This Selected Issues paper analyzes the developments in the labor market of Peru during the 1990s. The study assesses the relationship between the orientation of economic policy and export performance, in particular, export diversification over the last four decades. The paper describes the two-tier pension system and evaluates the long-term fiscal burden of this system. The study also reviews the design and implementation of monetary policy in Peru over the last decade.

Abstract

This Selected Issues paper analyzes the developments in the labor market of Peru during the 1990s. The study assesses the relationship between the orientation of economic policy and export performance, in particular, export diversification over the last four decades. The paper describes the two-tier pension system and evaluates the long-term fiscal burden of this system. The study also reviews the design and implementation of monetary policy in Peru over the last decade.

IV. The Design and Challenge of Monetary Policy in Peru65

78. In the last decade, Peru has adopted significant financial reform. It privatized public banks, adopted Basle-consistent prudential standards, enacted a new central banking law, embraced universal banking, fostered a competitive financial system, strengthened banking supervision, and introduced special legislation for the stock market and pension system. These sweeping changes have fundamentally altered the conduct of monetary policy since the early 1990s. This chapter discusses monetary policy in this new environment.

79. The first section outlines the design of monetary policy in the post-reform period. Section B discusses its implementation. Section C assesses its effectiveness against the central bank’s mandate of preserving monetary stability. It argues that the central bank has been effective, but that monetary policy could be more transparent, and views the volatility of the sol-denominated interbank rate as an inevitable consequence resulting from the central bank’s decision to conduct monetary policy by targeting base money. The last section concludes that a formal inflation targeting policy framework could help overcome these concerns.

A. The Design of Monetary Policy

Objective of monetary policy

80. The sole objective of the Central Reserve Bank of Peru (BCRP) established under the Central Reserve Bank Act of 1993 is to preserve monetary stability. The central bank interprets this mandate as a call to “defend the acquisitive power of the currency”66, translating it into the need to continue reducing inflation in Peru to developed countries’ level in the medium term. The act allows the central bank to set the inflation target and choose the instruments to achieve it, and prohibits it from earlier practices of financing the public sector, providing guarantees, directing credit to selected sectors, and establishing a multiple exchange rate regime.67 In addition, the act does not place on monetary policy the objectives of promoting high growth of output and employment, another departure from the previous law. To ensure consistency in economic policies, however, the central bank coordinates closely with the Ministry of Finance to formulate the annual monetary program.

Control of inflation

81. The central bank’s record on inflation has been reasonably strong since the act came into force. At the outset of each year, the central bank announces a target range for end-year inflation. Annual inflation has fallen within the targeted range twice, was lower than expected four times, and slightly overshot the targeted range once, see Figure 1, The target range for 2001 is set at 2.5–3.5 percent.

Figure 1.
Figure 1.

Peru: Target vs. Measured CPI, 1994 - 2000

Citation: IMF Staff Country Reports 2001, 051; 10.5089/9781451831009.002.A004

Source: Central Reserve Bank of Peru.

82. Given its control over base money, the central bank considers that a narrow concept of money is the appropriate immediate target for conducting monetary policy. Thus, the central bank uses a monetary model that is based on domestic currency aggregate only, although more than three-quarters of total bank deposits are held in dollars and about four-fifths of total bank credit are denominated in U.S. dollars. There are two justifications for this. First, central bank research shows that the demand for foreign currency (U.S. dollars) has been motivated largely for reasons of portfolio management while the demand for domestic currency was used mainly for transactions.68 Second, research also shows that since 1994 the demand for a narrow concept of money (i.e., excluding foreign currency aggregates) has been more stable than a broader concept of money, which can be used to bear on the rate of inflation.69 Thus, in the absence of currency substitution and with a reasonably stable demand for narrow money, targeting the quantity of narrow money would provide a basis for attaining the inflation target.

