During most of the period from the early1990s until 2003, keeping monetary value stableness and stabilisation of the exchange rate remained more or less the primary aims of pecuniary policy in Egypt. These aims were seen as indispensable for prolonging appropriate degrees of investing and advancing economic growing. Other on occasion conflicting ends including commanding liquidness growing, raising foreign fight, advancing exports and set uping assurance in the national currency were presumed as secondary policy marks at the clip.
The high rising prices rates that came approximately in the wake of the floatation- officially labeled ( de jure ) a free float-of the Egyptian lb at the terminal of January 2003 prompted the Central Bank of Egypt ( CBE ) to proclaim monetary value stableness and low rising prices rates as the primary and overruling pecuniary policy aim without endangering sustainable high rates of end product growing. By mid-2005, the CBE adopted the nightlong involvement rate on interbank minutess in stead of the extra bank militias as the chief operational instrument. To pull off the involvement rates ( including the nightlong interbank rate ) within the new pecuniary policy government, the CBE established an operational model early in June 2005, known as the corridor system, with a ceiling and a floor for the nightlong involvement rates on loaning from and sedimentations at the CBE, severally. However, an of import issue that as yet remains unsolved concerns the function of the exchange rate in the new pecuniary policy model.[ 1 ]
The argument is still unfastened refering the function of exchange rate in the preparation of pecuniary policy in an unfastened economic system model described by an rising prices aiming and a flexible exchange rate government ( Taylor 2001 ) . Hence, we will briefly reexamine some of the theoretical and/or empirical work done for some developed/developing states to measure whether the cardinal Bankss modify the scenes of their policy instruments ( i.e. short-run involvement rate ) in response to interchange rate fluctuations.
The recent theoretical pecuniary theoretical accounts under openness find no significant function for the exchange rate in policy direction. For illustration, Clarida et Al. ( 2001 ) show that the existent exchange rate is irrelevant to pecuniary determinations owing to the proportionate relation between the footings of trade and the end product spread ( Caputo and Liendo 2005 ) .
The empirical argument is still unfastened, nevertheless, refering its relevancy of exchange rate motions to policy determinations. Clarida et Al ( 1998 ) show utilizing structural appraisal of rising prices prognosis based pecuniary policy regulations for the US, Japan and selected European states that although the exchange rate has a instead little policy consequence in footings of magnitude, its consequence is statistically important. Furthermore, Lubik and Schorfheide ‘s ( 2003 and 2007 ) estimations of the parametric quantities of a generic Taylor-type policy regulation derived from a small-scale little unfastened economic system ( SOE ) dynamic stochastic general equilibrium ( DSGE ) patterning model unwrap that while the Bank of Canada and the Bank of England seem to aim the exchange rate the cardinal Bankss of Australia and New Zealand do non.
Adopting the Lubik and Schorfheide ( 2005 ) micro-founded patterning model, Caputo and Liendo ( 2005 ) provided grounds that the Chilean cardinal bank has non offset misalignments induced by exchange rate fluctuations ; in contrast, following Clarida et Al. ( 1998 ) , Caputo ( 2005 ) derived contradictory grounds for the Central Bank of Chile. The latter grounds is consistent with the major findings of Calvo and Reinhart ( 2002 ) who show-using a sample of 39 developed and developing states over the period January 1970-November 1999-that the pecuniary governments routinely absorb exchange rate fluctuations via involvement rate policy accommodations.
The chief aim of this paper is to measure the extent to which the CBE takes exchange rate motions into consideration in the preparation and execution of pecuniary policy determinations. In add-on, hence, we are able to verify whether “ fright of drifting ” induces the CBE to regularly utilize the involvement rate policy as a agency of smoothing exchange rate fluctuations and to measure the credibleness of its committedness to drifting the lb. We address these issue-given the worlds of the Egyptian economy-by analyzing the behavior of pecuniary policy in a SOE puting characterized by a flexible exchange rate, perfect capital mobility and an rising prices aiming mechanism that is described by a generic Taylor-type involvement reaction map ( Taylor 2001 ) .[ 2 ]The Taylor regulation adjusts the policy instrument in response to rising prices and end product every bit good as exchange rate motions. Within that model, it is possible non merely to prove whether the CBE does aim the exchange rate-i.e. , the function of the exchange rate in the policy rule-but besides to analyze the stochastic behaviour and dealingss between domestic and universe variables, to analyse the policy reactions of the pecuniary authorization to suit alterations in the international variables, e.g. , concern rhythm fluctuations owing to foreign dazes ( Lubik and Schorfheide 2007 ) and to measure the impact of foreign rising prices, end product and footings of trade dazes on involvement rate and other cardinal macroeconomic variables.
