## _wp07285

## Source details

**Canonical URL:** [_wp07285](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2007/_wp07285.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2007/_wp07285.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2007/_wp07285.pdf.json)

---

### I. INTRODUCTION — context and reform milestones
- Since completion of ERSAP in 1996, major structural changes:
  - abolition of the de jure exchange rate peg in 2000;
  - introduction of the domestic currency overnight interbank market in 2001;
  - launch of the foreign exchange interbank market in 2004;
  - introduction of the corridor for overnight facilities as main policy instrument in June 2005.
- In June 2005 the Central Bank of Egypt (CBE) announced intention to adopt inflation targeting (IT) over the medium term and declared price stability the CBE’s overriding policy objective.
- Transition to IT implies exchange rate would no longer be the primary instrument; other transmission channels must function well.
- Paper combines descriptive review of Egyptian monetary policymaking with a baseline VAR model and extensions to investigate monetary transmission mechanism (MTM) channels, subject to data constraints.

### Main empirical results
- Key empirical findings:
  - (i) the interest rate channel of monetary policy is still weak in Egypt;
  - (ii) the exchange rate channel is very pronounced, especially given infrequent but significant exchange rate movements over the last decade;
  - (iii) other transmission channels play a very limited role in transmitting monetary shocks to the economy; and
  - (iv) the introduction of the corridor for the overnight rate may have started to contribute to a firmer interest rate channel.

### Policy recommendations
- Primary policy recommendations:
  - (i) enhance competition in the banking system to reduce the spread between lending and borrowing rates and strengthen interest rate channel;
  - (ii) improve the CBE’s communication strategy to enhance the expectations channel;
  - (iii) add financial instruments to lengthen and deepen the yield curve (e.g., mortgage financing), possibly including instruments that reflect inflation expectations (inflation-indexed bonds);
  - (iv) allow greater flexibility in the exchange rate, a precondition for successful inflation targeting; and
  - (v) persevere with other governmental reforms, including resolving the remaining non-performing loan (NPL) issue and fiscal consolidation.

### Data, measurement, and recommended statistical improvements
- Statistical observations and needs:
  - improvements in price statistics would enhance empirical evaluations;
  - availability of an indicator of true economic activity at a monthly frequency and for a sufficient amount of time would strengthen analyses (current proxies like electricity are limited);
  - expanding available indicators is recommended from Egypt’s statistical agency.

### Monetary policy framework and structural context (1996–June 2005)
- Multiple, sometimes conflicting, objectives: high economic growth, low inflation, stable exchange rate.
- Increasing capital mobility (primarily inflows) reduced monetary policy independence and clouded measurement of stance.
- Operational/intermediate targets and instruments:
  - operational target: excess reserves of banks;
  - intermediate target: growth in M2;
  - instruments used: reserve requirements, government securities, repo/reverse repo, CBE discount rate.
- Banking-sector distortions:
  - public banks accounted for about two-thirds of sector assets in mid-1990s;
  - by 2006 privatization reduced public bank market share; privatization of the third-largest public bank was scheduled, expected to bring public bank market share significantly below 50 percent.
- Pre-2005 quantitative measures played a larger steering role than price instruments due to banking-sector rigidities and NPLs.

### Institutional and operational reforms toward inflation-targeting compatibility (Box 1)
- Institutional reforms:
  - Coordinating Council on Monetary Policy established January 2005, chaired by Prime Minister;
  - Monetary Policy Committee (MPC) at CBE convenes every six weeks; MPC of nine members;
  - Monetary Policy Unit established within CBE for analysis and communication;
  - MPC decisions communicated via monetary policy statement on CBE external web-site after each meeting.
- Operational reforms and instruments:
  - June 2, 2005: CBE introduced interest rate corridor defined by overnight lending and deposit facilities (ceiling/floor), steering overnight rate within corridor as operational target.
  - Starting August 2005: central bank notes added as primary liquidity management instrument; deposit auctions continued.
  - Alternative models to forecast inflation developed.

### Monetary transmission: empirical methodology and data
- Baseline VAR: monthly observations January 1996–June 2005; maximum observations 114 (some series start later).
- Baseline VAR variables:
  - economic activity measure (high-frequency monthly GDP measure by MMZ),
  - price level (WPI used often),
  - monetary policy stance (MMZ indicator, Litterman (1983) measure),
  - nominal effective exchange rate (NEER).
- Exogenous variables: (log) oil price, U.S. federal funds rate.
- Seasonality: X12 filter applied (except monetary stance/interest rates); three lags sufficient in most specifications.
- Stationarity: augmented Dickey-Fuller tests indicate variables are I(1) in levels but stationary in first differences; analysis conducted in levels for short-run focus.
- Identification: Choleski (lower diagonal) decomposition; ordering: output, prices, monetary policy stance, exchange rate in baseline.

### Granger causality and preliminary evidence (selected results)
- Baseline Granger causality (1995M1–2005M6), selected p-values and significance:
  - Prices excluded: Prices — Chi-square 9.333 — p-value 0.025 **;
  - Prices excluded: NEER — Chi-square 8.593 — p-value 0.035 **;
  - Prices jointly excluded — Chi-square 25.709 — p-value 0.002 ***;
  - Output excluded: Output — Chi-square 16.453 — p-value 0.001 ***;
  - Output excluded: NEER — Chi-square 14.123 — p-value 0.003 ***;
  - Output jointly excluded — Chi-square 32.039 — p-value 0.000 ***.
- Pairwise Granger tests (1995–2005; lag length three), selected results:
  - Three-month deposit rate Granger-causes lending rate: Observations 128 — F-Statistic 3.48 — Probability 0.02 **;
  - Six-month t-bill rate Granger-causes 3-month deposit rate: Observations 123 — F-Statistic 3.62 — Probability 0.02 **;
  - MP measure Granger-causes 3-month deposit rate: Observations 117 — F-Statistic 2.38 — Probability 0.07 *.
- Interpretation: interest rates are rather unrelated to each other; monetary policy stance appears to Granger-cause the 3-month deposit rate but not vice versa.

### Baseline VAR key results: impulse responses and variance decomposition
- Impulse response highlights:
  - Prices: strongly significant response to exchange rate shocks; response to MMZ monetary stance measure not quite significant in baseline but closer to significance if sample extended to end-2005.
  - Output: response to monetary policy stance shock shows expected sign (output starts rising after about a year in response to an easing) though not statistically significant.
  - Output: significant positive response to a nominal appreciation shock (noted as counterintuitive).
  - WPI: strong deflationary impact of a tightening.
- Variance decomposition (Baseline VAR, 1995–2005; in percent of total variance) — selected tables (Months and contributions):
  - Output variance decomposition (Output / WPI / MP measure / NEER):
    - 10 months: 100.0 / 0.0 / 0.0 / 0.0
    - 20 months: 96.0 / 1.7 / 0.0 / 2.3
    - 30 months: 83.8 / 5.7 / 0.2 / 10.2
    - 60 months: 78.1 / 4.0 / 0.3 / 17.6
    - 120 months: 68.4 / 2.9 / 0.4 / 28.4
    - 240 months: 66.0 / 3.7 / 3.4 / 26.8
    - 600 months: 71.2 / 4.5 / 7.2 / 17.2
  - NEER variance decomposition (Output / WPI / MP measure / NEER):
    - 10 months: 0.2 / 99.9 / 0.0 / 0.0
    - 20 months: 3.1 / 94.8 / 0.4 / 1.7
    - 30 months: 2.6 / 92.3 / 0.5 / 4.6
    - 60 months: 2.0 / 68.9 / 1.2 / 27.8
    - 120 months: 3.9 / 35.2 / 1.8 / 59.2
    - 240 months: 7.6 / 19.9 / 1.3 / 71.2
    - 600 months: 11.3 / 16.5 / 3.9 / 68.3
  - Monetary policy measure variance decomposition (Output / WPI / MP measure / NEER):
    - 10 months: 0.0 / 0.2 / 99.8 / 0.0
    - 20 months: 0.0 / 2.0 / 98.0 / 0.0
    - 30 months: 0.3 / 1.6 / 98.2 / 0.0
    - 60 months: 0.6 / 1.3 / 98.0 / 0.1
    - 120 months: 1.8 / 1.7 / 96.3 / 0.2
    - 241 months: 4.8 / 1.9 / 93.1 / 0.3
    - 601 months: 13.3 / 1.8 / 84.3 / 0.6
  - WPI variance decomposition (Output / WPI / MP measure / NEER):
    - 10 months: 0.0 / 2.3 / 0.0 / 97.7
    - 20 months: 0.0 / 1.4 / 0.5 / 98.1
    - 30 months: 0.1 / 2.0 / 1.6 / 96.4
    - 60 months: 0.1 / 2.0 / 2.0 / 95.9
    - 120 months: 0.4 / 1.3 / 1.3 / 97.0
    - 240 months: 0.6 / 1.0 / 3.5 / 94.9
    - 600 months: 1.4 / 0.9 / 8.8 / 88.9
- Interpretation: most variables are highly idiosyncratic; WPI variation is mainly explained by NEER after about one year (strong exchange rate pass-through to WPI).