B. Monetary Policy Implementation

Monetary policy formulation and instruments

83. The Board of the central bank approves a monetary program that includes an end-period inflation target range and an average rate of growth for base money. The central bank establishes monthly paths for inflation target and base money, but does not publish them, because these paths need to be adjusted (in reflection of the impact on prices stemming from the strength of internal demand, growth of bank credit, changes in the fiscal position of the public sector, adjustments of inflation expectations, and the depreciation of the currency).70 Like many central banks, the BCRP also does not publish an explicit model linking the inflation target range and the average rate of growth for base money. In the monetary program for 2001, however, the BCRP has provided a diagrammatic exposition of the monetary transmission mechanism for inflation and base money.

84. The BCRP uses three types of instruments for monetary policy and liquidity management: (a) market based instruments such as the auctions and repurchase operations of BCRPCDs (deposit certificates), the purchase and sale of foreign exchange, overnight swaps in foreign currency, and the auction of funds of Banco de la Natión (the government’s fiscal agent) deposited at the central bank, (b) BCRP-based facilities such as the rediscount window and overnight deposits in soles and dollars, and (c) regulatory instruments such as reserve requirements.71

Market based instruments

85. Central bank deposit certificates. The BCRP is increasingly active in using BCRPCD auctions to regulate excess liquidity, even though this is not carried out with regular periodicity. Bank bids are accepted on a best price offer basis (from highest to lowest price until the desired auction amount is satisfied). During tax payment periods (usually lasting about 2 weeks in the middle of the month), the central bank conducts repurchase operations to inject liquidity into the system. Repo bids are also accepted based on best price offers (from lowest to highest price until the desired repurchase amount is satisfied).

86. Foreign exchange intervention and swap operations. Prior to 2000, through the unsterilized purchase (sale) of foreign exchange in the interbank market, the BCRP injected (withdrew) domestic currency liquidity in the banking system. However, beginning in 2000, the monetary authorities no longer intervene in the foreign exchange market for the objective of injecting or withdrawing liquidity.72 The BCRP also injects temporary liquidity by swapping soles for foreign exchange from banks. These swap operations are normally for 24 hours. The cost of the swap to banks is either the commission established by the BCRP or the depreciation of the sol during the swap, whichever is higher, so that exchange rate risk is borne entirely by banks.

87. Auction of Banco de la Nación funds. The auctioning of funds from Banco de la Nación deposited at the BCRP as a liquidity management instrument is rare. Nevertheless, when the choice of instruments for the central bank is limited, as for example when the stock of outstanding BCRPCDs was low during the liquidity crunch in December 1998, the central bank injected liquidity by auctioning these deposits.

BCRP-based facilities

88. The central bank can extend domestic or foreign currency credit to banks with temporary liquidity problems. The BCRP credit rate in soles is set deliberately high, and never lower than the average interbank market rate, so that banks needing sol-denominated liquidity are encouraged to approach the interbank market first and access this credit facility as a last resort. For February 2001, the Board of the central bank set the sol-credit rate at 12.5 percent or the average of the top decil of interbank interest rates, whichever is higher. In the case of the U.S. dollar credit facility, which is infrequently used, a flat rate is set. In February, this rate was 8.5 percent. The Banking Law limits bank access to this “rediscount” window to no more than 90 days over a period of 180 days before triggering mandatory surveillance by the banking supervisor.

89. The central bank created a facility in 1998 for dollar funds in an overnight account that remunerates deposits at the average interest rate which it receives from investing such funds overseas (approximately LIBOR). The central bank extended the overnight facility to soles funds in September 2000. In February 2001, the interest rate for this facility was 4.5 percent.

Regulatory instruments

90. All deposits in domestic currency are subject to a 6 percent reserve requirement (reduced from 7 percent in September 2000), which are not remunerated. Deposits in foreign currency, remunerated at LIBOR minus a small fraction of one percent, are subject to a marginal reserve requirement of 20 percent. An average reserve requirement of 34 percent (following a three percentage point reduction in September 2000) is levied on the average stock of dollar deposits as of end-October 2000.

