CENTRAL BANK OF EGYPT

Transcription

1 CENTRAL BANK OF EGYPT ECONOMIC REVIEW Vol. 50 No /2010 Research, Development and Publishing Sector

2 The Economic Review is issued by the Research, Development and Publishing Sector at the Central Bank of Egypt (CBE) on a quarterly basis. It aims to make available to a broad readership of specialists and nonspecialists a wide range of information on the performance of the Egyptian economy during the reporting period. The CBE posts the Review on its website:

3 Contents Main Monetary and Financial Indicators Page The Leading Article: Egypt's Yield Curve: Application and Policy Implications Macroeconomic Performance 1/1 - Gross Domestic Product (GDP) 47 1/2 - Employment and Unemployment 54 1/3 - Inflation 56 1/4 - Tourism Monetary and Banking Developments 2/1 - Monetary Policy and Monetary Aggregates 68 2/1/1- Monetary Policy 68 2/1/2- Reserve Money 70 2/1/3- Banknote Issue 72 2/1/4- Domestic Liquidity (M2) and Counterpart Assets 73 2/1/5- Payment Systems and Information Technology (IT) 77 2/1/6- RRTGS and SWIFT Local Services 79 2/2 - Banking and Credit Developments 81 2/2/1- Banking Reform 81 2/2/2- Banking Supervision Sector 84 2/2/3- Banks' Aggregate Financial Position 88 2/2/4- Interbank Money Market in Egypt 90 2/2/5- Deposits 91 2/2/6- Lending Activity Stock Exchange 3/1 - Shares Market 97 3/1/1- Primary (Issue) Market 97 3/1/2- Secondary (Trading) Market 99 3/2 - Bonds Market 100 3/2/1- Primary (Issue) Market 100 3/2/2- Secondary (Trading) Market 101 3/3 - Mutual Funds 101

4 4 - Public Finance and Domestic Public Debt 4/1- Consolidated Fiscal Operations of the General Government 102 4/2 - Domestic Public Debt 108 4/2/1- Debt of the Government (Net) 108 4/2/2- Debt of Public Economic Authorities 111 4/2/3- Resources and Uses of the National Investment Bank (NIB) External Transactions 5/1 - Foreign Exchange Market 113 5/2 - Balance of Payments 115 5/2/1- Trade Balance 115 5/2/1/1- Merchandise Export Proceeds by Degree of Processing 117 5/2/1/2- Merchandise Import Payments by Degree of Use 117 5/2/1/3- Sectoral Distribution of Commodity Transactions 118 5/2/1/4- Geographical Distribution of Commodity Transactions 120 5/2/2 - Services Balance and Transfers 122 5/2/3 - Capital and Financial Account 125 5/3 - International Finance 127 5/3/1 - FDI in Egypt 129 5/3/2 - External Official Grants 132 5/3/3 - External Debt 135 Annex Statistical Section 141

5 Main Monetary and Financial Indicators GDP (LE bn) July/Dec. 2008/ /2010 GDP at Current and Market Prices Annual Growth Rate (%) Real GDP at Factor Cost Annual Growth Rate (%) GDP Growth at Factor Cost by Sector (%) A) Productive Sectors Of which: Construction & Building Electricity Water Manufacturing (Oil Refining & Others) Extractions B) Services Sectors Of which: Communications Restaurants and Hotels Transportation and Storage Wholesale and Retail Trade Financial Intermediaries and Supporting Services Price Index (%) 2008/ / Change in consumer price index (urban) (January 2007 = 100) Change in producer price index (2004/2005 =100) Monetary Survey (LE bn) July/Dec. 2008/ /2010 End of Period Domestic liquidity (M2) Growth rate (%) Reserve Money Growth rate (%) (1.2) 10.3

6 Money supply (M1) Growth rate (%) Currency in circulation/money supply (%) Banking system foreign assets, of which: CBE foreign assets Banking system foreign liabilities, of which: CBE foreign liabilities Total deposits with banks (excluding the CBE) In local currency In foreign currencies Foreign currency deposits/total deposits (%) Total lending and discount balances extended by banks (excluding the CBE), of which: To government and public economic authorities To business sector (public and private) Portfolio of securities and TBs with banks (excluding the CBE), of which: TBs and government bonds Loans/Deposits with banks (%) Investment in securities, TBs and equity participations/deposits (%) Annual Discount and Interest Rates (%) July/Dec. 2008/ /2010 End of Period CBE Lending and Discount Rate CBE Overnight Deposit and Lending Rates Deposit Lending Interest Rate on Less than 3-Month Deposits Interest Rate on Less than One Year Loans US Dollar Exchange Rate Announced by the CBE (PT/Dollar) - Buy and Sell Exchange Rates (Average of the period) End of the Period (Buy Rate)

7 Consolidated Fiscal Operations of 2009/2010 the General Government (Budget Estimates Actual Sector) FY (July/Dec.) LE bn - Total Revenues Total Expenditures Cash Deficit (or Surplus) Net Acquisition of Financial Assets Overall Deficit (or Surplus) Total Finance Domestic Finance Banking Non-Banking Foreign Borrowing Arrears Others Revaluation Differences Net Privatization Proceeds Difference between TBs Face Value and Present Value Foreign Debt Reclassification Diff. and Related FX Diff Discrepancy Cash Deficit (or Surplus) /GDP (%) 8.4% 4.9% - Overall Fiscal Balance /GDP (%) 8.4% 4.9% - Expenditures /GDP (%) 27.4% 12.9% - Revenues /GDP (%) 19.1% 8.0% Domestic Public Debt End of June 2009 Dec LE bn - Government (net) Public Economic Authorities NIB

8 Balance of Payments US$ bn July/Dec. 2008/ /2010 Current Account & Transfers (2.5) (1.3) Trade Balance (14.6) (11.9) Merchandise Exports Oil and its Products % Others % Merchandise Imports Intermediate Goods % Investment Goods % Consumer Goods % Fuel, Raw Materials and Others % Services Balance Receipts, of which: Transportation % Travel % Investment Income % Payments, of which: Transportation % Travel % Investment Income % Transfers Official % Private % Capital and Financial Account Overall Surplus/(Deficit) (0.5) 2.6 Outstanding External Debt (at End of Dec.)

9 The Leading Article

10 - 1 - Egypt s Yield Curve: Application and Policy Implications Abstract This paper acquires its importance from constructing Egypt's yield curve on the basis of the secondary market information for the government bonds. The purpose of the study is to probe the possibility of using the yield curve as an indicator of the monetary stance. This will positively reflect on the efficiency and transparency of the monetary policy. It tries to examine the predictive power of the yield curve with regards to inflation expectations by employing a structural approach, in order to provide a new macro-economic indicator to best conduct the monetary policy in Egypt. The coverage period of the study ranges from January 2005 throughout February Moreover, a non-structural approach introduced to explore the yield-macro dynamic relationship. The results indicate that the "level" of the yield curve is a good indicator of the stance of the monetary policy, yet the "spread" has registered a minor predictive power unless long-term predictive horizons were considered. Nevertheless, the evidence suggests the presence of a bilateral dynamic relationship between the yield curve and the macro-economy. However, this relationship was in favor of the yield curve's factors to the macro shocks. The study concludes that in order to increase the predictive power of the yield curve, the Central Bank of Egypt represented in the monetary policy should be more responsive to both the yield curve and macro economic shocks through adopting a formal inflation targeting policy. Also, there should be an economy-wide consensus that ensures consistency among the monetary, fiscal policies and the debt management issues in order to successfully achieve the macro-economic stability. Authors: Ahmed Nos'hy is the Assistant Sub-Governor, from the Central Bank of Egypt, Research, Development and Publishing Sector. Dina Rofael is an Economic Researcher from the Central Bank of Egypt, Research, Development and Publishing Sector. Acknowledgment: We would like to thank the Central Bank of Egypt's Team: Mr. Adel Abdel Latif, Mr. Hazim Shehata, Mr. Ahmed Ramy and Mrs. Shaimaa Mostafa, for their helpful assistance. Without their support, this paper would have never been finished; they helped in calculating the yield curve's slope and overcame a lot of the research process's obstacles. Additional thanks are due to Mr. Sameer El-Shenawi from the Central Bank for his valuable comments. Moreover, we are thankful to both Mr. Emad Abdel Hameed and Mr. Abdel Menaem Lotfy from the Ministry of Finance, the Debt Management Unit, for their valuable support in providing the data.

11 - 2 - Table of Contents Introduction: Section One: Literature Review of the Yield Curve and Its Application 1.1 Yield Curve in Theory: The Normal Yield Curve: The Flat or the Humped Yield Curve: The Inverted Yield Curve: 1.2 Yield Curve in Application: Section Two: The Evolution of the Yield Curve and Monetary Policy in Egypt ( ) 2.1 Evolution of the Yield Curve in Egypt over the Period ( ): 2.2 Evolution of the Monetary Policy in Egypt over the Period ( ): Section Three: Egypt's Yield Curve and the Macro-economy 3.1 Testing Egypt's Yield Curve's Predictive Power: 3.2 Testing the Dynamic Interaction between Egypt's Yield Curve and the Macro-economy: Macro Responses to Yield Curve Shocks: Yield Curve Responses to Macro Shocks: Conclusion Policy Recommendations References Appendix

12 - 3-1 Johansen Co-integration test List of Tables 2 Estimation Results of the Inflation Expectation based on the Yield Curve Information - [3months Lag] 3 Estimation Results of the Inflation Expectation based on the Yield Curve Information - [6months Lag] 4 Estimation Results of the Inflation Expectation based on the Yield Curve Information - [9months Lag] 5 Estimation Results of the Inflation Expectation based on the Yield Curve Information - [12months Lag] 6 Choice of VAR Lag Length (Carried out for the Basic Model of Inflation, Deposit, Real T-Bills and Spread) 7 Pair-wise Granger Causality Test 8 The Variance Decomposition of the Spread and the Inflation Rate 9 Estimation Results for Yields-Macro Transition Equation based on the Lag Specification (1 1) 10 The Calculated Share of the Government Bonds held by the Banking Sector 1 Common Yield Curves List of Figures 2 Government Bonds held by the Banking Sector as a Percentage of the Total Listed Government Bonds 3 Egypt's Yield Curve in Selected Time Points 4 Comparison between the Raw Spread Rate and the Calculated Spread Rate 5 Dynamism of the Yield-Macro Relationship 6 Impulse Response Functions for the Yield-Macro Equation based on the Lag specification (11) 7 7Residuals obtained from the Unrestricted VAR (1) Model

13 - 4 - Introduction It has been argued in a wide range of working papers, whether the yield curve has got a predictive power or not -based on its level and slope information- concerning the economic growth, inflation and the interest rate, so it can be considered as a macroeconomic indicator. Kotlán (2002) confirms that the predictive power of the yield slope (spread) is dependent upon the monetary policy. He concludes that, when the monetary authority pays more attention to inflation, i.e. when inflation targeting exists, the predictive power of the yield spread increases (through the monetary authority s reaction function). For Egypt, starting from the year 2005, the Ministry of Finance began to activate Law no.480 and Law no.723 issued in the year 2002, which organize the trading of Primary Dealers in the Treasury securities secondary market. This was to better reflect the true value of the assets and to free their rate of return. Correspondingly, in June 2005, the Central Bank of Egypt (CBE) has announced, that it would adopt a formal inflation targeting framework, upon the fulfillment of certain prerequisites. Over the transition period, the CBE is going to rely on "Overnight Interest Rates" on both the deposits and lending as the operational targets and the outer bounds of the corridor system, in order to curb inflationary pressures. In this context, the study sheds light on Egypt's Yield Curve as a macroeconomic indicator that helps implement the monetary policy in Egypt, especially when inflation targeting framework exists. The methodology used in this paper relies, primarily, on the "Spread" as being the difference between the rate of return on the long term security and that of the short term one. The spread will reflect the shape of the yield curve at each time point. It also incorporates the level of the yield curve as being a potential source of information about the stance of the monetary policy. To explore the yield curve's predictive ability regarding inflation expectations, we have employed the approach offered by Kozicki (1997), the European Central Bank (2006) following Stock and Watson (2003), and others; who use the h-step ahead forecasting regression. In addition, the Unrestricted Vector Auto-Regression (VAR) model and Impulse Response Functions were employed to characterize the dynamism and the direction of interactions between macroeconomic factors (inflation rate and the stance of monetary policy) and yield curve factors (level and slope).

