Chapter

3. Improving Monetary Policy Frameworks

Author(s):
International Monetary Fund
Published Date:
April 2014
Share
  • ShareShare
Show Summary Details

Monetary Policy Landscape in Sub-Saharan Africa

With inflation declining to single digits in many sub-Saharan African countries since the early 2000s, central banks currently face a different set of challenges. In particular, as is the case in other countries, the relationship between money and inflation has become weaker in a low-inflation environment, as countries have opened their capital accounts, attracting capital inflows, and deepened their financial markets. As advanced economies exit from unconventional monetary policies that were designed to provide economic stimulus in the wake of the global financial crisis, their tightening of monetary policy will increase funding costs for frontier market economies and heighten the risk of reversal of capital flows as recently experienced by a number of emerging markets. Against this backdrop, this chapter considers the conduct of monetary policy and its recent evolution in a select group of countries in sub-Saharan Africa, including the key factors driving the changes and the challenges they pose, and how policymakers are responding to them. The downside risks and policy recommendations set out in this chapter underscore the need for vigilant and effective monetary policy to help mitigate the risks to macroeconomic and financial stability.

In this chapter “monetary policy framework” refers to the process through which central banks with a floating or managed floating exchange rate regime conduct monetary policy aimed at achieving macroeconomic policy objectives and react to shocks to mitigate their impact on the economy. The process encompasses setting the objectives and intermediate targets for monetary policy, analyzing relevant developments, using instruments to achieve intermediate targets, and communicating about monetary policy.

Monetary policy frameworks in sub-Saharan Africa are reviewed against the backdrop of far-reaching changes in the global economic and financial landscape that are shaping the conceptual underpinnings, institutional arrangements, and implementation of monetary policy. The global financial crisis has challenged established underpinnings of monetary policy and provoked a reassessment of the objectives, decision rules, and tools of monetary policy. Monetary authorities around the world are pragmatically adapting monetary policy to achieve and maintain low inflation and robust economic growth, while promoting financial stability.

Until recently, central banks in sub-Saharan Africa relied mainly on controlling the money supply to achieve their inflation targets, but this link has now weakened substantially. In addition, despite the low inflation environment, monetary policy has become more challenging in the face of external shocks that frequently require policymakers to make trade-offs among inflation, growth, and exchange rate movements.

Implementation of monetary policy in sub-Saharan Africa has been complicated further by vulnerabilities to terms-of-trade shocks, ongoing financial deepening, and changes in exchange rate regimes. In addition, monetary policy implementation in the region is subject to a number of institutional and capacity constraints.1 These include the lack of adequate and timely data to accurately measure inflation given the urban/rural divide in the consumption basket, the state of the economy, limited understanding of the transmission channels, weak accountability regarding policy objectives and transparency of policy decisions, and insufficient attention to communication strategies.

As elsewhere, the weaker relationship between the money supply and inflation and changing financial landscape are forcing central banks in sub-Saharan Africa to use more eclectic approaches. The transition is being helped by reduced recourse to central banks for financing budgets and greater exchange rate flexibility.2 Thus, Ghana, South Africa, and more recently Uganda, have adopted a formal inflation-targeting regime, while other countries are deemphasizing the role of monetary aggregates and incorporating elements of the monetary policy practice of industrial and emerging market countries. This includes greater reliance on interest rates for the transmission of the monetary policy stance, improved liquidity management, and enhanced capacity for policy analysis, forecasting, and communication strategies.3

The analysis in this chapter suggests that as countries adopt a gradual approach toward more flexible and forward-looking monetary policy frameworks, they will need to strengthen policy formulation, operations, information flow, and communications to avoid engendering policy confusion and delaying action. Essentially, the appropriate policy framework lies along a continuum using money aggregates as an anchor at one extreme, hybrid frameworks with a significant role for inflation and the exchange rate (depending on a country’s structural characteristics) in between, and formal inflation targeting at the other extreme. At the same time, monetary policy analysis could be improved in countries that do not have evolving monetary policy frameworks, as well as in those that are transitioning toward more forward-looking frameworks, by considering the following factors:

  • Reduce overreliance on monetary policy to achieve competing goals. Monetary policy should not be overburdened by multiple objectives that impede its effectiveness and may compromise central bank credibility. As suggested by empirical studies, pass-through from the policy rate to interbank rates tends to be weaker in countries pursuing multiple goals that can, in some instances, be conflicting.

  • Improve data. The timeliness of high-frequency macroeconomic data is important for monitoring and evaluating developments in the economy and determining the appropriate policy response. IMF Article IV reports consistently evaluate the adequacy of data for surveillance and point to areas in need of improvement.

  • Strengthen analytical capacity. Analytical models are an important part of the policymaker’s toolkit in gauging the transmission of monetary policy and forecasting the medium-term path for inflation. Technical assistance will be needed to strengthen analytical capacity aimed at bolstering processes that guide the implementation of monetary policy.

  • Develop a communication strategy. Communicating the reasons for changes to the monetary policy stance or revisions to inflation projections is essential to building credibility with the public. Policy predictability enhances effectiveness in signaling central banks’ preferences regarding price stability and, consequently, the effectiveness of anchoring the private sector’s expectations regarding the future path for inflation. Similarly it is important for central banks to explain deviations from targets when they occur, and how they intend to correct policy misses.

  • Augment the range of policy instruments. Price-based instruments, such as the policy rate, will gain increasing importance as a means of communicating the marginal cost of funds to the private sector. The development of money market instruments will enhance the effectiveness of the signaling effect of the policy rate. This will require reforms to improve the efficiency of money and foreign exchange markets along with supportive legal infrastructure governing the collateral and autonomy of the central bank.

  • Improve liquidity management. Countries pursuing flexible money-targeting regimes will be challenged to manage short-term liquidity while attempting to stabilize and mitigate volatility in short-term money market interest rates, especially in the context of structural excess liquidity.

  • Stabilize the financial sector. An adequately regulated and supervised financial system will contribute to better transmission of monetary policy signals. Weaknesses in commercial banks’ balance sheets may prevent using monetary policy in a more decisive manner, given that such decisions may have detrimental effects on weaker banks.

