Our website has detected that you are using an outdated browser that will prevent you from accessing certain features. An upgrade is recommended to improve you browsing experience.
Research agenda 2021-2023
Policy design and effective implementation require high quality research, to identify unintended consequences of policy changes early and ensure policy decisions are based on accurate information.
This document sets out the South African Reserve Bank Research Agenda over the period 2021 to 2023. The agenda aims to generate research informing the three key policy areas of the Bank: monetary policy, macroprudential and micro-prudential policies.
The basic research questions are:
Poverty, inequality and unemployment remain the three main interlinked challenges facing the South African economy and society. While structural reforms are required to address them effectively, macro-economic policy design and coordination should support structural changes by providing a stable, growth-friendly, macroeconomic environment (Loewald, Faulkner, and Makrelov 2020). The projects in this research agenda aim to enhance macroeconomic policy design and coordination across the Bank’s policy areas, improve the depth, operation and stability of the financial sector, and work towards a stable, growth-friendly, macroeconomic environment.
The agenda has four main research areas, which are based on the 2020-2025 Strategic Plan. These are:
1. Enhancing monetary policy efficiency
2. Assessing climate change and the implications for the Bank
3. Macroeconomic Policy Coordination
4. Promoting financial inclusion and competition in the financial sector
Research areas discussed in detail below.
Theme 1: Enhancing monetary policy efficiency
Assess the effectiveness of the monetary policy transmission mechanism and the appropriate implementation framework
Structural drivers of policy transmission
The monetary policy implementation framework
Review the exchange rate policy and strategy for reserve use and accumulation
Enhancing the forecasting and policy analysis tools
Theme 2: Assess climate change and the implications for the Bank
Structural changes
Implications for monetary policy
Financial stability and regulation
Market development
Tools for analysis
Theme 3: Topics in macroeconomic policy coordination
The relationship between monetary and macro-prudential policy
The impacts of fiscal policy on monetary policy transmission and financial markets and stability
Global economic and policy changes and implications for South Africa’s macroeconomic policy
Theme 4: Promoting financial inclusion and competition in the financial sector
To download the full Research Agenda pdf
click here
The research agenda will remain centred on a set of questions about monetary policy itself, the target, the policy framework and the effectiveness of communications.[1] Ongoing work is needed to assess monetary policy in the context of broader macroeconomic considerations, including the role of fiscal policy and other macroeconomic developments and the trajectory of the global economy and global conditions and policy settings.
A macroeconomic balance framework has been worked on and will be further developed to provide a broader context for assessing required policy frameworks and the setting of policy.[2]
Critical to that framework, and the utility of the Taylor Rule embedded in the QPM model, is the role of real exchange rate and its position relative to equilibrium. Equilibrium itself however shifts due to trends in relative prices, technology and productivity, and other structural factors, and needs to be assessed on an ongoing basis. Some work has been done on this in 2020 and will be taken forward in 2021.[3]
Global conditions, global and domestic policy and time preference of consumption, and relative productivity of the domestic economy, among other factors, jointly determine equilibrium neutral real interest rates. Assessing changes to the neutral real interest rates will remain central to the research agenda and the effectiveness of monetary policy.[4]
Finally, while a Taylor Rule provides a macroeconomic perspective for setting policy rates, the detail of inflation dynamics, trends, and secular developments are critical to understanding movements within the CPI and PPI series and appropriate policy setting, and will remain an ongoing and core area of research, analysis and commentary.[5]
[1] Recent work on effectiveness of communication includes Reid and Siklos (2020) and Coco and Viegi (2020). The papers suggest that effectiveness of communication has improved and identify some important elements to be considered in the bank’s monetary policy communication strategy.
[2] See Loewald, Faulkner, and Makrelov (2020)
[3] See for example Rapapali and Steenkamp (2019) and Greenwood-Nimmo, Steenkamp, and van Jaarsveldt (2020)
[4] For previous work on neutral interest rates, see Loewald (2018) and Kuhn, Ruch, and Steinbach (2019).
[5] Several Reserve Bank working papers identify important structural drivers of inflation. For example, over the longer term, inflation dynamics in South Africa mainly reflect structural factors such as labour market rigidities and their impact on unit labour cost, growth in government expenditure, market power and terms of trade dynamics (Dadam and Viegi 2015; Fedderke and Liu 2016). Ruch, Rankin, and Du Plessis (2016) decompose goods inflation into a flexible and sticky-price inflation measure for South Africa at a product level from 2008 to 2015. They find that flexible-price inflation is more volatile, less persistent, and contributes the most to volatility in overall goods inflation. Sticky-price inflation is more persistent, less volatile and correlates well with future goods inflation. The relationship between the output gap and inflation is important for monetary policy analysis. The bank has improved its estimates of potential growth by using new and more advanced estimation techniques (Botha, Ruch, and Steinbach 2018; Fedderke and Mengisteab 2017). Philips curve estimates have consistently shown weak relationship between the output gap and inflation, mainly explained by labour market rigidities (Botha, Kuhn, and Steenkamp 2020; Dadam and Viegi 2015; Fedderke and Liu 2016; Kabundi, Schaling, and Some 2016). The relationship between the exchange rate and inflation has also been studied with the results suggesting a decreasing pass through (Botha 2014; Kabundi and Mbelu 2018; Kabundi and Mlachila 2018). New estimates of exchange rate pass through need to take into account how market structures of particular sectors, nature of domestic economic shocks, exchange rate hedging, government tax decisions, currency pricing of invoices and business cycle dynamics affect the pass-through.
Assess the effectiveness of the monetary policy transmission mechanism and the appropriate implementation framework
Measuring and understanding the effectiveness of the transmission mechanism is critical to monetary policy. Transmission however changes over time and under the impact of various long-term developments, including in global and South African natural interest rates, shifts in inflation expectations, changes in exchange rate pass through, greater integration with global financial markets and the introduction of Basel III regulatory requirements. Cyclical and one-off factors can also play an important role, such as tightening financial conditions and heightened uncertainty since the GFC or the Reserve Bank’s liquidity interventions and bond purchase programme since the Covid-crisis.
