Chapter 3: Moving toward a Market-Based Monetary Policy
- International Monetary Fund
- Published Date:
- March 2007
Reform of the monetary policy framework in Iran is needed in connection with the efforts to liberalize the financial system (Chapter 2) and the authorities’ objective of reducing the inflation rate to single digits. Drawing on academic research and the experience of other countries, as well as lessons learned from monetary policy management in Iran over the past 14 years, this chapter sketches a transition path from a system characterized by administrative controls and fiscal dominance toward one based on market incentives and signals. The chapter outlines a set of options for moving toward this objective by increasingly relying on indirect instruments of liquidity management consistent with Islamic finance principles.
The next section analyzes the main lessons from Iran’s experience in conducting monetary policy and highlights the difficulties encountered by the Central Bank of Iran in achieving its monetary policy goals with limited instrument independence. The following section underscores the need to clarify monetary policy objectives and targets, enhance central bank instrument independence, improve the coordination between monetary and fiscal policies, and develop indirect instruments of liquidity management. The conclusion recommends a transition toward a monetary aggregate targeting.
Lessons for Monetary Policy Implementation
Current monetary policy formulation and implementation rely to a large extent on administrative controls in the context of fiscal dominance (Box 3). Although administrative controls are in part used to alleviate the inflationary impact of fiscal dominance and to favor the redistribution of credit resources according to government priorities, they are also motivated by the slow progress toward developing money market instruments that are consistent with Islamic finance principles (i.e., Sharia).
Box 3.Fiscal Dominance
Fiscal dominance has been an important source of high liquidity growth and inflation in Iran. Fiscal dominance stipulates that “monetary policy is subordinated to fiscal financing requirements” (Sargent and Wallace, 1981) or that fiscal policy is active whereas monetary policy is passive (Leeper, 1991).
In the context of Iran, an oil-producing country, the government budget relies to a large extent on oil export revenue earned in foreign currency, and until recently, domestic and foreign bond financing has been very limited. Two separate channels of fiscal dominance are at play in Iran: (1) central bank financing of government deficits (“pure” seigniorage) and (2) spending out of foreign-exchange-denominated oil revenue, which results in an increase in high-powered money.
- The role of pure seigniorage has been steadily declining. Direct central bank credit to the government was virtually discontinued in 1998/99 (except for bank recapitalization operations and some quasi-fiscal subsidies financed by the central bank), although central bank credit to nonfinancial public enterprises (NFPEs) continues on a small scale (Figure 6).
- Spending out of export oil revenue is equivalent to a foreign-financed expenditure and thus has a substantial effect on base money and real rates of return, especially under less-than-perfect capital mobility (Barnett and Ossowski, 2003). In Iran, the variations in expenditure and in its liquidity effects have been high (Figure 6). In this context, fiscal dominance is manifested in the inability of the central bank to offset sudden large changes in liquidity conditions stemming from fluctuations in government sales of foreign-currency-denominated oil revenue—mainly owing to their sheer magnitude. Insufficient development of appropriate instruments of liquidity control has also complicated liquidity management.
Figure 6.Sources of Base Money Growth, 1991/92–2002/03
Sources: Iranian authorities; IMF staff estimates.
1 Estimated by subtracting all sources of financing from oil revenue.
The current approach to monetary policy formulation gives the government a decisive influence in setting specific monetary targets. In particular, FYDPs set annual targets for monetary growth and inflation that are approved by parliament and must be used by the central bank as benchmarks for formulating monetary programs. At the operational level, the Money and Credit Council (MCC) is responsible for key monetary policy decisions. Since 2005/06, the governor of the central bank has chaired the MCC; the minister of economy and finance is also a member, as are other ministers and government officials. Parliament and the government may issue directives for credit allocation, which could have implications for monetary policy implementation. In practice, the FYDP targets for broad money (M2) and inflation are usually revised by the MCC in its annual monetary guidelines. But even these revised targets are often inconsistent with fiscal financing requirements and other important decisions, in particular those on the administratively set rates of return and other direct controls on banking system activities.
Against this background, the central bank has not been able to meet its intermediate target for M2 since the inception of FYDPs. More important, these targets were exceeded by very large margins; as a result, the inflation rate objectives were not achieved during the first two FYDPs (Table 12).
|FYDP I1||Outcome||FYDP II||Outcome||FYDP III||Outcome2|
Five-year development plan.
