Introduction

I would like to thank KangaNews for the opportunity to discuss some important updates to the system for monetary policy implementation in Australia. The Reserve Bank Board discussed this late last year, and we are now ready to announce operational changes to our Open Market Operations (OMOs) that will support the transition to ample reserves.

Monetary policy implementation is at the core of the financial system’s plumbing. It is how we give effect to changes in the cash rate target, influence other money market rates and provide liquidity to the banking system. Importantly, it enables us to conduct monetary policy in a way that best contributes to both price stability and full employment.

The RBA achieves this by providing banks access to Exchange Settlement (ES) balances – otherwise known as reserves. Banks use these funds to settle payments with other banks and the RBA. Banks also hold reserves for precautionary and regulatory purposes. In response to various price signals, and to help manage their reserves and deal with their funding needs, banks borrow and lend reserves in money markets. These transactions underpin key interest rates in the Australian economy – such as the cash rate and short-term money market rates like bank bill swap rates.

An effective monetary policy implementation system is critical for all market participants. It aids in the smooth transmission of monetary policy, supports good functioning of money markets and hence other key financial markets, and encourages greater resilience in the financial system.

In March last year, the Reserve Bank Board endorsed the new system for implementing monetary policy. Banks’ demand for reserves would be satisfied in full at our OMOs, at a price near the cash rate target, using full allotment repurchase agreement (repo) auctions. We call this system ‘ample reserves with full allotment’ because it supplies as many reserves as banks demand at our OMOs.

In April last year, I discussed why the Board endorsed this framework.¹ In brief, it is a simpler and more robust system for us to operate compared with the alternatives. It is also similar to systems used by other central banks, including the European Central Bank and the Bank of England. Banks will determine the amount of reserves they hold to suit their liquidity needs. The system is resilient to structural changes affecting banks’ underlying demand for liquidity as well as policies that might affect the size of the RBA’s balance sheet (such as unconventional policies if they were to become necessary again). At the same time, it implies a materially larger steady-state balance sheet for the central bank compared with pre-pandemic times.

Over the past year or so, we have been working on the detailed design of this system, and today I am announcing some important changes. I stress that these changes are operational in nature. They do not represent or signal a shift in the stance of monetary policy. Nor do they have a bearing on the Monetary Policy Board’s current approach to allowing bond holdings acquired during the pandemic to mature.

Specifically, effective from 9 April 2025, we will:

• increase the price of all new OMO repos by 5 basis points to 10 basis points over the cash rate target; OMO will continue to be offered at a floating rate
• introduce a seven-day term, in addition to the existing 28-day term, at each weekly OMO.

Before outlining the Reserve Bank Board’s deliberations and explaining why we have decided to make these changes, I want to review recent market developments.

Recent developments in markets

Reserves have declined around $110 billion over the past year (Graph 1). Most of this reflected the final repayment of the Term Funding Facility (TFF) in June 2024. Subsequently, the level of reserves has fluctuated around $240 billion, and the cash rate has remained close to, but slightly below, the cash rate target.

Activity at our OMOs increased from around $3 billion a week in the June quarter of 2024 and has stabilised around $7 billion. This increase occurred shortly after the final repayment of the TFF, alongside a broader tightening in liquidity conditions in money markets globally. In response, banks accessed more reserves from OMO, and some of those funds appeared to have been recycled into other money markets. This was an early indication that the full allotment system was working as intended – reserves rose automatically in response to an increase in demand for liquidity while increases in money market rates were largely contained (Graph 2).

Current market conditions suggest that the transition to ample reserves – that is, a level of supply that is in balance with banks’ underlying demand – is ongoing. The stock of reserves remains elevated, reflecting the bonds still on the RBA’s books that we purchased during the pandemic. Our expectation is that reserves will continue to decline gradually for a time in response to the decline in the RBA’s bond holdings.² Eventually though, the supply of reserves will approach banks’ underlying demand, and thereafter banks’ participation in OMO should pick up to offset the effect of further declines in the RBA’s bond holdings.

Underlying demand for reserves is hard to estimate and it will only become evident as we approach ample reserves. We have done modelling work and banks have also provided us with estimates of their own demand for reserves. This suggests that underlying reserves could be anywhere between $100 and $200 billion.³ An advantage of our full-allotment system in the face of such uncertainty is that the transition to ample reserves can occur without us needing to know the level of banks’ underlying demand ahead of time. OMO use will rise automatically. Such a move, combined with an assessment of market conditions and liaison with the banks, will indicate when reserves have reached an ‘ample’ level. Private market activity may also increase as we approach this point – particularly in the short-term repo and cash markets. This is because banks wanting additional reserves on non-OMO days will seek to borrow them in private markets. Other banks can lend reserves if they have more than they need. The scale of this activity will depend in part on the extent to which banks choose to economise on their reserve holdings, given that obtaining reserves at OMO and leaving them in ES accounts comes at a cost to the banks. I will come back to this point in a moment.