Snapshot of Monetary Policy Frameworks in Sub-Saharan Africa

Besides the close linkage to inflation, anchoring monetary policy in most sub-Saharan Africa countries to a monetary aggregate had the advantage of readily available monetary data and a disciplining effect on fiscal policy. In the 1990s, some of the underlying assumptions of a money-targeting framework became increasingly questionable, especially in advanced economies (Box 3.1). In particular, the view that the central bank had full control of the nominal money stock—and that the long-run relationship between money growth and nominal income growth was stable—was widely challenged. At the same time, exchange rate pegs also started to fall out of favor in the face of growing international capital flows and the high cost of potential currency crises.

A dramatic decline in trend inflation has been a major factor in the way monetary policy is conducted in the region. In the 1990s, sub-Saharan African countries hung on to money- or exchange rate-based monetary policy regimes as inflation was still in double digits on average for most sub-Saharan African countries—except those belonging to the CFA franc zone—and closely controlling the money supply was seen as critical for bringing down inflation.4 However in the last decade, the region has seen a trend decline in inflation and its volatility (Figure 3.1), with diminished importance attached by policymakers to changes in the money supply. In money-targeting countries, inflation declined from 24 percent in 1989–2000 to 10 percent in 2001–12.5 In two of the three inflation-targeting countries (Ghana, South Africa, Uganda), inflation performance was mixed. In South Africa, inflation declined from 10 percent to 6 percent, in Ghana from 27 percent to 16 percent. At the same time, terms-of-trade shocks have tended to hamper the transition to disinflation in several sub-Saharan African countries (Box 3.2).

Figure 3.1.Sub-Saharan Africa: Trends in Inflation, Average Consumer Price Index, 1980–2012

Source: IMF, World Economic Outlook database.

Box 3.1.Global Evolution of Monetary Policy Frameworks

The adoption of money targeting by many countries in the mid-1970s followed the collapse of the Bretton Woods system of fixed exchange rates in the early 1970s and a widespread increase in inflation. In the late 1970s and 1980s, many central banks around the world built their monetary policy frameworks around money targeting as an alternative to exchange rate pegs (Goodhart, 1989). Money targets had the advantage of counting on rapidly available information, and of being a useful tool to induce fiscal policy discipline.

In the early 1980s major industrialized countries began abandoning money targeting as unstable money demand, rapid financial innovation, disintermediation from the banking system, and deregulation of financial markets began to weaken the relationship between monetary aggregates and inflation; monetary targeting became increasingly unsatisfactory as a useful nominal anchor for monetary policy. This was followed by the collapse of exchange rate pegs of various kinds in industrial countries—accounting for two-thirds of monetary policy frameworks in 1989—culminating in the Exchange Rate Mechanism crisis in 1992 that helped spur the adoption of inflation targeting by some European countries followed by several emerging markets in the early 2000s. New Zealand was the first advanced economy to introduce inflation targeting in 1989.

By the end of 1990, global inflation was at its lowest since the breakdown of the Bretton Woods system. Many countries brought down inflation to 25-year lows and were in search of a framework to anchor inflation gains more effectively (Mahadeve and Sterne, 2000). Countries in the European Union made a historic shift toward a currency union, while other countries were pushed by currency crises toward greater exchange rate flexibility. Many advanced and emerging market countries have adopted inflation targeting since then, incorporating key elements of successful money-targeting regimes.

The first real challenge to the credibility of inflation-targeting regimes was the 2008 commodity-price shock and how to integrate financial stability in the inflation-targeting framework after the global financial crisis. Inflation targets were overshot by most emerging market inflation-targeting countries, but inflation-targeting countries saw inflation rising by less than in other countries. Currency appreciation under flexible exchange rate regimes and a greater degree of central bank independence helped keep inflation in check (Habermeier and others, 2009). Some economists find that inflation targeting, like other regimes in place in advanced economies, did not pay enough attention to asset-price bubbles and their potential impact on financial stability before the crisis, meriting exploration of other monetary policy options (Frankel, 2012). However, many authors argue that inflation targeting will still be needed as “advanced economies work their way through today’s slow growth, rickety banks, and over indebted public sectors” (Reichlin and Baldwin, 2013, p. viii).

This box was prepared by R. Armando Morales.

Box 3.2.Monetary Policy and Terms-of-Trade Shocks in Sub-Saharan Africa

In recent years, terms-of-trade shocks (mainly reflecting food and fuel price shocks) have posed challenges to the conduct of monetary policy in sub-Saharan Africa (2007–08 and 2010–11; Figure 3.2.1):

Figure 3.2.1.World Commodity Prices

Source: IMF, World Economic Outlook database.

The 2007–08 Episode

In the 2007–08 episode, average inflation in sub-Saharan Africa increased from 9 to 15 percent, mainly because of an acceleration in food prices from 10 to 20 percent and in fuel prices from 12 to 19 percent. Inflationary pressures quickly abated because of the worsening of the global financial crisis, and interest rates remained low in most sub-Saharan African countries in anticipation of potential spillover effects from the global downturn on the region’s economic activity. The policy advice during this episode focused on accommodating the first-round effects of these shocks.

The 2010–11 Episode

During the 2010–11 episode, the impact on inflation was more varied across the region (Figure 3.2.2). In countries with formal inflation targeting regimes (for example, South Africa), the inflation impact was largely confined to the first-round effects of higher import prices. In Ghana, another inflation-targeting country, inflation was little changed compared to the year before. However, the picture was very different in some fast-growing countries with floating exchange rate regimes that had kept interest rates low for a prolonged time. The terms-of-trade shock combined with strong credit growth led to significant inflationary pressures in these countries. In several cases, exchange rate depreciation compounded the inflationary impact (Kenya, Tanzania, Uganda). Inflation rates approached or exceeded 20 percent by the fourth quarter of 2011 in Burundi, Ethiopia, Kenya, Tanzania, and Uganda. Both countercyclical monetary and fiscal policy were pursued in a number of countries but at varying speeds in response to the shocks.

Figure 3.2.2.Sub-Saharan Africa: Inflation by Region

Source: IMF, International Financial Statistics.

Note: Zimbabwe was excluded due to hyperinflation.

This box was prepared by Yibin Mu.