Four-year averages, 2000/01–2003/04.
At factor cost at constant 1997/98 prices.
Five-year development plan.
Four-year averages, 2000/01–2003/04.
At factor cost at constant 1997/98 prices.
Evolution of Monetary Policy Instruments
The role of direct instruments of monetary policy in Iran has gradually declined. Direct ceilings on refinancing facilities of commercial banks were abolished in 1991/92 and new directed, or “prearranged,” credits of the central bank to commercial banks were de facto discontinued in 1998/99. The requirement for banks to hold government bonds was gradually relaxed during 1993/94, when banks were authorized to sell these bonds to the central bank.
Sectoral credit allocation limits and control on rates of return of state-owned banks have been gradually eased but remain significant. They are approved at the beginning of the fiscal year and almost never adjusted in the course of the year. The share of banking credit subject to sectoral allocation limits has been gradually reduced to 55 percent. Although state-owned banks may allocate 45 percent of loans without sectoral restrictions, they are still bound by administered rates of return that are fixed for each sector (Chapter 2).
Controlled rates of return on both loans and deposits were negative in real terms for most of the period under review, contributing to inflationary pressures and lack of progress in financial deepening (Figure 7). Real rates of return display a clear procyclical pattern—they are negatively correlated with the output gap—which means that monetary conditions become tighter during economic slowdons and more relaxed during expansions (Figure 8). This is attributable to the limited flexibility of the rates of return combined with the liquidity impact of a procyclical fiscal policy.
Figure 7.Real Rates of Return, 1991/92–2002/03
Sources: Iranian authorities; IMF staff estimates.
Figure 8.Non-Oil Output Gap and Real Rates of Return, 1991/92–2002/03
Sources: Iranian authorities; IMF staff estimates.
Required reserves are high and differentiated by maturity of deposits, which has allowed the central bank to have better control of broader monetary indicators and to increase the demand for base money, thereby offsetting some of the impact of fiscal dominance. This situation in effect represents a tax on financial intermediation, because required reserves are only remunerated at 1 percent per year. The weighted average required reserve ratio declined to 16 percent in 2003/04 from 23 percent in 1990/91, which in part explains an upward trend of money multipliers.
Foreign exchange operations, excluding those with the central government, are an increasingly important element of monetary policy implementation. In 2002/03, the central bank’s sales of foreign exchange in both domestic and offshore interbank markets amounted to about $13 billion (almost 100 percent of beginning-of-period base money).
The overdraft facilities of the central bank do not support monetary policy objectives. These facilities are in great demand by commercial banks because the interbank market is virtually nonexistent. In practice, the central bank accommodates liquidity shortfalls in the payment system without consistently enforcing existing incentives against repeated large recourse to overdrafts. Until 1993/94, overdraft rates were set at only 2 percentage points above the directed credits; in that year, they were replaced by a progressive schedule of overdraft rates at 20, 24, and 30 percent, depending on access levels. If overdraft periods exceed three days, an additional 4 percentage points are added to each tier. Despite revisions to other administered rates in the banking system and sharp fluctuations in annual inflation since 1993/94, the overdraft rates have not been revised. Moreover, on some occasions, overdraft penalty payments have been waived. Frequent recourse to relatively cheap or even penalty-free overdraft facilities often puts the central bank in an accommodating position. There is also a standing credit facility providing financing up to one year.
The standing deposit facility (open deposit accounts) introduced in 1998/99 has played a marginal role in central bank operations. This facility has been used to regulate liquidity fluctuations within the year; but by year’s end, when most banks experienced liquidity pressures, the deposits were usually drawn down substantially. In 2003/04, the central bank attempted to auction deposit facilities to commercial banks, but there was no demand for such instruments, in part because of the unattractive terms.
