Most of the money in India is not printed by anyone. On 15 October 2026, India's broad money (M3) stood at ₹333.00 lakh crore, but currency with the public was only ₹42.56 lakh crore of it, about 13%. The other 87% was bank deposits, and most of those came into existence when a bank made a loan and credited the borrower's account. The RBI creates the base of the system, reserve money (M0): notes and coins and banks' balances with the RBI, ₹53.25 lakh crore on the same date. Commercial banks create the rest. This part explains how that happens, why the textbook "money multiplier" gets the direction of causation backwards, what actually limits money creation, how India measures money (M0, M1, M2, M3, M4 and the NM aggregates), and what the 2025–26 gap between credit growth and deposit growth does and does not mean.
The RBI's own balance sheet, which is where reserve money comes from, is covered in Part 13, currency in Part 12, CRR and SLR in Part 5, and the LAF corridor in Part 11. Capital and liquidity rules get their own parts in this series: Part 27 and Part 28. This part links to them rather than repeating them.
In the news
- 23 October 2026: broad money growing at 16%. The RBI's Weekly Statistical Supplement put M3 at ₹3,33,00,300 crore on 15 October 2026, up 16.0% from a year earlier (9.2% a year before that). Reserve money grew 10.3% to ₹53,24,881 crore. Why it matters for the exam: M3 ÷ M0 gives a money multiplier of about 6.25, and M3 growing faster than M0 is the multiplier rising.
- 23 October 2026: credit still ahead of deposits. In the same release, scheduled commercial banks' credit was ₹227.43 lakh crore, up 18.2% year on year, and aggregate deposits ₹278.76 lakh crore, up 16.7%. The credit-deposit ratio was 81.59%, and the incremental ratio for 2026-27 so far was 83.96%. Why it matters for the exam: these are the headline banking numbers for any current-affairs question on credit growth.
- 12 October 2026: banks turn to certificates of deposit. CareEdge Ratings reported credit growth of 19.4% against deposit growth of 17.8% for the fortnight ended 30 September, a gap of about 155 bp. Banks' CD issuance nearly doubled to ₹74.6 thousand crore in the fortnight, which CareEdge read as "greater reliance on wholesale funding". Why it matters for the exam: when deposits lag credit, banks fill the gap with costlier wholesale money such as CDs.
- 9 October 2026: tighter daily CRR. The RBI raised the share of the CRR that banks must keep on every day of the fortnight from 90% to 99%, from 16 October 2026. The CRR itself stayed at 3%. Why it matters for the exam: reserves are a constraint on each bank's day-to-day liquidity, even if they do not cap total lending.
- 7 October 2026: a hike citing money and credit growth. The MPC raised the repo rate by 25 bp to 5.50%. The Governor's statement said that while there was limited evidence of demand pressures, "risks in view of strong growth in monetary and credit aggregates exist", and that bank credit had grown 18.1% by 15 September, against 10.4% a year earlier. Why it matters for the exam: the price of money (the repo rate) is the RBI's main lever on how much money banks create.
- September 2026: rupees in through swaps, out through OMOs. By 21 September banks had reported US$143.6 billion of inflows under the RBI's special USD/INR swap facility, US$133.0 billion of it as FCNR(B) deposits. The RBI then sold ₹1,00,000 crore of government securities in three OMO auctions between 17 and 28 September to drain part of the resulting surplus. Why it matters for the exam: money is also created when the banking system buys foreign exchange, and destroyed when the RBI sells bonds.
- 15 December 2025: new dates for money data. Under the Banking Laws (Amendment) Act, 2025, the reporting fortnight now runs from the 1st to the 15th and from the 16th to month-end. The RBI now publishes money supply data for the 15th and the last day of each month. Why it matters for the exam: that is why M3 is quoted "as on 15 October", not on alternate Fridays.
- 29 November 2025: CRR at 3%. The last 25 bp tranche of the 100 bp CRR cut took effect. Banks' deposits with the RBI fell from ₹9.91 lakh crore at end-March 2025 to ₹8.10 lakh crore at end-March 2026. Why it matters for the exam: a lower CRR shrinks reserve money and so raises the measured M3/M0 ratio.
What counts as money
Three kinds of money
In a modern economy there are three kinds of money, and only one of them is created by commercial banks:
- Currency: notes issued by the RBI and coins issued by the Government. It is a claim on the RBI (or the Government), and anyone can hold it.
- Reserves: the balances banks keep in their current accounts with the RBI. Only banks and a few other institutions can hold them. Banks use them to pay each other and to meet the CRR.
- Bank deposits: what you hold in a savings, current or fixed deposit account. They are a claim on a commercial bank, not on the RBI.
Currency and reserves are central bank money; deposits are commercial bank money. A UPI or NEFT payment moves commercial bank money, and the banks settle with each other in reserves.