14 - 5 - It is quite obvious that the strength of the dynamic relationship between the yield curve and macro-economy relies intensively on the magnitude of the monetary policy reaction to macro-economic deviations from targets. This was due to the fact that macro-economic factors (e.g.: inflation rate) are affected by the yield curve factors only through the monetary policy tools. Worth mentioning, the slope of the yield curve has got a predictive power over the long term horizons as shown from the VAR analysis. However, it proves to have weak evidence at shorter horizons as illustrated by the structural model of the inflation expectation. On the contrary, the short end of the yield curve represented in the Treasury Bill rate, has proved to be a good indicator of the monetary policy stance; as it moves very close to the proxy of the monetary policy instrument (3 month deposit rate) 1 either in the short or the long predictive horizons. Evidence proves that Egypt's monetary policy has a negligible magnitude of its reaction to macroeconomic deviations, which may weaken the predictive ability of Egypt's yield curve that has been argued to rely intensively on the monetary policy reaction to macroeconomic deviations from targets. This may partly be explained by the fact that the time series covered short period on the one hand, and the main concern of the monetary policy during that period was to face the second round of the supply shock on the other hand. The paper is organized as follows: section one reviews the theories that explain the shapes of the yield curve in addition to featuring a group of working papers that deal with the applications of the yield curve. Section two deals with the evolution of both the yield curve and monetary policy framework in Egypt over the study period ranging from January 2005 to February Section three addresses the econometric approaches to examine the predictive power of Egypt's yield curve, in addition to determining the dynamic interaction between the yield curve and macro-economy. Finally, the study ends up with the conclusion and policy recommendations. 1 The calculations are employed on both overnight rate on deposits and the 3 months deposit rate, although the results are very close, the paper sticks to the 3 months deposit rate being a proxy of the monetary policy instrument in order to reflect the transmission mechanism.

15 - 6 - Section One Literature Review of the Yield Curve and Its Applications Much effort has been made to pick out the information embedded in the yield curve and the term structure in particular. It has been argued in a wide range of working papers, whether the yield curve has got a predictive power or not concerning economic growth, inflation and the interest rate; so it can be considered as a macroeconomic indicator. 1.1 Yield Curve in Theory: The yield curve 2 (also called the term structure of interest rates) is a graphical representation of instruments with various maturities. As illustrated in figure (1), the yield curve can have several shapes. It represents the returns for instruments with differing maturities but having the same risk structure. Figure (1): Common Yield Curves A complete theory of the yield curve must account for three empirical observations. First, the yield curve is typically positively sloped. Secondly, long and short rates tend to move together. Finally, When short-term rates are low, yield curves are more likely to have an upward slope; when short-term rates are 2 Cwik, Paul (May 2004), An Investigation of Inverted Yield Curves and Economic Downturns, Doctor of Philosophy, the Graduate Faculty of Auburn University, Auburn, Alabama.

16 - 7 - high, yield curves are more likely to slope downward and be inverted. The third factor restated is that long rates tend to remain stable relative to short rates The Normal (Regular) Yield Curve: The yield curve has usually been "normal", meaning that yields rise as maturity lengthens (i.e., the slope of the yield curve is positive). This positive slope reflects the "Liquidity Preference Hypothesis" and stems from Keynes's line of reasoning on the demand for money. According to this model, money, the most liquid asset has no interest rate and less liquid asset will yield higher return as long as the maturity of the asset increases. Moreover, the steepness of the yield curve stems from the investors' expectations that the economy would grow in the future and, importantly, for this growth to be associated with a greater expectation that inflation will rise in the future rather than fall. This expectation of higher inflation leads to expectations that the central bank will tighten monetary policy by raising short term interest rates in the future to slow economic growth and dampen inflationary pressure. It also creates a need for a risk premium associated with uncertainty about the future rate of inflation and the risk this poses upon the future value of cash flows. Investors price these risks into the yield curve by demanding higher yields for maturities further into the future. Worth mentioning, the degree of steepness of the curve will reflect the confidence level in the implemented monetary policy, i.e. the more steepness of a positive yield curve, the greater the degree of uncertainty and less confidence in the implemented monetary policy and vice versa The Flat or the Humped Yield Curve: A flat yield curve is observed when all maturities have similar yields, whereas a humped curve results when short-term and long-term yields are equal and medium-term yields are higher than those of the short-term and long-term. A flat curve sends signals of uncertainty in the economy. This mixed signal can revert back to a normal curve or could later result into an inverted curve. The flat yield curve will send mixed signals of uncertainty for the following two reasons:

17 - 8 - i. It is not clear to the market participants during these episodes, whether the yield curve will invert into upwards or downwards shape, i.e.: the market is not certain whether the economy will experience expansionary or tight conditions in the future. ii. The market environment is sending mixed signals to investors, who are interpreting interest rate movements in various ways. It is difficult for the market to determine whether interest rates will move significantly in either direction farther into the future The Inverted Yield Curve 3 : An inverted yield curve occurs when long-term yields fall below short-term yields. Under this abnormal and contradictory situation, long-term investors will settle for lower yields in this case, if they think the economy will slow or even decline in the future. An inverted curve may indicate a worsening economic situation in the future. In addition to potentially signaling an economic decline, inverted yield curves also imply that the market believes inflation will remain low. This is because, even if there is a recession, a low bond yield will still be offset by low inflation. In the framework of the interest rate and yield curve theories, six major theories have evolved; the Expectations Hypothesis Theory, the Liquidity Preference Hypothesis, the Segmented Markets Theory, the Preferred-Habitat Theory, a Stochastic-Process No-Arbitrage Approach, and a theory unique to Rothbard. Our concern is summarized in the Expectations Hypothesis Theory, where the framework of analyzing key economic factors determines how the yield curve responds to monetary policy. The paper is going to rely on the Expectations Hypothesis Theory for the reason that it has a vast range of applications especially in testing the predictive ability of the yield curve in the prediction of inflation and output in both the developed and the emerging economies. According to the Expectations Hypothesis Theory, the interest rate on the security contains 2 elements: information about financial market expectations of monetary policy over the life of the security and a term premium to compensate for risk. Moreover, The Expectations Hypothesis provides a simple way to link monetary policy actions to fluctuations of bond yields and forward rates, since for example, the 10-year rate will change whenever these rates change. 3

18 - 9 - The Expectations Hypothesis Theory assumes that all financial instruments along the yield curve are perfectly substitutable. The shape of the yield curve is dependent upon two factors: present short interest rates and expected future short interest rates. As described by Fisher s model, short interest rates are determined by both time-preference and the marginal productivity of capital. The n-period model is demonstrated by the following equation: i n, t = i i 1,t + i e 1,t+1 + i e 1,t i e 1,t+ n -1 n Where i n,t is the yield for an n-year bond for time period t, and i e 1,t+1 is the expected interest rate for a one-year bond at time period t+1. The long rates are determined based on the expected movements of future short rates. The theory assumes that investors are endowed with rational expectations, face low information costs and operate in a market where all assets prices are reflected in their fundamental value. In equilibrium, the price of a bond equals its discounted present value. A positively sloped yield curve will emerge only if economic actors believe that future interest rates will rise. With the expectation of higher future rates, the future returns are discounted by a larger interest rate. As a result, the present value of bonds falls. Investors enter into forward contracts to hedge against interest rate risk. The forward rates act as an unbiased estimate of future spot rates, causing the yield curve to have a positive slope. A limitation of this theory is that it can only explain the rotations of the yield curve but cannot explain the shifts. A second key factor influencing both near-term and longer-term rates is a term premium. The latter represents the extra compensation for risk that an investor may require for extending the maturity of his investment. In addition, there is considerable evidence suggesting that the term premia are time varying and so may contribute to changes in the yield curve over time. Term Premia may change for several reasons; institutional features such as variation in the relative supply and demand for treasury securities may be reflected in term premia, in addition to the investors' inflation perceptions 4. 4 Kozicki, Sharon and Gordon Sellon (2005), "Longer-Term Perspectives on the Yield Curve and Monetary Policy", Economic Review, Fourth Quarter 2005, Federal Reserve Bank of Kansas City.

19 Recent advances in yield curve theory focus on the relationship between the yield and macroeconomic variables such as growth, inflation, and future interest rates. The yield curve is often used as an indicator of the type of monetary policy being pursued, i.e. the stance of the monetary policy. 1.2 Yield Curve in Application: The Expectations Hypothesis Theory leads to the idea of the yield curve as a predictor of future inflation, positing that long-term rates are based upon the investors expectations of future interest rates (and therefore inflation). An increase in current long rates indicates that investors expect future short rates to increase. Several authors have found considerable evidence of a relationship between the yield curve and inflation. Despite these positive results in this literature, there are those who have found that the relationship does not hold. In addition to those who have found results that support and deny the relationship between the yield curve and inflation, there are those who find mixed results. They all find some predictive power of the yield curve to forecast inflation, but the results are relatively weak. A recent study by Blake, Henry and Robertson (2002), finds some informational content at the longer rate segments, but it admits that these results are not robust. Instead of using the Expectations Hypothesis Theory, Ferderer and Shadbegian (1993) add a term premium to the theory. They find that market participants gradually learned about changes in monetary policies. In other words, authors claim that market forecasts will accurately reflect market beliefs, but only after a period of time when investors figure out the latest Central Bank policy 5. Current monetary policy has a significant influence on the yield curve spread and hence on real activity over the upcoming several quarters. A rise in the short rate tends to flatten the yield curve as well as slow real growth in the near term. This relationship, however, is only one part of the explanation for the yield curve s usefulness as a forecasting tool. Expectations of future inflation and real interest rates contained in the yield curve spread also seem to play an important role in the prediction of economic activity. The yield curve spread variable examined here corresponds to a forward interest rate applicable from three months to ten years into the future. As explained in Mishkin (1990a, 1990b), this rate can be decomposed into expected real interest rate and 5 The paper will show that the evidence in Egypt has proved this argument, since the yield curve is able to expect inflation after 16 months of the CBE reaction to inflation shocks; this will be illustrated in detail in part three of the study.

