Bank deposits are not growing at a fast pace because people are investing in the stock market and buying shares directly as well as indirectly through mutual funds and insurance. Or so we have been told over and over again the past few months. But as I explained in the 28 August issue of the Easynomics newsletter, this is basically bunkum.
For a buyer to buy shares, someone has to sell. Hence, every time someone buys shares, someone sells, and money moves from one bank account to another, and thus stays within the banking system. There are other nuances to the whole issue. For a detailed argument I suggest you read the 28 August newsletter.
Now, when more individuals buy shares, the money may continue to stay within the banking system but there are other systemic level effects. In this newsletter we will look at these effects. Of course, before I get to the most important points, I have to look at other important points to make my case and say what I want to say.
So, dear reader, stay with me, because this might get slightly detailed and a little complicated before it gets simple. Here we go.
Banks raise three kinds of deposits—current, savings, and term deposits. Current and savings deposits can be withdrawn at any point of time. Term deposits come with maturity periods or a tenure. Fixed deposits and recurring deposits are the two kinds of term deposits. Of course, depositors can withdraw money from term deposits also at any point of time, but there is a certain penalty attached to it, which is a way of discouraging early withdrawals.
Look at the following chart which plots the proportion of current and savings accounts (CASA) deposits and term deposits as a proportion of overall bank deposits.
This chart tells us that the share of CASA deposits had peaked at 47.2% of the total deposits in 2021-22 (not taking non-resident rupee deposits, foreign currency deposits, and deposits from banks into account), and have fallen since then. Why did this happen?
It happened primarily because in the pandemic years people didn’t like the idea of locking away their money in term deposits like fixed deposits. They wanted to have the flexibility of withdrawing their money if a medical or a financial emergency came up.
One impact of this was that banks were able to raise a greater proportion of their deposits at very low interest rates. In the process, they made greater profits, given that no interest is paid on money deposited in a current account and savings accounts typically pay an interest of 2.7-4%.
But that insecurity has now lessened and hence, in 2023-24, the proportion of deposits in CASA fell to around 42.9%—similar to the level it was at in the years before the pandemic.
What this means is that a greater proportion is being saved in term deposits than in the pandemic years. In 2023-24, 57.1% of total deposits were in term deposits. That was a great improvement over 2021-22, when the proportion had fallen to a low of 52.8%. At 57.1%, the proportion is similar to what it was before the pandemic.
So, isn’t this good news, given that banks now have a greater proportion in term deposits?
On the face of it this suggests that more money is locked up for the longer-term. But that’s not really the case, given that term deposits can have a tenure of as low as seven days.
Take a look at the following chart. It plots the share of term deposits as per their tenure.
Before analysing this chart, let me make the disclaimer first. The data was originally published in a lot more detail, but I have bunched it up here to visually represent it in a simpler manner. I have used three brackets here—term deposits maturing in less than a year; those maturing in 1-3 years; and those maturing beyond 3 years.
In 2021-22, term deposits of less than one year made up for more than 31% of the term deposits. This was the highest in the 15-year period considered for the chart. By 2023-24, this had fallen to 20.4%. As mentioned earlier, in the pandemic years depositors did not like the idea of locking in their money for the longer term. The anomaly has since been corrected, hence, the fall.
In 2023-24, more than 66% of term deposits were in the 1-3-year tenure. This was the highest in the 15 years considered for the chart. In 2021-22, this had stood at 50.4%.
Further, in 2023-24, only 13.1% of term deposits had a maturity period of greater than 3 years. I checked data as far back as 1989-90, and this is the lowest that term deposits of a maturity greater than 3 years have been at between then and 2023-24. In fact, in 2015-16, they were at 25.8% of total deposits.
Let’s look at another chart. In the first chart we looked at the breakdown of overall deposits. In the second chart, we looked at term deposits. Now, let’s try combining these two charts and looking at the overall picture. To do that, I need to define a few terms. Let’s start with what I like to call short-term deposits.
Short-term deposits can be defined as the sum of deposits in CASA and term deposits of less than 1 year. Medium-term deposits can be defined as term deposits of a tenure of 1-3 years. Long-term deposits can be defined as term deposits of a tenure beyond 3 years.
And this is how things look.
Short-term deposits, which I have defined as the sum of CASA and term deposits of less than 1 year, formed 54.5% of the overall deposits in 2023-24. In that sense, they are back to the levels they were at in the pre-pandemic years. In 2021-22, they had stood at 63.7%.
The medium-term deposits, which I have defined as term deposits with tenure of 1-3 years, form 38% of the overall deposits, the highest in the 15-year period for which the data has been considered.
The long-term deposits, which I have defined as term-deposits with a tenure of more than 3 years, stood at 7.5%, the lowest in 15 years.
Why is this happening?
There can be multiple reasons for this.
