Monetary Policy and Operations in Countries with Surplus Liquidity
Monetary policy in most developed countries is conducted with the system being kept marginally short of liquidity. In contrast, many emerging market monetary authorities have been facing surplus liquidity due to factors such as capital inflows, privatisation programmes or fiscal surpluses. This paper provides a cross-country experience of monetary policy and operations conducted in some of these countries and attempts to draw policy inferences concerning policy stances, operating frameworks and procedures in this regard.
MRIDUL SAGGAR
T
I Contemporary Methods of Conduct of Monetary Policy
The 1990s has witnessed a marked transformation in the way in which monetary policy is formulated and implemented. These changes include: (i) a shift in reliance from direct to indirect instruments of monetary control, (ii) a more transparent way of formulation and communication of monetary policy, with a shift from monetary policy surprises to time consistency of policies, and (iii) recognition that a trade-off between inflation and unemployment may not exist, allowing monetary authorities to target low inflation through market operations by seeking to control short-term interest rates.
Direct instruments of monetary control have a one-to-one correspondence between instruments (such as credit ceilings) and objectives (such as credit outstanding). They set or limit either the prices (such as interest rates) or the quantities (such as credit). In contrast, indirect instruments act through markets by adjusting underlying demand and supply of liquidity, operating on bank reserves or on short-term interest rates. Interest rate controls, credit ceilings and directed lending are typical examples of direct instruments, while open market operations and standing facilities in the form of marginal lending or deposit facilities constitute the indirect instruments. Reserve requirements do not fall in any of these classifications in a clear manner. They do not have a direct correspondence between instrument and objective (with the impact depending upon money multipliers which are often not stable), nor do they typically adjust bank reserves through a market process. Nevertheless, reserve requirements are being increasingly considered more as a direct instrument of monetary control on the ground that they are a tax on the banking system.
Alexander, et al (1995) explain that the transition from direct instruments to indirect instruments is a difficult one. It requires the development of primary and secondary markets to support full fledged open market operations. Countries in their sample required an average of 3.7 years for the transition to take place. Seven of the 19 countries in their sample required five or more years to complete this transition. Also, a major factor in the decision to switch to indirect instruments was the initial conditions which set off the transition. It appears that the initial condition of surplus liquidity has been a major trigger, as 79 per cent of the countries in their sample that switched to indirect instruments of monetary control were characterised by surplus liquidity.
Important changes have also occurred in respect of institutional developments relating to monetary policy formulation and implementation. Following Rogoff (1985), central bank independence has found its raison d’ être. A large number of central banks have been provided legal autonomy, generally in terms of setting instruments. Goals are generally set by the governments. Increased independence has been accompanied with improved monetary transparency, reflected in the rate setting exercise being done in the monetary policy committee (MPC) framework. So market based monetary operations have gained added importance as short-term interest rates are sought to be kept in line with the target interest rates or bands set by the central banks or their institutional bodies such as the MPC.
Yet another difference in the conduct of monetary policy has been that the empirical finding of the Phillips curve is no longer taken for granted. There is an increasing recognition that
Graph 1: FDI Inflows to Croatia Graph 2: Impact of Croatian National Bank FX-Interventionson the Exchange Rate
Per Cent
2,000.0 1,800.0 1,600.0 1,400.0 1,200.0 1,000.0 800.0 600.0 400.0 200.0 0.0 Million euros
10/1/9617/5/9620/6/9725/8/9827/1/9922/11/9918/10/019/1/0314/19/05 1,500.0 1,000.0 500.0 0.0 -500.0 -1,000.0 -1,500.0 -2,000.0 2.0 1.5 1.0 0.5 0.0 -0.5 -1.0 -1.5199319941995199619971998199920002001200220032004
Year
FDI
2005(Q1-3)
— — CNB Intervention amount
....
.... Exchange rate change on intervention date relative to
4 days prior (in per cent)
inflation-unemployment (or output) trade-off may not occur in the way envisaged earlier. The Phillips curve could be vertical or horizontal. Following Friedman and Phelps, it is generally accepted that there is no long-run trade-off between inflation and output (vertical Phillips curve). However, King (2005) argues that by credible monetary policy action, which reigns in inflation expectations, it is possible to have a horizontal Phillips curve and indeed the same has been delivered by the Bank of England at the targeted inflation rate. In order to do so, it is necessary to have a credible reaction function that sets the short-term interest rate based on a rule, so that market interest rates react to what the central bank is expected to do. Open market operations are the most practical way to deliver such reaction. Most central banks, however, seek to do so by providing marginal liquidity to banks at an interest rate set by them. Monetary policy, therefore, has increasingly taken the form of setting an official interest rate for the central bank’s dealings with the banking system.
As most central banks through the 1990s shifted to controlling the short-term interest rates, the operating frameworks have changed accordingly. Borio (1997) provides the first authentic documentation of such frameworks used in industrialised countries. However, the application of such a framework to countries in surplus liquidity and the efficacy of such transmission have not been specially examined in the extant literature.
II Managing Surplus Liquidity
In the 1990s, a large number of central banks faced the problem of liquidity surpluses as capital accounts were liberalised and large capital flows occurred from developed to developing countries. While increased capital inflows have been a major source of surplus liquidity in many emerging markets, surpluses could arise from different factors. In case they arise from capital inflows, they are reflected in a reserves build up. This is typically the case when country receives large direct or portfolio investments, or remittances in form of foreign currency bank deposits. Liquidity surpluses may also arise from privatisation receipts. If foreign investors are permitted to participate in the privatisation programme, the liquidity may build up through reserve accretion. Alternatively, if domestic agents provide the privatisation receipt, liquidity surpluses may take the form of fiscal surpluses, which may lead to build up of government cash balances. Besides, monetary financing of government deficits may also result in surplus liquidity.
Managing surplus liquidity poses substantial difficulties if there is a continuous inflow of liquidity, leading to accretion in surpluses, the stock of surplus carried from the past could be absorbed through monetary policy tools such as outright transactions or changes in reserve requirements, but continuous flows can lead to depletion in the stock of securities available with the central bank or increased dependence on repo windows, where the central bank has to roll over large amounts in reverse repo windows. Repo windows are useful for absorbing marginal liquidity and for absorbing enduring surplus liquidity it is better to rely on outright transactions by creation of additional securities, if necessary.
Draining surplus liquidity often poses balance sheet concerns for central banks. If the system is in deficit, the central bank can earn interest on the marginal liquidity it injects, but draining liquidity is a cost to central banks and can have serious repercussions for their balance sheets. A large money market disequilibrium, therefore, may elicit an asymmetric response from the monetary authorities. The central bank may be more eager to provide an additional liquidity than to drain it. In case it chooses to accommodate surplus liquidity for a long time, it could risk sustained losses. The structure of the balance sheet becomes a crucial factor in determining the impact on its balance sheet. In case a central bank does incur large losses, it can potentially impact on central bank autonomy as the central bank seeks to recapitalise with fiscal support. Surplus liquidity may also force central banks to increase their dependence on non-market tools and increase reserve requirements to drain surplus liquidity as an easy way out. A steep rise in reserve requirements could lead to disintermediation. Banks in search of returns could alter their portfolio behaviour resulting in moral hazard of lending to high risk, high return projects or to retail lending, resulting in the build up of household debt financed consumption. If surpluses are large and enduring, they can even have adverse macroeconomic consequences, such as nominal interest rates falling so low as to affect saving behaviour and maybe even the viability of financial intermediation. If the nominal interest rates fall to the zero lower bound (ZLB) as a result of large surpluses, the economy could head for a liquidity trap, making monetary policy ineffective. Even though the likelihood of such an event is low, it is important to note that enduring surpluses in themselves attenuate monetary control with central banks being able to Graph 3: Interest Rates on Credit and Deposit Facilities
of the Croatian National Bank
exercise little control over short-term rates and with the transmission mechanism also turning weak.