91. For a highly dollarized economy such as Peru, the reserve requirement on foreign currency has acted to reduce the upward pressure on the sol at times of strong capital inflow and moderated the ensuing monetary growth in foreign currency. It also provides the central bank with substantial reserves that cushion the financial system from the risk of capital outflow.73

92. While the BCRP does not generally rely on the reserve requirement on sol-denominated or dollar-denominated deposits to inject or withdraw liquidity, there have been exceptional cases where this has been done. For example, in late 1998, the BCRP reduced the average reserve ratio on dollar deposits on three occasions (reducing required reserves by US$420 million in total) to help offset the cutback in foreign lines of credit (of a similar amount) at the time of the Russian and Brazilian crises.

Daily monetary management

93. The central bank’s daily monetary management is based on a daily estimate of banking system liquidity. This estimate is based on a set of indicators such as banks’ cash balances, bank reserves (required reserves as well as those freely available) in domestic currency held at the central bank, the level of interbank interest rates, and the net check-clearing position of banks.74 The central bank then attempts to influence the level of bank reserves through open market operations and foreign currency swaps, while accounting for other factors such as the expected level of tax payments and public sector expenditures during the day, which are outside its control.

94. The BCRP makes a daily projection of bank reserves that is consistent with its monthly base money target and compares this with the estimated daily bank reserves. The latter is obtained by adding the observed bank reserves of the previous day and the net change in bank reserves estimated for the day (before open market operations). The decision to inject or withdraw liquidity during the day is made after comparing the projected bank reserves and the estimated bank reserves. To get an indication of the daily change in bank reserves (before open market operations), the BCRP looks at bank transactions that have a major impact on their reserve positions. Outflows from banks that reduce their reserve position would include transfers (in soles) to Banco de la Nación, the expiration of sol-denominated loans from the central bank rediscount window, the unwinding of repo transactions in BCRPCDs, reverse swaps in foreign exchange, and repayments of deposits of the Banco de la Nación auctioned by the central bank. Inflows to banks that increase their reserve positions would include the expiration of BCRPCDs, transfers of cash in bank vaults to reserve accounts in the central bank, net purchases of foreign exchange by the BCRP, auctions of Banco de la Nación funds held at the central bank, repo transactions in BCRPCDs, access to the central bank’s rediscount window, and swaps in foreign currency.

95. Interest rates and the exchange rate are market determined. As monetary management is geared at controlling a monetary aggregate, the central bank does not target any particular interest rate, even though it has from time to time expressed concerns about its level and volatility. The central bank lets market forces determine the level of the exchange rate, and does not have, as an objective, a particular target exchange-rate level. However, it has managed liquidity as needed, to smooth out sudden and sharp fluctuations in the exchange rate when the authorities considered that pressures in the exchange market were not reflecting underlying fundamentals; at times, this resulted in long periods of significant stability in the exchange rate.

C. Effectiveness of Monetary Policy

96. The conduct of monetary policy has been reasonably effective, but lacked transparency and has had an unintended impact on interbank interest rate volatility. Together with fiscal consolidation and wide-ranging structural reforms (including privatization, liberalization of the exchange rate and interest rates, and the elimination of various distortionary measures that stifled financial intermediation), prudent monetary policy management since the early 1990s has led to a sustained reduction in inflation, a significant accumulation of net international reserves, and an increase in financial intermediation. Real interest rates which were negative in the late 1980s and 1990 turned positive, and extremely high nominal sol-denominated interest rates of the early 1990s fell as inflation declined (see Table 1).

Table 1.

Peru: Selected Financial Indicators, 1990–2000

article image
Source: BCRP.

Monetary policy transparency

97. Recently the central bank took steps to improve transparency and accountability of monetary policy. Monetary policy procedures and transparency in goals, instruments, and decisions in the past fell short of the best practices recommended by the Code of Good Practices on Transparency in Monetary and Financial Polices. In previous years, apart from the publication of an annual monetary program, the central bank did not systematically report and account for its policy decisions.75 However, in early 2001, it has taken steps to address this deficiency.

a. In January, details of the monetary program for the year were published that make the formulation of monetary policy more transparent. In addition, the central bank will also publish in May and September an update of the annual monetary program.

b. Beginning in February, monthly information is being provided to guide the market on monetary operations for the month. On the first Friday of every month, the central bank will publish a report76 providing: (a) the daily average target range on the reserve balance banks held at the central bank and (b) interest rates that the central bank will charge for monetary regulation credits and the interest rate it will pay on overnight deposits held by banks at the central bank.