Principles for an ample reserves system

Over the past year, the RBA has consulted banks, estimated the underlying demand for reserves, and considered the ways in which the new ample reserves system might operate. We have published a summary of consultation responses on our website today; thank you to those who contributed. This work informed discussions at the Reserve Bank Board late last year at which three key principles for the ample reserves system were considered:

1. Sufficient monetary control. The Board agreed that the primary objective for monetary policy implementation was to achieve sufficient ‘monetary control’. This involves the cash rate trading close enough to the target with other short-term interest rates tethered to the cash rate to be consistent with the desired stance of monetary policy.
2. Supporting private markets. The Board agreed that we could achieve the primary objective of monetary control while still allowing deviations of the cash rate from target. Allowing the cash rate to trade within a modest range will avoid the RBA having an overly large presence in markets and thereby encourage banks to use private markets. Well-functioning private markets will help banks to better manage their funding needs in normal times and times of stress. Banks can be encouraged to use private markets by setting the price for OMO in a way that avoids the RBA having an overly large presence in the repo market.⁴ Using a mix of different operations to supply reserves could also be used to avoid an overly large presence in any one market.
3. Minimising risk to the RBA balance sheet. Providing reserves carries risks for the RBA – both financial and operational. The size and nature of the risks depend on the quantity of reserves as well as the characteristics of the operations used to supply them. Under an ample system, the RBA will provide more reserves compared with the earlier corridor system. OMOs do not carry interest rate risk because the floating rate of our OMOs is linked directly to the rate we pay on our liabilities. However, the use of other operations to supply reserves could entail financial risk.

A key question we considered was how to balance these principles given there is some tension between them. For example, we could have a high degree of monetary control by setting a low price for OMO close to the ES rate. But that would encourage banks to obtain a lot of reserves via OMO, crowding out private market activity and implying a large balance sheet for the RBA. Decisions on the configuration of OMO as well as the mix of other operations to supply reserves will need to balance these various trade-offs.

Changes to the configuration of our OMOs

We have been running full-allotment OMOs since the onset of the pandemic. We switched these from daily to weekly auctions from October 2021. We then offered a term of 28 days and at a price 5 basis points above overnight indexed swaps from early 2022. We then switched this price to a floating rate that was 5 basis points above the cash rate target from February of last year. The system has worked well under an excess reserves system and has delivered an acceptable degree of monetary control. However, as reserves will decline further, and demand for OMO will pick up when reserves are no longer in excess of banks’ underlying needs, we judged that some further changes were warranted.

A key issue is that at a price of 5 basis points above the cash rate target, meeting a large increase in the demand for funds at OMO might impair, at least at the margin, the health of other private money markets. Similarly, this low price for OMO will lead to a larger RBA balance sheet than otherwise and implies a tighter degree of monetary control that we judged to be necessary. At the same time, the current 28-day tenor is too long for those banks that may need additional reserves for only short periods, and it is much longer than the tenor of some key markets, particularly for overnight cash.

The changes I have announced will better allow us to balance the various trade-offs between meeting the three principles I have outlined. The two changes effective from 9 April 2025 are:

• We will increase the price of all new OMO repos from 5 basis points to 10 basis points over the cash rate target.⁵
• We will offer a seven-day tenor in addition to the current 28-day tenor.

Auctions will continue to take place once a week (generally on a Wednesday morning).

An OMO rate of 10 basis points over the cash rate target remains consistent with the Board’s desired degree of monetary control. Under this higher OMO price, we expect the cash rate will trade within a reasonable range of the cash rate target. Accordingly, the cash rate, and other money market rates, will be consistent with the desired stance of monetary policy.

Importantly, this higher price for OMO implies a lower overall demand for reserves than otherwise. The higher price will provide more of an incentive for participants to recycle reserves in private markets. Banks can still come to OMO to acquire reserves to meet their payment needs and obtain ‘precautionary reserves’ for unexpected liquidity needs or to lend to others. But the higher price will reduce banks’ incentives to obtain more reserves at OMO than necessary. A bank can make good use of private markets as a source of reserves if they face an unexpected need for funds.

Offering a seven-day tenor has a couple of benefits. OMO will provide a closer substitute to overnight cash and funding from other short-term money markets. By itself, this will strengthen the degree of monetary control over those key markets. This decision is also consistent with feedback from market participants that a shorter tenor would help them to better manage their liquidity needs. However, respondents to the consultation also expressed an interest in the 28-day tenor. Retaining that longer tenor allows banks and the RBA to more efficiently manage their OMO activity by reducing operational burdens associated with more frequent rolling of positions.

During consultation some market participants wanted more frequent operations, but we believe the current weekly auction is enough to anchor the cash rate and other money market rates to the target. This setup also encourages banks to use private markets, especially on non-OMO days. In line with APRA’s standards, banks must have strong frameworks for forecasting their liquidity demands and managing their liquidity risks. These processes are becoming more important as banks need to increasingly engage in private money markets to meet their liquidity needs.

As we transition to the ample reserves system, the RBA and market participants will gain valuable insights. We will actively monitor market conditions, engage with banks, and respond if needed, including by adjusting our OMO or other administered rates.