This research theme explores how these development have affected optimal monetary policy settings and the transmission of monetary policy, and develops recommendations to enhance the implementation framework.
The proposed programme is divided according to the following sub-themes:
1. Structural drivers of policy transmission
2. The monetary policy framework
Structural drivers of policy transmission
A key question in monetary policy is the degree of pass-through from changes in policy rates to lending rates, and understanding conditions under which transmission changes. There are several factors that affect and determine this relationship. For example, changes in risk premia, bank margins and demand for loans can affect the pass through. Rapapali and Steenkamp (2020) show a significant increase in the funding costs of banks that affects their lending rates but is unrelated to repo rate changes. In addition, market operations can also affect the transmission of repo rate changes. Another important driver of monetary policy transmission is competition among banks, in particular the way in which it affects the supply of credit and the sensitivity of supply to changes in policy. Structural factors such as labour markets dynamics also affect the transmission mechanism. For example the responsiveness of wage settings to economic shocks impact the transmission process (Bhattarai 2016).
The key questions to be addressed under this topic include:
1. What has been the pass through to different lending rates post GFC?
2. What factors have affected the transmission mechanism and how? The analysis will be split into three groups of structural factors:
a. Global factors such as cross border funding.
b. Domestic policy factors including government policy decisions and SARB interventions to sterilise or provide liquidity in domestic markets and,
c. Other factors such as structural changes to financial markets driven for example by changes in competition and technological innovation.
3. What structural changes are required to improve the transmission mechanism?
The monetary policy implementation framework
Understanding the transmission mechanism and the efficiency of domestic liquidity management is key to assessing the effectiveness of the monetary policy framework and to further provide insight into how monetary policy should be implemented. In practice, policy is implemented through rate setting and actions across several connected but separate markets, including for repurchase agreements, other short-term money, and currency markets, obscuring the clarity of policy signals and from time to time generating inefficiencies and volatility in specific markets. In addition, the absence of representative benchmark rates makes it difficult to accurately assess the consistency of monetary policy implementation with the stance of monetary policy as decided by the Monetary Policy Committee or the extent to which overnight rates reflect repurchase rate settings.
This area of the research programme will assess the optimality of the SARB liquidity management policy and the current reference rate proposal and develop recommendations to enhance the monetary policy framework. It will draw on research produced across the research agenda, including, for instance, work coming out of the climate change research stream.
Key questions include:
1. Are the current liquidity management policies optimal?
2. What are the costs and benefits of the current reference rate proposal?
3. What changes are required to the monetary policy framework to increase its effectiveness?
Review the exchange rate policy and strategy for reserve use and accumulation
The exchange rate shapes macroeconomic outcomes and is impacted by endogenous and exogenous shocks. In a floating rate system, the exchange rate tends to be more volatile, but also less of a shock transmitter and more of a shock absorber. This is also reflected in higher short-term volatility but more stability over the longer term, which is important for trade and investment (Hassan 2015). Nonetheless, many countries seek ways to manage exchange rates to reduce costs of floating and enhance benefits of greater stability (such as very large currency swings and external sector pricing volatility), although success in these efforts depends critically on starting conditions and endowments, such as depth of local currency funding markets, savings-investment balances, patterns of trade and trade invoicing, degree of price rigidities, among other factors.
This part of the agenda will look at our floating rate policy empirically and in comparative perspective to understand how policy can be improved. It will develop policy recommendations on exchange rate policy and reserve accumulation.
Key questions include:
1. Should there be more intervention, and of what kind in exchange rate markets? What externalities or market failures can be identified and addressed by interventions?
2. Should reserve purchases and sales be conducted and should other interventions be considered, such as to address short-run currency volatility, preparing for asset sales by investors (insurance) and/or to increase financial stability?
3. What are the implications or unintended consequences of these interventions for other policy areas?
Enhancing the forecasting and policy analysis tools
Since the implementation of the inflation targeting framework almost two decades ago, the Bank has, in line with best practice, developed a suite of models as part of its modelling and forecasting strategy. Continuous review, development and enhancement of the existing suite of models is critical to ensuring their relevance. Globally, most central banks employ a suite of models to fulfil their mandates. Usually, the suite of models includes some short-term as well as assumptions models using ARIMA or now-casting techniques and medium term forecasting tools such as large econometric and DSGE models. Rapid development of storage capacity and increases in computational power have allowed for the creation of large data sets and their analysis, opening new frontiers in our ability to forecast and analyse economic activity and behaviour. Over the period 2021 -2023 , the Bank will develop capacity and build models based on large and high frequency data sets.
We propose steps to explore new model directions, strengthen our array of models to solve specific forecasting concerns, and ensure we have the technical capacity and processes to keep existing models, especially the QPM, up to date. The discussion below reflects learnings from the reviews conducted by ERD of research activities, the use of the QPM in the forecasting process, and a process engineering exercise of the forecasting process, all completed in the last two years.
As central banks began to shift across to Dynamic Stochastic General Equilibrium (DSGE) models, we developed a simpler policy model, the QPM model, that became the main forecasting model in September 2017. While this model has no direct “links” to the data (similar to the DSGE type), its strength rests in a more consistent theoretical framework with forward-looking expectations and a “Taylor-rule” to close the model in a general equilibrium system. The QPM generates medium-term projections which enable policy makers to communicate and quantify how policy rates impact on inflation and the economy generally.
The 2018 peer review recommended enhancements to the original structure of the model as well as extensions to the model. Recommendations were also made on improving the QPM forecast performance and on how it could be used to improve communication by the MPC. These enhancements include: improving the expectations channel in the model; improving the labour market channel to include some of SA’s labour characteristics; getting rid of some of the embedded AR processes; and subjecting the Monetary Policy Rule to sensitivity analysis and testing. These will be addressed over the next two years.