### Robustness checks
- Substitute reserve money for MMZ monetary policy stance:
  - Key results unchanged: output responds positively (but insignificantly) to expansionary shocks; WPI no longer shows initial (insignificant) increase.
  - Reserve money limited as stance indicator: NEER appreciates in response to an expansionary monetary shock.
- Inversion of exchange rate and monetary stance ordering: baseline results robust.

### Exploration of individual transmission channels

- Interest rate channel
  - Historical assessment (1996–2005):
    - Overnight interbank market introduced in 2001; overnight interest rate very volatile initially.
    - Alternative indicators: 3-month treasury bill rate and CBE discount rate; movements appear secular with no evident cyclical pattern.
    - Coefficient of variation in nominal policy interest rates (treasury bill and discount rate) ranged between 0.1 and 0.3; nominal retail rates variability between 0.04 and 0.1 during 1995–2005.
    - Granger causality tests suggest a weak interest rate transmission channel during the period.
  - Empirical pairwise Granger tests (1995–2005; selected):
    - Six-month t-bill rate does not Granger-cause lending rate: Observations 123 — F-Statistic 0.230 — Probability 0.87
    - Three-month deposit rate Granger-causes lending rate: Observations 128 — F-Statistic 3.48 — Probability 0.02 **
    - MP measure Granger-causes 3-month deposit rate: Observations 117 — F-Statistic 2.38 — Probability 0.07 *
  - Re-estimated VAR with lending and deposit rates:
    - Significant reduction of 3-month deposit rate after expansionary shock to monetary policy measure.
    - Almost-significant response of lending rate to policy stance shock.
    - Absence of effective policy interest rate before overnight corridor introduction hampered short-end transmission.
    - Limited evidence of interest rate mechanism beyond the very short end.

- Exchange rate channel
  - Exchange rate channel central where financial markets underdeveloped; changes in monetary stance reflected in exchange rate and transmitted to domestic prices via pass-through.
  - Four phases versus U.S. dollar (selected facts):
    - Phase I: January 1997–December 2000 — pound de jure/de facto pegged; foreign reserves fell from US$16.8 billion (early 1997) to US$12.9 billion (late 2000); step devaluations from April 1999 amounting to 8.4 percent; WPI/CPI annual averages: 2.1 percent and 3.2 percent respectively.
    - Phase II: January 2001–December 2002 — crawling bands ±1 percent widened to ±3 percent in August 2001; parallel market at 15 percent premium in 2002; WPI/CPI annual growth averaged 3.7 percent and 2.6 percent.
    - Phase III: January 2003–December 2004 — float allowed January 2003; speculative hoarding and parallel market activity until interbank FX market launched December 2004; WPI and CPI inflation peaked at 21.7 percent (February 2004) and 17.2 percent (April 2004); two-year Phase III average WPI/CPI reported as 15.7 percent and 10.2 percent respectively.
    - Phase IV: Since January 2005 — pound appreciated about 7 percent within one quarter after launching FX interbank market in December 2004; WPI 3.1 percent (April 2005) and CPI 4.1 percent (April 2005); between February 2005 and July 2007 nominal exchange rate versus U.S. dollar broadly stable.
  - Empirical test: treating exchange rate as exogenous shows active exchange rate channel doubles or triples (in response to deposit rate shock) the magnitude of the price response due to depreciation effects on import prices.
  - For effective exchange rates, an active transmission channel broadly doubles the impact of a monetary shock on prices.
  - When exchange rate channel is active, price impact of monetary shock is smaller after about 1½ years (initial depreciation may give way to small appreciation).

- Asset price channel
  - Main documented asset market: CASE stock exchange.
  - CASE 30 index movements:
    - Between March 2003 and February 2006 CASE 30 increased by about 12 times.
    - CASE 30 lost about 35 percent within four months during regional stock market correction.
  - Inclusion of stock price in VAR (limited sample since January 1998):
    - Response of output to expansionary monetary policy muted.
    - Response of output to appreciation no longer significant.
    - Significance of price response to monetary stance lost.
    - Functioning asset price channel intensifies price response to exchange rate shock (increases amplitude but shortens significant response by about one year).
  - Two tentative mechanisms: wealth effects reducing consumption/investment; deterioration of export outlook lowering profit expectations.

- Bank lending channel
  - Aggregate and disaggregated evidence:
    - Total commercial bank lending to private sector as share of GDP increased until 2001 and decreased since.
    - Late 1990s boom-bust in lending mirrored business cycle; bank lending channel affected aggregate demand even with weak interest rate channel.
    - Lending to private sector, especially households, began to pick up again since 2006.
    - Between 2000 and 2006 banks built sizeable government securities positions, crowding out private lending.
  - Selected bank lending statistics (In percent of GDP; end of fiscal year July–June; selected years):
    - Total bank loans: 1996 — 60.3; 2001 — 72.8; 2006 — 55.8; 2007 — 51.8
    - Private sector: 1996 — 38.4; 2001 — 59.9; 2006 — 46.6; 2007 — 44.3
    - Business: 1996 — 30.3; 2001 — 49.3; 2006 — 36.9; 2007 — 35.0
    - Household: 1996 — 7.5; 2001 — 9.3; 2006 — 9.1; 2007 — 8.8
    - Holdings of government securities (memo): 1996 — 16.1; 2001 — 14.8; 2006 — 22.5; 2007 — 16.2
  - VAR results with credit aggregates:
    - Domestic credit expands significantly after monetary easing or NEER appreciation shocks, but impulse responses of output and prices similar to baseline.
    - Disaggregated credit: active credit channel increases positive output response to monetary easing; public sector credit transmission yields output response about twice as large as private sector credit; corporate lending more important for output than household lending.
    - Initial response of public sector credit to monetary easing is negative, pointing to decoupling of monetary stance and public lending decisions.

### Impact of the overnight corridor (post-June 2005)
- Corridor: overnight deposit and lending rates define floor and ceiling; CBE steers overnight interbank rate mid-corridor and manages liquidity via open market operations.
- Since corridor launch, overnight interbank rate substantially less volatile and potentially better indicator of stance.
- Granger causality tests for 2005–06 (short sample; Observations 16; lag length three), selected p-values/significance:
  - Overnight interbank rate Granger-causes 3-month deposit rate: F-Statistic 10.150 — Probability 0.00 ***
  - Overnight interbank rate Granger-causes lending rate: F-Statistic 5.600 — Probability 0.02 **
  - Six-month t-bill rate Granger-causes lending rate: F-Statistic 3.940 — Probability 0.05 *
- Interpretation: despite short sample, overnight interbank rate has started to Granger-cause other interest rates; after late-2006 increase the overnight rate remained close to corridor floor for most of 2007 amid surplus liquidity.