III
Recent Experiences of Countrieswith Surplus Liquidity
30.00 25.00 20.00 15.00 10.00 5.00 0.00 Per cent per annumCroatia
Croatia had proclaimed sovereignty along with Slovenia in
1990 seceding from the six-member Yugoslavian state. It followed this up with the declaration of independence in 1991. Stabilising the economy required about four years and the inflation rate as measured by the consumer price index was brought down to about 4.0 per cent in 1995 from a four digit hyper inflation
Jan 1994Jan 1995Jan 1996Jan 1997Jan 1998Jan 1999Jan 2000Jan 2001Jan 2002Jan 2003Jan 2004Jan 2005
of 1910 per cent in 1993. Foreign investment came in a big way once the economy stabilised. Real GDP growth has been around 3 to 5 per cent for last six consecutive years. Croatia is an interesting case study of one of the transition economies where macroeconomic stability has been maintained with low inflation, low variability in output and a stable exchange rate, in spite of a high degree of dollarisation, once stabilisation was achieved. This has been the case in spite of a relatively high rate of monetary expansion arising from large foreign capital inflows. Money market liquidity has been characterised by surpluses. However, overnight interest rates have been quite volatile.
Capital inflows have been dominated by foreign direct investment (FDI). Net FDI inflows started in 1996 and had risen to
1.8 billion euros by 2003, though they moderated somewhat over the next two years (Graph 1). The stock of direct investment in Croatia had risen to 9.4 billion euros by 2004 and, net of such Croatian investments abroad, was placed at 7.8 billion euros. This is in addition to the stock of net portfolio investments of euro
4.4 billion and net other foreign investments of euro 9.7 billion, which are mainly in form of official and banking loans. Thus, there was a net investment position of euro$15.5 billion by 2004 in the economy.
Foreign investment inflows of this magnitude for an economy of Croatia’s size has translated into monetary pressures. Yearon-year growth in money supply increased sharply in 2002 and by August of that year was running at 53.4 per cent (in euro terms), due largely to runaway expansion in international reserves, between mid-2000 and mid-2002. Inflation which had jumped from 3.4 per cent in 1996 to 6.1 per cent in 1998, averaged close to 5 per cent during 2000-2001. It declined, thereafter to an average of about 1.5 per cent during 2002-2003, before climbing again to surpass a 3 per cent mark by 2005, due mainly to pressures on energy prices.
Croatian monetary management experience in recent years has been an interesting one. It is a case where the monetary authorities have pursued the twin goals of exchange rate and liquidity management by employing tools of foreign currency auctions and open market operations. However, its dependence on reserve requirements as a traditional monetary policy tool has been an important factor in the success of overall monetary policy.
On the exchange rate front, Croatia implements a managed floating rate without a peg or basket. The Croatian National Bank (CNB) announces daily values of the kuna against other major currencies and intervenes frequently with a view to prevent excessive fluctuations in exchange rates. The main policy
Months
CNB’s overnight deposit facility
instrument used for the purpose is foreign exchange auctions, through which the central bank purchases or sells foreign exchange to commercial banks. CNB generally leans against the wind. Central bank interventions are of moderate size. In 2003, buying interventions accounted for 1.8 per cent of the total volumes in the market, while selling interventions accounted for
0.3 per cent. CNB lays emphasis on exchange rate stability with a view to maintain price stability, as sharp movements are known to affect exchange rate expectations considerably. The effect of CNB’s forex interventions on the exchange rate has been considerable and can be seen from Graph 2. During May-September 1998, small interventions had a large impact on the exchange rate However, the impact was muted thereafter till the first quarter of 2000. From the second quarter of 2000, large interventions have been seen to generate a significant exchange rate impact (Graph 2). On the whole, since October 1996, there has been a statistical correlation of about -0.70 between the interventions amount and exchange rate changes calculated from four days prior to the interventions. The correlation of these two series improves to -0.78 when worked out from the 2000:Q2. CNB’s balance sheet has got dollarised to a considerable extent, with 95 per cent of its assets being in the form of foreign assets. Another macroeconomic consequence has been that its external debt to GDP ratio has risen from about 25 per cent in 1996 to 81 per cent in 2004.
CNB has been extensively relying on open market operations in addition to foreign exchange auctions to deal with surplus liquidity arising from these large capital flows. It has adopted an operating framework, which has large similarities to that of the European Central Bank (ECB).1 While the ECB keeps the system in deficit at the margin, Croatia has been characterised by large liquidity surpluses. As such, its regular operations are based on weekly one week liquidity absorbing reverse repo transactions. The fine-tuning operations are provided for both reverse repos and repos and are conducted in non-regular manner for non-standard tenors. Outright sales/purchases of T-bills are also used for fine tuning operations. The structural operations are also undertaken in the form of repos / reverse repos or outright purchase/sale of T-bills. Auction procedures are used for all kinds of open market operations, but bilateral procedures are also used for fine-tuning or structural operations.
Graph 4: Foreign Investment Inflows and sometimes marginally breaching them. A major difficulty in conducting open market operations has been problems posed
18000
16000
government deposits. These deposits have frequently fallen to
14000
below 500 million kunas and surpassed 1,500 million kunas at
12000 10000
least on five occasions since 2003. On three of these occasions,
deposits crossed 2,000 million kunas mark. Government deposits
8000
bulge towards the end of the month and the start of the month
6000
4000
due to tax inflows or on occasions of bond issuance, but have
2000
a poor predictability otherwise. Apart from the errors in liquidity
US $ mn US $ mn
forecasting, the open market operations also remain a blunt tool
-
because of the failure of the repo rate to emerge as a reference
interest rate due to illiquidity in the kuna market. The CNB has
Year s
also auctioned some of its own kuna denominated bills through
Graph 5: Monthly Foreign Investment Inflows
3500 3000 2500 2000 1500 1000 500 0
-5 0 0Apr-95 Apr-96 Apr-97 Apr-98 Apr-99 Apr-00 Apr-01 Apr-02 Apr-03 Apr-04 Apr-05
Months
-1000
multiple price auctions at a discount, but the instrument has not been very successful in sterilising liquidity.
As open market operations are still to become an effective monetary policy tool, CNB has continued to depend heavily on reserve requirements as a monetary policy tool. Given the high degree of dollarisation, where residents hold large amounts of cash or deposits in euros, the CNB has chosen to apply reserve requirements separately to the kuna and foreign liabilities.2 The reserve maintenance period in Croatia is one month, starting from second Wednesday. In addition to the reserve requirements, the CNB imposes unremunerated marginal reserve requirements, at a rate of 55 per cent, on the incremental average daily balance of specific funds and specific off-balance sheet liability, which is maintained in foreign currency. Also, special reserve requirements are placed in kunas on incremental average daily balance of issued kuna denominated securities, as also in
foreign currency on incremental foreign currency denominated
Open market operations are supplemented by three types of standing facilities – standing deposit, Lombard loan and intraday loan facilities. An overnight deposit facility was introduced lately in April 2005 and is provided at a low interest rate of a mere 0.5 per cent. These deposits are not eligible for reserve maintenance calculations. Lombard loans are collateralised with a hair cut that provides 90 per cent of the T-bills valuation. They currently carry an interest rate of 7.5 per cent after the rate was lowered from 9.5 per cent, effective December 14, 2005. Intraday loans are also collateralised with the same hair cut, but are provided interest free with end of the day maturity and the limits are linked to the bank’s settlement account during the day. The CNB reserves the right to suspend or limit recourse to any or all of the standing facilities. The CNB also uses the discount rate inside the interest rate corridor as an additional interest rate signal, but the corridor has been generally asymmetric to this policy rate. Before the introduction of the deposit facility, the corridor floor was weak, as only statutory reserves deposited were remunerated at rates which ranged between 4.5-6.0 per cent during May 1994 to February 2001 before being lowered in several steps to the current rate of 0.75 per cent (Graph 3).