The Core model is a medium-term forecasting model, estimated using the Error-Correction Model (ECM) technique. The basic problem with the core model has been its lack of a monetary policy rule that shapes the trajectory of the economy through the forecast. Its strength has been the detail of the price formation channel, the disaggregation of the different drivers of expenditure components of growth (GDP), and a forecasted balance of payments. The core model also provides alternative simulations outside the scope of the QPM, such as the fiscal scenarios and is also used to perform stress testing. Core models are being superseded by next generation hybrid models, which more explicitly combine core components with long-run equilibriums, forward-looking expectations and policy rules. ERD intends to develop a hybrid model, using the core model as a foundation, and as a key initiative in the ERSA-run modelling network.
Rising fiscal risks and the 2008 financial crisis have highlighted the importance of incorporating fiscal and financial dynamics in economic models. These dynamics will be developed in the current and new models over the period 2021-2023.
Other models also need development, such as Integrated Assessment models to develop climate change scenarios and a dynamic stochastic general equilibrium model (DSGE) with fiscal and financial dynamics that will be pursued through support to the ERSA model network.
Assess the effectiveness of the monetary policy transmission mechanism and the appropriate implementation framework
Measuring and understanding the effectiveness of the transmission mechanism is critical to monetary policy. Transmission however changes over time and under the impact of various long-term developments, including in global and South African natural interest rates, shifts in inflation expectations, changes in exchange rate pass through, greater integration with global financial markets and the introduction of Basel III regulatory requirements. Cyclical and one-off factors can also play an important role, such as tightening financial conditions and heightened uncertainty since the GFC or the Reserve Bank’s liquidity interventions and bond purchase programme since the Covid-crisis.
This research theme explores how these development have affected optimal monetary policy settings and the transmission of monetary policy, and develops recommendations to enhance the implementation framework.
The proposed programme is divided according to the following sub-themes:
1. Structural drivers of policy transmission
2. The monetary policy framework
Structural drivers of policy transmission
A key question in monetary policy is the degree of pass-through from changes in policy rates to lending rates, and understanding conditions under which transmission changes. There are several factors that affect and determine this relationship. For example, changes in risk premia, bank margins and demand for loans can affect the pass through. Rapapali and Steenkamp (2020) show a significant increase in the funding costs of banks that affects their lending rates but is unrelated to repo rate changes. In addition, market operations can also affect the transmission of repo rate changes. Another important driver of monetary policy transmission is competition among banks, in particular the way in which it affects the supply of credit and the sensitivity of supply to changes in policy. Structural factors such as labour markets dynamics also affect the transmission mechanism. For example the responsiveness of wage settings to economic shocks impact the transmission process (Bhattarai 2016).
The key questions to be addressed under this topic include:
1. What has been the pass through to different lending rates post GFC?
2. What factors have affected the transmission mechanism and how? The analysis will be split into three groups of structural factors:
a. Global factors such as cross border funding.
b. Domestic policy factors including government policy decisions and SARB interventions to sterilise or provide liquidity in domestic markets and,
c. Other factors such as structural changes to financial markets driven for example by changes in competition and technological innovation.
3. What structural changes are required to improve the transmission mechanism?
The monetary policy implementation framework
Understanding the transmission mechanism and the efficiency of domestic liquidity management is key to assessing the effectiveness of the monetary policy framework and to further provide insight into how monetary policy should be implemented. In practice, policy is implemented through rate setting and actions across several connected but separate markets, including for repurchase agreements, other short-term money, and currency markets, obscuring the clarity of policy signals and from time to time generating inefficiencies and volatility in specific markets. In addition, the absence of representative benchmark rates makes it difficult to accurately assess the consistency of monetary policy implementation with the stance of monetary policy as decided by the Monetary Policy Committee or the extent to which overnight rates reflect repurchase rate settings.
This area of the research programme will assess the optimality of the SARB liquidity management policy and the current reference rate proposal and develop recommendations to enhance the monetary policy framework. It will draw on research produced across the research agenda, including, for instance, work coming out of the climate change research stream.
Key questions include:
1. Are the current liquidity management policies optimal?
2. What are the costs and benefits of the current reference rate proposal?
3. What changes are required to the monetary policy framework to increase its effectiveness?
Review the exchange rate policy and strategy for reserve use and accumulation
The exchange rate shapes macroeconomic outcomes and is impacted by endogenous and exogenous shocks. In a floating rate system, the exchange rate tends to be more volatile, but also less of a shock transmitter and more of a shock absorber. This is also reflected in higher short-term volatility but more stability over the longer term, which is important for trade and investment (Hassan 2015). Nonetheless, many countries seek ways to manage exchange rates to reduce costs of floating and enhance benefits of greater stability (such as very large currency swings and external sector pricing volatility), although success in these efforts depends critically on starting conditions and endowments, such as depth of local currency funding markets, savings-investment balances, patterns of trade and trade invoicing, degree of price rigidities, among other factors.
This part of the agenda will look at our floating rate policy empirically and in comparative perspective to understand how policy can be improved. It will develop policy recommendations on exchange rate policy and reserve accumulation.
Key questions include:
1. Should there be more intervention, and of what kind in exchange rate markets? What externalities or market failures can be identified and addressed by interventions?
2. Should reserve purchases and sales be conducted and should other interventions be considered, such as to address short-run currency volatility, preparing for asset sales by investors (insurance) and/or to increase financial stability?
3. What are the implications or unintended consequences of these interventions for other policy areas?
Enhancing the forecasting and policy analysis tools
Since the implementation of the inflation targeting framework almost two decades ago, the Bank has, in line with best practice, developed a suite of models as part of its modelling and forecasting strategy. Continuous review, development and enhancement of the existing suite of models is critical to ensuring their relevance. Globally, most central banks employ a suite of models to fulfil their mandates. Usually, the suite of models includes some short-term as well as assumptions models using ARIMA or now-casting techniques and medium term forecasting tools such as large econometric and DSGE models. Rapid development of storage capacity and increases in computational power have allowed for the creation of large data sets and their analysis, opening new frontiers in our ability to forecast and analyse economic activity and behaviour. Over the period 2021 -2023 , the Bank will develop capacity and build models based on large and high frequency data sets.
