CS/TCM/MERP/VI/5

Page 1

Distr.

LIMITED

CS/TCM/MERP/VII/5

October 2009

Original: ENGLISH

COMMON MARKET FOR EASTERN

AND SOUTHERN AFRICA

Seventh Meeting of the COMESAMonetary and Exchange

Rates Policies Sub-Committee

Lusaka, Zambia

12-14 October, 2009

REPORT OF THE SEVENTH MEETING OF THE COMESA

MONETARY AND EXCHANGE RATES POLICIES SUB-COMMITTEE

09 (IZ-mmn)

A.INTRODUCTION

  1. The Seventh Meeting of the COMESA Monetaryand Exchange Rates Policies Sub-Committee was held from October 12 to 14, 2009 in Lusaka, Zambia.

B.ATTENDANCE, OPENING OF THE MEETING, ADOPTION OF THE AGENDA AND ORGANISATION OF WORK

  1. The meeting was attended by representatives from Central Banks of Burundi, Congo (D.R.) Egypt, Kenya, Mauritius, Libya,Rwanda, SwazilandSudan, Uganda and Zambia. The list of participants is at Annex I.

Opening of the Meeting (Agenda item 1)

  1. The Chairperson of the meeting, Mrs. Nkusi Rose Kamariza, Board Advisor forBanque de la Republique du Burundi, called the meeting to order. She welcomed the delegates and expressed confidence that the Sub-Committee would come up with concrete proposals to move the COMESA Monetary Cooperation Programme forward.

Adoption of the Agenda and Organisation of Work

  1. The Meeting adopted the following agenda:
  1. Opening of the Meeting
  1. Adoption of the Agenda and Organisation of Work
  1. A Draft Charter of the COMESA Monetary Institute
  1. A Study on the Choice of Monetary Policy Regimes- Case Study of Selected Countries
  1. A Study on the Extent and Symmetry of Shocks in Selected COMESA member Countries
  1. Global Financial Crisis: Impact, Response and the Way forward; and
  1. Work Plan of the Monetary and Exchange Rate Policies Sub-Committee for 2010.
  1. Any Other Business
  1. Adoption of the Report and Closure of the Meetings
  1. The meeting agreed on the following hours of work:

Morning: 9.00-12.30 hours

Afternoon: 14.00-17.00 hours

C.ACCOUNTS OF PROCEEDINGS

A Draft Charter of the COMESA Monetary Institute ( Agenda item3)

6.The Secretariat highlighted salient points of the following articles of the Draft Charter of the Institute:

(i)Definitions;

(ii)Establishment;

(iii)Membership of the Institute;

(iv)Objectives of the Institute;

(v)Functions of the Institute;

(vi)Organs of the Institute;

(vii)Functions of the Committee of Governors of Central Banks;

(viii)Meetings of the Committee of Governors of Central Banks;

(ix)Bureau of the COMESA Committee of Governors of Central Banks;

(x)Finance and Monetary Affairs Committee;

(xi)Secretariat of the Institute;

(xii)Powers to Seek Assistance;

(xiii)Relations with the COMESA Secretariat;

(xiv)Status, Immunities and Privileges of the Institute;

(xv)Budget of the Institute; and

(xvi)Reports of Accounts; Financial regulations; Headquarters, Official Languages; Channels of Communication; Amendments of the Charter; Dispute Settlement; Dissolution of the Institute; Entry into force; Transitional Provisions; withdrawal and Conflict of Provisions.

RECOMMENDATIONS

7. The meeting adopted the Charter of the Institute with amendments which is contained in annex I of this report for onward transmission to the COMESA Finance and Monetary Affairs Committee.