Reserve money (M0)
The RBI defines reserve money as:
M0 = currency in circulation + bankers' deposits with the RBI + 'other' deposits with the RBI
It is also called base money or high-powered money, because it is the base on which deposits are built. On 15 October 2026:
| Component of M0 | ₹ lakh crore | Share |
|---|---|---|
| Currency in circulation | 43.47 | 81.6% |
| Bankers' deposits with the RBI | 8.50 | 16.0% |
| 'Other' deposits with the RBI | 1.28 | 2.4% |
| Reserve money (M0) | 53.25 | 100% |
"Other" deposits are balances held with the RBI by entities other than banks and the Government, such as foreign central banks and some public institutions. Reserve money is almost entirely the RBI's own liabilities, so it changes only when the RBI's balance sheet changes: when it buys foreign exchange or bonds, lends to banks, or absorbs money through the SDF and reverse repos (Part 13).
Broad money (M3)
Broad money, M3, is the money the public holds:
M3 = currency with the public + demand deposits with banks + time deposits with banks + 'other' deposits with the RBI
Currency with the public is currency in circulation minus the cash that banks hold in their own vaults. Bank cash is not counted, because a bank's own cash is not money held by the public. On 15 October 2026:
| Component of M3 | ₹ lakh crore | Share |
|---|---|---|
| Currency with the public | 42.56 | 12.8% |
| Demand deposits with banks | 34.54 | 10.4% |
| Time deposits with banks | 254.63 | 76.5% |
| 'Other' deposits with the RBI | 1.28 | 0.4% |
| Broad money (M3) | 333.00 | 100% |
Two things stand out. First, deposits are 87% of M3. The Bank of England made the same point for the UK in 2014, where deposits were 97% of broad money; India uses far more cash, but the principle is the same. Second, bankers' deposits with the RBI are in M0 but not in M3, while time deposits are in M3 but not in M0. The two measures overlap only in currency and 'other' deposits.
India's money measures: M1 to M4
The RBI's Handbook of Statistics lists these measures and definitions:
| Measure | Definition | Nature |
|---|---|---|
| M0 (reserve money) | Currency in circulation + bankers' deposits with the RBI + 'other' deposits with the RBI | Central bank money |
| M1 (narrow money) | Currency with the public + demand deposits with the banking system + 'other' deposits with the RBI | Money used for payments |
| M2 | M1 + savings deposits of post office savings banks | Narrow money plus post office savings |
| M3 (broad money) | M1 + time deposits with the banking system | The headline measure |
| M4 | M3 + all deposits with post office savings banks (excluding National Savings Certificates) | The broadest of the old measures |
Liquidity falls from M1 to M4. On 15 October 2026, M1 was about ₹78.37 lakh crore, only 23.5% of M3: most Indian money sits in fixed deposits.
The "banking system" here means commercial banks and co-operative banks. M3 is the measure the RBI publishes every fortnight in its Weekly Statistical Supplement and the one quoted in the news. Broad money was the intermediate target of policy under monetary targeting from 1985 (the Chakravarty Committee) until the multiple indicators approach of April 1998 (Part 3).
The NM aggregates and liquidity aggregates
The Working Group on Money Supply: Analytics and Methodology of Compilation (chaired by Dr Y. V. Reddy, report of June 1998) proposed a new set of measures. The RBI uses the prefix "N" to tell them apart from the old ones:
| Measure | Definition |
|---|---|
| NM1 | Currency with the public + demand deposits with the banking system + 'other' deposits with the RBI (same as M1) |
| NM2 | NM1 + short-term time deposits of residents (contractual maturity up to and including one year) |
| NM3 | NM2 + long-term time deposits of residents + call/term funding from financial institutions |
| L1 | NM3 + all deposits with post office savings banks (excluding National Savings Certificates) |
| L2 | L1 + term deposits with term-lending and refinancing institutions (FIs) + term borrowing by FIs + certificates of deposit issued by FIs |
| L3 | L2 + public deposits of non-banking financial companies |
The key differences: NM2 splits time deposits by maturity (one year), the NM measures count residents' deposits, and NM3 adds call and term funding from FIs. The L measures are liquidity aggregates that reach beyond banks to post offices, FIs and NBFCs.
The textbook story: the money multiplier
Deposit expansion, step by step
The older textbook story runs like this. Suppose the reserve ratio is 10%, people hold no cash, and the RBI adds ₹100 of reserves:
- Bank A receives a ₹100 deposit. It keeps ₹10 as reserves and lends ₹90.
- The ₹90 is spent and deposited in Bank B, which keeps ₹9 and lends ₹81.
- Bank C receives ₹81, keeps ₹8.10 and lends ₹72.90, and so on.
Total deposits = 100 + 90 + 81 + 72.9 + … = 100 ÷ 0.10 = ₹1,000. Reserves of ₹100 have been "multiplied up" ten times. In this story, deposits come first, then loans, and the central bank controls the money supply by controlling reserves.