The approach to more flexible monetary policy frameworks varies across countries in sub-Saharan Africa. Key features of the monetary policy frameworks for 11 countries—for which information is readily available and that represent varied monetary and exchange rate regimes, per capita incomes, dependence on natural resources, and central bank institutional arrangements as of 2013—are shown in Table 3.1. All countries report the use of some form of policy rate, and all but one pay attention to the interbank rate as an operating target.6 Still, 8 out of 10 countries rely on broad money as an intermediate target, and the same number report a floating exchange rate regime.

Table 3.1.Sub-Saharan Africa: Key Features of Monetary Policy Frameworks, Selected Countries
CountryMain InstrumentsOperating TargetsIntermediate targetsObjectivesDe Facto Monetary RegimeDe Facto Rate RegimeModalities of Communication
AngolaOpen market operations



Reserve requirement



Foreign exchange sales



Policy rate



Standing facilities
Exchange rateBroad moneyPreservation of the value of national currency







Price stability



Single-digit inflation
Exchange rate anchor to the U.S. dollarStabilized arrangementMPC in operation since 2012



MPS published infrequently
GhanaPolicy rate (prime rate)



Open market operations



Reserve requirement



Foreign exchange purchase/sale
Interbank rateInflation forecastLow inflation



Inflation target of 8.5 ± 2 percent



Employment



Growth
Inflation targetingFloatingMPC meets bimonthly; PR/PC







Central bank accountable to the parliament and the wider public
KenyaPolicy rate (central bank rate)



Open market operations



Reserve requirements



Standing facilities



Foreign exchange purchase/sale
Interbank rate



Reserve money



Net domestic assets of central bank
Broad moneyLow and stable inflation



Inflation target of 5 ± 2.5 percent



Growth



Employment



Financial stability
Hybrid inflation targeting liteFloatingMPC meets bimonthly; PR/PC



MPS published twice a year
MauritiusPolicy rate (key repo rate)



Open market operations



Standing facilities



Reserve requirement
Interbank rateInflation forecastPrice stability



Orderly and balanced development



Exchange rate
Hybrid inflation targetingFloatingMPC meets quarterly; PR



Inflation report published twice per year
MozambiqueOpen market operations



Standing facilities



Reserve requirement



Policy rate
Reserve moneyBroad money



Credit to private sector
Price stability.



Medium term inflation target of 5–6 percent



Financial stability
Money targetingFloatingMPC meets monthly; PR
NigeriaOpen market operations



Standing facilities



Policy rate



Reserve requirement
Reserve money



Interbank rate
Broad moneyPrice stability



Financial stability



Sustainable growth
Money targetingOther managed arrangementMPC meets bimonthly; PR/PC
RwandaOpen market operations



Policy rate (key repo rate)



Reserve requirement



Foreign exchange sales
Reserve money



Interbank rate
Broad moneyPrice stability and low inflation



Maintain a stable and competitive financial system without exclusion
Money targetingCrawl-like arrangementMPC meets quarterly; PR/PC



MPS published twice a year
South AfricaPolicy rate (repo rate)



Standing facilities; repo



Open market operations
Interbank rateInflation forecastAchievement and maintenance of price stability



Inflation target 3–6 percent
Inflation targetingFloatingMPC meets bimonthly; PR/PC



Central bank accountable to the parliament and the wider public
TanzaniaOpen market operations



Policy rate (central bank rate)



Reserve requirement



Foreign exchange sales
Reserve money



Interbank rate
Broad moneyPrice stability



Balanced and sustainable growth
Money targetingFloatingMPC meets bimonthly; PR/PC



MPS published twice a year
UgandaOpen market operations



Policy rate (central bank rate)



Foreign exchange purchase/sales

Standing facilities; repo
Interbank rateInflation forecastLow and stable inflation



Inflation target of 5 percent over the medium term
Inflation targeting liteFloatingMPC meets monthly; PR/PC



MPS published monthly
ZambiaOpen market operations



Policy rate (central bank rate)

Foreign exchange purchase/sales Reserve requirement
Reserve money



Interbank rate
Broad moneyPrice stability



Financial stability



Balanced macroeconomic development
Money targetingFloatingMPC meets monthly; PR/PC



MPS published twice a year
Sources: IMF staff survey; Hammond (2012); and websites of central banks.Note: PR= press release; PC= press conference; MPS= monetary policy statement; MPC=monetary policy committee.

A few specific observations are noteworthy in Table 3.1:

  • Achieving and maintaining low inflation is the primary objective of monetary policy frameworks in most countries (Table 3.1) and in most cases the specific level or the horizon over which it is to be achieved is not defined. Some money-targeting countries also use a policy rate as a complement (Kenya, Nigeria). However, under this regime while money targets may not play a systematic role in monetary policy, the adoption of greater flexibility permits the use of target ranges for monetary aggregates.

  • A similar set of monetary policy instruments is employed across all monetary policy frameworks, although the extent to which they are used varies across countries.

  • Monetary policy committees are now becoming an integral part of the institutional arrangements (Angola, Kenya, Uganda, among others). This has increased the transparency and communication of monetary policy.

Key Drivers of the Evolution of Monetary Policy Frameworks in Sub-Saharan Africa

Weaker relationship between money and inflation

As inflation has stabilized, its relationship with broad money has shifted in the face of financial deepening and increased capital inflows. Although the relationship remains positive and significant, recently it has weakened considerably (Figure 3.2). In sub-Saharan African countries, the correlation between inflation and M2 has declined from 0.88 in1989–2000 to 0.20 in 2001–12. This is consistent with the behavior observed in low-income countries as a whole: cross-section regressions show that the coefficient of money growth has declined from 0.64 in 1990–2002 to 0.29 in 2002–12 (IMF, 2014d). Also, the IMF study finds that the regression coefficient for money growth in high-inflation countries (inflation higher than 10 percent) is between 30 and 50 percent larger than for low-inflation countries.7 In addition, the positive correlation between changes in the velocity of money and inflation in sub-Saharan African countries is considerably lower, suggesting money demand is no longer influenced significantly by inflation developments.

Figure 3.2.Sub-Saharan Africa: Inflation, Money Growth, and Velocity, 12-year Average

Sources: IMF, African Department database, International Financial Statistics, and World Economic Outlook database.