We take up the conventionalized New Keynesian DSGE rational outlooks concept framed by Lubik and Schorfheide ( 2003 and 2007 ) for a SOE as a templet to pattern the behavioural kineticss of pecuniary policy and cardinal variables in the Egyptian economic system and to gauge the policy map parametric quantities. The chief advantage of the Lubik and Schorfheide attack is that it allows gauging the parametric quantities of the policy reaction map from within a structural multivariate macroeconomic theoretical account. Hence, it is possible to see the cross-equation limitation effects between the structural-including the policy-parameters and the determination regulations of the different agents in the economic system while keeping the independency of the structural coefficients that characterize the economic system from pecuniary policy ( Lubik and Schorfheide 2007 ; Taylor 2001 ; Caputo and Liendo 2005 ) . We besides follow the Bayesian appraisal process proposed by Lubik and Schorfheide ( 2007 ) .
Our findings show that the CBE adopts an aggressive rising prices aiming policy and reacts to divergences of end product from its expected value. However, our results-using a posterior odds ratio test-suggest that the CBE policy does non aim the exchange rate by on a regular basis reacting to foreign exchange motions. The consequences besides reveal a comparatively important grade of involvement rate smoothing that renders the CBE policy responses unnecessarily transparent. This is in line with old consequences for Egypt every bit good as other states.
The remainder of the paper is structured as follows. Section 2 gives a brief overview of the specification of the SOE structural DSGE macroeconomic theoretical account estimated on informations for Egypt. In subdivision 3, we describe the information and the pick of priors and show the Bayesian estimations for the structural parametric quantities in the theoretical account. Section 4 evaluates the implied behavioural kineticss of the estimated theoretical account, identifies the chief policy drivers and trials whether the CBE considers the exchange rate when explicating pecuniary policy. The hardiness of our findings are discussed in subdivision 5. Section 6 concludes.
2. Overview of the Model
We adopt the now all excessively good known micro-founded DSGE theoretical account proposed by Lubik and Schorfheide ( 2003 2007 ) based on the seminal work of Gali and Monacelli ( 2005 ) . That stylized mold model has been used extensively for policy analysis in assorted developed and developing states including US, Canada, New Zealand, UK, Chile, Portugal, Nigeria and Mozambique ( Lubik and Schorfheide 2007 ; Caputo and Liendo 2005 ; Almeida 2009 ; Richard 2009 ; Peiris and Saxegaard 2007 ) . Consequently, we present merely a brief overview of the theoretical account with the inside informations kept to a lower limit, merely sufficient for the reader to logically follow the consequences.
The theoretical account consists of four individual equations ( 1-4 ) and four autoregressive procedures ( 5-8 ) that drive the development of the exogenic variables. There are eight variables in the theoretical account ; four are endogenous ( end product, nominal involvement rate, rising prices and nominal exchange rate ) . The staying variables-terms of trade, universe end product, engineering growing and universe inflation-are considered exogenic. The theoretical account includes four policy and nine non-policy parametric quantities every bit good as five stochastic dazes ( including the policy daze IµtR ) . the theoretical account equations and the definitions of the different variables, parametric quantities and dazes are portrayed below.
Open economic system IS curve
yt: end product ; Rt: nominal involvement rate ; Iˆt: rising prices rate
qt: footings of trade ; yt* : universe end product ; At: engineering growing
i?? : coefficient of comparative hazard antipathy ; I± : import portion ( degree
of openness )
Et: conditional outlooks operator based on information up until clip T ;
a?† : first difference operator
Open economic system new-Keynesian Philips Curve ( NKPC ) :
E?t: possible end product
i?« : NKPC slope coefficient ; R: steady province existent involvement rate ; I? : price reduction factor
et: nominal exchange rate
Iˆt* : universe rising prices
I?R: involvement rate smoothing parametric quantity ; I?1, I?2, I?3: Taylor regulation rising prices, end product,
exchange rate coefficients
IµtR: pecuniary policy daze
Stochastic daze procedures
I?q, I?y* , I?Iˆ* , I?A: footings of trade, universe end product, universe rising prices and engineering
growing smoothing parametric quantities
Iµtq, Iµty* , IµtIˆ* , IµtA: footings of trade, universe end product, universe rising prices and
engineering growing dazes
The advanced unfastened economic system IS map ( 1 ) links the cyclical development of domestic end product to the expected values of end product and engineering growing. Furthermore, the behavioural kineticss of end product are governed by the ex-ante existent involvement rate and the values of future realisations of alterations in the footings of trade and universe end product, whose influence on domestic production is regulated by the whirls I± and i?? ( import portion and snap of intertemporal permutation, severally ) to account for the inter- and intratemporal ingestion smoothing effects of foreign trade every bit good as comparative hazard antipathy and wont formation.