### Summary and conclusions
- CBE upgraded framework toward inflation targeting once prerequisites are met.
- Effective IT requires clear identification/communication of policy objective and tool, and understanding intensity/lags of interest rate transmission.
- Main empirical conclusions:
  - Interest rate channel provides correct signs but lacks sufficient significance/amplitude for IT move.
  - Exchange rate channel remains important and magnifies policy shocks drastically.
  - Asset price channel generally subdued; modeling intensifies price response to exchange rate shocks.
  - Bank lending channel shows stronger transmission to output via public sector credit than private sector; initial public sector credit response to easing negative despite positive medium-run response.
- Preconditions not yet fully met for successful IT: competitive banking system and strengthened monetary policy framework.
- Since June 2005 corridor launch, overnight interbank rate is less volatile and a better stance indicator, but empirical evidence is still emerging.
- Policy implications/priorities:
  - Improve CBE communication strategy to strengthen expectations channel.
  - Allow further exchange rate flexibility (important given large share of tradable goods in CPI).
  - As interest rate channel emerges, exchange rate channel expected to lose prominence and bank lending channel to grow with enhanced banking competition.
  - Weaken relative strength of transmission via public sector credit as private lending re-emerges.
  - Enhance asset price channel via mortgage finance, benchmark yield curve, and inflation-indexed instruments.
  - Strengthen link between mortgage rates and monetary stance.
  - Government priorities—clean up remaining NPLs and lower fiscal burden—should strengthen transmission mechanisms (bank lending and balance sheet channels).

_Italic: Source — _wp07285 - References.............................................................................................................._

### References..............................................................................................................

### _wp07285 - References..............................................................................................................

### I. INTRODUCTION — context and reform milestones
- Since the completion of the Economic Reform and Structural Adjustment Program (ERSAP) in 1996, Egypt has undertaken major structural changes in its monetary policy framework, including:
  - the abolition of the de jure exchange rate peg in 2000;
  - the introduction of the domestic currency overnight interbank market in 2001;
  - the launch of the foreign exchange interbank market in 2004;
  - the introduction of the corridor for overnight facilities as main policy instrument in June 2005.
- In June 2005 the Central Bank of Egypt (CBE) announced its intention to adopt inflation targeting (IT) over the medium term and declared price stability as the CBE’s overriding policy objective.
- The transition to IT implies that the exchange rate would no longer be the primary instrument to engineer a certain inflationary outcome; other transmission channels must function well for successful attainment of the inflation target.
- The paper combines a descriptive review of Egyptian monetary policymaking over the last decade with a baseline VAR model and extensions to investigate specific monetary transmission mechanism (MTM) channels, subject to data constraints.

### Main empirical results
- The main empirical results of this paper are:
  - (i) the interest rate channel of monetary policy is still weak in Egypt;
  - (ii) the exchange rate channel is very pronounced, especially in light of the infrequent but significant exchange rate movements in Egypt over the last decade;
  - (iii) other transmission channels play a very limited role in transmitting monetary shocks to the economy; and
  - (iv) the introduction of the corridor for the overnight rate may have started to contribute to a firmer interest rate channel.

### Policy recommendations
- The main policy recommendations include:
  - (i) enhancing competition in the banking system to reduce the spread between lending and borrowing rates and to strengthen the link between the monetary stance and macro variables via the interest rate channel;
  - (ii) improving the CBE’s communication strategy to enhance the expectations channel;
  - (iii) adding financial instruments to lengthen and deepen the yield curve (e.g., mortgage financing), possibly including instruments that reflect inflation expectations (inflation-indexed bonds);
  - (iv) allowing greater flexibility in the exchange rate, a precondition for successful inflation targeting; and
  - (v) persevering with other governmental reforms, including resolving the remaining non-performing loan (NPL) issue in the banking system and fiscal consolidation.

### Data, measurement, and recommended statistical improvements
- From a statistical perspective the paper notes:
  - improvements in price statistics would enhance empirical evaluations;
  - availability of an indicator of true economic activity at a monthly frequency and for a sufficient amount of time (as opposed to a production factor like electricity) would strengthen analyses;
  - expanding available indicators could be a useful contribution from Egypt’s statistical agency.

### Structure of the paper (as presented)
- The remainder of the paper is structured as follows:
  - a brief distinction of conventionally described channels of MTMs and a review of related literature on Egypt’s monetary policy;
  - a descriptive analysis of the major transmission channels in Egypt;
  - development of a baseline VAR model describing the Egyptian economy and investigation of several model extensions to probe MTM channels;
  - a final section summarizing findings and presenting policy recommendations.

*Source: _wp07285 - References..............................................................................................................*

### conclusions.

### conclusions.

### Background and literature on monetary transmission mechanisms
- Vector autoregressive models (Sim, 1980) facilitated a surge in empirical research on the monetary transmission mechanism (MTM).
- Commonly discussed MTM channels:
  - The (direct) interest rate channel.
  - The exchange rate channel.
  - The asset price channel.
  - The bank lending channel.
  - The balance sheet channel.
  - The expectations channel.
- For industrial countries, the main channel is found to work through interest rates transmitted via a competitive financial sector.
- For emerging markets, transition economies, and developing countries, many studies find a limited role for the direct interest rate channel, often due to a weak banking system.

### Empirical evidence specific to Egypt
- Exchange rate pass-through:
  - Rabanal (2005) finds WPI reacts significantly to changes in the nominal exchange rate after 6–12 months; CPI reacts after 12–24 months but not significantly (interpreted as structural weaknesses of the CPI measure used until 2003 and a significant share of administered prices).
  - Internal CBE research updating the CPI series shows that 26 percent of the shock is passed on to consumer prices after 12 months of the shock.
- Money gap and inflation:
  - Movements in the money gap explain 90 percent of WPI inflation dynamics between 1999 and late 2005.
  - The explanatory power of the output gap for WPI fluctuations over the same period is limited to 12 percent.
- Interest rate channel:
  - Hassan (2003) finds the nominal interest rate does not have a significant impact on real domestic credit to the private sector.
  - Interest rates play a small role in explaining movements in international reserves of the CBE relative to real domestic credit and the real exchange rate.
- MMZ (Moursi, Mossallamy, and Zakareya, 2007):
  - Estimate a structural VAR and simulate monetary policy rules.
  - Conclude the direct impact of monetary policy shocks on real output is negligible (supporting money neutrality).
  - Argue for an indirect positive growth effect via targeting long-run price stability.

### Data and channel coverage limitations
- Analysis focuses on the first four channels (interest rate, exchange rate, asset price, bank lending) due to lack of reliable data for:
  - Balance sheet channel: no detailed balance sheet data by currency over sufficient time/frequency covering a sizeable fraction of the corporate sector; CBE’s reporting to IMF SRFs does not date back far enough.
  - Expectations channel: limited data on inflation expectations (no inflation-indexed bonds; no representative survey comparable to those in some advanced economies).

### Monetary policy framework and structural context (1996–June 2005)
- Period marked by multiple, sometimes conflicting, objectives: attaining high economic growth, maintaining low inflation, and preserving a stable exchange rate.
- Increasing capital mobility (primarily inflows) made independent monetary policy virtually impossible and clouded measurement of monetary policy stance during 1996–2005.
- Operational and intermediate targets:
  - Operational target: excess reserves of banks.
  - Intermediate target: growth in M2 (given strong link between monetary aggregates and inflation).
- Policy instruments used at different times included reserve requirements, government securities, repo and reverse repo operations, and the CBE discount rate.
- Treasury bill rates provided some market indications but, given their fiscal role, were misleading as a monetary policy instrument.
- Banking sector structure and distortions:
  - Dominance of the state in the banking sector until recently created rigidities in the interest rate structure.
  - Nonperforming loans (NPLs) created distortions that weakened links between price measures and macro outcomes.
  - In the mid 1990s, public banks accounted for about two-thirds of sector assets.
  - By 2006, privatization reduced public bank market share; privatization of the third-largest public bank was scheduled, expected to bring market share of public banks significantly below 50 percent.

### Transition toward an inflation-targeting-compatible framework
- The exchange rate served as the key nominal anchor during 1996–2005; exchange rate channel proved more effective than the interest rate channel in this context.
- Under inflation targeting (IT):
  - Central banks commonly use price instruments and rely on a functioning and competitive banking system for interest-rate transmission to be effective.
  - The exchange rate is abandoned as the nominal anchor and replaced by an explicit price-stability objective; interest-rate decisions become important and easier to communicate.
- To prepare for an IT regime, the CBE launched a comprehensive banking-sector reform program in 2004, including:
  - Restructuring and privatization of banks with state participation (NPL-related).
  - A new banking law and other regulatory reforms.
  - Liberalization of the foreign exchange and money markets.
  - Efforts to strengthen bank supervision.
- These institutional and operational changes aimed to address previous banking-sector shortcomings and lay groundwork for formally adopting inflation targeting over the medium term.

