Managing monetary and financial stability in a

dynamic global environment: Bank Indonesia’s

policy perspectives

Juda Agung, Solikin M Juhro, Harmanta, Tarsidin1

Abstract

As the Indonesian economy is becoming progressively more integrated with the

global economy, the impact of global economic shocks on the domestic economy is
becoming more pronounced. Capital inflows, which trigger excessive liquidity and
exacerbate the risk of a sudden reversal, pose a serious threat to the Indonesian
economy, especially in terms of financial stability. Recent crisis episodes have
indicated that monetary policy alone is insufficient to maintain macroeconomic
stability; it should be accompanied by macroprudential policy. This paper explores
the dynamics of the external and financial sectors as well as the optimal policy mix
in order to maintain monetary and financial stability. We use an enhanced or
modified small open-economy New Keynesian model to discuss the operation of a
flexible inflation targeting framework (ITF). The simulations show that the model’s
impulse response functions are in line with theoretical and empirical predictions, in
which external shocks have significant impacts on both monetary and financial
stability. The simulations also show that the adverse macroeconomic and financial
effects of external shocks can be mitigated by a mix of monetary and
macroprudential policies.
JEL classification: E17, E52, F47.
Keywords: inflation targeting framework, monetary policy, macroprudential policy,
policy instrument mix.

1
Juda Agung is the Head of the Economic and Monetary Policy Department; Solikin Juhro is the Head of
the Monetary Policy Group of the Economic and Monetary Policy Department; Harmanta is a Senior
Economist and Tarsidin is an Economist at the Economic and Monetary Policy Department – Bank
Indonesia. The authors thank Idham and Aditya Rachmanto for their excellent research assistance. The
views expressed in this paper are solely those of the authors and do not necessarily represent the
views of Bank Indonesia.

BIS Papers No 88
1
1. Introduction

The Global Financial Crisis (GFC) of 2008–09 showed that keeping inflation in check
is not, by itself, sufficient to preserve macroeconomic stability. Several crisis
episodes over the past decade have shown that most macroeconomic instability
stems from shocks in the financial sector. Financial markets are inherently prone to
excessive procyclicality, which ultimately manifests itself in macroeconomic
instability. In addition, risk-taking behaviour among economic agents also
strengthens financial accelerator mechanisms.
Amidst global economic shocks and more dynamic capital flows, high
procyclicality in the financial sector in many emerging markets requires that
monetary policy and macroprudential policy be coordinated in order to mitigate
excessive economic fluctuations. On the one hand, conventional monetary policy
has the potential to bolster financial system stability through its influence on
financial conditions and behaviour in financial markets, even if it is focused on
financial stability. On the other hand, macroprudential policy is designed to directly
ensure financial stability. Given the interactions between them, it is important to
adopt a flexible monetary policy regime that can accommodate both monetary and
financial system stability. In the case of Indonesia, this takes the form of a flexible
inflation targeting framework (ITF), one that is constructed to take account of the
wisdom gained from the unconventional monetary policy in the post-GFC era.
In the context of a small open economy, global financial market integration and
large capital flows complicate the implementation of monetary policy. There has
been a tendency for monetary authorities to shift their preferences from “corner
solutions” to “middle solutions” to the classic open economy trilemma, particularly
in developing countries. It is widely argued that the policy response should manage
exchange rate movements within a certain range (without adopting full flexibility)
and restrict capital flows, in addition to targeting domestic inflation. A flexible ITF,
incorporating a mix of monetary and macroprudential instruments, can
accommodate a compromise between the three intermediate goals of (1)
maintaining monetary policy autonomy: (2) stabilising exchange rates; and (3)
managing capital flows.
In practice, to optimally support the implementation of flexible ITF in Indonesia,
the Bank Indonesia Forecasting and Policy Analysis System (FPAS) uses a model that
captures interactions between the financial sector and the real sector as well as the
dynamics of the external sector. Bank Indonesia’s response to financial and external
sector shocks necessitates a mixture of monetary and macroprudential policy tools.
To do this requires that we further develop Bank Indonesia’s macroeconomic model
(ARIMBI).2 In future, ARIMBI is expected to capture the dynamics of the financial and
external sector more fully, thereby improving the accuracy of policy simulations and
projections through the FPAS.
This study aims to explore the linkages between monetary and financial
stability, especially in the context of a dynamic global environment; to simulate
policy and analyse several external shocks to the Indonesian economy and their
implication for both monetary and financial stability; and to search for an
optimal policy mix in

2
ARIMBI is a semi-structural New Keynesian model adopted from the IMF’s Quarterly Projection Model
(QPM), as further developed by Harmanta et al (2013, 2015) and Wimanda et al (2013).

2 BIS Papers No
response to global economic dynamics. We use a novel modelling approach, in
which the financial sector is highly susceptible to financial accelerators. We consider
a number of key variables, including real credit volume growth, the spread between
lending rates and deposit rates as well as the banking sector’s default risk. The
macroprudential policies included in the model are loan-to-value (LTV) policy as
well as the reserve requirement (RR). As regards the external sector, we focus on the
current account (CA) gap and the capital flow (CF) gap. The model is then used to
simulate Bank Indonesia’s policy response to a number of shocks and explore the
implications for optimal policy.
We find that the flexible ITF is well suited to managing monetary and financial
stability in Indonesia. Using the framework, Bank Indonesia can mitigate the impact
of external shocks as well as shocks to the exchange rate, current account and
capital flows, while simultaneously maintaining both monetary and financial
stability. In addition, the integration of monetary and macroprudential policies
provides better results in terms of mitigating excessive output and credit
fluctuations, as compared with any single policy instrument. We conclude that, for
the Indonesian economy, flexible ITF is superior to the standard ITF.
The paper is structured as follows. Section 2 of this paper presents the
dynamics and challenges of the post-GFC Indonesian economy. Section 3 discusses
Bank Indonesia’s policy framework for managing monetary and financial stability,
emphasising the flexible ITF. Section 4 concludes.

2. Dynamics and challenges of the post-GFC Indonesian

economy

The GFC provided a number of valuable lessons, including illustrating that

maintaining price stability alone through monetary policy is insufficient. In addition
to price stability, financial system stability is also a prerequisite for macroeconomic
stability. And, in line with increasing economic openness and integration, the
external sector requires considerable attention.

2.1. The post-GFC challenges

As a small open economy, Indonesia faces a number of challenges in the

implementation of monetary policy relating to persistent capital flows arising from
quantitative easing (QE) in advanced economies. From Q3 2009 to Q2 2011, these
inflows precipitated rupiah appreciation and a widening current account deficit. An
open capital account, coupled with an influx of capital flows, ensured that capital
flows, rather than the current account, predominantly determined exchange rate
behaviour. Accordingly, capital inflows drove nominal rupiah appreciation of 15.9%
in 2009 and 4.5% in 2010. In real terms, the value of the rupiah appreciated by
17.8% in 2009 and 11.4% in 2010, even though the currency remained relatively
competitive compared with those of some other Asian countries. Combined with the
end of the commodity supercycle and a growing middle-income population in
Indonesia, rupiah appreciation contributed to a current account (CA) deficit that
surpassed 4.27% in the second quarter of 2014.

BIS Papers No 88
3
 Capital flows, exchange rate, and inflation Graph 2.1

Second, capital flow volatility created financial system vulnerability. Capital

flows that fluctuated widely, amid ubiquitous herding behaviour, might reverse
suddenly if market sentiment changed. They also threatened to increase financial
market volatility and, in turn, act as a shock amplifier. Such consequences were
further exacerbated by weak infrastructure and a lack of financial deepening, as is
often the case in developing countries such as Indonesia. Furthermore, a significant
portion of the capital inflows was invested in short-term financial instruments, such
as SBIs, government bonds (Surat Utang Negara/SUNs) and stocks, which are
particularly vulnerable to sudden reversals. As the Federal Reserve began to “taper”
in January 2014, domestic liquidity shrank. Investors withdrew their money from
emerging markets, including Indonesia, and switched their investments to US
markets.

Credit and GDP growth Graph 2.2

Third, financial sector procyclicality was amplified by foreign capital inflows. The
influx of capital drove more liquidity into the banking system and more credit was

4 BIS Papers No
channelled to the real sector. Credit growth induced overheating in the economy.
As a result, an asset price bubble emerged, especially in housing prices. The
financial sector tended to exacerbate economic fluctuations. In Indonesia,
procyclicality is reflected in the performance of bank credit during expansionary and
contractionary phases. Observing credit growth during periods of expansion and
contraction revealed the magnitude of procyclicality in the Indonesian banking
system. Risk behaviour also contributed to procyclicality in the financial sector.
Optimism about the Indonesian economy and diminishing concerns about the Fed’s
tapering may have contributed to high portfolio investment in 2014.