Equation ( 2 ) is a SOE version of the expectational NKPC that explains the dynamic development of the rising prices rate. It reduces to the closed economic system NKPC with Calvo-type strictly advanced monetary value outlooks when I± peers zero ( Gali and Monacelli 2005 ) . Hence, the rising prices rate is a map of the future realisation of monetary values loaded by the price reduction factor ( i?? ) . The rising prices rate is besides driven by the end product spread ( yt – E?t ) , where, whose consequence on the rising prices rate depends on the incline of the NKPC adjusted by I± and i?? and the divergence of the discounted realisation of the future alteration in footings of trade from its contemporary value whose burden is I± .
The advanced orientation of the NKPC is motivated by the nonstructural nature of indexation to past rising prices rate and the empirical grounds provided from surveies in which the rising prices mark alterations over clip ( Benati 2008 and 2009 ; Cogley and Sbordone 2008 ; Castelnuovo 2009 ) . In add-on, Moursi and El Mossallamy ( 2009 ) show-within a closed economic system setting-that the monetary value indexation coefficient for the instance of Egypt is highly low, diagnostic of limited rising prices continuity.
Assuming PPP, equation ( 3 ) defines the difference between domestic ( Iˆt ) and universe ( Iˆt* ) rising prices rates in footings of the grasp or depreciation of the nominal exchange rate and existent exchange rate fluctuations ( i?„qt ) loaded by 1- I± ( Richard 2009 ) . In add-on, the universe rising prices dazes are treated in the theoretical account as an unobservable ; hence, Lubik and Schorfheide argue that Iˆt* may capture possible divergences from PPP ( Lubik and Schorfheide 2005, 2007 ) .
The CBE pecuniary policy reaction map ( equation 4 ) is approximated in the theoretical account by a Taylor-type regulation described by the short-run involvement rate. CBE reacts in response to a leaden norm of involvement rate inactiveness and the rising prices and end product spreads and exchange rate fluctuations.
The staying four AR ( 1 ) equations ( 5 ) – ( 8 ) , which characterize the exogenic procedures for the footings of trade, universe end product and rising prices and engineering growing subject to relevant inventions, near the theoretical account ( Lubik and Schorfheide 2003 2007 ) .
3. Anterior Selection and Empirical Estimation
In this subdivision we present a description of the discernible variables employed in appraisal and of the choice of priors chosen for standardization. The posterior estimations are presented along with a study of the empirical Bayesian appraisal process and some basic nosologies for measuring the numerical solution algorithm and the statistical consequences and their sensitiveness to priors.
3.1 Data Description
With the exclusion of the footings of trade, the exogenic variables in the model-world end product, rising prices and productiveness growing and possible output-are non-observable ( latent ) . The theoretical account was calibrated utilizing monthly observations-mainly retrieved from the IMF-IFS 2009 CD-ROM-spanning the sample period January 2002-July 2008 on end product, rising prices, the policy instrument, nominal exchange rate and footings of trade alterations.
The end product series was obtained as the month-to-month per centum existent GDP per capita growing ( derived by temporally disaggregating the corresponding one-year existent GDP per capita series ) scaled by a factor of 100.[ 3 ]Inflation was measured as the annualized per centum rate of alteration of monthly CPI. The policy instrument was denoted by the annualized monthly nightlong involvement rates on interbank minutess ( CBE unpublished ) . The nominal exchange rate was defined in Egyptian lbs per dollar and the footings of trade were represented by the per centum alteration in the existent exchange rate ( with mention to the US dollar ) . All the series were tested for seasonality ( none detected ) and demeaned prior to appraisal.
3.2 Prior Choice
The Bayesian processs employed in appraisal require the specification of the anterior chance distributions for the structural parametric quantities. Our pick of priors was based on a-priori information derived from historical grounds ( utilizing pre-sample informations ) and old research. Furthermore, relevant size and mark limitations on the different parametric quantities were imposed by paring the sphere of the distribution and choosing plausible scopes for the priors ( Lubik and Schorfheide 2007 ) . Importantly, at the beginning, a benchmark version of the theoretical account was estimated assuming no limitations on the response of the CBE to interchange rate fluctuations i.e. , policy regulation i??3i‚?0. An alternate restricted version-calibrated assuming i??3=0-was estimated subsequently to prove the hypothesis that the CBE does non respond to interchange rate motions. The distribution denseness and its sphere every bit good as the first and 2nd minutes for the anterior coefficients under the policy regulation i??3i‚?0 are depicted in columns 2-5 of table 1.