We propose steps to explore new model directions, strengthen our array of models to solve specific forecasting concerns, and ensure we have the technical capacity and processes to keep existing models, especially the QPM, up to date. The discussion below reflects learnings from the reviews conducted by ERD of research activities, the use of the QPM in the forecasting process, and a process engineering exercise of the forecasting process, all completed in the last two years.
As central banks began to shift across to Dynamic Stochastic General Equilibrium (DSGE) models, we developed a simpler policy model, the QPM model, that became the main forecasting model in September 2017. While this model has no direct “links” to the data (similar to the DSGE type), its strength rests in a more consistent theoretical framework with forward-looking expectations and a “Taylor-rule” to close the model in a general equilibrium system. The QPM generates medium-term projections which enable policy makers to communicate and quantify how policy rates impact on inflation and the economy generally.
The 2018 peer review recommended enhancements to the original structure of the model as well as extensions to the model. Recommendations were also made on improving the QPM forecast performance and on how it could be used to improve communication by the MPC. These enhancements include: improving the expectations channel in the model; improving the labour market channel to include some of SA’s labour characteristics; getting rid of some of the embedded AR processes; and subjecting the Monetary Policy Rule to sensitivity analysis and testing. These will be addressed over the next two years.
The Core model is a medium-term forecasting model, estimated using the Error-Correction Model (ECM) technique. The basic problem with the core model has been its lack of a monetary policy rule that shapes the trajectory of the economy through the forecast. Its strength has been the detail of the price formation channel, the disaggregation of the different drivers of expenditure components of growth (GDP), and a forecasted balance of payments. The core model also provides alternative simulations outside the scope of the QPM, such as the fiscal scenarios and is also used to perform stress testing. Core models are being superseded by next generation hybrid models, which more explicitly combine core components with long-run equilibriums, forward-looking expectations and policy rules. ERD intends to develop a hybrid model, using the core model as a foundation, and as a key initiative in the ERSA-run modelling network.
Rising fiscal risks and the 2008 financial crisis have highlighted the importance of incorporating fiscal and financial dynamics in economic models. These dynamics will be developed in the current and new models over the period 2021-2023.
Other models also need development, such as Integrated Assessment models to develop climate change scenarios and a dynamic stochastic general equilibrium model (DSGE) with fiscal and financial dynamics that will be pursued through support to the ERSA model network.
Climate change mitigation and adaptation will have far-reaching, rapid, and most likely irreversible structural implications. Without concerted global mitigation policy, mean projected warming is expected to rise by 3 to 4 °Cover pre-industrial levels by 2100 with great costs to human and ecological welfare and stress to economic and social systems (IEA 2014; IPCC 2014, 2018).
Price and quantity adjustments of various kinds – energy costs, pricing of externalities, growing demand for more energy-efficient products and services, revaluation of real and financial assets among others – will be far-reaching. Importantly, emerging and developing economies are almost invariably more exposed to climate change events because of the relatively large size of directly weather-dependent sectors (e.g., agriculture and food-processing). They also are typically less resilient to climate change events and more vulnerable to associated shocks (Farid et al. 2016).
Adaptation and mitigation measures need to be economically and financially manageable in the sense of minimizing damage and maximizing opportunity. This also requires dealing with possible adverse impacts on particular groups and ensuring that any transition to a lower emissions development path is inclusive.
“Green swan events” will be disruptive, with complex transmission mechanisms, including and perhaps primarily through financial systems, and as stress impacts on firm and household balance sheets and risk profiles, with implications for financial stability (Carney 2015) .[1] The breadth of policy risk faced by central banks in dealing with climate change and its consequences is becoming increasingly apparent.
The CDRC has initiated a structured programme into climate change and its implications for the Reserve Bank:
1. Understand the impacts and design policy responses to improve the bank’s ability to deliver on its key mandates and support the transition.
2. Engage more effectively in domestic and international forums, which aim to inform decision-makers and develop responses to climate change risks, and
3. Avoid bad solutions that might be imposed on the Reserve Bank due to lack of appropriate analytical work.
The programme is divided according to the following groups: structural changes, implications for monetary policy, financial stability and regulation, market development, and tools for analysis.
[1] Bolton et al. (2020)
Climate change and its structural implications will generate transition and physical risks, changing the composition of growth and inflation dynamics in the economy. These, in turn, will generate balance sheets effects and impact the monetary policy and financial stability mandates of the Reserve Bank. These structural changes will also be driven by policy decisions in other countries, which affect the global economy and the flow of capital into emerging markets. Understanding these structural changes and developing interventions to minimise negative impacts, while supporting positive developments, can improve economic and financial resilience and support low and stable inflation. Some of the key questions under this topic are:
1. What are the likely policy developments (in the areas of climate change) in the global economy and how would they affect the domestic economy?
2. What is the impact of climate change on particular sectors of the economy and the composition of economic growth?
3. How would mitigation efforts impact the economy?
4. What are the major technological developments over the next 2 to 10 years that are likely to generate large transition risks?
The economic literature on climate change is evolving rapidly and some of these questions are being addressed. In this case, the role of Reserve Bank staff will be to stay abreast of developments and identify implications for Reserve Bank policy.
Implications for monetary policy
The projects under this heading aim to inform the monetary policy responses to physical and transition risks, and identify possible mechanisms to support the transition.
1. What is the optimal response for central banks to large supply side output and price shocks, especially those which are long lasting or permanent?
2. How is inflation affected by different instruments that aim to facilitate the transition to a greener economy?
3. How can central banks use their balance sheets to support the transition?
4. How should the Reserve Bank support the flow of green funding?
5. What are the implications of using green assets as collateral in SARB’s market operations?
Financial stability and regulation
Physical and transition risks will impact the balance sheets of financial sector institutions. Some of the key questions are:
1. How should banks and insurers manage physical and transition risks?
2. What are the impacts of different climate change paths and events on financial stability?
3. What changes are required to micro and macro-prudential policy to facilitate the transition and manage physical risks, particularly to reduce the relative cost of ‘green’ funding to ‘brown’ funding? What are their costs and benefits?