2.2. The optimal policy response

Persistent foreign capital inflows undermine the efficacy of monetary management,

given that measures to manage liquidity in the economy, such as an interest rate
increase, could ultimately be offset by the sheer magnitude of the capital inflows. To
manage upward exchange rate pressures, high capital inflows demand intensive
intervention, which causes the amount of excess liquidity in the banking system to
increase significantly. Such capital flow dynamics could reduce the degree of
autonomy in monetary policy and shift its orientation from a sole focus on inflation
control towards mitigating rupiah appreciation through intensive intervention.
The orientation of monetary policy in the midst of high global uncertainty is
tactically directed towards not only controlling inflation but also to managing
exchange rates in line with macroeconomic fundamentals through active
intervention in the foreign exchange market. In addition, it simultaneously manages
international reserves at a safe level in accordance with best international practice.
This has the logical consequence that exchange rate dynamics will not be
completely influenced by market forces but also by domestic monetary policy.
Post-GFC challenges have revealed some valuable lessons for monetary policy
implementation in Indonesia. First, the multiple challenges facing monetary policy
imply that Bank Indonesia should employ multiple instruments. In the face of capital
flows, while the exchange rate should remain flexible, it should also be maintained
in such a way that the exchange rate is not misaligned from its fundamental value.
Concomitantly, measures are required to accumulate foreign exchange reserves as
self-insurance given that short-term capital flows are particularly vulnerable to a
sudden stop. In terms of capital flow management, a variety of policy options are
available to deal with the excessive procyclicality of capital flows, especially short-
term and volatile capital. In terms of monetary management, the dilemmas have
been partially resolved by applying a quantitative-based monetary policy to support
the standard interest rate policy instrument. In addition, macroprudential policies
aimed at maintaining financial system stability should also be adopted to mitigate
the risk of asset bubbles in the economy.
Second, while price stability should remain the primary goal of Bank Indonesia,
the GFC showed that keeping inflation in check is not, by itself, sufficient to preserve
macroeconomic stability. A number of crises in recent decades have also shown that
macroeconomic instability is primarily rooted in financial crises. Therefore, the key
to managing macroeconomic stability is to manage not only the imbalance of
goods (inflation) and externalities (balance of payments) but also imbalances in the
financial sector, such as excessive credit growth, asset price bubbles and the cycle of
risk- taking behaviour in the financial sector. In this regard, Bank Indonesia
would be

BIS Papers No 88
5
effective in maintaining macroeconomic stability if also mandated to promote
financial system stability. Hence, the monetary policy framework of ITF requires
enhancement by including the substantial role of the financial sector.
Third, exchange rate policy should play an important role in the ITF of a small
open economy. Under a standard ITF, Bank Indonesia would not attempt to manage
the exchange rate. This benign view argues that the exchange rate system should be
allowed to float freely, thus acting as a shock absorber for the economy. However,
in a small open economy with open capital movements, exchange rate dynamics are
largely influenced by investor risk perception, which triggers capital movements. In
this environment, there is a case for managing the exchange rate in order to avoid
excess volatility that could push the exchange rate beyond a level conducive to
achieving the inflation target.
Based on the aforementioned rationale, there is a justification for implementing
a less rigid ITF, otherwise known as flexible ITF. Flexible ITF requires monetary and
macroprudential policy to be integrated, including capital flow management and
exchange rate policy. The policy mix should be an optimal response to tackling
multiple challenges in managing monetary and financial stability.

The policy mix under several circumstances Graph 2.3

: tight : loose

The formulation of an optimal policy mix in Indonesia depends on what kinds

of shocks hit the economy. A fall in world GDP would elicit an accommodative
monetary policy response and looser macroprudential measures. An increase in
global interest rates would be followed by tighter monetary and macroprudential
policy. Meanwhile, a broader current account deficit would require tighter monetary
policy and looser macroprudential measures. On the other hand, capital outflows
would require raising

6 BIS Papers No
 the policy rate and looser macroprudential measures. As Indonesia faces multiple
challenges, for which there are multiple shocks, the formulation of a policy mix is
significantly more complex. In Graph 2.3 we describe the policy mix of Bank
Indonesia under specific circumstances.

Macroprudential measures in Indonesia Table 2.1

No Measure Objectives
1 Minimum holding period on BI bills To “put the brake” on short-term and speculative
capital inflows and mitigate the risk of a sudden
reversal.
2 Lengthen auctions and offer longer To enhance the effectiveness of domestic liquidity
maturity of BI bills. management, including capital inflows, by locking
investments into the longer term and helping
develop domestic financial markets.
3 Non-tradable rupiah term deposits for To lock domestic liquidity into the longer term and
banks limit the supply of BI bills on the market.
4 Limits on short-term offshore  To limit short-term and volatile capital inflows.
borrowing by banks  To limit FX exposure of the banking system
stemming from capital inflows.
5 Mandatory reporting of foreign To increase dollar supply.
exchange originating from export
earnings
Primary rupiah reserve requirement
6 To help absorb domestic liquidity.
(checking accounts held at BI)
7 Secondary rupiah reserve requirement To absorb liquidity and to strengthen the banking
(checking accounts held at BI, SBI and system.
government bonds)
8 FX reserve requirements of the banks  To strengthen FX liquidity management, and
thereby banking system resilience, in the face of
increasing FX exposure stemming from capital
inflows
 To help absorb domestic liquidity.
9 LDR-based reserve requirement To absorb domestic liquidity and enhance liquidity
management at banks without exerting negative
impacts on lending that is needed to stimulate
growth.
10 Loan-to-value (LTV) ratio for the To control accelerating credit growth in consumer
property sector and downpayments sectors (especially the property and automobile
on automotive loans sectors).
11 LTV for second and third properties To slow the rate of increase of credit risk
concentration in the property sector and to foster
prudential principles.

As a result of the global financial crisis that hit the global economy in 2008–09,
Indonesia’s GDP growth dropped to 4.6% in 2009, while nominal credit growth fell
to its lowest level, namely 5%. Under such circumstances, it was optimal for Bank
Indonesia to lower its policy rate in order to catalyse economic activity, while
loosening macroprudential measures (required reserve ratio (RR)). From 2010–12,
however, as the economy strengthened and inflation was well managed, Bank
Indonesia maintained a low policy rate. Regarding credit growth, which skyrocketed

BIS Papers No 88
7
to around 25%, macroprudential measures (loan-to-value ratio (LTV)) were
tightened in 2012. Furthermore, rapid credit growth was spurred by an influx of
capital into the country as investors regarded Indonesia as a prospective investment
destination. To curb credit growth, Bank Indonesia continued to tighten
macroprudential measures in 2013 by regulating LTV policy for second and third
properties and by raising the secondary RR. Despite decelerating GDP growth, Bank
Indonesia raised its policy rate as inflation increased on volatile food and
administered prices. From 2014–15, Bank Indonesia maintained a high policy rate in
order to control inflation. Simultaneously, Bank Indonesia loosened macroprudential
measures (LTV and RR) to stimulate waning credit growth that had sunk to 10%. At
the time, the LTV policy was targeted on specific sectors, such as property, so that
the divergent stances of macroprudential policy and monetary policy did not
confuse the market (by conveying misleading signals). Such conditions are evidence
of the advantages of macroprudential tools, which clearly require the support of
good policy communication. Table 2.1 presents a number of macroprudential
measures implemented by Bank Indonesia, while Appendix 1 presents the same but
in chronological order.

3. Framework for managing monetary-financial stability

Bank Indonesia currently implements a de facto flexible inflation targeting

framework (ITF) as its policy framework. It is an enhanced framework, given that the
Indonesian economy is confronting multiple challenges and that merely achieving
the inflation target is insufficient. The framework requires monetary and
macroprudential policy to be integrated, which is believed to be the optimal
response from a monetary and financial stability viewpoint.

3.1. The framework

Bank Indonesia has operated an inflation targeting framework (ITF) since July 2005.
This is a “standard” ITF. Bank Indonesia perceives ITF as a reliable monetary policy
strategy, although capable of further enhancement by refining the future ITF
implementation strategy. There are two rationales for this enhancement. First,
evaluations of ITF implementation in Indonesia have evidenced the requirement for
a number of adjustments and refinements, which have been undertaken according
to the conventional monetary policy wisdom. In this case, there is justification for
implementing a less rigid ITF as an ideal format for the Indonesian economy.
Second, Indonesian economic performance during the GFC instilled confidence
concerning the aptness of ITF as a reliable monetary policy strategy for Indonesia.
However, considering the dynamics and complexity of challenges faced, the
framework requires further enhancements.

3.1.1 Integration of monetary and macroprudential policy

The macroeconomic stability attained during the Great Moderation of 1987–2007
did not protect the global economy from the impact of a crisis propagated by
financial sector fragility. This experience suggests that monetary policy should
anticipate macroeconomic instability risk stemming from the financial system, and
that financial system stability is the foundation for a sustainable macroeconomic
environment.

8 BIS Papers No
 Within this policy perspective, the central bank requires flexibility in responding
to emerging uncertainties within the economy. Such flexibility is crucial in
overcoming the potential conflicts or trade-offs between targeting monetary
stability and financial system stability. It can be achieved through, among other
means, additional instruments (in this case macroprudential policies) and by
extending the horizon for attaining the inflation target in order to accommodate
near-term output stabilisation. To overcome potential policy conflict, it is also
important to prioritise the policy goal, for example, by setting price stability as the
overarching aim.
The pressing need to strengthen the monetary and financial system stability
framework requires a strong financial infrastructure coupled with an effective
supervisory function. In this regard, Borio (2003) emphasises the need to strengthen
the regulatory framework or macroprudential policy, thereby limiting the risk that
prolonged financial markets instability would undermine real economic output.
Conceptually, macroprudential policy aims at enforcing financial system
stability as a whole, instead of the wellbeing of individual financial institutions.
“Macroprudential policy seeks to develop, oversee and deliver an appropriate policy
response to the financial system as a whole. It aims to enhance the resilience of the
financial system and dampen systemic risks that spread through the financial
system” (G30). In maintaining the stability of financial intermediation,
macroprudential policy is thus a key factor in backing the monetary policy goal of
price and output stability.
Especially after the 2008–09 crisis, many central banks have applied
macroprudential policy instruments more broadly. Consequently, several
instruments previously considered to be microprudential (such as loan-loss
provisioning requirements or loan-to-value) or monetary instruments (such as
reserve requirements) have been utilised to curb systemic risk and maintain financial
system stability. Rather than focusing on efforts to deal with risk at individual banks,
such policy instruments have encompassed a wider macroprudential perspective.
Strengthening the monetary and financial system stability framework requires
appropriate monetary and macroprudential policy integration. It is generally
accepted that the main goal of monetary policy is to maintain price stability.
Accordingly, central banks traditionally use interest rates as their primary instrument
to attain that goal. Maintaining price stability, however, is still not sufficient to
guarantee macroeconomic stability because the financial system, with its procyclical
behaviour, triggers excessive economic fluctuations. Meanwhile, the goal of
macroprudential policy is to safeguard overall financial system resilience in a bid to
support financial intermediation in the economy as a whole. With its countercyclical
role, macroprudential policy supports the goal of monetary policy by preserving
price and output stability.
The objectives achieved through monetary and macroprudential policies should
be mutually reinforcing. Steps to reinforce financial system resilience will also
strengthen monetary policy, by protecting the economy from sharp fluctuations in
the financial system. On the other hand, macroeconomic stability will lessen the
vulnerability of the financial system, with its procylical characteristics. Therefore, the
interest rate may not require adjusting to the extent that would be needed in the
absence of policy integration or coordination. Meanwhile, macroprudential policy
affects credit supply conditions and, consequently, monetary policy transmission.
The efficacy of policy coordination relies on the macroeconomic environment,
financial conditions, the intermediation process and the level of capital and assets in
the banking system. Hence, it is not realistic to expect the combination of monetary
BIS Papers No 88
9
 and