The end product, rising prices and exchange rate policy parametric quantities ( i??1, i??2 and i??3, severally ) follow a gamma distribution. Their anterior mean and standard divergence are in line with the values reported in the literature and used in earlier surveies for Egypt ( Lubik and Schorfheide 2003 2007 ; Caputo and Liendo 2005 ; Taylor 2001 ; Moursi and El Mossallamy 2009 ) . It has been shown that pecuniary policy in Egypt implies a high grade of involvement smoothing ( Moursi, El Mossallamy and Zakareya 2007 2008 ; Moursi and El Mossallamy 2009 ) . Consequently, we use a beta distribution for I?R with average 0.8-which is comparatively higher in comparing with correspondent priors proposed in the literature for other countries-and a low standard divergence of 0.2.
Since the theoretical account is parameterized with mention to the ( annualized ) steady province existent involvement rate ( R ) instead than the price reduction factor, we need to put its priors. We choose a gamma distribution for the coefficient R with a anterior mean of 4.2 based on the estimated value of the steady province existent involvement rate reported in Moursi, El Mossallamy and Zakareya ( 2007 2008 ) . The corresponding implied price reduction factor is about 0.996506, which is reasonably standard in the literature. We choose a comparatively broad sphere for R by stipulating a standard divergence of 2.0 to let for a considerable grade of generalization about information on existent involvement rates in Egypt.
Table 1. Anterior Distributions
[ 0, +i‚? )
[ 0, +i‚? )
[ 0, +i‚? )
[ 0,1 ]
[ 0,1 ]
[ 0, +i‚? )
[ 0, +i‚? )
[ 0,1 )
[ 0,1 )
[ 0,1 )
[ 0,1 )
[ 0,1 )
Notes: Table shows prior mean, standard divergence ( Stdev ) and domain scope for
gamma, beta and reverse gamma distributions and the upper and lower bounds of
the support for the unvarying distribution.
Following Moursi and El Mossallamy ( 2009 ) , the value of the anterior mean for the Philips curve incline coefficient ( i?« ) was set below the usual values ( around 0.5 ) reported in the literature for other states ( Lubik and Schorfheide 2003 ; Caputo and Liendo 2005 ) . However, because Moursi and El Mossallamy ‘s anticipation of i?« is unrealistically low ( .01 ) , we centered the anterior mean at 0.2 with a high criterion divergence of 0.1 to be every bit uninformative as possible, allowing the informations free to bespeak its value.
The anterior first minute of i?? was set to 0.3 in line with the mean portion of entire Egyptian imports in GDP during the period 1980-2001 ( IMF-IFS 2009 ) . The standard divergence was set really low at 0.03. Alternatively, in the absence of dependable information on comparative hazard aversion-the cardinal determiner of the intertemporal snap of permutation between foreign and domestic goods-we centre the anterior mean and standard divergence for i?? at values coherent with the literature ( 0.5 and 0.2, severally ) .
All the exogenic dazes, save the pecuniary policy invention ( i??tR ) , evolve harmonizing to an autoregressive procedure AR ( 1 ) . The priors for the autoregressive parametric quantities i??q and i??y* were estimated utilizing arrested development analysis on pre-sample period historical informations. Likewise, the anterior mean and standard divergence for the footings of trade were obtained from an AR ( 1 ) arrested development of the ( demeaned ) monthly alteration in existent exchange rate series during the period 1995-2000. The anterior minutes for foreign end product were besides specified by suiting an AR ( 1 ) for the ratio of foreign to domestic end product series measured as the proportion between Egypt and the US existent GDP growing in per capita term.[ 4 ]The anterior distribution for the remainder of the universe rising prices daze ( i??i?°* ) and the engineering growing smoothing parametric quantity ( I?A ) were borrowed from old empirical surveies ( Lubik and Schorfheide 2003 2007 and Caputo and Liendo 2005 ) . Finally, all the measurement mistakes were assumed to be independently distributed reverse gamma with anterior agencies runing between 1.0-2.0. We opted to be uninformative about the measurement mistakes taking a standard mistake of 4.0.
