ADBI Working Paper Series

DEALING WITH QUANTITATIVE

EASING SPILLOVERS IN EAST ASIA:
THE ROLE OF INSTITUTIONS AND
MACROPRUDENTIAL POLICY

Ayako Saiki
Pornpinun Chantapacdepong
Ulrich Volz

No. 604
October 2016

Asian Development Bank Institute

Ayako Saiki is currently a visiting scholar at SOAS, University of London. Pornpinun

Chantapacdepong is a research fellow at the Asian Development Bank Institute (ADBI).
Ulrich Volz is head of the Department of Economics at SOAS, University of London; a
senior research fellow at the German Development Institute; and an honorary professor
of Economics at the University of Leipzig.
The views expressed in this paper are the views of the author and do not necessarily
reflect the views or policies of ADBI, ADB, its Board of Directors, or the governments
they represent. ADBI does not guarantee the accuracy of the data included in this paper
and accepts no responsibility for any consequences of their use. Terminology used may
not necessarily be consistent with ADB official terms.
Working papers are subject to formal revision and correction before they are finalized
and considered published.
The Working Paper series is a continuation of the formerly named Discussion Paper series;
the numbering of the papers continued without interruption or change. ADBIs working
papers reflect initial ideas on a topic and are posted online for discussion. ADBI encourages
readers to post their comments on the main page for each working paper (given in the
citation below). Some working papers may develop into other forms of publication.
Suggested citation:
Saiki, A., P. Chantapacdepong, and U. Volz. 2016. Dealing with Quantitative Easing
Spillovers in East Asia: The Role of Institutions and Macroprudential Policy. ADBI Working
Paper 604. Tokyo: Asian Development Bank Institute. Available:
https://www.adb.org/publications/dealing-quantitative-easing-spillovers-east-asia/
Please contact the authors for information about this paper.
Email: ayako@alumni.brandeis.edu (corresponding author)

Asian Development Bank Institute

Kasumigaseki Building 8F
3-2-5 Kasumigaseki, Chiyoda-ku
Tokyo 100-6008, Japan
Tel:
Fax:
URL:
E-mail:

+81-3-3593-5500
+81-3-3593-5571
www.adbi.org
info@adbi.org

2016 Asian Development Bank Institute

ADBI Working Paper 604

Saiki, Chantapacdepong, and Volz

Abstract
This paper explores the impact of advanced countries quantitative easing on emerging
market economies (EMEs) and how macroprudential policy and good governance play a role
in preventing potential financial vulnerabilities. We used confidential locational bank statistics
data from the Bank for International Settlements to examine whether quantitative easing has
caused an appreciation of EMEs currencies and how it has done so, and whether this has
in turn boosted foreign-currency borrowing, thus making EMEs vulnerable to balance sheet
and maturity mismatch problems. While focusing our analysis on East Asian economies,
we compare them with Latin American economies, which were also major recipients of
quantitative easing capital inflows. We found that government effectiveness plays an
important role in curbing excessive borrowing when the exchange rate is overvalued.
JEL Classification: E44, E58, F31, F32, F34

ADBI Working Paper 604

Saiki, Chantapacdepong, and Volz

Contents
1.

INTRODUCTION ....................................................................................................... 1

2.

LITERATURE REVIEW ............................................................................................. 2

2.1
2.2

3.

STYLIZED FACTS..................................................................................................... 3
3.1
3.2

4.

Capital Flows ................................................................................................. 4

Exchange Rates ............................................................................................ 5

THE ROLE OF MACROPRUDENTIAL POLICY AND GOOD GOVERNANCE .......... 6

4.1
4.2

5.

Some Features of Cross-Border Bank Lending in Emerging

Market Economies ......................................................................................... 2
Currency Appreciation and Borrowing Behaviors ........................................... 3

Macroprudential Measures ............................................................................. 6

Macroprudential Policy Implementations in East Asia and Other
Emerging Market Economies ......................................................................... 7

EMPIRICAL ANALYSIS ............................................................................................. 8

5.1
5.2

Data: Sources and Summary Statistics .......................................................... 8

Variables ........................................................................................................ 9

6.

MODEL ................................................................................................................... 12

7.

RESULTS................................................................................................................ 13

8.

CONCLUSIONS AND POLICY IMPLICATIONS ...................................................... 17

REFERENCES ................................................................................................................... 18
APPENDIXES
1
2
3
4
5

Sample Data Countries ................................................................................ 21

Variables and Data Sources ........................................................................ 22
List of Tax Havens and Banking Center Countries ....................................... 23
Capital InflowsOutflow Restriction Measures and Macroprudential Index ... 24
Government Effectiveness of Sample Countries
(Baseline Asia and Latin America, Mean and 2 Standard Deviations) ....... 28

ADBI Working Paper 604

Saiki, Chantapacdepong, and Volz

1. INTRODUCTION
The quantitative easing of advanced economies central banks in the aftermath of the
global financial crisis (GFC)the Federal Reserve Bank, the European Central Bank,
the Bank of England, and the Bank of Japan in particularhave contributed to largescale capital inflows to emerging market economies (EMEs), including those in East
Asia. While the capital surge into EMEs underpinned financial stability of EMEs to
some extent, for some EMEs, capital inflows turned out to be pro-cyclical, causing
asset bubbles and appreciation of exchange rates. This posed a new challenge to
monetary policy.
For example, Brazilian President Rousseff in 2012 mentioned that a monetary
tsunami was making its way to EMEs. Indias Central Bank Governor Raghuram Rajan
(2013) called for prudent measures to avoid an excess buildup of credit. Indeed,
given the volatile nature of short-term capital flows and the danger of sudden stops
(reversals of capital flows), many EMEs implemented macroprudential policies and
(sometimes temporary) measures of capital account restrictions to better cope with
capital flows. The speed at which the direction of flows can change was illustrated
by Fed Chairman Bernankes tamper talk, which, just by expectation, prompted a
reversal of capital flows in some EMEs and stirred fear of financial instability in a
number of EMEs.
Figure 1 shows annual percentage change in global liquiditydefined here as
cross-border capital flow (debt securities and bank credit). While a sharp reversal of
capital flows occurred during the GFC, the growth in cross-border capital flow in United
States (US) dollars turned positive as early as December 2008. However, the growth
of cross-country bank and debt securities capital flows are still lower than they were
before the GFC, demonstrating the severe abundance of global liquidity before
the GFC.
Figure 1: Percentage Change in Cross-Border Credit (Debt Securities
and Bank Lending by Major Currencies) (US$ trillion)

Source: Bank for International Settlements. Global Liquidity Indicators. Table E-1. http://www.bis.org/statistics/gli.htm

ADBI Working Paper 604

Saiki, Chantapacdepong, and Volz

Against this backdrop, this paper investigates the impact of quantitative easing of major
advanced economies on foreign-currency borrowing of EMEs in East Asia to examine
if quantitative easings had led EMEs to overborrow. In particular, it focuses on the role
of institutions in dealing with capital account management and macroprudential policies
adopted by East Asian EMEs 1 in response to capital inflows. Of the different types of
capital inflows, we focus on cross-border bank lending, using data from the Bank for
International Settlements (BIS) that consist of comprehensive lending information
including details on currency breakdown. While the paper focuses on EMEs in
East Asia, we compare our findings with Latin American countries.

2. LITERATURE REVIEW
This paper focuses on how EMEs responded to capital inflows, particular in the form
of cross-border bank borrowing, seeing if and how the prudential policies and
institutional setting dampened the foreign exchange borrowing caused by exchange
rate appreciation. As a meta-survey by Frankel and Saravelos (2010) shows, a surge in
capital inflows and real exchange rate appreciation (especially the short-term hot
money flows) are among the most reliable early warning indicators of a financial crisis.
Hence, preventing overborrowing is an important element in preventing financial crisis.

