In June 2022, the SA Reserve Bank moved away from implementing monetary policy through a classical cash reserve or shortage system, to a surplus or tiered floor system. In this new system, banks are provided with a larger supply of reserves (also called settlement balances) than they need to satisfy reserve requirements and make interbank payments. The SA Reserve Bank then offers a deposit facility where banks are able to place these excess balances, earning the repo rate, on an overnight basis. To preserve some lending in the interbank market, the SA Reserve Bank places certain limits (quotas) on deposits, with excess balances earning a lower rate.
The core logic of the system is that the ample supply of reserves exerts a downward pressure on interest rates, while the deposit facility forms a floor which prevents rates from falling below the policy rate.
This paper provides a full description of the SA Reserve Bank’s monetary policy implementation framework (MPIF), which was implemented in June 2022.
Most modern economies, including South Africa, are using a short-term interest rate to achieve the goals of monetary policy. In South Africa, that short-term rate is the repo rate, which is set by the Monetary Policy Committee (MPC) to deliver on the flexible inflation target. The role of the MPIF is to transmit the repo rate decisions of the MPC to financial markets and the economy at large.
The repo rate affects a range of financial assets, including mortgage and vehicle loans, which is where many individuals see its effects most clearly. However, it describes in the first instance the cost of financing a very specific asset, namely bank reserves. These are issued exclusively by the SA Reserve Bank and can be held only by institutions with accounts on the South African Multiple Option Settlement (SAMOS) payment system. Most of these institutions are banks that use these reserves for two purposes – to meet regulatory reserve requirements and to process interbank payments. Since the SA Reserve Bank is the only issuer of bank reserves, it can determine their price. This power is the basis for monetary policy, which is exerted through the MPIF.
There are two basic frameworks for controlling short-term interest rates. One is to restrict the supply of reserves so that banks have neither too many nor too few. The central bank calculates the likely requirements of the whole system, and either injects or drains reserves to deliver on the expected demand, leaving the distribution of reserves among banks to the interbank market. Banks still have the option to borrow or lend from the central bank, each day, but deposits earn a lower rate while loans are charged at a higher rate. To avoid these punitive rates, banks will typically prefer to trade reserves, so that banks with excess funds lend them to banks with shortages. The objective is that interbank rates end up closely aligned to the policy rate. These systems are typically referred to as scarce reserves or mid-corridor frameworks – the latter term reflecting the way that the policy rate sits in the middle of a corridor, bordered by the lending and deposit facilities.
The main MPIF alternative is known as a surplus or floor system. Banks are provided with ample reserves, so their demand is completely saturated. The central bank then accepts deposits of excess reserves, reimbursing these deposits at the policy rate. The option to deposit with the central bank prevents banks from lending out excess reserves at lower rates, as the central bank is the safest borrower in the system. In contrast to scarce reserve systems, this arrangement is much less sensitive to the volumes of bank reserves: as long as demand is satiated, interest rate control remains possible with different size surpluses.
The SA Reserve Bank uses a modification of a floor system, known as a tiered floor or quota floor. Under this arrangement, banks can deposit excess reserves at the SA Reserve Bank and earn the policy rate, but deposit volumes are subject to quotas. If banks exceed these quotas, the marginal portion earns a lower interest rate.
This system encourages banks to recirculate some reserves, maintaining interbank market functioning. The system of quotas is aimed at avoiding challenges observed in some other economies, where reserves were hoarded by some banks and other banks ended up short, leading to weakened interest rate transmission. This problem can also be addressed by providing a larger supply of reserves, but the SA Reserve Bank would not want to have its MPIF force it into substantial and unpredictable balance sheet expansions.
The SA Reserve Bank’s MPIF is aimed at implementing monetary policy efficiently, effectively and resiliently.
The effectiveness requirement means that the framework should be successful in transmitting the policy rate to the broader economy and financial markets. In other words, it should set the base for all other interest rates.
Efficiency means that the system should deliver this interest rate control without imposing excess costs in terms of time or money for users, and without requiring unnecessary expense from the SA Reserve Bank.
Resilience means it should work under a wide variety of scenarios.
Globally, there is no ‘one size fits all’ rule for policy implementation frameworks. The SA Reserve Bank prefers to use a tiered-floor framework because it fits well with the South African context and the SARB’s mandates. In particular, this framework is better suited to the structural surplus of bank reserves seen in South Africa since the mid-2000s (mostly related to the accumulation of foreign exchange reserves). The ability to supply liquidity, without losing control of interest rates, is also likely to support financial stability.