3.3 Bayesian Estimation Results
The information and priors described above were used to acquire the posterior distribution of the policy and non-policy parametric quantities of the theoretical account for Egypt utilizing Bayesian econometric techniques.[ 5 ]Following Schorfheide ( 2000 ) , An and Schorfheide ( 2005 ) and Lubik and Schorfheide ( 2003 2007 ) , the anterior ( pre-sample ) beliefs about the structural coefficients are first described by a denseness map. Then the observed informations are used to update the likeliness map conditional on the theoretical account parametric quantities and specification. The anterior denseness and the likeliness map together charecterize the posterior denseness and, accordingly, the fringy informations density- conditional on the model-from which the buttocks minutes, the extension mechanisms for the structural dazes ( impulse responses ) , the comparative importance of the structural inventions ( discrepancy decomposition ) and prognosiss can be derived ( Griffoli 2007 ) . Furthermore, the fringy denseness can be used for calculating hypotheses trials about different parametric quantities in the theoretical account ( odds ratio ) e.g. , i??3=0 ( Lubik and Schorfheide 2007 ) . In line with the literature, the likeliness map was generated with the Kalman filter and the ( un-normalized ) posterior denseness was simulated with the Metropolis-Hastings ( MH ) Monte Carlo trying method.
A set of ocular diagnostic trials is utilised to measure the statistical unity of the Bayesian appraisal process. The esthesia of the MH simulations is examined utilizing the univariate Monte Carlo Markov Chain MCMC nosologies charts displayed in figure A1 for each coefficient. The horizontal axis in each chart represents the figure of MH loops and the perpendicular axis measures the parametric quantity minutes. Three steps are reported for the different parametric quantities: an 80 % assurance interval around the parametric quantity mean ( interval ) , the discrepancy ( M2 ) and the 3rd minute ( M3 ) . For each step, the charts portray ruddy and bluish lines mensurating the parametric quantity vectors within and between ironss, severally. The esthesia of the MH algorithm requires that the simulations are similar within and across the ironss. In that instance, both lines should expose small variableness and converge ( Griffoli 2007 ) . The diagram shows- likely with the exclusion of universe end product measuring mistake and to a lesser extent the intertemporal snap of substitution-that the minutes for all the parametric quantities of the theoretical account appear stable and convergent.
The multivariate MCMC nosologies demonstrated in figure A2 are correspondent to the univariate trials. The multivariate trials, nevertheless, represent an aggregative step of stableness and convergence based on the characteristic root of a square matrixs of the discrepancy covariance matrix of the coefficients ( Juillard 2008 ; Griffoli 2007 ) . The multivariate MCMC tests back up the plausibleness of the MH optimizer solution.
Figure A3 juxtaposes the anterior ( Grey ) and posterior ( black ) distributions for the parametric quantities of the theoretical account. The perpendicular green line in each chart identifies the posterior manner from the numerical optimisation simulations. The anterior and posterior distributions are reasonably near. The aforethought posterior distributions do non look to divert well from normalcy and the optimisation manner is rather similar to the posterior manner indicating that both the informations and the selected priors are enlightening about the parametric quantities.
Table 2. Posterior Appraisal Consequences
Policy Rule I?3a‰?0
Notes: Mean and 5th and 95th percentiles for posterior distributions are obtained utilizing the MH sampling algorithm
using 150000 draws with 5 analogue ironss. All the reported estimations are based on the last 75000 draws from
each concatenation. The maximized log-likelihood map is represented by the log of the Laplace estimate and the
Harmonic Mean i.e. , the fringy informations denseness ( Lubik and Schorfheide 2007 ) . Credence shows the norm
acceptation rate in each concatenation. The ideal credence rate is 25 % . The reported acceptation rates across the analogue
ironss, which provide a numerical rating of the public presentation of the MH algorithm are comparatively stable across
To complement the above proof cheques, we plot the subtraction of the posterior denseness for values around the computed manner for each parametric quantity. Successful convergence of the optimizer is obtained when manner is at the lower limit of the subtraction of the posterior distribution ( Griffoli 2007 ) . Figure A4 reveals that the optimizer has non run into numerical trouble.
Finally, figure A5 illustrates the estimated smoothened dazes. The horizontal axis in each secret plan denotes the length of the sample period. Ocular review supports the consistence of the expected way of the dazes with the realized estimations of the inventions, demonstrated by the bunch of the smoothened dazes estimates around zero.
To sum up, different ocular diagnostic trials were conducted for the Bayesian appraisal process. They all show that the theoretical account provides sensible parametric quantity estimations.
