Aug 28, 2026 - RBI has [announced](https://rbi.org.in/scripts/BS_PressReleaseDisplay.aspx?prid=63467) a 15-day VRRR auction for 6 lakh crore on the final day of the forex swap facility, to manage the secondary impact of absorbing capital flows.
RBI has many tools to manage the 'undesired capital flows.' Here, we focus on how RBI absorbs (sterilises) the extra rupee liquidity created by its own swap window.
As RBI swaps dollars, it pays banks by increasing balances in their current accounts with RBI known as "excess reserves". From here, these balances can then get drained in various ways like:
### Autonomous/non-policy leakages
1. cash withdrawals by the public (CiC)
2. changes in CRR requirements due to change in NDTL
3. forex transactions (other than forex swaps) - dollar supply in days of short-supply of foreign exchange and unwind/settle short term net drains on foreign currency assets due within next 3 months will absorb system liquidity.
4. govt. of India cash balances - advance tax and GST outflows in September, or
5. NNML variations - reduction in non-monetary assets
#### Breakdown of short-term net drains on foreign currency assets
Here we consider the period "Up to 1 month" and "More than 1 and up to 3 months" from the ["Data Template on International Reserves/Foreign Currency Liquidity](https://rbi.org.in/scripts/SDDS_ViewDetails.aspx?SDDSID=285), as on July 31, 2026 is:
1. II.1 – Foreign currency loans (taken by RBI/govt.), securities (issued by govt./rbi = 0), deposits (like foreign central banks' with RB) (net drains in $ million):
1. Up to 1 month: $406
2. 1–3 months: $2,115
3. Subtotal (up to 3 months): $2,521
2. II.2 – Forwards/Futures (net of short & long):
1. Short: Up to 1 month -$15,585 | 1–3 months -$7,318
2. Long: Up to 1 month $2,429 | 1–3 months **0**
3. Net up to 1 month: -$13,156
4. Net 1–3 months: -$7,318
5. Subtotal (up to 3 months): -$20,474
3. II.3 – Repos/Reverse repos (self-cancelling, net = 0):
1. Outflows and inflows are equal and offsetting each month and so net impact = 0
4. III.1 – Contingent liabilities (collateral guarantees):
1. Up to 1 month: $9 (shown as a positive stated liability figure, not netted against reserves the same way)
2. 1–3 months: $77
3. Subtotal (up to 3 months): $86
5. Total, up to 3 months (II.1 + II.2 + II.3 + III.1) $=$ -2,521 + (-20,474) + 0 + 86 ≈ -22,909 (US$ million)
So roughly **$22.9 billion** in net predetermined/contingent short-term drains due within the next 3 months and the bulk of it (~$20.5bn) coming from the forward/swap book, not loans or contingent liabilities, which are comparatively small.
### Absorption by RBI - VRRR and SDF, forex swaps, OMO/MSS/CMBs, and I-CRR
1. VRRR under the LAF or SDF - The daily MMO data shows banks have parked surplus funds under reverse repos. However, a cost is borne by the RBI via interest pay-outs, directly reducing its surplus transferable to the govt.
2. forex swaps - a short-term sell/buy forex swap is another possibility but it will add cost in terms of forward premium.
3. RBI's operations in government securities:
1. [Cash Management Bills (CMBs)](Cash%20Management%20Bills.md) - short tenor instruments, and They can also used as a liquidity management tool to adjust the rupee liquidity conditions in the market not on durable basis but rather transient/frictional basis.
2. OMO sales - Selling government securities from the RBI’s portfolio permanently absorbs liquidity and shrinks the RBI's balance sheet. But it reduces the central bank’s holdings of yield-earning asset stock and risks hardening sovereign bond yields.
1. 10-year G-sec yield marginally increased to 6.85% on August 21, 2026, from 6.84% on July 31, 2026, and 6.75% as at end-June 2026, as mentioned in RBI's August Bulletin.
2. G-sec yields hardened in July reflecting renewed geopolitical tensions in West Asia, volatile crude oil prices, rise in US treasury yields and higher domestic CPI inflation for June.
3. In mid-August, yields rose across tenors, reflecting a rise in crude oil prices and hardening of US treasury yields.
4. Market Stabilisation Scheme (MSS) - this is an innovative option where RBI sells MSS bonds on behalf of govt.
1. So govt bears interest cost directly. RBI simply manages the scheme. Funds move from "excess reserves" to "govt. balances with the RBI". So only the composition of liabilities shifts, and it does not change size of RBI's balance sheet. But the scheme will ruffle feathers in the Ministry.
2. The loss of interest income to the RBI (and hence cost to govt.) on sale of securities under the OMOs and the cost to govt. in the MSS scheme can be higher than paying interest under VRRR.
3. So here VRRR stands out as the most preferred option if the liquidity is transient in nature, which is not the case here. Liquidity from swap facility is durable in nature and hence, RBI will have to undertake longer-term transactions after the 15-day VRRR auction.