Croatian monetary operations have had a limited success due to a number of factors. First, the corridor set has been wide and weak. A corridor of 9.0 percentage points till December 14 and
7.0 per cent thereafter leaves scope for considerable volatility in short-term rates. The call rate has been very volatile, frequently oscillating and gyrating by touching the edges of the wide band securities. In sum, Croatian monetary policy has been a reasonable success mainly due to direct instruments and not monetary policy operations.
India
India’s experience with surplus liquidity is rooted in large capital flows after the capital account was liberalised as part of the wide ranging reforms launched since mid-1991. Starting 1993-94, India has received large portfolio and direct investment flows. These flows were in t he region of US$ 4-6 billion per year during 1993-94 to 1997-98, before the flows dropped next year in the aftermath of the east Asian crisis. The flows renewed once again on the back of credible reforms, restructuring of the corporate sector, and new opportunities in sectors such as information and communication technologies. Total capital inflows have averaged US$ 15 billion a year during 2003-04 and 2004-05, posing serious problems in sterilising them (Graph 4). Not only have these flows been large but, with preponderance of portfolio flows, have become quite volatile. Portfolio flows have accounted for more than two thirds of the total foreign investment flows in the form of direct or portfolio flows and this ratio stands out in comparison with experiences of other countries. High dependence on portfolio flows meant that monetary authorities faced a more serious challenge for monetary management. The graph on monthly foreign investment flows (Graph 5) clearly shows that these flows have been volatile, due largely to volatility in the portfolio component. The Reserve Bank of India (RBI) has been absorbing capital flows for a larger part of the period, except the recent building up of foreign exchange reserves, which
rose from less than a billion US dollars at the time of 1991 crisis Graph 6: Sterilisation through Liquidity Adjustment Facility
and Market Stabilisation Scheme
to over US$ 140 billion.
120000 100000 80000
Average daily absorption in Rs crore
Liquidity surpluses got built up following the huge capital
inflows and monetary accommodation provided through con
scious monetary policy stance of soft interest rate bias practised
to overcome slow industrial growth, which averaged below 5.0
60000
40000
per cent for 12 quarters in succession till Q4 of 1999-00. After
a temporary recovery over the next three quarters, industrial
20000
growth slumped back to below 5.0 per cent over the next five
0
quarters. However, the monetary stimulus helped in industrial
recovery thereafter. Sustained buoyancy was observed, with the
industrial growth rate exceeding 5.0 per cent over the next 14
-20000 -40000
Months
quarters in succession.
While excess liquidity was left in the system, the monetary authorities had to nevertheless undertake sizeable sterilisation
LAF
MSS
Graph 7: Call Money Rates
operations. In 2003, about Rs.47,000 crore of securities were sold 40 in outright transactions to mop up surplus liquidity. As a result, 35 the stock of securities got depleted. In a forward looking assess
30 ment, the RBI took a close look at the options it had to carry 25out sterilisation ahead. RBI (2003) recommended several steps to improve its liquidity adjustment facility (LAF), which was 20 set up in June 2000 as a principal tool for monetary management 15 and constituted daily repos and reverse repo bids to inject or 10 absorb surplus liquidity. RBI (2004) recommended instituting 5 sterilisation bonds in the form of a market stabilisation scheme 0
Dec 23, 1988 May 7, 1990 Sep 19, 1991
Jan 31, 1993
June 15, 1994
Oct 28, 1995
Mar 11, 1997
July 24, 1998
Dec 6, 1999
April 19, 2001
Sep 1, 2002
Jan 14, 2004
May 28, 2005
Oct 10, 2006
Feb 1, 2006
Feb 22, 2008
(MSS). Making a clear choice in favour of issuing government bonds instead of central bank bills, MSS became operational from April 1, 2004.3 Proceeds of the MSS are immobilised in a separate identifiable cash account maintained and operated by the Reserve Bank, which is used only for redemption and/or buyback of MSS securities. Initially, a ceiling of Rs 60,000 crore was fixed on the outstanding amount of MSS, but in accordance with the MOU provisions, the ceiling was enhanced to Rs 80,000 crore (US$17.5 billion) on October 14, 2004. The total outstanding amount absorbed under the MSS had increased to about Rs.80,000 crore in September 2005. It can be seen from Graph 6 that while MSS
Graph 8: Short-term Movements of Money Market Ratesand Liquidity Adjustment Facility
9.00
8.00
7.00
6.00
helped bring down the LAF absorptions, overall liquidity sur
pluses continued to rise and were in excess of Rs 1,00,000 crore
Q1 of 2004-05, before it began to unwind and decline to about
Per cent
5.00
4.00
half the earlier levels by November 2004 (Graph 6). The unwind-3.00 ing of liquidity was largely the result of resurgence of inflation, 2.00 which after having dropped to below the 5.0 per cent mark over the last four years, reappeared in the face of the global oil price 1.00
0.00
shock and even began to feed into manufacturing inflation.
However, RBI changed its monetary policy stance and moved away from easy monetary policy to calibrated hardening, making it clear to the markets that it would prefer to provide “appropriate” rather than “adequate” liquidity. However, large capital flows continued and surplus liquidity again got built up and exceeded Rs 1,00,000 crore for most of 2005-06 till October. As absorption
Apr 1, 2003Jun 1, 2003Aug 1, 2003
Oct 1, 2003
Dec 1, 2003 Feb 1, 2004 April 1, 2004
Jun 1, 2004Aug 1, 2004Oct 1, 2004
Dec 1, 2004
Feb 1, 2005 April 1, 2005 Jun 1, 2005
Aug 1, 2005
Oct 1, 2005 Dec 1, 2005
Call Money Reverse Repo Rates
Market Repo CBLO REPO
picked up in the economy, reflected in all round credit growth interest rates in uncollateralised as well as collateralised segmentscovering industrial as well retail sectors, RBI continued with have been by and large anchored to the reverse repo rates. Thegradual tightening, raising its liquidity absorbing reverse repo call volatility has significantly come down after introduction ofrate from 4.5 per cent in October 2004 to 5.5 per cent by January the LAF (Graph 7).2006. Four hikes of 25 bps along with accelerating credit growth Interest rate corridors have been set by LAF repo/reverse reposucked away most of the surplus marginal liquidity. Anticipating rates, which predominantly have characteristics of standingthese trends, RBI had begun to unwind MSS since September
2005 and about Rs 45,000 crore were released through phased facilities, even while they aid open market operations.4 The floor unwinding of this instrument till February 2006. corridors were breached on occasions during the last two years Monetary operations in India have been reasonably successful when faced with large surplus liquidity, but the overnight call in delivering monetary policy objectives and the short-term rates (uncollateralised), as well as CBLO and market repo rates
(collateralised) remained by and large anchored to the floor set Diagram: Corto System of Banco de Mexico
by the reverse repo rates (Graph 8). As liquidity tightened since October 2005, short-term rates hardened and quickly moved to the upper edge of the band. The IMD redemptions of US$ 7.1 billion during December 27-29, 2005 accentuated this transition. The large build up of government cash and investment balances with RBI also contributed significantly to this tightening. As a result, call money rates ruled above the corridor for most part of January and February 2006 and those in the collateralised segments of CBLO and market repo also hardened close to the upper edge of the corridor. Open market operations have a crucial role in such circumstances, specially when balance sheet considerations are mitigated after the system has transited out of the enduring surplus.