1 BIS Papers No
macroprudential policy to fully eliminate economic cycles. The main goal of such
policy integration is to moderate cycles and bolster financial system resilience at a
macro level.
Several conditions are required to ensure that monetary and macroprudential
policy integration runs smoothly. First, there is a need to understand the framework
of linkages amongst monetary, macroprudential and microprudential policies. This
is to take into account potential trade-offs when pursuing policy objectives. That is
why the use of an instrument mix or adding new instruments can be considered
desirable. Second, there is a need to understand the workings of monetary and
macroprudential policy transmission in terms of catalysing economic activity. This
requires a more integrated analytical framework, especially when evaluating the
important role of the financial sector. Third, there is a need to measure appropriate
risk behaviour indicators in monitoring system risk. Measuring the risk indicators in
addition to supporting the right monitoring system will also strengthen the analysis
of transmission mechanisms through the risk-taking channel.

3.1.2. Managing the monetary policy trilemma

The purpose of a flexible ITF is to manage the monetary policy trilemma (as
presented in Graph 3.1), namely to achieve three intermediate goals as follows: (1)
maintaining monetary policy autonomy in achieving price stability by employing a
monetary and macroprudential policy (instrument) mix; (2) stabilising the movement
of the exchange rate in line with its fundamental value by employing exchange rate
management; and (3) managing capital flow dynamics to support macroeconomic
stability by implementing capital flow management.

Bank Indonesia monetary policy trilemma management Graph 3.1

Monetary and macroprudential policy (instrument) mix

Maintain monetary policy autonomy in achieving price stability

Interest rate, to provide a signal and manage inflation expectations;
Macroprudential instruments to manage liquidity and prevent financial sector risks.

Exchange rate management Capital flow management

Stabilise exchange rate movements in line with its fundamental value Manage capital flow dynamics in supporting macroeconom
Foreign exchange intervention to reduce short-term volatility; Macroprudential to manage capital flows and prevent externa
Promoteand
Strike an optimal balance between providing space for appreciation/ depreciation financial deepening
managing of the
adequate foreign
foreign exchange
exchange market;
reserves.
Support foreign exchange reserves management as a form of

BIS Papers No 88
11
 There are five principles of enhancement, as follows:
a. Continuing the adherence of a policy framework to the inflation target as the
overriding objective of monetary policy. The main characteristics of an ITF will
remain, ie pre-emptive, independent, transparent and accountable policy
implementation.
b. Integrating monetary and macroprudential policy. Appropriate monetary and
macroprudential policy integration is required in order to buttress monetary
and financial system stability.
c. Managing the dynamics of capital flows and exchange rates. To support
macroeconomic stability, coordinated implementation of a policy instrument
mix is ultimately part of an important strategy to optimally manage the
monetary policy trilemma.
d. Strengthening the policy communication strategy as part of the policy
framework. Policy communication is no longer merely for the sake of
transparency and accountability but also serves as a monetary policy
instrument.
e. Strengthening Bank Indonesia and government policy coordination. Policy
coordination is crucial, given that inflation stemming from the supply side
creates the majority of inflation volatility.
Monetary policy complexity stemming from the interest rate can be partially
resolved through quantitatively tighter monetary policy by raising the reserve
requirement. In addition, macroprudential policy aims to avoid financial risks, such
as asset bubbles and excessive credit growth, which could trigger potential financial
system instability. This type of macroprudential policy is effective if banks
intermediate the majority of capital flows. Nevertheless, if the capital flows originate
directly from unregulated sectors, such as direct loans from the private sector,
measures to control capital inflows are another option, for example, by limiting
private loans.
In terms of the exchange rate, the rupiah should be managed to remain flexible,
with scope to appreciate/depreciate, but the currency should also be managed so
that it avoids misalignment with the economic fundamentals, as this will jeopardise
macroeconomic stability. Consequently, Bank Indonesia’s presence is required on
the foreign exchange market to ensure that the rupiah does not incur excessive
volatility. Of course, this option is no longer available if the rupiah becomes
overvalued. Simultaneously, efforts to accumulate foreign exchange reserves are
vital as a form of self-insurance, given that short-term capital flows are particularly
vulnerable to the risk of a sudden reversal.
Regarding capital flows, by continuing to adhere to a free foreign exchange
regime, macroprudential measures also consist of policy options designed to reduce
excessive short-term capital flows, which could potentially lead to financial risks
from the external side. Such measures have been introduced by Bank Indonesia
through regulations that require investors to hold Bank Indonesia Certificates (SBI)
for a minimum period of one month. This policy has helped diversify foreign
portfolio capital flows and extend the duration of SBIs, which consequently nurtured
financial deepening, especially of the foreign exchange market.
The coordinated implementation of a policy instrument mix is ultimately part of
an important strategy to manage the monetary policy trilemma in the current

1 BIS Papers No
uncertain climate. Coordination is critical, not only to address sources of external
and internal imbalances but also to optimally manage the impact of monetary
policy.
According to the above policy perspective, the achievement of macroeconomic
stability is tied not only to monetary stability (price stability) but also to its
interaction with financial system stability. Under a flexible ITF, the flexibility of policy
implementation is achieved through macroprudential instruments in addition to
monetary instruments that are mutually reinforcing. While monetary instruments are
utilised to influence monetary variables, such as the interest rate, exchange rate,
credit and expectations, macroprudential instruments are utilised primarily to
manage potential risk or risk perception in financial markets. Concerning the
measures to overcome potential policy conflict, it is imperative to prioritise policy
objectives by setting price stability (inflation) as the overriding objective.
Graph 3.2 shows schematically how the monetary framework under a flexible
ITF can be enhanced through a mix of monetary and macroprudential policy
instruments.

Monetary policy framework under a flexible ITF Graph 3.2

In response to the aforementioned challenges, the tasks faced by Bank

Indonesia are becoming increasingly complex, particularly in terms of maintaining
financial system stability. Consequently, Bank Indonesia strives to consistently
implement a flexible ITF. This is achieved in the form of macroprudential policy in
addition to monetary policy (interest rate). Concerning macroprudential policy,
Agung (2010) recommends monetary and macroprudential policy be conducted
within the confines of the same institution considering the close interconnectedness
between the two, in this case Bank Indonesia. Furthermore, at the practical level,
Agung (2010) recommends several alternative macroprudential instruments for
Bank Indonesia, namely countercyclical CAR, forward-looking provisioning (so that
when a bank is appropriating reserves, expected losses are also included), the LTV
ratio (as an upper limit for credit to asset value that can be offered to a borrower)
and the reserve requirement (RR).

BIS Papers No 88
13
 Meanwhile, for a small open economy such as Indonesia, the exchange rate
plays a central role in the economy. Monetary policy is transmitted partly through its
impact on the exchange rate. Changes to the policy rate will influence the rupiah
exchange rate through interest rate parity (IRP). Raising the policy rate (which
subsequently increases deposit rates) will cause the rupiah to appreciate and vice
versa. Furthermore, changes in the value of the rupiah will have direct pass-through
and/or indirect pass-through effects on exports, imports, GDP and inflation.

3.2. Modelling a flexible ITF for the Indonesian economy

Here we use an enhanced or modified small open-economy New Keynesian model

to explain the flexible ITF and how the policy mix works. We then present policy
simulations on the impact of external shocks on the Indonesian economy, especially
the impact on monetary and financial stability, and the Bank Indonesia response
using monetary and macroprudential policy.

3.2.1. Modelling strategy

Some recent literature explored the integration of monetary and macroprudential
policy using quantitative models. Galati and Moessner (2011) state that there is lack
of clarification and consensus regarding a definition of financial stability and
effective models to explain interactions between the financial system and
macroeconomy. A selection of the literature tries to include financial frictions in the
corresponding models, in this context relating to credit constraints of loans and
non-financial sectors, which are built based on the financial accelerator mechanism
of Bernanke et al (1996). Furthermore, efforts have also been taken to include
financial frictions relating to financial intermediaries.
Angelini et al (2011) argued that macroprudential policy is expected to have a
direct and indirect influence on the monetary policy transmission mechanism. Based
on their research, it was found that incorporating macroprudential policy is most
beneficial when the economy experiences shocks stemming from the money market
or households, where both types of shock affect the supply of credit. As suggested
by Gerali et al (2010), banks accumulate capital from retained earnings and strive to
maintain a capital-to-assets ratio close to that of the regulated target. According to
Angelini et al (2011), using capital requirements as a macroprudential policy tool is
based on the argument that systemic crises affect bank capital and the supply of
credit. Capital requirements increase when economic conditions are good and,
conversely, decrease when economic conditions deteriorate.
Beau et al (2011) identified circumstances where monetary policy and
macroprudential policy had a complementary, independent or conflicting effect on
price stability. Their findings, amongst others, showed that the best results for price
stability were achieved by combining monetary policy focused on price stability with
macroprudential policy centred on credit growth. Such a policy mix generates
several types of Taylor rule, namely the plain vanilla Taylor rule (using the standard
Taylor rule to achieve the overarching goal of price stability) or the augmented
Taylor rule (to the original Taylor rule is added the argument that short-term
nominal interest rates must be raised in line with stronger credit growth).
Independent macroprudential policy can use the augmented Taylor rule
accompanied by separate macroprudential policy.

