چارچوب اقتصادسنجی کلان برای ارزیابی سیاست های پولی: مطالعه موردی از پاکستان

This paper attempts to establish the quantitative importance of the various channels of monetary transmission by constructing, estimating and simulating a small macroeconometric model of Pakistan's monetary sector, while using data from the monetary statistics and the monetary survey of the State Bank of Pakistan over 1976–2007. The paper elucidates that the key feature of the study of monetary policy in Pakistan has been preoccupied with neglect either of the demand or the supply function of money and shows how this may lead to imprecise policy actions and mistaken conclusions. Accordingly, we delineate the transmission mechanism of monetary policy by taking into consideration all structural money demand and money supply linkages along with the historically implied identifying assumption in the framework of a marginalized macroeconometric model. The within-sample and out-of-sample evaluations of the model are found satisfactory. The paper presents results of three policy simulations from the estimated model that highlight the impact of alternative monetary policy instruments on the monetary variables under a rule-based and a discretionary policy environment. We find that (i) the SBP subscribes to an unannounced monetary policy rule, (ii) the determination of the policy rate under the announced rule environment stabilizes the monetary sector in that convergence to full equilibrium is smooth and rapid, (iii) a 100 bps reduction in the discount rate, ceteris paribus, decreases money supply by 4.97%, and (iv) the long term implication of reducing (increasing) the reserve requirement ratio on time (demand) deposits, ceteris paribus, is only higher inflation. Finally, we establish that a 100 bps increase in interest rate increases money supply by 3.14% in full equilibrium.

Pakistan economy has recently experienced a steady fall in GDP growth, a deterioration in its current account balance and balance of payments, a fall in the currency exchange rate, a rise in the fiscal deficit, an unprecedented rise in inflation, a rise in the profits of the banking firms underscored by a sizeable divergence between the lending and borrowing rates and a fall in the stock of foreign exchange reserves (State Bank of Pakistan, 2007a, State Bank of Pakistan, 2008a and Government of Pakistan, 2008). The State Bank of Pakistan (henceforth referred as SBP) has responded to these macroeconomic imbalances with a steady increase in the policy interest rates, an increase (decrease) in the reserve requirements against time (demand) deposits and an overall increase in the statutory reserve requirements. In addition, SBP has also endeavored to place constraints on public sector borrowing for budgetary support. Is this the correct policy response? In fact, and as we show in this paper, these policies have only intensified macroeconomic pressures and inflation has sharply increased over the period 2007–2009.2 While many researchers, and especially those at the SBP, believe that monetary policy has no role in generating these outcomes (see State Bank of Pakistan, 2007b, State Bank of Pakistan, 2007c and State Bank of Pakistan, 2008b), the role of SBP in not delivering upon its fundamental promise of keeping inflation low and stable cannot be disregarded. Given SBP's commitment to maintaining low and stable inflation (State Bank of Pakistan, 2008b), we identify that it is the incorrect belief of the SBP about the interest elasticity of money supply3 that inhibits SBP to know the exact effects of its actions on monetary equilibrium. This wrong belief, we conjecture, stems from the absence of the correct identifying assumption about monetary policy4 and from the absence of use of a detailed macroeconometric model that spells out the transmission mechanism of monetary policy. The SBP reckons that it is targeting nominal money supply growth using interest rates as the key policy instrument (State Bank of Pakistan, 2008b). However, regardless of the corroborated significance of the credit channel in monetary transmission (State Bank of Pakistan, 2001, State Bank of Pakistan, 2008b and Agha et al., 2005), or the endogeneity of money supply (Ahmad and Ahmed, 2006 and Omer and Saqib, 2008), the SBP continues to remain ignorant of the effect of interest rates on the supply of money. The SBP predominantly relies on planned rates of growth of net credits to public and private sectors (while ignoring their interdependence, or their dependence on the interest rates and the level of economic activity) along with only an estimated money demand equation to derive its target rate of growth of the money supply (State Bank of Pakistan, 2008b). This approach not only happens to be quite naïve, but inevitably implies the incorrect identification of monetary policy. The ignorance of the understanding as to what drives monetary policy coupled with the lack of use of a detailed macroeconometric model thus makes monetary policy stand in a state of chaos. Accordingly, we set the objective of this paper to establish the quantitative importance of the various channels of monetary transmission in Pakistan by estimating and simulating a small macroeconometric model of Pakistan's monetary sector, while explicitly incorporating an identifying assumption for monetary policy. The model seeks to determine the short term rate of interest as an equilibrium process generated through the interaction of money demand and money supply. It also establishes a precise estimate of the interest elasticity of money supply. The rest of the paper is organized as follows: Section 2 provides a brief review of the theoretical issues in the study of monetary policy and evaluates the study of monetary policy in Pakistan. In Section 3, we discuss our methodology and model specification. This is followed by results of estimation and simulation exercises in Section 4. The paper concludes with a discussion of the results and policy implications in Section 5.

This paper establishes the quantitative importance of the various channels of monetary transmission in Pakistan by constructing, estimating and simulating a small macroeconometric model of Pakistan's monetary sector that may be used as a prototype for designing and/or evaluating monetary policy responses in the face of macroeconomic developments. Our monetary sector model estimates the money demand and money supply functions using all structural linkages thereby determining the short term rate of interest as an endogenous equilibrium process. We conjecture that while the demand for monetary assets may be known through monetary statistics, changes in money supply get accounted for in the monetary survey. Finally, in order to trace out the effect of central bank policy distinctively from other macroeconomic developments, we use an empirically motivated Taylor-type policy rule equation as our identifying assumption. We find that the policy interest rate is very closely related to the macroeconomic variables indicating that the State Bank of Pakistan is in fact subscribing to some unannounced Taylor type monetary policy rule. One important difference of the SBP policy rule with the standard Taylor rule equation is in the sign of the coefficient on the growth rate of output. The SBP rule tells us that the State Bank of Pakistan cuts the interest rate (by about 48 bps) when output increases by 1%. This is contrary to what central bankers practice all over the world, and we contend that this reflects the confusion of the SBP policymakers about the design of monetary policy. Nevertheless, we make use of this rule for carrying out policy simulations. The within-sample and out-of-sample evaluations of the model are found satisfactory. The static and dynamic forecasts of all endogenous variables consistently track their paths and the RMPSE indicates a very close fit of the model. Dynamic forecasts establish the stability and credibility of the model in that they reflect how might endogenous variables have behaved had this model been constructed in the year 2000 and then used for policy simulations with the paths of all exogenous variables known with certainty up to year 2007. The model can thus be convincingly used for policy simulations. The paper presents the results of three such policy simulations that highlight the impact of alternative monetary policy instruments on the monetary variables under a rule-based and a discretionary policy environment. The simulations show that (i) A 100 bps reduction in the discount rate, ceteris paribus, decreases money supply by 4.97%, (ii) The long term implication of reducing (increasing) the reserve requirement ratio on time (demand) deposits, ceteris paribus, is only higher inflation, and that (iii) a 100 bps increase in interest rate increases money supply by 3.14% in full equilibrium.