Mexico
=
ChaCha
ngenge
inin 
MM
oo
nene
tt
aa
ryry
baba
ss
ee 
demdem
andand
mm
ee
tt 
by BoMby BoM 
aa
tt 
mm
aa
rkrk
etet
ii
nn
tt
ee
rere
stst 
rara
tt
ee
+
ChaCha
ngenge
inin 
SS
uu
pppp
lyly
ofof 
MonMon
ee
tt
aa
rr
yy 
BasBas
ee
byby
BoMBoM
Open market operations Amount of Corto or short
ChaCha
ngenge
inin 
BoMBoM
’’
ss 
netnet 
domdom
eses
tt
ii
cc 
crcr
eded
itit 
ChaCha
ngenge
inin 
MM
oo
nene
tt
aa
ryry
baba
ss
ee 
demdem
andand
mm
ee
tt 
by BoMby BoM 
aa
tt 
ii
nn
tt
ee
rere
stst
rara
tt
ee
AA
bb
ovov
e me m
aa
rkerke
tt 
levlev
ee
ll
ss
((
tt
wiwi
cc
ee 
the OIR)the OIR)
Graph 9: Banco de Mexico’s Sterilisation OperationsMexico is an interesting case for managing surplus liquidity for Creating Corto
because of its choice of monetary policy instruments. It relied
mainly on ‘Corto’ – the negative target of current account balances
maintained by banks with the central banks – which was directly
targeted as a policy instrument.5 The instrument choice is analogous to the one being currently used by the Bank of Japan (BoJ)
to come out of the liquidity trap, except that BoJ has for long
tried to keep the system in surplus liquidity by flushing money
markets with liquidity through remunerated current account balance
accretions. In contrast, in Mexico, the system has been kept
marginally short of liquidity by the central bank. Currently, there
are indications and financial market speculation that Japan may
End of the year outstandings inmillions of Mexican pesos
300000 250000 200000 150000
100000 50000
0
1999 2000 2001 2002 2003 2004 2005
abandon its current policy of quantitative easing sooner than later. Similarly, Banco de México or the Bank of Mexico (BoM) is expected to move away from Corto to directly target short-term interest rates over a year or two, specially as current reforms to develop the bond market progress. Other interesting experiences of monetary management in Mexico are its conscious effort to avoid foreign exchange reserves build up by adopting informal rules, conducting frequent exchange rate auctions to drain excessive reserves and maintaining a credible floating rate regime by allowing considerable exchange rate flexibility.
On January 18, 2000, BoM introduced a new operating framework of keeping the market short of liquidity by raising the targeted current account balances for banks from 160 to 180 million pesos for the monthly period.6 The design of the monetary policy as aimed to first lower inflation to a targeted level of 10 per cent and then to bring it down further to drift towards its main trading partners (which includes the US) by 2003. The system of managing liquidity in the money markets in Mexico is explained in the Diagram.
The distinguishing feature of Corto, a Spanish word which literally means “short”, is a negative target for daily balances of the commercial bank’s deposits at the central bank. In other words, the BoM supplies all the liquidity needed, except for the amount of the Corto. BoM allows commercial banks to draw negative balances or “overdraft” in their current account balances at the end of the day as long as the overdraft are offset at the end of the 28-day period so that banks register a non-negative balance at the end of the period. Any shortages in the form of negative balances at the end of the period after netting overdrafts are charged interest rates twice the market interest rates. The benchmark for market interest rates was initially the yields on 28-day Cetes, but was later changed to the overnight interbank
Years
rate (OIR). No remuneration is provided to positive balances. The BoM provides normal liquidity through open market operations in the form of daily short-term credit auctions, the quantity of which is fixed by the bank such that the system is brought to the targeted Corto. The operations amount to keeping the markets short of liquidity, even while surpluses otherwise would exist. The shortage is created through long-term sterilisation of excess liquidity by BoM, accepting remunerated deposits from banks for an unspecified maturity and also by BoM issuing debt instruments (Graph 9).
The monetary authorities in Mexico have tightened monetary policy considerably over the last two years, reflected in a larger amount of targeted shortage in current account balances. The Corto has remained at 79 million pesos per day for the larger part of 2005, after being raised, in several small steps, from 20 million pesos in February 2004. Inflation expectations had risen in 2004, following supply side shocks, and the authorities moved in a credible way to anchor headline inflation rate at around 3 per cent, right in the middle of the inflation target band of 24 per cent, by emphasising that the upper half of the band will not be tolerable. By mid-2005, Core inflation as well as inflation expectations had declined. During the restrictive cycle, OIR changed significantly only when either the Corto target was changed or when the Federal funds target was changed, except in the latter half of 2005 when the link to US Fed was abandoned. Monetary conditions have eased in this latter phase without changes in the Corto. Short-term interest rates, which had risen to 6 per cent, have begun to decline. While overnight interest rates have fallen, the Mexican authorities have not changed the
Graph 10: Foreign Direct and Portfolio Investment In Pakistan
(July-Jan)
Year
Graph 11: Trade and Current Account Deficits in Pakistan
official policy instrument – Corto.7 This is seen by a segment of the market as a pre-cursor for abandoning the Corto in favour of interest rate targeting. The operating procedure currently in place has seemed to work reasonably well for monetary policy in delivering lower inflation. However, certain deficiencies are seen in the system. First, while changes in the Corto do indicate whether the monetary policy is being made less or more restrictive, it does not indicate any interest rate target. Second, one level of the Corto may be consistent with more than one level of interest rates, leaving markets somewhat confused. Third, the transmission of the Corto to inflation rates is not well established or understood. Fourth, as the relative amount of targeted Corto is short, the BoM is expected to satisfy a large amount of liquidity through its daily market operations. Fifth, BoM has to sterilise large amounts in order to create the Corto. Sixth, creation of the Corto results in a game theoretic environment in which at least one bank is left short of meeting its current account balance target and is penalised at excessive rates at twice the OIR.
Pakistan
Pakistan’s tryst with surplus liquidity arose in the backdrop of September 11, 2001 terrorist strikes. Prior to the strike, there were meagre flows, mainly in th e form of debt creating capital inflows. The current account remained in deficit and foreign exchange reserve levels were inadequate. The State Bank of Pakistan (SBP), which is the central bank of the country, faced a tough challenge of meeting the govern ment borrowing target, while at the same time stabilising the Pakistan rupee, which was under substantial devaluation pressure, with market expectations for the same clearly reflected in the kerb market rates. The SBP indulged in frequent selling interventions and relied on the discount rate, the cash reserve ratio (CRR) and the statutory liquidity ratio (SLR) to restraint liquidity for monetary policy purposes.
However, post-September 11, 2001 monetary policy was eased and capital flows were liberalised as part of the reforms as a sequel to the IMF programme. FDI was liberalised with permission for 100 per cent equity participation and removal of restrictions on profit repatriation. Liberalisation was not restricted to the financial sector or manufacturing and included even the agricultural, infrastructure and social sectors as thrust points. In addition to FDI flows, workers remittances grew rapidly as non-resident Pakistan nationals and even persons of Pakistani origin preferred to send money to Pakistan as a relatively safe haven for such persons in the post-September 11 period. Also, foreign economic assistance increased as Pakistan became the frontier state in the US war against terrorism. Even with increasing remittances post-September 11, 2001, capital flows remained dominated by foreign economic assistance till 2002-03, a phenomenon which began in 1992-93. However, thereafter, capital flows have been clearly dominated by workers’ remittances.