98. While this increase in transparency is welcomed, more could be done. For example, the central bank could announce the base money target for the month. It could explain in the monthly report how the evolution of financial indicators, such as interest rates, credit growth, and the exchange rate, affected its operations. In updating the monetary program in May and September, the central bank could comment on the outcome of the inflation and base money paths and explain any modifications to monetary policy, if any, as conditions change. These steps will further help the market anchor inflation expectations and enhance market confidence that the BCRP is committed to keeping inflation low.

99. Limited publicly available information in the past impeded a rigorous scrutiny and evaluation of monetary policy. In the past six years, even though yearly inflation has been broadly in line with the target, the (ex ante) intermediate targets for average base money growth have largely been out of line with the monetary growth outturns that, supposedly, ought to have been consistent with the inflation target. While the central bank was clearly more concerned with meeting the inflation target than the base money intermediate target, explaining why the latter had to be readjusted would have improved transparency.

100. The relationship linking the average growth rate for base money and the end-period inflation target (or the relationship between the daily operational target, bank reserves, and inflation target) has not been made public. Independent verification of whether the average base money growth rate target had been met was not possible because this target was not announced prior to 2000. Ex-post comparisons of the upublished targets for 1994-99 reveal sizeable deviations from their outturns; albeit, with a decreasing margin over time (see Table 2).

Table 2.

Peru: Unpublished Average Annual Base Money Growth Target vs. Outturn

article image
Source: BCRP.

In 2000, the BCRP announced, for the first time, the average base money growth rate.

101. In the absence of information on the target for average growth of base money, the public may have been tempted in the past to evaluate monetary policy by comparing the publicly available end-period base money projection and its outcome; however, actual base money growth (end-of-period basis) also deviated significantly from the projection (see Table 3).

Table 3.

Peru: Publicly Available End-Period Annual Base Money Growth Target vs. Outturn

article image
Source: BCRP.

Interbank interest rate volatility

102. The second concern regarding the conduct of monetary policy relates to the volatility of sol-denominated interbank interest rates (both over the course of the month and on a daily basis). Sol-denominated interbank rates exhibit a clear intramonthly pattern. The fluctuation is closely linked to the monthly tax payment schedule. In the two weeks when the main taxes are paid (starting around the 9th working day of each month), the fall in sol-denominated bank liquidity (as proxied by a decline in bank reserves at the central bank) is accompanied by an increase in the average interbank rate in soles. The cycle is reversed after taxes are paid (around the 18th working day of the month). Figure 2 shows that from October 1997–June 2000, average bank reserves increased in the first part of the month leading up to the tax payment period, followed by a decline in the days when taxes were paid, and then rose slightly again toward the end of the month. Figure 3 shows the inverse relationship of the average interbank rate in soles over the month.

Figure 2.
Figure 2.

Average Bank Reserves at BCRP in Soles

(October 1997-June 2000)

Citation: IMF Staff Country Reports 2001, 051; 10.5089/9781451831009.002.A004

Figure 3.
Figure 3.

Average Interbank Rate in Soles

(October 1997-June 2000)

Citation: IMF Staff Country Reports 2001, 051; 10.5089/9781451831009.002.A004

Source: BCRP; and Fund staff estimates.

103. The daily interbank rate in soles exhibited considerable volatility in the last decade, even though reducing its volatility has long been a concern of monetary policy.78 To help achieve this objective, the central bank started to remunerate overnight sol-denominated deposits of commercial banks for the first time in September 2000, in an effort to set a floor for the overnight sol-denominated interbank rate. Stability in the interbank rate in soles should help lower the average sol-denominated lending rate and increase financial intermediation in soles.79 The dollar-denominated interbank rates were more stable than their sol-denominated counterpart (Figure 4 and Table 4).