First, the composition of who holds bank deposits has changed over the last few years. Data from the Reserve Bank of India show that individuals held 56.4% of term deposits as of March 2019 and 56.7% as of March 2021. As of March 2024, this had fallen to 51.7%.
Individuals typically tend to invest in deposits that have a longer tenure than others. As RBI governor Shaktikanta Das put it in a July 2024 speech: “Households and consumers who traditionally leaned on banks for parking or investing their savings are increasingly turning to capital markets and other financial intermediaries. While bank deposits continue to remain dominant as a percentage of financial assets owned by households, their share has been declining with households increasingly allocating their savings to mutual funds, insurance funds and pension funds. To be precise, households are increasingly turning to other avenues for deploying their savings instead of banks.”
This can be seen in a chart recently shared by Uday Kotak, the founder of Kotak-Mahindra Bank, on X.
As can be seen in the above chart, as of 2019, deposits (CASA + fixed deposits) formed around 51% of the household financial assets. This has since fallen to 42% in 2024, primarily because individuals are investing more money in stocks, directly, and indirectly through mutual funds, insurance and collective investment schemes.
As I had explained in the newsletter published on 28 August, this money that is used to buy shares again ends up with a bank as deposits, but the constitution of deposits changes. This is one reason why the deposits held by individuals in banks has come down.
A greater proportion of deposits is being held by what RBI calls “others” in its data. These “others” do not like to hold deposits in long-term tenures. A detailed reading of RBI data over the years clearly suggests that. The “others” possibly include corporate firms and other institutions.
Another reason for this change can be the significantly higher profits being earned by companies in recent years. The aggregate net profit of over 5,000 listed firms in 2018-19, before the pandemic broke out, stood at ₹3.4 trillion. In 2023-24, this had jumped to ₹13.1 trillion. These higher profits need to be held as deposits with banks and thus companies are holding more deposits with banks, and those deposits are of shorter tenures.
Third, many companies have sold shares through initial public offerings (IPOs). Others have come up with follow-on public offerings (FPOs). Promoters of already listed firms have also sold shares to the public. This also changes the constitution of deposits, making them more corporate, than individual- and household-driven.
Fourth, looking specifically at the interest rate data of the State Bank of India, it can be said that interest rates being offered by banks in the 1-3 year tenure are higher or similar to the interest rates on offer for periods of greater than 3 years. This wasn’t the case earlier. This basically implies that banks are incentivising prospective depositors to save money in fixed deposits with tenures of 1-3 years than in those with longer tenures.
This basically means that the average tenure for deposits held by banks has dropped over the years. What’s the impact of this? The dynamic at the heart of banking is maturity transformation, wherein deposits taken on by banks for the short-term help balance loans given for significantly longer terms. This happens primarily because not all depositors land up at the bank at the same time on any given day wanting their money back.
To explain this in simple English, let’s consider this: banks give out home loans. The typical tenure of the repayment of a home loan when it’s given out is 15-20 years. But no one deposits money in a fixed deposit for that kind of period. So banks borrow short and lend long. This is the maturity transformation that happens and keeps banking going.
The trouble is that maturity transformation has its limits and its risks. Let’s try and understand this through a very simple example. Let’s say a bank gives a loan that has to be repaid over 10 years. This loan is balanced by deposits that are supposed to mature in 2 years. At the end of 2 years the bank will have to repay these deposits.
How will the bank repay the deposits in their entirety after 2 years, given that the loan will be repaid only over 10 years? One way out is to convince the depositor to extend the deposit for a further period. Another way is to hope that someone else will come and deposit money and that money can then be used to pay off the earlier depositor.
The point being that while the tenure of the loans and the tenure of the deposits of a bank cannot be totally matched because then banking as a business would never exist, nonetheless, there has to be some sense of balance. The larger the difference between the tenure of loans and the tenure of deposits, the greater the asset-liability mismatch.
There is one more point that needs to be made. CASA deposits tend to be stable but at the same time the risk of such deposits being withdrawn quickly always remains. With a huge chunk of deposits now in the 1-3-year tenure, Indian banks are facing an asset-liability mismatch.
As far as the tenure of loans is concerned, the latest data available in the public domain is as of end March 2023. At that point, close to 36% of loans and advances given by banks had a tenure of greater than 3 years. In comparison, deposits with a tenure of more than 3 years stood at 8.7%. In March 2019, loans and advances with a tenure of greater than 3 years were at around 33%, whereas deposits of a similar tenure stood at 11.5%.
If we look at loans and advances of greater than 5 years and deposits of greater than 5 years, a graver situation emerges.
In March 2023, loans and advances of greater than 5 years stood at 22.5% of overall loans and advances. In comparison, deposits of greater than 5 years stood at just 4.6% of overall deposits. In March 2019, loans and advances of greater than 5 years had stood at 20.4% of overall loans and advances. In comparison, the deposits of greater than 5 years had stood at 7.7% of overall deposits.
Clearly, the asset-liability mismatch has become worse.