Post-September 11, 2001, the SBP had to make large buying interventions in the foreign exchange market, which resulted in substantial reserves accretion that required to be sterilised through open market operations. The kerb market premium, which had hovered in 4-7 per cent range between mid-1999 to mid-2001, dipped sharply and has generally hovered in +1 to -1 per cent range since 2002. There were episodes where there were pressures of currency appreciation. The SBP tried to balance by allowing gradual exchange rate appreciation along with sterilisation.
Over the last five years there has been a steady and steep rise in direct investment flows, while portfolio investment has oscillated with outflows in 2001-02 and 2003-04 before picking up in next two years (Graph 10). On the whole, portfolio investment has accounted for only 10 per cent of the foreign investment flows, with direct investment accounting for the other 90 per cent. The SBP has been intervening heavily to sterilise these flows, purchasing foreign exchange in the interbank market and kerb market. The interventions were particularly large in 2001-02 to 2003-04 when net purchases ranged in the region of US$ 3-5 billion a year. In spite of increased foreign investments in 2004-05, the SBP relatively restrained from interventions and allowed greater exchange rate flexibility with total net purchases of about US$ 1.3 billion. Faced with still larger capital inflows, net purchases of about US$ 2.5 billion were made in 2005-06. Foreign exchange reserves were less than US$2.0 billion in 1999-00, but incessant capital inflows saw them surpass the US$ 10 billion mark in 2002-03 to touch US$ 12.7 billion in 2004-05. They have since moderated to just below US$ 12.0 billion. While, Pakistan ran a current account deficit (CAD) for 10 years from 1991-92 to 2000-01, averaging 4.1 per cent of GDP, capital flows sustained current account surpluses over the next four years – 2001-02 to 2004-05. The current account balance (CAB) hit a surplus of 5.7 per cent of GDP in 2003-04 (Graph 11). Trade balance has been in deficit throughout the period starting 1991-92 and has averaged 3.6 per cent of GDP a year. The current account gap has suddenly widened in 2005-06, registering a deficit of 1.9 per cent of GDP. This is largely due Graph 12: Repo Rates in Pakistan to increased absorption reflected in a jump in the trade deficit from 1.6 per cent in 2004-05 to 5.6 per cent in 2005-06.8 Foreign 18
16
investment flows continue to be robust. Recent changes in the
14
balance of payments and resurgence in inflation, which shot up
to 9.3 per cent in 2004-05 after staying below the 5.0 per cent
12
10
Per cent per annum
mark for five years in succession, has however changed the
macroeconomic climate. Furthermore, the accommodative mon
etary policy stance has contributed to an asset bubble creation
in the economy. The SBP has responded by a clear shift in the
8
6
4
monetary policy stance from accommodation to aggressive tight-2 ening since Q2 of 2004-05.
0
The SBP has been using a mix of instruments to sterilise the capital flows. These include open market operations, partly taking recourse to sterilisation bonds in the form of Pakistan Investment Bonds (PIB) launched in October 2003. PIBs have been floated
24/7/1998
15/3/2000
11/5/2001
28/6/2003
17/2/2005
10/10/2006
Dates
as long-term instruments with tenors of 3, 5, 10, 15 and 20 years,
Market Repo SBI Repo
carrying coupon rates in the range of 6-10 per cent. The SBP has also been using shorter tenor open market operations of late,
Graph 13: Inflation and Growth in Russia
using 3, 6 and 12-month T-bills, as with interest rate cycle hardening, demand for long-term paper became thin. Monetary 2000
15
Real GDP growth in per cent per annum
operations in Pakistan have had a limited success. Interest rate
1800 1600 1400
Inflation per cent per annum
1200 1000 -5800
10
volatility remains high with overnight interest rates widely
oscillating in the 2-9 per cent band throughout last year and
discount rates also likewise spiking frequently in the same region.
Repo rates at which market trades have also been very volatile,
though they have responded to the SBP’s repo rate changes, they
remain rather volatile with frequent trades at off-market rates
(Graph 12). Also, the market repo rates have responded, but
generally stayed below the SBP repo rates in a wide zone. Pakistan has reduced its dependence on direct instruments of monetary control since 2002-03 onwards, but it would require considerable efforts to make its open market operations a principal and ef
5
0
600
-10
400
-15200
-20
0
Year
fective tool of monetary management. As these tools develop, it may also require to keep the exchange rate more flexible, as currently, it is pursuing the twin targets of exchange rate and monetary stability.9
GDP Growth
Inflation Rate GDP growth
to changing economic and financial circumstances.11 However,
it must be said that the Russian economy has transformed itself by stabilising the inflation rate and recording good growth in recent years (Graph 13).
Russia
The monetary policy framework in Russia is based on a monetary The Russian experience with managing surplus liquidity is of targeting framework under which inflation is explicitly targeted.
special interest not just because of its position in the world economy and political order, but also because of its innovative use of a wide array of sterilisation instruments of which establishment of the stabilisation fund is the most important one. After independence in 1991, the Russian Federation had a long history of difficult reforms which included hyperinflation and debt legacy problems. It inherited $69 billion of foreign debt from the Soviet period. By 1995, chronic inflation totalled 2,00,000 per cent. Between 1991 and 1995, the official gross domestic product (GDP) contracted (negative growth) every year and every quarter. Industrial production also fell every quarter during 1992-95. The big reform of January 1992 was the direct source of macroeconomic deterioration. In March 2002, the chairman of the central bank of the Russian Federation (CBR), Viktor Gerashchenko had to resign and was replaced by Sergei Ignateiv.10 There was a view in that period that the CBR did not understand the basic principles of money and banking and did not know how to operate the standard tools of monetary policy in a way that will bring about price stability and create an economic climate conducive to growth. The CBR was said to be simply captive The broad money aggregate, M2, is used as an intermediate target. Russia has tried moving to setting interest rate policy for monetary policy and exchange rate objectives with limited success. In the aftermath of the 1998 debt crisis, the refinance rate was set at 60 per cent. In 2000, this rate was cut five times to bring it down to 25 per cent. In 2001, it was reduced further to 21 per cent. However, refinance operations were formalised only at the start of 2003. Lack of marketable securities also prevented the CBR from carrying out open market operations till 2003. So, for all practical purposes the Russian experience with market-based monetary policy operations began in 2003. Somewhat strangely, the CBR also did not use reserve requirements as an active tool of monetary policy till 2003, adjusting it only twice during 19992003 to increase it by 2-3 percentage points on various types of liabilities.
In 2003, the first half was characterised by a high level of surplus liquidity, which the CBR tried sterilising through deposit operations and reverse repo modified operations. The latter absorbed 141 billion ruble (US$3.7 billion). But by May 2003, the repo was no longer attractive as a sterilisation instrument, and of the 6-8 auctions that were announced only 1-2 were carried Graph 14: MIACR and CBR’s Interest Rate Corridor out over the next few months. The introduction of market-based
monetary operations in 2003 has, however, contributed to lowering short-term interest rate volatility. The volatility of the overnight interbank rates has been traditionally quite high in Russia. Graph 14 plots the weighted Moscow interbank actual credit rate (MIACR). It can be seen that there have been frequent spikes in the range of 20-55 per cent during 1999-2003. These spikes have come down after the CBR established an interest rate corridor between its credit and deposit facilities. The former is on an overnight basis and the latter is on a tom next basis. The rates breached the corridor on several occasions in 2004, but the volatility has been well contained thereafter, though in a wide band that remains large at nearly 12 bps even after some shrinking.