2.1 Some Features of Cross-Border Bank Lending in Emerging

Market Economies
The literature finds that the quantitative easing spillover and its determinants are no
different from conventional monetary easing (Glick and Leduc 2013). The stance of the
International Monetary Fund (IMF) on the quantitative easing spillover to EMEs is
overall positive, albeit with some nuances: quantitative easing not only supported
EMEs by preventing the contagion-driven market meltdown, but also underpinned their
growth. At the same time, the IMF expressed concern about the consequences of
a prolonged period of capital inflows and cheap foreign financing for future financial
stability. Indeed, in many EMEs, we observed a rapid expansion of credit, which
historically has been followed by a period of financial instability (Schularick and
Taylor 2012).
Typically, EMEs cannot borrow their own currency in cross-border borrowing, a
phenomenon that has been referred to as the original sin problem (Eichengreen,
Hausmann, and Panizza 2003). While some EMEs, as a result of the catching-up
process, are able to raise funds in their own currencies (the recent launch of the
renminbi bond index in the London Stock market is a good example), there are
reasons, especially at the micro level, why agents in EMEs prefer borrowing in foreign
currencies, especially in the dollar. First, many developing countries depend on
commodity exports as their main source of revenue; since commodities are almost
exclusively priced in the dollar, exporters in those countries prefer to use the dollar
rather than local currencies in their transactions (Frankel and Saiki 2002). Even if they
are exporting industrial goods because of the lower pricing power of exporters and the
thin financial markets, they tend to price in the importers currency (Goldberg and Tille
2009; Friberg and Wilander 2008). Finally, when EMEs trade with each other, the dollar

Our sample of East Asian countries includes India. India is not part of East Asia, but we included it in
our analysis as it is one of the countries that experienced a quantitative-easing monetary tsunami and
took several measures to combat it, especially under its Central Bank Governor Rajan.

ADBI Working Paper 604

Saiki, Chantapacdepong, and Volz

is often used because of its role as the worlds vehicle currency. The need to settle
trade in dollars makes them prefer dollar debt to home currency debt.
Compared with advanced countries, EMEs typically suffer from poor protection of
property rights and corruption. Eichengreen, Hausmann, and Panizza (2003) claimed
that the currency mismatch problem and institutional weakness (they call it the debtintolerance school) are analytically distinct. We need to bridge the two by looking at
how a countrys prudent behavior, or a good institution in a broader sense, would make
a country less vulnerable to the currency mismatch problem that led to balance-ofpayments crises in EMEs in the past. In other words, we think a prudent government
will not overborrow in foreign currencies.

2.2 Currency Appreciation and Borrowing Behaviors

Rajan (2013) claimed that strong currency and asset prices reduce agents loan-tovalue ratios, which lessens perceived risks of the leverage. This often feeds into an
oversupply of credit from cross-border lending, which sometimes results in balance-ofpayment problems. Against this backdrop, Rajan called on EMEs to design and
implement prudential policy, the effectiveness and efficacy of which has not yet been
strongly established in the literature.
The meltdown of foreign-currency debt overhang has been well noted in in the past
(Latin Americas Tequila Crisis, Asian currency crisis, etc.). In floating exchange rate
regime countries, such as Poland and Hungary, the expectation of continuing
appreciation has led them to borrow heavily in euro and in Swiss franc, which was
pegged to the euro between September 2011 and January 2015. During the GFC, the
Polish zloty and the Hungarian forint depreciated by almost 65% vis--vis the euro and
20% vis--vis the Swiss franc. This increased the foreign-currency loans of Hungary to
47% of gross domestic product (GDP) (of which 31% was denominated in Swiss
francs) and 16% of GDP in Poland. Afterward, these two countries put macroprudential
tools in place. Arteta (2005) found that floating exchange rate regime countries are
more likely to deposit or borrow in the US dollar, because floating regimes are
positively associated with deposit dollarization more strongly as investors seek to keep
their asset value stable in US dollar terms. However, a stronger home currency will
make it easier for EMEs to borrow in a foreign currency. To the best of our knowledge,
no studies have so far assessed how the valuation of exchange rates is related to
foreign bank lending, and how they interact with institutions and macroprudential
measures. Overall, existing literature seems to imply that foreign-currency lending
and/or borrowing (financial dollarization) can be detrimental to an economy. Some go
one step further and suggest that it was one of the main causes of the GFC (Akca
2011).

3. STYLIZED FACTS
Before going into the formal analysis, it may be useful to see some stylized facts
concerning the GFC and the various quantitative easings that followed. Since we are
particularly interested in the effect of exchange rate appreciation, and possibly
overvaluation, in what follows we examine the actual data of capital flow exchange
fluctuations of major EMEs (not just East Asian EMEs) around the time of the GFC and
subsequent quantitative easing policies.

ADBI Working Paper 604

Saiki, Chantapacdepong, and Volz

3.1 Capital Flows

Figure 2 shows both gross and net capital inflows to selected East Asian countries. 2
After the start of quantitative easing in 2008, we can see an increase in portfolio
inflows, in terms of both gross and net. However, EMEs became disproportionally
(relative to their economic size) large recipients of capital inflows in the run up to the
GFC even in net terms (not reported in the graph). This indicates that, when the GFC
hit their currencies, these countries were likely to have suffered from a currency
mismatch problem. According to the updated systemic banking crisis database (Laeven
and Valencia 2012), the following EME countries experienced currency crises after
2007: Belarus, Democratic Republic of the Congo, Ghana, Serbia, Turkmenistan,
Uganda, Serbia, Ukraine, and Zambia.
Figure 2: Net and Gross Capital Flows to East Asia

Note: The total capital flow to India, Indonesia, the Republic of Korea, Malaysia, the Philippines, and Thailand.
Source: International Monetary Fund, Balance of Payment Statistics.

The plunge of capital flows to East Asian economies during the GFC was less severe
compared with other EMEs because of several factors, mainly: macroeconomic
management by central banks, strong economic fundamentals and high levels of
financial reserves, and geographic proximity to the Peoples Republic of China with its
2

We also included India in our baseline sample. Strictly speaking, India is not geographically categorized
under East Asia, but because of its experience in receiving quantitative easing spillover money and
capital openness, as well as the various macroprudential policy (MPP) measures it has taken, we
included it in our list of sample countries.

ADBI Working Paper 604

Saiki, Chantapacdepong, and Volz

high economic growth rate. Nevertheless, with greater financial integration, the external
economic environment has a growing impact on the domestic economy.

3.2 Exchange Rates

Figure 3a shows our measurement of exchange rate valuation, which is discussed in
the next chapter. Figure 3b shows the real effective exchange rate (consumer price
index [CPI] based) from the BIS.
Figure 3(a): Exchange Rate vis--vis the US$ of Selected EMEs

Figure 3(b): Real Effective Exchange Rate of Selected EMEs

EME = emerging market economy, REER = real effective exchange rate.

Notes:
1. In Figure 3(a), the higher value means greater overvaluation.
2. In Figure 3(a), the degree of overvaluation is calculated by the authors. See Chapter 4 for explanation.
3. The value reported is the average for the following countries:
Latin America: Brazil, Chile, Colombia, Mexico, and Peru.
Eastern Europe and the Russian Federation: Bulgaria, Czech Republic, Hungary, Poland, Romania, and the Russian
Federation.
East Asia: India, Indonesia, the Republic of Korea, Malaysia, the Philippines, and Thailand.
Others: South Africa and Turkey.
Source: Bank for International Settlements.