4. How should we manage currency risks (or macroeconomic and financial imbalances) associated with a large demand for foreign capital to finance investment in mitigation and adjustment?
5. What are the effects on financial stability arising from unevenness of climate change effects between global regions and countries and transmitted through exchange rates and international capital markets? How can financial stability policy take them into account?
Many of these questions are part of the research agenda of other central banks, the Network for Greening of the Financial Sector and the National Treasury. It is important for the Reserve Bank to develop its own thinking around these questions in order to engage effectively in local and international debates.
South Africa has had some success with developing green markets, which, from a central banking perspective, should be welcomed as a broadening and deepening of local debt markets.[1] Market liquidity and reduced transaction costs should be supported, alongside strengthening green certification systems and issuers’ accountability for the use of funds raised. Market development needs risk-sharing mechanisms to improve agents’ risk/return calculations. South Africa has substantial experience of risk-sharing between the public and private sectors (through, for example, credit and debt guarantees, public-private partnerships). These mechanisms will be important in mobilizing private sector funding and providing appropriate insurance products.
Key questions under this area include:
1. How do we improve market liquidity and reduce transaction costs?
2. How do we develop green certification systems and issuers’ accountability for the use of funds raised?
3. What mechanisms can be put in place to improve risk sharing?
4. What forms of market regulation will be appropriate for markets in green instruments?
[1] The JSE’s Green Bond Segment, established in 2014, has raised finance for green investment by municipalities (e.g. Cape Town and Johannesburg) and corporations (e.g. Greenpoint and Nedbank).
Central banks will be required to develop new tools to assess the impacts of climate change related events on the economy as well as the resilience of the financial system. The Network for Greening the Financial System is already developing scenarios, using Integrated Assessment Models (IAMs). These, however, do not provide physical risk estimates and region-specific impacts. In South Africa, it is expected that climate change will impact different regions very differently. On an aggregate level, the impacts appear small, but at the regional level the impacts can be large.
The Reserve Bank needs to develop its capacity to analyse the transitional and physical impacts of climate change. As mentioned before, these cut across the different mandates of the Reserve Bank. IAM frameworks or other similar models, for instance, are complex and require specialised skills. This part of the agenda will aim to evaluate and develop these tools, or simpler alternatives, in collaboration with institutions that have relevant technical expertise, taking into account the approaches of other central banks.
Topics of interest include:
1. Modification of IAM frameworks to cater for the needs of central banks.
2. Linking central banks’ models to climate change impact models.
3. Developing physical risk models.
4. Developing local modelling networks.
5. Micro level models to provide granular impacts on specific financial institutions.
Assess the effectiveness of the monetary policy transmission mechanism and the appropriate implementation framework
Measuring and understanding the effectiveness of the transmission mechanism is critical to monetary policy. Transmission however changes over time and under the impact of various long-term developments, including in global and South African natural interest rates, shifts in inflation expectations, changes in exchange rate pass through, greater integration with global financial markets and the introduction of Basel III regulatory requirements. Cyclical and one-off factors can also play an important role, such as tightening financial conditions and heightened uncertainty since the GFC or the Reserve Bank’s liquidity interventions and bond purchase programme since the Covid-crisis.
This research theme explores how these development have affected optimal monetary policy settings and the transmission of monetary policy, and develops recommendations to enhance the implementation framework.
The proposed programme is divided according to the following sub-themes:
1. Structural drivers of policy transmission
2. The monetary policy framework
Structural drivers of policy transmission
A key question in monetary policy is the degree of pass-through from changes in policy rates to lending rates, and understanding conditions under which transmission changes. There are several factors that affect and determine this relationship. For example, changes in risk premia, bank margins and demand for loans can affect the pass through. Rapapali and Steenkamp (2020) show a significant increase in the funding costs of banks that affects their lending rates but is unrelated to repo rate changes. In addition, market operations can also affect the transmission of repo rate changes. Another important driver of monetary policy transmission is competition among banks, in particular the way in which it affects the supply of credit and the sensitivity of supply to changes in policy. Structural factors such as labour markets dynamics also affect the transmission mechanism. For example the responsiveness of wage settings to economic shocks impact the transmission process (Bhattarai 2016).
The key questions to be addressed under this topic include:
1. What has been the pass through to different lending rates post GFC?
2. What factors have affected the transmission mechanism and how? The analysis will be split into three groups of structural factors:
a. Global factors such as cross border funding.
b. Domestic policy factors including government policy decisions and SARB interventions to sterilise or provide liquidity in domestic markets and,
c. Other factors such as structural changes to financial markets driven for example by changes in competition and technological innovation.
3. What structural changes are required to improve the transmission mechanism?
The monetary policy implementation framework
Understanding the transmission mechanism and the efficiency of domestic liquidity management is key to assessing the effectiveness of the monetary policy framework and to further provide insight into how monetary policy should be implemented. In practice, policy is implemented through rate setting and actions across several connected but separate markets, including for repurchase agreements, other short-term money, and currency markets, obscuring the clarity of policy signals and from time to time generating inefficiencies and volatility in specific markets. In addition, the absence of representative benchmark rates makes it difficult to accurately assess the consistency of monetary policy implementation with the stance of monetary policy as decided by the Monetary Policy Committee or the extent to which overnight rates reflect repurchase rate settings.
This area of the research programme will assess the optimality of the SARB liquidity management policy and the current reference rate proposal and develop recommendations to enhance the monetary policy framework. It will draw on research produced across the research agenda, including, for instance, work coming out of the climate change research stream.