Features of the ample reserves system

Private markets

As we transition to ample reserves, some banks may need more liquidity than their current ES balances. One option is to borrow reserves from a bank with a surplus, benefiting banks on both sides of such transactions. This private activity may be associated with short-term volatility in money markets as prices adjust to supply and demand changes. Within reasonable bounds, this is a sign of healthy markets. Weekly full allotment OMOs will help banks meet their liquidity needs. But to limit volatility, banks should be ready to transact in various markets, including the cash market. Banks might use OMOs to acquire reserves for precautionary reasons or to lend into other markets when prices are high. Over time, banks will refine their reserve management approaches in the ample reserves system.

The RBA’s overnight standing facility

If banks face unexpected liquidity needs on a non-OMO day or after OMO has taken place, and cannot find liquidity on suitable terms in private markets, we would expect and encourage them to use the RBA’s overnight standing facility (OSF). This facility provides reserves overnight at 25 basis points above the cash rate target, thereby limiting deviations in money market rates from the cash rate target set by the Monetary Policy Board. While the price is set to avoid displacing private market activity, it provides an incentive for banks to use the facility when other sources are more expensive.

Historically, market participants have been reluctant to use this facility. However, both the RBA and APRA expect that banks should use the OSF as part of their liquidity management if they fall short on their daily liquidity needs. We will encourage its use as part of the new normal.

In the rare case of broader stress across the banking system, the RBA could run an unscheduled OMO. But that would not be the standard approach in the case of a few banks requiring additional liquidity that could otherwise be provided in the market or via the OSF.

Other operations

In addition to our open market repo operations, the RBA plans to use other operations to provide reserves across a range of markets, including foreign exchange swaps and purchases of short-dated government bonds. We would not use these to influence rates or liquidity in those markets. Rather, they will help the RBA to limit the extent of our footprint in any one market, particularly the repo market, and manage operational risks. The use of these operations is expected to be some time away since reserves supplied via OMO should gradually rise to meet demand as the supply of reserves from our existing bond holdings declines. We will outline our plans for these operations before actively using them to manage monetary policy implementation.

The rate paid by the RBA on reserves

When the RBA moved to an excess reserves system in March 2020, banks had little need to borrow in the cash market, and the cash rate became closely anchored to the ES rate (Graph 3). The Reserve Bank Board narrowed the spread between the cash rate target and ES rate to 10 basis points and announced the ES rate in its monetary policy decisions. As we continue to transition to ample reserves, borrowing rates in private markets will rise as demand for liquidity from those sources increases, partly due to the higher rate at our weekly OMO. Consequently, the ES rate will be less significant as an anchor. Because of this, starting in May the Monetary Policy Board will announce the cash rate target in its decisions but not the ES rate.

Moreover, from time to time the RBA may adjust the ES rate if that will help to better meet the objectives of the ample reserves system. For example, we may need to provide market participants with more of an incentive to recycle excess reserves by altering the ES rate, thereby changing the opportunity cost of holding reserves. Any such adjustments would be purely operational in nature and would not represent a shift in the stance of monetary policy. Indeed, such changes in the ES rate could occur as needed. While we would convey these clearly to the market, such changes would not require the approval of, or announcement by, the Monetary Policy Board.

Next steps

To reiterate, the changes to our operations will take effect on 9 April 2025.

It is important that banks focus on their liquidity management practices as we continue to transition to the ample reserves system. During the excess reserves period, many did not need to top up their reserves, but now all banks must be ready to use our facilities and transact in private markets.

The RBA and APRA will encourage banks to use the overnight standing facility as needed as part of their routine liquidity management. Today we have released a joint statement to emphasise this commitment and together we will engage with banks to ensure they understand the role of the OSF and are comfortable and ready to use it to manage liquidity as the system transitions to an ample level of reserves.

Meanwhile, we will continue to monitor conditions in key markets, including by talking regularly with market participants.

Finally, I stress that these changes have no implications for the stance of monetary policy. They do, however, represent important changes in the plumbing that supports the transmission of monetary policy and underpins critical activities across the financial system.

Endnotes

I thank Sean Dowling, Gian-piero Lovicu, Thomas van Florenstein Mulder and Aidan Penman for their excellent assistance in helping to prepare this speech. *

See Kent C (2024), ‘The Future System for Monetary Policy Implementation’, Bloomberg Australia Briefing, Sydney, 2 April. 1

In particular, the Australian Government will repay the RBA by issuing new bonds to the private sector (or receiving more in revenue than it spends), which will drain reserves from the system. Note that for the bonds issued by the governments of the states and territories this reduction in reserves happens on the days that semis mature. 2

Bristow L (2024), ‘Modelling Reserve Demand with Deposits and the Cost of Collateral’, RBA Research Discussion Paper No 2024-08. 3

Other operations could include foreign exchange and cross-currency swaps as well as purchases of near-maturity Australian Government Securities. RBA (2024), ‘Minutes of the Monetary Policy Meeting of the Reserve Bank Board’, Hybrid, 9 and 10 December. 4

The price on existing OMO repos will remain 5 basis points over the cash rate target. 5