ADBI Working Paper 604

Saiki, Chantapacdepong, and Volz

Roughly speaking, currency overvaluations of selected Asian countries were severe

before the Asian currency crisis in 1997, but not so much before the GFC when global
liquidity was very high probably due to their active management of exchange rate.
On the contrary, the exchange rates of Latin America, South Africa, and Turkey
have been relatively more overvalued. Looking at the development of real effective
exchange rate (REER), Latin American countries tend to have higher REER than the
rest of the EMEs, and the REER of the Eastern European countries has been
appreciating.

4. THE ROLE OF MACROPRUDENTIAL POLICY

AND GOOD GOVERNANCE
Macroprudential thinking begins with the observation that the whole of the financial
system is more than the sum of the parts. Jeffrey Frankel (2015)

4.1 Macroprudential Measures

The idea of macroprudential policy is not new. According to Galati and Moessner
(2013) the word and concept were already used in an undisclosed document of the BIS
in the late 1970s, which defined it as a systematic orientation of regulation and
supervision linked to the macroeconomy.
Macroprudential policies take macroeconomic cycles into consideration and deploy
countercyclical policy. Examples include cyclically varying provisioning requirements,
cyclically varying loan-to-value (LTV) ratios, countercyclical capital buffers,
capital/liquidity surcharges/levies on systemically important financial institutions, taxes
on volatile funding, caps on credit growth, and higher reserve requirements (Ostry et al.
2011). Before the GFC, monetary authorities tried to maintain financial stability mainly
through microprudential policies, which aim to cope with idiosyncratic risks of individual
financial institutions and mitigate externalities arising from the behavior of individual
institutions (such as the forward-looking provisioning of expected losses, caps on
LTV ratios or minimum collateral haircuts, higher-risk weights on specific types of
exposures, and leverage ratios (Ostry et al. 2011).
The GFC made it clear that microprudential policy alone is not sufficient to maintain
financial stability, as financial institutions are interconnected in a very complex manner.
In addition, traditional monetary policy (changing interest rates, money supply, etc.)
affects not only the finance sector but also non-finance sectors, including households.
On the other hand, macroprudential measures are more targeted to the finance sector
and look at the finance sector as a whole, rather than the individual financial institution.
There are some overlapping aspects of macroprudential policy (MPP) and capital flow
management (CFM). 3 However, Forbes, Fratzscher, and Straub (2015) claimed that
since capital restriction measures do not discriminate among different measuresthey
may hurt some lenders (such as small and medium-sized enterprises) who are actually
in need of foreign money. They also argued that capital flow restrictions suffer from
leakage, making MPPs a preferable option, as these are more targeted to the finance
3

CFM refers to measures that are designed to limit capital flows. It aims to discourage transactions in
foreign currency as well as residency-based capital flow management measures. The form and extent
of CFM in Asia has varied widely, reflecting three main issues: (i) the limit to macroeconomic policy
adjustment each economy faced, (ii) the specific finance sector pressures each economy faced, and
(iii) concerns about the volatility of inflows.

ADBI Working Paper 604

Saiki, Chantapacdepong, and Volz

sector. Overall, they concluded that MPPs are a better tool to employ when there is a
massive inflow of capital, especially when it is a pro-cyclical one. There is a caveat,
however. MPP also can leak via regulatory arbitrageso international coordination is
an urgent task.
In much of the earlier literature it was argued that capital flow restrictions alone can be
welfare reducing (e.g., Fischer 1998) while more recent literature (Korinek 2010;
Bianchi 2011) showed that carefully calibrated MPP can increase welfare under certain
circumstances. Technically, one MPP tool to deal with cross-capital inflows involves the
ratio of foreign-currency debt: for example, restricting agents to borrow in foreign
currencies, especially mortgage loans, or higher requirements for the foreign reserve
ratio of commercial banks (Frankel 2015).

4.2 Macroprudential Policy Implementations in East Asia

and Other Emerging Market Economies
Appendix 4 shows capital inflow and outflow openness of the selected countries
(Schindlers index), along with the IMFs macroprudential index (MPI). There is an
interesting pattern across different regions: in East Asia, although some economies
became more restrictive toward capital flows after the Asian crisis in 1997, MPPs
were strengthened, both after the Asian crisis and the start of quantitative easing
in advanced economies. Latin American countries, especially Brazil, reacted to
quantitative easing spillover by restricting capital flows, especially outflows. Economies
in Central and Eastern Europe loosened capital flow restriction but strengthened
MPP measures.
Pradan et al. (2011) summarized the tools of MPP and CFM of selected East Asian
countries, along with their motivation and objectives. They found that (i) the MPP and
CFM stances vary significantly among East Asian countries, partly due to the fact that
some East Asian economies (such as Hong Kong, China and Singapore) with an open
capital account and an exchange rate targeting and/or currency board system cannot
have independent monetary policy (impossible trinity). These countries mostly used
MPP instruments, such as caps on LTV ratios and debt-to-income (DTI) ratios, as well
as tax measures. The Republic of Korea, an East Asian economy with relatively high
financial development, has been an active user of both MPP and CFM. Banks in the
Republic of Korea have high exposure to short-term external debt because of growing
demand for foreign exchange derivative transactions. Financial imbalances have
emerged among domestic banks as well as in foreign exchange transactions, along
with the housing boom. Thus, the Republic of Korea implemented MPPs for property
financing. It also imposed a levy on bank non-deposit foreign exchange liabilities and a
ceiling on bank foreign exchange derivative positions to deter speculative flows.Korea
also implemented the CFM measure to deter foreign fund flows into their bond market.
On the other hand, developing East Asian economies such as the Peoples Republic
of China and India have a relatively closed capital account, which has enabled them
to rely heavily on traditional monetary policy tools like reserve requirement ratios
and some housing measures. The economies of the Association of Southeast Asian
Nations (ASEAN) have relied on a mixture of traditional monetary policy tools, MPP,
and CFM, which can complement each other. Indonesia and Thailand introduced
restrictions on non-residence access to the domestic bond market, minimum bond
holding periods, and withholding tax on bonds as part of their CFMs.

ADBI Working Paper 604

Saiki, Chantapacdepong, and Volz

5. EMPIRICAL ANALYSIS
5.1 Data: Sources and Summary Statistics
Appendix 1 lists our sample countries and their regional classifications. Most
importantly, our baseline East Asian countries are current members of ASEAN and the
so-called four tigers. We exclude the PRC as a unique case (very fast growth and
restricted financial activities) and Hong Kong, China because it is classified as a tax
haven by the Organisation for Economic Co-operation and Development (OECD),
and because it provides a loophole of speculative money in and out of the Peoples
Republic of China. This makes our baseline East Asian economies as follows:
Brunei Darussalam; Cambodia; India; Indonesia; the Republic of Korea; Lao Peoples
Democratic Republic; Malaysia; Myanmar; the Philippines; Singapore; Taipei,China;
Thailand; and Viet Nam. This is based primarily on ASEAN plus Singapore and the
Republic of Korea, which were affected by capital inflows after the quantitative easings.
We excluded the Peoples Republic of China because its domestic financial market
is heavily regulated and highly distorted, and its capital market is still very closed
therefore, there is strong pressure to circumvent the restriction to invest in foreign
countries for higher returns. We did not include Hong Kong, China because it is a
global banking center, often acting as a gateway of the Peoples Republic of Chinas
capital inflow and outflow, and it is designated as a tax haven by the OECD.
Table 1: Summary Statistics of Key Variables
Growth per Region
2000 2005
Baseline East Asia
7.0
6.4
Developing Europe
5.4
7.3
Latin America
2.8
4.9
South Asia
5.5
5.3
MENA
4.5
5.7
Sub-Saharan Africa
3.5
5.2
Non-OECD Members
4.1
5.5
OECD Members
4.3
2.5
Macroprudential Index
2000 2005
Baseline East Asia
Developing Europe
Latin America
South Asia
MENA
Sub-Saharan Africa
Non-OECD Members
OECD Members

