A. Introduction

1. The Central Bank of Paraguay (BCP) has a clear legal mandate to preserve the stability of prices and the financial system. The BCP’s policy monetary framework includes instruments that are in line with conventional practices, such as open-market operations with sterilization bills (Instrumentos de Regulation Monetaria, IRM), reserve requirements that vary by type of deposit and currency, a short-term collateralized liquidity facility,² and, sporadically, foreign exchange intervention. Under this framework, the BCP seeks to influence liquidity conditions through operations with IRM. It also maintains a flexible exchange regime, relying on foreign exchange intervention to smooth seasonal and erratic movements in the exchange rate. Moreover, as part of the transition to an inflation targeting (IT) framework, the BCP has announced for the past few years an annual inflation objective of 5 percent with a range of +/- 2.5 percent.

2. However, the BCP may need to adapt its monetary policy framework as it moves towards a full IT framework. One key area of reform is to strengthen the effectiveness of the policy interest rate to provide a clear signal of the stance of monetary policy. To that end the BCP should consider moving away from current methods to influence liquidity conditions, which can muddy perceptions of the direction of monetary policy. The BCP transacts in IRM through auctions in which it tailors both the quantity and maturity to the perceived needs of banks, with IRM at maturities ranging from 35 to 546 days, creating a yield curve of policy interest rates. Also the BCP has in the past conducted large-scale spot purchases of foreign exchange to conduct monetary policy. In 2009, it purchased US$1 billion of reserves, when injecting liquidity in the banking system through purchases of IRM proved to be especially challenging. This approach can confuse the market by signaling—perhaps erroneously—that the BCP also has an objective for the exchange rate.

3. This paper analyzes the effectiveness of some of the different policy instruments and offers several recommendations to enhance the role of the policy interest rate. Section B analyzes the effect of different instruments of monetary policy with a standard VAR analysis. Section C discusses several recommendations to enhance the effectiveness of the policy interest rate. Section D offers concluding remarks.

B. Evidence on the Transmission Mechanisms of Monetary Policy

4. The transmission mechanism of monetary policy in Paraguay is analyzed using a VAR model. In particular, the following structure for the VAR is assumed:

where A(L) and B(L) are a n x n and a n x k polynomial matrix in the lag operator L, respectively, Y_t is a n x 1 vector of endogenous variables and X_t is a k x 1 vector of exogenous variables. The latter vector is included to control for exogenous disturbances that may affect the dynamics of the model.³ Finally, U_t is a n x 1 vector of (possibly correlated) estimated residuals. The vectors Y_t and X_t include the following variables

where R_t is the policy rate, represented as the average interest rate on outstanding BCP’s sterilization bills, the IRM; IP_t is the industrial production index; P_t is the consumer price index; M_t is money in circulation; E_t is the nominal bilateral exchange rate against the U.S. dollar; FF_t is the U.S. Federal Reserve fund rate; WCPI_t is the world commodity price index and finally ${IP}_t^{bra}$ is the industrial production index of Brazil. Structural shocks are identified using a Cholesky decomposition, with the variables ordered as in vector Y_t. This implies that R_t is contemporaneously affected only by its own shock whereas E_t, the most endogenous variable, is affected on impact by all structural innovations. All variables in the VAR are in logs and seasonally adjusted, with the only exception of interest rates, and the model is estimated in levels.⁴ The sample period is 2003M1 to 2009M12, and the model assumes 2 lags.⁵

5. The key results include:⁶

• The policy interest rate does not yet appear to be the most effective policy instrument. The impulse response functions (Appendix 1) for a 100 basis point increase in the policy interest rate leads to small and insignificant effects on output and prices. However, results also suggest that changes in the policy rate have a small but statistically significant effect on bank deposit and lending rates.⁷

• Changes in the quantity of central bank sterilization bills and net international reserves seem to be more effective instruments. Alternative VARs are estimated with the policy interest rate replaced by either the stock of BCP’s sterilization bills (IRM), or by the stock of net international reserves (NIR). The impulse response functions for the stock of IRM suggest that an increase in the stock of IRM (a contractionary monetary policy) leads to a decline in both the industrial production index and the price level, although the statistical significance of the responses is limited. The impulse response functions also suggest that an increase in the stock of NIR has an expansionary effect, raising industrial production as well as the price level.