Instability of money velocity and money multipliers

Since 2001, average velocity in sub-Saharan Africa has declined, largely because of the increased demand for money, reflecting the greater monetization of the economies and sensitivity to interest rates with the deepening of the financial sectors (Figure 3.3). At the country level, money velocity has declined even further in most money-targeting countries since 2007. Over a longer horizon, low-income countries show significant financial deepening between 1990–2010 (IMF, 2014d).

Figure 3.3.Sub-Saharan Africa Money Targeters: Evolution of Money Multipliers

Sources: IMF, African Department database, International Financial Statistics and World Economic Outlook database.

The evolution of money multipliers also reflects increased monetization and financial intermediation (Figure 3.4) accompanied by changes in transactions technologies and greater sensitivity to interest rates. Money multipliers have generally trended upward since 2001 for money-targeting countries, although there are variations across countries. Recent studies for member countries of the East African Community (EAC) find that money multipliers varied significantly in Kenya, Tanzania, and Uganda in the short run, but increased over time—consistent with financial deepening and greater demand for real money balances (Davoodi, Dixit, and Pintor, 2013). The variation of the multiplier likely reflects the effect of technological changes such as mo-bile banking on the use of currency, changes in reserve requirements, and the effect of interest rate movements on banks’ willingness to hold reserves. In addition, changes in reserve money in sub-Saharan Africa countries have been erratic, being affected by variable capital flows, including aid inflows, consistent with findings for other low-income countries (Batini, Kuttner, and Laxton, 2005).

Figure 3.4.Sub-Saharan Africa Money Targeters: Evolution of Money Velocity

Sources: IMF, African Department database, International Financial Statistics, and World Economic Outlook database.

Reduced fiscal dominance

The burden of fiscal dominance has been greatly reduced in the region, increasing the scope for central bank independence (Figure 3.5). Central bank financing to the government has declined throughout the region, from an average of more than 12 percent of GDP between 1990 and 2000, to nearly 2 percent of GDP in 2012, in line with a reduction in overall government debt. Some variation, however, exists across country groupings. For example, central bank financing remains high among fragile economies, particularly in Eritrea, where it has averaged some 50 percent in the past five years. Conversely, the largest improvements have taken place among middle-income countries.

Figure 3.5.Sub-Saharan Africa: Evolution of Central Bank Credit to the Government and Government Debt

Sources: IMF, International Financial Statistics, and World Economic Outlook database.

Higher exchange rate flexibility

The trend toward greater exchange rate flexibility in sub-Saharan African countries allowed room for a more independent monetary policy as the exchange rate was allowed to be a shock absorber (Figure 3.6). The number of soft-peg regimes has declined from 28 in 2007 to 11 in 2012 as many countries abandoned soft-peg regimes when the recent crisis unfolded.8 This has allowed monetary authorities to pay greater attention to fighting inflation rather than preventing currency crises. Over time, this has resulted in a lower elasticity of prices to changes in the exchange rate, reflecting improvements in the independence of monetary policy.

Figure 3.6.Sub-Saharan Africa: Exchange Rate Arrangements

Sources: IMF, Annual Report on Exchange Arrangements and Exchange Restrictions database.

Notes: South Sudan is excluded due to data limitations. Countries that belong to CEMAC and WAEMU (14 countries) are also excluded. In 2008, 14 countries previously classified as managed floaters (soft peg) were reclassified as floaters.

The adoption of more flexible exchange rate regimes combined with easing of restrictions on the capital account in many countries has enhanced the role of interest rates in monetary policy, for example, in Ghana, Kenya, and Uganda. Overall, the correlation between policy interest rates and inflation has strengthened, despite operational problems in the use of policy rates. This also reflects a shift toward a more intensive use of market-based instruments, such as open market operations, for the conduct of monetary policy by most sub-Saharan African countries. The correlation between policy rates and exchange rates has also increased slightly.

Higher exchange rate flexibility has made the use of market-based monetary instruments more intensive. Outright sales of government securities by central banks are used by 17 out of 25 reporting countries for monetary policy purposes, while 16 countries report the use of securities repos. By contrast, only 5 countries (Comoros, Eritrea, Namibia, Swaziland, Zambia) still report the use of some form of direct instrument (Table 3.2).

Table 3.2.Sub-Saharan Africa: Central Bank Monetary Policy Instruments, Selected Years
Direct InstrumentsStanding FacilitiesGovernment activities for monetary purposesOpen market operations
CountryYearInterest rate controlsCredit ceilingsDirected creditSpecific lending requirementsReserve requirementsStatutory liquidity requirementsStanding credit facilityStanding deposit facilityOther standing facilitiesSale of government securitiesTransfer of government depositsOutright sale/purchase of securitiesOutright sale/purchase of foreign exchangeReverse transactions-securities repoReverse transactions-foreign exchange swapsCollateralized lendingTake deposits or sell central bank billsOther OMO instrumentsSame list of collaterals for OMOs and SFsOwnership restriction of central bank bills
Angola2010XXXXXXXXXXX
Botswana2013XXXXXXXXX
Burundi2013XXXXXXXX
2010XX
Cabo Verde2010XXXXXXXXXXX
Comoros2013XXXXXX
2010XXXXXX
Eritrea2010XXXXX
Ethiopia2010XXXXXXXX
Ghana2010XXXXXXXX
Kenya2013XXXXXXX
Lesotho2013XXXX
2010XXXXXX
Liberia2013XXX
2010XXXXXX
Madagascar2013XXXXXXXXXX
Malawi2010XXXXXXXXXXX
Mauritius2013XXXXXXXXXXXX
2010XXXXXXXX
Mozambique2010XXXXXXXXXX
Namibia2013XXXXXXXXXX
2010XXXXXXXXX
Nigeria2013XXXXXXXXXXXX
2010XXXXXXXXXXXXX
Rwanda2013XXXXXXXXXX
2010XXXXXXXX
Seychelles2010XXXXXXXXXX
Sierra Leone2013XXXXXXX
2010XXXXXXXXX
South Africa2013XXXXXXXXXXXXXX
2010XXXXXXXXXXXX
Swaziland2013XXXXXXXXXXX
2010XXXXXXXXXXXXXX
Tanzania2010XXXXXXXXXXX
Uganda2010XXXXXXXXXX
Zambia2013XXXXXXXXXX
2010XXXXXXXXXXXXXX
Source: IMF, Information System for Instruments of Monetary Policy (ISIMP) database.Note: OMO = open market operations; SFs = standing facilities.