20 expected inflation components, each of which may be helpful in forecasting. The expected real rate may be associated with expectations of future monetary policy and hence of future real growth 6. Moreover, because inflation tends to be positively related to activity, the expected inflation component may also be informative about future growth 7. Evans and Marshall (1998) find evidence that contraction monetary policy increases term premia for shorter maturities, raising real interest rates. A money supply shock raises the level of the yield curve but reduces its slope and curvature. The effects of the monetary shock on the slope and curvature dissipate in 4-6 months and the yield curve returns to its original level within 6 months. The authors results for long-term rates fit the Expectations Hypothesis. Estrella (1998) had conducted a rational expectation model based on the following economic components: the Expectations Hypothesis, a short-run Phillips curve, the IS curve, a monetary authority reaction function, and the Fisher Equation. The model suggests that the empirical regularities are not structural but are significantly affected by national monetary policy. When the monetary authority reacts to levels of national income below full employment, the yield curve is an optimal predictor, for recession prediction purposes. Kotlán (2002) confirms that the predictive ability of the yield spread is dependent on the monetary policy. However, his results oppose Estrella (1998). Kotlán (2002) concludes that, when the monetary authority pays more attention to inflation, i.e. when inflation targeting exists, the predictive ability of the yield spread increases (through the monetary authority s reaction function). Kim (2000) based his analysis on the Expectations Hypothesis; he decomposes the yield curve into an expectations effect and a term premium effect. While both variables are significant, Kim (2000) performs a Wald test to determine that the expectations effect is slightly stronger 8. 6 This conclusion can be applied to developed countries but perhaps not upon Egypt's case, since Egypt's economy historically suffered from "Inflationary Recession", that's why each case of recession and inflation prediction should be separately examined. 7 Estrella, Arturo and Frederic S. Mishkin (June 1996), The Yield Curve as a Predictor of U.S. Recessions, Federal Reserve Bank of New York, Current Issues in Economics and Finance,VOL.2, Number 7. New York, U.S.A. 8 Cwik, Paul (May 2004), An Investigation of Inverted Yield Curves and Economic Downturns, Doctor of Philosophy, the Graduate Faculty of Auburn University, Auburn, Alabama.

21 It was argued by (Estrella, 2005) that the slope of the yield curve has empirically shown to be a significant predictor of inflation and real economic activity, but there is no standard theory as to why the relationship exists. Estrella constructed an analytical rational expectations model to investigate reasons behind the empirical results; the model suggests that relationships are not structural but are rather influenced by the monetary policy regime. However, the yield curve should have predictive power in all circumstances. A conclusion frequently cited, is that the yield curve tends to flatten when there is tightening of monetary policy. This is likely to happen due to the fact that when the monetary authority decides to adopt a tightening policy, this implies that short term interest rate will increase followed by an increase in the short end of the yield curve (a good indicator of the stance of the monetary policy). This action will induce the market participants to have a higher degree of confidence in the implemented policy, therefore, they will expect that the monetary authority is capable to curb the inflationary pressures in the future; consequently, they will not demand an extra risk premium for the longer maturities, that will keep the interest rate on these maturities at the same level, and at the meantime, short term interest rates are increasing, that's why a tightening monetary policy is likely to flatten the shape of the yield curve. Importantly, if the monetary policy is essentially reactive to deviations of inflation from target and output from potential, the predictive relationships for output and inflation depend primarily on the magnitudes of the reaction parameters. If the monetary authority systematically optimizes to achieve certain goals with regard to inflation and output variability, the predictive power of the yield curve is more directly dependent on the structure of the macro-economy 9, i.e. the more responsiveness of the monetary policy in terms of higher magnitudes to inflation and output deviations from targets, the more the predictive power of the yield curve. As mentioned above the evidence of the predictive role of the yield curve has been justified for the industrial and developed countries, the standard economic rationale for these findings is that the slope of the yield curve is a monetary policy indicator. Again the monetary tightening results in short-term interest rates are high relative to long-term interest rates. According to Fisher equation, the nominal interest rate reflects market expectations of both future 9 Estrella, Arturo (February 2005), "Why Does the Yield Curve Predict Output and Inflation?", Federal Reserve Bank of New York, New York, U.S.A.

22 inflation and the real rate for a given maturity. The slope of the yield curve should therefore reflect expected inflation, in line with this, Mishkin (1990) finds predictive content of the US yield curve for domestic inflation. However, the absence of evidence for emerging economies reflects the very reason that the bond markets have started to deepen significantly only since the turn of the millennium. That's why Stock and Watson have concluded that "Universality of the predictive power of the yield curve in predicting recession and inflation is unresolved". The development of domestic debt security markets in these economies in the very recent years reflects their efforts to self-insure against "sudden stops" in international capital flows. Many economies - including Egypt's - have managed to extend debt duration and to give emphasis on longer maturities as an aspect of government securities diversification 10, parallel to the availability of the short term instruments for liquidity purposes. The following part of the study will be dealing with the evolution of the monetary policy and the formation of Egypt's Yield Curve over the study period starting from the beginning of primary dealers' system activation in 2005 which coincides with the formal inflation targeting framework announced by the Central Bank of Egypt since June Mehl, Arnoud (2006), "The Yield Curve as a Predictor and Emerging Economies", Bank of Finland, Institute for Economies in Transition, BOFIT Discussion Papers.

23 Section Two The Evolution of the Yield Curve and Monetary Policy in Egypt ( ) Revisiting the Treasury bonds market in Egypt is very important to have a clue about the degree of activeness of these bonds and consequently to have a preliminary overview about Egypt's yield curve and its predictive ability. In other words, do the information incorporated in the yield curve reflect the market anticipations and perceptions or not? If not, what should be done to increase the predictability of this indicator to better conduct the monetary policy in Egypt? Tracing the government bonds held by the banking sector (except the Central Bank of Egypt) as a percentage of the total listed government bonds, the paper notices that although this share began to decline in 2007 and 2008 as obvious in figure (2) to about 62% and 64% on average respectively, compared to 89% on average in 2005, it is still large in magnitude, implying inactive treasury bonds market. This gives an impression of a modest predictive ability for Egypt's yield curve (see table (9) in the appendix). Figure (2): Government Bonds held by the Banking Sector as a Percentage of the Total Listed Government Bonds 100 (%) Feb-05 Apr-05 Jun-05 Aug-05 Oct-05 Dec-05 Feb-06 Apr-06 Jun-06 Aug-06 Oct-06 Dec-06 Feb-07 Apr-07 Jun-07 Aug-07 Oct-07 Dec-07 Feb-08 Source: Capital Market Authority and Central Bank of Egypt, Monthly Bulletin, Various Issues.

24 Less active treasury bonds implies they have a modest ability to reflect the market anticipations about future inflation or recession, therefore less ability to reflect the investors preferences towards long or short term securities. Unfortunately, this situation will lessen the predictive ability of the yield curve, unless more active treasury bonds market will exist in the future. Introducing the yield curve and the construction of this important indicator is a very recent issue in developing countries in general and in Egypt in particular. Although Egypt's yield curve could have been constructed since the activation of primary dealers system in November 2004, the paper's coverage period starts from January 2005, given that no significant amount of government securities have been traded in the last two months of Evolution of the Yield Curve in Egypt over the Period ( ): In early 2005, Egypt's yield curve has experienced instability. Throughout years 2005 and 2006, it used to have irregular flat shapes, implying uncertainty in the economy. Throughout 2007 and the beginning of 2008, it takes the normal shape being upward sloped as illustrated in figure (3), indicating that there is some anticipation in the economy of an upcoming inflation. That's why it used to be steeper reflecting less confidence in the implemented monetary policy. Worth noting, flat yield curves posses uncertainty due to the reason early mentioned in p.10. However, the more steeper upward sloping yield curve would suggest less confidence in the implemented monetary policy; consequently, this will create an uncertain situation represented in the anxiety of the investors related to the risk poses upon their future value of cash flows, leading to a need for a risk premium associated with uncertainty about the future rate of inflation. Investors price these risks into the yield curve by demanding higher yields for maturities further into the future. In this context, after the construction of the yield curve at different time points starting from January 2005 until February 2008, the paper went on calculating the spread as an indicator on the yield curve's slope to be able to test its predictive power in the next part of the study. The paper stuck to the two maturities that most of the economic and financial literature have settled down on and have proved to have a good predictive power, those are the 10 years Treasury Bonds Rate and the 3 months Treasury Bills Rate, so the spread is calculated using the following formula: Spread t = (10 years Treasury Bonds Rate) t (3 Months Treasury Bills Rate) t

25 In this regard, the paper faced some obstacles, since it found differentiated time points in our sample where the Treasury bonds with 10 years maturity were not traded; instead, two alternative solutions have been conducted: 1. The rate of return of the nearest maturity to 10 years bonds has been used, based on the raw data. 2. The paper calculated a hypothetical rate for the 10 years based on the Yieldto-Maturity formula (Kolb and Rodriguez,1996), which states that: Figure (3): Egyptian Yield Curves at Selected Time Points (%) Egyptian Yield Curve in December days T-bills 182 days T-Bills 364 days T-bills Days T-Bills Egyptian Yield Curve in June Days T-Bills 364Days T-Bills Egyptian Yield Curve in May 2007 Egyptian Yield Curve in January days T-bills 182 days T-Bills 364 days T-bills days T-bills 182 days T-Bills 364 days T-bills Source: Capital Market Authority and Central Bank of Egypt, Monthly Bulletin, Various Issues.

26 Yield-to-Maturity = C + [(F-P)/n] [(F+2P)/3] Where: C: Coupon Rate F: Nominal Value of the Bond P: Market Value of the Bond n: Bond's Maturity 11 However, it has been noticed that there is no big difference between the two spreads, whether raw or estimated on the basis of the yield to maturity formula. Instead, they coincide in some time points as illustrated in the following chart. Figure (4): Comparison between the Raw Spread Rate and the Calculated Spread Rate (%) Jan 05 Mar 05 May05 Jul05 Sep05 Nov05 Jan 06 Mar 06 May06 Jul06 Sep06 Nov06 Jan 07 Mar 07 May07 Jul07 Sep07 Nov07 Jan08 Spread (based on calculated 10 year bonds rate ) Spread (raw data) Source: Capital Market Authority and Central Bank of Egypt, Monthly Bulletin, Various Issues, and the calculated series is based upon the authors' calculations. 11 Kolb, W.Robert and Richard J.Rodriguez (1996), "Financial Management", Second Edition, Blackwell Publishers, pp

27 Evolution of the Monetary Policy in Egypt over the Period ( ): The following part presents a summary on monetary policy implemented during the FY 2004/2005 till the FY 2006/2007. In FY 2004/2005, the CBE intended to set and implement a formal inflation targeting framework upon the fulfillment of certain prerequisites, in order to anchor the inflation expectations a matter that will enhance the transparency of the monetary policy in Egypt. It planned to steer short term interest rates to meet inflation targets in the transition period. In this context, the CBE has developed a new framework for implementing the monetary policy that relies on the "Overnight Interbank Interest Rate" as being the operational target, which provides the outer bounds of the corridor. The ceiling of this corridor is the overnight rate on lending from the CBE and the floor is the overnight interest rate on deposits at the CBE. This system has been put into action since 5 June 2005, as the Monetary Policy Committee (MPC) determined deposit and lending rates to be 9.5% and 12.5% respectively, in its first meeting. In FY 2005/2006, the CBE continued on realizing price stability as the overriding objective of its monetary policy. The overnight deposit and lending rates have been reduced a number of times to reach 8% and 10% respectively in June 2006, narrowing the corridor to 2% compared to 3% since these rates have been set for the first time in June Then in the MPC's meeting, held on 2 November 2006, the overnight deposit and lending rates increased to reach 8.5% and 10.5% respectively. In addition, the CBE continued to absorb the excess liquidity in the banking sector through Open Market Operations (OMO). It began to issue a new instrument namely; Certificates of Deposits (CDs) with maturities extending across one year and the CBE notes of maturities one to two years, the CBE sells these two instruments to banks through outright sales. In FY 2006/2007, the MPC decided to raise the CBE's key policy rates (the overnight deposit and lending rates). This was to curb the inflationary pressures resulted from the accelerated economic growth and more importantly, to firmly face the second round effects of the supply shocks stemming from the avian flu crisis and oil subsidy cuts. According to CBE's annual report of 2006/2007, the monetary policy transmission mechanism showed an improvement as the interest rates on clients' deposits and loans became more responsive to changes in the CBE's overnight interest rates. The interest rates on three-month deposits reached about 6.6% by the end of July 2006, and 6.8% in October 2006; then ranging from 6.7% to 6.8% from November 2006 till June Meanwhile, interest rates on loans of one year and less were relatively less flexible Central Bank of Egypt, "Annual Report", Various Issues.