A Study on the Choice of Monetary Policy Regimes- Case Study of Selected Countries ( Agenda item 4)

8.Under this agenda item, a presentation on a study on the choice of monetary policy regimes in selected COMESA member countries was made by Dr. Christopher K. Kiptoo, who is currently Deputy Advisor, Economic Affairs, in the Office of the Prime Minister of the Republic of Kenya. The study pointed out at the outset that the choice of an appropriate nominal anchor - the variable the central bank uses to discipline its policy decisions and convince agents in the economy that it can deliver price stability- is critical in the choice of the monetary policy framework adopted. Therefore, adopting the right monetary policy framework is a key decision to be taken by the authorities. It also indicated that international experience over the recent past has sparked an active discussion in COMESA about a change in the monetary policy strategy with a view to adopting an inflation targeting (IT) framework, hence the decision by COMESA Committee of Governors of Central Banks held in Cairo, Egypt, in October 2008 for the study to be undertaken.

9.In the study, existing monetary policy frameworks were reviewed. It indicated that all contemporary monetary policy frameworks are ‘inflation targeting’ frameworks in the sense that one of the objectives of monetary policy is to establish a credible nominal anchor for domestic prices. The study however, indicated that ‘inflation targeting’ frameworks differ in terms of the choice of anchor and, as a consequence: choice of instruments, mode of operation and communication and engagement. They also differ in terms of how far concerns about other objectives are reconciled with inflation objectives. The study pointed out that there are at least four types of nominal anchors that exist, namely; commodity targeting: exchange rate targeting interest rate targeting and monetary targeting. It also pointed out that while in general, the choice of monetary policy framework adopted by a country depends on economic, financial and institutional environment within which policy is operating apart from other constraints in policy formulation, the choice of exchange rate arrangement by a country determines the degree of independence of its monetary policy. It indicated that there are at least three alternative monetary policy frameworks that countries have adopted, namely; exchange rate targeting, monetary targeting and inflation targeting. The study pointed out that exchange rate targeting is currently employed by at least 115 countries with the following exchange rate regimes: Exchange arrangement with no separate legal tender;Currencyboard arrangement; Other conventional fixedpeg arrangement;Pegged exchange rate within horizontal bands; Crawling peg; Crawling band. Djibouti was pointed out as the only COMESA country implementing exchange rate targeting framework..

10.Under theMonetary Targeting framework, the study indicated that a stable and well understood relationship is assumed between inflation and a chosen monetary aggregate. It also indicated that to attain the intermediate target and ultimate objectives of monetary policy, an operational framework is adopted that specifies operational variables of the monetary policy. Under this framework, monetary policy may aim at controlling the interest rates or monetary base of the banking system. The study pointed out that there are at least 22 Countries in the world currently implementing the monetary targeting framework. Most of the COMESA countries fall under this framework. With regard to the inflation targeting (IT), the study indicated that public announcement of medium-term numerical targets for inflation, with an institutional commitment by the monetary authority to achieve these targets is critical. Additional key features include; increased communication with the public and the markets about the plans and objectives of monetary policymakers and increased accountability of the central bank for its inflation objectives. Monetary policy decisions are guided by the deviation of forecasts of future inflation from the announced inflation target, with the inflation forecast acting (implicitly or explicitly) as the intermediate target of monetary policy. The paper pointed out that at least 44 countries use inflation targeting as a framework for monetary policy. None of these countries is in COMESA with the majority falling under developed countries or emerging markets. South Africa and Ghana are among the few countries in Africa that have adopted inflation targeting monetary policy framework.

11.The study also covered the pros and cons of different monetary policy regimes in addition to the institutional arrangements for implementing different monetary policy regimes. The study pointed out three elements of the central bank’s institutional arrangements that are particularly relevant in implementing different monetary policy regimes generally. They are the degree of; central bank autonomy, accountability mechanisms; and procedures and decision-making arrangements. The study also made empirical examination of money demand functions for selected COMESA countries using cointegration analysis technique. The results showed that income and interest rate as well as the nominal exchange rate elasticity’s were significant and carried expected signs. Given that the monetary targeting framework is used by most COMESA countries, the study tested for the stability of the demand for money estimate for a selected group of COMESA countries using conventional methods such as; CUSUM, CUSUM Squares and Recursive Estimates and coefficient tests. The results were mixed. Some countries like Egypt and Kenya had generally stable money demand functions while others like Zambia and Mauritius had unstable ones.