The formula
With a currency-deposit ratio c (cash the public holds per rupee of deposits) and a reserve-deposit ratio r (reserves banks hold per rupee of deposits), the multiplier is:
m = M3 ÷ M0 = (1 + c) ÷ (c + r)
Worked example with India's actual numbers (15 October 2026). Ignore the small 'other' deposits for now.
- Deposits D = demand + time deposits = 34.54 + 254.63 = ₹289.17 lakh crore
- Currency with the public C = ₹42.56 lakh crore, so c = 42.56 ÷ 289.17 = 0.147
- Bank reserves R = bank cash (43.47 − 42.56 = 0.91) + bankers' deposits with the RBI (8.50) = ₹9.41 lakh crore, so r = 9.41 ÷ 289.17 = 0.033
- m = (1 + 0.147) ÷ (0.147 + 0.033) = 1.147 ÷ 0.180 = 6.4
Adding back the 'other' deposits to both M3 and M0 gives the measured ratio: 333.00 ÷ 53.25 = 6.25.
The simple version, m = 1 ÷ r, would give 1 ÷ 0.03 = 33 at a 3% CRR. The real figure is much smaller because Indians hold a lot of cash: every rupee that leaves the banks as currency stops the chain. The CRR matters far less than the currency ratio: c is about four and a half times r.
How money is actually created: loans create deposits
The Bank of England's 2014 paper
In March 2014 the Bank of England's Quarterly Bulletin (2014 Q1) carried "Money creation in the modern economy" by Michael McLeay, Amar Radia and Ryland Thomas. It is now the standard short statement of the modern, "endogenous money" view. Its core claims, in its own words:
- "Whenever a bank makes a loan, it simultaneously creates a matching deposit in the borrower's bank account, thereby creating new money."
- "Rather than banks receiving deposits when households save and then lending them out, bank lending creates deposits."
- "In normal times, the central bank does not fix the amount of money in circulation, nor is central bank money 'multiplied up' into more loans and deposits."
- Banks "cannot do so freely without limit": profitability, prudential regulation and the behaviour of borrowers all constrain them, and "monetary policy acts as the ultimate limit on money creation".
The paper concludes that "the theory of the money multiplier operates in the reverse way to that normally described". More lending and more spending lead banks and the public to demand more reserves and currency, and the central bank supplies them at its policy rate. Causation runs from loans to deposits to reserves, not from reserves to loans.
Worked example 1: a loan creates a deposit
Asha, a small manufacturer in Pune, borrows ₹10 lakh from SBI to buy a machine. SBI approves the loan and credits her current account. Nothing is taken from any other depositor.
| SBI: Assets | SBI: Liabilities |
|---|---|
| Loan to Asha +₹10 lakh | Asha's deposit +₹10 lakh |
| Asha: Assets | Asha: Liabilities |
|---|---|
| Deposit at SBI +₹10 lakh | Loan from SBI +₹10 lakh |
Both balance sheets grow by ₹10 lakh at the same moment. M3 rises by ₹10 lakh, because Asha's deposit is money and nobody else's deposit fell. M0 does not change: no reserves or currency were needed to make the loan.
Worked example 2: a payment between banks
Asha now pays ₹10 lakh to the machine supplier, who banks with HDFC Bank, through RTGS.
| Bank | Assets | Liabilities |
|---|---|---|
| SBI | Reserves with the RBI −₹10 lakh | Asha's deposit −₹10 lakh |
| HDFC Bank | Reserves with the RBI +₹10 lakh | Supplier's deposit +₹10 lakh |
| RBI | No change | SBI's current account −₹10 lakh; HDFC Bank's +₹10 lakh |
M3 is unchanged: the deposit has moved from one bank to another. M0 is unchanged: reserves have moved between banks' accounts at the RBI. But SBI has lost reserves. If it is short, it must borrow them back, in the call money market, through a repo with the RBI or, in an emergency, at the MSF. This is why an individual bank cares about deposits even though the system as a whole creates them by lending: a bank that lends faster than others loses deposits and reserves to them.
If both Asha and the supplier banked with SBI, the payment would just move a deposit within SBI, and SBI would lose no reserves at all.
Worked example 3: cash and repayment
- The supplier withdraws ₹1 lakh in cash. HDFC Bank's deposits fall ₹1 lakh and its vault cash falls ₹1 lakh. M3 is unchanged (a deposit has become currency with the public). But the bank must top up its vault by drawing on its RBI balance, so its reserves fall. This is the currency drain that makes India's multiplier small.
- Asha repays ₹2 lakh of her loan. SBI's loans fall ₹2 lakh and Asha's deposit falls ₹2 lakh. M3 falls by ₹2 lakh. Repaying a bank loan destroys money, just as making one creates it.
Other ways banks create money
Lending is the main channel, but not the only one. Any time the banking system buys an asset from the public and pays by crediting a deposit, money is created.