Central bank participation papers (CBPPs)—which are compatible with Islamic finance principles—were first issued in March 2001 (Table 13). The bearer securities are issued in parcels of Rls 1, 2, 5, and 10 million, which reflects their retail focus. Maturities are 6 or 12 months, with nontaxable quarterly coupon payments. The CBPPs are issued based on a portfolio of completed infrastructure projects previously financed by the central bank credit to government; they yield a predetermined rate of return presumed to approximate the returns on underlying assets. This means that the CBPPs are backed by underlying central bank claims on the government. The objective was to create a marketable money market instrument that would empower the central bank to regulate liquidity and provide a viable instrument for liquidity management by commercial banks, leading to the development of an interbank market (Ul Haque and Mirakhor, 1998). However, the final design turned out to be somewhat different.
|Government Participation Papers (national)|
|Amount issued (gross)||2,174||2,500||1,927||0||0||2,400|
|Average maturity (years)||3.0||3.0||4.0||…||…||5.0|
|Average rate of return (in percent, per year)||20.0||20.0||19.0||…||…||15.0|
|Central Bank Participation Papers|
|Average maturity (years)||…||…||…||…||0.8||1.0|
|Average rate of return (in percent, per year)||…||…||…||…||17.5||17.0|
|In percent of beginning-of-period base money||…||…||…||…||10.5||17.6|
CBPPs have been issued only to nonbanks in the primary market, at preannounced rates of return. Commercial banks are obliged to rediscount them in the secondary market at par and to guarantee the initial yield to maturity. In other words, CBPPs cannot be traded in the secondary market at prices different from par and thus represent a highly liquid instrument held by nonbanks. Although to some extent these instruments have helped absorb liquidity created by government sales of foreign-currency-denominated oil revenue, their high cost to the central bank raises questions about the sustainability of their growing stock. Moreover, at times, the central bank has not been able to achieve the targeted amount of issues at the rate of return fixed by the MCC. These design features of CBPPs have limited their effectiveness.
Government participation papers (GPPs) were first issued in 1998 (Table 13), but they did not play a major role in attenuating the impact of fiscal dominance or stimulating the development of money markets. (Participation papers of municipalities and various ministries and public enterprises have been authorized since 1994.) A GPP is an instrument used to finance nonspecific government infrastructure projects by providing investors a temporary (equal to the maturity of the paper) equity stake in the underlying assets. The government promises to pay on maturity a return that approximates the rate of return on the underlying asset, which should be at least equal to the private sector rate. GPP is a different instrument than the government bonds issued in the 1980s. GPPs are not designed for use by the central bank to manage liquidity but were primarily intended to finance central government infrastructure projects. GPPs are issued in the primary market at preannounced fixed rates of return with a five-year maturity, and the outstanding stock is modest. Banks are obliged to rediscount GPPs in the same manner as CBPPs (see above). The tax-adjusted rates of return on GPPs are below those on CBPPs, despite the much longer maturity of the former. This implies a negatively sloped yield curve of rates of return, which has reduced the attractiveness of GPPs in the presence of high uncertainties over future inflation developments.
Monetary Policy since 2002
The 2002 exchange rate unification and the establishment of a managed float exchange rate regime raised the issues of the appropriate nominal anchor and related supporting policies. Although the central bank has increasingly focused monetary policy implementation on M2, this policy has not yet achieved sufficient credibility to effectively anchor inflationary expectations. In the aftermath of the 2002 exchange rate unification, exchange rate considerations continued to dominate, initially out of concern for the stability of the nominal rate and then to preserve competitiveness through a gradual nominal effective depreciation of the exchange rate to compensate for past inflation differentials. However, this dual-objective policy became difficult to sustain in the face of an increased supply of foreign exchange stemming from fiscal relaxation and FDI inflows. In an attempt to contain the growth of monetary aggregates while continuing with nominal exchange rate depreciation, the central bank began to use CBPPs to mop up excess liquidity at relatively attractive fixed rates of return, thereby directly bearing the cost of sterilization operations. With the rapid increase in the CBPP stock, these operations became costly and less effective at offsetting large injections of oil revenue into the system. As a result, the amount of unsterilized purchases became a function of exchange rate objectives, and the control over monetary aggregates weakened. The policy of nominal depreciation did not prevent the REER from appreciating by 7.5 percent during 2002/03–2003/04, 1 reflecting a large inflation differential between Iran and its trading partners.
During 2003/04, the policy on rates of return conflicted with monetary policy targets. Negative real rates of return on loans for the major sectors of the economy—such as industry, mining, and agriculture—helped fuel credit demand growth, which the central bank accommodated through its overdraft facilities. It is clear that the rapid credit growth, together with large unsterilized purchases of foreign exchange from the government, also contributed to nominal exchange rate depreciation.
Monetary Policy Framework
There is a pressing need to improve the monetary policy framework and to enhance coordination between fiscal and monetary policies. Drawing on the extensive academic literature and operational experience of central banks in market-based financial systems (Walsh, 2003), this subsection highlights the general principles of an effective monetary policy framework; the next subsection elaborates on the implications of using Sharia-compliant monetary policy instruments.