• Evidence also suggests that the policy interest rate has little effect on expected inflation (Box 1). According to the analysis in this paper, the market adjusts its expectations for inflation in light of the past difference between expected and actual inflation, the deviation of the exchange rate from trend, and inflation in world commodity prices. Changes in the policy interest rate have no significant effect on expected inflation.

• The exchange rate pass-through is high from a regional perspective. A slightly modified version of the VAR model is estimated for Paraguay as well as Bolivia, Chile, Peru and Uruguay.⁸ Using the impulse response functions for the price level and the exchange rate, derived from a shock to the exchange rate, we could compute a “proxy” for the exchange rate pass-through. That proxy is estimated as the cumulative impulse response of Pt relative to that of E(NC/$)t, according to the following formula:⁹

  $\underset{t=1}{\overset{j}{\mathrm{\Sigma }}}{P}_t/\underset{t=1}{\overset{j}{\mathrm{\Sigma }}}E(NC/\${)}_t$

  where j is the relevant horizon at which the pass-through is computed, and j=2,6 and 12 months. Results in Table 1 indicate that Paraguay’s short-run pass-through is relatively large by regional standards, although it remains well below one. Also, the pass-through in Paraguay is more important at longer horizons. This probably reflects the high degree of openness of the economy, as well as the high degree of dollarization.

Box 1.

What is Driving Inflation Expectations in Paraguay?

To examine the main drivers of inflation expectations in Paraguay, we estimate a regression of the monthly expected inflation rate ${(\mathrm{\Pi }}_t^e)$ as a function of the information set available to economic agents in the previous month. We assume that the agents’ information set in t-1 includes their forecast error in t-1, (i.e. the deviation between the expected inflation rate ${{(\mathrm{\Pi }}_{t-1}}^e)$ and the actual level of inflation (Π_t-1) in t-1), the change in the policy rate (r_t-1 - r_t-2), the output gap $\left(y_{t-1},y_{t-1}^{\mathrm{potential}}\right)$, the deviation of the bilateral real exchange rate against the U.S. dollar from its trend $\mathrm{(}{\mathrm{e}}_{\mathrm{t}\mathrm{-}\mathrm{1}}\mathrm{-}{{\mathrm{e}}_{\mathrm{t}\mathrm{-}\mathrm{1}}}^{\mathrm{t}\mathrm{r}\mathrm{e}\mathrm{n}\mathrm{d}}\mathrm{)}$, and the world commodity price index (wcpi_t-1). The regression is estimated by transforming all variables into first order differences. Econometric results suggest that in Paraguay the policy rate has not any influence in determining inflation expectations. In addition, the negative and significant coefficient of $\mathrm{(}{{\mathrm{\Pi }}_{\mathrm{t}\mathrm{-}\mathrm{1}}}^{\mathrm{e}}\mathrm{-}{\mathrm{\Pi }}_{\mathrm{t}\mathrm{-}\mathrm{1}}\mathrm{)}$ suggests that economic agents learn from their forecast errors, and reduce their inflation expectations in t when the forecast error in t-1 is positive. Moreover, empirical results suggest that the exchange rate depreciation, measured as the positive deviation of the exchange rate e_t with respect to its trend value, and the world commodity price index, wcpi_t-1, are important factors in driving inflation expectations in Paraguay.

Determinants of Inflation Expectations in Paraguay ^1/

Source: IMF staff estimates.

^1/t-statistics in parenthesis. All variables are transformed into first-order differences.

Determinants of Inflation Expectations in Paraguay ^1/

	${{\mathrm{\Pi }}^e}_t$
Constant	-0.02
	(-1.38)
$({{\mathrm{\Pi }}^e}_{t-1}-{\mathrm{\Pi }}_{t-1})$	-0.55
	(4.35)
(rt-1–rt-2)	-0.01
	(-0.83)
$(y_{t-1}-{y_{t-1}}^{potential})$	0.07
	(0.21)
$(e_{t-1}-{e_{t-1}}^{trend})$	0.57
	(1.61)
wcpit-1	0.37
	(2.15)
Number of Observations	43
Adjusted R Squared	0.42
Durbin Watson	1.87

Source: IMF staff estimates.

^1/t-statistics in parenthesis. All variables are transformed into first-order differences.