1.1
0.9
1.9
1.8
1.6
0.8
1.3
0.9

1.9
1.6
2.7
2.5
1.9
1.0
1.9
0.9

Quinns Capital Openness Measure

2000
2005
2012

2012
5.1
2.5
2.9
5.9
9.9
3.7
4.3
0.0
2012
2.8
2.7
3.2
3.3
2.8
1.4
2.6
1.8

Baseline East Asia

52.9
53.8
Developing Europe
60.3
67.4
Latin America
79.7
77.1
South Asia
40.0
37.5
MENA
58.7
64.4
Sub-Saharan Africa
59.0
59.0
Non-OECD Members
97.2
97.2
OECD Members
61.5
62.2
Institutional Quality
2000
2005

2012

Baseline East Asia

Developing Europe
Latin America
South Asia
MENA
Sub-Saharan Africa
Non-OECD Members
OECD Members

0.0
0.4
0.6
0.5
0.1
0.6
0.5
1.4

0.2
0.2
0.3
0.6
0.4
0.7
0.5
1.4

0.1
0.5
0.6
0.6
0.1
0.5
0.5
1.3

72.2
85.4
77.9
41.7
73.2
71.6
98.9
73.5

MENA = Middle East and North America, OECD = Organisation for Economic Co-operation and Development.
Note: For institutional quality, a higher number means better quality.
Sources: Authors calculation based on the World Banks World Development Indicators; Quinns Capital Openness
Measure; Macroprudential Index by the International Monetary Fund; and Institutional Quality (see main text
for definition).

ADBI Working Paper 604

Saiki, Chantapacdepong, and Volz

To get a better perspective, rather than simply focusing on East Asia, we compare
other EMEs for key macroeconomic and institutional variables in Table 1. These
groups are mutually exclusive, i.e., a country belongs to only one regional group. The
definition of developed and developing (EME) countries follows the World Banks
definition. For the rest of this paper, we chose Latin American countries (see
Appendix 1 for the full list) for the comparison group. As commonly known, baseline
East Asia enjoyed a constant relatively high growth over the sample period, while its
capital openness is lower and its institutional quality is better than other developing
countries. As Cerutti, Claessens, and Laeven (2015) pointed out, the MPP is used
more often in developing countries than in advanced countries (see Table 1, lower
left panel).
In Table 2, we show the correlation of key variables on the cross-section basis as of
2005. Capital openness measures have a high correlation with institutional quality. In
particular, the correlation with the legal institutional quality is as high as about 0.5 for
both Quinn and ChinnItos capital openness measures. Financial openness has been
documented to have a close connection with governance. For this reason, we analyze
the role of capital openness separately from governance.
Table 2: Correlation of Capital Openness, Governance, and Real GDP per Capita
Quinn

Chinn
Ito

WB
EDB

MPI

Quinns Measure: Capital Openness (Quinn)

1.00

ChinnIto Capital Openness Index

0.91

1.00

Government Effectiveness (WBEDB)

0.52

0.46

1.00

0.54

0.42

0.34

1.00

0.51
0.51

0.50
0.50

0.83
0.83

0.29
0.29

Macroprudential Index (MPI)

Financial Depth
Real GDP per Capita

Fin. Depth

RGDPPC

1.00
0.35

1.00

GDP = gross domestic product, WBEDB = World BankEase of Doing Business.

Source: Authors.

5.2 Variables
We used a panel regression model with annual data covering the period 20012013
for Asian EMEs (excluding low-income countries with an income per capita of less
than US$1,460 per year as defined by the World Bank). Appendix 2 lists the variables,
definitions, and the sources.

5.2.1 Cross-Border Borrowing in Foreign Currency (Dependent Variable)

Our dependent variables are annual change of cross-border lending in foreign
currencies (in net terms) and exchange rate-adjusted change. The source is a
confidential version of the BIS locational banking statistics. This dataset comprises
loansboth in gross and net termsof BIS reporting banks to the rest of the world
(on a country level). We use the residential base version of cross-border lending data,
which is broken down by currencies (lenders currency versus borrowers currency). In
our analysis, we used the BIS reporting banks cross-border lending to Asian EMEs in
lenders currency (i.e., for borrowing EMEs, it is a foreign currency).

ADBI Working Paper 604

Saiki, Chantapacdepong, and Volz

The original source is on a quarterly basis, but because quarterly frequency data for
other key variables (especially the governance and capital openness data) 4 were
lacking, we summed the change over four quarters and defined it as an annual change.
The data enable us to break down, among other things, currency composition (i.e.,
lenders currency versus borrowers currency) of bank lending. Since interbank lending
and/or borrowing contains overnight market activities, or to cover the cash shortage on
a very short-term basis, we focus on cross-border lending to non-bank sectors in the
lenders currency. In what follows, we explain the variables of interest.
We assume that, other things being equal, countries with strong institutional settings
will attract more capital (because of more reliable rule of law, enforcement of
contracts, etc.).

5.2.2 Currency Valuation

Exchange rate appreciation would also make the borrowers currency stronger vis--vis
the borrowing currencies, thus a borrower has more incentive to borrow in lenders
currency, while a lender has more incentive to lend as the borrowers balance sheet
(asset side) looks strong. 5
However, when the home currency depreciates, a currency crisis occurs and
companies and household become over-leveraged.
Although there is no consensus on how to measure the equilibrium value of exchange
rates, the traditional approach to gauge real exchange rate overvaluation is as follows:
ln (, ) = ln

[1]

XRAT is the nominal exchange rate, and PPP is the purchasing power parity conversion
factor (both from the World Banks World Development Indicators database). The
rough interpretation is that if the log of the ratio of nominal exchange rate (local
currency per foreign) to PPP conversion factor is greater than 0, the currency is
undervalued, and vice versa. However, this suffers from the BalassaSamuelson effect
(Frankel 2006), which is one of the first studies to stress the importance of including the
GDP per capita into the exchange rate valuation. Thus, following Frankel (2006) and
Rodrik (2008), we regress this traditional measure of overvaluation on real GDP per
capita (in natural logs) with year fixed effects in panel-data setting, and take the error
term as the measure of undervaluation:

,
,

= + (, ) + + ,,

[2]

RGDPCH is GDP per capita. We confirm that the error term is normally distributed.
We define negative values of the residual as the degrees of undervaluation after
controlling for the BalassaSamuelson effect. Since we are interested in the effect of
overvaluation, we use ,, as our measure of overvaluation.
4

The Quinns openness measure is based on the qualitative information of the IMFs Annual Report
on Exchange Arrangement and Exchange Restrictions, which measures the financial openness
as of the end of the respective year. This makes conversion of the annual data to quarterly data
quantitatively difficult. Also, empirically, this means that it is appropriate to take one lag for the capital
openness measure.
Shin (2015) pointed out that exchange rate appreciation is becoming expansionary, not contractionary
as the traditional MundellFleming model claims, because the financial channel is dominating the
goods channel.