Key questions include:
1. Are the current liquidity management policies optimal?
2. What are the costs and benefits of the current reference rate proposal?
3. What changes are required to the monetary policy framework to increase its effectiveness?
Review the exchange rate policy and strategy for reserve use and accumulation
The exchange rate shapes macroeconomic outcomes and is impacted by endogenous and exogenous shocks. In a floating rate system, the exchange rate tends to be more volatile, but also less of a shock transmitter and more of a shock absorber. This is also reflected in higher short-term volatility but more stability over the longer term, which is important for trade and investment (Hassan 2015). Nonetheless, many countries seek ways to manage exchange rates to reduce costs of floating and enhance benefits of greater stability (such as very large currency swings and external sector pricing volatility), although success in these efforts depends critically on starting conditions and endowments, such as depth of local currency funding markets, savings-investment balances, patterns of trade and trade invoicing, degree of price rigidities, among other factors.
This part of the agenda will look at our floating rate policy empirically and in comparative perspective to understand how policy can be improved. It will develop policy recommendations on exchange rate policy and reserve accumulation.
Key questions include:
1. Should there be more intervention, and of what kind in exchange rate markets? What externalities or market failures can be identified and addressed by interventions?
2. Should reserve purchases and sales be conducted and should other interventions be considered, such as to address short-run currency volatility, preparing for asset sales by investors (insurance) and/or to increase financial stability?
3. What are the implications or unintended consequences of these interventions for other policy areas?
Enhancing the forecasting and policy analysis tools
Since the implementation of the inflation targeting framework almost two decades ago, the Bank has, in line with best practice, developed a suite of models as part of its modelling and forecasting strategy. Continuous review, development and enhancement of the existing suite of models is critical to ensuring their relevance. Globally, most central banks employ a suite of models to fulfil their mandates. Usually, the suite of models includes some short-term as well as assumptions models using ARIMA or now-casting techniques and medium term forecasting tools such as large econometric and DSGE models. Rapid development of storage capacity and increases in computational power have allowed for the creation of large data sets and their analysis, opening new frontiers in our ability to forecast and analyse economic activity and behaviour. Over the period 2021 -2023 , the Bank will develop capacity and build models based on large and high frequency data sets.
We propose steps to explore new model directions, strengthen our array of models to solve specific forecasting concerns, and ensure we have the technical capacity and processes to keep existing models, especially the QPM, up to date. The discussion below reflects learnings from the reviews conducted by ERD of research activities, the use of the QPM in the forecasting process, and a process engineering exercise of the forecasting process, all completed in the last two years.
As central banks began to shift across to Dynamic Stochastic General Equilibrium (DSGE) models, we developed a simpler policy model, the QPM model, that became the main forecasting model in September 2017. While this model has no direct “links” to the data (similar to the DSGE type), its strength rests in a more consistent theoretical framework with forward-looking expectations and a “Taylor-rule” to close the model in a general equilibrium system. The QPM generates medium-term projections which enable policy makers to communicate and quantify how policy rates impact on inflation and the economy generally.
The 2018 peer review recommended enhancements to the original structure of the model as well as extensions to the model. Recommendations were also made on improving the QPM forecast performance and on how it could be used to improve communication by the MPC. These enhancements include: improving the expectations channel in the model; improving the labour market channel to include some of SA’s labour characteristics; getting rid of some of the embedded AR processes; and subjecting the Monetary Policy Rule to sensitivity analysis and testing. These will be addressed over the next two years.
The Core model is a medium-term forecasting model, estimated using the Error-Correction Model (ECM) technique. The basic problem with the core model has been its lack of a monetary policy rule that shapes the trajectory of the economy through the forecast. Its strength has been the detail of the price formation channel, the disaggregation of the different drivers of expenditure components of growth (GDP), and a forecasted balance of payments. The core model also provides alternative simulations outside the scope of the QPM, such as the fiscal scenarios and is also used to perform stress testing. Core models are being superseded by next generation hybrid models, which more explicitly combine core components with long-run equilibriums, forward-looking expectations and policy rules. ERD intends to develop a hybrid model, using the core model as a foundation, and as a key initiative in the ERSA-run modelling network.
Rising fiscal risks and the 2008 financial crisis have highlighted the importance of incorporating fiscal and financial dynamics in economic models. These dynamics will be developed in the current and new models over the period 2021-2023.
Other models also need development, such as Integrated Assessment models to develop climate change scenarios and a dynamic stochastic general equilibrium model (DSGE) with fiscal and financial dynamics that will be pursued through support to the ERSA model network.
Assess the effectiveness of the monetary policy transmission mechanism and the appropriate implementation framework
Measuring and understanding the effectiveness of the transmission mechanism is critical to monetary policy. Transmission however changes over time and under the impact of various long-term developments, including in global and South African natural interest rates, shifts in inflation expectations, changes in exchange rate pass through, greater integration with global financial markets and the introduction of Basel III regulatory requirements. Cyclical and one-off factors can also play an important role, such as tightening financial conditions and heightened uncertainty since the GFC or the Reserve Bank’s liquidity interventions and bond purchase programme since the Covid-crisis.
This research theme explores how these development have affected optimal monetary policy settings and the transmission of monetary policy, and develops recommendations to enhance the implementation framework.
The proposed programme is divided according to the following sub-themes:
1. Structural drivers of policy transmission
2. The monetary policy framework
Structural drivers of policy transmission
A key question in monetary policy is the degree of pass-through from changes in policy rates to lending rates, and understanding conditions under which transmission changes. There are several factors that affect and determine this relationship. For example, changes in risk premia, bank margins and demand for loans can affect the pass through. Rapapali and Steenkamp (2020) show a significant increase in the funding costs of banks that affects their lending rates but is unrelated to repo rate changes. In addition, market operations can also affect the transmission of repo rate changes. Another important driver of monetary policy transmission is competition among banks, in particular the way in which it affects the supply of credit and the sensitivity of supply to changes in policy. Structural factors such as labour markets dynamics also affect the transmission mechanism. For example the responsiveness of wage settings to economic shocks impact the transmission process (Bhattarai 2016).