Changes in the financial landscape

Sub-Saharan African financial systems have seen significant deepening in the past decade, as reflected in the increase in credit to the private sector and commercial bank deposits (Figure 3.7). Given the relatively low level of financial development, this deepening should translate into potentially stronger monetary transmission, although interest rate spreads have not declined as observed in many emerging market countries during this period. While the sensitivity of economic agents to changes in interest rates has likely increased, lending rates remain high and sluggish because of structural impediments that adversely affect the risk perceptions of lenders.

Figure 3.7.Sub-Saharan Africa: Financial Deepening Indicators

Sources: IMF, Financial Access Survey, and World Economic Outlook database.

Objectives and targets

The evolution of monetary and financial systems and its impact on volatility of monetary aggregates has translated into weak adherence to reserve-money targets. In turn, these misses have been a poor guide to policy decisions. In fact, no statistical correlation has been found between reserve-money target deviations and inflation deviations (IMF, 2014d).

In light of this experience, the IMF has adapted its approach to monetary policy conditionality to provide flexibility to certain countries with IMF programs that are implementing more forward-looking monetary policy frame-works. The existing framework is being enhanced by introducing a monetary policy consultation clause for countries that have the capacity to adjust their policy settings in a flexible way to achieve their objectives (IMF, 2014d).

Responding to Challenges

Constraints to policymaking

The exchange rate and monetary aggregate anchors have been important for achieving macroeconomic stability and lowering inflation in sub-Saharan Africa. However, as discussed above, ongoing structural changes in sub-Saharan African economies have underscored the limitations of strict adherence to money targeting and led to the quest for more “flexible” monetary policy frameworks. At the same time, the demands on policymakers in sub-Saharan African countries have increased. Monetary policy is part of the first line of defense against a variety of shocks, including volatile fuel and food prices, and lumpy capital flows. Apart from responding to these shocks, in some sub-Saharan African countries monetary policy is also burdened with other objectives that go beyond maintaining price stability, such as supporting higher credit growth or offsetting expansionary fiscal policy.

Central banks in sub-Saharan Africa also face a number of constraints that make the policymaking process more challenging. Limited operational and financial independence of the central banks in some countries and persistent structural excess liquidity reduce the scope for effective use of interest rates and market-based instruments for monetary policy implementation. Government cash-management problems, including in aid inflows, complicate the conduct of monetary policy. High levels of dollarization in a few countries, particularly large currency mismatches on banks’ balance sheets, make policymakers more concerned about the impact of exchange rate swings on banks’ balance sheets.

A number of practical and technical aspects also play a role: (i) in several cases, considerably limited information is available to policymakers regarding developments in the economy, particularly forward-looking indicators and surveys; (ii) uncertainty about functioning monetary transmission mechanisms, including transmission channels at play, leads to delays in policy responses; (iii) recognizing structural relationships and mechanisms in the economy is more difficult when the economy is undergoing structural changes; (iv) credibility of the monetary policy framework is still weak in countries with relatively short track records of meeting objectives, compounded by the absence of accountability frameworks in some cases; and (v) countries with shallow financial markets lack the benefit of the “feedback loop” for the monetary policy stance, meaning that policymakers can distill little information from possible changes in market sentiment.

Thus, several sub-Saharan African central banks (for example, in Kenya, Mauritius, Seychelles, Tanzania, Uganda, and Zambia) are reviewing their frameworks to better understand factors influencing inflation. In particular, countries are reexamining the usefulness of strict adherence to monetary aggregates to anchor inflation expectations and signal the stance of monetary policy. On the one hand, allowing more flexibility in responding to liquidity shocks has become more important as countries in sub-Saharan Africa have opened and become more exposed to external influences. On the other hand, promoting financial deepening and market development requires relatively stable interest rates.

Transitioning toward market-based instruments

Some countries have explored the possibility of giving a greater role to short-term interest rates in guiding their monetary policy actions (for example, Kenya and Uganda). In particular, several countries have already moved to a combination of interest rate instruments with monetary aggregate targets, and they are working toward developing a set of forward-looking indicators. The need for such indicators became particularly noticeable during large external shocks to the economy (for example, Zambia). Frequent misses of money reserve targets forced policymakers to look for alternative sources of information that would allow for better assessment of a quickly evolving macroeconomic environment. Although the transition to more “flexible” monetary policy frameworks is taking place in a number of countries across the region, the central banks’ approaches display some important differences (Box 3.3).

The choice of two operating targets—interest rate and reserve money or a price and a quantity—seems problematic, but such a flexible use of operating targets is possible if movements in monetary aggregates and interest rates send a consistent signal to market participants. Such consistency is typically assessed by the monetary policy committee and involves taking into consideration a broader array of information beyond monetary aggregates. Several central banks in the region have successfully introduced a policy rate and are using adjustments in the rate to signal their monetary policy stance (for example, Uganda, Kenya, Tanzania, Mauritius, and Zambia, among others).

It should be noted, however, that the role played by policy rates in the monetary frameworks differs across countries. In some countries (for example, Tanzania), the rate itself tends to play a less important role than in others (Kenya, Uganda). This shows partly that changes in the policy rate are thought to have a smaller impact on inflation and economic activity (because of a weaker transmission mechanism). The Bank of Tanzania is mindful of limitations on strict adherence to reserve-money targeting, and its implications for interest rate volatility and signaling of monetary policy stance. Thus, some preliminary work is underway on the design of a more forward-looking monetary policy framework, with more active use of the policy rate.

In countries where the use of policy rates is more ingrained (Uganda, Kenya), market participants tend to focus more on the policy interest rate announced by the monetary policy committees, and less on the money aggregates. Moreover, setting the policy rate at the short end of the yield curve associated with central bank operations at short-term maturities (e.g., overnight or seven days) strengthens the signaling role, because it de-links monetary operations from longer-term interest rates (for example, 90-day securities). Such a change from a longer-term instrument to a shorter-term instrument has had a positive impact in Kenya in terms of strengthening policy signals.