1 BIS Papers No
 Cúrdia and Woodford (2009) found a need to accommodate the response to
variations in aggregate credit in the Taylor rule, with explorations based on a New
Keynesian model with financial friction. Monetary policy should be used to help
stabilise aggregate private credit by tightening policy during periods of abnormally
robust credit growth and, conversely, by loosening policy when credit contracts.
Efforts to model macroprudential policy and incorporate it into monetary policy
were also undertaken by Peñaloza (2011), namely by adding a financial block to a
standard semi-structural small open-economy Neo Keynesian model. The
simulations benefited greatly from the inclusion of macroprudential tools (in this
case the CAR rule), enabling the monetary authority to better mitigate output gap
shocks, as compared with just using the standard Taylor rule. Therefore, financial
shocks could be isolated and their adverse impact on macroeconomic variables
alleviated. In this context, the financial block basically represented a set of reduced
form equations that facilitate analysis of lending spreads, the delinquency index and
credit volume, which can be integrated into the core model. Such a model
accommodates the feedback effect from the core model to the financial sector.
Regarding capital flows, Unsal (2011) states that the challenge to policymakers,
concerning the influx of capital flows, is preventing the domestic economy from
overheating with implications for inflation, as well as mitigating the risks associated
with the impact on financial stability, which would be undermined as credit and
financing became more accessible. Monetary policy could be utilised to overcome
the effect on inflation; however, macroprudential policy is required to mitigate the
impact on financial stability. According to Capistrán et al (2011), emerging
economies face the very real threat of a capital flow reversal.
Juhro and Goeltom (2012) state that, in response to capital flow dynamics,
amidst inflationary pressures, Bank Indonesia should implement unconventional
policy using multiple instruments. The framework applied is a flexible ITF, where the
overriding objective is the inflation target. However, a flexible ITF is more flexible
than its standard counterpart. The central bank is focused not only on achieving the
inflation target but also takes into account a number of other considerations,
including financial sector stability, the dynamics of capital flows as well as the
exchange rate. With such a policy perspective, the achievement of macroeconomic
stability is not only related to monetary stability (price stability) but also to financial
system stability.
The importance of balance between the current account and capital flows is
highlighted by Ghosh et al (2008), who focus on five cases: (i) conditions where
capital inflows respond to the CA financing requirement; (ii) conditions where
capital inflows are merely due to higher yields; (iii) conditions where pressures
emerge in the balance of payments due to a current account surplus; (iv) conditions
where the current account surplus is offset by capital outflows; and (v) pre-crisis and
crisis conditions (that transpire due to a current account deficit and/or capital
outflows that are not offset by capital inflows and/or a current account surplus). The
illustrations developed by Ghosh et al (2008) reveal ideal current account and
capital flow conditions, where both are found in a state of equilibrium. According to
Lee et al (2008), based on a macroeconomic balance approach, there is a certain
level of exchange rate in line with CA norms, known as the equilibrium real
exchange rate.
In general, macroeconomic models utilised by countries adhering to an
inflation targeting framework tend to institute monetary policy based on the Taylor
rule. The basic version stipulates that a central bank only responds to changes in the
inflation gap and output gap. Another version specifies that in addition to
BIS Papers No 88
15
responding to both

1 BIS Papers No
Structure of Bank Indonesia’s core model (ARIMBI) Graph 3.3

Interest
Rate
Spread

Default Credit
Risk Growth

Macro-
LTV, RR pruden
tial

Expected World Inflation

Real Inflation Inflation
GDP
Interest
Growth
Rate

Domestic Taylor Real

Interest Rate UIP Expect Rule Exchan
ed
ge
_ Apr./De
+
=
World Risk Premium Curr
Interest Rate ent
Acco

Nomin World
al GDP Growth
Exchan

Capital
Flows

Macro
Risk

BIS Papers No 88
17
the aforementioned gaps, a central bank also responds to exchange rate dynamics,
marked by the inclusion of a variable for the exchange rate in the Taylor rule.
Despite its inclusion in the Taylor rule, the exchange rate is not a policy instrument.
In that context, monetary policy is implemented solely through the Taylor rule. As
stated by Taylor (2001), there are several research papers dedicated to the inclusion
of the exchange rate in the monetary policy rule, including Ball (1999), Svensson
(2000) and Taylor (1999).
To support implementation of a flexible ITF, Bank Indonesia developed several
macroeconomic models for use in its Forecasting and Policy Analysis System (FPAS).
The core model used to make forecast and policy simulations is known as ARIMBI.
Besides ARIMBI, there are a number of supporting satellite models, namely SOFIE
(short-term forecast of GDP components and inflation by category), MODBI
(medium-term forecast of macroeconomic variables), BIMA (short-term forecast of
balance of payments) and ISMA (short-term forecast of sectoral GDP). In addition,
near-term forecasts of GDP, inflation and exchange rates are also provided based
on assessments and anecdotal information.
In the following section, we use equations from the ARIMBI model to explain
Bank Indonesia’s flexible ITF. Originally, ARIMBI was a standard small open-
economy New Keynesian model, consisting of four main equations, namely IS –
output gap, inflation – New Keynesian Phillips Curve (NKPC), Uncovered Interest
Parity (UIP) and the Taylor rule. When the flexible ITF was introduced, we modified
the model to incorporate financial accelerators, as well as procyclicality between the
real and financial sectors and the risk-taking channel. We further enriched the
model by including the ability to capture the dynamics of the current account and
capital flows. In addition, the policy mix of Bank Indonesia is also modelled,
including its monetary policy (Taylor rule and optimal exchange rate) and
macroprudential policy (LTV rule and RR rule). The structure of the model is
presented in Graph 3.3, with further elaboration provided in Appendix 2.

Block 1: The real sector and monetary policy

There are four main equations, namely IS – output gap, inflation – NKPC, the Taylor
rule and Uncovered Interest Parity (UIP). The four equations represent a
macroeconomy or real sector. The output gap represents the size of the disparity
between real GDP and its potential level. The credit growth gap is added to this
equation to reinforce the correlation between the macroeconomy and the financial
block. Meanwhile the second equation shows that CPI inflation is determined by its
expected value, output gap and real exchange rate gap. The Taylor rule is
determined by its long-term trend, inflation gap and output gap. The UIP equation
shows that it holds when the interest rate differential is the same as the summation
of expected nominal exchange rate depreciation/appreciation and risk premium.

Block 2: The financial block and macroprudential policy

There are three equations in the financial block, namely the credit growth gap
equation, the interest rate spread gap equation and the default risk gap equation.
Three additional equations were included because when an economy experiences a
boom/bust episode, real credit growth increases/decreases, accompanied by an
increase/decrease in default risk. Meanwhile, the inclusion of the interest rate
spread gap equation is required to capture the dynamics of lending rates, given that
the core equations do not include the lending rate as a variable. There are
two

1 BIS Papers No
macroprudential tools in the model, namely the LTV and RR ratios, which are
modelled together. In the macro model, the macroprudential instrument
mechanism of the RR resembles the LTV. The current LTV regulation has a direct
effect on mortgages and automotive loans, and ultimately influences total credit.
The reserve requirement affects total credit through its impact on loanable funds. In
the model, both macroprudential tools respond to total credit.

Block 3: The external block and exchange rate policy

There are three equations in the external block, namely the current account (CA)
gap equation, the capital flow (CF) gap equation, and several equations
representing the rest of the world. As mentioned previously, the CA gap is the
difference between the CA to GDP ratio and CA norms. Meanwhile, the CF gap is the
difference between the CF to GDP ratio and the optimum level of CF. The rest-of-
the-world equations consist of world IS – output gap, world inflation – NKPC and
the world Taylor rule. It is a simple model of the global economy and a
representation of what central banks do in response to shocks of world inflation and
GDP. Bank Indonesia’s exchange rate policy is basically a combination of responses
to current economic conditions and a drift towards gradually bringing the economy
to its internal and external balance. If there are no other shocks in the near term
(about one to two years ahead), the path of the short-term fundamental exchange
rate will be the same as path of the medium- term fundamental exchange rate. The
path resembles the concept of permanent equilibrium exchange rate (PEER), in
which there are responses to both temporary and permanent shocks.

Block 4: Macro risk and the risk-taking channel

In order to capture the role of risk perception in the model, we endogenise variables
of risk, using the International Country Risk Guide (ICRG) index as a proxy. The risk is
called macro risk to represent risk at the macro level. A higher output gap would
induce lower macro risk, while a higher inflation gap would raise macro risk. On the
other hand, real exchange rate depreciation would raise macro risk in a similar way
to a deteriorating current account. In the financial sector, higher default risk would
escalate macro risk. The determinants of macro risk are basically composed of
macroeconomic and financial variables. Furthermore, macro risk influences other
variables in the model. Its impact affects not only real exchange rate
depreciation/appreciation but also the credit growth gap (or risk-taking channel),
default risk gap, risk premium and capital flow gap.