The posterior parametric quantity mean and 5th and 95th percentiles estimated by MH sampling-corresponding to the posterior distributions portrayed in figure A3-are reported in table 2. The tabular array besides presents the values of the fringy likeliness calculated with the Laplace estimate and the harmonic mean calculator every bit good as the mean acceptation rate used for measuring the public presentation of the MH algorithm. The statistical indexs found in table 2 support the plausibleness of the MH simulations and the rationality of our anterior distribution picks.
4. Policy Analysis
This subdivision discusses the estimated coefficients and the implied dynamic behaviour derived from the theoretical account. In peculiar, we investigate the policy relevancy of the different parametric quantity estimations to economic world in Egypt, study the response of the economic system to different dazes and prove the reaction of the CBE to interchange rate motions.
4.1 Appraisal Consequences
The posterior Bayesian estimations and assurance intervals of the structural policy response parametric quantities ( I?1, I?2, I?3 and I?R ) for the unrestricted version of the theoretical account ( I?3i‚?0 ) are presented in table 2. Our findings show that the CBE has maintained a strong anti-inflation place with the estimated posterior coefficient of I?1 transcending 2. Alternatively, the estimated value of I?2=0.17 reveals a more indulgent policy concern about end product spread aiming at least in comparing with the policy response to rising prices. The estimation of I?3 ( 0.04 ) is besides declarative of weak response to nominal exchange rate fluctuations. In general, the buttocks means for the three estimated policy coefficients are well different from their false priors bespeaking that they draw on of import information from the information. The posterior estimation of I?R with an norm of 0.74 reveals a really high grade of involvement rate smoothing. This consequence is peculiarly in line with old consequences depicting involvement rate inactiveness in Egypt ( Moursi, El Mossallamy and Zakareya 2007 2008 ; Moursi and El Mossallamy 2009 ) . But table 2 shows that there is n’t much difference between the posterior mean of the smoothing parametric quantity and the anterior implying predomination of a priori strong beliefs about the magnitude of I?R. However, because the posterior distribution is more concentrated,[ 6 ]it is reasonable to interpret that the involvement smoothing coefficient is driven to a sensible extent by the informations.
The openness parametric quantity estimation ( I±=0.20 ) is significantly lower than the ascertained portion of imports in GDP. Lubik and Schorfheide ( 2007 ) argue that such a state of affairs can originate as the appraisal process efforts to choose a value for I± that reconciles the volatilities in the footings of trade and rising prices related by the rising prices equation ( 3 ) while obeying the cross-coefficient limitations imposed via the IS and NKPC equations 1 and 2, severally. In contrast, the long-run existent involvement rate r-estimated at 4.19 percent-seems mostly driven by the anterior. This makes sense for the Egyptian economic system where the existent involvement rate in many cases turns negative so that the behavioural premises underlying the intetemporal picks of manufacturers and consumers implied by the theoretical account become inconsistent with the informations. To do sense, the appraisal solution is coerced, hence, to take a value for R in the vicinity of the positive prior.
The buttockss for the incline of the Phillips curve ( i?« ) and the coefficient of comparative hazard antipathy ( i?? ) appear rather divergent from the false priors. The norm of i?« ( estimated at 0.51 ) is markedly higher than the anterior mean ; this estimation is in line with old surveies for Egypt and, furthermore, is consistent with the grounds for other states found in both closed and unfastened economic system surveies ( Lubik and Schorfheide 2005 ) . The snap of intertemporal permutation is around i?? -1=3.23. This value is perceptibly higher than the correspondent estimations for the UK, Canada, Australia and New Zealand ( Lubik and Schorfheide 2005 ) and a shade above the value reported by Caputo and Liendo ( 2005 ) for Chile.
Table 3. Discrepancy Decomposition
Footings of Trade
The discrepancy decomposition estimates-portrayed in table 3-are utile to estimate the importance of each of the five structural dazes for the fluctuations each of the endogenous variables. The tabular array underscores the exposure of the economic system to universe end product dazes. Domestic end product, CPI rising prices and the nominal involvement rate volatility are chiefly driven ( 80.4, 64.3 and 86.6 per centum, severally ) by foreign end product inventions.[ 7 ]The part of pecuniary policy dazes to GDP ( 8.5 per centum ) is in line with the literature. However, while pecuniary policy besides explains a considerable portion ( 16.6 per centum ) of the discrepancy in rising prices, it is accountable for merely 0.20 per centum of the exchange rate fluctuations. Alternatively, the nominal exchange rate motions are dominated by foreign monetary value dazes that account for over 78 per centum of its variableness. Real exchange rate dazes in Egypt have a strong consequence on rising prices transcending 15 per centum. The portion of discrepancy in the policy variable and in domestic end product induced by motions in the footings of trade is little. Similarly, proficient inventions do non hold a significant function in driving each of the four endogenous variables.