5. Incremental Cash Reserve Ratio (I-CRR) - Here we further analyse the I-CRR measure.
#### Why RBI may prefer I-CRR?
1. Now since March 31, 2007, CRR balances earn zero interest.
2. Unlike VRRR where a cost is borne by the RBI via interest pay-outs, directly reducing its surplus transferable to the govt., the central bank pays no interest on required reserves.
3. It also reduces the need for costly sterilisation through sale of bonds in OMOs or MSS, 2004.
4. RBI has used I-CRR successfully in the past, like in April 1997, when it announced the imposition of a 10% CRR on all incremental NRI deposits over the level outstanding on April 11, 1997.
#### Why RBI and banks may not prefer it?
1. <span style="color:#0047AB;">Credit is already outrunning deposits. So I-CRR would make this worse.</span>
1. SCBs' credit and deposit growth stood at 18.3% year-on-year and 14.6% year-on-year respectively as on August 15, 2026. Growth in deposits since April 1, 2026 has been ₹7,02,559 crore and growth in credit is ₹6,47,867 crore putting the incremental credit-deposit ratio at 92.2% for FY26-27, though it is lower than the peak value of the period.
2. Now locking up incremental deposits via I-CRR can tighten the supply of loanable short-term funds, and especially push the 3-month CD and CP rates up.
3. Now, locking up incremental deposits via I-CRR can tighten the supply of loanable short-term funds, which could push up the CD and CP rates, particularly at the 3-month tenor, where they are most actively traded.
4. Rather, on June 8, 2026, it exempted CRR on these deposits mobilised between June 08, 2026 and September 30, 2026, to lower their cost for banks.
2. <span style="color:#0047AB;">WACR is already broadly within the policy corridor despite system and durable liquidity surplus.</span>
1. WACR moderated to 5.13% in August (up to 21st) from 5.29% in July and is comfortably above the SDF floor of 5.00%, despite average SDF balances rising to ₹2.12 lakh crore during August (till 21st), from ₹1.60 lakh crore in July, largely reflecting banks that raised large FCNR(B)/NRI deposits.
2. WACR has not collapsed toward that floor of 5% because most of this surplus wasn't parked in SDF but went through VRRR auctions instead, where the RBI has been setting the cut off (~5.24%), well above the flat 5.00% SDF rate.
1. But banks can use both VRRR and SDF (lower rate) on the same day. For instance, if a bank doesn't want to lock funds beyond overnight when VRRR is offered for a longer tenor, or if its surplus is only confirmed after inter-bank settlements or past the VRRR cutoff timing. In all such cases, that leftover cash simply gets parked in SDF instead.
2. Just to recall the timings, VRRR timings are announced in the press release, and usually between 9:30 and 11:30 am. SDF is available between 7:00 PM and 11:59 PM on all days of the week, and the reversal is allowed even on weekends and holidays. The settlement of the same occurs in real-time. The reversal happens early morning next day. The call money market operates between 9:00 AM to 5:00 PM.
3. <span style="color:#6f1fb5;">Annex-1 has the daily data of MMO operations from Aug-1 to Aug-21, 2026 with call money market rates, volume and range.</span>
3. Since surplus banks naturally prefer the higher, competitively-priced VRRR return over the passive SDF rate, VRRR effectively became the rate that anchored the system, pulling WACR toward it rather than down to the SDF floor. This is visible in the sharp rise in total daily average net absorption under the LAF (SDF + VRRR), which is up to ₹3.48 lakh crore in August (till 21st), from ₹1.19 lakh crore in July and ₹0.89 lakh crore in June.
4. In effect, VRRR is keeping the WACR, which is the operating target of monetary policy, anchored close to the repo rate, without RBI needing to reach for a more disruptive tool like I-CRR.
3. <span style="color:#0047AB;">An Incremental-CRR applies to all banks but surplus liquidity is never evenly distributed.</span>
1. Large public or foreign banks may hold excess cash from swap legs, and they park it under reverse repo. When CRR is hiked, they will move them from "excess reserves" to "required reserves". In this case, there is no change in RBI's balance sheet. But as funds get locked instead of lending to economy, a hike in CRR carries the risk of misallocation of resources when credit growth is strong.
2. Banks lose VRRR rate income. Hence I-CRR is treated strictly as an emergency, temporary shock absorber rather than a permanent liquidity management tool.
1. I-CRR was used on August 10, 2023 to manage liquidity surge following the return of ₹2,000 banknotes, and even in times of BoP crisis of 1991.
3. Some banks might face tight liquidity due to lower deposits mobilisation or strong credit demand. If such a deficit bank is forced to borrow via the LAF repo window at the policy rate to meet mandatory I-CRR targets, the RBI's balance sheet expands on the asset side (loans to banks) while raising the bank’s effective cost of funds. But this helps RBI earn interest income on repos while absorbing overall liquidity. So it is a win for the central bank’s cost of sterilization. In this case, banks pay the repo rate/VRR rate.