3.2.2. Policy simulation

In this policy simulation, some external shocks are simulated, namely a shock to
world GDP, world interest rate, the current account and capital flows. We
differentiate two scenarios in the simulations as follows: (i) Bank Indonesia only uses
monetary policy in response to the shocks (indicated by the broken red line); and (ii)
Bank Indonesia utilises both monetary and macroprudential policy (indicated by the
solid dark blue line).

a) A decline in world GDP

The slowdown in world GDP growth in 2010–13 had a significant impact on the
domestic economy. Using the model, we simulate a shock in the form of a 1% drop
in the world output gap, accompanied by declines in both world inflation and the

BIS Papers No 88
19
world nominal interest rate. Such a shock would precipitate a decrease in the output
gap (economic growth) of Indonesia by around 0.10%, followed by a lower rate of
inflation, prompting Bank Indonesia to lower its policy rate based on the standard
Taylor-type rule mechanism.
A decrease in the world output gap would have a 0.20% impact on the current
account deficit due to a larger decline in exports (stemming from the decrease in
the world output gap and rupiah real exchange rate appreciation) than imports
(because of the decrease in the output gap of Indonesia). Meanwhile, a falling
output gap and current account as well as escalating default risk would trigger a
limited increase in macro risk and also impact the nominal and real exchange rates
as well as other variables.
A more pronounced decline in the world nominal interest rate (in response to a
drop in world output gap) compared to the BI rate, coupled with the inherent lag
associated with reducing the BI rate, would trigger capital inflows to the domestic
economy, thereby increasing the capital flow (CF) gap by around 0.18%.
Consequently, the rupiah would appreciate at the onset of the shock but
subsequently depreciate as the falling world output gap starts to influence
macroeconomic variables, for instance, through domestic economic moderation, a
current account deficit and BI rate reductions.

IRF – World output gap shock Graph 3.4

: Monetary policy only

: Policy mix

A decline in the output gap would subsequently lead to slower real credit
growth due to procyclicality and the presence of a financial accelerator, which could
also be attributable to a wider interest rate spread caused by higher default risk, in
line with

2 BIS Papers No
 economic moderation. Real credit growth would also be suppressed slightly due to
declining liquidity as a result of the ensuing balance of payments (BOP) deficit. In
response, Bank Indonesia would need to raise the LTV ratio and lower the RR to
maintain financial stability.
It can be observed from the simulation that the integration of monetary and
macroprudential policy is superior in terms of slowing the pace of credit growth, as
compared with using just one of the policies. It is unnecessary to lower the policy
rate dramatically to boost the economy, or even to aggressively raise LTV or lower
RR in order to spur credit growth. Implementing the two policies simultaneously
necessitates only moderate shifts.

b) World interest rate increase

Normalisation of the Fed’s monetary policy stance would compel other central
banks to raise their own policy rates. Using the model, we simulated a shock in the
form of a 1% increase in the world interest rate. The shock would induce a decline in
the world output gap and lower world inflation. The shock would also spur an
outflow of capital from Indonesia and cause the rupiah to depreciate, both in
nominal and real terms. Nominal rupiah depreciation would bring higher CPI
inflation and prompt the central bank to raise its policy rate. On the other hand, real
exchange rate depreciation would close the CA gap and subsequently boost the
output gap. Furthermore, real credit growth would increase in the first quarter as
the output gap increased, causing Bank

IRF – World interest rate shock Graph 3.5

: Monetary policy only

: Policy mix

BIS Papers No 88
21
Indonesia to lower the LTV ratio and raise the RR. Here we see that Bank Indonesia
responds to the world interest rate shock with an appropriate policy mix.
Accordingly, monetary policy is directed towards stabilising domestic inflation, while
macroprudential policy would be used to control credit growth. The simulations
show that using a policy mix would lessen fluctuations amongst economic and
financial variables.

c) Widening current account deficit

Bank Indonesia faces the challenge of a large current account (CA) deficit, peaking
at more than 4% of GDP in the previous period. A 1% drop in the CA gap (widening
CA deficit) would undermine economic growth as the current account (CA)
represents net exports, which is a component of GDP. Weaker net exports would
clearly undermine GDP and the shock would also slow real credit growth, appearing
as an impact of the decline in the CA gap on liquidity and as a result of a lower
output gap. Ultimately, a decline in real credit growth would prompt Bank Indonesia
to raise the LTV ratio and lower the RR ratio. Meanwhile, mounting macro risk due
to the shock would cause the rupiah to depreciate and contribute to higher
inflation, prompting Bank Indonesia to raise its policy rate. That combination of
outcomes would also precipitate capital outflows. The impact of a drop in the CA
gap would demand vigilance, with the balance of payments experiencing
reinforcing pressures from both the current account and capital account. A
depreciating rupiah constitutes an optimal response to restore the external sector
because it would support current account adjustments. The simulations show that it
would be better for Bank Indonesia to respond to the multitude of challenges
through an appropriate policy mix.

IRF – Current account gap shock Graph 3.6

: Monetary policy only : Policy mix

2 BIS Papers No
 d) Capital outflow
The risk of a sudden reversal following large capital inflows seems to be a serious
threat, as experienced by Indonesia due to the 1997–98 crisis. A shock in the form of
a 1% drop in the capital flow (CF) gap would reduce real credit growth, following a
decline in liquidity flowing into the domestic economy. A substantial deceleration in
real credit growth would reduce the output gap due to the presence of a financial
accelerator, thus triggering a further decline in real credit growth. Subsequently, the
fall in real credit growth would prompt Bank Indonesia to raise the LTV ratio and
lower the reserve requirement (RR). The drop in the CF gap would eventually cause
the rupiah to depreciate, both in nominal and real terms, inducing higher domestic
inflation and compelling the central bank to raise its policy rate. The simulations
show that a policy mix is more optimal than monetary policy alone when managing
monetary and financial stability.

IRF – Capital flow gap shock Graph 3.7

: Monetary policy only

: Policy mix

4. Conclusion

As the Indonesian economy is becoming progressively more integrated with the

world economy, the impact of global shocks on the domestic economy are
becoming increasingly pronounced. The influx of capital to the domestic
economy, which

BIS Papers No 88
23
triggers excessive liquidity and exacerbates the risk of sudden reversal, poses a
serious threat to the Indonesian economy, especially in terms of financial stability.
Recent crisis episodes have indicated that monetary policy alone is insufficient to
maintain macroeconomic stability; it needs to be accompanied by macroprudential
policy. Against the backdrop of a dynamic global environment, the multitude of
challenges confronting the Indonesian economy demand a policy mix response
utilising multiple instruments. To that end, a flexible ITF is considered more suitable
than the standard ITF in terms of managing monetary and financial stability in
Indonesia as well as dealing with the dynamics of the financial and external sectors.
Using the framework, Bank Indonesia could mitigate the impact of external shocks
and simultaneously maintain both monetary and financial stability.
This paper finds that the integration of monetary and macroprudential policy
provides better results in terms of mitigating excessive macroeconomic (output) and
financial sector (credit) fluctuations, as compared with any single policy instrument.
By modelling the financial block, the model is better able to capture Indonesian
economic dynamics in both the real sector and financial sector, including
procyclicality and the presence of a financial accelerator.
More comprehensive external sector modelling provides increasingly accurate
analysis of several issues that occur in Indonesia’s external sector. External sector
dynamics, namely shocks affecting the exchange rate, current account and capital
flows, have a significant impact on macroeconomic stability in Indonesia. The model
has proved itself useful in helping Bank Indonesia to formulate an appropriate
policy mix to mitigate the adverse effects of external shocks.

References

Agung, J (2010): ”Integrating monetary and macroprudential policy: towards a new

paradigm of monetary policy in Indonesia after the global crisis”, Bank Indonesia
Working Paper, no WP/07/2010.
Altunbas, Y, L Gambacorta and D Marques-Ibanez (2009): “An empirical assessment
of the risk-taking channel”, Paper presented on the BIS-ECB Workshop on Monetary
Policy and Financial Stability, Basel, September.
Angelini, P, S Neri and F Panetta (2011): “Monetary and macroprudential policies”,
Bank of Italy Working Papers, no 801, March.
Ball, L (1999): “Policy rules for open economies”, in J Taylor (ed), Monetary Policy
Rules, University of Chicago Press.
Bank Indonesia (2011): Strengthening Monetary Policy Framework in Indonesia:
Flexible Inflation Targeting Framework, October.
Beau, D, L Clerc and B Mojon (2011): “Macro-prudential policy and the conduct of
monetary policy”, Bank of France Working Paper, January.
Bernanke, B, M Gertler and S Gilchrist (1996): “The financial accelerator and the
flight to quality”, The Review of Economics and Statistics, vol 78, no 1, February.
Borio, C, C Furfine and P Lowe (2001): “Procyclicality of the financial system and
financial stability: issues and policy options”, BIS Papers, no 1.

2 BIS Papers No
Borio, C (2003): “Towards a macroprudential framework for financial supervision and
regulation?”, BIS Working Papers, no 128, December.
Borio, C and H Zhu (2008): “Capital regulation, risk-taking, and monetary policy: a
missing link in the transmission mechanism?”, BIS Working Papers, no 268,
December.
Capistrán, C, G Cuadra and M Ramos-Francia (2011): “Policy response to external
shocks: lessons from the crisis”, Bank of Mexico Working Papers, no 2011-14,
December.
Cúrdia, V and M Woodford (2009): “Credit spreads and monetary policy”, NBER
Working Paper Series. no 15289, August.
De Nicolò, G, G Dell’Ariccia, L Laeven and F Valencia (2010): “Monetary policy and
bank risk taking”, IMF Staff Position Note, SPN/10/09, July.
Galati, G and R Moessner (2011): “Macroprudential policy – a literature review”, BIS
Working Papers, no 337, February.
Gerali, A, S Neri, L Sessa and F-M Signoretti (2010): “Credit and banking in a DSGE
model of the euro area”, Bank of Italy Working Papers, no 740, January.
Gertler, M and P Karadi (2011): “A model of unconventional monetary policy”,
Journal of Monetary Economics, vol 58.
Ghosh, A, M Goretti, B Joshi, U Ramakrishnan, A Thomas and J Zalduendo (2008):
“Capital inflows and balance of payments pressures—tailoring policy responses in
emerging market economies”, IMF Policy Discussion Paper, no PDP/08/2, June.
Harmanta, Tarsidin, M Bathaluddin and Idham (2013): “ARIMBI with macroprudential
policy”, Revised Edition (not published), Bank Indonesia Working Paper, no
WP/1/2013, October.
Harmanta, Tarsidin, Idham and A Rachmanto (2015): “ARIMBI 2015”, (not published).
Bank Indonesia Research Paper, no LHP/1/DKEM/2015, June.
Issing, O (2003): “Monetary and financial stability – is there a trade-off?”, speech at
the BIS Annual Conference on Monetary Stability, Financial Stability and the
Business Cycle, Basel.
Juhro, S (2014): “The linkages between monetary and financial stability: some policy
perspectives”, Bank Indonesia Occasional Paper, March 2014.
——— (2015): “The role of the central bank in promoting sustainable growth:
perspectives on the implementation of flexible ITF in Indonesia”, Paper presented in
the CCBS/BoE-Columbia University Joint Seminar on Inflation Targeting, New York,
April.
Juhro, S and M Goeltom (2015): “Monetary policy regime in Indonesia”, in A
Kohsaka (ed), Macro-Financial Linkages in the Pacific Region, Routledge.
Lee, J, G Milesi-Ferretti, J Ostry, A Prati and L Ricci (2008): “Exchange rate
assessments: CGER methodologies”, IMF Occasional Paper, no 261.
Peñaloza, D (2011): “In the quest of macroprudential policy tools”, Bank of Mexico
Working Paper, March.
Siregar, R (2011): “The concepts of equilibrium exchange rate: a survey of literature”,
SEACEN Staff Paper, no 81, April.