Table 2 besides reports the continuity and volatility posterior estimations for the structural inventions. The estimations for all the stochastic dazes, save engineering growing, unwrap moderate continuity. The autocorrelation coefficients for import monetary value rising prices and footings of trade volatility are 0.43 and 0.30, severally. It is sensible to presume that in an unfastened economic system, there are more channels for the extension of daze urges, therefore cut downing the grade of continuity. The continuity of foreign end product dazes is significantly high averaging at 0.79. Yet none of these procedures exhibit inordinate continuity to justify concerns about the presence of a unit root since the posterior distribution ‘s 95th percentile for each of those dazes is less than 0.93 ( Almeida 2009 ) . On the other manus, the information supported by the informations show that engineering growing exhibits an highly low degree of continuity ( 0.07 ) . The estimated posterior mean volatility of engineering dazes measured by I?A ( 5.78 ) is markedly high both in absolute and comparative footings as suggested by the anterior. Egypt ‘s increasing dependence on foreign markets for high tech trade goods, unequal human capital and labour accomplishments and limited spillover effects associated with foreign direct investing ( possibly with the exclusion of selected services sectors e.g. , touristry, fiscal, banking, etc. ) are likely major subscribers to the sporadic growing of proficient advancement peculiarly in the industrial sector. The highest grade of volatility is associated with foreign rising prices dazes ( 8.40 ) . Its consequence is restrained by a moderate grade of continuity ( I?Iˆ*=.43 ) that limits the exposure of the Egyptian economic system to comparative foreign monetary value dazes.
4.2 Impact of Policy Shocks
The posterior impulse response maps ( IRFs ) illustrated in figure 1 show the dynamic consequence of positive exogenic dazes on the discernible endogenous variables in the theoretical account. The diagram reveals that all variables returned to their steady-state value after the daze therefore corroborating the stableness of the solution of the theoretical account.
In general, the IRFs do non look to connote perplexing responses. Adopting a tight pecuniary policy consequences in the contraction of end product and take downing the rate of CPI rising prices that in bend leads to an grasp of the domestic currency. The lessening in money supply associated with the contractionary pecuniary policy raises the nominal involvement rate as shown in figure 1.
An betterment in the footings of trade increases the degree of domestic end product, which provokes an grasp of the Egyptian lb. The grasp induces a decrease in the degree of rising prices to suit the betterments in the footings of trade and the GDP. It besides induces the pecuniary authorization ( CBE ) to take down the involvement rate. This expansionary pecuniary policy reinforces the end product growing.
A stationary productiveness sweetening lowers the fringy costs of production for domestic goods taking to a lessening in the degree of domestic rising prices. The lower domestic monetary values spell higher comparative monetary values of imported and foreign goods. Consequently, for a given value of comparative hazard antipathy, the domestic currency appreciates to queer a lessening in the penchant parametric quantity ( below the estimated I±=20 per centum ) and inordinate exportation owing to higher comparative monetary values of foreign goods.[ 8 ]The grasp of the lb brings on an extra supply of end product that the pecuniary authorization tends to absorb by loosen uping policy therefore reenforcing the expansionary consequence of the technological invention on the economic system with farther decrease of domestic monetary values. We have seen in the old subdivision, nevertheless, that engineering dazes exhibit low degrees of continuity and explicate small of the variableness in the endogenous variables in the theoretical account. The IRFs for the engineering dazes in figure 1 reflect these findings. The addition in universe end product has an inauspicious consequence on domestic production that intensifies inflationary force per unit area.[ 9 ]Following the statement in the old paragraph, the higher domestic monetary values lead to a depreciation of the exchange rate to suit the 20 per centum import portion. The pecuniary authorization raises the involvement in response to the depreciation therefore reenforcing the contractionary consequence on the economic system. With similar logical thinking, a universe comparative monetary value daze triggers an grasp of the Egyptian lb and an addition in domestic monetary values. As the lb appreciates, the cardinal bank lowers its nominal rates therefore reenforcing the expansionary consequence of the daze on end product.