10

ADBI Working Paper 604

Saiki, Chantapacdepong, and Volz

5.2.3 Macroprudential Index

Another key variable we brought into our model is the MPI developed by Cerutti,
Claessens, and Laeven (2015). This dataset covers 119 countries over 20002013.
The MPI covers many instruments such as loan-to-value ratio, reserve requirements,
and debt-to-income ratio. We use the borrower-targeted macroprudential tool index
(henceforth BTM), as it is heavily used in East Asia. The higher the MPI, the more
macroprudential measures are in place and/or strictly implemented. This is based
on the IMFs Global Macroprudential Policy Instruments survey of each countrys
executives (Cerutti, Claessens, and Laeven 2015). The MPI is the
sum of a (different combination of) binary valuation of 12 tools (LTV, DTI, etc.). As
mentioned above, EMEs employ MPIs more frequently than advanced countries, and
MPIs are less effective for countries with more open and deeper financial markets
(as one will find in advanced economies), because there is more room for regulatory
arbitrage, i.e., investors simply investing somewhere else. This is also in line with the
argument of Engel (2015), who stressed the importance of international coordination in
macroprudential policies. However, the very nature of MPPtied to the economic cycle
of an individual countrymakes it very difficult to coordinate the policy across
countries. Although Ceruttis database was the first attempt to comprehensively
quantify macroprudential measures, it has some shortcomings, which may be improved
in future versions. First, it only targets domestic (residence based) banks (not
insurance companies, hedge funds, shadow banking, etc.). Second, it is a binary
variable (1 when the measure is in place, 0 otherwise), so it fails to capture the reach
and depth of its implementation and enforcement, let alone efficacy. 6 Various MPP
measures were taken in East Asian EMEs, in different forms, and often combined with
capital flow measures, but LTV is the most frequently used measure in East Asia
(for details, see Lee, Asuncion, and Kim 2015).
Macroprudential policies can be divided into borrower-based (caps on LTV and DTI
ratio) and lender-based, but it has been shown that for East Asia, borrower-based
income has been used extensively and showed some impact (Cerutti, Claessens, and
Laeven 2015; Bruno, Shim, and Shin 2015).

5.2.4 Institutional Quality Database

Among the various indicators for institutional quality, the most popular one is the
International Country Risk Group database. But in 1997 the World Bank started issuing
its own estimations of institutional quality (now called Ease of Doing Business [World
Bank EDB]). It was published biannually until 2002, and annually thereafter. We use
this World Bank measure as it is more focused on economic factors, rather than the
International Country Risk Group index, which contains many other factors such as
crime rate and ethnic tensions. 7
The World Bank data consist of several components that mainly relate to business
climate (getting credit, tax systems, etc.). Here, we use the governments effectiveness
as strong legal framework and enforcement are relevant for overseeing and regulating,
if necessary, cross-border lending. The indicator is compiled from many sources, such
as the Economist Intelligence Unit and the Asian Development Bank, and defined as
capturing perceptions of the quality of public services, the quality of the civil service
6

The measures developed by Bruno, Shim, and Shin (2015) also capture the scope of each instrument,
but as the sample size is rather small (60 countries), we use the index of Cerutti, Claessens, and
Laeven (2015).
That said, there is a very high correlation between these two measures. (The correlation between the
principal component of ase of Doing BusinessT and our measure is 0.97).

11

ADBI Working Paper 604

Saiki, Chantapacdepong, and Volz

and the degree of its independence from political pressures, the quality
of policy formulation and implementation, and the credibility of the governments
commitment to such policies (World Bank, Doing Business). The credibility and
commitment of public institutions are at the core of this index. This can also be
interpreted as how prudent the government is perceived to be.
A strong institutional and governance framework helps MPPs to be more effective, as
good governance would result in a healthy central bank, regulatory authorities, and
relevant government institutions via strength of powers and clear accountability (IMF
2013). Numerous studies have been done on the positive relationship between good
governance and per capita income growth especially for lower-income countries;
however, to the best of our knowledge, no scholarly studies examine how institutional
factors would safeguard financial stability. In Appendix 5, we present the mean and
2 standard deviations of baseline Asia and Latin America. On average, baseline
Asia has higher government effectiveness, but perhaps because of a smaller sample
size compared with Latin American countries, the standard deviation is higher for
baseline Asia.

5.2.5 Other Variables

We also included other variables to control country-specific factors that affect
cross-border foreign-currency borrowing by EMEs such as financial depth; the real
interest rate differential from the United States benchmark rate; economic cycle;
stock of foreign direct investment (FDI) per GDP; trade openness defined as (export
+ import)/GDP; global liquidity (source: BIS) defined as total credit outstanding of
international banks; and VIX (the Chicago Board Options Exchange [CBOE] Volatility
Index). The source and description are presented in Appendix 2.

6. MODEL
Because of the weakness of the MPI as discussed above, we included the government
effectiveness index to measure a governments prudency. The basic idea is that a
prudent government would closely monitor foreign-currency liabilities and take effective
actions. Our model is as follows:
, = , + 1 , + 2 , + 3 ,
, + , + ,

[3]

Y is the annual change in foreign-currency cross-border borrowing (exchange rate

adjusted, BIS reporting banks lending to sample countries in lenders currency) per
GDP of the recipient country.
We used clustered standard errors (country level) to allow for unobserved correlation
in errors within a country: if we do not control this, the standard errors would be
underestimated and the significance overestimated (Baum, Nichols, and Schaffer
2011). We conducted a panel unit-root test to make sure that all the variables used are
stationary. We divided samples into two time periods: (i) the pre-quantitative easing
period (20012007, so-called global saving glut period); and (ii) the quantitative
easing period (20082013). The sample period is divided this way because, before the
GFC, capital was flowing from emerging economies to advanced economies (Lucas
paradox), largely to the US, fueling housing bubbles, which was a major contributor to
the GFC. As discussed, global liquidity after the quantitative easing started was still
high, but it is a result of quantitative easing policy, so the source of global liquidity was

12

ADBI Working Paper 604

Saiki, Chantapacdepong, and Volz

very different. Going further, global liquidity is likely to decline as the Feds monetary
policy has been in a tightening cycle since December 2015. More importantly, after the
quantitative easing started, the capital flow reversedit is now flowing from advanced
economies to EMEs.
Dividing our sample into these two periods gives us a clear picture of how EMEs coped
with capital inflows as a result of quantitative easing. We looked at selected East Asian
countries (Baseline East Asia) and Latin American countries (for comparison). We
excluded banking centers and tax haven countries (Appendix 3) from our sample,
as capital flows to these countries are driven by different motives, rather than
macroeconomic fundamentals or institutions.
Second, we looked at the role of capital openness, using Quinns measure of
capital openness. 8
, = , + 1 ,1 + 2 , + 3 , , + , + ,

[4]

, = , + 1 , + 2 , + 3 , , + , + ,

[5]

Further, to test the effectiveness of macroprudential policy, we use the MPI (borrowers
component) by Cerutti, Claessens, and Laeven (2015).

7. RESULTS

Table 3 presents the estimation results (columns 3 and 4 for pre-crisis and post-crisis
East Asia, and columns 7 and 8 for Latin America). Interestingly, capital openness
is negative, perhaps because this is net bank lending. One potential explanation is
that countries with open capital markets like Singapore (a BIS-reporting country) was
lending to other countries. Or, more capital-open countries have better governance
(as we saw in the correlation table), so the respective government was taking
measures to stop excessive capital inflows, which leads to an asset bubble. For the
MPI, as we predicted from the shortcomings of the data discussed above, we did
not get significant results. The coefficients of overvaluation and its cross-term with
MPIBTM are significant before the crisis, possibly indicating that the MPI was actively
employed before the crisis in East Asia. Since we are interested in how our
sample countries reacted to quantitative easing, we look at the second column. There,
we can clearly see the negative correlation between government effectiveness
and foreign-currency cross-border borrowing. Notably, the cross-term between
overvaluation and government effectiveness has been negative and statistically
significant for both regions since 2008. This indicates that the incentive to borrow in
foreign currency when the home currency is strong was reduced probably because
efficient governments take action against overborrowing.