BIS Papers No 88
25
Svensson, L (1998): “Open-economy inflation targeting”, Journal of International
Economics, no 50, November.
Taylor, J (2001): “The role of the exchange rate in monetary-policy rules”, American
Economic Review, vol 91, no 2, May.
Unsal, D (2011): “Capital flows and financial stability: monetary policy and
macroprudential responses”, IMF Working Paper, no WP/11/189, August.
Wimanda, R, Tarsidin, T Winarno and Idham (2013): “ARIMBI with extended external
sector”, (not published). Bank Indonesia Working Paper, no WP/2/2013, December.

2 BIS Papers No
 Appendix 1

Macroprudential measures in Indonesia

Period of 2010–15 Table A1

Period Measures
July 2010 Minimum holding period on BI bills, one-month holding
period. July 2010 Introduce non-tradable rupiah term deposits for banks.
Nov 2010 Increase the primary rupiah reserve requirement from 5% to 8%, effective from June
2011.
Jan 2011 Reinstate limits on short-term offshore borrowing by banks
 Maximum of 30% of capital;
 Effective end of January 2011 with a three-month transition period.
March 2011 Increase the FX reserve requirements on banks from 1% of FX deposits to 5%,
effective from March 2011.
March 2011 Impose the LDR-based reserve requirement.
Jan 2011 Lengthen (from weekly to monthly) auctions and offer longer maturity (three, six
and nine months) for BI bills.
May 2011 Introduce a six-month holding period for BI bills.
June 2011 Increase the FX reserve requirement from 5% to
8%.
Sept 2011 Mandatory reporting of foreign exchange originating from export earnings.
March 2012 Loan-to-value (LTV) ratio for the property sector (max 70%)
Dec 2012 Mandatory reporting of foreign exchange originating from export earnings.
 Adjustment of the deadline for receipt;
 Limiting the difference between the report and the value based on
the declaration of exported goods
Sept 2013  LTV for second property 60%;
 LTV for third property 50%.
Sept 2013 Secondary reserve requirement raised from 2.5%:
 to 3% from 1 to October 31, 2013.
 to 3.5% of from November 1 to December 1, 2013
 to 4% from December 2, 2013.
Sept 2013 Adjustments of LDR-based reserve requirement
 The upper limit of the LDR-based RR was reduced from 100% to 92%;
 The lower limit remained at 78%;
 Disincentives imposed on banks with an LDR ratio above 92% and CAR of
less than 14%
 Disincentives is also imposed on banks that have LDR less than
78% June 2015 LTV ratio for property sector:
 LTV for first property 80%;
 LTV for second property 70%;
 LTV for third property 60%.
Down payments (DP) for automobiles (min 25%), for commercial vehicles (min 20%)
and for motorcycles (min 20%).
June 2015 Adjustments to LDR-based reserve requirement:
 Redefinition: to include bank securities in the calculation. LDR to be renamed
to Loan-to-Funding Ratio (LFR);
 Commencing August 2015, the upper limit of LFR permitted at 94% if the bank
fulfils an NPL ratio < 5%.

BIS Papers No 88
27
Appendix 2

Bank Indonesia’s Core Model (ARIMBI)

Block 1: The real sector and monetary policy

There are four main equations, namely IS – output gap, inflation – NKPC, the Taylor
rule and Uncovered Interest Parity (UIP). The four equations represent a
macroeconomy or real sector.

IS – Output gap
𝑦�𝑡 = 𝛽1 𝑦�𝑡−1 + 𝛽2 𝑦�𝑡+1 −
𝛽3 𝑟𝑡 + 𝛽6 𝑑�𝑐𝑟𝑡 + 𝛽7 𝑐�𝑎𝑡 + 𝑒𝑒 𝑦� A2.1
�
The output gap (𝑦�) represents the size of the disparity between real GDP and
its potential level. It is determined by the output gap in the previous period, its
value in the next period, real interest rate gap (𝑟̂ ), credit growth gap (𝑑�𝑐𝑟) and
current account (CA) gap (𝑐�𝑎). The credit growth gap (𝑑�𝑐𝑟) is added to
reinforce the correlation between the macroeconomy and the financial block. The
addition of this variable is necessary considering that the dynamics of real credit
volume growth (that subsequently affect the output gap) cannot be fully
represented by real interest rate gap (𝑟̂). Other factors also influence real credit
volume growth. Therefore, it would be more appropriate to directly input the
impact of real credit volume growth into the output gap equation. On the other
hand, investment not only stems from or is financed through credit but also
through direct investment, the magnitude of which is determined by the real
interest rate gap. Meanwhile, the CA gap represents the level of exports and
imports, which is modelled in detail in the external sector. A positive CA gap
increases the output gap.

Inflation – NKPC
𝜋 𝐶𝑃𝐼 = 𝑤𝑤 𝑎𝑑𝑚 𝜋 𝑎𝑑𝑚 + (1 − 𝑤𝑤 𝑎𝑑𝑚 )𝜋 𝑛𝑛𝑛𝑡 + 𝑒𝑒 𝜋𝐶𝑃𝐼 A2.2
𝑡 𝑡 𝑡 𝑡
𝜋 𝑛𝑛𝑛𝑡 = 𝜆 𝜋 𝑛𝑛𝑛𝑡 + (1 − 𝜆 )𝐸 𝜋 𝑛𝑛𝑛𝑡 + 𝜆
𝑦� + 𝜆 𝑧 + 𝑒𝑒 𝜋𝑛𝑛𝑛𝑛𝑛 A2.3
𝑡 𝑡−1 1 𝑡 𝑡+1 3 4 𝑡 𝑡
𝑡
1

The first equation represents CPI inflation and its components, while the second is
NKPC. The CPI inflation (𝜋𝐶𝑃𝐼) is formed by two components, ie administered price
inflation (𝜋 𝑎𝑑𝑚 ) and core inflation (𝜋 𝑛𝑛𝑛𝑡 ), including volatile food inflation. The NKPC
shows that inflation is determined by its value in the previous period, its expected
value, output gap (𝑦�), and real exchange rate gap (𝑧̂ ). We can see that it is a
forward- looking specification of inflation and shows the significance of inflation
expectation. Meanwhile, the output gap represents the level of inflation pressure in
which a higher output gap indicates more intense inflationary pressures. The
variable of the real exchange rate gap represents sources of inflation from abroad, ie
exchange rate pass- through to inflation from imported goods.

Monetary policy – Taylor rule

𝑖𝑖𝑡 = 𝛾𝛾1 𝑖𝑖𝑡−1 + (1 − 𝛾𝛾1 )(𝚤𝚤𝑡 + 𝛾𝛾2 𝜋� 𝐶𝑃𝐼 + 𝛾𝛾3 𝑦�𝑡 ) + 𝑒𝑒 𝑖𝑖 A2.4
𝑡 𝑡

The Taylor rule is determined by value of the policy rate in the previous period, its
long-term trend (𝚤𝚤), inflation gap (𝜋� 𝐶𝑃𝐼 ) and output gap (𝑦�). It is a standard Taylor
rule. Bank Indonesia responds to the inflation gap (deviation of inflation expectation

2 BIS Papers No
from

BIS Papers No 88
29
 its target) and output gap. There is no real exchange rate variable in the equation, in
line with ITF, where the exchange rate is free-floating in nature.

Uncovered interest parity

𝑖𝑖𝑡 − 𝑖𝑖 ∗ = 𝐸 𝐷𝑆
+ 𝑝𝑟𝑒𝑒𝑚 A2.5
� � �
�
The UIP equation shows that it holds when the interest rate differential (𝑖𝑖 − 𝑖𝑖 ∗ ) is the
same as the summation of expected nominal exchange rate
depreciation/appreciation (𝐸𝐷𝑆) and risk premium (𝑝𝑟𝑒𝑒𝑚). The expected nominal
exchange rate appreciation/depreciation is calculated by comparing the expected
nominal exchange rate level (𝑆𝑛𝑛) and the actual current nominal exchange rate level
(𝑆). The expected nominal exchange rate level is calculated based on the exchange
rate level in the period t + 1 and t – 1, with the addition of a drift that is twice the
nominal exchange rate appreciation/depreciation trend. Meanwhile, risk premium
represents the amount of premium asked by investors to invest in domestic assets.

Block 2: The financial block and macroprudential policy

There are three equations in the financial block, namely the credit growth gap
equation, the interest rate spread gap equation and the default risk gap equation.
Three additional equations were included because when an economy experiences a
boom/bust episode, real credit growth increases/decreases, accompanied by an
increase/decrease in default risk. Meanwhile, inclusion of the interest rate spread
gap equation is required to capture the dynamics of lending rates, considering that
the core equations does not include lending rate as its variable. The equations refer
to Peñaloza (2011), with some modification.