View Table

Determinants of Inflation Expectations in Paraguay ^1/

	${{\mathrm{\Pi }}^e}_t$
Constant	-0.02
	(-1.38)
$({{\mathrm{\Pi }}^e}_{t-1}-{\mathrm{\Pi }}_{t-1})$	-0.55
	(4.35)
(rt-1–rt-2)	-0.01
	(-0.83)
$(y_{t-1}-{y_{t-1}}^{potential})$	0.07
	(0.21)
$(e_{t-1}-{e_{t-1}}^{trend})$	0.57
	(1.61)
wcpit-1	0.37
	(2.15)
Number of Observations	43
Adjusted R Squared	0.42
Durbin Watson	1.87

Source: IMF staff estimates.

^1/t-statistics in parenthesis. All variables are transformed into first-order differences.

View Full Size

Paraguay

Citation: IMF Staff Country Reports 2010, 170; 10.5089/9781455207534.002.A003

• Download Figure
• Download figure as PowerPoint slide

Table 1.

Exchange Rate Pass-Through

(In percent)

Table 1.

Exchange Rate Pass-Through

(In percent)

	Bolivia	Chile	Paraguay	Peru	Uruguay
2-months	2.0	1.3	5.0	1.9	3.5
6-months	14.2	5.2	13.8	13.8	17.2
12-months	18.6	10.6	21.9	18.4	17.2

View Table

Table 1.

Exchange Rate Pass-Through

(In percent)

	Bolivia	Chile	Paraguay	Peru	Uruguay
2-months	2.0	1.3	5.0	1.9	3.5
6-months	14.2	5.2	13.8	13.8	17.2
12-months	18.6	10.6	21.9	18.4	17.2

C. Enhancing the Effectiveness of the Policy Interest Rate

6. The results suggest that the BCP needs to strengthen the effectiveness of the policy interest rate as the main instrument of monetary policy. Efforts could focus on several areas:

• Continue to develop domestic financial markets. Currently domestic money markets are thin, with very little secondary market trading of government securities. Also the banking system has a relatively high level of structural liquidity, especially when compared to similar countries. This excess liquidity reflects concerns about counterparty risk in the interbank market as well as the short average maturity of bank deposits (over 70 percent are sight deposits). These constraints on the domestic money market limit the scope to conduct monetary policy through repurchase operations in which the central bank sets the interest rate and the market determines the volume. Another factor that undermines the credit channel is Paraguay’s low degree of financial intermediation, under which personal savings, retained earnings and the use of informal credit markets have become the key sources to finance consumption and investment.

• Continue to strengthen the BCP’s operational framework. The BCP could continue to enhance its capacity to monitor and control liquidity, partly by promoting better information sharing with the ministry of finance and the cooperatives. The BCP is considering the introduction of auctions of IRM that would allow for market-based open market operations. This would represent an important improvement from the current practice of tailoring the quantity and maturity of IRM operations to the perceived needs of banks. It would be important to reduce the current high concentration of open-market operations in IRM of one year of maturity and beyond. This would allow open-market operations to focus on determining only very short-term interest rates, instead of determining the whole yield curve as it is the case now. Moreover, the BCP should adopt a more efficient payment and clearing system to reduce the risks associated with the transfer of securities and foster the development of the interbank and foreign exchange markets.

• Improve further the communication strategy. The recent decision to publish the minutes of the meetings of the BCP’s open market committee will improve the communication and transparency of monetary policy. Over time, the BCP could rely on an explicit, forward-looking framework to convey its views, especially when actual inflation deviates from the target, and explain the actions needed to correct such a deviation.

• Continue to discourage financial dollarization. With Paraguay’s highly dollarized financial system, financial institutions could perceive that the central bank could pay too much attention to the exchange rate, especially if there are concerns about currency mismatches on balance sheets.

• Adopt more transparent mechanisms to intervene in the foreign exchange market. Intervening through auctions that allow for a market-based exchange rate could help alleviate concerns that the central bank is concerned about the level of the exchange rate.

• Implement the recapitalization of the BCP.¹⁰ In coming months, the central bank and the ministry of finance will implement the recently approved law by deciding on key details of the transfer of Treasury bonds to capitalize the central bank, including their maturity and interest rates. This will allow the BCP to focus more on monetary policy issues rather than on the implications of its actions on its own balance sheet, thus enhancing its credibility.

D. Concluding Remarks

7. The analysis has shown that there is scope to enhance the effectiveness of the policy interest rate as the key instrument of monetary policy. Currently, the evidence suggests that the policy interest rate is not the most effective instrument at this time, reducing the value of this signal of the stance of monetary policy. The BCP is already adopting several reforms that—if sustained and accompanied by other modifications—will shift the balance among its monetary policy instruments and enhance the effectiveness of monetary policy.