Policymakers expect that short-term interest rates, which are being more effectively anchored by the policy rate, will replace reserve money or net domestic assets as the operating target over time, and the policy rate will be the primary monetary policy instrument, supported by proper liquidity management. Empirical studies indicate the interest rate channel can in turn become stronger and, if coupled with greater exchange rate flexibility, could help improve the monetary transmission mechanism, thereby providing a greater scope for independent monetary policy. Such effects through interest rate and exchange rate channels seem to be at play in East Africa as the framework for monetary policy was adapting more flexibly when inflation began soaring in late 2011 (Box 3.4).

A study by Mishra and Montiel (2012), illustrated important differences in the monetary policy transmission among various regimes.9 The main findings are as follows (Figure 3.8):

  • In countries with a full-fledged inflation-targeting regime (Ghana, South Africa) or some form of inflation-targeting lite (Uganda), or where money targeting is applied flexibly (Mozambique), interbank rates responded strongly and positively to changes in policy rates.

  • In countries pursuing some type of inflation-targeting regime or flexible money targeting, the policy rate also had a larger impact on lending rates, and banks reacted faster to changes in costs of funding. This was particularly the case for Ghana, Kenya, Mozambique, South Africa, and Uganda.

  • In countries with a “hybrid” monetary policy framework (Mauritius), the interest rate pass-through from the policy rate to the interbank rate appeared to be weak. This is largely due to excess liquidity in the system, which the central bank is reluctant to absorb because of profitability concerns. As a result, the interbank rates remain well below the policy rate, and the latter does not serve as a very effective anchor.

Figure 3.8.Sub-Saharan Africa: Impact of Changes in Policy Rates on Other Interest Rates-Experience of Selected Countries, 2003–13

Sources: IMF, African Department database; and International Financial Statistics; Mishra and Montiel (2012); and IMF staff calculations.

Note: The correlations are calculated by estimating the equation Δyit01 Δyit-12Δyit-23 Δxit4 Δxit-15 Δxit-2+ εit where y is the interbank rate (lending rate) and x is the policy rate (interbank rate) and in the final panel bottom right y is the lending rate and x is the policy rate. The long-term effect is calculated as (β345)/(1− β1− β2) and the short-term effect is given by β3. The sample period is from 2003 to early 2013 at monthly frequency.

IT = inflation targeting; MT = money targeting.

Introducing a policy rate, which serves as an anchor for short-term interest rates, can be accompanied by an interest rate corridor, with the intention of steering short-term rates toward the middle of the corridor. In this setting, the policy rate becomes the lowest rate for liquidity injection operations and the highest rate for liquidity absorption operations. A number of countries in the region have successfully launched such corridors (Cabo Verde, Lesotho, Mauritius, Mozambique, Namibia, Rwanda, South Africa, Uganda, Zambia, among others). This, however, requires more active liquidity management, with availability of both repo and reverse repo operations.

It is worth noting that introducing a corridor around the policy rate is not limited to countries with an inflation-targeting regime or a flexible money-targeting framework. For example in Cabo Verde, where the exchange rate remains an anchor for monetary policy, establishing a corridor around the policy rate was aimed at improving the monetary transmission mechanism, enhancing liquidity management, and activating the interbank money market. In other countries that struggle with excess domestic liquidity, the corridor usually helps to limit the volatility of short-term interest rates, and create demand for the central bank’s eligible collateral.

While deciding on the width of the corridor, the central banks typically balance different objectives, namely: (i) developing an active interbank market, which would be incentivized by a broad corridor; or (ii) reducing volatility in the overnight interbank interest rates by bringing them closer to the policy rate (narrow corridor). Too much recourse to the central bank facilities indicates the interbank market is not playing its role, and the central bank should widen the corridor. Conversely, particularly volatile interbank interest rates, with no recourse to the central bank overnight facilities, indicate that the central bank should consider narrowing its corridor. For example, the Bank of Zambia, in attempting to develop an active interbank market, initially established a wide corridor of 400 basis points.

Managing short-term liquidity

In countries with flexible money-targeting frameworks, short-term liquidity management aims at stabilizing short-term money market interest rates (Kenya, Zambia). Fine-tuning operations are conducted at a given interest rate, compared to achieving a volume target in the strict reserve-money-targeting framework. Some countries impose a limit on the total amount of liquidity auctioned, based on their assessment of liquidity conditions (Cabo Verde, Seychelles, Tanzania, Uganda).

This puts a premium on the quality of the central bank’s liquidity forecast, although the accuracy of the forecast is also important for all monetary regimes. Errors in the daily forecast are likely to be reflected in the deviations in money market interest rates from the policy rate. High volatility in money market rates and persistent, large deviations can undermine the signaling role of the policy rate.

Timely and better-quality data

Central banks in the region have also responded to lack of timely and reliable high-frequency statistics for making monetary policy decisions. This has enabled them to make their money-targeting regimes more flexible and forward looking. In some cases, it has facilitated transitioning away from money-targeting regimes, or strengthened their current inflation-targeting frameworks. The central banks of Ghana, Rwanda, South Africa, and Uganda are among the central banks in the region that have taken a lead in this area while supplementing these efforts with regular surveys of inflation expectations and economic conditions between banks and the private sector. These surveys and high-frequency indicators of coincident and leading economic indicators have become essential ingredients of the monetary policy committee decision-making process.

Despite challenges, central banks seem to be better positioned now to start implementing monetary policy in a forward-looking manner based on a thorough assessment of the economic outlook, underpinned by analysis of a broad group of indicators (movements in market interest rates, monetary aggregates, credit to the private sector, fiscal stance, exchange rate, capital flows, output gap, and external developments). However, this in turn requires a supporting institutional infrastructure and adequate analytical capacity for modeling.

How to Make Monetary Policy in Sub-Saharan African Countries More Effective

As underscored earlier in the chapter, the relationship between money and inflation has weakened over time in a number of developing countries. This reflects in part greater international capital mobility and financial innovations that give rise to instability in velocity and money multipliers and questions whether money targets remain effective components of monetary policy frameworks. Consequently many developing countries have increasingly adopted more flexible monetary policy frameworks, underpinned by forward-looking policies. Specifically, between 2003 and 2011 about 40 percent of emerging markets and 20 percent of low-income countries have moved away from strict money targeting (IMF, 2014d). In sub-Saharan Africa, countries including Uganda, Kenya, and others are already moving away from strict money targeting, in addition to the two earlier sub-Saharan Africa inflation targeters (Ghana and South Africa).