4. In both cases, there is a drag on banking sector margins. This will then also take away the advantage of lower cost on these funds, and perhaps spillover to the borrowers.
5. Balance sheet impact - Though both absorb rupee liquidity, OMO sales reduce balance sheet, while CRR hikes may keep balance sheet unchanged or expand it if deficit banks tap the LAF repo window.
4. <span style="color:#0047AB;">Another challenge is unwinding of I-CRR</span>.
1. This has to be done again phase-wise or sudden release of impounded liquidity could lead to a liquidity shock and push down the WACR to the lower bound.
2. Also VRRR have been for tenor as short as 1 day or for up to 15 days, repeatable and helps banks to manage liquidity. I-CRR, on the other hand, is typically imposed with no clear end-date and needs a separate future announcement to unwind as seen in the 2023 I-CRR episode, which took 3 phased steps to unwind from September to October.
5. <span style="color:#0047AB;">G-sec yields are already firming and I-CRR would add pressure</span>
1. 10-year G-sec yield rose from 6.75% (June) to 6.85% (Aug 21), driven by crude and US treasury yields, as mentioned in the RBI bulletin report of August 2026.
2. There is also some uncertainty surrounding US interest rates which risks pushing up domestic bond yields.
3. A CRR-type liquidity squeeze risks pushing yields up further, which could be an unwelcome side effect RBI would want to avoid.
The RBI Aug-26 bulletin says "Overall, daily average net absorption under the LAF increased significantly to ₹3.48 lakh crore during August (till 21st)...". But [my calculation as shown below](https://docs.google.com/spreadsheets/d/18zwRlE-MsODlxIy6L_uMC1Jk9sCzW7lmUZp_Yw1zpC0/edit?usp=sharing) (without SLF) comes out to 3.46 lakh crore. It usually matches. I've attached the raw data sheet below.
Annex-1
So, what's the better tool here? Will RBI keep leaning on a mix of tools used gradually over time or will it eventually reach for a CRR hike?
## Related Notes
1. [Reserve Money](Reserve%20Money.md)
2. [Market Stabilisation Scheme (MSS), 2004](https://iloverbi.com/The+Reserve+Bank+of+India/Market+Stabilisation+Scheme+(MSS)%2C+2004)
3. [Cash Reserve Ratio (CRR)](https://iloverbi.com/The+Reserve+Bank+of+India/Cash+Reserve+Ratio+(CRR)#Maintenance%20of%20CRR)
4. [Money Market Operations (MMO)](https://iloverbi.com/The+Reserve+Bank+of+India/Money+Market+Operations+(MMO))
5. [Liquidity Adjustment Facility (LAF)](https://iloverbi.com/The+Reserve+Bank+of+India/Liquidity+Adjustment+Facility+(LAF))
1. [Standing Deposit Facility (SDF), 2018](https://iloverbi.com/The+Reserve+Bank+of+India/Standing+Deposit+Facility+(SDF)%2C+2018)
6. [Forex Swaps](https://iloverbi.com/The+Reserve+Bank+of+India/Forex+Swaps)
7. [Capital Flows - Account & Management](https://iloverbi.com/The+Reserve+Bank+of+India/Capital+Flows+-+Account+%26+Management)
8. [Forex Interventions and Sterilisation](https://iloverbi.com/The+Reserve+Bank+of+India/Forex+Interventions+and+Sterilisation)
## References
1. Duvvuri Subbarao. (Jan 31, 2011). ==Implications of the Expansion of Central Bank Balance Sheets==. (Comments of Dr. Duvvuri Subbarao, Governor, Reserve Bank of India at the Special Governors’ Meeting in Kyoto, Japan, on January 31, 2011.). [Link](https://rbi.org.in/scripts/BS_SpeechesView.aspx?Id=547) | [pdf](RBI_Speeches_20110131_Implications%20of%20the%20Expansion%20of%20Central%20Bank%20Balance%20Sheets.pdf)
2. RBI. (2014, January 21). [Report](https://rbidocs.rbi.org.in/rdocs/PublicationReport/Pdfs/ECOMRF210114_F.pdf) of the Expert Committee to Revise and Strengthen the Monetary Policy Framework. Chairman: Dr. Urjit R. Patel, Deputy Governor, Reserve Bank of India.
3. RBI. (Aug, 2018). Table 3: Variation in Reserve Money and Main Durable Liquidity Drivers. RBI Bulletin (August-2018). [Link](https://rbi.org.in/scripts/BS_ViewBulletin.aspx?Id=17703)
4. RBI. (2025, May). Report of the Working Group on Comprehensive Review of Trading and Settlement Timings of Markets Regulated by the Reserve Bank \[Report\]. [Link](https://rbi.org.in/scripts/PublicationReportDetails.aspx?ID=1292)