- Buying securities. When a bank buys government bonds from a non-bank investor, it credits the seller's deposit. M3 rises. (When a bank buys bonds from another bank or from the RBI, no deposit is created.)
- Buying foreign exchange. When an exporter or NRI converts dollars into rupees, the bank credits a rupee deposit. If the bank sells the dollars on to the RBI, the RBI credits the bank's reserves, so M0 rises too. The 2026 FCNR(B) swap facility worked this way on a huge scale: banks handed dollars to the RBI and received rupees. In the year to 15 October 2026, the RBI's net foreign exchange assets rose ₹6.34 lakh crore, more than the whole ₹4.98 lakh crore rise in reserve money.
- Government spending. When the Government pays a contractor out of its account with the RBI, the contractor's bank deposit and the bank's reserves both rise. Tax payments do the opposite: deposits and reserves move into the Government's account at the RBI, which is outside M3.
Where reserves come from
If banks do not need reserves before they lend, where do the reserves for the CRR come from? From the RBI, at a price. The RBI steers the overnight call rate inside the LAF corridor: banks short of reserves can borrow at the repo rate in the RBI's auctions, or at the MSF rate of 5.75%, and banks with spare reserves can park them at the SDF rate of 5.25% (Part 11). Because the RBI targets an interest rate, it supplies whatever quantity of reserves the system needs at that rate. That is the practical reason the multiplier story fails: the RBI does not ration reserves; it prices them.
In October 2026 the system needed more reserves after the CRR daily minimum rose to 99%, and the RBI injected them through variable rate repo auctions from 19 October. A year of record OMO purchases (₹7.39 lakh crore in 2025-26) and the 2026 swap inflows show the same thing from the other side: the RBI adds or drains reserves to keep the call rate near the repo rate, not to hit a quantity of money.
Is the multiplier useless?
The multiplier view is not useless. It is an accounting identity: M3 is always M0 × (M3/M0). It also still describes what happens when the central bank changes c or r sharply, as in a cash withdrawal or a big CRR change. What it gets wrong is the direction of causation in normal times.
What limits money creation
If banks can create deposits by lending, why do they not lend without end? Because a loan has to be profitable, safe and funded, and the borrower has to want it.
Demand and price
A loan needs a borrower who will pay the interest. When the RBI raises the repo rate, lending rates rise and fewer projects pass the test. On 7 October 2026 the MPC raised the repo rate to 5.50% partly because of "strong growth in monetary and credit aggregates". Monetary policy works on money creation mainly by changing the price of credit, as described in Part 9.
Capital
Every loan uses capital. Under Basel III as applied in India, a bank needs total capital of at least 11.5% of risk-weighted assets (a 9% CRAR plus a 2.5% capital conservation buffer).
Worked example. A bank lends ₹100 crore to a company with a 100% risk weight. Risk-weighted assets rise by ₹100 crore, so the bank needs at least 11.5% × ₹100 crore = ₹11.5 crore of capital against it. A bank with ₹1,000 crore of spare capital above the minimum could support at most ₹1,000 ÷ 0.115 ≈ ₹8,696 crore of such loans before it must raise more capital or slow down. Capital, not reserves, is usually the binding constraint (Part 27).
Liquidity and funding
As Worked example 2 showed, a bank that lends money loses deposits and reserves when borrowers spend it. It must replace them by attracting deposits, issuing certificates of deposit or borrowing, and each source has a cost. It must also keep a liquidity coverage ratio of at least 100%: enough high-quality liquid assets to survive 30 days of stressed outflows (Part 28).
CRR and SLR
The reserve requirements still bite on each new deposit (Part 5).
Worked example. A bank's lending adds ₹100 crore of deposits to its books. At a CRR of 3% and an SLR of 18% of NDTL, it must eventually hold:
- CRR: 3% × ₹100 crore = ₹3 crore in its RBI current account, earning nothing.
- SLR: 18% × ₹100 crore = ₹18 crore in government securities, cash, gold or SDF balances.
- Total: ₹21 crore of the ₹100 crore of new deposits.
The CRR has to be met in reserves, which the bank can only get from the RBI or from other banks. The SLR ties up part of its balance sheet in low-yielding assets. Both raise the cost of lending, but neither sets a hard limit on how much the system as a whole can lend, because the RBI supplies reserves at its policy rate.
Borrowers and money holders
People who receive new money can destroy it, by repaying loans, or move it out of M3, by withdrawing cash, paying taxes into the Government's RBI account, or buying bank shares and bonds (bank capital and bonds are non-monetary liabilities). These leakages are why deposit growth can lag credit growth, as the next sections show.
India's money in 2026
M3 and M0 on 15 October 2026
| Outstanding (₹ lakh crore) | Change in a year (₹ lakh crore) | Growth | |
|---|---|---|---|
| Reserve money (M0) | 53.25 | +4.98 | 10.3% |
| Broad money (M3) | 333.00 | +45.86 | 16.0% |
| M3 ÷ M0 | 6.25 |
A year earlier, M3 growth was 9.2% and reserve money growth 3.5%.