Monetary Policy Objectives
Since Iran has chosen a managed float exchange rate regime,2 price stability should be an overriding objective of monetary policy. Possible intermediate targets consistent with this objective include a monetary aggregate or a measure of consumer price inflation. Targeting a monetary aggregate such as broad (M2) or narrow (M1) money is perhaps more familiar to policymakers in Iran and has less stringent institutional and policy requirements than targeting, say, an annual percentage change in the CPI.3 The M2 intermediate target could be set jointly by the government and the central bank, consistent with the desired outcomes for inflation. Greater simplicity in the transparency and accountability requirements under monetary targeting is a distinct advantage. In contrast to inflation targeting (Schaechter, Stone, and Zelmer, 2000), adherence to monetary targets is easier to monitor. Timely publication of data on monetary aggregates is an important accountability requirement, but it is easier to prepare, interpret, and understand than subtle explanations of monetary policy actions needed to achieve inflation objectives in the presence of complex transmission mechanisms.
A transition toward monetary aggregate targeting could be considered notwithstanding money velocity instability in the recent past. The empirical study of Celasun and Goswami (2002) finds that inflation is affected by real money and output growth but also depends on exchange rate developments and the degree of deviation of real money demand from its equilibrium level. The implication of these findings is that inflation forecasts are subject to uncertainty, and it would be difficult for the central bank to meet an inflation target by adhering strictly to intermediate M1 or M2 targets. This does not mean, however, that M1 or M2 targeting should not be considered an option. An indicative inflation objective could be formulated as a band in order to accommodate forecast errors in setting an annual target for M1 or M2 growth.
The central bank needs to determine an operational target that it can directly control, such as base money or money market rates of return. The latter, however, are not readily available because of the slow development of money markets in Iran. Thus, base money could be initially selected as an operational target. Given the instability of money multipliers,4 the base money target should be revised periodically in light of new information to maximize the chances of hitting the intermediate target for M1 or M2. It is also important to ensure that fiscal policy is consistent with the need to achieve the base money operational target, which would require a careful assessment of the liquidity impact of fiscal operations at the budget preparation stage. Specifically, the authorities would need to ensure that the size and composition of the non-oil deficit financing is consistent with the operational and intermediate targets of monetary policy.
Instrument Independence of the Central Bank
As mentioned earlier, current legislation provides the central bank with limited authority to use monetary policy instruments—such as rates of return or the amount of CBPP issuance—without prior approval from the MCC. This institutional arrangement does not give the central bank the needed flexibility to deal rapidly with changes in money and credit conditions during the year. Moreover, the fact that the government or parliament can issue credit directives undermines the ability of the central bank to meet its intermediate targets. Granting the central bank instrument independence—defined as day-to-day independence in using all relevant instruments needed to achieve intermediate targets, subject to a possible override provision for the government—is essential for successful implementation of monetary aggregate targeting. Such independence will need to be combined with stringent accountability requirements to various layers of authority and the public. There is also a need to establish transparent procedures for resolving potential conflicts between monetary policy and broader economic policy objectives.
The range of instruments at the disposal of the central bank should be broadened, and the existing instruments should be adapted to the new framework. The required reserve ratios should be unified and reduced to lower the cost of financial intermediation, provided that offsetting measures are put in place to mop up excess liquidity. Access to the standing credit facilities should be tightened and made more onerous to discourage frequent use. Indirect instruments of monetary policy need to be redesigned to gradually become the preferred instruments in order to facilitate the emergence of a benchmark rate of return (the following subsection includes a description of options that could be implemented in the short run). Once money markets grow in depth and experience, the operational target could be changed from base money to the rate of return on an appropriate money market instrument. Foreign exchange operations will continue to be important, but the central bank needs to gradually shift the emphasis in these operations from the exchange rate to base money by more actively using indirect instruments of monetary policy and tolerating greater fluctuations in the exchange rate.5 Monetary instruments alone, however, are unlikely to be sufficient to sterilize the liquidity impact of injections of government oil revenue into the system (see Box 3) or of large capital inflows; thus, fiscal policy actions would also be needed. In the long run, incorporating market-based mechanisms into the design of government participation papers and increasing their outstanding volumes and liquidity would also facilitate the deepening of financial markets.