28 As noticed, it appears that the short-term interest rates have used to move close to the key rates of the monetary policy (i.e. overnight deposit rate). In addition the CBE's overnight deposit rate and short term deposit rate in banks which have been econometrically examined to guarantee that they are cointegrated using the "Johansen Co-integration Test". This test has been employed over the time period December 2001 till February 2008, to reflect the relationship between the monetary policy instrument (overnight interbank interest rate) and the short term 3-months deposit rate in banks. Moreover, the test covered a longer time period to overcome the limited dataset that the study has relied upon due to institutional considerations. The test's statistics: max Eigen-value and the trace statistics have indicated one co-integrating equation at both 5% and 1% levels [Refer to table (A)]. This result is likely to reflect the presence of a long-run equilibrium relationship between the two interest rates, i.e. they can replace each other, reflecting the existence of the transmission mechanism. The implemented monetary policy over the study period that ranges from January 2005 to February 2008 was characterized by facing the second round of the supply shocks that Egypt's economy has faced. That's why the monetary policy appears to have a negligible reaction to the macroeconomic deviations, and consequently less predictive ability for Egypt's yield curve.

29 Table (A): Johansen Co-integration Test Sample(adjusted): 2002: :02 Included observations: 73 after adjusting endpoints Trend assumption: Linear deterministic trend Series: Overnight Rate, 3-months Deposit Rate Lags interval (in first differences): 1 to 1 Unrestricted Co-integration Rank Test Hypothesized Critical Value No. of CE(s) Eigen-value Trace Statistic 5 Percent 1 Percent None ** At most *(**) denotes rejection of the hypothesis at the 5%(1%) level Trace test indicates 1 co-integrating equation(s) at both 5% and 1% levels Hypothesized Critical Value No. of CE(s) Eigen-value Trace Statistic 5 Percent 1 Percent None ** At most *(**) denotes rejection of the hypothesis at the 5%(1%) level Max-Eigen-value test indicates 1 co-integrating equation(s) at both 5% and 1% levels Note: Lag interval has been determined according to the Schwartz criterion.

30 Section Three Egypt's Yield Curve and the Macro-economy For the purpose of testing empirically whether Egypt's yield curve is a monetary policy indicator, in addition to having a predictive power or not regarding the inflation expectation based on the information embedded in the yield curve (the level and slope), two approaches have been employed: i. First, regressing the monthly inflation rate (annualized) on the yield curve information (lagged) to determine the time horizon at which the yield curve has a relevant predictive power based on the approach carried out by Kozicki (1997), the European Central Bank (2006) following Stock and Watson (2003) and others; who use the h-step ahead forecasting regression which uses the k th lag of the slope of the yield curve. ii. Second, employing Unrestricted Vector Auto-Regression (VAR) models and Impulse Response Functions to characterize the dynamism and the direction of interactions between the macroeconomic factors (inflation rate and the stance of monetary policy) and the yield curve factors (level and slope) simulating the popular concept of the latent factor model of Nelson and Siegel (1987). 3.1 Testing Egypt's Yield Curve's Predictive Power: Testing the predictive power of the yield curve in expecting both the real activity and inflation is not a recent issue. However, a group of developed countries and recently the emerging economies have employed statistical and econometrical approaches to meet this regard 13. The main concern of this paper is to examine how far the yield curve is able to clearly reflect the inflation expectations relying on the information embedded in the yield curve. The approach simply utilizes the most popular financial market variable: the "Yield Spread" between yields on long-term and short-term government instruments, representing the shape of the yield curve at the same time Mehl, Arnaud (November 2006), "The Yield Curve as a predictor and Emerging Economies", The European Central Bank, Working Paper Series, Number 691, Frankfurt, Germany. 14 Kozicki, Sharon (1997), "Predicting Real Growth and Inflation with the Yield Spread", Economic Review, Fourth Quarter, Federal Reserve Bank of Kansas City.

31 The paper uses monthly data for both the Treasury Bill Rate and the Treasury Bonds Rate traded in the secondary market based on the Yield-To- Maturity (YTM) rate, where the time series ranges from January 2005 till February The paper uses a short time series on the backdrop of the following: To employ this approach, the paper sticks to those securities which are traded in the secondary market to make sure that the rate of return reflects the bonds market mechanism and free interest rates. Moreover, Law No. 480 and Law No.723 issued in the year 2002 by the Ministry of Finance, which organizes the Primary Dealers trading in the bonds secondary market, were only activated in November For the aforementioned reasons, the obtained results are to be considered with caution. The implemented methodology incorporates both the yield curve level and slope as being both potential sources for information about future economic conditions. Several hypotheses argue that the information in the yield curve is forward-looking and therefore should have predictive power, among these arguments are the following: The yield spread reflects the stance of the monetary policy. The yield spread reflects the direction of future inflation changes. The level of the yield curve as measured by its short end might help expect inflation, because short term interest rate may provide information on the stance of the monetary policy as they move closely to the monetary policy instrument. Moreover, fluctuations in the yield curve may either reflect shifts in policy or shifts in the risk premium. That's why there is an argument that the yield curve level can provide a better measure of the stance of monetary policy than the one offered by the yield spread 16. This section of the study examines whether Egypt's yield curve helps reflect inflation expectations. The Ordinary Least Squares (OLS) method is employed, whereas the inflation rate is treated as the endogenous variable, which is a function of the spread lagged, a proxy for real Treasury Bill Rate 17 (lagged) the interaction between the lagged spread with the real bill rate, and 15 Ministry of Finance (MOF): 16 Ibid, Sharon Kozicki, Real rate is calculated as the difference between the nominal bill rate (3-month maturity) at time (t) and the annualized inflation rate at time (t).

32 the inflation rate (lagged). [Refer to Tables (1-4) in the appendix]. Worth noting, the paper controls for the heteroskedasticity and autocorrelation (HAC) by applying Newey-West Correction following Stock and Watson (2003). The implemented approach utilizes the h-step ahead forecasting regression which uses the k th lag of the slope of the yield curve carried out by Kozicki (1997), the European Central Bank (2006) following Stock and Watson (2003). Evidence suggests that the Treasury Bill Rate embedded in Egypt's yield curve proves to have a predictive power, or more precisely, it is a good indicator of the monetary policy stance over the 6, 9 and 12 prediction horizons that have been examined, since estimated results indicate significant estimates and the signs are correct and coinciding with both the economic and the financial theories. It has got a negative sign with the expected inflation; implying that if it increases at any horizon, the inflation is expected to decline significantly. On the contrary, the slope factor of the yield curve is unable to reflect the inflation expectations over the tested horizons, since its estimated coefficients appear to be insignificant, in all regressions, except in the 9 lag regression. If things are going on the right track, the slope should have a positive sign, implying that if it increases over a past period of time, this will be a signal of market players' anticipation of an upcoming inflation, which will induce the monetary authority to have some protective actions. Regarding the interaction term between the spread and the Treasury bill rate that reflects the indirect prediction of these two components of the yield curve. It appears to be significant only in the 12 lag regression; it reflects the indirect prediction only for the T-bill rate as the spread used to be insignificant in all regressions except for the 9 lag regression. Worth noting, in the latter regression, although the spread shows some significance, the interaction term reflects no effect. The above mentioned results do conform to the theory because it has been argued that the short end of the yield curve will reflect the stance of the implemented monetary policy, while the slope of the yield curve predictive power in inflation expectations has registered weak evidence unless long term horizons are considered starting from 2 to 3 years.

33 Unfortunately, our sample is so limited, so our structural analysis was restricted to 1 year prediction horizon only. However, the rest of the analysis will deal with the dynamic interaction between the yield curve factors and the macro-economic indicators in a nonstructural framework to get a clue about the ability of the yield curve expectations at longer horizons. 3.2 Testing the Dynamic Interaction between Egypt's Yield Curve and the Macro-economy: Theoretically, the framework to analyze the dynamic interaction is far from being a settled issue. On one hand, there is no consensus on the way to estimate the term structure of the interest rates. On the other hand, the incorporated macroeconomic variables create fundamental differences in the conclusions obtained from empirical analysis. One approach is to provide a structural representation of macro-economy, usually by means of a small (linear) model or consistently modeling long run expectations about future inflation as what we have carried out in the previous section. The other possibility is to use a non-structural representation of the macro-economy, considering observed macro variables or latent macro factors (level, curvature and the slope of the yield curve) 18. The paper examines the dynamics of the yield curve factors (level and slope) along with the macro-indicators (stance of the monetary policy and inflation rate) via the impulse response functions as recommended by a group of studies 19. We consider two groups of impulse response in order to get a clue about the direction of the dynamic interactions between the macro-economy and the yield curve in Egypt; those are: 1. Macro response to yield curve shocks. 2. Yield curve responses to macro shocks. 18 Morales, Marco (March 2007), "The Yield Curve and Macroeconomic Factors in the Chilean Economy", Center of Economics and Finance, Diego Portales University, Chile. 19 We estimate the VAR model incorporating the inflation rate, deposit rate, real t-bill rate and the spread in the level form in spite of being I(1) according to Augmented Dickey Fuller results. This approach simulates many studies that concentrated on the dynamic interaction between the yield curve and the macro-economy which employ the VAR analysis with level non-stationary variables, due to the fact that there is a trade-off between the statistical efficiency and the potential loss of information that takes place when economic time series are differenced.