Where the new money came from
The RBI also publishes M3 from the sources side, the banking system's assets:
M3 = net bank credit to Government + bank credit to the commercial sector + net foreign exchange assets of the banking sector + Government's currency liabilities to the public − banking sector's net non-monetary liabilities
In the year to 15 October 2026:
| Source of M3 | Outstanding (₹ lakh crore) | Change in a year (₹ lakh crore) |
|---|---|---|
| Bank credit to the commercial sector | 236.03 | +35.56 |
| Net foreign exchange assets of the banking sector | 75.66 | +9.77 |
| Net bank credit to Government | 94.54 | +7.61 |
| Government's currency liabilities to the public | 0.43 | +0.04 |
| Less: net non-monetary liabilities | 73.66 | +7.11 |
| M3 | 333.00 | +45.86 |
Bank credit to the commercial sector accounted for ₹35.56 lakh crore of the ₹45.86 lakh crore increase, or 78%. That is "loans create deposits" in India's own data. Foreign inflows (largely the FCNR(B) swap) added another ₹9.77 lakh crore. The rise in non-monetary liabilities (bank capital, reserves and similar items) took ₹7.11 lakh crore out.
The multiplier over time
| End-March | M0 (₹ lakh crore) | M3 (₹ lakh crore) | M3 ÷ M0 | M0 growth | M3 growth |
|---|---|---|---|---|---|
| 2016 | 21.81 | 116.18 | 5.33 | 13.1% | 10.1% |
| 2017 | 19.00 | 127.92 | 6.73 | −12.9% | 10.1% |
| 2018 | 24.19 | 139.63 | 5.77 | 27.3% | 9.2% |
| 2019 | 27.70 | 154.32 | 5.57 | 14.5% | 10.5% |
| 2020 | 30.30 | 168.00 | 5.55 | 9.4% | 8.9% |
| 2021 | 36.00 | 188.45 | 5.23 | 18.8% | 12.2% |
| 2022 | 40.69 | 204.94 | 5.04 | 13.0% | 8.8% |
| 2023 | 43.87 | 223.44 | 5.09 | 7.8% | 9.0% |
| 2024 | 46.31 | 249.40 | 5.38 | 5.6% | 11.6%* |
| 2025 | 48.29 | 272.87 | 5.65 | 4.3% | 9.4% |
| 2026 | 51.08 | 314.53 | 6.16 | 5.8% | 15.3% |
| 15 Oct 2026 | 53.25 | 333.00 | 6.25 | 10.3% (y-o-y) | 16.0% (y-o-y) |
*Includes the merger of HDFC Ltd (a non-bank) with HDFC Bank on 1 July 2023. Excluding it, M3 growth in 2023-24 was 11.1%. Growth rates are computed from the RBI's end-March levels.
Why the multiplier is not fixed
The ratio moved between 5.0 and 6.7 in a decade. Each move has a clear cause, and few of them are about banks lending more or less:
- Demonetisation (2016-17). Old ₹500 and ₹1,000 notes were deposited into banks, and currency was slow to be replaced. By end-March 2017 reserve money had fallen 12.9% while M3 still grew 10.1%, so the ratio jumped to 6.73. Remonetisation in 2017-18 (M0 +27.3%) pulled it back to 5.77. A fall in c raises m.
- The pandemic (2020-21). The CRR was cut to 3% in March 2020, which on its own should have raised the multiplier. Instead it fell to 5.23, because people hoarded cash (c rose) and the RBI flooded the system with reserves. Reserve money grew 18.8%, M3 only 12.2%.
- CRR cuts and fast credit (2025-26). The CRR fell from 4% to 3% between September and November 2025, and banks' deposits with the RBI fell 18.3% over the year. At the same time, credit and deposits grew quickly. Both pushed the ratio up to 6.16, and to 6.25 by October 2026.
In each case the multiplier was the result of choices by the public, banks and the RBI, not a fixed number that determined them. That is why the RBI does not forecast M3 by multiplying M0.
Credit growth vs deposit growth: the 2025–26 debate
What the numbers show
| Date | Credit growth | Deposit growth | Gap (pp) | Credit-deposit ratio |
|---|---|---|---|---|
| Mar 2025 | 11.0% | 10.3% | 0.7 | 80.8% |
| Jun 2025 | 9.5% | 10.1% | −0.6 | 78.9% |
| Sep 2025 | 10.4% | 9.5% | 0.9 | 80.3% |
| Dec 2025 | 14.5% | 12.7% | 1.8 | 81.7% |
| Mar 2026 | 16.1% | 13.5% | 2.6 | 81.4% |
| Jun 2026 | 18.6% | 13.3% | 5.3 | 82.6% |
| 30 Sep 2026 | 19.4% | 17.8% | 1.6 | 81.0% |
| 15 Oct 2026 | 18.2% | 16.7% | 1.5 | 81.6% |
Sources: CareEdge Ratings from RBI data (to 30 September 2026); RBI Weekly Statistical Supplement (15 October 2026). The credit-deposit ratio peaked at 82.7% in May 2026.
