositors and still does in my village. To them, this is baffling! The Medici could have gone about in a variety of ways to explain this interest on deposits. They might have estimated a value to the depositor’s postponement of consumption, which is equivalent to interest on deposits. They can also explain that the survival of a bank relies on attracting deposits and lending loans at a higher interest rate. The gaps between the two rates are (or is?) where the bank hopes to make profits.
It would only take any number of exogenous shocks to this operation for the assets to become non-performing assets (NPAs) and show up in the data as fall in the bank – the Medici family bank lasted roughly 100 years. The series of bank failures on account of non-payment by borrowers would create mistrust in the minds of depositors which a simple hike in deposit rate and lending rate cannot solve. In other words, interest hikes are not credible enough to induce depositors to continue deposit after a shock, which to them looks like the bank’s bad decision.
This brought with it the first injection of family or own capital into the banks liability to support the advent of NPAs, or so we conjecture. In early days of Banco di San Giorgio, its treasurer had to inject significant capital and adhere to impeccable standards of double entry bookkeeping for the safety of the bank’s operations. Giuseppe Felloni, the ardent professor from University of Genoa tell us:
“…He then had to lodge 16,000 lire (worth about €275,000 in today’s money) with the bank– it was raised to 40,000 lire in 1634 – and to provide the names of sponsors who would guarantee a further 90,000 lire for the duration of the treasurer’s six-month tenure. It seems that in those days you had to pay the bank in order to run it, rather than the other way around…”
Such injections of capital in the bank would have been cyclical to the ups and downs of the business. With every looming threat of a higher than desired NPAs, depositors would err towards withdrawal or hesitation, which could compromise the operations of the banks. Thus, banks would react by injecting more capital to alleviate the fears of the depositors.[5] However, the fear of NPAs are not the only reason for injecting capital into a bank. Own capital is also required to establish trust among depositors of a “reasonable risk”. The idea is that depositors would be vary of the bank’s investment practices especially in the face of NPAs. The only solace they would have then is that some of its own capital is put into the same investments as that of the depositors. Now the balance sheet looks even more familiar – deposits, loans, and bank capital. This also brings to us a measure of riskiness in the bank’s operations based on its capital adequacy or leverage.
How much capital is good reasonable enough to allude to the fears of our modern depositors? In modern parlance, the Basel III norms require Common Equity Tier 1 (CET1) capital requirement of 4.5%, minimum Tier 1 capital requirement of 6% etc. Where does this number come from? Perhaps those years of bank failures inform these adequacy norms. I still do not have very concrete answers to this question so will leave this issue to tackle in the future.
Fractional Reserve, Maturity Transformations, and the Limit Theorem
The last piece of the banking puzzle, it can be conjectured, comes from the mismatch of lending and withdrawal needs. While the Medici accepted deposits, gave interest on deposits, provided loans at a higher interest rate, they must have kept some money back to service the transaction needs of the depositors. Let us understand what happens if they hadn’t under the assumption that they are the only bankers around. When a depositor deposits 100 florins they are immediately lent out to the borrower who promises the most return, assuming complete information. He in turn deposits this loan amount in the Medici bank. Thus, the deposit to loans transaction is merely an accounting entry without the actual money changing hands – this makes Medici an important clearing house. This process goes on. The loans get deposited back to the Medici and they lend it out to the next in line. The initial 100 florins deposited can create an infinite amount of wealth – which is contentiously attributed to banks.
Who owns these 100 florins? The one who first deposited or the ones who were lent out these on successive loans? We do not really know who has a higher claim. What we know is that anyone can withdraw any proportion of their deposits anytime they want. This creates the other big problem in the banking system – the problem of bank runs. Bank runs can be caused by a variety of factors, including rumors of financial instability, a downturn in the economy, or a loss of confidence in the banking system. In this case, even a day-to-day withdrawal can cause initiate this lack of confidence purely because the bank is hyper-illiquid. This can lead to a crisis of liquidity – there are only 100 florins but the claimants to these are many. In fact, even a partial withdrawal by many enough to overrun existing liquidity in the bank would cause a crisis.
This simple lending cycle which creates infinite wealth can be broken if part of the deposits can be kept as reserves in anticipation of this day-to-day withdrawal. This is what makes modern banks adhere to fractional reserve principles in modern times. This completes the basic picture of a bank’s balance sheet – one with deposits, loans, bank capital and fractional reserves. The fractional reserves do something else too – they help converge the wealth that can be created in this deposit-lending-deposit cycle to a finite amount. The idea is that every time a part of the deposit is kept as reserve, a smaller part gets lent out, which generates further deposits, of which a part of reserve is kept out before lending even a smaller part out. You can immediately observe that with each lending cycle, the amount of the loan shrinks. In fact, we can apply simple limit theorems and calculate the exact wealth created from this process based on what proportion of reserves are kept aside as a rule.
You can also think of the banks as borrowers in the first place who are forced to run a business without it being a Ponzi scheme. Their business involves balancing maturity transformations. A bank may borrow money from depositors, who expect to be able to withdraw their money on demand and use those funds to make long-term loans to businesses. The bank earns a spread between the short-term interest rate it pays to depositors and the long-term interest rate it charges to borrowers, which is its main source of profit. So, the story of banking is not merely accepting deposits and giving out loans. The character of these transactions is engrained in different maturity structures which create risk as well as need to continuously attract a constant flow of depositors. This completes the basic idea of banking without going into issues such as term structure of interest rates, maturity mismatches and the crisis of banking. We will come back to these issues later.
[1] Did you know that paper money existed almost 1000 years back, but then miraculously disappeared as if people had forgotten this idea. The idea of banking and the use of money feed each other throughout history.
[2] “…my greatest wish is that these claims should lead to further research elsewhere. Whatever the findings might be – favourable to these theses or not – such research would serve only to further enrich and benefit our historical knowledge…” Felloni, Giuseppe. https://www.ft.com/content/6851f286-288d-11de-8dbf-00144feabdc0
[3] On the origin of specie, The Economist, 18 August 2012. https://www.economist.com/finance-and-economics/2012/08/18/on-the-origin-of-specie. This particular piece talk about the idea that trade brought with it a payments delay. An involved discussion is here Jonathan Stray http://jonathanstray.com/the-origin-of-banking who uses Hick’s analysis to draw upon the origins of banks. Both are very important reads.
[4] If I get a reference on this, I will update it in a latter post, but a lot of the discussion on early origins of banks point to the irrelevance of this argument. For instance - Goodhart, C. A. (1998). The two concepts of money: implications for the analysis of optimal currency areas. European Journal of Political Economy, 14(3), 407-432. https://doi.org/10.1016/S0176-2680(98)00015-9
[5] I haven’t got a clear idea of the proportion of bank failures in the early banking period, but one could assume that they would be plenty.