The MPC pursues its objectives by setting a policy rate called the repo rate. In turn, this repo rate pins down the short end of the South African yield curve.
In the SARB’s MPIF, the repo rate does this by establishing the return on a maximally safe and liquid rand asset, which is a rand deposited at the central bank overnight. This rate then feeds into the pricing of other assets further out on the yield curve, with adjustments made by markets to reflect differences in the riskiness of the borrowers, the time period of investments and the expected repo rate.
In general, other assets will typically have yields above the repo rate because they are not as safe and liquid as cash balances at the central bank.
Transmission of the repo rate is enforced by an arbitrage condition. If a given asset is mispriced relative to the repo, banks will have incentives to exchange it for excess reserves. They would want to buy it if it is too cheap (reducing their stock of excess reserves) or sell it if it is too expensive (receiving reserves instead).
Because payments are made in rands – more technically, the rand is the unit of account and the means of exchange for the South African economy – the price of rands has powerful effects on other asset prices. In turn, these effects shape activity in the real economy.
Given that the SA Reserve Bank implements monetary policy by administratively paying interest on deposits rather than by steering a specific market rate, it is not necessary to identify a single benchmark interest rate for verifying transmission. Instead, the SA Reserve Bank uses a pluralist approach, consulting a range of indicators. Specifically, the SA Reserve Bank considers various short-term market rates, primarily in the interbank market, as well as banks’ weighted cost of funding with the SA Reserve Bank. The SA Reserve Bank also draws on market intelligence and data to monitor the transmission process.
One advantage of this approach is that it is robust to changing market structures, such as the shift in the interbank market from unsecured to secured lending. It is also well suited for navigating benchmark reforms, such as the replacement of the South African Benchmark Overnight Rate (Sabor) and the Johannesburg Interbank Average Rate (Jibar) with new reference rates in the future.
The SA Reserve Bank’s tiered floor system limits the size of deposits banks can place at the SA Reserve Bank overnight. Each bank faces a quota. Funds in excess of quotas earn the repo rate less one percentage point. Banks have no obligation to fill quotas.
To ensure transparency, both the rules determining quotas and actual quota amounts are public.
In determining quotas, the SA Reserve Bank starts with the amount of liquidity it desires to place in the system. This is known as the liquidity target. This amount is supplemented with a shock buffer to absorb unexpected movements in liquidity driven by factors such as notes and coin. The sum of the liquidity target and the shock buffer is then divided between SAMOS banks.
In determining banks’ shares of this amount, the SA Reserve Bank uses bank liability data from the BA900 forms. Every six months, the SA Reserve Bank calculates the total liabilities of SAMOS banks as well as the share of each SAMOS bank in that total. For smoothing, these calculations are made with an average of liabilities for the most recent three months. Banks are allocated quotas based on their shares of total liabilities. This is similar to the method used for determining banks’ required cash reserves.
To ensure quotas are easy to use, and the amounts are non-trivial even for small banks, quotas are rounded up to whole numbers. Large banks are rounded up to the nearest R1 billion; medium banks are rounded up to the nearest R500 million; small banks are rounded up to the nearest R200 million.
Quotas are designed so that they are always large enough to comfortably accommodate all excess liquidity in the system. If some banks have filled their quotas, other banks will have quota space. If they do not, the SA Reserve Bank will provide supplementary facilities at square off to balance the system.
The SA Reserve Bank has flexibility to adjust quota volumes as needed. Such adjustments can happen at any time; they are not tied to the six-monthly updating process described above. Adjustments are made by changing the liquidity target. It would be appropriate to change the liquidity target following a material change to the supply of bank reserves. The shock buffer is also at the discretion of the SA Reserve Bank and can be recalibrated if needed.
Quotas are published on the SARB website (see here for an example).
From 1998 until 2022, the SA Reserve Bank used a shortage or classic cash reserve system for implementing monetary policy. In this framework, the SA Reserve Bank ensured banks did not have sufficient reserves to satisfy their cash reserve requirements, draining liquidity as needed to engineer the targeted shortage. Banks borrowed the missing funds at Wednesday repo auctions, paying the repo rate. The current framework is in many ways the mirror image of the shortage system, with banks in a surplus rather than a short position, and the SA Reserve Bank implementing policy through a deposit facility rather than a lending facility.
Further details of the shortage framework may be found here.