### Implications for transmission-channel effectiveness and policy assessment
- Over 1996–2005, monetary policymaking underwent structural breaks; each transmission channel faces specific constraints related to instrument identification and data availability.
- Quantitative measures likely played a more important role in transmission than price instruments (interest rates) during the pre-2005 period due to banking-sector distortions.
- Understanding the intensity and lags of policy interest-rate impacts is critical for effective IT adoption (Schmidt-Hebbel and Tapia, 2002).

*Source: _wp07285 - conclusions.*

### Box 1. A New Monetary Policy Framework: Steps Toward Inflation Targeting

### Box 1. A New Monetary Policy Framework: Steps Toward Inflation Targeting

### Institutional reforms to support inflation targeting
- Coordinating Council on Monetary Policy headed by the Prime Minister established in January 2005 to ensure government policies are consistent with objectives of monetary policy; first meeting discussed objectives of monetary policy and the importance of CBE independence.
- Monetary Policy Committee (MPC) established at the CBE; convenes on Thursdays every six weeks to decide on key policy rates; MPC consists of nine members: the Governor, the two Deputy Governors, and six members of the CBE’s Board of Directors.
- Monetary Policy Unit established as a satellite unit within the CBE to provide objective monetary policy analysis, assessment, and modalities of communication with the market through research and other functions.
- To enhance transparency and anchor inflation expectations, MPC decisions are communicated via a monetary policy statement released on the CBE’s external web-site after each meeting.

### Operational reforms and instruments
- On June 2, 2005 the CBE introduced an interest rate corridor defined by the CBE’s two standing facilities, the overnight lending and a deposit facility; the interest rates on the two standing facilities define the ceiling and floor of the corridor respectively, making steering the overnight rate within this corridor the operational target of the CBE.
- Starting in August 2005, central bank notes were added to the CBE’s toolkit as the primary instrument for liquidity management via open market operations, in addition to deposit auctions.
- Alternative models to forecast inflation have been developed.

### The interest rate channel: historical assessment (1996–2005)
- The overnight domestic currency interbank market was introduced in 2001; the overnight interest rate was extremely volatile at the beginning and thus not a reliable measure of the monetary stance.
- Several alternative interest rates could serve as indicators: the 3-month treasury bill rate (issued in coordination with the CBE to sterilize capital inflows, with the amount issued substantially exceeding the financing needs of the government) and the CBE’s discount rate (which cannot be considered a key signaling device between 1996 and 2005 due to a continued downward trend).
- Movements in observed interest rates appear secular with no evident cyclical pattern, suggesting a limited role of the interest-rate channel in economic fluctuations in Egypt during the period.
- During 1995–2005, the coefficient of variation in nominal policy interest rates (treasury bill rate and discount rate) ranged between 0.1 and 0.3 throughout the period; nominal retail rates showed low variability ranging between 0.04 and 0.1.
- Granger causality tests (see Section IV. C) suggest a weak interest rate transmission channel for Egypt during the period under investigation.
- The CBE’s operational target at the time was banks’ excess reserves, which were very volatile; the intermediate target—M2—grew at a broadly constant rate. Quantitative measures were considered more effective in steering aggregate demand.
- Example shocks and outcomes:
  - Exogenous shocks in 1996/97 (decline in international oil price, the East Asian crisis, and the Luxor attack) triggered a rapid weakening of Egypt’s external position as capital inflows and tourism revenues subsided.
  - These shocks combined with loose monetary and fiscal policies led to a widening of the current account deficit to about 3 percent of GDP in 1997/98 as the trade balance worsened.
  - Domestic credit continued to grow at around 25 percent annually through end-1999 on average while interest rates remained flat; real GDP growth started to slow after 1997/98 with a collapsing share of investment in GDP.
  - Drastically slowing goods imports (by volume), together with a decrease of the value (in U.S. dollars) of investment and intermediate goods between 1998/99 and 2001/02, point to growing difficulties importing intermediate goods vital for production.
- Policy responses and banking sector effects:
  - To address exchange rate pressure and financing needs, the CBE resorted to quantitative trade measures including requirements for importers to put up a 100-percent cover when opening import letters of credit.
  - The CBE reduced the reserve requirement ratio on domestic-currency deposits from 15 to 14 percent and changed the calculation of the ratio (removing the 3-year saving certificates from the denominator) to free resources and encourage investment.
  - The maintenance period was lengthened from one to two weeks in 2002 to support banks in liquidity management.
  - These measures exacerbated lending conditions and led to a “liquidity crunch” in the banking sector.
- Empirical approach: because the CBE did not have a continuous indicator of the monetary policy stance over 1996/97–2005, the empirical section uses a constructed measure of the policy stance derived in MMZ (2007). As a robustness test, reserve money will also be employed.

### The exchange rate channel: phases and dynamics (1997–2005)
- The exchange rate channel is central where financial markets are underdeveloped; changes in the monetary policy stance are reflected in the exchange rate and transmitted to domestic prices via exchange rate pass-through.
- For descriptive purposes, four phases against the U.S. dollar are distinguished:

  Phase I: January 1997 to December 2000
  - Egyptian pound de jure and de facto pegged to the U.S. dollar.
  - To maintain the peg amid substantial balance of payments deficits, the CBE intervened and reduced foreign reserves from US$16.8 billion in early 1997 to US$12.9 billion in late 2000.
  - A series of step devaluations starting in April 1999 amounted to 8.4 percent (against the U.S. dollar).
  - Wholesale and consumer price inflation—annual average of 2.1 percent and 3.2 percent respectively—remained low with minimal volatility.

  Phase II: January 2001 to December 2002
  - Exchange rate set to crawl within horizontal bands beginning in 2001 (initially ±1 percent, widened to ±3 percent in August 2001).
  - Supply of foreign exchange in the official market dropped sharply in 2002 as exporters and holders sought parallel market rates.
  - Parallel market traded at a 15 percent premium over the official rate in 2002.
  - WPI and CPI annual growth rates averaged 3.7 percent and 2.6 percent respectively over the period.

  Phase III: January 2003 to December 2004
  - Exchange rate allowed to float in January 2003; lack of credibility and expectations of further depreciation led to hoarding of foreign exchange receipts and speculative activity amid an inoperative interbank market, causing official channel shortages and reemergence of the parallel market.
  - By mid-2004, a formal interbank foreign exchange market was essential; in December 2004 the CBE launched a new interbank foreign exchange market accommodating all foreign exchange transactions between banks.
  - As the interbank market became fully functioning, parallel market activity diminished due to ample dollar liquidity from current account inflows reflecting competitiveness gains from prior real effective depreciation.
  - Turbulence during this phase was reflected in highly volatile domestic prices: WPI and CPI inflation peaked at 21.7 percent (February 2004) and 17.2 percent (April 2004) respectively.
  - On average over the two-year Phase III period, WPI and CPI inflation amounted to (figure truncated in source).

- Empirical modeling note: while the U.S. dollar relationship is prominent given past pegs, the nominal effective (trade-weighted) exchange rate is used in the baseline empirical model in the next section.

*Source: Box 1. A New Monetary Policy Framework: Steps Toward Inflation Targeting (excerpt).*

### 15.7 percent and 10.2 percent, respectively. The persistently high inflation levels that

### _wp07285 - 15.7 percent and 10.2 percent, respectively. The persistently high inflation levels that

### Phase IV: Since January 2005
- The pound appreciated by about 7 percent within one quarter of launching the foreign-exchange interbank market in December 2004, with a corresponding disinflationary impact on domestic prices.
- Between February 2005 and July 2007, the nominal exchange rate versus the U.S. dollar has been broadly stable, limiting exchange rate effects on domestic prices.
- Annual inflation reversed to lower levels in early 2005, reaching:
  - 3.1 percent for the WPI in April 2005
  - 4.1 percent for the CPI in April 2005
- Starting in early 2006, inflation began to rise and spiked in March 2007, driven largely by:
  - the impact of an avian flu outbreak,
  - adjustments in administered prices,
  - some domestic demand pressures emanating from high economic growth.