The system was in huge liquidity surpluses during Q1 and Q4 of 2004, though liquidity provisions were required during Q2 and Q3 of 2004. The tightening of liquidity during Q2 and Q3 of 2004 was largely due to appreciation of the US dollar, which
MIACR CBR overnight credit rates CBR tom next deposit rates Per cent per annum 60.00 50.00 40.00 30.00 20.00 10.00 0.00 Jan 1, 1999 May 1, 1999 Sep 1, 1999 Jan 1, 2000 May 1, 2000 Sep 1, 2000 Jan 1, 2001 May 1, 2001 Sep 1, 2001 Jan 1, 2002 May 1, 2002 Sep 1, 2002 Jan 1, 2003 May 1, 2003 Sep 1, 2003 Jan 1, 2004 May 1, 2004 Sep 1, 2004 Jan 1, 2005 May 1, 2005 Sep 1, 2005
Dates
resulted in increased demand for foreign exchange and reduction in budgetary and extra-budgetary balances maintained at CBR. Reflecting the underlying balances, MIACR averaged 1.3 per cent in the Q1, hardened to 7.3 per cent in Q2, before dropping back to 3.5 per cent in Q3 and 1.1 per cent in Q4. The increase in overnight interest rates in Q2 was accentuated by money market frictions caused by segmentation – not only regionally, but also within Moscow, where redistribution of short-term surpluses suffered on account of increased lending risk problems. On many days the overnight inter-bank rate surpassed the CBR’s overnight loan rate. The CBR, faced with sudden tightening of liquidity conditions and the failure of money markets to clear the same, resorted to cuts in reserve requirements. The required reserve ratio on corporate liabilities in rubles or in foreign currency, as also on individual liabilities in foreign currency, was reduced from 10 per cent to 9 per cent on April 1, to 7 per cent on June 15, and further to 3.5 per cent on July 8. During June and July 2004, excess reserves of more than 150 billion rubles were released by the CBR through an unscheduled regulation. Since July 1, 2004, reserve requirement averaging was allowed within an averaging rate of 0.2. Reserves averaging also improved the liquidity of the credit institutions, releasing an estimated 9 billion rubles from the required reserves. With improved liquidity, the CBR floated its own bonds (OBR) worth 34.5 billion rubles in September 2004; the first such placement since 2001. In the subsequent period, the CBR went for open market operations on both sides, buying and selling bonds at yields ranging from 4.5-5.5 per cent. In the process, the CBR could improve its holdings to enable it to conduct open market operations to address future volatility. The CBR also broadened the collateral for its Lombard facilities and extended this facility to many more Russian regions.
In 2005 again, there were sharp swings in liquidity conditions, with large surpluses continuing through Q1 of 2005, followed by relative tightening in Q2. In Q2, free funds of the banking sector shrunk and some of the banks borrowed from CBR. CBR also opened two way quotes for open market operations in its own bonds, improving the liquidity of its instrument in the secondary market and enhancing its sterilisation capabilities. These OBR market operations were carried at moderately submarket yields of 3.5-4.5 per cent with a view to further push the money market rates southwards. The MIACR hardened towards
Graph 15: Current Account Surpluses in Russia
Year
the end of Q2 of 2005 and again towards the end of October 2005, when the average rates were seen in the 8-12 per cent band, very close to the upper edge of the CBR interest rate corridor.
The main source of enduring surplus liquidity in the Russian economy has been foreign investment inflows and oil export receipts, resulting in a sustained current account surplus (Graph 15). Current account surpluses now exceed US$ 100 billion and have averaged over 11 per cent of GDP over the last six years. The sheer challenge of sterilisation for the monetary authorities has been huge. The CBR, which was founded only in 1990, has been intervening in the foreign exchange markets to lean against ruble appreciation. This has resulted in a sharp accretion in international reserves, which have reached nearly US$ 188.5 billion by the end of January 2006. The CBR has attempted to drain the excess liquidity so flowing in order to contain inflationary pressures, but faced limits on instruments for sterilisation. As a result, sterilisation has been partial and a substantial amount of unsterilised interventions have exacerbated monetary pressures.
In 2003 the Russian government conceived the creation of an oil stabilisation fund (OSF), partly to finance social welfare activities and to meet emergency expenditures, such as those on flood relief and partly to build a buffer against a possible oil price reversal in the future.12 The law for the establishment of such a fund stipulated that the funds could only be used after its corpus reached 500 billion rubles (US$17.3 billion). The fund was to be financed by extra revenues coming from the Russian oil price exceeding US$ 20 per barrel. It became operational at the start of 2004 as part of the federal budget. Unspent fiscal surpluses amounting to 106 bn rubles in 2003 and 218 bn rubles in 2004 were also transferred to the OSF. By the end of 2005 its corpus had reached 1.46 trillion rubles (US$ 52.2 billion). The fund is currently being used to prepay part of the foreign debt, including that to the IMF and to the Paris Club. Russia also intends to use it to finance the gap in pension funding. Though, initially the IMF had opposed the establishment of a stabilisation fund in Russia, the fund has actually become a very effective instrument for imparting long-term stability to the federal budget as well as for sterilising excess liquidity in the Russian economy for monetary policy purposes and has now got a global acceptance. Sustained budget surpluses could have been a major source of surplus liquidity in Russia, but the OSF has helped dampen their impact. The fiscal surpluses have also helped lower the cost of borrowing for the Russian government as yields have continued to fall in the GFK-OFZ bond markets for a long time. Against the background of budget surpluses, time and again the Russian fiscal authorities have slowed down the government borrowing programme, further affecting the yields in the GKO and OFZ bond market.
While OSF has reduced the pressure on monetary policy, CBR continues to use several other instruments to provide or suck liquidity from the system. It has been using a range of instruments to provide liquidity to the banks whenever needed. These include free of charge intra-day loans, overnight repo auctions conducted twice daily, in the morning and the afternoon. In addition, the CBR conducts weekly longer-term repo auctions for seven days and 90 days tenors. It also conducts one and two-week Lombard loans auctions. At the end of the day, banks can access standing facilities such as overnight loans and currency swaps at rates equal to refinance rates. Deposit auctions have been a major source of absorbing surplus liquidity, but the CBR has also been making active use of open market operations through outright sales of bonds in the GKO-OFZ market. From time to time, the CBR has been using longer-term repos, such as the operations in February 2004, when modified reverse repo auctions were made for tenors longer than four months in the absence of comparable instruments of sterilisation before the CBR. The CBR hopes to bring down the inflation rate in Russia over the next three years in 5.0-6.5 per cent band.
Notwithstanding, noticeable improvement in monetary management in Russia, the economy remains susceptible to highly volatile cross-border capital flows. With the economy slowing down a bit in 2005, the risks may have increased somewhat. As Russia increases capital account liberalisation, it is necessary that the CBR allows increased exchange rate flexibility, something which it has been reluctant to do so far. It finds that with monetary expansion predominantly taking the form of net foreign assets, it becomes necessary to target interest differentials for exchange rate management and as such the efficacy of the interest rate as a monetary policy tool becomes limited. Russia has set a goal for full capital account liberalisation by 2007 and in this context would need to maintain a stable ruble in its run up to this event. Even if the difficult tasks of maintaining exchange rate and monetary stability is carried out over the next two years, a successful transition to a fully convertible ruble could invite another round of large capital inflows, which could be bigger in size and pose new difficulties for the CBR to sterilise these flows.