12.The study looked at the effects of emerging issues such as dollarization as well as structural changes in the real economy on effectiveness of the conduct of monetary policy. It pointed out that money demand has proved unstable in many countries, limiting its usefulness as an indicator of the appropriate stance of monetary policy. The study also reviewed the performance and challenges of the existing monetary policy regimes in selected COMESA countries. It established that the monetary targeting framework is employed by most COMESA countries as a framework of monetary policy with price stability as the final target, monetary aggregates as intermediate operating target and reserve money as operation target of monetary policy. The study pointed out that indirect instrument of monetary policy instruments are applied by COMESA countries. They include: Open market operations -Treasury bill auction, deposit auction arrangements and foreign exchange auction; repurchase and discount window facility; Minimum Reserve Requirement (MRR) and Standing Credit Facility (SCF) - an overnight collateralized loan facility that provides funds to the commercial banks, so as to cover temporary end-of-day liquidity shortages. The study established that since the adoption of the monetary targeting framework, more specifically the reserve money program, inflation has been brought under control. Inflation in most countries has declined from double-digit levels of in early 1990s to single digit levels. The good inflation outturn has been largely a result of the continued pursuance of prudent monetary policies by most COMESA countries. Close co-ordination between the monetary and fiscal authorities also contributed significantly to bringing down inflationary expectations.

13.The study identified the following as some of the challenges faced by COMESA countries in the implementation of monetary policy: Weak and underdeveloped financial system that curtail the effectiveness of the monetary policy; money market imperfections evidenced through interest rate volatility ,and fiscal dominance; impact of external shocks such as international fluctuation of commodity and oil prices; an unstable relationship between money and inflation due to financial deregulation, and an increasingly unstable money multiplier and to some extent velocity which are making it difficult to derive a credible path for money supply and challenges of forecasting liquidity within a short-time period that constrain effective implementation of open market operations. This mostly manifests in excessive fluctuation of short-term interest rates and challenges due to increasingly globalized World. The study also presented country experiences of different monetary policy regimes as well as the requirements for implementing different monetary policy regimes.

Discussions

14. In the ensuring discussions, participants commended Dr. Kiptoo for making an elaborate and excellent presentation. The meeting in particular commended Dr. Kiptoo for attempting to address all the terms of reference (TOR) for the study in spite of the fact that they were too many and in fact constituted many studies compressed into one. In this respect, the meeting requested the COMESA secretariat to in future ensure too many TORs are not given in one study unless there is sufficient resources and time.

15.The meeting observed that the results showing the insignificance of interest rate in the estimated money demand model for Swaziland may be a reflection of Swaziland’s membership to the Common Monetary Area (CMA) arrangements in which there is free flow of capital. In this case, changes in interest rates may not have a significant influence in money demand. Swaziland is in many ways following the Inflation Targeting framework that is practiced by the Reserve Bank of South Africa. It was, however, observed that the coefficient on income that was above unity was too high. It was therefore proposed that the data on Swaziland GDP data should be checked to confirm its accuracy. The same case also applied to the GDP variable for Uganda.

16.The meeting also made the following comments in respect of the study:

(i)The channels of monetary policy need to be well understood. The study should make reference to extensive work already done in COMESA on monetary policy transmission and other related studies;

(ii)Consider using dummies to reflect structural breaks in respective COMESA countries;

(iii)Consider using deposit interest rates in addition to the treasury bill rates to capture for the opportunity cost for holding money;

(iv)Consider inclusion of inflation expectations in the money demand model using, for instance, a proxy such as a relevant yield curve;

(v)Consider using quarterly data to avoid the noise that often accompany monthly data;

(vi)Need to check for possibility of multicollinarity problem given the use of nominal interest rate as well as nominal exchange rate both of which affect one another.