We did not use the ChinnIto index because it is slow moving by design (because of its 5-year average
calculation), and it does not capture the sudden change in capital flow management policy, which
frequently occurred during this period. Since the IMFs Annual Report on Exchange Arrangements and
Exchange Restrictions, on which Quinns measure is based, takes the snapshot of capital openness at
the end of the year (December every year), we take a lag of capital openness.

13

ADBI Working Paper 604

Saiki, Chantapacdepong, and Volz

Table 3: Cross-Country Foreign Currency Borrowing

and the Role of Good Governance and Capital Openness
[1]
Exchange rate overvaluation
Government effectiveness
Overvaluation x Government
effectiveness

0.00615
[.9393]
0.00508
[1.569]
0.00164
[.2866]

Quinns capital opennness

(1 lag)
Overvaluation x Quinns
capital opennness (1 lag)
Financial depth
Interest rate difference from
the US real interest rate
Cycle (HP filtered)
FDI per GDP
Trade Openness
(EX+IM)/GDP
Global Liquidity
VIX
Number of observations
R-squared

Exchange rate overvaluation

Government effectiveness
Overvaluation x Government
effectiveness
Quinns capital opennness
(1 lag)
Overvaluation x Quinns
capital opennness (1 lag)

0.0000474
[1.227]
0.000121
[.5008]
0.177*
[1.781]
0.0000136
[.5503]
0.00000779
[.646]
0.00230
[.7563]
0.000347
[1.512]
63
0.2866
[5]
0.0173
[1.01]
0.00691
[.6764]
0.0135
[.7469]

[2]
[3]
Baseline Asia Pre-crisis
0.00871
0.00505
[1.074]
[.4148]
0.00779**
[2.263]
0.0189**
[1.974]
0.000119*
[1.791]
0.0000386
[.1934]
0.000104***
0.0000363
[3.281]
[1.242]
0.000106
0.0000791
[1.239]
[.5839]
0.0267
0.142
[.4919]
[1.296]
0.0000678***
0.0000321
[2.812]
[.9941]
0.0000716***
0.0000126
[4.022]
[.7218]
0.0102
0.00590
[.9556]
[1.477]
0.00162
0.0000215
[.9399]
[0.1065]
40
64
0.2269
0.29
[6]
[7]
Latin America Pre-crisis
0.198**
0.0119
[2.155]
[.9112]
0.00209
[.075]
0.402***
[2.956]
0.00000408
[.0372]
0.00000495
[.0336]

[4]
0.0200
[.6874]

0.0000807
[.6344]
0.000428
[.7332]
0.0000385
[1.287]
0.000127
[.2708]
0.0990*
[1.83]
0.0000124
[.2617]
0.0000157
[.539]
0.0132
[.7094]
0.00192
[.6882]
30
0.2085
[8]
0.0265
[.3333]

0.000632**
[2.078]
0.000444
[.3713]

continued on next page

14

ADBI Working Paper 604

Saiki, Chantapacdepong, and Volz

Table 3 continued
[5]
Financial depth
Interest rate difference from
the US real interest rate
Cycle (HP filtered)
FDI per GDP
Trade Openness
(EX+IM)/GDP
Global Liquidity
VIX
Number of observations
R-squared

[6]
[7]
Latin America Pre-crisis
0.000328
0.000160
[.3513]
[.7414]
0.000372
0.0000807
[.4387]
[.3803]
0.0760
0.00973
[.2004]
[.1046]
0.000117
0.000231
[.4348]
[.8417]
0.000101
0.000151
[.2596]
[1.011]
0.0292
0.00300
[.7572]
[.3872]
0.00448
0.00136
[.9272]
[1.233]
85
132
0.372
0.1446

0.000138
[.3954]
0.000102
[.3819]
0.133
[.8202]
0.00000620
[.0516]
0.0000892
[.5917]
0.00678
[.7523]
0.00129
[1.373]
132
0.0816

[8]
0.000402
[1.096]
0.00191**
[2.525]
0.380
[.5783]
0.000655
[1.154]
0.00147***
[4.019]
0.00536
[.4698]
0.00224*
[1.862]
53
0.6341

Cycle (HP filtered) = HP-filtered cycle of log of real GDP in local currency term (lambda: 6.25), EX = export,
FDI = foreign direct investment, Financial depth = domestic private credit to the real sector by deposit money banks as
percentage of local currency, GDP = gross domestic product, IM = import, VIX = CBOE Volatility Index.
Notes:
1. t statistics in brackets.
2. ***, **, * denote statistical significance at 1%, 5%, and 10% level, respectively.
3. Raw data from the World Bank; Bank for International Settlements; International Monetary Fund-International
Financial Statistics; Federal Reserve Economic Data database; Organisation for Economic Co-operation and
Development; Dharmapala and Hines (2006); and Cerutti, Claessens, and Laeven (2015).
Source: Authors.

Table 4: Coss-Country Foreign Currency Borrowing

and the Role of Macroprudential Index

Overvaluation
MPI-BTM
(Borrower targeted measure)
MPI-BTM x Overvaluation
Real interest rate difference
from the US
HP filtered cycle
FDI per GDP

[1]
[2]
Baseline Asia
Pre-crisis
After Crisis
0.0101*
0.00128
[1.702]
[.1716]
0.00431
0.00113
[1.01]
[.2173]
0.0177*
0.00130
[1.706]
[.2235]
0.000320**
0.000146
[2.083]
[1.339]
0.164
0.0159
[1.552]
[.288]
0.0000206
0.0000313
[.6761]
[.6469]

[3]
[4]
Latin America
Pre-crisis
After Crisis
0.00445
0.0511
[.3389]
[.8074]
0.00470
0.00403
[.9968]
[.4823]
0.0194
0.000373
[.9287]
[.0079]
0.00000784
0.000578
[.0422]
[1.317]
0.315
0.0970
[.7779]
[.3392]
0.0000223
0.0000271
[.1327]
[.1097]
continued on next page

15

ADBI Working Paper 604

Saiki, Chantapacdepong, and Volz

Table 4 continued

Trade (EX+IM)/GDP
Financial depth
log(logal liquidity)
VIX
R-squared
Observation

[1]
[2]
Baseline Asia
Pre-crisis
After Crisis
0.0000360
0.0000502
[1.501]
[1.058]
0.0000655
0.0000634
[.8639]
[.7187]
0.00433
0.0113
[1.191]
[.8392]
0.0000637
0.00190
[.5255]
[.8081]
0.3581
0.1987
64
36

[3]
[4]
Latin America
Pre-crisis
After Crisis
0.0000235
0.0000412
[.2981]
[.295]
0.000386
0.000450
[1.208]
[.7609]
0.00463
0.0515
[.5754]
[1.01]
0.000166
0.00596
[1.038]
[.9354]
0.1057
0.1101
94
54

Cycle (HP filtered) = HP-filtered cycle of log of real GDP in local currency term (lambda: 6.25), EX = export,
FDI = foreign direct investment, Financial depth = domestic private credit to the real sector by deposit money banks as
percentage of local currency, GDP = gross domestic product, IM= import, VIX = CBOE Volatility Index.
Notes:
1. t statistics in brackets.
2. ***, **, * denote statistical significance at 1%, 5%, and 10% level, respectively.
3. Raw data from the World Bank; Bank for International Settlements; International Monetary Fund-International
Financial Statistics; Federal Reserve Economic Data database; Organisation for Economic Co-operation and
Development; Dharmapala and Hines (2006); and Cerutti, Claessens, and Laeven (2015).
Source: Authors.