34 An unrestricted VAR model has been employed to trace these dynamics [Refer to the appendix Table (5) for lag specification criteria], the model has specified a lag interval (1 1) based on the information criterion provided by Schwarz (SIC), Final Prediction Error (FPE) and Hannan-Quinn (HQ) criteria. The Generalized impulse-response functions for unrestricted VAR were carried out to avoid the ordering problem: changing the order produces different impulse response functions for each variable as proposed by Pesaran and Shin (1998) Macro Responses to Yield Curve Shocks: Generally speaking, the monetary policy instrument represented in the 3- months deposit rate, registers very negligible responses in terms of magnitude to shocks either from the yield curve factors (level and slope) or from the inflation rate as illustrated in Table (8). It reacts positively and insignificantly to inflation shocks over a period that ranges from the upcoming 1 to 10 months, with peak effect in the fifth month. i.e. the monetary policy reaction function is less sensitive to the YC shocks. This evidence might not be in favor of the Central Bank of Egypt's reaction function, implying that the latter lacks the link between the anticipations of bond market players and the implemented monetary policy. For the inflation rate responses, the paper easily observes that both Treasury Bill Rate (level of YC) and the spread (slope of YC) shocks have a far lasting effect on inflation rate, since inflation responses extend over the upcoming 10 months. In addition, for the spread shock, the reaction pattern reversed to be positive starting from the fourteenth months to reach its peak after approximately 2 years and half of the shock. [See Impulse-Response Estimates Figure (6) in the appendix] Meanwhile the inflation rate responds negatively to the deposit rate shocks, the reason for that stems from the following: For the deposit rate, when the Central Bank of Egypt (CBE) decides to increase the short run rates, the inflation will respond directly to this action and the inflation decline accelerates to reach its trough in the eighth month, it continues until its effect disappears completely within 2 years and half Yield Curve Responses to Macro Shocks: Now consider the response of the yield curve factors to the macro variables shocks. While the slope factor shows minor reactions to the inflation and the deposit rate, the level factor reacts directly to these shocks.

35 Regarding the Treasury bill reaction to the monetary policy instrument, one should expect one of two alternative effects: either a large degree of credibility and transparency in the central bank to suppress inflation and likely lower the level factor, or a tightening monetary policy reflecting the anxiety of the central bank authority which will increase inflation expectations and consequently increase the level factor of the yield curve 20. Evidently, in the sample, the second effect has dominated and the real Treasury bill rate didn't react to the inflation hikes directly. However, it affects inflation through its positive and significant reaction to the deposit rate shocks to reflect the stance of the monetary policy as illustrated in the estimated results of both the transition equation and the impulse-response function [Refer to the appendix], that can be attributed to the following: First, a surprising tightening policy will indicate that the central bank is worried about overheating and inflationary pressures in the economy that boost future inflation expectations and induced the investors to demand higher rate of return on short run securities. Second, as a matter of fact, 3-month deposit is considered a strong rival (perfect substitute) for short run securities. That's why any increase in the deposit rate would encourage the investors to shift from the Treasury securities -supported by the high degree of liquidity- to the bank deposits to enjoy the additional return. 20 Diebold, X. Fancis, Glenn D. Rudebusch and S. Boragan Aruoba (October 2003), "The Macro-economy and the Yield Curve: A Nonstructural Analysis", Federal Reserve Bank of San Francisco.

36 Figure (5): Dynamism of the Yield-Macro Relationship Inflation (t) +ve Monetary Policy Reaction to Inflation Shock Deposit Rate (t+5) -ve Treasury Bills Rate (t+6) to (t+16) Spread (t+7) to (t+28) -ve Inflation (t+7) to (t+17) +ve Inflation (t+21) to (t+30) Stance of Monetary Policy Expected Inflation Estimates show some interesting conclusions about the macro-yield dynamics: inflation starts to go down after 7 months of the inflation shock and its decline extends over the following 30 months before it restores its initial rate once more, i.e. it takes a period from CBE that extends to 2 years and half to dampen inflationary pressures. In addition, inflation reaches its trough 8 months after the monetary policy shock which is equivalent to 13 months of the inflation shock, as illustrated in the appendix Figure (6). The mechanism starts by a rise in inflation that will induce the Central Bank to raise the deposit rate as being the monetary policy instrument 5 months after the shock. This monetary action will be followed by two effects [Refer to figure (5)]: 1. Treasury Bill Rate will follow the deposit rate increase as being a good indicator of the monetary stance; this action will push the inflation to decline over a period that ranges from 7 to 17 months of the inflation rise.

37 Spread will react negatively to the T-bill rate increase, cause the yield curve to flatten, implying the market anticipations of a future decline in the inflation rate; these expectations will cause an inflation decline over a period ranging from 21 to 30 months of the inflation shock. It is obvious that the strength of the dynamic relationship between the yield curve and macro-economy intensively relies upon the magnitude of the monetary policy reaction to macro-economic deviations from targets, since macro-economic factors (e.g. inflation rate) are affected by the yield curve factors only through monetary policy tools. The slope of the yield curve has got a predictive power over long term horizons as shown in the VAR analysis. However, it proves to have weak evidence at shorter horizons as illustrated by the structural model of the inflation expectation. On the contrary, the short end of the yield curve represented in the Treasury bill rate, has proved to be a good indicator of the monetary policy stance; as it moves very closely to the monetary policy tool in short or long predictive horizons. Moreover, to cross check the robustness of the previously estimated results, the paper introduced the Pair-wise Granger Causality test and the Variance Decomposition estimates for both the spread and the inflation rate [Refer to tables (6) and (7) in the appendix]. The former shows almost the same conclusions that have been early mentioned. In the framework of this approach, the inflation appears to granger cause the spread but the reverse doesn't apply. In addition, the Treasury bill rate granger causes the spread, however, the spread doesn't granger cause the Treasury bill. This test also suggests that there is no granger causality relationship between the Treasury bill and the inflation rate. Regarding the variance decomposition analysis, it has been conducted twice: once for the spread and the other for the inflation rate. For the spread, the Treasury bill rate appears to have the major and an everlasting effect on the spread; this can be explained from the fact that T-bill rate is a good indicator for the monetary stance, so if it experiences any change, this will affect the spread, i.e. the shape of the yield curve. Meanwhile, both the inflation rate and the deposit rate have a peak effect in the 9 th and the 23 rd month respectively, on the spread. On the other hand, the spread and the Treasury bill appear to have a very minor effect on the inflation rate variation. At the meantime, both the inflation and the deposit rate are responsible for about 90% of the variation experienced by the inflation rate over the upcoming two years. These conclusions suggest that the yield-macro bi-directional dynamic relationship moves from the macroeconomy to the yield curve.

38 Conclusion The incorporated time series were in level form because differenced time series in VARs will really generate efficient estimates but it will ignore potential long-run relationships that are of great importance 21. Moreover, estimation results based on our sample that ranges from January 2005 till February 2008 support the dynamic interaction between yield curve factors (level and slope) and macroeconomic variables (inflation and the stance of monetary policy). The evidence suggests the following conclusions: Despite that the yield curve represented in both its level and slope is able to expect the upcoming inflation at higher prediction horizons as shown in the transition unrestricted VAR equation and the impulse-response functions, the results are not so robust and should be taken with great caution due to data limitations. Moreover, the estimated results indicate that the magnitude of the YC slope's effect on the deposit rate is lesser than its effect on the inflation rate, i.e. the monetary policy reaction function is less sensitive to the YC shocks; apparently, this evidence might not be in favor of the Central Bank of Egypt's reaction function, implying that the latter lacks the link between the bond market players anticipations and the implemented monetary policy. However, this situation resulted from the inactive bonds market where 89%, 75.5% and 64% of the listed bonds are held by the banking sector in 2005, 2006 and 2007 respectively. Evidence has proved that Egypt's monetary policy has a negligible magnitude concerning its reaction to macroeconomic deviations where the CBE doesn't react to inflation rises immediately; the reason for that stems from the fact that the implemented monetary policy over the study period was characterized by facing the second round of the supply shocks. That's why the monetary policy appears to have a negligible reaction to the macroeconomic deviations, and consequently less predictive ability for Egypt's yield curve. 21 Ramaswamy, Ramana and Torsten Sloek (December 1997), "The Real Effects of Monetary Policy in the Euro Area: What are the Differences?", International Monetary Fund, Research Department, IMF Working Papers Number 97/160.

39 Macro variables have strong effect on the level factor of the yield curve but not on its future movements (i.e. term-structure or spread). However, much weaker evidence upon the reverse influence of the YC factors on the macro variables exist, unless higher lag lengths are considered. This indicates that the monetary policy reaction function does not take into account the bonds market expectations about the macro-economy in the near prediction horizons. This result is consistent with most empirical results in both developed and emerging economies. The incidence of a structural positive relationship between the monetary policy and the level factor of the yield curve implies that the short end of the yield curve is determined by and follows the monetary policy which assures the argument that the YC level can reflect the stance of the monetary policy than the YC slope does. This can be easily observed from the positive and significant reaction of the real Treasury Bill Rate to the deposit rate shock.

40 Policy Recommendations To increase the predictive power of Egypt's yield curve, the paper recommends that there should be an economy-wide consensus, integrated, complementary fiscal and monetary frameworks. In this context, there are some prerequisites that are recommended to be fulfilled to enhance the yield curve predictive ability, in order to be used as a fiscal and monetary indicator to realize macro-economic stability in Egypt. First, for the fiscal policy and the debt management strategy, three aspects should prevail: Working on the minimization of the fiscal deficit as one of the major prerequisites of adopting the formal inflation targeting framework. Having an active Treasury bonds market that relies on the market mechanism through diversifying debt securities in terms of maturities. Increasing the holding base of T-bonds to eliminate risk stems from the concentration of government debt with certain entities. Second, for the monetary policy, the central bank should implement two types of responses: Responsive Monetary Policy: The CBE should instantly and firmly respond to macro-economic deviations from targets through a formal inflation targeting framework. Protective Monetary Policy: The CBE should increase its reaction magnitude to yield curve shocks; to construct a direct link between the bonds market players anticipation about the future inflation and real activity and the implemented monetary policy to increase its transparency and credibility in the future.