### The Asset Price Channel
- The asset price channel operates via Tobin’s q theory and wealth effects on consumption and output; it can work through securities markets and real estate.
- For Egypt, the most documented asset market is the stock exchange (Cairo and Alexandria Stock Exchanges - CASE).
- Key observations on stock market developments:
  - After subdued activity since the mid-1990s, activity picked up from 2003.
  - Between March 2003 and February 2006, the CASE 30 index increased by about 12 times.
  - The CASE 30 index then lost about 35 percent within four months during the regional stock market correction.
  - Anecdotal evidence indicates a strong increase in real estate prices across Egypt, but data are sketchy and empirical analysis is not possible in this paper.
- Selected CASE indicators (1996–2006, as reported in Table 1):
  - Number of companies listed: 64 65 48 70 103 107 111 0115 197 879 574 459 5
  - Number of companies traded: 35 44 16 55 16 63 65 96 43 67 15 40 50 34 41 40 7
  - Market capitalization (L.E. billions): 48 71 82 112 121 112 122 172 234 456 534
  - Market capitalization (percent of GDP): 19 27 29 36 36 30 29 35 43 74 80
  - Total volume of stocks (L.E. billions): 0.2 0.4 0.6 1.1 1.1 1.3 0.9 1.4 2.4 5.3 9.1
  - Total value traded (L.E. billions): 1 1 2 4 2 3 3 9 5 3 3 4 2 8 4 2 1 6 1 2 8 7
  - Turnover ratio (%): 23 29 23 29 35 14 10 13 13 15 31 51
  - Egyptian investors in terms of value traded (percent): 83 80 78 79 84 81 80 73 70 70
  - Foreign investors in terms of value traded (percent): 17 20 22 21 13 17 13 21 16 17
  - Arab investors in terms of value traded (percent): 3 2 8 7 14 14
  - CASE 30 index (selected points): 100 066 0114 0716 4904 9311 5625 6863 2569 73
  - (Source cited in table: Capital Market Authority.)
- Note: The tabulated indicators reflect rapid development in the stock market since 2003 and large swings in valuation and turnover.

### Bank Lending Channel
- Under the bank lending channel, changes in credit availability resulting from monetary policy (e.g., reserve requirements, credit ceilings) affect economic activity and amplify interest rate channel effects.
- Evidence on bank lending and activity is mixed:
  - Total lending from commercial banks to the private sector as a share of GDP increased until 2001 and decreased since then (see Table 2).
  - The late 1990s boom-bust cycle in bank lending mirrored the business cycle, implying the bank lending channel affected aggregate demand even with a weak interest rate channel.
  - The economic expansion since 2004 was not accompanied, until recently, by a sizeable increase in commercial banks’ lending to the private sector.
  - Low variability in lending rates and government ownership of a large share of bank assets contributed to a disconnect between borrowing decisions and prevailing interest rates.
- Historical factors and policy actions:
  - After abolishing credit ceilings on lending to the private business sector (PBS) in October 1992, loans to the PBS accelerated, reaching 33 percent (y/y) growth in 1994/95.
  - By end-1999 a credit crunch occurred; slowdown in lending was driven by credit rationing measures including:
    - 100-percent cover imposed by the CBE on letters of credit for imports,
    - drastic reduction of bank lending to the construction sector,
    - significant government borrowing from the private corporate and household sector,
    - tighter lending requirements after large NPLs emerged, partly tied to a real estate bubble.
  - As a consequence of the real estate lending crisis, the CBE adopted a prudential regulation limiting the share of banks’ lending to the real estate sector to 5 percent of total lending.
- Crowding out and bank behavior:
  - Between 2000 and 2006, commercial banks built up sizeable positions of government securities, crowding out private-sector lending.
  - Stable investment income from government securities helped banks address NPL issues, forming the basis for a “quiet” recapitalization.
  - Lending to the private sector, especially households, began to pick up again since 2006.
  - A recent substantial drop in yields of government securities exerted pressure on banks’ income, cost of funding, and interest rate spreads; the decrease in yields is linked to increased integration into global financial markets and foreign investor interest in Egyptian government securities.
- Banking sector reform program (initiated 2003):
  - Increased minimum bank capital and strengthened supervision.
  - Number of banks declined from 57 in 2004 to 39 in 2007, with an increase in average bank size.
  - CBE tackled resolution of remaining NPLs, cleaning up public banks’ balance sheets.
  - State banks still account for a large share of sector assets, but reforms should improve the bank lending channel going forward.
- Selected bank lending statistics (Table 2; In percent of GDP; end of fiscal year July–June):
  - 1996 / 2001 / 2006 / 2007:
    - Total bank loans: 60.3 72.8 55.8 51.8
    - Private sector: 38.4 59.9 46.6 44.3
    - Business: 30.3 49.3 36.9 35.0
    - Household: 7.5 9.3 9.1 8.8
    - Foreign: 0.6 1.0 0.5 0.6
    - Public sector: 21.9 12.9 9.2 7.4
    - Public economic authorities: 3.4 3.5 3.5 3.8
    - State-owned enterprises: 13.4 8.8 5.6 3.6
    - Government: 5.1 0.5 0.1 0.0
  - Memo item:
    - Holdings of government securities: 16.1 14.8 22.5 16.2
  - (Sources: Central Bank of Egypt; and authors' calculations.)

### Interest Rate Spreads and Banking Sector Dynamics
- The drop in government security yields pressured banks’ income and interest rate spreads.
- The reform program and recapitalization through government securities holdings have reduced NPLs and led to consolidation (fewer banks, larger average size).

### Monetary Transmission in Egypt: Some Empirical Results — Data and Choice of Variables
- Baseline VAR specification uses monthly observations from January 1996 to June 2005, when the overnight corridor was introduced.
- The maximum number of observations for any given model is 114, but some series start later, reducing observations slightly in selected models.
- VAR variables include:
  - a measure of economic activity,
  - the price level,
  - a measure of the monetary policy stance,
  - the exchange rate.
- Monthly-frequency construction issues:
  - National accounts data (consumption, investment, output) are not available monthly; the authors use a high-frequency measure of economic activity developed by MMZ, which distributes lower-frequency real GDP data to monthly frequency.
  - The monetary policy stance measure used is the MMZ indicator (Litterman (1983) measure), which captures both anticipated and surprise components of monetary policy and yields the most significant results in most specifications.
  - The 3-month deposit rate was also used in alternative specifications and produced qualitatively similar but less significant results, pointing to the CBE’s limited ability to control price developments and signal monetary stance during the sample period (i.e., before the interest rate corridor introduction).
- Methodological notes:
  - MMZ (2007) develop two high-frequency measures following distribution techniques of Litterman (1983) and Chow and Lin (1971); the Litterman (1983) measure is used in what follows.
- Time series illustration in the text:
  - The monetary policy stance series and the 3-month deposit interest rate are plotted over 1995–2006 to show movements and relative behavior.

*Italic: Source — _wp07285 (excerpt) as provided in the supplied content.*

### Box 2. The Monetary Policy Stance Index

### Box 2. The Monetary Policy Stance Index

### MMZ methodology and VAR structure
- MMZ (Moursi, Mossallamy, and Zakareya, 2007) apply a six-variable semi-structural VAR methodology pioneered by Bernanke and Mihov (1998) to compute a monetary policy stance index for Egypt and assess its impact on the economy.
- BM framework: identify monetary instruments that best describe policy shocks and choose a model to illustrate behavioral dynamics (see Favero, 2001).
- Market for bank reserves represented in innovation form by equations (1)–(3):
  - (1) d IRTR = α - νa + uu  (banks’ demand for total reserves; IR affects negatively)
  - (2) b IRBR = β + uuν  (banks’ demand for borrowed reserves; IR affects positively)
  - (3) sbbdd NBR = φν + φν + νν + u  (supply of non-borrowed reserves; sν represents monetary policy shock)
- MMZ identify policy variables: total reserves, non-borrowed reserves, short-term interest rates. Non-policy variables: real GDP, GDP deflator, index of commodity prices.
- Data: monthly data for Egypt, 1985–2005.
- Interpretation: A higher value of the measure indicates monetary easing. The measure is based on the Bernanke and Mihov (1998) model; MMZ find it fits Egyptian data best.