Figure 1. Impulse Responses Posterior Means ( – ) and 90 % Posterior
Assurance Intervals ( — ) to One Standard Deviation Shocks
4.3 Policy Response to Exchange Rate Motions
To prove whether the CBE marks exchange rate, the theoretical account is re-estimated while shuting the foreign policy transmittal channel-i.e. , policy regulation I?3=0-and maintaining the same value of all the other priors. The diagnostic tests-for the maximal likelihood process and the MH trying solver-for the restricted theoretical account are satisfactory and rather similar to the comparable trials of the benchmark appraisal. Table 2 studies the alternate posterior mean and chance interval for the different parametric quantities. Probably with the exclusion policy parametric quantities, and the incline of the Phillips curve and grade of comparative hazard antipathy coefficients, the posterior estimations under the two policy regulations ( I?3i‚?0 and I?3=0 ) are besides qualitatively similar. The alternate policy regulation suggests less aggressive end product aiming. This consequence is in line with the estimations obtained by Moursi and El Mossallamy ( 2009 ) for the Egyptian economic system ( I?1=1.30 ) within a closed economic system puting. The posterior mean of the involvement smoothing parametric quantity is smaller ( 0.59 ) but still declarative of considerable policy transparence. The incline of the NKPC additions from 0.51 to 0.74 under the alternate policy regulation and this is accompanied by a bead in the intertemporal snap of permutation from 3.03 to about 2.63. In general, the discrepancy decomposition estimations do non differ between the unrestricted and the restricted versions of the theoretical account. However, the absence of the exchange rate policy channel somewhat increases the impact of existent exchange rate inventions on domestic rising prices and to a lesser extent on GDP. Finally, the IRFs of the two theoretical accounts are similar except for the foreign rising prices daze where the invention in i?°* is wholly absorbed by nominal exchange rate go forthing all the values all the other endogenous variables integral ( Lubik and Schorfheide 2007 ) .
Table 4. Posterior Odds Test for Exchange Rate Targeting
Log Marginal Data Densities
H0: N±3=0 vs. H1: N±3a‰?0
Table 4 studies the posterior odds ratio for proving the void N±3=0 against the alternate the hypothesis N±3a‰?0 ( Lubik and Schorfheide 2005 2007 ) . As shown in the tabular array, the fringy informations denseness of the unrestricted theoretical account is 13.66 smaller on a log graduated table connoting an inordinate odds ratio. Consequently, we unambiguously reject sensible opportunities that the CBE policy responds to interchange rate fluctuations in the exchange rate. This consequence does non prevent the uncorroborated allegations-which on occasion surface in Egypt within academic community concern circles-that the CBE intervenes in the market for foreign exchange. However, if such an allegation were true, our consequences confirm that such intercession could non be from within the exchange rate policy aiming model.
We adopt Lubik and Schorfheide little unfastened economic system DSGE patterning model to analyze the policy response of the CBE to interchange rate fluctuations. Although Egypt does non follow a formal policy aiming scheme we believe that the pecuniary behavior of the CBE can be approximated by a Taylor-type targeting map. Our findings show that the CBE has late adopted a strong anti-inflationary policy and has besides resorted to end product aiming. Conversely, our trials show that the CBE has non been responding consistently to dazes in the market for foreign exchange via policy accommodations. Besides, the CBE resorts to considerable involvement rate smoothing. This consequences is in line with earlier grounds obtained for Egypt every bit good as for other states.
The theoretical account estimates the part of policy, footings of trade, proficient growing, universe end product and import monetary value alterations on cardinal policy and non-policy variables in the economic system. Apparently, universe end product dazes contribute significantly to the fluctuation in GDP, domestic rising prices and involvement rates. Real exchange rate dazes have a ample effects on alterations in domestic monetary values. But engineering dazes do non act upon alterations in end product, rising prices, involvement rate and nominal exchange rates. We have besides examined the impact of the different dazes on the macroeconomic variables by agencies of impulse response maps. The estimated impulse responses are in line with the literature and do non bespeak mystifiers. Relaxing pecuniary policy has an expansionary consequence and reduces inflationary force per unit areas. Similarly, technological growing consequences in higher degrees of end product and an grasp of the Egyptian lb which lead to a decrease in involvement rate thereby reenforcing GDP growing. Despite that combined consequence, our findings show that the expansionary impact of technological advancement in Egypt is instead limited. World end product and comparative foreign monetary value dazes have a strong consequence on the economic system. A positive foreign end product daze lowers domestic production and deprecate the Egyptian lb taking to higher involvement rate and rising prices. In contrast, a rise in import monetary values raises end product, appreciates the currency, lower involvement rate and increases domestic monetary values.