Credit growth rose from about 10% in mid-2025 to 18–19% in 2026, across agriculture, industry, services and personal loans. Deposits followed with a lag, then jumped in July–August 2026 with the FCNR(B) inflows.
If loans create deposits, how can credit outrun deposits?
At the level of the whole system, every new loan does create a deposit. But deposits then leak:
- Into cash. Currency with the public rose ₹5.27 lakh crore (14.1%) in the year to 15 October 2026. Each rupee withdrawn is a rupee of deposits gone.
- Into the Government's account. Tax payments move deposits into the Government's balance with the RBI, outside M3. CareEdge noted that net RBI credit to the Government fell ₹2.62 lakh crore in the second half of September 2026 alone, as mid-September tax payments built up Government cash.
- Into bank capital and bonds. When savers buy bank shares, bonds or other non-monetary liabilities, deposits are converted into something that is not money. These rose ₹7.11 lakh crore in the year.
At the same time, deposits can rise without bank credit: through foreign inflows and bank purchases of government bonds. So the credit-deposit gap measures leakages and other flows, not "banks running out of savings to lend".
Why banks still chase deposits
For one bank, the loans-create-deposits logic does not save it from funding pressure. Its new borrowers usually pay people at other banks, so it loses deposits and reserves, as in Worked example 2. To lend faster than its rivals it must win deposits from them, or pay for wholesale funds:
- Certificates of deposit: issuance nearly doubled to ₹74.6 thousand crore in the second half of September 2026.
- Bulk deposits and market borrowing, which cost more and run off faster, so they count against the LCR.
Rising repo and CD rates mean this funding will cost more. CareEdge expects bank credit growth of about 15–16% in 2026-27 as the base effect and the rate hike take hold.
Worked example: the incremental credit-deposit ratio
The incremental credit-deposit ratio compares new credit with new deposits over a period. For 2026-27 up to 15 October 2026:
- New credit = ₹13.83 lakh crore
- New deposits = ₹16.47 lakh crore
- Incremental CD ratio = 13.83 ÷ 16.47 = 84.0% (the RBI reports 83.96%)
So for every ₹100 of deposits added this year, banks added ₹84 of loans. Over the same period, with 18% SLR and 3% CRR, ₹21 of each ₹100 of new deposits is already committed to reserves and liquid assets. That leaves ₹79, less than the ₹84 lent, so the rest has to come from existing liquid assets above the SLR minimum or from other funding. That is why bankers worry about a high incremental ratio even when the stock ratio is steady.
Common exam traps
M0 vs M3 components. Bankers' deposits with the RBI are in M0 only. Time deposits are in M3 only. Currency appears in M0 as currency in circulation and in M3 as currency with the public.
M1 vs M3. The difference between them is time deposits. M2 and M4 add post office deposits; they are not the headline measures.
NM2. It adds short-term (up to one year) time deposits of residents, not all time deposits.
L3 adds public deposits of NBFCs. L2 adds deposits and borrowing of FIs. L1 adds post office deposits.
"Banks lend out deposits." Wrong in the modern view: lending creates deposits. A single bank, though, does need to fund the reserves it loses when borrowers spend.
Repayment. Repaying a bank loan destroys money. A payment between two bank customers changes neither M3 nor M0.
The multiplier. 1 ÷ CRR (33 at 3%) is not India's multiplier. The measured M3/M0 ratio is about 6.25, mainly because of the public's high cash holdings.
Credit-deposit ratio vs gap. The ratio is credit ÷ deposits (81.6%). The gap is the difference in growth rates (about 1.5 pp on 15 October 2026).
Key facts at a glance
| Broad money (M3), 15 Oct 2026 | ₹333.00 lakh crore, up 16.0% y-o-y |
| Reserve money (M0), 15 Oct 2026 | ₹53.25 lakh crore, up 10.3% y-o-y |
| M3 ÷ M0 (money multiplier) | 6.25 (15 Oct 2026); 6.16 (end-March 2026) |
| Currency with the public | ₹42.56 lakh crore, 12.8% of M3 |
| Time deposits | ₹254.63 lakh crore, 76.5% of M3 |
| Currency in circulation | 81.6% of reserve money |
| M3 growth, 2025-26 | 15.3% (M0: 5.8%) |
| SCB credit and deposits, 15 Oct 2026 | ₹227.43 and ₹278.76 lakh crore; up 18.2% and 16.7% |
| Credit-deposit ratio | 81.59% (15 Oct 2026); peak 82.7% (May 2026) |
| Incremental CD ratio, 2026-27 so far | 83.96% |
| Highest multiplier in the last decade | 6.73 (end-March 2017, after demonetisation) |
| New aggregates | NM1–NM3 and L1–L3, Working Group chaired by Y. V. Reddy, 1998 |
| Bank of England paper | "Money creation in the modern economy", Quarterly Bulletin 2014 Q1 |
| Money data dates | 15th and last day of the month, from 15 Dec 2025 |
Exam corner: what to remember
- M0 = currency in circulation + bankers' deposits with the RBI + 'other' deposits with the RBI. Also called reserve money, base money or high-powered money.