### Granger causality (preliminary evidence) — Table 3 (Baseline VAR, 1995M1–2005M6)
- Dependent variable: Prices (Excluded variable — Chi-square (dof) — p-value)
  - Prices: 9.333 0.025 **
  - Monetary policy stance measure: 0.503 0.919
  - Three-month deposit rate: 4.323 0.229 ***
  - NEER: 8.593 0.035 **
  - Jointly: 25.709 0.002 ***
- Dependent variable: Prices (alternate column)
  - Prices: 9.693 0.021 **
  - Monetary policy stance measure: [blank in source]
  - Three-month deposit rate: [blank in source]
  - NEER: 13.053 0.005
  - Jointly: 35.729 0.000
- Dependent variable: Output (Excluded variable — Chi-square (dof) — p-value)
  - Output: 16.453 0.001 ***
  - Monetary policy stance measure: 4.933 0.177
  - Three-month deposit rate: 4.993 0.173
  - NEER: 14.123 0.003 ***
  - Jointly: 32.039 0.000 ***
- Dependent variable: Output (alternate column)
  - Output: 18.243 0.000 ***
  - Monetary policy stance measure: [blank in source]
  - Three-month deposit rate: [blank in source]
  - NEER: 18.723 0.000 ***
  - Jointly: 42.599 0.000 ***
- Note: Block Granger causality test for exclusion is based on a Wald test and follows a χ² distribution; *, **, and *** denote rejection at the 10, 5, and 1 percent level.

### Baseline VAR key results: impulse responses and variance decomposition
- Impulse responses (selected):
  - Prices: strongly significant response to exchange rate shocks; response to MMZ monetary stance measure is not quite significant in baseline but closer to significance if sample extended until end-2005.
  - Output: response to monetary policy stance shock is in line with expectations (output starts rising after about a year in response to an easing stance) though not statistically significant.
  - Output: significant positive response to a nominal appreciation shock (counterintuitive and noted for further attention).
  - WPI: strong deflationary impact of a tightening is intuitive and consistent with earlier findings (Rabanal, 2005).
- Variance decomposition (Baseline VAR, 1995–2005; in percent of total variance):
  - Variance Decomposition of Output (Months — Standard error — Output — WPI — MP measure — NEER)
    - 10 — 0.0081 — 100.0 — 0.0 — 0.0 — 0.0
    - 20 — 0.0083 — 96.0 — 1.7 — 0.0 — 2.3
    - 30 — 0.0092 — 83.8 — 5.7 — 0.2 — 10.2
    - 60 — 0.0118 — 78.1 — 4.0 — 0.3 — 17.6
    - 120 — 0.0148 — 68.4 — 2.9 — 0.4 — 28.4
    - 240 — 0.0180 — 66.0 — 3.7 — 3.4 — 26.8
    - 600 — 0.0251 — 71.2 — 4.5 — 7.2 — 17.2
  - Variance Decomposition of NEER
    - 10 — 0.0085 — 0.2 — 99.9 — 0.0 — 0.0
    - 20 — 0.0117 — 3.1 — 94.8 — 0.4 — 1.7
    - 30 — 0.0132 — 2.6 — 92.3 — 0.5 — 4.6
    - 60 — 0.0178 — 2.0 — 68.9 — 1.2 — 27.8
    - 120 — 0.0269 — 3.9 — 35.2 — 1.8 — 59.2
    - 240 — 0.0365 — 7.6 — 19.9 — 1.3 — 71.2
    - 600 — 0.0409 — 11.3 — 16.5 — 3.9 — 68.3
  - Variance Decomposition of Monetary Policy Measure
    - 10 — 0.3427 — 0.0 — 0.2 — 99.8 — 0.0
    - 20 — 0.4708 — 0.0 — 2.0 — 98.0 — 0.0
    - 30 — 0.5787 — 0.3 — 1.6 — 98.2 — 0.0
    - 60 — 0.7717 — 0.6 — 1.3 — 98.0 — 0.1
    - 120 — 0.9575 — 1.8 — 1.7 — 96.3 — 0.2
    - 241 — 1.0762 — 4.8 — 1.9 — 93.1 — 0.3
    - 601 — 1.1734 — 13.3 — 1.8 — 84.3 — 0.6
  - Variance Decomposition of WPI
    - 10 — 0.0226 — 0.0 — 2.3 — 0.0 — 97.7
    - 20 — 0.0355 — 0.0 — 1.4 — 0.5 — 98.1
    - 30 — 0.0439 — 0.1 — 2.0 — 1.6 — 96.4
    - 60 — 0.0604 — 0.1 — 2.0 — 2.0 — 95.9
    - 120 — 0.0791 — 0.4 — 1.3 — 1.3 — 97.0
    - 240 — 0.0911 — 0.6 — 1.0 — 3.5 — 94.9
    - 600 — 0.0956 — 1.4 — 0.9 — 8.8 — 88.9
- Interpretation: Most variables are highly idiosyncratic; even after five years, three of four variables’ variation explained largely by their own innovations. Exception: WPI—after barely one year its variation is mainly explained by NEER (strong exchange rate pass-through to WPI).

### Robustness checks
- Substitute reserve money for MMZ monetary policy stance measure:
  - Key results unchanged: output continues to respond positively (but insignificantly) to expansionary monetary shocks; WPI no longer shows the initial (insignificant) increase from baseline.
  - Reserve money as an indicator of stance is limited: nominal effective exchange rate appreciates in response to an expansionary monetary shock.
- Inversion of exchange rate and monetary stance ordering: no effect on results—baseline specification robust.

### Exploring MTM channels

- The interest rate channel
  - Functioning interest rate channel crucial for transmission; without it central bank capacity to influence real activity is limited.
  - Pairwise Granger causality tests between interest rate series and monetary policy stance measure (lag length three). Table 5 (1995–2005):
    - Six-month t-bill rate does not Granger-cause lending rate: Observations 123 — F-Statistic 0.230 — Probability 0.87
    - Lending rate does not Granger-cause 6-month t-bill rate: 0.350 — 0.79
    - Three-month deposit rate does not Granger-cause lending rate: 128 — 3.48 — 0.02 **
    - Lending rate does not Granger-cause 3-month deposit rate: 1.30 — 0.28
    - MP measure does not Granger-cause lending rate: 117 — 0.01 — 1.00
    - Lending rate does not Granger-cause MP measure: 0.52 — 0.67
    - Three-month deposit rate does not Granger-cause 6-month t-bill rate: 123 — 0.69 — 0.56
    - Six-month t-bill rate does not Granger-cause 3-month deposit rate: 3.62 — 0.02 **
    - Mp measure does not Granger-cause 6-month t-bill rate: 112 — 0.29 — 0.83
    - Six-month t-bill rate does not Granger-cause MP measure: 0.36 — 0.78
    - MP measure does not Granger-cause 3-month deposit rate: 117 — 2.38 — 0.07 *
    - Three-month deposit rate does not Granger-cause MP measure: 0.99 — 0.40
    - Note: *, ** imply significance at the 10, 5 percent level.
  - Findings:
    - Interest rates in Egypt are rather unrelated to each other; non-causal relationship rejected in only three cases at the 10-percent level.
    - Monetary policy stance measure appears to Granger-cause the 3-month deposit rate but not vice versa.
    - The 3-month deposit rate has some impact on the lending rate.
  - Re-estimated baseline VAR including lending rate, 3-month deposit rate, and monetary policy stance between price measure and exchange rate:
    - Significant reduction of the 3-month deposit rate after an expansionary shock to the monetary policy measure (consistent with Granger causality).
    - Almost-significant response of the lending rate to the policy stance shock (not mirrored in Granger tests).
    - Absence of an effective policy interest rate before the introduction of the overnight corridor hampers transmission at the short end.
    - Results provide limited evidence of an interest rate mechanism beyond the very short end.

- The exchange rate channel
  - Given Egypt’s exchange rate experience, the exchange rate channel is important.
  - Test: shut down transmission by treating the exchange rate as exogenous (include in Z_t).
  - Using NEER (baseline) and comparing monetary policy measure vs. 3-month deposit rate:
    - Exchange rate channel plays an important role: after a monetary easing, an active exchange rate channel doubles or triples (in response to a deposit rate shock) the magnitude of the price response due to depreciation effects on import prices.
    - For output: an active exchange rate channel appears to delay the expansionary reaction for both monetary policy measures, possibly due to depreciation-induced import price rises reducing availability of critical imported inputs—this effect could be driven by monetary easing in the late 1990s that coincided with a drop in investment and intermediate imports.
  - Results for alternative exchange rates (REER, bilateral U.S. dollar) are largely similar.