Recommendations

17,The meeting agreed on the following recommendations of the study:

(i)Since many of the COMESA central banks face limitations in respect of technical and institutional capacity to implement the alternative framework of inflation targeting, it is necessary for them to adopt modified reserve money framework while in the meantime putting in place the necessary preconditions of inflation targeting that include; central bank independence, availability of high frequency data and forecasting models.

(ii)Need for deepened financial markets; strengthened regulatory and supervisory function of deposit taking financial institutions; monetary and fiscal policy coordination;coordination and consistency; as well as access to financial services.

(iii)COMESA countries should prepare to move in the medium-term to an inflation targeting framework for conducting monetary policy only when the prerequisites for inflation targeting are achieved.

A Study on the Extent and Symmetry of Shocks in Selected COMESA member Countries (agenda item5)

18.This agenda item was presented by Dr. Charles Abuka of Bank of Uganda, Dr. Jacob Opolot of Bank of Uganda and Dr. Lucas Njoroge of the Central Bank of Kenya,

19.They informed the meeting that this study is important in that it contributes to the on-going debate on the suitability of forming a monetary union in the COMESA region. In particular, it examines the variability, symmetry/asymmetry, transmission mechanisms and the persistence of shocks in the region..

Demand and supply shocks

20.FollowingBlanchard and Quah, 1989; Bayoumi and Eichengreen, 1992, 1993; Bayoumi and Taylor, 1995; Ramaswamy and Slok, 1998; and Frenkel and Nickel, 2002 among others, the study uses a finite two variable Structural VAR to identify and recover the underlying supply and demand shocks. The authors noted that although there have been modifications to this identification scheme, for example the use of a three variable VAR, the underlying assumptions remain largely consistent with a two variable VAR as applied in this study. The basic premise of the methodology is that while supply shocks have permanent effect on output, demand shocks have only temporary effects. However, both demand and supply shocks are assumed to have permanent effects on prices.

21.The analysis indicates that shocks vary across countries in terms of both magnitude and frequency with demand shocks showing a greater dispersion. Specifically, relatively small magnitudes and equally distributed demand and supply shocks are observed suggesting a higher likelihood of shocks convergences within COMESA countries. Correlation analysis suggests that there is some symmetry of demand and supply shocks among COMESA member countries implying possibly, some increasing degree of macroeconomic policy coordination following a number of initiatives that have been implemented to further policy coordination such as adopting and implementing macro-economic convergence criteria. This may generally suggests increasing trade and financial linkages among a majority of COMESA countries.

22.The results also show that the impulse response of price to a positive supply shock leads to permanently lower prices. The effect of a supply shock on output in most COMESA countries is in general, positive as expected. In addition, following a demand shock, the output of almost all countries rises temporarily but the effects wanes relatively quickly. The countries have similar magnitudes and speed of adjustments to the shocks suggesting symmetry of shocks. This finding concurs with the descriptive and correlation analysis. The few exceptional countries that show marked difference in size of the shock and speeds of adjustments seem to be confined to those that have experienced major civil strife (DRC, Rwanda, Burundi etc). Variance decomposition results reveal that supply shocks account for most of the variability of real output, while demand shocks accounts for most of the variability in prices, though with a few exceptions. On the other hand, demand shocks account for a much higher proportion (over 50 percent) of the variability in the price level relative to its contribution to the variability in real output, but with greater spread across countries.

Real exchange rate shocks

23.The bilateral real effective exchange rates (RERs) were computed by taking Kenya to be the reference country. The effect of seasonality on the monthly real exchange rate (RER) was removed by regressingthe observed changes in the RER on a set of 12 monthly dummies. The monthly deseasonalized RER changes are the short-run RER shocks, while the quarterly deseasonalized RER changes obtained by consecutively aggregating over the seasonally adjusted monthly RER changes to obtain typical four quarters in each year depict the long-run case.