Evaluations of different monetary policy regimes for a small developing open economy

Abstract

Motivated by recent experiences in Iran, this paper incorporates structural characteristics that reinforce the reluctance of central banks to commit to free floating exchange rates (‘fear of floating’) into optimal policy formulation to determine the extent of exchange rate involvement in monetary policy. We utilize a small stylized model of a small developing open economy calibrated for the Iranian economy to compare the performance of alternative policy regimes, namely flexible domestic inflation targeting, flexible consumer price index inflation targeting, and real exchange targeting in handling higher openness, higher exchange rate pass through, and financial vulnerability. Evaluation of the above alternative policy regimes and relative stability of key macroeconomic variables are conducted through an optimal Ramsey policy approach. The results suggest that, in financially vulnerable small open developing economies, consumer price IT may not be the optimal policy and giving a higher positive weight to the real exchange rate in the central bank’s loss function and allowing policy reaction to this variable via the instrument rate can result in a lower social loss and more stability for the key macroeconomic variables, and hence a superior policy regime. Policy evaluation results based both on stabilization and welfare measures obtained in this paper imply that for the developing commodity (oil) exporting economies with high degrees of financial vulnerability and relatively high pass-through rates, an IT policy framework in which the real exchange rate is also targeted is a superior regime.

Keywords:

• inflation targeting policy
• exchange rate pass through
• financial vulnerability
• fear of floating

JEL Classification:

• E52
• F31
• F41
• O24

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. In the absence of fiscal dominance the CB has the choice of not buying extra oil revenues, thus limiting the institutional link between fiscal operations and CB balance sheet.

2. In this setting we assume the government continues its pro-cyclical fiscal policy.

3. In Fleming–Mundell type models price of imported goods are in the currency of the producer (exporting) country, the exchange rate pass through to price of imports is complete, and the law of one price for tradable applies. Devaluation or depreciation of a currency results in proportional increase in import prices. The ensuing relative price increase switches expenditures from imports towards home-made goods. To the extent that the Marshall–Lerner condition is met, currency depreciation has stabilizing properties. For More details, see Duarte and Obstfeld (2008).

4. For a different interpretation regarding the results of this model see Duarte and Obstfeld (2008), Devereux and Engel (2003).

5. For more details see Céspedes, Chang, and Velasco (2004).

6. According to Backus and Smith (1993) complete international risk sharing can be shown in the following formula:where and are domestic and foreign consumption respectively, and stands for real exchange rate. In the market structure where representative households has the possibility to trade an contingent domestic and foreign bonds denominated in units of home currency, complete risk sharing reads as:where stand for the aggregate price level in domestic and foreign economies and is the nominal exchange rate. Considering the above Euler relationship for domestic and foreign bonds in this structure, we have the following uncovered interest rate parity relationship: and stands for domestic and foreign nominal interest rate on bonds. An indeterminacy problem is associated with this UIP form when the domestic and foreign bond are perfect substitutes and have equal yields (complete international risk sharing). A stationarity problem arises in the sense that a determinate asset allocation cannot be obtained. This would prevent a proper analysis of small deviations around a deterministic steady state. One solution is to make the return on the two securities unequal. In order to break the monotonic relationship between the real exchange rate and relative consumptions (mirroring UIP), UIP relationship can be expressed in terms of a time varying risk premium so as to allow deviations from the above form. This risk premium can be endogenized by different methods, e.g. balance sheet effect which changes the UIP relationship as follows:where is a time varying risk premium function. Equation (5) in our model is based on this approach. In our paper the risk premium is dependent on (or endogenized) by the real exchange rate to capture the balance sheet effect as in Equation (6).

7. Several studies reject the hypothesis that exchange rate expectations are rational. For instance, see Cavaglia, Verschoor, and Wolff (1993), Ito (1998), Leitemo and Söderström (2005).

8. Benigno and Thoenissen (2008)

9. An instance is any change in the value of letters of credit issued by the central bank.

10. Cavoli (2009).

11. A ‘sudden stop' type shock can result in a very large increase in the risk premium.

12. As discussed by Mehrara and Oskoui (2007) the main source of business cycle fluctuations is due to terms of trade shocks. As indicated by (15) the terms of trade fluctuations influence the risk premium.

13. See Khan, King, and Wolman (2003), Levin et al. (2006), and Schmitt-Grohé and Uribe (2007) for detailed discussions in New Keynesian models.

14. The authors then evaluate the performance of alternative simple policy rules relative to this benchmark.

15. Kam et al. (2009) raises three arguments in defense of this type of ad-hoc loss function. (1) It might be impossible to map a second-order approximation of the social welfare maximizing CB loss function based on household preferences. (2) Empirical studies indicate that the above loss function includes the goal variables of many developed and developing commodity exporting economies. (3) Finally, the above loss function covers the alternative monetary policy literature which that strives to assess the usefulness of alternative monetary policy rules using quadratic loss functions. For more details see Kam et al. (2009).

16. For more details, see Schmitt-Grohé and Uribe (2007).

17. Note that in (24) real exchange rate does not enter directly or indirectly because our model does not incorporate differentiation of intermediate and final goods as well as imported and domestic intermediate goods.

18. Cavoli (2008) examined the monetary policy rules derived from the above mentioned policy regimes. He shows that the underlying policy rules for FDIT and FCIT react to real exchange rate even though real exchange rate does not appear in the loss function. For RERT real exchange rate appears in the loss function and the underlying policy rule reacts to it directly and indirectly. His study shows that reaction feedback coefficient to real exchange rate in the FDIT regime is the lowest and is the highest for RERT. It must be mentioned that reaction to real exchange rate in FDIT and FCIT monetary policy rules are to be expected, because real exchange rate contains information that will help policy-maker in achieving the inflation target. In other words, the effect of the exchange rate movements filters through domestic inflation and also CPI inflation.

19. These parameters, along with other parameters of the model, are estimated with Bayesian techniques.

20. The parameters for Indonesia, Philippines, and South Korea see Cavoli (2009).

21. Along with CPI and output gap.