Credit growth gap

𝑑�𝑐𝑟𝑡 = 𝛿𝛿1 𝑑�𝑐𝑟𝑡−1 + (1 − 𝛿𝛿1 )�−𝛿𝛿2 𝑟𝑡 − 𝛿𝛿3 𝑠𝑝�𝑟𝑒𝑒𝑎𝑑𝑡 + 𝛿𝛿4 𝑦�𝑡 + 𝛿𝛿5 𝑙�𝑡𝑣𝑡 − 𝛿𝛿6 𝑟�𝑟𝑡 +
𝛿𝛿7 𝑐�𝑎𝑡−1 +
𝛿𝛿8 𝑐�𝑐𝑐𝑡 − 𝛿𝛿9 Υ�𝑡 � + 𝑒𝑒 𝑑𝑑𝑑𝑑𝑑 A2.6
�
The credit growth gap (𝑑�𝑐𝑟) indicates the size of disparity between real credit
volume growth and potential real credit volume growth. The credit growth gap is
determined by the credit growth gap in the previous period as well as by the real
interest rate gap (𝑟̂ ), the interest rate spread gap (𝑠𝑝�𝑟𝑒𝑒𝑎𝑑) and output gap (𝑦�). A
wider real interest rate gap implies a correspondingly narrower credit growth gap,
and a wider spread gap leads to a narrower credit growth gap. On the other hand, a
larger output gap will exacerbate the credit growth gap. Meanwhile, a narrow credit
growth gap is also the result of macroprudential variables, in this instance LTV and
RR. Moreover, a wider LTV gap (𝑙�𝑡𝑣) will broaden the credit growth gap.
Conversely, a wider RR gap reduces loanable funds and thereby narrows the credit
growth gap. Escalating default risk will also precipitate a narrower credit growth gap
as banks opt to hold their credit allocation. However, considering that default risk is
already implicitly represented by spread (which indicates that higher lending rates
tend to escalate default risk) and output gap (which denotes that a larger output
gap leads to lower default risk), default risk no longer appears in the credit growth
gap equation.
A balance of payments (BOP) surplus/deficit is added to the credit growth gap
equation, represented by the total of the CA gap and CF gap. The inclusion of the
BOP variables intends to capture the impact of more/less liquidity in the economy
stemming from the external sector. The combination of the variables, CA gap and
CF gap, in the credit growth gap equation can be expressed as a single variable,

3 BIS Papers No
namely

BIS Papers No 88
31
the BOP surplus/deficit. However, in order to accommodate differences between the
characteristics of the current account and capital flows with regards to their impact
on liquidity, a lag is applied to the CA gap, while the CF gap has no lag.

Interest rate spread gap

𝑠𝑝�𝑟𝑒𝑒𝑎𝑑𝑡 = 𝑣1 𝑠𝑝�𝑟𝑒𝑒𝑎𝑑𝑡−1 + (1 − 𝑣1 )𝑣2 𝑑�𝑒𝑒𝑐𝑐𝑡 + 𝑒𝑒 𝑠𝑝𝑟𝑛𝑛𝑎𝑑 A2.7
�
The interest rate spread gap equation (𝑠𝑝�𝑟𝑒𝑒𝑎𝑑) shows that the gap
between the lending rate and deposit rate is not only determined by the spread gap
in the previous period but also by the default risk gap (𝑑�𝑒𝑒𝑐𝑐) faced by the banks.
In addition to the range of variables previously mentioned, the interest rate spread
gap is also affected by the bank market structure. This, for instance, is observable
based on the fact that the wide spread in Indonesia, amongst others, is attributable
to monopolistic competition in terms of market structure. However, considering
that the bank market structure has remained relatively unchanged in the near term
(one to two years), the variable of bank market structure is omitted from the model.

Default risk gap

𝑑�𝑒𝑒𝑐𝑐𝑡 = 𝜃𝜃1 𝑑�𝑒𝑒𝑐𝑐𝑡−1 + (1 − 𝜃𝜃1 )�𝜃𝜃2 𝑠𝑝�𝑟𝑒𝑒𝑎𝑑𝑡 + 𝜃𝜃3 𝑑�𝑐𝑟𝑡−1 − 𝜃𝜃4 𝑦�𝑡 + 𝜃𝜃5 Υ�𝑡 � + 𝑒𝑒 𝑑𝑛𝑛𝑑𝑑 A2.8
�
The default risk gap equation (𝑑�𝑒𝑒𝑐𝑐) indicates that the default risk gap is
determined by the default risk gap in the previous period, the interest rate spread
gap (𝑠𝑝�𝑟𝑒𝑒𝑎𝑑), credit growth gap (𝑑�𝑐𝑟) in the previous period and the output
gap (𝑦�). A wider spread gap leads to greater pressures on the cost of capital faced
by the customer, which will clearly intensify default risk. Meanwhile, a higher credit
growth gap increases the likelihood of default. Conversely, a larger output gap
ameliorates business conditions and eases default risk as the economy experiences
robust growth. Gross non- performing loans (NPL) data are used as a proxy for
default risk gap, which reveals the level of risk faced by the bank (and the requested
premium).

Macroprudential rule
Macroprudential policy can constitute a separate policy, in other words
macroprudential policy and monetary policy are independent. Notwithstanding,
macroprudential policy can be incorporated into monetary policy, explicitly using
the augmented Taylor rule or implicitly through variables in the Taylor rule.
Interaction between macroprudential policy and monetary policy must be modelled
accurately, considering that the effect of such interaction can be complementary,
neutral/independent or indeed conflicting. In this case, modelling macroprudential
policy and monetary policy falls under the auspices of Bank Indonesia.
There are two macroprudential tools in the model, namely the LTV ratio and the
reserve requirement (RR), which are modelled together. In the macro model, the
macroprudential instrument mechanism of the RR resembles the LTV. The current
LTV regulation has a direct effect on mortgages and automotive loans, and
ultimately influences total credit. The reserve requirement affects total credit
through its impact on loanable funds. In the model, both macroprudential tools
respond to total credit.

LTV rule
LTV policy intends to limit the provision of credit/financing by banks when an
economy is experiencing boom conditions and, conversely, expand the allocation of
credit/financing when an economy is in recession. Limits are placed on credit

3 BIS Papers No
availability, where banks are only permitted to extend credit up to the specified LTV
ratio, and, on the other hand, households and the corporate sector are only allowed
to borrow up to the prevailing LTV ratio. Thus, LTV ratio intends to control the pace
of credit growth. If the credit growth gap widens, LTV will need to be lowered in
order to curb credit growth. To this end, an LTV rule is required as follows:
𝑙�𝑡𝑣𝑡 = 𝜇1 𝑙�𝑡𝑣𝑡−1 + (1 − 𝜇1 )�−𝜇2 𝑑�𝑐𝑟4𝑡 � + 𝑒𝑒 𝑙𝑡𝑣 A2.9
�

RR rule
The reserve requirement ratio (RR) is fundamentally designed to manage the
amount of loanable funds. This can be achieved, for example, by raising/lowering
the reserve requirement at times when the economy is facing excess/insufficient
liquidity. The reserve requirement is countercyclical in nature and helps control
procyclicality in the financial sector, thereby avoiding excessive credit growth. The
dynamic reserve requirement refers to the following rule:
�𝑟𝑟𝑡 = 𝜅1 𝑟�𝑟𝑡−1 + (1 − 𝜅1 )𝜅2 𝑑�𝑐𝑟4𝑡 + 𝑒𝑒 𝑟𝑟 A2.10
�

Block 3: The external block and exchange rate policy

There are three equations in the external block, namely the current account (CA)
gap equation, the capital flow (CF) gap equation and several equations representing
the rest of the world. As mentioned previously, the CA gap is the difference
between the CA to GDP ratio and CA norms. Meanwhile, the CF gap is the difference
between the CF to GDP ratio and the optimum level of CF. CA norms are calculated
by regressing the variable CA to GDP ratio against the fundamental variables of an
economy, applying the Macroeconomic Balance Approach cited by Lee et al. (2008).
In contrast, an approach to calculate the optimal level of capital flows is yet to be
determined. There are only depictions of an optimal CF to GDP ratio in relation to
the CA to GDP ratio, as quoted by Ghosh et al (2008).

Current account gap

𝑐�𝑎𝑡 = 𝖯1 𝑐�𝑎𝑡−1 + (1 − 𝖯1 )(𝖯2 𝑧̂𝑡 − 𝖯3 𝑦�𝑡 + 𝖯 𝑦� ∗ ) + 𝑒𝑒 𝑐�𝑎 A2.11
4 𝑡 𝑡

The equation of the current account gap (𝑐�𝑎) shows the size of the gap between
the CA to GDP ratio and CA norms. This variable is influenced by the previous
current account gap, real exchange rate gap (𝑧̂ ), output gap (𝑦�) and global output
gap (𝑦� ∗ ). As the real exchange rate gap increases (depreciates), the current account
gap also increases, while a higher output gap will result in a lower current account
gap (considering its impact on higher imports). On the other hand, a larger global
output gap (representing external/foreign demand) will exacerbate the current
account gap (considering its impact on higher exports).