In November 2021, the SA Reserve Bank published a consultation paper proposing a new framework for implementing monetary policy. This paper is available here. Following a period of public consultation, the SA Reserve Bank finalised the framework reform and started the transition to the new framework in June 2022. Details of the transition as well as the major themes arising from the consultation period, are available here.
What’s the difference between a monetary policy implementation framework and a monetary policy framework?
The SA Reserve Bank has a mandate, enshrined in the Constitution, to protect the value of the currency in the interests of balanced and sustainable growth. It pursues this mandate using a flexible inflation target, set in agreement with National Treasury. The MPC decides on the appropriate repo rate setting to deliver on the inflation target, over time, while taking care to avoid excessive volatility in output. This is the monetary policy framework.
The monetary policy implementation framework (MPIF) is simply the means whereby the SA Reserve Bank gives effect to the decisions of the MPC. It is effectively a plumbing operation, carried out by the Financial Markets Department.
A monetary policy framework can be judged based on macroeconomic outcomes such as the medium-term inflation rate. By contrast, an MPIF succeeds if it transmits the interest rate set by the MPC to markets, reliably, efficiently and without imposing undue costs.
How can the SA Reserve Bank afford to pay interest on reserves?
Up until 2022, the SA Reserve Bank used a shortage system to implement monetary policy. The system was structurally in surplus, and the SA Reserve Bank was therefore forced to drain liquidity to create shortage. These draining operations were expensive and sometimes caused distortions in financial markets. When the SA Reserve Bank moved to a surplus system, these draining activities were unwound. The shift allowed the SA Reserve Bank to use a cheaper tool – interest on reserves – to implement monetary policy instead. Paying interest on reserves has therefore been a cost saving.
Over the longer term, when assessing costs it is important to consider not just the liabilities side of the balance sheet – with reserves that earn interest – but also the assets side. When the SA Reserve Bank expands the supply of liquidity, it also acquires assets. For this reason, paying more interest on reserves does not automatically impose a net cost. For instance, if the SA Reserve Bank lends money at repo auctions at the repo rate, and also pays interest on that money at the repo rate, there is no net cost: the interest on the asset equals the payment on the liability.
It is also important to appreciate that the SA Reserve Bank’s primary objective is not profitability. The SA Reserve Bank is required to deliver on its mandates. Costs are relevant mainly because they might affect the SA Reserve Bank’s credibility and ability to achieve its mandates. Loss-making operations may be acceptable so long as they are consistent with the SA Reserve Bank’s mandates.
Does this MPIF mean the SA Reserve Bank is doing quantitative easing?
In popular usage, ‘quantitative easing’ (QE) is sometimes used to describe any money creation by a central bank. This simple definition is problematic, as central banks have a long history of creating money for diverse purposes, including facilitating payments in the economy, protecting financial stability by acting as a lender of last resort, and controlling short-term interest rates. Such practices long predate the term QE, which was first used in Japan in the 1990s.
Rather than focusing on money creation, it is more useful to think of QE as a specific monetary policy tool used to provide stimulus when the conventional short-term interest rate tool is constrained by the zero lower bound. This is done by buying longer-term assets such as government bonds. As long as the SA Reserve Bank can achieve its monetary policy goals using the repo rate, it is not necessary to resort to QE.
Many banks that use floor systems have done so in the context of QE policies, as they allow for interest rate control despite a surplus of bank reserves. But there are countries that have adopted floor-style systems without undertaking QE, including New Zealand and Norway, the two models for the SA Reserve Bank’s framework. In these cases, floor-type systems were adopted to enhance the payment system and simplify the monetary policy framework. The zero lower bound constraint was not binding.
Ultimately, ample-reserve frameworks like the SA Reserve Bank’s represent a technology for controlling interest rates, but do not require or imply QE.
Do other emerging markets also use floor-style systems?
From time to time, a number of emerging markets (EMs) have had interest rates at the bottom of their policy corridors. However, it appears that South Africa is the first EM to explicitly adopt a floor-style system, with the policy rate used as the overnight deposit facility rate.
A critical consideration for adopting a floor-style system has been South Africa’s structural surplus of reserves, which has hampered the smooth working of a system predicated on a shortage. In fact, many other EMs have run surplus liquidity positions for extended periods, often related to purchases of foreign exchange reserves. This liquidity has usually been dealt with using other tools, such as cash reserve requirements, longer-term bond issues or agreements with fiscal authorities.