*Source: Box 2, "_wp07285 - Box 2. The Monetary Policy Stance Index_", authors' calculations and estimates as presented in the source content.*

### conclusion, however, can be drawn from comparing the responses in the top two panels of

### _wp07285 - conclusion, however, can be drawn from comparing the responses in the top two panels of

### Exchange rate channel
- For effective exchange rates, an active transmission channel broadly doubles the impact of a monetary shock on prices.
- The U.S. dollar exchange rate channel does not affect the amplitude for the transmission of the monetary policy stance (as measured by the MMZ variable) to the WPI, consistent with the pound being de facto mostly pegged to the U.S. dollar throughout the period.
- For all three exchange rates, an active exchange rate channel results in a much smaller price impact of a monetary shock after about 1½ years compared to an inactive exchange rate channel, i.e., once the initial depreciation has given way to a small appreciation.
- The appreciation of the NEER over the medium term in response to a monetary easing explains why the response of the WPI to the MP measure dies out much more quickly when the exchange rate channel is alive (top left panel in Figure 5).

### Asset price channel
- Major asset classes in Egypt include real estate and equity; statistical measures are insufficient except for equity quoted on CASE (CASE 30 index).
- Stock market index data are available since January 1998; the baseline VAR was re-estimated for this limited period with the (log) stock price ordered after output and the WPI but before the monetary policy stance.
- Key differences when including asset prices (Figure 7):
  - The response of output to expansionary monetary policy is muted (top left panel).
  - The response of output to an appreciation of the exchange rate is no longer significant (top right panel).
  - The borderline significance of the response of prices to shocks in the monetary policy stance is lost (bottom left panel).
  - A functioning asset price channel intensifies the price response to an exchange rate shock: it increases the amplitude but shortens the time during which the response is significant by about one year (bottom right panel).
- Two tentative explanations for the intensified but shorter price response to exchange rate shocks:
  - Wealth effect: appreciation raises cost of LE-based stock market investments, foreign investors reduce positions, stock market index decreases, investors experience lower net worth and adjust consumption and investment.
  - Export outlook: appreciation could deteriorate outlook for export-oriented sectors (tourism, non-oil goods exports), lowering profit expectations and exerting pressure on stock market quotations.
- VAR specification notes:
  - One lag used as indicated by Schwarz and Hannan-Quinn information criteria; Akaike criterion pointed to 12 lags but one lag is a local optimum according to Akaike.
  - Results robust to ordering the stock market index before the monetary policy stance.

### Bank lending channel
- A real credit aggregate (total private domestic and public credit, excluding the government proper but including state-owned enterprises and public economic authorities) was included in the baseline VAR, ordered between the price index and the monetary policy measure.
- Domestic credit expands significantly after shocks corresponding to monetary easing and an appreciation of the nominal effective exchange rate (Figure 8), but impulse responses of output and prices to these shocks are very similar to the baseline scenario.
- Disaggregated analysis (Figure 9):
  - An active credit channel increases the positive response of output to a monetary easing in all cases.
  - Transmission via public sector credit yields an output response broadly twice as large as transmission via private sector credit.
  - Transmission via corporate lending is quantitatively more important for output than lending to households.
  - Responses of public and private credit to a monetary easing are both positive in the medium run, but the initial response of public sector credit is negative, pointing to decoupling of the monetary stance and lending decisions in the public sector.
- Credit aggregates were deflated with the WPI and seasonally adjusted (X-12 filter).

### The impact of the target corridor for the CBE’s overnight standing facilities
- The CBE launched the overnight corridor system in June 2005, shifting from a quantity-based target (excess reserves) to a price-based target (overnight interbank rate).
- The overnight deposit rate and the overnight lending rate serve as the floor and ceiling of the corridor; the CBE steers the domestic currency overnight interbank rate in the middle of the corridor and manages market liquidity through open market operations.
- Since the corridor launch, the overnight interbank rate has become substantially less volatile and potentially a better indicator of the monetary policy stance.
- Granger causality tests for 2005–06 (short sample; Table 6; lag length is three; 1/ *,** imply significance at the 10, 5 percent level):
  - Lending rate does not Granger-cause 3-month deposit rate — Observations 16 — F-Statistic 1.190 — Probability 0.37
  - Three-month deposit rate does not Granger-cause lending rate — Observations 16 — F-Statistic 1.280 — Probability 0.35
  - Overnight interbank rate does not Granger-cause 3-month deposit rate — Observations 16 — F-Statistic 10.150 — Probability 0.00 ***
  - Three-month deposit rate does not Granger-cause Overnight interbank rate — Observations 16 — F-Statistic 0.970 — Probability 0.45
  - Six-month t-bill rate does not Granger-cause 3-month deposit rate — Observations 16 — F-Statistic 2.420 — Probability 0.14
  - Three-month deposit rate does not Granger-cause 6-month t-bill rate — Observations 16 — F-Statistic 1.230 — Probability 0.36
  - Overnight interbank rate does not Granger-cause lending rate — Observations 16 — F-Statistic 5.600 — Probability 0.02 **
  - Lending rate does not Granger-cause Overnight interbank rate — Observations 16 — F-Statistic 1.680 — Probability 0.24
  - Six-month t-bill rate does not Granger-cause lending rate — Observations 16 — F-Statistic 3.940 — Probability 0.05 *
  - Lending rate does not Granger-cause 6-month t-bill rate — Observations 16 — F-Statistic 1.380 — Probability 0.32
  - Six-month t-bill rate does not Granger-cause Overnight interbank rate — Observations 16 — F-Statistic 1.180 — Probability 0.38
  - Overnight interbank rate does not Granger-cause 6-month t-bill rate — Observations 16 — F-Statistic 2.180 — Probability 0.17
- Interpretation:
  - Despite the extremely short sample, the overnight interbank rate has started to assume a strong role in the interest rate channel, Granger-causing interest rates on 3-month deposits and new bank borrowing (the lending rate).
  - After the increase in the overnight interest rate in late 2006, the rate remained close to the floor of the corridor for most of 2007 amid surplus market liquidity.

### Summary and conclusions
- The CBE has upgraded Egypt’s monetary policy framework with a view to adopt inflation targeting once prerequisites are fulfilled.
- Effective monetary policy under inflation targeting requires clear identification and communication of the policy objective and tool, and an understanding of the intensity and lags of interest rate transmission to the economy and the policy objective.
- Main empirical findings:
  - The interest rate channel provides correct signs in Egypt, but significance and amplitude are not satisfactory for the planned move toward inflation targeting.
  - The exchange rate channel continues to play an important role and magnifies the impact of policy shocks drastically.
  - The asset price channel is generally subdued, but explicit modeling intensifies the response of prices to exchange rate shocks.
  - The bank lending channel points to stronger transmission of the monetary policy stance on output through credit to the public sector compared to private sector lending; initial public sector credit response to monetary easing is negative despite positive medium-run responses.
- Preconditions not yet fulfilled for successful IT (not exhaustively discussed): a competitive banking system and a strengthened monetary policy framework.
- Since the corridor launch in June 2005, the overnight interbank rate is less volatile and has proven to be a better indicator of the monetary policy stance, though empirical evidence is slow to emerge.
- Policy implications and priorities:
  - Further improve the CBE’s communication strategy to strengthen the monetary policy framework and the expectations channel.
  - Allow further exchange rate flexibility as an important precondition for successful inflation targeting, given a large share of tradable goods in the CPI basket.
  - As the interest rate channel emerges in the transition to IT, expect the exchange rate channel to lose prominence and the bank lending channel to grow stronger with enhanced banking competition.
  - The relative strength of transmission via public sector credit should weaken once private sector lending re-emerges.
  - Enhance asset price channel effects via mortgage finance introduction, establishment of a benchmark yield curve, and financial instruments that lengthen and deepen the yield curve (e.g., mortgage financing) and reflect inflation expectations (e.g., inflation-indexed bonds).
  - Strengthen the link between mortgage interest rates and the monetary stance.
  - Government priorities—cleaning up remaining NPLs and lowering the fiscal burden—should strengthen transmission mechanisms, especially the bank lending channel and the balance sheet channel, by improving banks’ and firms’ capacity to respond to interest rate changes.