IV Policy Inferences and Suggestions
The diverse experiences of the five countries are illuminating in many ways. Croatia faced surplus liquidity largely on account of FDI inflows. Though it has moved from direct to indirect instruments to control excess liquidity, the transition is far from complete. The CNB is forced to rely on direct instruments like reserve requirements. The Croatian example tells us that forcing the pace of transition in favour of open market operations without developing a complete institutional mechanism of requisite standing facilities could hamper success. India’s experience has been far more successful and is clearly seen in the lowest volatility of short-term rates amongst the sample countries. However, the framework is again hampered by legal and institutional constraints. Occasional spikes suggest that further improvements could be attempted by reducing the segmentation in the different overnight market segments and by targeting open market operations towards the middle of the corridor. Nevertheless, considering the source of liquidity lying in large and volatile portfolio flows, monetary operations, specially through the sterilisation bonds in form of MSS, can be considered a reasonable success. The mexican experience of keeping marginal liquidity short through open market operations aimed at Corto is innovative in many respects. While it did help in averting some of the ill-effects associated with the surplus, the operating framework provides confusing signals and the transmission channels become blurred. Pakistan’s experience with surplus liquidity arising from large direct investment flows has been one of limited success. In the absence of effective open market operations, it may be useful to more directly focus on monetary policy, if necessary by relaxing competing exchange rate objectives. The Russian experience has been a mixed one. The CBR’s efforts to develop a monetary operations framework on the lines of developed countries has helped in bringing down interest rate volatility. However, the magnitude of foreign flows has been extraordinarily large and constraints on monetary policy are very much in evidence, suggesting that market operations are not a substitute for addressing the fundamental disequilibrium through monetary policy actions. The Russian experience also suggests that the innovative use of monetary policy instruments could be helpful. The stabilisation fund appears to be a preferable alternative instrument to sterilisation bonds, which could be considered by countries which face surplus liquidity amidst fiscal surpluses.
The experience sighted above also suggests certain general guidelines for effective monetary operations. First, it is important to clearly understand the sources of liquidity and tailor instruments accordingly. Second, effort to improve liquidity forecasting could contribute a long way to managing surplus liquidity and to conduct monetary operations in line with the monetary policy stance and parameters. Several countries including Croatia and India have faced serious problems arising from unpredictability of these flows. It is also important to understand seasonal patterns, as also the day-of-the week effects in liquidity flows. Third, if surplus liquidity is enduring and an imbalance is expected to persist for sometime, maybe about a year or even more, it is best that the bulk of the adjustment is undertaken by changes in monetary policy instruments such as the policy rates or reserve requirements. If there are compelling reasons which limit the use of main monetary policy instruments, it would be best to use market instruments such as volume based open market operations. The load of adjustment in such cases would have to be on outright transactions to adjust the bank reserves. Foreign exchange swaps are more useful instruments under conditions and when liquidity imbalances are expected to persist at best for a quarter or so. Repos as an instrument for market operations is best suited for short-run adjustments where liquidity mismatches are of a marginal nature.
Fourth, if central bank holdings of securities emerge as a constraining factor for open market operations, issuing of additional bonds for sterilisation purposes becomes a compulsion. In such cases there is a choice whether additional securities are issued as central bank bills or as government bills. The choice depends on a number of factors of which the nature of the existing central bank balance sheet is a critical factor. However, on balance it is better if such bonds are issued by government rather than central banks. The Indian experience is the best testimony to this. Fifth, greater the volatility in capital flows or in surplus liquidity arising from other sources, greater is the advantage in issuing short-term bills rather than long-term securities. Short-term bills, such as T-bills up to six-month tenor provides the monetary authorities greater degrees of freedom in liquidity management as they could roll over these instruments or unwind depending upon how the liquidity situation transforms. However, short-term instruments come with greater issuance cost and uncertainty about how markets would respond at the time of the issuance. As such, issuing long-term bonds in case of sustained enduring surplus could have its merit, specially if the secondary debt market is liquid and the central bank indulges in active open market operations on a day-to-day basis enabling it to buy or abstain or sell securities if swings in liquidity conditions occur. The RBI’s experience of issuing short-term MSS securities, in contrast with the SBP’s experience with long-term PIBs makes an interesting case for comparison. Sixth, an important element for successful monetary operations to combat surplus liquidity is the central bank developing control over short-term interest rates, inter alia, by lowering their volatility. Rogers (2005) and Mac Gorain (2005) discuss several policy issues in this regard, such as allowing reserves averaging. Taking into account the experiences provided in this paper, as also those of developed countries documented elsewhere, it can be said that it is important to develop interest rate corridors in the first place. This could be done by putting in place effective overnight standing deposit and credit facilities, of which the deposit facility may or may not be collateralised, while the lending facility should necessarily be collateralised. The standing facilities should ordinarily provide a limited recourse, but towards the end of the reserve maintenance period it should allow banks to borrow funds through repos on eligible securities, so that targeted reserves are attained on a reserves averaging basis without unduly disturbing the market rates. Open market operations should preferably be conducted in the middle of the interest rate corridor around a credible single policy rate in a narrow undeclared range of ±25 bps of the policy rate, while the corridor itself could be wider with width ranging from two times to four times the target range of market operations. Two things are necessary for this. First, marginal liquidity should be kept small and enduring surpluses or deficits should be corrected through monetary policy and operations using outright transactions and other instruments. Second, it is necessary to over time evolve a credible single policy rate in the middle if it does not exist, such as that in case of India, so that demand for reserves becomes flatter in a small segment around this rate. Fixed rate repos could then become a more effective tool for adjusting marginal liquidity and could be generally preferred to variable rate repos as the latter can be read as signals away from the policy rates. Discretionary open market operations could become useful in case it is not found feasible to contain interest rates in this segment. However, it is a second best option. Also, monetary operations should be contained at the short-end of the yield curve, so that the information content of the yield curve is not lost. The range of facilities developed by developing countries has been rather limited due to balance sheet constraints arising from absorption of surplus liquidity and lack of technical expertise in some cases. As a result of this, monetary operations had limited success in these countries. It is worthwhile to experiment with operating procedures as there is no common practice which can serve all at all times. Innovative measures such as cash management of government balances through auctions or absorptions in stabilisation funds have been helpful innovations for many.13
Lastly, it must be said that the first best policy choice in most cases should be appropriate exchange rate adjustment, so that the need for sterilisation is minimised. Calvo and Reinhart (2000) provide evidence that a lot many developing countries fear floating. It is important to understand that sterilisation at times could lead to increase in domestic interest rates which can lead to sustained capital inflows over and above the absorption capacity of the economy. The choice to sterilise or not is often a “devils alternative”. If the monetary authorities do not sterilise and the interest rate falls due to large enduring liquidity surpluses, it can fuel an asset bubble in the bond market. As portfolio adjustment takes place with refinancing of loans and mortgages, the financial surpluses of the household sector and corporate incomes rise, which can translate into an equity bubble as well. The monetary authorities then face the dilemma of giving weight to asset price considerations in monetary policy. Soft-landing becomes a crucial issue, as any bubble burst could have a large impact on the real variables such as consumption, saving, investment, employment and output, not only because of the wealth impact on some of these variables, but also because the value of collaterals fall, impacting the ability to obtain bank loans. In any case, protracted sterilisation could be hazardous. Mohanty and Turner (2005) point out that such policies pose risks to monetary balance and financial stability, besides entailing large costs.