- M1 = currency with the public + demand deposits + 'other' deposits with the RBI. M3 = M1 + time deposits. M2 = M1 + post office savings deposits; M4 = M3 + all post office savings deposits (excluding NSCs).
- NM2 = NM1 + residents' short-term time deposits (up to one year). NM3 = NM2 + long-term time deposits of residents + call/term funding from FIs. L1 adds post office deposits, L2 adds FI deposits, borrowing and CDs, L3 adds NBFC public deposits.
- Money multiplier m = (1 + c) ÷ (c + r) = M3 ÷ M0. India's is about 6.25.
- Loans create deposits (Bank of England, 2014: McLeay, Radia and Thomas). The RBI supplies reserves at the policy rate; it does not ration them.
- Limits on money creation: loan demand and interest rates, capital (11.5% of RWAs including the CCB), liquidity and funding (LCR 100%), CRR (3%) and SLR (18%), and borrower behaviour.
- Repayment destroys money. Inter-bank payments shift reserves but leave M3 and M0 unchanged.
- 15 Oct 2026: M3 +16.0%, M0 +10.3%, credit +18.2%, deposits +16.7%, CD ratio 81.59%.
- Demonetisation pushed the multiplier to 6.73 (March 2017); the pandemic pulled it to 5.23 (March 2021).
For exam-specific syllabus and pattern details, see our pages for RBI Grade B, SBI PO and IBPS PO.
Test yourself
According to the Bank of England's 2014 paper "Money creation in the modern economy", when a commercial bank makes a loan it:
Show answer
Answer: (b) The paper says that "whenever a bank makes a loan, it simultaneously creates a matching deposit in the borrower's bank account, thereby creating new money".
Which of these is NOT a component of reserve money (M0)?
Show answer
Answer: (a) Time deposits are part of M3; M0 consists only of currency in circulation, bankers' deposits with the RBI and 'other' deposits with the RBI.
NM2, one of the new monetary aggregates proposed by the 1998 Working Group on Money Supply, equals NM1 plus:
Show answer
Answer: (c) NM2 adds residents' short-term time deposits; post office deposits come in L1, FI instruments in L2 and NBFC public deposits in L3.
If the currency-deposit ratio is 0.15 and the reserve-deposit ratio is 0.05, the money multiplier is:
Show answer
Answer: (b) m = (1 + 0.15) ÷ (0.15 + 0.05) = 1.15 ÷ 0.20 = 5.75; 20 would be the simple 1 ÷ r, which ignores currency.
On 15 October 2026, M3 was about ₹333.00 lakh crore and reserve money about ₹53.25 lakh crore. The M3/M0 ratio was closest to:
Show answer
Answer: (d) 333.00 ÷ 53.25 = 6.25.
A customer of SBI pays ₹5 lakh by RTGS to a customer of HDFC Bank. What happens to M3 and M0?
Show answer
Answer: (a) The deposit and an equal amount of reserves move between banks; the totals do not change.
According to the RBI's Weekly Statistical Supplement of 23 October 2026, year-on-year growth of broad money (M3) on 15 October 2026 was:
Show answer
Answer: (c) M3 grew 16.0%; 10.3% was reserve money growth, 18.2% bank credit growth and 9.2% M3 growth a year earlier.
When a household repays ₹1 lakh of a bank loan out of its savings account, broad money:
Show answer
Answer: (b) The loan and the deposit both disappear, so repaying a bank loan destroys money.
The credit-deposit ratio of scheduled commercial banks for the fortnight ended 15 October 2026 was about:
Show answer
Answer: (b) The RBI reported 81.59%; it peaked at 82.7% in May 2026, and 78.9% was the June 2025 level.
A bank's lending adds ₹200 crore of new deposits. With a CRR of 3% and an SLR of 18%, how much must it eventually hold as CRR balances and SLR assets together?
Show answer
Answer: (d) CRR: 3% × 200 = ₹6 crore; SLR: 18% × 200 = ₹36 crore; total ₹42 crore.
Further reading
- Bank of England, "Money creation in the modern economy" (Quarterly Bulletin 2014 Q1): the 14-page paper that made "loans create deposits" mainstream, with balance-sheet diagrams.
- Bank of England, "Money in the modern economy: an introduction" (Quarterly Bulletin 2014 Q1): the companion piece on what money is, from IOUs to central bank reserves.