I feel that this kind of mismatch would have only continued in the year to March 2024. Of course, this is the overall average, with some banks facing a greater mismatch than others.
A possible reason for this could be that the business model of banks has gradually moved towards issuing more and more retail loans. Incrementally, a good chunk of these retail loans has a repayment tenure of greater than 3 years.
This is why there has been so much talk around banking deposits in the recent past though no one is really talking directly about the worsening asset-liability mismatch, and those in the business of talking about such things are largely pussyfooting around the issue.
RBI governor Das talked about the system being exposed to structural liquidity issues in his speech referred to earlier. RBI’s State of the Economy report for August talked about banks issuing bonds at interest rates higher than those on offer on fixed deposits to manage their asset-liability mismatches.
Along similar lines, Das said at the time of the last monetary policy meeting in August: “Banks are taking greater recourse to short-term non-retail deposits and other instruments of liability to meet the incremental credit demand. This, as I emphasised elsewhere, may potentially expose the banking system to structural liquidity issues.”
That leaves us with the question: How did we get here? The main reason for this situation is the change in the constitution of household financial savings and the fact that the share of bank deposits held by individuals has come down.
How did this happen? RBI actively cut interest rates in the aftermath of the pandemic. The idea was to help the government, companies and individuals to borrow money at lower rates, and thus encourage them to spend money and pump-up economic growth.
This is a standard operating procedure as far as central banks are concerned. The trouble is that there is a flip side to it. Interest rates are the price of money. And any price has two sides to it: Someone who pays the price and someone who earns it. And those earning an interest lose when interest rates fall.
When this happens, many of these people try looking for other avenues of investment where they can hope to earn a higher rate of return. This has been proven over and over again. It was proven true during the Japanese stock market and real estate bubble of the 1980s, the dotcom bubble of the 1990s, and the sub-prime real estate bubble of the 2000s.
Whenever central banks try to manage interest rates in any economy extensively, it forces people to look beyond bank deposits and consider more risky modes of investment. This is what happened in India post March 2020.
People went looking for higher returns and ended up investing more and more in stocks, directly or indirectly. Of course, RBI has raised interest rates since but with people addicted to huge returns from the stock market, it hasn’t made much of a difference as far as investing habits go.
Indeed, monetary policy around the world, where central banks cut interest rates to pump economic growth, seems to be analysed and thought about largely based on the first-order effect of low interest rates—encouraging borrowing and spending and thus helping revive economic growth. There is almost no serious talk about all the bubbles that monetary policies fuel.
The current IPO mania in India wouldn’t have happened if interest rates hadn’t been cut to as low a level as they were. Of course, other reasons, like cheap smartphones, availability of very cheap internet bandwidth, and new apps with easy-to-use interfaces that let you buy stocks and mutual funds at the click of a button, have also helped fuel the bubble.
There is another possible angle here. In September 2019, the government cut the corporate income tax rate. This—along with RBI cutting interest rates and the increasing formalization of the economy following the implementation of the goods and services tax—helped corporate profits soar to never-before-seen levels. This led to companies depositing more money in banks, effectively changing the constitution of deposits with banks. That, in turn, led to a greater asset-liability mismatch.
A lot of this analysis is basically the benefit of hindsight. I can’t claim that I had all this clarity in thinking a few years back. I didn’t. But one thing that I have had clarity about for quite a while now is that economics is all about understanding second order and nth order effects. But professionals who decide on economic policy seem to be lagging on this front, despite history telling them otherwise.
At a personal level there is a lesson for all of us. As Olivier Burkeman writes in Meditations for Mortals—Four Weeks to Embrace Limitations and Make Time for What Counts: “The conservative American economist Thomas Sowell summed things up with a bleakness I appreciate, insisting that there are no solutions, only trade-offs. The only two questions, at any moment of choice in life, is what the price is, and whether or not it’s worth paying.”
I think this is an important lesson that us Indians—brought up on answering questions in exams with only one right answer—need to understand. Most complex decisions in life are about making trade-offs. They don’t have right or wrong answers.
So, what can be done to address this issue?
First, banks can hope and pray for the stock market mania to subside, which would make bank deposits a more attractive investment proposition for individuals.
Second, banks need to offer higher interest rates on deposits of longer tenures. But that will impact their net interest margin, which is why they seem to be happy with a higher asset-liability mismatch. As I said: trade-offs.
Third, the government needs to make bank deposits a more attractive proposition for households. Currently, interest on deposits is taxed at the marginal rate of income tax, whereas long-term capital gains tax on stocks and equity mutual funds is taxed 12.5%. And the long-term is a period of holding just greater than 1 year. The insurance policy on maturity is tax free.
What needs to be realised is that it is very easy to invest in stocks now in comparison to how things were even 5 years earlier, and that different kinds of investing need to have a level-playing field. Otherwise, this asset-liability mismatch will persist.