The key questions to be addressed under this topic include:
1. What has been the pass through to different lending rates post GFC?
2. What factors have affected the transmission mechanism and how? The analysis will be split into three groups of structural factors:
a. Global factors such as cross border funding.
b. Domestic policy factors including government policy decisions and SARB interventions to sterilise or provide liquidity in domestic markets and,
c. Other factors such as structural changes to financial markets driven for example by changes in competition and technological innovation.
3. What structural changes are required to improve the transmission mechanism?
The monetary policy implementation framework
Understanding the transmission mechanism and the efficiency of domestic liquidity management is key to assessing the effectiveness of the monetary policy framework and to further provide insight into how monetary policy should be implemented. In practice, policy is implemented through rate setting and actions across several connected but separate markets, including for repurchase agreements, other short-term money, and currency markets, obscuring the clarity of policy signals and from time to time generating inefficiencies and volatility in specific markets. In addition, the absence of representative benchmark rates makes it difficult to accurately assess the consistency of monetary policy implementation with the stance of monetary policy as decided by the Monetary Policy Committee or the extent to which overnight rates reflect repurchase rate settings.
This area of the research programme will assess the optimality of the SARB liquidity management policy and the current reference rate proposal and develop recommendations to enhance the monetary policy framework. It will draw on research produced across the research agenda, including, for instance, work coming out of the climate change research stream.
Key questions include:
1. Are the current liquidity management policies optimal?
2. What are the costs and benefits of the current reference rate proposal?
3. What changes are required to the monetary policy framework to increase its effectiveness?
Review the exchange rate policy and strategy for reserve use and accumulation
The exchange rate shapes macroeconomic outcomes and is impacted by endogenous and exogenous shocks. In a floating rate system, the exchange rate tends to be more volatile, but also less of a shock transmitter and more of a shock absorber. This is also reflected in higher short-term volatility but more stability over the longer term, which is important for trade and investment (Hassan 2015). Nonetheless, many countries seek ways to manage exchange rates to reduce costs of floating and enhance benefits of greater stability (such as very large currency swings and external sector pricing volatility), although success in these efforts depends critically on starting conditions and endowments, such as depth of local currency funding markets, savings-investment balances, patterns of trade and trade invoicing, degree of price rigidities, among other factors.
This part of the agenda will look at our floating rate policy empirically and in comparative perspective to understand how policy can be improved. It will develop policy recommendations on exchange rate policy and reserve accumulation.
Key questions include:
1. Should there be more intervention, and of what kind in exchange rate markets? What externalities or market failures can be identified and addressed by interventions?
2. Should reserve purchases and sales be conducted and should other interventions be considered, such as to address short-run currency volatility, preparing for asset sales by investors (insurance) and/or to increase financial stability?
3. What are the implications or unintended consequences of these interventions for other policy areas?
Enhancing the forecasting and policy analysis tools
Since the implementation of the inflation targeting framework almost two decades ago, the Bank has, in line with best practice, developed a suite of models as part of its modelling and forecasting strategy. Continuous review, development and enhancement of the existing suite of models is critical to ensuring their relevance. Globally, most central banks employ a suite of models to fulfil their mandates. Usually, the suite of models includes some short-term as well as assumptions models using ARIMA or now-casting techniques and medium term forecasting tools such as large econometric and DSGE models. Rapid development of storage capacity and increases in computational power have allowed for the creation of large data sets and their analysis, opening new frontiers in our ability to forecast and analyse economic activity and behaviour. Over the period 2021 -2023 , the Bank will develop capacity and build models based on large and high frequency data sets.
We propose steps to explore new model directions, strengthen our array of models to solve specific forecasting concerns, and ensure we have the technical capacity and processes to keep existing models, especially the QPM, up to date. The discussion below reflects learnings from the reviews conducted by ERD of research activities, the use of the QPM in the forecasting process, and a process engineering exercise of the forecasting process, all completed in the last two years.
As central banks began to shift across to Dynamic Stochastic General Equilibrium (DSGE) models, we developed a simpler policy model, the QPM model, that became the main forecasting model in September 2017. While this model has no direct “links” to the data (similar to the DSGE type), its strength rests in a more consistent theoretical framework with forward-looking expectations and a “Taylor-rule” to close the model in a general equilibrium system. The QPM generates medium-term projections which enable policy makers to communicate and quantify how policy rates impact on inflation and the economy generally.
The 2018 peer review recommended enhancements to the original structure of the model as well as extensions to the model. Recommendations were also made on improving the QPM forecast performance and on how it could be used to improve communication by the MPC. These enhancements include: improving the expectations channel in the model; improving the labour market channel to include some of SA’s labour characteristics; getting rid of some of the embedded AR processes; and subjecting the Monetary Policy Rule to sensitivity analysis and testing. These will be addressed over the next two years.
The Core model is a medium-term forecasting model, estimated using the Error-Correction Model (ECM) technique. The basic problem with the core model has been its lack of a monetary policy rule that shapes the trajectory of the economy through the forecast. Its strength has been the detail of the price formation channel, the disaggregation of the different drivers of expenditure components of growth (GDP), and a forecasted balance of payments. The core model also provides alternative simulations outside the scope of the QPM, such as the fiscal scenarios and is also used to perform stress testing. Core models are being superseded by next generation hybrid models, which more explicitly combine core components with long-run equilibriums, forward-looking expectations and policy rules. ERD intends to develop a hybrid model, using the core model as a foundation, and as a key initiative in the ERSA-run modelling network.
Rising fiscal risks and the 2008 financial crisis have highlighted the importance of incorporating fiscal and financial dynamics in economic models. These dynamics will be developed in the current and new models over the period 2021-2023.
Other models also need development, such as Integrated Assessment models to develop climate change scenarios and a dynamic stochastic general equilibrium model (DSGE) with fiscal and financial dynamics that will be pursued through support to the ERSA model network.
Anaya, Pablo, Michael Hachula, and Christian J. Offermanns. 2017. 'Spillovers of U.S. unconventional monetary policy to emerging markets: The role of capital flows', Journal of International Money and Finance, 73: 275-95.