Figure 4: The Coefficients of Interactive Term

(overvaluation governance)

Source: Authors.

16

ADBI Working Paper 604

Saiki, Chantapacdepong, and Volz

Interpreting the coefficients on cross-terms of two continuous variablesgovernance

and overvaluationrequires caution, as the coefficients would change as the value of
either continuous variable changes. To check this, we drew a scatter plot (Figure 4).
The line denotes the coefficient, and the x-axis and y-axis denote two continuous
variables. The dotted line shows values of two standard deviations. From the graph,
the significance is clearly robust to different values of either one of the continuous
variables. This means that more effective governments curbed cross-border foreigncurrency borrowing. Government effectiveness itself was only significant for East Asia
after the crisis, indicating a more proactive approach on the government side. For the
MPI, we observed a significant effect only before quantitative easing for East Asia.
This may be a result of the governments ongoing effort to prevent a repeat of the 1997
Asian financial crisis. Before the GFC, East Asia was facing high pressure of capital
inflow because of its robust economic growth. But this needs further examination.

8. CONCLUSIONS AND POLICY IMPLICATIONS

In this paper, we looked into the role, governance, and macroprudential measures to
deal with cross-border bank lending to EMEs against the backdrop of quantitative
easing of advanced economies. In this context, we paid particular attention to the effect
of quantitative easing-related capital inflows on exchange rate valuations and foreigncurrency borrowing of EMEs. We also investigated the role of institutions to curb
overborrowing in foreign currencies. Our key findings are as follows: (i) After the GFC,
there is a clear indication that a more effective government can curb foreign-currency
borrowing; this holds in East Asia and Latin America; and (ii) East Asian countries have
been more proactive in responding to excessive credit buildup, i.e., the overvaluation
effect on foreign-currency borrowing. The policy implication is that, since East Asia has
a higher score in government effectiveness, it is less likely, especially compared with
Latin American countries, to suffer from the negative result of overborrowing (currency
mismatch, sudden stop, balance-of-payment crisis, etc.) in the near future.
In sum, countries with good governance and the right macroprudential measures in
place are less likely to suffer from currency mismatch as well as deleveraging as a
result of overborrowing in foreign currencies.

17

ADBI Working Paper 604

Saiki, Chantapacdepong, and Volz

REFERENCES
Akca, O. 2011. Exchange Rate Fluctuations and Constraints of the Global Financial
Architecture: Implications during the 2008 Global Financial Crisis. PhD
Dissertation. http://gradworks.umi.com/34/71/3471756.html (accessed
7 September 2016).
Arteta, C. 2005. Exchange Rate Regimes and Financial Dollarization: Does Flexibility
Reduce Bank Currency Mismatches? In Berkeley Electronic Journal of Macroeconomics, edited by T. Cavalcanti and K. Mertens.
http://www.bepress.com/bejm/topics/vol5/iss1/art10 (accessed
7 September 2016).
Baum, C. F., A. Nichols, and M. E. Schaffer. 2011. Evaluating One-way and Two-way
Cluster Robust Covariance Matrix Estimates. German Stata Users Group
Meeting. July. http://fmwww.bc.edu/repec/dsug2011/desug11_schaffer.pdf
(accessed 7 September 2016).
Bianchi, J. 2011. Overborrowing and Systemic Externalities in the Business Cycle.
American Economic Review 1017: 34003426.
Bruno, V., I. Shim, and H. S. Shin. 2015. Comparative Assessment of Macroprudential
Policies. BIS Working Papers 502. Basel, Switzerland: Bank for International
Settlements.
Cerutti, E., S. Claessens, and L. Laeven. 2015. The Use and Effectiveness of
Macroprudential Policies: New Evidence. IMF Working Papers 15/61.
Washington, DC: International Monetary Fund.
Chantapacdepong, P., and I. Shim. 2014. Correlations across Asia-Pacific Bond
Markets and the Impact of Capital Flow Measures. BIS Working Papers 472.
Basel, Switzerland: Bank for International Settlements.
Chinn, M. D., and H. Ito. 2008. A New Measure of Financial Openness. Journal of
Comparative Policy Analysis 103: 309322.
Dharmapala, D., and J. R. Hines, Jr. 2006. Which Countries Become Tax Havens?
NBER Working Papers 12802. Cambridge, MA: National Bureau of Economic
Research.
Eichengreen, B. J., R. Hausmann, and U. Panizza. 2003. Currency Mismatches, Debt
Intolerance and Original Sin: Why They are Not the Same and Why it Matters.
NBER Working Paper 10036. Cambridge, MA: National Bureau of Economic
Research.
Engel, C. 2015. Macroprudential Policy in a World of High Capital Mobility: Policy
Implications from an Academic Perspective. NBER Working Paper 20951.
Cambridge, MA: National Bureau of Economic Research.
Fischer, S. 1998. Capital-Account Liberalization and the Role of the IMF. In Should the
IMF Pursue Capital-account Convertibility? Essays in International Finance No.
207, edited by S. Fischer et al. Princeton, NJ: Princeton University Department
of Economics.
Forbes, K., M. Fratzscher, and R. Straub. 2015. Capital-flow Management Measures:
What are They Good for? NBER Working Paper 20860. Cambridge, MA:
National Bureau of Economic Research.

18

ADBI Working Paper 604

Saiki, Chantapacdepong, and Volz

Frankel, J. 2006. The BalassaSamuelson Relationship and the Renminbi. Harvard

University Working Paper. Cambridge, MA: Kennedy School of Government,
Harvard University.
. 2015. The International Finance and Macroeconomic Program. NBER
Reporter 2.
Frankel, J., and A. Saiki. 2002. A Proposal to Anchor Monetary Policy by the Price of
the Export Commodity. Journal of Economic Integration 17: 417448.
Frankel, J., and G. Saravelos. 2010. Are Leading Indicators of Financial Crises Useful
for Assessing Country Vulnerability? Evidence from the 200809 Global Crisis.
NBER Working Paper 16047. Cambridge, MA: National Bureau of Economic
Research.
Frankel, J., and A. Rose. 2002. An Estimate of the Effect of Common Currencies on
Trade and Income. The Quarterly Journal of Economics 117(2): 437466.
Friberg, R., and F. Wilander. 2008. The Currency Denomination of ExportsA
Questionnaire Study. Journal of International Economics 751: 5469.
Galati, G., and R. Moessner. 2013. Macroprudential PolicyA Literature Review.
Journal of Economic Surveys 27(5): 846878.
Glick, R., and S. Leduc. 2013. The Effects of Unconventional and Conventional US
Monetary Policy on the dollar. FRB San Francisco Working Paper 201311.
San Francisco, CA: Federal Reserve Bank of San Francisco.
Goldberg, L. S., and C. Tille. 2009. Micro, Macro, and Strategic Forces in International
Trade Invoicing. Staff Reports 405. New York, NY: Federal Reserve Bank of
New York.
International Monetary Fund (IMF). 2013. Global Impact and Challenges of
Unconventional Monetary Policies. IMF Policy Paper 2013. Washington, DC:
Washington, DC: International Monetary Fund.
. 2013. Key Aspects of Macroprudential Policy. IMF Policy Report. Washington,
DC: International Monetary Fund.
. 2011. The Multilateral Aspects of Policies Affecting Capital Flows. IMF Policy
Report. Washington, DC: International Monetary Fund.
. Various issues. Spill-Over Report. Washington, DC: International Monetary
Fund.
Korinek, A. 2010. Regulating Capital Flows to Emerging Markets: An Externality View.
Mimeo.
Laeven, L., and F. Valencia. 2012. Systemic Banking Crises Database: An Update.
IMF Working Paper 12/163. Washington, DC: International Monetary Fund.
Lee, M., R. C. Asuncion, and J. Kim. 2015. Effectiveness of Macroprudential Polices in
Developing Asia: An Empirical Analysis. ADB Economics Working Papers 409.
Manila: Asian Development Bank.
Ostry, J., A. Ghosh, K. Habermeier, L. Laeven, M. Chamon, M. Qureshi, and
A. Kokenyne. 2011. Managing Capital Inflows: What Tools to Use? IMF Staff
Discussion Notes 11/06. Washington, DC: International Monetary Fund.
Quinn, D. P., M. Schindler, and A. M. Toyoda. 2011. Assessing Measures of Financial
Openness and Integration. IMF Economic Review 59(3): 488522.