*Source: Authors' calculations.*

### Appendix I. Impulse Responses, Baseline VAR

### Appendix I. Impulse Responses, Baseline VAR

### Methodology
- Response to Cholesky one standard deviation innovations ± 2 standard errors

### Responses of Output
- Response of Output to Output
  - Axis/labels text: -.004-.002
  - Axis/labels text: .0 0 0.0 0 2.0 0 4.0 0 6.0 0 8.0 1 0
  - Period markers: 5 10 15 20 25 30 35
- Response of Output to Prices
  - Axis/labels text: -.004-.002
  - Axis/labels text: .0 0 0.0 0 2.0 0 4.0 0 6.0 0 8.0 1 0
  - Period markers: 5 10 15 20 25 30 35
- Response of Output to Reserve Money
  - Axis/labels text: -.004-.002
  - Axis/labels text: .0 0 0.0 0 2.0 0 4.0 0 6.0 0 8.0 1 0
  - Period markers: 5 10 15 20 25 30 35
- Response of Output to Exchange Rate
  - Axis/labels text: -.004-.002
  - Axis/labels text: .00 0.00 2.00 4.00 6.00 8.01 0
  - Period markers: 5 10 15 20 25 30 35

### Responses of Prices
- Response of Prices to Output
  - Axis/labels text: -.015-.010-.005
  - Axis/labels text: .0 0 0.0 0 5.0 1 0
  - Period markers: 5 10 15 20 25 30 35
- Response of Prices to Prices
  - Axis/labels text: -.015-.010-.005
  - Axis/labels text: .0 0 0.0 0 5.0 1 0
  - Period markers: 5 10 15 20 25 30 35
- Response of Prices to Reserve Money
  - Axis/labels text: -.015-.010-.005
  - Axis/labels text: .0 0 0.0 0 5.0 1 0
  - Period markers: 5 10 15 20 25 30 35
- Response of Prices to Exchange Rate
  - Axis/labels text: -.015-.010-.005
  - Axis/labels text: .00 0.00 5.01 0
  - Period markers: 5 10 15 20 25 30 35

### Responses of Reserve Money
- Response of Reserve Money to Output
  - Axis/labels text: -.010-.005
  - Axis/labels text: .0 0 0.0 0 5.0 1 0.0 1 5.0 2 0
  - Period markers: 5 10 15 20 25 30 35
- Response of Reserve Money to Prices
  - Axis/labels text: -.010-.005
  - Axis/labels text: .0 0 0.0 0 5.0 1 0.0 1 5.0 2 0
  - Period markers: 5 10 15 20 25 30 35
- Response of Reserve Money to Reserve Money
  - Axis/labels text: -.010-.005
  - Axis/labels text: .00 0.00 5.01 0.01 5.02 0
  - Period markers: 5 10 15 20 25 30 35
- Response of Reserve Money to Exchange Rate
  - Axis/labels text: -.02-.01
  - Axis/labels text: .0 0.0 1.0 2.0 3.0 4
  - Period markers: 5 10 15 20 25 30 35

### Responses of Exchange Rate
- Response of Exchange Rate to Output
  - Axis/labels text: -.02-.01
  - Axis/labels text: .00.01.02.03.04
  - Period markers: 5 10 15 20 25 30 35
- Response of Exchange Rate to Prices
  - Axis/labels text: -.02-.01
  - Axis/labels text: .00.01.02.03.04
  - Period markers: 5 10 15 20 25 30 35
- Response of Exchange Rate to Reserve Money
  - Axis/labels text: -.02-.01
  - Axis/labels text: .0 0.0 1.0 2.0 3.0 4
  - Period markers: 5 10 15 20 25 30 35
- Response of Exchange Rate to Exchange Rate
  - Axis/labels text: -.02-.01
  - Axis/labels text: .0 0.0 1.0 2.0 3.0 4
  - Period markers: 5 10 15 20 25 30 35

---

### Appendix II. Impulse Responses, Modified Baseline VAR (including Reserve Money)

### Note
- Reserve money is included instead of the monetary policy stance measure as a robustness check.
- Response to Cholesky one standard deviation innovations ± 2 standard errors

### Responses of Output (Modified VAR)
- Response of Output to Output
  - Axis/labels text: -.004-.002
  - Axis/labels text: .0 0 0.0 0 2.0 0 4.0 0 6.0 0 8.0 1 0
  - Period markers: 5 10 15 20 25 30 35
- Response of Output to Prices
  - Axis/labels text: -.004-.002
  - Axis/labels text: .0 0 0.0 0 2.0 0 4.0 0 6.0 0 8.0 1 0
  - Period markers: 5 10 15 20 25 30 35
- Response of Output to Exchange Rate
  - Axis/labels text: -.004-.002
  - Axis/labels text: .0 0 0.0 0 2.0 0 4.0 0 6.0 0 8.0 1 0
  - Period markers: 5 10 15 20 25 30 35
- Response of Output to MP Measure
  - Axis/labels text: -.004-.002
  - Axis/labels text: .0 0 0.0 0 2.0 0 4.0 0 6.0 0 8.0 1 0
  - Period markers: 5 10 15 20 25 30 35

### Responses of Prices (Modified VAR)
- Response of Prices to Output
  - Axis/labels text: -.016-.012-.008-.004
  - Axis/labels text: .0 0 0.0 0 4.0 0 8.0 1 2
  - Period markers: 5 10 15 20 25 30 35
- Response of Prices to Prices
  - Axis/labels text: -.016-.012-.008-.004
  - Axis/labels text: .0 0 0.0 0 4.0 0 8.0 1 2
  - Period markers: 5 10 15 20 25 30 35
- Response of Prices to Exchange Rate
  - Axis/labels text: -.016-.012-.008-.004
  - Axis/labels text: .0 0 0.0 0 4.0 0 8.0 1 2
  - Period markers: 5 10 15 20 25 30 35
- Response of Prices to MP Measure
  - Axis/labels text: -.016-.012-.008-.004
  - Axis/labels text: .0 0 0.0 0 4.0 0 8.0 1 2
  - Period markers: 5 10 15 20 25 30 35

### Responses of Exchange Rate (Modified VAR)
- Response of Exchange Rate to Output
  - Axis/labels text: -.02-.01
  - Axis/labels text: .00.01.02.03.04
  - Period markers: 5 10 15 20 25 30 35
- Response of Exchange Rate to Prices
  - Axis/labels text: -.02-.01
  - Axis/labels text: .00.01.02.03.04
  - Period markers: 5 10 15 20 25 30 35
- Response of Exchange Rate to Exchange Rate
  - Axis/labels text: -.02-.01
  - Axis/labels text: .00.01.02.03.04
  - Period markers: 5 10 15 20 25 30 35
- Response of Exchange Rate to MP Measure
  - Axis/labels text: -.2-.1
  - Axis/labels text: .0.1.2.3.4.5
  - Period markers: 5 10 15 20 25 30 35

### Responses of MP Measure (Modified VAR)
- Response of MP Measure to Output
  - Axis/labels text: -.2-.1
  - Axis/labels text: .0.1.2.3.4.5
  - Period markers: 5 10 15 20 25 30 35
- Response of MP Measure to Prices
  - Axis/labels text: -.2-.1
  - Axis/labels text: .0.1.2.3.4.5
  - Period markers: 5 10 15 20 25 30 35
- Response of MP Measure to Exchange Rate
  - Axis/labels text: -.2-.1
  - Axis/labels text: .0.1.2.3.4.5
  - Period markers: 5 10 15 20 25 30 35
- Response of MP Measure to MP Measure
  - Axis/labels text: -.2-.1
  - Axis/labels text: .0.1.2.3.4.5
  - Period markers: 5 10 15 20 25 30 35

---

### Appendix III. Impulse Responses, Baseline VAR, Inverted Variable Ordering

### Note on Ordering
- Response to Cholesky one standard deviation innovations ± 2 standard errors
- Endogenous variables ordered as follows: X = (output, prices, exchange rate, monetary policy stance measure)'

*Source: Appendix I–III, _wp07285 - Appendix I. Impulse Responses, Baseline VAR*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2007/_wp07285.pdf_