The adoption of a forward-looking framework requires capacity building among central banks to strengthen their institutional frameworks. A number of pillars are critical to this endeavor, including improving the functioning of money and foreign exchange markets to enhance the transmission of monetary policy signals to markets. Central banks also need to develop a number of instruments to manage excess liquidity in the financial system while balancing the market’s demand for money and mitigating the emergence of inflationary pressures. Flexibility in daily liquidity management would depend on the stages of money market development. Where money markets are shallow, greater flexibility in short-term liquidity management may be desirable because money market rates may not reflect changes in daily liquidity conditions. In countries with fairly developed money markets where interest rates are responsive to market developments, short-term liquidity management should be geared toward stabilizing money market rates.

Complementary to these reforms of monetary policy operations is development of analytical tools for forecasting inflation and analyzing the transmission of monetary policy. Understanding the channels through which monetary policy affects inflation will clarify the trade-offs between price and output stability. The IMF has pro-vided training to a number of central banks in sub-Saharan Africa in monetary policy analysis, the development of inflation forecasting models, and a broad range of issues related to monetary and foreign exchange policies, central bank governance, reserve management, and the development of financial market infrastructure. How-ever, more needs to be done because strengthening institutional capacity takes time.

Sub-Saharan African central banks’ capacity for monetary policymaking depends critically on the timely availability of high-frequency macroeconomic data. Although many countries have reliable series on monthly inflation and interest and exchange rates, few have developed high-frequency indicators of real activity. Many countries have only annual GDP series. This shortcoming highlights the need for high-frequency indicators covering a broad spectrum of domestic and external sectors critical to evaluating the state of the economy throughout the business cycle. Timely policy intervention is crucial to mitigating the adverse impact of macroeconomic imbalances. This lesson was forcefully brought to the fore by the responses of a number of sub-Saharan African frontier markets in managing surges in capital inflows that were driven by both cyclical and structural factors (see the October 2013 Regional Economic Outlook: Sub-Saharan Africa).

A key pillar in the transition toward more forward-looking monetary policy frameworks comprises increased reliance on the central banks’ policy interest rates. The use of price-based instruments to communicate the marginal cost of funds to the market would provide greater clarity and transparency in signaling the policy stance. Along the continuum of countries headed for this transition, the Bank of Uganda has adopted a policy rate as the instrument by which it would achieve a low and stable inflation objective.

Given that macroeconomic policies are the primary tools to correct for macroeconomic imbalances, countries that move toward more forward-looking monetary policy frameworks will need to achieve greater coordination of fiscal and monetary policies. This would encompass greater political commitment at the highest level to fiscal discipline over time to build policy credibility by signaling to the public that the central bank has the full backing of the government to pursue its mandated primary objective of price stability. Equally important are communications to the public regarding current and future inflation and greater clarity on why targets were missed so as to shape inflation expectations and to strengthen the central bank’s credibility.

Removal of the impediments to financial sector deepening augurs well for improving the effectiveness of the monetary transmission mechanism. A major concern of central banks during the transition is whether banks can safely respond to monetary policy signals without endangering financial sector stability. A key lesson from the Great Recession is that financial stability is as important as the pursuit of price stability. Central banks must reform their banking systems through adoption of a risk-based approach to banking supervision, recapitalization of weaker banks, better provisioning and correct reporting, and recognition of nonperforming loans.

Conclusion

With the changing monetary policy landscape, reflecting in part greater integration with the global economy and financial deepening, a number of sub-Saharan African countries have begun to augment their monetary policy frameworks with forward-looking elements. This trend has gained impetus with the sustained decline in inflation and an apparent weaker relationship between monetary aggregates and inflation. However, the path toward more effective monetary policy requires a number of reforms that makes capacity strengthening among central banks necessary.

The way forward will require development of a tool kit that includes instruments to manage excess liquidity, high-frequency data to assess macroeconomic conditions and to guide timely policy intervention, and analytical models for forecasting inflation that are underpinned by a clear understanding of the monetary transmission process. A clear communications strategy that conveys the rationale for the policy stance and subsequent changes will augur well for anchoring inflation expectations and fostering policy credibility over time.

Finally these developments will also have implications for IMF programs with sub-Saharan African countries in their assessment of the stance of monetary policy. Increasingly, the focus of the evaluation of monetary policy will shift to a deeper assessment of monetary conditions that explain the evolution of monetary aggregates and inflation. Not all countries will be able to move toward this framework on account of underdeveloped financial markets, fiscal dominance, and limited institutional capacity. In such cases these countries are better advised to continue with money-targeting regimes.

Box 3.3.Adapting Monetary Policy Frameworks to Challenges

Uganda: After experiencing large volatility in short-term interest rates in the aftermath of the reversal of the carry trade in late 2008, the Bank of Uganda adopted a more flexible approach to the implementation of the reserve-money program in 2009. Flexibility was achieved through a separation of operations into two components: a structural program for liquidity management set monthly on the reserve-money target and a fine-tuning of liquidity operations with a daily to weekly horizon, which focused on stabilizing money market rates. Later, in July 2011, the Bank of Uganda announced a transition to inflation targeting lite and introduced a central bank policy rate to guide interbank rates toward the middle of the interest rate corridor set by standing facilities.

The Bank of Uganda achieves its inflation objective through taking into account developments in a broad range of macroeconomic variables, including inflation, interest rates, and exchange rates. Because focusing on monetary targets alone has been increasingly insufficient, the Bank of Uganda also considers high-frequency indicators of economic activity. Introduction of the central bank rate has facilitated and contributed to enhancing monetary policy signaling.

Kenya: In the last few years Kenya has faced two large shocks from international food prices, which were quickly translated into domestic inflation. Monetary policy was also challenged by the inability to accommodate expansion in domestic demand. As a response to unstable money demand and weakening correlation between money growth and inflation, the Central Bank of Kenya decided to move gradually toward an inflation-targeting framework, with inflation becoming the overriding objective of the monetary policy.