Figure 4.
Figure 4.

Peru: Interbank Overnight Rates, 1997 - 2000

Citation: IMF Staff Country Reports 2001, 051; 10.5089/9781451831009.002.A004

Source: BCRP.
Table 4.

Peru: Interbank Rate Volatility. 1997–2000

article image
Source: BCRP.

104. Interbank interest rate volatility has declined in recent months, presumably reflecting an effort by the BCRP to reduce its fluctuation. The standard deviation of this rate has fallen from 6.7 percent in 1998 to 4.9 percent in 1999 and to 2.7 percent in 2000. Fluctuation in the sol-denominated interbank rate cannot be totally eliminated because the use of bank reserves (quantity) as the daily operating instrument for achieving the monthly base money target affects the price of these reserves, namely, the interbank rate, which is freely determined in the interbank market. Thus, if eliminating sol-interbank-rate volatility is an important goal of monetary policy, an alternative to targeting base money in the form of directly targeting the interbank rate should be considered.80

105. High volatility in sol-denominated interbank rates can lead to volatility in sol-denominated lending rates; thus, discouraging borrowing in the domestic currency, even as inflation falls.81 Furthermore, in 1998 and 1999, sol-denominated lending remained below 20 percent of total lending and was significant82 only in sectors dominated by the public sector (public administration, health, and electricity and water services). The share of total bank credits extended to these sectors was less than five percent (see Table 5).

Table 5.

Peru: Percent of Outstanding Bank Credits to Sectors by Types of Currency

article image

Share of total outstanding bank credits accounted for by each sector.

Source: SBS.

106. The lack of long-term lending in soles is influenced by the volatility of the interbank interest rates in soles. Confronted with the frequent need to roll over loans, whose future interest rates are uncertain, investors may choose to borrow in dollars, where there is less interest-rate volatility. Smoothing the sol-denominated interbank rate can, therefore, reinforce the authorities’ recent efforts to lengthen the sol-denominated yield curve (for the first time in August 2000, the central bank issued one year CDs while COFIDE, a second-tier public financial institution, issued two-year, fixed-interest-rate instruments in soles).

D. Inflation Targeting as a Framework for Monetary Policy

107. The authorities have expressed interest in inflation targeting and, together with the Fund, they are organizing a seminar on the topic to assess the suitability of this approach for Peru. Without prejudging the conclusion of the seminar, it is widely recognized that inflation targeting is an approach that would increase transparency and strengthen accountability, as well as providing an anchor for inflationary expectations and increasing the predictability of market reactions to monetary policy announcements. Also, countries adopting formal inflation targeting normally use the interbank interest rates as a daily operational target of monetary policy. In the case for Peru, this would lead to lower volatility in interest rates.83

108. Peru has already met the main prerequisites for inflation targeting, namely, central bank independence and policy effectiveness in controlling inflation.84 On transparency, the central bank through its monthly report is moving in the direction of establishing a forum to explain the design and implementation of monetary policy. The May and September reports would update the annual monetary programs as conditions change in the course of the year. Over time, these reports could be expanded and form a basis of a future inflation report.85

109. The central bank is in a good position to tackle nearly all technical issues of inflation targeting, but needs to strengthen its inflation-forecasting model. The central bank would have no problems addressing the choices of a relevant price index (general or core inflation), an inflation target range (point target or band), and a target horizon (one or more years), and the central bank has a firm understanding of the transmission mechanism of monetary policy.86 However, as inflation targeting uses a forward-looking framework to predict the lagged impact of monetary policy and sets conditional inflation forecasts as intermediate targets, the ability to produce a path of credible inflation forecasts is vital. Such forecasts would be used to judge, ex-post, the effectiveness of monetary policy actions. More needs to be done in this area as work on a highly reliable inflation model is still at an early stage.

List of References

  • Amato, J. and T. Laubach,The Value of Interest Rate Smoothing: How the Private Sector Helps the Federal Reserve,Economic Reviews—Federal Reserve Bank of Kansas City, pp. 113, 1999.