- Jakab and Kumhof, "Banks are not intermediaries of loanable funds – and why this matters" (Bank of England Staff Working Paper, 2015): the formal model behind the endogenous money view, for readers who want the economics.
- RBI, Handbook of Statistics on the Indian Economy, Table 36: Components of Money Stock: end-March reserve money, M1 and M3 from 1972-73, the series behind this part's multiplier chart.
- RBI, Handbook of Statistics, Notes on Tables: the official definitions of M0 to M4, NM1 to NM3 and L1 to L3, and the sources-side identity.
- RBI, Weekly Statistical Supplement: Money Stock, 23 October 2026: the latest M3 components and sources, updated every fortnight.
- RBI, Governor's Statement, 7 October 2026: the RBI's own reading of credit and deposit growth and why it raised rates.
Sources
- RBI, Weekly Statistical Supplement, Money Stock: Components and Sources, 23 October 2026: M3 ₹3,33,00,300 crore on 15 October 2026, up 16.0% (9.2% a year earlier); currency with the public, demand and time deposits, 'other' deposits; sources of M3 and their changes; note on the 2023 merger (accessed 24 October 2026)
- RBI, Weekly Statistical Supplement, Reserve Money: Components and Sources, 23 October 2026: reserve money ₹53,24,881 crore, up 10.3% (3.5% a year earlier); currency in circulation, bankers' deposits and 'other' deposits; RBI's net foreign exchange assets up ₹6.34 lakh crore (accessed 24 October 2026)
- RBI, Weekly Statistical Supplement, Scheduled Commercial Banks – Business in India, 23 October 2026: aggregate deposits ₹2,78,75,910 crore (16.7%), bank credit ₹2,27,42,777 crore (18.2%), changes in 2026-27 so far; fortnight definition from 15 December 2025 (accessed 24 October 2026)
- RBI, Weekly Statistical Supplement, Ratios and Rates, 23 October 2026: credit-deposit ratio 81.59% and incremental credit-deposit ratio 83.96% for 15 October 2026; CRR 3%, SLR 18%, repo 5.50%, SDF 5.25%, MSF 5.75% (accessed 24 October 2026)
- RBI, Handbook of Statistics on the Indian Economy 2025-26, Table 36: Components of Money Stock: end-March reserve money and M3 from 2015-16 to 2025-26 used for the multiplier series; bankers' deposits with the RBI of ₹9,91,488 crore (2025) and ₹8,10,124 crore (2026); merger-adjusted M3 for 2023-24 (accessed 24 October 2026)
- RBI, Handbook of Statistics on the Indian Economy, Notes on Tables: definitions of reserve money, M1, M2, M3, M4, NM1, NM2, NM3, L1, L2 and L3; the WGMS (Chairman Dr Y. V. Reddy, June 1998); coverage of co-operative banks; money data for the 15th and last day of the month from 15 December 2025 (accessed 24 October 2026)
- Bank of England, "Money creation in the modern economy", Quarterly Bulletin 2014 Q1: authors, published 14 March 2014; all quotations; three sets of constraints; deposits 97% of UK broad money; the multiplier operating "in the reverse way" (accessed 24 October 2026)
- RBI, Governor's Statement, 7 October 2026: repo raised to 5.50%; "risks in view of strong growth in monetary and credit aggregates exist"; bank credit growth 18.1% on 15 September 2026 against 10.4% a year earlier; SCB credit and deposits up 18.13% and 17.31% (accessed 24 October 2026)
- CareEdge Ratings, "Bank Credit Growth Jumps to 19.4% on Quarter-End Surge, Outpacing Deposits", 12 October 2026: quarterly credit and deposit growth and loan-to-deposit ratios from March 2025 to 30 September 2026; May 2026 peak of 82.7%; CD issuance; net RBI credit to Government down ₹2.62 lakh crore; FY27 credit forecast of 15–16% (accessed 24 October 2026)
- Angel One, 16 October 2026: credit growth of 19.38% and deposits of ₹283.95 trillion at end-September 2026; credit of ₹16.35 trillion and deposits of ₹21.67 trillion added in April–September (accessed 24 October 2026)
- Pareto Academy, Part 10: Open market operations and liquidity: swap inflows of US$143.6 billion by 21 September 2026; OMO sales of ₹1,00,000 crore between 17 and 28 September 2026; record OMO purchases of ₹7.39 lakh crore in 2025-26 (accessed 24 October 2026)
- Pareto Academy, Part 5: CRR and SLR: CRR history including 3% from 28 March 2020 and 29 November 2025; daily CRR minimum of 99% from 16 October 2026 (accessed 24 October 2026)
- Pareto Academy, Part 18: RBI regulation of banks and NBFCs: 9% CRAR plus 2.5% capital conservation buffer; LCR minimum of 100% (accessed 24 October 2026)
- Pareto Academy, Part 3: How monetary policy works in India: monetary targeting from 1985 (Chakravarty Committee) and the multiple indicators approach from April 1998 (accessed 24 October 2026)