It is not obvious why interest-on-reserves has not been used more widely among EMs, given its simplicity and cost-effectiveness. It may be that the reliability of this tool was not previously well-established or that financial markets in EMs were not sufficiently developed to cater for such a system. Floor-type systems have become standard among the major advanced economies over the past decade, however, providing ample evidence of their effectiveness.
Won’t extra money be inflationary?
A common misconception about bank reserves is that banks lend them out into the economy. This sometimes leads people to think that creating more money leads to higher prices (inflation) through a mechanism of ‘more money chasing the same goods’.
However, bank reserves can only be held by institutions with accounts at the central bank. These balances can move around between banks, but the total supply of reserves is determined by the central bank. Because banks cannot lend these reserves out to firms, households, or non-residents, it is not the quantity of reserves that matters for monetary policy transmission, but instead the price (i.e., the interest rate). This is one reason why central banks use interest rates as their key policy tool. As long as a central bank can control the interest rate on bank reserves, the quantity of reserves need not have an independent effect on prices.
This principle was demonstrated in the wake of the global financial crisis, when central banks significantly expanded the supply of bank reserves without causing any corresponding increase in inflation.
What does this mean for ordinary households and firms?
The SA Reserve Bank’s monetary policy implementation reform is not meant to intrude on the lives of most firms or households. Like any good plumbing, it is intended to function smoothly and unobtrusively. The framework reform does not have any impact on the way the MPC goes about achieving its mandate, which it does by setting interest rates.
That said, the reform will have some impacts on the larger economy. Because the new framework is expected to be more efficient and less distortionary than its predecessor, it will save the broad public sector some money and improve the transmission of monetary policy. Floor-style systems also tend to facilitate credit extension, by expanding liquidity and therefore reducing liquidity risk for lenders. They further tend to put an upward pressure on deposit rates, as banks at a minimum have the option to place funds at the central bank and earn the policy rate, an option not available in mid-corridor systems.
These effects are likely to be modest, and therefore not highly visible to the general public. Monetary policy implementation will be improved but not transformed.
The daily calculation and publication of Jibar rates are essential for the efficient functioning of the South African money, capital and interest rate derivatives markets. These rates are important benchmark rates in South Africa and are used in determining the reset rate for over-the-counter swaps and forward-rate agreements. The JSE has, since the inception of Jibar, assumed the responsibility for providing the operational infrastructure in order to calculate and release the Jibar on a daily basis. The SARB is the administration agent for Jibar and monitors compliance with the Jibar Code of Conduct.
View Jibar rates here: Current market rates
The Sabor is a transactions-based benchmark rate for commercial banks’ overnight cost of funds. Sabor is made up of three components. As such, contributing commercial banks are required to submit transactions-based data on overnight funding raised in the interbank market, call deposit funding raised from their top 20 non-bank corporate clients and overnight funding raised from the foreign exchange forward market. Sabor enhances transparency and price discovery and serves as a reliable indicator of liquidity conditions in the overnight market. The SARB, in its capacity as administrator, calculates and publishes the Sabor rate on a daily basis.
View Sabor rates here: Current market rates
The MMIS is used for the electronic submitting of bids for Repos, Reverse repos, Treasury bills and SARB debentures.
For more information contact:
Alfred Erasmus: +27 12 313-4472
Francois Aylward: +27 12 313-4085
Louise Saayman: +27 12 313-4353
The Basel III liquidity framework requires banks to adhere to a new liquidity coverage ratio. In 2012, the Bank approved the provision of a committed liquidity facility (CLF) to commercial banks to assist them in meeting their liquidity coverage ratio (LCR) as there is limited availability of high quality liquidity assets (HQLA) in South Africa. The Bank also approved that statutory cash reserves could be included in the banks’ high-quality liquidity assets (HQLA) in order to calculate the liquidity coverage ratio . Initially, the BSD published Guidance Note 5/2012 on 10 May 2012, which detailed acceptable collateral for the CLF. On 2 August 2013, it then issued Guidance Note 6/2013 (GN 6/2013) which contained revised guidelines in terms of acceptable collateral for the CLF, as well as further details on the operational requirements thereof. The Financial Markets Department formalised the operational arrangements pertaining to both the application for a CLF by a bank as well as draw-down on the facility in an Addendum to the Operational Notice.
The SARB has implemented a new framework that introduces modifications to the Bank’s accommodation/refinancing system in order to promote a more active domestic money market in South Africa.
If you have further questions about monetary policy implementation framework, please do not hesitate to contact us.