An important element in macroeconomic policy choices in case of large capital flows is to avert falling into a trilemma trap or the impossible trinity. Enough literature exists in the Mundell-Fleming tradition to convince macroeconomists that it is impossible to simultaneously achieve free capital mobility, a fixed exchange rate and monetary policy independence. Yet, history is replete with examples of countries falling into problems of trying to achieve all at the same time. Countries on a credible fixed exchange rate system sacrifice their monetary policy independence. However, such an option may not be suitable for countries which face asymmetric shocks. For small economies, the shocks are often transmitted from abroad and could be symmetric to those faced in the countries with whom they choose to peg their exchange rate. This need not be the case with large economies – large in a geographical sense, large in terms of domestic markets or large in terms of the economic power they command in global markets. If such economies have a low degree of openness, there is less of a policy dilemma, as capital controls make it possible to maintain monetary policy independence with exchange rate stability. However, possible gains from risk sharing that the capital account liberalisation brings remain unrealised.14 If, however, countries liberalise and open their capital account, policy choice becomes focused on the choice between a fixing exchange rate and coordinating monetary policy in sync with the country of the peg currency or allowing exchange rate flexibility and practising monetary policy according to requirements of the domestic economy. For large economies facing asymmetric shocks, who seek to benefit from capital account liberalisation, a very practical policy option is to allow considerable exchange rate flexibility and substantial capital account liberalisation, which falls short of complete capital account liberalisation and allows for occasional interventions in foreign exchange markets if there is large volatility or fundamental disequilibrium in exchange rate. Calibrating monetary policy to the domestic economy could be placed on the top of the hierarchy in policy choices for such economies.
rr;Email: mksaggar@hotmail.com
Notes
[The author is grateful to Simon Gray, Chandan Sinha, Roger Clew, Kazi Abdul Muktadir, Riaz Riazuddin and Shaghil Ahmed for clarifying some finer points relating to issues covered in this paper. The views expressed are strictly personal and should not be attributed to any of the persons acknowledged or to the institution to which the author belongs.]
1 ECB conducts four types of operations: (i) main liquidity providing 1week repos conducted once a week, (ii) liquidity providing long-term operations once a month using three-month repos, (iii) fine-tuning operations, as and when required, in a non-standardised manner of liquidity injections through quick tenders or absorptions through bilateral procedures, with a view to smooth and steer interest rates in face of unexpected liquidity fluctuations, and (iv) liquidity providing repos on a regular or non-regular basis for non-standard tenors using standard tenders in case of a structural mismatch of liquidity in the eurosystem.
2 The current reserve requirement ratio is 17 per cent. Half of the reserve requirements on foreign liabilities is required to be met in kuna, while half is maintained in euros or in US dollars. The kuna component of reserves maintenance is remunerated at the rate of 0.75 per cent. Reserves maintained in euros and US dollars are remunerated at half of the ECB and US Fed policy rates, respectively.
3 The main consideration in issuing government bonds in preference to central bank bills was to avoid competing securities with risk free characteristics which could distort yield curve, segment the market and affect liquidity of either securities. The balance sheet considerations may have also played some role.
4 Legal constraints arising from Sec 17 of the RBI Act have come in the way of setting up of effective standing facilties. However, LAF proxies the objective of standing facilities in a very liberal way, with the quantum generally limited by only the availability of collateral.
5 BoM has also established a floor to the monetary conditions. 6 The target was raised over the years to 700 million pesos by March 2003 before the targets were set for daily balances. 7 As BoM relaxed monetary conditions, the OIR fell by 75 bps to 9.0 per cent in the third quarter of 2005.
8 Increased absorption has been led by a retail credit boom, with housing loans growing by 20 per cent and auto finance as well as personal loans growing by 38 per cent each in FY05.
9 On this aspect, see Husain (2005).
10 Gerashchenko was earlier chairman of CBR in the first half of 1990s, during which the macroeconomic record was perhaps even more dismal. However, his second term, which began in September 1998 after the Russian default of August 1998, came under intense global scrutiny prompting Jeffrey Sachs to describe him as the “the world’s worst central banker”.
11 See Hoover Institution public policy inquiry on the Russian Economy at http://www.imfsite.org/.
12 The concept of stabilisation fund was discussed even earlier. See for instance, P Kadotchnikov (2002): ‘Establishing a Stabilisation Fund in Russia’,Russian Economic Trends, January 2002, 11(1): pp 8-17. However, it was formalised in official circles in 2003 after Norway has set up such a fund.
13 Apart from Russia, Azerbaijan, Kuwait, Qatar and some other countries have also benefited from stabilisation funds.
14 While gains exist from capital account liberalisation, it is important to understand that they could be of second order, when compared with trade account liberalisation. Gray and Woo (2000) cautions against the received wisdom of seeking indiscriminate capital flows to lower the cost of borrowing. Edwards (2001) provide empirical evidence to suggest that capital account openness and productivity performance only manifest after a country in question reaches a certain degree of development.
References
Alexander, W E, T J T Balino and C Enoch (1995): ‘The Adoption of Indirect Instruments of Monetary Policy’, IMF Occasional Paper, No 126, Washington DC.
Borio, Claudio (1997): ‘The Implementation of Monetary Policy in Industrial Countries’,BIS Economic Papers, No 47, Bank of International Settlements.
Calvo, G A and C M Reinhart (2000): ‘Fear of Floating’, NBER Working Paper, No 7993.
Clew, Rogers (2005): ‘Implementing Monetary Policy: Reforms to the Bank of England’s Operations in Money Market’, Bank of England Quarterly Bulletin, Summers, pp 211-20.
Edwards, Sebestian (2001): ‘Capital Mobility and Economic Performance: Are Emerging Economies Different?’, NBER Working Paper, No 8076.
Gray, Simon and David Woo (2000): ‘Reconsidering External Financing of Domestic Budget Deficits: Debunking Some Received Wisdom’, IMF Policy Discussion Paper, DPD/00/8, July.
Husain, Ishrat (2005): ‘Monetary-cum-Exchange Rate Regime: What Works Best for Emerging Market Economies’, keynote address by governor at SBP conference on monetary and exchange rate regime held at Karachi on November 14.
Kaminski, Garciela L (2005), ‘International Financial Flows, Financial Stability and Growth’, DESA Working Paper, No 10, December.
King, Mervyn (2005): ‘Monetary Policy: Practice Ahead of Theory’, Mais Lecture 2005, May 17, Cass Business School, London.
Kohli, Renu (2001): ‘Capital Account Liberalisation: Empirical Issues and Policy Issues-I’, Economic and Political Weekly, April 14, pp 1201-06.
Mac Gorain, Seamus (2005): ‘Stabilising Short-term Interest Rates’, Bank of England Quarterly Bulletin, Winter, pp 462-70.
Mohanty, M S and Philip Turner (2005): ‘Intervention: What Are the Domestic Consequences’ in Foreign Exchange Market Intervention in Emerging Markets: Motives, Techniques and Implications, BIS Paper, No 24, May.
RBI (2003): Report of the Internal Group on Liquidity Adjustment Facility, December, Reserve Bank of India, Mumbai.
– (2004): Report of the Working Group on Instruments of Sterilisation (chairman: Usha Thorat), Reserve Bank of India, Mumbai.
Riazuddin, Riaz (2005): ‘Sterilisation of Capital Flows in Pakistan’, SAARCFINANCE seminar on conduct of monetary policy and management of capital flows, August 29-31, National Institute of Banking and Finance, State Bank of Pakistan, Islamabad.
Rogoff, Kenneth (1985): ‘The Optimal Commitment to an Intermediate Monetary Target’, Quarterly Journal of Economics, 100, pp 1169-89.