    • Search Google Scholar
    • Export Citation
  • Berg, A. and E. Borensztein,The Choice of Exchange Rate Regime and Monetary Target in Highly Dollarized Economies,IMF Working Paper (WP/00/29), 2000.

    • Search Google Scholar
    • Export Citation
  • Blejer, M., Ize, A., and Leone, A. (eds), Inflation Targeting in Practice: Strategic and Operational Issues and Application to Emerging Market Economies, IMF, 2000.

    • Search Google Scholar
    • Export Citation
  • Bringas, P. and Tuesta, V., El Superavit de Encaje y los Mecanismos de Transmision de la Politica Monetaria en el Peru, Central Reserve Bank of Peru, 1996.

    • Search Google Scholar
    • Export Citation
  • Central Reserve Bank of Peru, Notas de Estudios, No. 6, La Politica Monetaria del Banco Central de Reserva del Peru, December 1999.

  • Central Reserve Bank of Peru, Programa Monetario para el año 2000, 2000.

  • Choy Chong, M., Monetary Policy Operating Procedures: the Peruvian Case, Bank of International Settlement, 1998.

  • De la Rocha, J., The Transmission of Monetary Policy in Peru, Bank of International Settlement, 1998.

  • Friedman, B.,Federal Reserve Policy, Interest Rate Volatility, and the U.S. Capital Raising Mechanism,Journal of Money, Credit, and Banking, pp. 721745, 1982.

    • Search Google Scholar
    • Export Citation
  • Quispe, Z., Monetary Policy in a Dollarized Economy: The Case of Peru, Central Reserve Bank of Peru, 1998.

  • Quispe, Z., Una Aproximacion a la Demanda de los Principles Agregados Monetarios en el Peru: junio 1991–mayo 1997, Central Reserve Bank of Peru, 1999.

    • Search Google Scholar
    • Export Citation

Annex: Summary of the Tax System as of December 2000

Peru: Summary of the Tax System

(December 2000)

article image
article image
article image
article image
article image
article image

Table 1.

Peru: Aggregate Supply and Demand

article image
Sources: Central Reserve Bank of Peru.

Investment of central government and nonfinancial public enterprises.

Computed as a percentage of domestic demand.

Table 2.

Peru: Saving, Investment, and External and Fiscal Balances

(In percent of GDP, unless otherwise indicated)

article image
Sources: Central Reserve Bank of Peru.

Includes exceptional financing.

Includes net borrowing by the financial public sector, and errors and omissions.

Excludes grants.

Table 3.

Peru: National Accounts at Current Prices

(In millions of new soles)

article image
Source: Central Reserve Bank of Peru.

Central government and nonfinancial public enterprises.

Table 4.

Peru: National Accounts at Constant Prices

(In millions of 1994 new soles)

article image
Source: Central Reserve Bank of Peru.

Central government and nonfinancial public enterprises.

Table 5.

Peru: Sectoral Origin of Gross Domestic Product at Constant (1994) Prices

article image
Source: Central Reserve Bank of Peru.
Table 6.

Peru: Growth in Agricultural and Livestock Production

(Annual percentage change)

article image
Sources: Ministry of Agriculture; and Central Reserve Bank of Peru.
Table 7.

Peru: Agricultural Production

(Production in thousands of metric tons; area in thousands of hectares; and yield in metric tons per hectare)

article image
Sources: Ministry of Agriculture; and Central Reserve Bank of Peru.
Table 8.

Peru: Production, Consumption, and Exports of the Fishing Industry

article image
Sources: Ministry of Fishing; and Central Reserve Bank of Peru.
Table 9.

Peru: Mineral Production by Major Products

article image
Sources: Ministry of Energy and Mines; and Central Reserve Bank of Peru.

Percentage change in output volume.

Percentage change in output valued at 1994 prices.

Table 10.

Peru: Sources and Uses of Petroleum and Petroleum Derivatives

article image
Sources: Ministry of Energy and Mines; and Central Reserve Bank of Peru.

Includes own consumption by refineries and sales of bunker fuel.