Capital flow gap

𝑐�𝑐𝑐𝑡 = 𝜙𝜙1 𝑐�𝑐𝑐𝑡−1 + (1 − 𝜙𝜙1 )�𝜙𝜙3 (𝑖𝑖𝑡 − 𝑖𝑖 ∗ − (𝜙𝜙2 𝐸𝐷𝑆𝑡−1 + (1 − 𝜙𝜙2 )𝐸𝐷𝑆𝑡 ) − 𝑝𝑟𝑒𝑒𝑚𝑡 ) + 𝜙𝜙4 𝑦�𝑡 −
� �
𝑑𝑑 − 𝜙𝜙 Υ � + A2.12
∗
𝜙𝜙𝑐�𝑦�
𝑒𝑒
5 𝑡 6 𝑡 𝑡

The capital flow gap equation (𝑐�𝑐𝑐) indicates the magnitude of the gap between
the CF to GDP ratio and the optimum level of CF. This variable is driven by the
previous capital flow gap, uncovered interest rate parity (UIP), output gap (𝑦�) and
global output gap (𝑦� ∗ ). A higher domestic nominal interest rate will attract capital
flows (increasing the capital flow gap), while a higher global nominal interest rate
will trigger foreign

BIS Papers No 88
33
capital outflows (reducing the capital flow gap), and higher expected exchange rate
depreciation will also lead to foreign capital outflows (reducing the capital flow
gap). Conversely, a greater level of risk in Indonesia will prompt foreign capital
outflows (lowering the capital flow gap). In this case, it is assumed that UIP will not
hold because of a lag in the formation of expected nominal exchange rate
appreciation/depreciation.
An increase in the output gap represents improvements in the economy of
Indonesia and will attract foreign capital (increasing the capital flow gap).
Meanwhile, a larger global output gap denotes improvements in the global
economy, thereby diverting foreign capital from Indonesia (reducing the capital flow
gap). Considering the Indonesian economy is affected by improvements in the
global economy, the impact on capital flows is sometimes mixed, depending on the
most dominant factor.
It should be emphasised here that the CA gap and CF gap were brought back
to their optimal path (where the gap is zero) through appreciation/depreciation of
the nominal and real exchange rates and adjustments to other influencing variables.
Furthermore, there is no policy to control capital flow in this model. The CF gap
equation shows that capital moves freely depending on its determinants. However,
some shocks can be added to the model to capture the impact of BI regulations on
capital movements.

Rest of the world

𝑦� ∗ = 𝛽𝑐𝑐 𝑦� ∗ + 𝛽𝑐𝑐 𝑦� ∗ − 𝛽𝑐𝑐 (𝑟 ∗ − 𝑟�∗ ) + 𝑒𝑒 𝑦�∗ A2.13
𝑡 𝑡−1 2 𝑡+1 3 𝑡 𝑡 𝑡

1
𝜋 ∗ = 𝜆𝑐𝑐 𝜋 ∗ + (1 − 𝜆𝑐𝑐 )𝐸 𝜋 ∗ + 𝜆𝑐𝑐 𝑦� ∗ + 𝑒𝑒 𝜋∗ A2.14
𝑡 𝑡−1 1 𝑡 𝑡+1 3 𝑡 𝑡

1
𝑖𝑖𝛾𝛾∗ = 𝛾𝛾𝑐𝑐 𝑖𝑖 ∗ + �1 −
��𝑟̅∗ + 𝜋4∗ + 𝛾𝛾𝑐𝑐 (𝜋4∗ − 𝜋∗ ) + 𝑦� ∗ � + 𝑒𝑒 𝑖𝑖∗ A2.15
𝛾𝛾
𝑡 1 𝑡−1 𝑑𝑑1 𝑡+3 2 𝑡+4 𝑠𝑠 𝑑𝑑3 𝑡 𝑡

The rest of the world equations consist of world IS – output gap, world inflation –
NKPC and the world Taylor rule. It is a simple model of the world economy and a
representation of what central banks do in response to shocks of world inflation and
GDP. In addition, there is a residual in each equation to represent an external shock.
As required, covariance shocks could be added to ensure a more appropriate
magnitude of external shocks.
Bank Indonesia actively responds to external sector dynamics, such as through
foreign exchange market intervention and capital flow management. Even though
Bank Indonesia’s exchange rate regime is free floating, the rupiah exchange rate is
actually managed in line with economic fundamentals and to be less volatile. By
intervening on the foreign exchange market, Bank Indonesia adheres to the
exchange rate policy framework, which is continuously developed in line with the
current issues and challenges.
The Indonesian economy is not currently at an external balance, as indicated by
a current account (CA) deficit that surpassed 4.27% in second quarter of 2014. On
the other hand, the Indonesian economy is nearly close to its internal balance, which
is demonstrated by the low level of unemployment (around 5.7% in February 2014)
as well as low and stable inflation (3.99% in August 2014, in line with the inflation
target of 4.5%± 1%). Internal and external balance cannot be achieved with rapidity,
it should be achieved gradually over a sufficiently long horizon, for example five
years. In the near term, however, Bank Indonesia should seek to achieve favourable

3 BIS Papers No
economic conditions, while simultaneously pursuing the internal and external
balance.

BIS Papers No 88
35
 Bank Indonesia policy does not aim to achieve internal and external balance
immediately but to achieve it optimally, which requires a rupiah exchange rate path
that responds favourably to current economic conditions and concomitantly seeks
the internal and external balance. This is not a medium-term fundamental exchange
rate path (ie an exchange rate path that is consistent with the internal and external
balance). If Bank Indonesia sticks to the medium-term fundamental exchange rate
path; the economy would be forced to adjust drastically to the internal and external
balance, which could trigger macroeconomic instability as the exchange rate is
distorted from its optimal level. An optimal exchange rate level must be consistent
with macroeconomic and financial variables at their fundamental value. Therefore,
Bank Indonesia should maintain the rupiah exchange rate in line with short-term
fundamentals, while simultaneously aiming for the medium-term fundamental
exchange rate path. The optimal exchange rate path is consistent with attaining the
inflation target as outlined by the flexible ITF.
Bank Indonesia’s exchange rate policy is basically a combination of responding
to current economic conditions while gradually shifting the economy back to its
internal and external balance. If no other shocks occur in the near term (about one
to two years), the short-term fundamental exchange rate path will be the same as
the medium-term fundamental exchange rate path, resembling the concept of
permanent equilibrium exchange rate (PEER), in which there are responses to both
temporary and permanent shocks. The path is depicted in Graph A2.1, in which
there are two exchange rate paths, ie nominal exchange rate trend values ( 𝑆̅,
broken straight line) and Bank Indonesia’s exchange rate policy path (𝑆, solid curve
line).

Path of exchange rate policy Graph A2.1

Nominal
Exchange
Rate

S=S

S
S: Nominal exchange rate
S: Trend of nominal exchange rate

Period

In the ARIMBI model, 𝑑𝑧 represents short-term fundamental (real) exchange

rate depreciation/appreciation, while 𝑑� 𝑧 � represents medium-term
fundamental (real) exchange rate depreciation/appreciation. Meanwhile, in nominal
terms it is represented by 𝐷𝑆 and 𝐷� �𝑆� . In the model, the exchange rate is
modelled as it depreciates/appreciates over time. In the model, the nominal and
real exchange rates are determined based on the uncovered interest parity (UIP)
and purchasing power parity (PPP) equations as follows:

3 BIS Papers No
Uncovered interest parity
𝑖𝑖𝑡 − 𝑖𝑖 ∗ = 𝐸𝐷𝑆
+ 𝑝𝑟𝑒𝑒𝑚 A2.16
� � �

Purchasing power parity

𝑑𝑧𝑡 = (𝐷𝑆𝑡 + 𝜋∗) − 𝜋𝑡 A2.17
�
where:
trend of real exchange rate depreciation/appreciation:
𝑑� 𝑧 � = �𝑑�𝑑 �
�𝑑�𝑧�𝑑𝑑�𝑟�𝑑 �𝑤𝑤� 𝑡�ℎ +�𝜄𝜄 − 𝜄𝜄 + 𝑐�𝑐𝑐 �+𝑒𝑒 A2.18
� Υ �𝑐�
𝑡 𝑡 1 2 𝑡−1 𝑡 𝑡
𝑡

From both the UIP and PPP equations, several determinants of nominal and real
exchange rates are observed in the model, ie (i) interest rate differential (considering
expected nominal exchange rate depreciation/appreciation and risk premium), (ii)
terms of trade (represented by domestic and world inflation), (iii) risk (Υ), and (iv)
net foreign assets represented by current account gap (𝑐�𝑎) and capital flows
gap (𝑐�𝑐𝑐). Higher risk leads to greater exchange rate depreciation. On the other
hand, larger net foreign assets would lead to exchange rate appreciation.

Block 4: Macro risk and the risk-taking channel

In the model, we endogenise variables of risk in order to capture the role of risk
perception, using the International Country Risk Guide (ICRG) index as a proxy. The
risk is called macro risk to represent risk at the macro level and modelled as follows:
Υ�𝑡 = 𝜂𝜂1 Υ�𝑡−1 + (1 − 𝜂𝜂1 )�−𝜂𝜂2 𝑦�𝑡 + 𝜂𝜂3 𝜋� 𝐶𝑃𝐼 + 𝜂𝜂4 𝑧̂𝑡 − 𝜂𝜂5 𝑐�𝑎𝑡 + 𝜂𝜂6 𝑑�𝑒𝑒𝑐𝑐𝑡 � + 𝑒𝑒 Υ� A2.19
𝑡 𝑡

From the equation, we notice that the level of macro risk in the previous period
determines macro risk along with the output gap ( 𝑦� ), inflation gap ( 𝜋� 𝐶𝑃𝐼 ),
real exchange rate gap (𝑧̂ ), current account gap (𝑐�𝑎) and default risk gap (𝑑�𝑒𝑒𝑐𝑐).
A higher output gap would induce lower macro risk, while a higher inflation gap
would raise macro risk. On the other hand, real exchange rate depreciation would
raise macro risk in a similar way to a deteriorating current account. In the financial
sector, higher default risk would escalate macro risk.
The determinants of macro risk are basically composed of macroeconomic and
financial variables. Furthermore, macro risk influences other variables in the model.
Its impact not only affects real exchange rate depreciation/appreciation but also the
credit growth gap (called the risk-taking channel), default risk gap, risk premium
and capital flow gap.

BIS Papers No 88
37