The medium-term inflation target was set at 5 percent, with a range of ± 2.5 percent. The Central Bank of Kenya introduced the central bank rate, and since 2012 monetary policy incorporates policy analysis and a forecasting model to support monetary policy committee decisions. The central bank rate is based on monetary policy operations and is expected to guide movements in short-term interest rates. The Central Bank of Kenya conducts regular open market operations to ensure stability in the interbank market and short-term interest rates. The monetary policy stance is signaled by changes in the central bank rate. To ensure that future sustainability is not derailed by excessive expansion in credit to the government, the growth in net domestic assets is subject to a limit. The reserve-money target is not binding in Kenya; therefore, the operational framework for monetary policy implementation provides room for judgment calls. This means that policy can be adjusted without revisions to the reserve-money target. Judgment calls are based on market information and guided by inflation forecasts.

Zambia: Highly dependent on copper exports, Zambia’s economy experienced large deterioration in terms of trade and sharp outflow of foreign investors from the domestic government securities market during the global financial crisis. Domestic banks decisively curbed their lending to the private sector and sharply increased demand for liquid assets. Reallocation of assets in the banking sector, together with a slowdown in the growth of broad money, contributed to a decrease in the money multiplier. The buildup of liquidity, which mirrored increased risk aversion in domestic banks rather than growing inflationary pressure, sent the wrong signal on the monetary policy stance. This heightened the need to enhance the monetary policy framework by developing a broader set of forward-looking indicators, allowing for better assessment of quickly changing macroeconomic conditions during large external shocks to the economy. Missing the money reserve targets at that time underscored the need to develop a more “flexible” monetary policy framework.

To improve the signals of monetary policy stance the Bank of Zambia introduced a policy rate in April 2012. The intention was to replace reserve money with a policy rate as the main monetary policy tool. The policy rate was initially set at 9 percent. Simultaneously, the Bank of Zambia introduced a 400 basis point band for the interbank rates, as part of its strategy to activate the interbank money market and encourage banks’ liquidity management. These actions increased emphasis on stabilizing short-term interest rates and expanding the forward-looking aspect of monetary policy to stabilize inflation. The changes in monetary policy implementation are expected to lead to a transition to a full-fledged inflation-targeting regime in the future.

This box was prepared by Bozena Radzewicz-Bak.

Box 3.4.The Monetary Transmission Mechanism in the Tropics: A Narrative Approach

Many central banks in sub-Saharan Africa are looking to modernize their frameworks to make monetary policy more forward looking to promote macroeconomic stability, financial development, and, ultimately, economic growth. However, how the monetary transmission mechanism operates in low-income countries is not well understood. Evidence analyzed through statistical techniques conveys a sense of monetary policy “pessimism,” because the results typically do not find significant effects of policy on relevant macro variables. However, such analysis is often constrained by ambiguous identification techniques and noisy and scarce data (Mishra and Montiel, 2012). Thus, a case study approach can be more insightful.

In October 2011, in the face of high and rising inflation, the central banks of four east African countries—Kenya, Rwanda, Tanzania, and Uganda (EAC4 for short)—decided to tighten monetary policy in a coordinated manner. While the tightening took place in response to economic events—a commodity price shock, vigorous economic activity, balance of payments pressures, and accommodative policy—it was both unexpected and unusual in its magnitude. It therefore serves as a useful event to learn about the transmission mechanism, especially given variations across countries in terms of the regime in place and the extent of the tightening (Davoodi, Dixit, and Pinto 2013).

A review of the developments in EAC4 indicates a robust transmission mechanism. After a large policy-induced rise in the short-term interest rate, lending and other interest rates rose, the exchange rate appreciated, output fell, nominal credit growth slowed, and inflation declined (Figure 3.4.1). This narrative is corroborated by model-based analysis, both in terms of the initial inflation surge and the role of policy in stabilizing inflation (Benes and others, 2013). The case of Rwanda is instructive, where tighter monetary policy throughout 2011 and a stable exchange rate kept inflation from taking off.

Figure 3.4.1.East Africa: Selected Macro Variables, January 2010–13

Sources: IMF, International Financial Statistics; and IMF staff estimates.

The policy regime is critical to the nature of the transmission, regardless of the structure of the financial system. The clearest transmission occurred in Uganda, where the financial system is not as developed as in Kenya but where the inflation-targeting lite regime is simpler and more transparent. In Kenya, which has the most advanced financial sector in the region, an incipient tightening earlier in the year was unsuccessful. Transmission was effective once the Kenyan authorities explicitly signaled the monetary policy stance with their policy rate and described their intentions in terms of their inflation objective. In other cases, such as Tanzania, which conducts monetary policy under a de jure monetary-targeting regime, and Rwanda, which has a de facto exchange rate peg, the transmission to lending interest rates is less evident.

This box was prepared by Andrew Berg, Luisa Charry, and Rafael Portillo based on Berg and others (2013).

Although these challenges are not unique to sub-Saharan African countries, they are relatively more pronounced, reflecting the differences in the region’s economic and trade structure.

The breakdown in the relationship between money and inflation partly reflects changes in money velocity and multiplier. Money velocity is defined as the frequency at which one unit of currency is used to purchase domestically produced goods and services in a given time, measured by the ratio of GDP and broad money (M2). The money multiplier measures the amount of commercial bank money that can be created by a given unit of central bank money, measured as the ratio between broad money (M2) and base money (M0).

For a discussion see “Monetary Policy Issues in Sub-Saharan Africa,” by Berg and others (2014).

The CFA (African Financial Community) franc zone comprises two monetary unions pegged to the euro: the West African Economic and Monetary Union (WAEMU), comprising eight members, and the Economic and Monetary Community of Central Africa (CEMAC), comprising six members.

Excluding the Democratic Republic of the Congo, which experienced hyperinflation in 1993–95.

In some cases the policy rate is an announced rate and may not always signal the stance of monetary policy.

In 2008, the IMF undertook a reclassification of soft-peg regimes.

The study examined how changes in the policy rate are passed through to interbank rates and commercial banks’ lending rates for 10 countries representing a variety of monetary policy frameworks and shows important differences for different types of regimes (strict money targeting, flexible money targeting, inflation-targeting lite, hybrid inflation targeting, full-fledged inflation targeting).

    Other Resources Citing This Publication