Sharia-Compliant Indirect Instruments of Monetary Policy
This subsection reviews options for developing Islamic money market instruments in Iran. A number of difficulties arise in designing short-term financing instruments that are Sharia-compliant (that is, are interest-free, rely on profit and loss sharing linked to real transactions, or are based on purchase and resale contracts) and whose value can be determined at a high frequency to facilitate short-term trading and money market operations. Several central banks—notably in Malaysia, Sudan, and Bahrain—have developed Islamic financial instruments to facilitate liquidity management (by the central bank as well as by commercial banks) and public borrowing (Majid, 2003). At the same time, there has been a growing use of asset securitization techniques to design Islamic securities for issuance in regional and international capital markets (Hassan, 2002). This has opened the door for developing short-term instruments for monetary operations.
Effective market-based monetary operations require an instrument with the following characteristics:
- The instrument must be relatively risk-free. It must be able to serve as a benchmark to price more risky instruments of varying maturities and strongly influence the marginal cost of funds for banks.
- A sufficient supply of the instrument must be available to meet both monetary policy needs and the portfolio needs of investors.
- The instrument must be widely held by both banks and nonbanks to support a liquid market.
- The payment settlement system must be robust and reliable to facilitate trading in the instrument.
On the basis of these criteria, a structure that securitizes a range of Islamic financial contracts seems the most promising for monetary operations; all other market instruments do not meet one or more of the criteria for effective monetary operations.
Purely equity-based instruments (Musharaka), or structures containing only cash-flow rights from government ownership in enterprises can carry high returns that raise costs to the government. Also, the volume of issuance of such instruments may not be sufficient for monetary policy purposes, insofar as the issue amounts are limited by the extent of government ownership in high-quality enterprises.
Purely commodity-resale-type instruments (Murabaha) and participation papers with guaranteed minimum returns may not trade at prices different from par under Islamic finance principles and thus cannot be a reliable basis for developing interbank money markets.
Purely debt-type contracts (Mudarabah), such as interbank deposit placements linked to bank profits, are not suited for the liquidity-absorption operations of central banks owing to the difficulties of linking returns to central bank profits. Moreover, differences in perceived bank risks might limit the volume of interbank placements.
Thus, securities based on a mixture of contracts representing equity-(Musharaka), debt- (Mudarabah), and leasing-type financing (Ijara) have the best chance of being issued in sufficient volume, achieving adequate market liquidity, and providing sufficient flexibility in the mix of risks and return. The mix of contracts should be transparent to allow investors to assess risks and form expectations of returns based on expected performance of the underlying cash flow.6
The ongoing issues of CBPPs in Iran can be transformed over time to become an effective instrument of monetary management by
- identifying a wider range of government assets and cash flows that can be securitized,
- strengthening the coordination of public expenditure management and the government financing program to ensure an optimal combination of assets that can be securitized,
- adopting a high-quality and transparent accounting and disclosure framework for communicating the value and returns on the underlying assets,
- adopting auction-based primary issuance that reflects market expectations in the price of the security,
- supporting the liquidity of the instrument in the secondary market through repurchase facilities, and
- organizing efficient trading and payment settlement arrangements.
Such newly designed Sharia-compliant instruments would overcome the current constraints on Iran’s CBPPs and allow more flexible rates of return and better-functioning secondary markets to emerge, thereby facilitating more effective monetary and public debt management.
The current system of monetary policy formulation and implementation still largely relies on administrative controls in a context of fiscal dominance. Although administrative controls are used to alleviate the inflationary impact of fiscal dominance and to direct credit resources according to government priorities, they are also motivated by the slow progress in developing money market instruments consistent with Islamic finance principles.
The need for a properly sequenced financial liberalization and the stated objective of reducing inflation call for major changes in the monetary policy framework in Iran. Initial steps in this area could include the development of monetary aggregate targeting. Central bank instrument independence, stringent accountability requirements, and the development of indirect instruments of monetary policy are the major ingredients of success in this area. Although developing liquid money market instruments consistent with Islamic finance principles may be a difficult undertaking, the obstacles are surmountable, as evidenced by the experience of other countries.
Reforming monetary policy alone will not remove inflationary pressures or enhance financial intermediation. The elimination of fiscal dominance; the restructuring of the banking system, with a greater emphasis on private sector participation and competition; and other institutional reforms are key to achieving sustainable low-inflation growth.