19

ADBI Working Paper 604

Saiki, Chantapacdepong, and Volz

Rajan, R. 2013. A Step in the Dark: Unconventional Monetary Policy after the Crisis.
First Andrew Crockett Memorial Lecture at the Bank for International
Settlements. 23 June.
Rodrik, D. 2008. The Real Exchange Rate and Economic Growth. Brookings Papers on
Economic Activity 39(2): 365439.
Schularick, M., and A. M. Taylor. 2012. Credit Booms Gone Bust: Monetary Policy,
Leverage Cycles, and Financial Crises, 18702008. American Economic
Review 102(2): 10291061.
Shin, H. S. 2015. Exchange Rates and the Transmission of Global Liquidity. Speech at
the Bank of KoreaIMF Conference in Seoul. 11 December.
World Bank. Doing Business. various issues. http://www.doingbusiness.org/
reports/global-reports/doing-business-2016 (accessed 1 December 2015).

20

ADBI Working Paper 604

Saiki, Chantapacdepong, and Volz

APPENDIX 1: SAMPLE DATA COUNTRIES

Baseline East Asia

Latin America

Brunei Darussalam

Argentina

Guyana

Cambodia

Barbados

Honduras

India

Belize

Jamaica

Indonesia

Brazil

Mexico

Lao Peoples Democratic Republic

Chile

Nicaragua

Malaysia

Colombia

Panama

Myanmar

Costa Rica

Paraguay

Philippines

Dominica

Peru

Republic of Korea

Dominican Republic

St. Vincent and Grenadines

Singapore

Ecuador

Suriname

Taipei,China

El Salvador

Trinidad and Tobago

Viet Nam

Grenada

Uruguay

Guatemala

Venezuela

Source: Authors compilation.

21

ADBI Working Paper 604

Saiki, Chantapacdepong, and Volz

APPENDIX 2: VARIABLES AND DATA SOURCES

Variable
Net foreign currency
borrowing

Description

Source

Exchange rate adjusted change in net cross-border

borrowing from BIS reporting bank countries in lenders
currency, on residential basis.
See equation [1] in the main text for the definition

BIS Locational Banking

Statistics

Government
Effectiveness

Government effectiveness captures perceptions of the

quality of public services, the quality of the civil service
and the degree of its independence from political
pressures, the quality of policy formulation and
implementation, and the credibility of the governments
commitment to such policies.

The World Banks Ease of

Doing Business

Financial Depth

Domestic private credit to the real sector by deposit

money banks as percentage of local currency GDP

World Bank WDI

Interest Rate relative to

US

Real interest rate of home country minus US real interest

rate

IMFIFS

HP-filtered cycle

HP filtered cycle of log of real GDP in local currency term

(lambda: 6.25)

World Bank WDI

FDI/GDP
(EX+IM)/GDP

Inflow of FDI divided by GDP

Exports and Imports divided by GDP (Trade openness)

World Bank WDI

World Bank WDI

Log (Global Liquidity)

Global credit it aggregates (banks + nonbanks)

BIS Global liquidity database

VIX

CBOE Volatility Index

FRED database

Offshore

Offshore market according to OECD definition

OECD

Tax Haven
MPIBTM

Tax Haven
Borrower targeted macroprudential measure

Dharmapala and Hines (2006)

Cerutti et al. (2015)

Overvaluation

Calculated based on data

from the World Bank WDI

BIS = Bank for International Settlements, WDI = World Development Indicators, US = United States, IMF = International
Monetary Fund, IFS = International Financial Statistics, HP = Hodrick-Prescott, GDP = gross domestic product, FDI =
foreign direct investment, VIX = Implied volatility of S&P500 index, CBOE = Chicago Board Options Exchange, FRED =
FRED economic data by the Federal Reserve Bank of St. Louis, OECD = Organisation for Economic Co-operation and
Development, MPIBTM = Borrower targeted macroprudential measure.
Source: Authors compilation.

22

ADBI Working Paper 604

Saiki, Chantapacdepong, and Volz

APPENDIX 3: LIST OF TAX HAVENS AND BANKING

CENTER COUNTRIES
Tax Havens

Banking Center

Andorra

Lebanon

Ireland

Anguilla

Liberia

Liberia

Antigua and Babuda

Liechtenstein

Luxembourg

Bahamas

Luxembourg

Malta

Bahrain

Macao

Panama

Barbados

Maldives

Samoa

Belize

Malta

Salomon Ireland

Bermuda

Marshall Islands

Singapore

British Virgin Islands

Monaco

South Sudan

Cayman Islands

Montserrat

St. Kitts and Nevis

Channel

Netherlands Antilles

Switzerland

Cook Islands

Panama

Tuvalu

Cyprus

St. Kitts and Nevis

Vanuatu

Dominica

St. Vincent and Grenadines

Gibraltar

Seychelles

Grenada

Singapore

Hong Kong, China

Turks and Islands

Ireland

Vanuatu

Isle of Man
Jordan
Note: See Dharmapala and Hines (2006) for a definition of tax havens and Quinn et al. (2011) for banking center.
Sources: Dharmapala and Hines (2006); and Quinn et al. (2011).

23

ADBI Working Paper 604

Saiki, Chantapacdepong, and Volz

APPENDIX 4: CAPITAL INFLOWSOUTFLOW

RESTRICTION MEASURES AND MACROPRUDENTIAL
INDEX
East Asia and India

continued on next page

24

ADBI Working Paper 604

Saiki, Chantapacdepong, and Volz

Figure continued

continued on next page

25

ADBI Working Paper 604

Saiki, Chantapacdepong, and Volz

Latin America

continued on next page

26

ADBI Working Paper 604

Saiki, Chantapacdepong, and Volz

Emerging Europe

MPI = macroprudential index, RHS = right-hand side.

Note: Higher capital index corresponds to higher restriction.
Source: International Monetary Fund Schindlers Index and Macroprudential Index.

27

ADBI Working Paper 604

Saiki, Chantapacdepong, and Volz

APPENDIX 5: GOVERNMENT EFFECTIVENESS

SCORES OF SAMPLE COUNTRIES (BASELINE ASIA
AND LATIN AMERICA, MEAN AND 2 STANDARD
DEVIATIONS)

Source: Authors calculation based on the World Banks Ease of Doing Business.

28