41 References Capital Market Authority, "Monthly Report", Various Issues. Central Bank of Egypt, "Annual Report", Various Issues for Fiscal Years 2004/2005, 2005/2006, and 2006/2007, Cairo, Egypt. Cwik, Paul (May 2004), An Investigation of Inverted Yield Curves and Economic Downturns, Doctor of Philosophy, the Graduate Faculty of Auburn University, Auburn, Alabama. Diebold, X. Fancis, Glenn D. Rudebusch and S. Boragan Aruoba (October 2003), "The Macro-economy and the Yield Curve: A Nonstructural Analysis", Federal Reserve Bank of San Francisco, San Francisco, U.S.A. Estrella, Arturo and Frederic S. Mishkin (June 1996), The Yield Curve as a Predictor of U.S. Recessions, Federal Reserve Bank of New York, Current Issues in Economics and Finance,Volume.2, Number 7, New York, U.S.A. Estrella, Arturo (February 2005), "Why Does the Yield Curve Predict Output and Inflation?", Federal Reserve Bank of New York, New York, U.S.A. Estrella, Arturo and Mary R. Trubin (2006), " The Yield Curve as a Leading Indicator: Some Practical Issues", Federal Reserve Bank of New York, Current Issues in Economics and Finance, Volume 12, Number 5, New York, U.S.A. International Monetary Fund, "International Financial Statistics", CD-ROM, April Kolb, W.Robert and Richard J.Rodriguez (1996), "Financial Management", Second Edition, Blackwell Publishers, pp Kozicki, Sharon (1997), "Predicting Real Growth and Inflation with the Yield Spread", Economic Review, Fourth Quarter, Federal Reserve Bank of Kansas City, Kansas City, U.S.A. Kozicki, Sharon and Gordon Sellon (2005), "Longer-Term Perspectives on the Yield Curve and Monetary Policy", Economic Review, Fourth Quarter 2005, Federal Reserve Bank of Kansas City, Kansas City, U.S.A. Mehl, Arnaud (November 2006), "The Yield Curve as a predictor and Emerging Economies", The European Central Bank, Working Paper Series, Number 691, Frankfurt, Germany. Ministry of Finance (MOF), Decree Number 480 for the year 2002, Establishing the Primary Dealers System, Cairo, Egypt. Ministry of Finance (MOF), Decree Number 723 for the year 2002, (Executive Regulations for Decree Number 480 for 2002), Primary Dealers System, Cairo, Egypt. Morales, Marco (March 2007), "The Yield Curve and Macroeconomic Factors in the Chilean Economy", Center of Economics and Finance, Diego Portales University, Chile. Nelson, R. Charles & Siegel, Andrew F (October 1987), "Parsimonious Modeling of Yield Curves" Journal of Business, University of Chicago Press, vol. 60(4), pages

42 Ramaswamy, Ramana and Torsten Sloek (December 1997), "The Real Effects of Monetary Policy in the Euro Area: What are the Differences?", International Monetary Fund, Research Department, IMF Working Papers Number 97/160. Shelile, Teboho (December 2006), "The Term Structure of Interest Rates and Economic Activity in South Africa", Thesis to obtain the Master Degree in Commerce and Financial Markets, Department of Economics and Economic History, Rhodes University, Grahamstown, South Africa. Stock, H. James & Mark W. Watson (September 2003), "Forecasting Output and Inflation: the Role of Asset Prices" Journal of Economic Literature, American Economic Association, vol. 41(3), pages Wright, H. Jonathan ( ), "Yield Curve and Predicting Recessions", Finance and Economics Discussion Series, Divisions of Research & Statistics and Monetary Affairs, Federal Reserve Board, Washington, D.C. Websites: Answers Encyclopedia: Central Bank of Egypt (CBE): Ministry of Finance (MOF): Wikipedia Encyclopedia:

43 Appendix Table (1): Estimation Results of the Inflation Expectation based on the Yield Curve Information - [3months Lag] Dependent Variable: INFLATION_Y Method: Least Squares Sample (adjusted): 2005M M02, Included observations: 35 after adjustments Newey-West HAC Standard Errors & Covariance Explanatory Variables Coefficient Std. t- Error Statistic Prob. Spread(-3) R_Bill Rate(-3) Spread R_Bill Rate(-3) Annualized Inflation Rate(-3) Constant R-squared Mean dependent var Adjusted R-squared S.D. dependent var S.E. of regression Akaike info criterion Sum squared resid Schwarz criterion Log likelihood Hannan-Quinn criter Prob F-statistic Durbin-Watson stat 0.726

44 Table (2): Estimation Results of the Inflation Expectation based on the Yield Curve Information - [6months Lag] Dependent Variable: INFLATION_Y Method: Least Squares Sample (adjusted): 2005M M02, Included observations: 32 after adjustments Newey-West HAC Standard Errors & Covariance Explanatory Variables Coefficient Std. t- Error Statistic Prob. Spread(-6) R_Bill Rate(-6) Spread R_Bill Rate(-6) Annualized Inflation Rate(-6) Constant R-squared Mean dependent var Adjusted R-squared S.D. dependent var S.E. of regression Akaike info criterion Sum squared resid Schwarz criterion Log likelihood Hannan-Quinn criter Prob F-statistic Durbin-Watson stat 0.596

45 Table (3): Estimation Results of the Inflation Expectation based on the Yield Curve Information - [9months Lag] Dependent Variable: INFLATION_Y Method: Least Squares Sample (adjusted): 2005M M02, Included observations: 29 after adjustments Newey-West HAC Standard Errors & Covariance Explanatory Variables Coefficient Std. t- Error Statistic Prob. Spread(-9) R_Bill Rate(-9) Spread R_Bill Rate(-9) Annualized Inflation Rate(-9) Constant R-squared Mean dependent var Adjusted R-squared S.D. dependent var S.E. of regression Akaike info criterion Sum squared resid Schwarz criterion Log likelihood Hannan-Quinn criter Prob F-statistic Durbin-Watson stat 0.936

46 Table (4): Estimation Results of the Inflation Expectation based on the Yield Curve Information - [12months Lag] Dependent Variable: INFLATION_Y Method: Least Squares Sample (adjusted): 2006M M02, Included observations: 26 after adjustments Newey-West HAC Standard Errors & Covariance Explanatory Variables Coefficient Std. t- Error Statistic Prob. Spread(-12) R_Bill Rate(-12) Spread R_Bill Rate(-12) Annualized Inflation Rate(-12) Constant R-squared Mean dependent var Adjusted R-squared S.D. dependent var S.E. of regression Akaike info criterion Sum squared resid Schwarz criterion Log likelihood Hannan-Quinn criter Prob F-statistic Durbin-Watson stat 1.278

47 Table (5): Choice of VAR Lag Length (Carried out for the Basic Model of Inflation, Deposit, Real T-Bills and Spread) VAR Lag Order Selection Criteria Endogenous Variables: Annualized Inflation Rate, Deposit Rate, Real Treasury Bill Rate, Spread Exogenous Variables: C Sample: 2005: :02 Included observations: 33 Lag Log L LR FPE AIC SC HQ NA * * 6.518* * * * indicates lag order selected by the criterion LR: sequential modified LR test statistic (each test at 5% level) FPE: Final prediction error AIC: Akaike information criterion SC: Schwarz information criterion HQ: Hannan-Quinn information criterion

48 Table (6): Pair-wise Granger Causality Test Sample: 2005: :06 Lags: 1 Null Hypothesis: Obs F-Statistic Probability R_BILL does not Granger Cause Inflation Inflation does not Granger Cause R_BILL Spread does not Granger Cause Inflation Inflation does not Granger Cause Spread Spread does not Granger Cause R_BILL R_BILL does not Granger Cause Spread

49 Table (7): The Variance Decomposition of the Spread and the Inflation Rate Period S.E. Variance Decomposition of the Spread Variance Decomposition of the Inflation Rate Inflation Deposit R_Bill Spread Inflation Deposit R_Bill Spread

50 Table (8): Estimation Results for Yields-Macro Transition Equation based on the Lag Specification (1 1) Vector Auto-Regression Estimates Sample (adjusted): 2005: :02 Included Observations: 37 after adjusting endpoints Standard Errors in ( ) & T-Statistics in [ ] Inflation Deposit R_Bill Spread Annualized Inflation Rate(-1) (0.24) (0.03) (0.28) (0.21) Deposit Rate(-1) R_Bill Rate(-1) Spread(-1) [ 3.596] [ 1.297] [ 0.324] [-2.462] (0.39) (0.06) (0.46) (0.35) [-2.798] [ ] [ 2.264] [ 1.472] (0.27) (0.04) (0.31) (0.24) [-0.099] [ 0.934] [ 3.238] [-3.368] (0.22) (0.03) (0.26) (0.19) [-0.937] [ 0.676] [ 0.568] [-0.792] (2.32) (0.33) (2.69) (2.04) Constant [ 3.610] [ 0.152] [-2.865] [ 1.358] R-squared Adj. R-squared Sum sq. residuals S.E. equation F-statistic Log likelihood Akaike AIC Schwarz SC Mean dependent S.D. dependent Determinant Residual Covariance Log Likelihood (d.f. adjusted) Akaike Information Criteria Schwarz Criteria 7.423

51 Figure (6): Impulse Response Functions for the Yield-Macro Equation based on the Lag specification (1 1) Response to Generalized One S.D. Innovations ± 2 S.E. Response of INFLATION_Y to INFLATION_Y Response of INFLATION_Y to DEP Response of INFLATION_Y to R_BILL Response of INFLATION_Y to S Response of DEP to INFLATION_Y Response of DEP to DEP Response of DEP to R_BILL Response of DEP to S Response of R_BILL to INFLATION_Y Response of R_BILL to DEP Response of R_BILL to R_BILL Response of R_BILL to S Response of S to INFLATION_Y Response of S to DEP Response of S to R_BILL Response of S to S

52 Figure (7): Residuals obtained from the Unrestricted VAR (1) Model INFLATION_Y Residuals DEP Residuals :07 06:01 06:07 07:01 07:07 08: :01 05:07 06:01 06:07 07:01 07:07 08:01 3 R_BILL Residuals 4 S Residuals :07 06:01 06:07 07:01 07:07 08: :01 05:07 06:01 06:07 07:01 07:07 08:01

53 Table (9): Government Bonds held by the Banking Sector (except the Central Bank of Egypt) as a Percentage of the Total Listed Government Bonds Listed Government Bonds in the Stock Market Government Bonds in Banking Sector Share of the Listed Gov. Bonds held by the Banking Sector (%) Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Year Average 89% Jan Feb Mar Apr May Jun Jul Aug Sep

54 Listed Government Bonds in the Stock Market Government Bonds in Banking Sector Share of the Listed Gov. Bonds held by the Banking Sector (%) Oct Nov Dec Year Average 75.5% Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Year Average 64% Jan Feb Year Average 62% (Continued)

55

56 - 47-1/1- Gross Domestic Product (GDP) 1: Macroeconomic Performance Egypt's economy continued to recover from the spillovers of the global financial crisis, with the acceleration of the annual real GDP growth at factor cost to 4.8 percent during the first half of FY 2009/2010, up from 4.4 percent in the second half of FY 2008/2009. That rate was faster than the average real GDP growth of the emerging economies as a whole, though it remained slightly slower than its pace in the first half of FY 2008/2009 (4.9 percent) (%) Egypt's Real GDP vs. Emerging Economies Q3: 07/08 Q4: 07/08 Q1: 08/09 Q2: 08/09 Q3: 08/09 Q4: 08/09 Q1: 09/10 Q2: 09/10 Egypt Emerging Economies Source: Ministry of Economic Development. As concerns emerging economies, "JP Morgan World Financial Markets Report, Dec. 2009". On the supply side, the gradual improvement in the economic performance was a main result of the higher contributions of domestic demand-driven sectors which accounted for 80 percent of the above-said real GDP growth rate, particularly manufacturing; construction and building; wholesale and retail trade; agriculture; communications and IT; and the general government.