Bhattarai, Keshab. 2016. 'Unemployment–inflation trade-offs in OECD countries', Economic Modelling, 58: 93-103.
Borio, Claudio, and Haibin Zhu. 2012. 'Capital regulation, risk-taking and monetary policy: A missing link in the transmission mechanism?', Journal of Financial Stability, 8: 236-51.
Botha, Byron. 2014. 'Exchange Rate Pass-Through Post-2007.', SARB Economic Notes Series, EN/14/21.
Botha, Byron, Lauren Kuhn, and Daan Steenkamp. 2020. 'Is the Phillips curve framework still useful for understanding inflation dynamics in South Africa?', South Africa Reserve Bank Working Paper Series, WP/20/07.
Botha, Byron, Franz Ruch, and Rudi Steinbach. 2018. 'Short-lived supply shocks to potential growth', South African Reserve Bank Working Paper, No 8/02.
Bussière, Matthieu, Jin Cao, Jakob de Haan, Robert Hills, Simon Lloyd, Baptiste Meunier, Justine Pedrono, Dennis Reinhardt, Sonalika Sinha, and Rhiannon Sowerbutts. 2020. 'The interaction between macroprudential policy and monetary policy: Overview', Review of International Economics.
CGFS. 2016. 'Regulatory change and monetary policy', Committee on the Global Financial System Paper No 54.
Coco, Alberto, and Nicola Viegi. 2020. 'The monetary policy of the South African Reserve Bank: stance,communication and credibility', South Africa Reserve Bank Working Paper Series, No WP/20/06.
Dadam, Vincent, and Nicola Viegi. 2015. 'Labour Market and Monetary Policy in South Africa', Fourteen Years of Inflation Targeting in South Africa the Challenge of a Changing Mandate: 113.
Dell'Ariccia, Giovanni, Caio Ferreira, Nigel Jenkinson, Mr Luc Laeven, Alberto Martin, Camelia Minoiu, and Alex Popov. 2018. Managing the sovereign-bank nexus (International Monetary Fund).
Eller, Markus, Florian Huber, and Helene Schuberth. 2020. 'How important are global factors for understanding the dynamics of international capital flows?', Journal of International Money and Finance, 109: 102221.
Fedderke, J. W., and Yang Liu. 2016. 'Inflation in South Africa: An assessment of Alternative Inflation Models', South Africa Reserve Bank Working Paper Series, No. WP/16/03.
Fedderke, Johannes W, and Daniel K Mengisteab. 2017. 'Estimating South Africa's output gap and potential growth rate', South African Journal of Economics, 85: 161-77.
Gaspar, Vitor, Maurice Obstfeld, Ratna Sahay, Douglas Laxton, Dennis Petrus Johannes Botman, Kevin Clilnton, Romain Duval, Kotaro Ishi, Zoltán Jakab, and Laura Jaramillo. 2016. "Macroeconomic management when policy space is constrained: a comprehensive, consistent and coordinated approach to economic policy." In, edited by IMF Discussion Note SDN/16/09. International Monetary Fund.
Greenwood-Nimmo, M, D Steenkamp, and R van Jaarsveldt 2020. 'Risk and Return Spillovers in a Global Model of the Foreign Exchange Networks', South Africa Reserve Bank Working Paper Series (forthcoming ).
Hassan, Shakill 2015. 'Speculative Flows, Exchange Rate Volatility and Monetary Policy: the South African Experience', South Africa Reserve Bank Working Paper Series, No. WP/15/02.
Kabundi, Alain, and Asi Mbelu. 2018. 'Has the Exchange Rate Pass‐Through changed in South Africa?', South African Journal of Economics, 86: 339-60.
Kabundi, Alain N, and Mr Montfort Mlachila. 2018. 'Monetary policy credibility and exchange rate pass-through in South Africa', South Africa Reserve Bank Working Paper Series, WP/18/04.
Kabundi, Alain, Eric Schaling, and Modeste Some. 2016. 'Estimating a time-varying Phillips curve for South Africa', South African Reserve Bank Working Paper 16/05.
Kuhn, Lauren, Franz Ruch, and Rudi Steinbach. 2019. 'Reaching for the (r)‐stars: estimating South Africa’s neutral real interest rate', South Africa Reserve Bank Working Paper Series, Working Paper No 19/01.
Loewald, Christopher. 2018. "Making sense of neutral real interest rates." In Occasional Bulletin of Economic Notes, edited by South African Reserve Bank. South African Reserve Bank.
Loewald, Christopher, David Faulkner, and Konstantin Makrelov. 2020. 'Time consistency and economic growth-A case study of South African macroeconomic policy', South African Reserve Bank Working Paper No/20/12.
Makrelov, Konstantin, Rob Davies, and Laurence Harris. 2019. 'The impact of capital flow reversal shocks in South Africa: a stock-and flow-consistent analysis', South Africa Reserve Bank Working Paper Series, No1905.
Noss, Joseph, and Priscilla Toffano. 2016. 'Estimating the impact of changes in aggregate bank capital requirements on lending and growth during an upswing', Journal of Banking & Finance, 62: 15-27.
Rapapali, Mpho, and Daan Steenkamp. 2019. 'Is There a Relationship between Productivity and Relative Prices in South Africa?', Journal of Development Perspectives, 3: 164-90.
———. 2020. "Developments in bank funding costs in South Africa." In.
Reid, Monique, and Pierre Siklos. 2020. 'Building Credibility and Influencing Expectations-The Evolution of Central Bank Communication', South Africa Reserve Bank Working Paper Series, No WP/20/08.
Ruch, Franz, Neil Rankin, and Stan Du Plessis. 2016. "Decomposing inflation using micro-price data-Sticky-price inflation." In.
Soobyah, Luchelle, and Daan Steenkamp. 2020. 'Term premium and rate expectation estimates from the South African yield curve', South Africa Reserve Bank Working Paper Series, WP/20/03.
Woodford, Michael 2010. 'Financial intermediation and macroeconomic analysis', Journal of Economic Perspectives, 24: 21-44.