57 GDP Growth by Domestic and External Demand-Driven Sectors in the First Half of FY 2009/2010 Sector Domestic Demand-Driven Sectors Growth Rate Share in Real GDP Growth (4.8 %) Agriculture, Irrigation & Fishing Manufacturing Electricity Construction & Building Transportation & Storage Communications and IT Wholesale & Retail Trade Financial Intermediaries & Supporting Services General Government Total 3.9 External Demand-Driven Sectors Sector Growth Rate Share in Real GDP Growth Extractions Suez Canal Restaurants and Hotels Total 0.4 Source: According to the Ministry of Economic Development data for the 1 st half of 2009/2010. External demand-driven sectors contributed only 0.4 percentage point of GDP growth, mainly generated by extractions and restaurants and hotels, given that the share of extractions was less than its level in the first half of FY 2008/2009. On the other hand, the contribution of restaurants and hotels went up, because of the increase in tourism revenues in the reporting period above the period of comparison. Moreover, the share of the Suez Canal sank to a negative 0.5 percentage point, driven by the fall in its receipts during the period influenced by the slowdown in international trade. The chart below shows a continued negative gap of 0.4 percentage point between real GDP growth rate (seasonally adjusted) and real long-term growth rate which can be deduced from the curve of trend growth. This gap was largely attributed to the persistence of the negative gap of the Suez Canal growth (due to the decline in its tolls), though it narrowed in the second quarter of FY 2009/2010. Another affecting factor was that the sectors of extractions, financial intermediaries, and manufacturing lagged behind their normal growth levels because of their economic recession. (%)

58 Development of the Real GDP Growth Rate Divided into Potential GDP and GDP Gap (%) Q1: 02/03 Q3: 02/03 Q1: 03/04 Q3: 03/04 Q1: 04/05 Q3: 04/05 Q1: 05/06 Q3: 05/06 Q1: 06/07 Q3: 06/07 Q1: 07/08 Q3: 07/08 Q1: 08/09 Q3: 08/09 Q1: 09/10 GDP Gap_Business Cycle GDP Growth Rate_Seasonally Adjusted Potential GDP_Trend After estimating the trend growth of real GDP and analyzing the results, the main economic performers in the second quarter of FY 2009/2010 were divided into three main categories of sectors: sectors under recession; sectors in the process of recovering from recession; and sectors undergoing economic expansion and contributing to the reduction of the economic gap. Economic expansion was led by the sector of restaurants and hotels which showed a conspicuously strong performance in the second quarter of FY 2009/2010, after having suffered from a sharp recession in the first quarter of the same year. Communications and IT came second, followed by construction and building, and the general government sector which is mainly concerned with rendering government services. The real estate sector came last, being by far reliant on administrative procedures, facilities and services.

59 Real & Potential Growth Rates of Main Categories of Economic Sectors (Seasonally Adjusted) in Q 2 of FY 2009/2010 First: Sectors under Economic Recession Sector Real Growth (%) (1) Potential Growth (%) (2) Difference (Percentage Point) (1) (2) Transportation & Storage Manufacturing Extractions Financial Intermediaries Second: Recovering Sectors Suez Canal Third: Sectors Achieving Economic Expansion Restaurants & Hotels Communications & IT Construction & Building Social Insurance General Government Real Estate Source: According to the data of the Ministry of Economic Development. In spite of its stronger pace of growth and increased share in the real GDP growth, the manufacturing sector s business cycle is still in economic recession and the sector has not yet restored its normal growth rates. As regards the public and private sectors contributions to GDP growth (4.8 percent) in the reporting period, the public sector generated 1.1 percentage points (against 1.9 points in the first half of FY 2008/2009), compared with 3.7 points contributed by the private sector (against 3.0 points). This indicated that the private sector continued to play a major role in development. Contribution of the Private Sector to GDP Growth by Economic Sector (Percentage Point) Restaurants & Hotels Financial Intermediaries Wholesale & Retail Trade Construction & Building Manufacturing Extractions st half. 2008/2009 (3.0%) 1st half 2009/2010 (3.7%)

60 The main drivers of growth on the level of the public sector were the general government, social insurance, construction and building and financial intermediaries. In contrast, the Suez Canal's contribution was negative. Turning to the private sector, the main contributors to GDP growth were the sectors of manufacturing, construction and building, wholesale and retail trade, and restaurants and hotels. In contrast, the contribution of extractions decreased. Contribution of the Public Sector to GDP Growth by Economic Sector (Percentage Point) General Government Social Solidarity Financial Intermediaries Wholesale & Retail Trade Suez Canal Construction & Building Manufacturing Extractions st half 2008/2009 (1.9%) 1st half 2009/2010 (1.1%) On the demand side, the gradual improvement in the economic performance was essentially ascribed to the private consumption s higher contribution to GDP growth (3.6 percentage points, against 2.9 points in the period of comparison). Likewise, the contribution of net external demand to economic growth went up (0.9 percentage point against 0.8 point), driven by the fall in the imports of goods and services to a higher degree than in the exports of goods and services. Those factors compensated for the poor contribution of total investment (including change in stock) to GDP growth, which fell to a negative 0.2 percentage point, down from a positive 0.7 point in the period of comparison.

61 Share of Demand Components in Real GDP Growth Rate during July/December (Percentage Point) Growth Rates in 1 st Half Share in GDP Growth 2008/ / / /10 Real GDP Growth Rate (1+2) Domestic Demand (A+B) A- Final Consumption Private Public B- Capital Formation (Including Change in the Stock) Net External Demand (A-B) A- Exports of Goods & Services B- Imports of Goods & Services Shares of Consumption, Investment and Net Exports in Real GDP Growth Rate during July / December (Percentage Point) / /2010 Consumption Investment Net Exports Total investments (at constant prices) declined by 3.0 percent in the first half of FY 2009/2010 compared with the corresponding period of the preceding FY. That was ascribed to the negative contribution of the private sector (4.8 percentage points) to the real growth rate of total investment. In contrast, the public sector recorded a positive contribution of 1.8 percentage points (against 2.7 points a year earlier). This emphasizes the conclusion drawn earlier in the context of the demand side analysis, with respect to the drop in the contribution of public investment to the real GDP growth rate.

62 Contribution of the Public Sector to the Real Investment by Sector (Percentage Point) Others Sanitation Health Services Educational Services Real Estate Restaurants & Hotels Communications Transportation & Storage Water Electricity Oil Refining Natural Gas st half 2008/2009 (2.7%) 1st half 2009/2010 (1.8%) The decline in the real growth rate of total investment came largely from the negative contribution of the private sector to the manufacturing sector (excluding oil refining), natural gas, and health and education services. In contrast, the contribution of the private sector to real estate, communications and IT, and crude oil increased. On the other hand, the public sector s contribution to the real growth rate of total investment was focused on the sectors of transportation and storage, real estate, and health services. In contrast, the contribution of the public sector to water, electricity and natural gas declined. Contribution of the Private Sector to the Real Investment by Sector (Percentage Point) Others Health Services Educational Services Real Estate Restaurants & Hotels Wholesale & Retail Trade Communications Transportation & Storage Other Manufacturing Oil Refining Natural Gas Crude Oil Agriculture, Irrig.& Reclam st half 2008/2009 (-1.7%) 1st half 2009/2010 (-4.8%)

63 The breakdown of total implemented investments (at constant prices 2006/2007=100) by economic sector during the period indicated that crude oil, natural gas, and other extractions accounted for 22 percent, agriculture and manufacturing 16 percent, electricity and water 10 percent, construction and building 2 percent, and services 50 percent (productive services 27 percent, and other social services 23 percent). The reporting period witnessed the foundation of 3439 new companies with capitals of LE 8.3 billion, of which 45 percent belonged to the services sector. According to the geographical distribution, around 63 percent of those companies are located in Cairo and Giza. 1/2-Employment and Unemployment According to data of the Labor Force Survey in the second quarter of FY 2009/2010, unemployment accelerated to 9.4 percent in the reporting period, from 8.8 percent in the second quarter of FY 2008/2009. That was an outcome of the increase in the number of the unemployed by 2.4 million persons during the second quarter of FY 2009/2010, rising 7.5 percent over the same period a year earlier. The heightening of unemployment was more pronounced in female unemployment at a rate of 22.9 percent, since the rise in male unemployment was as slight as 5.3 percent during the period. Development of Annual Growth Rates in the Labor Market (%) FY 2008/2009 FY 2009/2010 Q 1 Q 2 Q 3 Q 4 Q 1 Q 2 Labor Force Employment Unemployment Source: CAPMAS, Labor Force Survey. The distribution of employment by business sector revealed that the private and investment sector came in the forefront of the labor-absorbing sectors, attracting 72.3 percent of the number of employees. This reflected the key role played by the private sector in accelerating development and reducing unemployment.

64 Distribution of Employment by Economic Sector during October/December (%) Q2 2008/2009 Q2 2009/2010 Public Sector Private Sector Others Source: CAPMAS, Labor Force Survey. The distribution of employment by economic sector in the second quarter of FY 2009/2010 showed that the commodity sectors absorbed 53.4 percent, the productive services 20.4 percent, and the social services 26.2 percent. Sectoral Distribution of Employees during Oct./Dec. 2009/2010 Social Services Sectors 26.2% Sectoral Distribution of Employees during Oct./Dec. 2008/2009 Social Services Sectors; 24.8% Productive Services Sectors 20.4% Commodity Sectors 53.4% Productive Services Sectors; 20.0% Commodity Sectors; 55.2% Source: Ibid.

65 - 56-1/3- Inflation A - Consumer Price Index (CPI) In the period July/December of FY 2009/2010, the annual headline CPI inflation (urban) remarkably rose to about 6.7 percent, from 3.5 percent in the corresponding period of the preceding FY. The increase was particularly pronounced in the prices of food and non-alcoholic beverages that added 5.3 percentage points to headline inflation (compared with 0.8 point in the corresponding period). % Annual CPI and Price Index of Food and Non-Alcoholic Beverages (Urban) Jun.2008 Aug. Source: CAPMAS. Oct. All Items Dec.2008 Feb. Apr. Jun.2009 Aug. Food and Non -Alcoholic Beverages Oct. Dec.2009 The share of food and non-alcoholic beverages in inflation rose to 10.9 percent during the period under review (against 1.8 percent in the corresponding period). The rise came on the back of the increase in international food prices (16.6 percent) during the year ending December 2009, especially sugar and palm oil. However, the world price hikes were not fully reflected on domestic food prices, because of price rigidities in the domestic market.

66 Source : IMF. % The Change in International Prices of Basic Foodstuffs Wheat Maize Palm Oil Meat Sugar Coffee Tea Rice Food Prices Dec.2008/Dec Nov.2009/Dec The rise in the share of food and non-alcoholic beverages was ascribed to the larger contribution of vegetables (3.2 percentage points, against -0.1 point), fruit (0.8 point against 0.1 point), meat (0.8 point against 0.4 point), oils and fats (0.2 point against -0.1 point) and sugar and confectionary (0.3 point against nil), due to their higher inflation rates during the period under review especially of vegetables (42.1 percent) and sugar (17.2 percent). However, declines were observed in the groups of fish (-0.2 point, against 0.2 point) and milk, cheese and eggs (0.2 point against 0.7 point) with the drop in their inflation rates, thus subduing the rise in headline inflation. Contribution of Main Items of Food to Headline Inflation (Annually) during July/Dec. Percentage Point Bread & Cereals Meat Fish and Seafood Milk,Cheese and Eggs Oils & Fats Fruit Vegetables Sugar 2008/ /2010

67 The following table illustrates the shares of CPI groups (urban) in headline inflation during the period under review and the period of comparison: Main CPI Groups Weights Inflation Rate in the First Half of FY (%) Share in Headline Inflation in the First Half of FY (Percentage Point) 2008/ / / /10 General Index Food & non-alcoholic beverages Alcoholic beverages, tobacco and narcotics Clothing and footwear Housing, water, electricity, gas & fuel Furnishings, household equipment and routine maintenance Health care Transportation Communications Culture & recreation Education Restaurants & hotels Miscellaneous goods & services The rise in miscellaneous goods and services (0.6 percentage point against nil) also contributed to the increase in headline inflation during the period under review, due to the rise in health insurance fees in October Add to this the increase in the share of the education group (0.5 point against 0.2 point) with the acceleration of the inflation rate of tertiary education (15.1 percent against 7.2 percent) and school tuition (6.6 percent against 3.4 percent). However, the effect of such seasonal increases is confined to the beginning of the school year in Egypt.

68 According to the CPI (urban), the monthly headline inflation notably retreated during the last two months of the period under review, registering a record low of -1.3 percent in December 2009 ever since January That decrease was largely traced to the decline in the share of food and nonalcoholic beverages to -0.2 percentage point in November 2009 and -1.3 percentage points in December % Monthly CPI (Urban) Jun.2008 Jul. Aug. Sep. Oct. Nov. Dec.2008 Jan. Feb. Mar. Apr May Jun.2009 Jul. Aug. Sep. Oct. Nov. Dec.2009 The significant decreases in the monthly inflation rates during November and December 2009, were mainly driven by the fall in the shares of the respective groups of vegetables, fruit and meat to negative levels during these two months. Percentage Point Contribution of Main Items of Food to Headline Inflation (Monthly) Meat Fish and Seafood Fruit Vegetables Nov Dec. 2009

69 B - Producer Price Index (PPI) Showing the same upward trend as the CPI, the annual headline PPI inflation accelerated in the period under review, registering 4.7 percent, up from percent in the corresponding period a year earlier % Jun.2008 Jul. Annual PPI ( 2004/2005 = 100 ) Aug. Sep. Oct. Nov. Dec.2008 Jan. Feb. Mar. Apr May Jun.2009 Jul. Aug. Sep. Oct. Nov. Dec.2009 The rise in PPI inflation was largely ascribed to the higher share of mining and quarrying (1.5 percentage points against negative 18.6 points), in view of the noticeable rise in crude oil (2.3 percentage points against negative 28.9 points). That was the result of the high increase of 10.3 percent in the inflation rate of crude oil in the period under review (against negative 73.6 percent in the corresponding period) % Jun.2008 Jul. Aug. Annual PPI of Mining and Quarrying Section Sep. Oct. Nov. Dec.2008 Jan. Feb. Mar. Apr May Jun.2009 Jul. Aug. Sep. Oct. Nov. Dec.2009

70 Another factor that explains the acceleration in headline PPI inflation in July/December 2009/2010, is the pickup in the inflation rate of agriculture and fishing (1.8 percentage points against negative 0.2 point). That was mainly attributed to the increase in the share of vegetables (2.2 percentage points against 0.4 point) and poultry and eggs (0.2 percentage point against a negative 0.1 point). This increase was accompanied by the rise in the inflation rates of these groups to 40.9 and 5.8 percent, respectively, (against 8.9 and negative 3.6 percent). Moreover, the manufacturing group contributed to the rise of headline inflation during the period under review (1.1 percentage points against a negative 0.5 percent), with the rise in the inflation rate of this group (2.9 percent against negative 1.5 percent). That was mainly due to the increase in the headline inflation of iron and steel (0.1 percentage point against negative 1.5 percentage points) and the inflation rate thereof by 1.4 percent against negative 25.4 percent. The following table shows inflation rates and the shares of PPI groups in headline inflation during the two periods of review and comparison:

71 Main PPI Groups Share of PPI Groups in Headline Inflation (Jan = 100) Inflation Rate (%) July/Dec. Share in Headline Inflation (Percentage Point) July/Dec. 2008/09 Share in Headline Inflation (Percentage Point) July/Dec. 2009/ / /10 General Index Agriculture, Forestry and Fishing, of which: Cereals and leguminous crops Rice Vegetables Fruit Poultry and eggs Fish Mining and Quarrying, of which: Crude oil Stone, sand and clay Manufacturing, of which: Processed food products, of which: Oils and fats Dairy products Fertilizers Wood & products Cement Iron and steel Electricity and Gas, of which: Electric power generation, transmission and distribution Water Supply Activities Transportation and Storage, of which: Land transport Accomodation and Food Services, of which: Meal serving services in limited service facilities Information and Communications

72 - 63-1/4- Tourism In the light of the data of the Ministry of Tourism, the indicators of arrival tourism in Egypt manifested a slight rebound in July/Dec. of FY 2009/2010, compared with the previous corresponding period. The total number of arrivals rose by 3.7 percent to some 6.8 million against 6.6 million, and so did the number of tourist nights (up by 4.7 percent to 70.7 million against 67.5 million). Though the tourism sector was the hardest hit by the crisis, it began to show mild recovery and revert to its pre-crisis levels in the reporting period, after the negative repercussions of the global financial crisis had almost faded away. million nights Number of Tourists & Tourist Nights (July/Dec.) million tourists / /2010 Tourist Nights of Departures Number of Tourists Among the signs of recovery was the pickup in tourism revenues influenced by the rise in the number of tourist nights, even though the average spending per tourist a night remained unchanged at US$ 85 a night. Hence, tourism revenues went up by 4.7 percent to US$ 6.0 billion (against US$ 5.7 billion) or 2.8 percent of GDP at current prices during FY 2009/2010 (against 3.2 percent), making up 25.3 percent of total BOP visible and invisible receipts during the same year (against 21.1 percent). Indicators of Tourism Revenuse during July/Dec. % / /2010 Tourism Revenues / Visible & Invisible Receipts Tourism Revenues / GDP

73 Investments in the tourism sector (restaurants and hotels) amounted to LE 2.4 billion during July/Dec. of FY 2009/2010, constituting 2.4 percent of total investments. The private sector undertook the major part of these investments, with a share of about 93.1 percent. Statistical Indicators July/Dec. 2008/ /2010 Change + (-) % Number of arrivals (000s) Number of nights for departures (000s) Estimated average spending per tourist a night (US$ ) Tourism revenues (US$ mn) Average tourist stay (night) GDP at current prices (LE bn) GDP at current prices(us$ bn) Source: The CBE and the Ministries of Tourism & Economic Development. Number of Tourists Arrivals from all tourist exporting markets during the period under review totaled 6.8 million, with a modest increase of thousand or 3.7 percent in comparison with the previous FY. The growth in the number of tourists, though weak, denoted an incipient tourism recovery that followed the sharp downturn sparked by the global financial crisis. Number of Tourist Arrivals (Thousand) July/Dec. 2008/ /2010 Number Relative Weight Number Relative Weight Total Europe Middle East Africa The Americas Asia and the Pacific Others Source: Ministry of Tourism. Change + (-) %

74 Relative w eight of Tourist Arrivals July/Dec. 2008/2009 Relative w eight of Tourist Arrivals July/Dec. 2009/2010 African countries 3.2% The Americas 3.8% Asian & Pacific countries 4.6% Others 0.4% African countries 3.5% The Americas 3.8% Others 0.3% Asian & Pacific countries 4.5% Middle East countries 14.7% European countries 73.3% Middle East countries 12.8% European countries 75.1% With a relative weight of 75.1 percent of total tourist flows, the European group remained in the lead, registering a rise of thousand tourists or 6.1 percent. Russia accounted for the bulk of 23.2 percent (1.2 million, up by 44.4 percent); followed by the UK (714 thousand, up by 21.3 percent); Germany (623 thousand, up by 2.6 percent); and Italy (542 thousand at a rate of negative 0.6 percent). Despite the 9.8 percent decline in tourist flows from the Middle East group, it occupied the second position, with a share of 12.8 percent of the total number of tourists. Arrivals came mostly from Libya (220 thousand), Saudi Arabia (205 thousand), and Jordan (92 thousand). The number of tourists from these countries decreased by 22.1 percent, 19.5 percent and 2.2 percent, respectively. However, there was a marked increase in the number of tourists from Palestine (53.5 percent), and Kuwait (9.1 percent). Ranking third, the Asian and Pacific group accounted for 4.5 percent of the total number of tourists, as the number of arrivals therefrom slightly rose by 1.6 percent. The increase was mainly in arrivals from China (40.0 percent), followed by India (5.4 percent). However, there was a decline in the number of tourists from Australia, Japan, and Philippines by 6.8 percent, 3.2 percent, and 3.1 percent, respectively. Arrivals from the Americas group ranked fourth with a relative weight of 3.8 percent of the total number of arrivals, up by 5.2 percent. Arrivals came mostly from the USA (165 thousand or 63.0 percent), and Canada (19.1 percent).

75 In spite of occupying the last position with a relative weight of 3.5 percent of the total number of arrivals, the African group recorded the highest rate of growth relative to the other groups (12.9 percent). The Sudan came first with a share of 32.2 percent, followed by Nigeria with 13.6 percent, Morocco with 11.9 percent, Algeria with 9.7 percent and South Africa with 8.1 percent. Tourist Nights In line with the increase in the number of tourists, tourist nights exhibited an upward trend during July/Dec. 2009/2010. The number of nights spent by all departure groups totaled some 70.7 million during the relevant period, up by 3.2 million or 4.7 percent above the period of comparison. Number of Tourist Nights by Departure Group (Thousand) 2008/ /2010 Number Relative Weight Number Relative Weight Change + (-) % Total Europe Middle East Africa The Americas Asia and the Pacific Others Source: Ibid. The European group ranked first with a relative weight of 69.2 percent of the total and a rise of about 3.7 million nights or 8.3 percent above the period of comparison. This was largely attributed to the increase in the number of nights spent by departures of Russia by 88.6 percent, the UK by 23.0 percent, and the Netherlands by 5.0 percent. In the meantime, there was a decrease in the number of tourist nights of other countries; mainly France by 12.7 percent, and Poland by 11.0 percent.

76 % European countries Tourist Nights by Departure Groups (July/Dec.) Middle East countries African countries The Americas Asian & Pacific countries Others 2008/ /2010 The Middle East group ranked second with a relative weight of 18.1 percent; however, tourist nights by departures of this group decreased by 0.6 million or 4.7 percent. In the forefront of these countries came Saudi Arabia (3.3 million nights), down by 15.4, followed by Libya (3.1 million nights with a negative 8.6 percent). However, the number of tourist nights by departures of Palestine markedly rose by 25.8 percent, followed by Kuwait by 6.2 percent, and Jordan by 4.6 percent. The Americas group recorded a limited increase of 3.0 percent in the number of tourist nights by departures (3.3 million), mostly spent by departures of the USA (2.2 million up by 1.8 percent), and Canada (0.6 million up by 11.4 percent). As for the Asian and Pacific markets, tourist nights scantly scaled up by 0.9 percent to 2.7 million nights. Departures of China accounted for 44.3 percent of this increase, followed by Japan 14.7 percent. However, the number of tourist nights of most countries of this group retreated, especially Indonesia (20.0 percent) and India (2.4 percent). The number of tourist nights by departures of the African markets rose by 8.2 percent to 2.8 million nights, mostly contributed by the Sudan (1.5 million nights). Moreover, Nigeria recorded the highest growth rate in the number of tourist nights (68.8 percent). However, the number of tourist nights by departures for Tunisia declined by 0.8 percent.