Asian Economic Integration Report 2022

Despite uncertainties about the coronavirus disease (COVID-19) pandemic, financial markets remained relatively calm in the first half (H1) of 2021, compared with 2020. Supportive fiscal and monetary policy measures and vaccination rollout lifted growth prospects and sustained favorable financial conditions in Asia and the Pacific and elsewhere. Global financial stress has trended downward since the second quarter (Q2) of 2020. And financial stress indexes in advanced and selected Asia and Pacific economies have declined since early 2021 as investor sentiment improved and accommodative policy measures remained. The financial stress indexes in the euro area, the United Kingdom, and the United States (US)— and selected Asia and Pacific economies, including Hong Kong, China; India; Indonesia; Japan; the People’s Republic of China (PRC); the Philippines; Singapore; and Thailand—showed no signs of stress in financial markets during March to July 2021 (Figure 4.1). Investor risk appetite also improved. The Chicago Board Options Exchange’s volatility index (VIX), a measure of risk aversion, has likewise continued to trend downward, approaching its pre-pandemic level (Figure 4.2).


4
Continued accommodative policy and stronger growth prospects due to vaccine rollout in Asia and the Pacific and elsewhere buoyed financial conditions in the first half of 2021, but financial uncertainties emerged in the latter part of the year.  Sovereign credit default swaps of selected Asian economies have also declined from peaks in March-May of 2020 (Figure 4.3). As of July 2021, sovereign credit default swaps have dropped below pre-pandemic levels for Japan and the Republic of Korea, and stayed above pre-pandemic levels for others. In addition, short-term US dollar funding markets have returned to pre-pandemic levels for most of 2021, along with offshore US dollar funding costs as measured by foreign currency basis swaps versus the US dollar (Figure 4.4). Consequently, these measures indicate favorable financial market conditions in the first 7 months of 2021.
However, financial uncertainties emerged in the second half of 2021. Strong growth in advanced economies, such as the US, and inflation concerns signaled earlier monetary policy normalization than in emerging and developing economies (Knightley and Garvey 2022). Such a scenario could lead to tighter liquidity conditions in emerging and developing economies, including those in Asia and the Pacific, and could result in lower capital inflows or capital flow reversals and further weakening of the region's currencies. In addition, the ongoing financial woes of Evergrande in the PRC property and housing sector adds to uncertainties over its domestic and cross-border financial spillovers. Consequently, the regional financial stress indexes, global Notes: A credit default swap is a financial derivative that insures against the risk of default by one party. A higher index value reflects a higher spread, which is associated with higher default risk.
Source: ADB calculations using data from Bloomberg. Prices of financial assets in the region have diverged in 2021 but the region's total nonresident capital inflows remained robust, while its currencies weakened.
Stock prices in the region have recovered from their lowest point in 2020. However, equity prices have diverged across the region in 2021. Benchmark stock price indexes in Australia; India; Kazakhstan; Sri Lanka; Taipei,China; Thailand; and Viet Nam have grown by more than 10% since the start of 2021 up to 10 December 2021. Those in Indonesia, Japan, the Philippines, the PRC, the Republic of Korea, and Singapore have grown less than 10%; while share prices in Malaysia and Hong Kong, China have declined in value since the start of 2021 (Figure 4.5). The prices of sovereign bonds of selected Asia and Pacific economies also diverged in 2021, following their recovery in late 2020 from a slight drop in March 2020 (Figure 4.6).    -. . -. -. -. -.
In particular, the values of sovereign bonds of Australia; Hong Kong, China; Kazakhstan; Malaysia; the Philippines; the Republic of Korea; Singapore; and Thailand slightly dipped in 2021, while those for India, Indonesia, Japan, and the PRC slightly increased, suggesting diverging bond price movements due to various economic factors associated with uneven economic recoveries, varying pace of vaccine rollout, and differences in policy support measures.
The region's nonresident capital inflows continued to increase in 2021, reaching around $372 billion for select Asia and Pacific economies in Q2 2021, a 175% increase from Q2 2020 (Figure 4.7). Nonetheless, the volatility of nonresident capital inflows for select economies in the region increased slightly in H1 2021 compared with H1 2020. The sustained increase in nonresident capital inflows in 2021, follows the increase of capital inflows in 2020 to $1.6 trillion from $1.2 trillion in 2019, mainly due to increases in other accounts payable, currency and deposits, as well as debt inflows including portfolio debt and loans. In contrast, equity inflows including foreign direct investment (FDI) and portfolio equity decreased by 30% in 2020, compared with 2019. Moreover, the volatility of capital inflows inched higher in 2020 compared with 2019 as volatilities for loans and portfolio inflows have gone up.
Regional currencies have mostly weakened against the US dollar in 2021 on expectations of stronger recovery in the US compared with other economies and softening regional growth prospects in the second half of 2021 ( Figure 4.8). Regional currencies have weakened against the US dollar on a year-to-date basis in 2021, with the Australian dollar, baht, Sri Lankan rupee, yen, and won dropping by more than 5%; while the Hong Kong dollar, the Indian rupee, peso, ringgit, rupiah, Singapore dollar, som, and tenge have declined by less than 5% since the start of 2021.
Monetary policy in the region was mostly accommodative in 2021 as economies weathered the uncertainties of the ongoing pandemic.
Despite continuous efforts to curb the pandemic through containment measures and improved vaccine rollouts in 2021, Asia and Pacific economies and elsewhere continued to face uncertainties over the outcome of the pandemic as novel variants of the COVID-19 virus emerged, including the highly transmissible Delta and Omicron variants. To keep economies afloat and ease investor concerns, central banks in Asia kept policy rates low. With the exception of Taipei,China, policy rates in selected Asian economies in mid-2021 were mostly lower compared with March 2020. Taipei,China held its policy rate at 1.1% ( Figure  4.9). But rising inflation concerns in the second half of 2021 prompted some regional central banks to raise policy rates. The Bank of Korea raised its policy rate from 0.5% to 0.75% in August 2021, then to 1.0% in November 2021, while the Central Bank of Sri Lanka raised its policy rate from 4.5% to 5.0% in the same period. The Reserve Bank of New Zealand also raised its policy rate to 0.5% in October 2021, after keeping it at 0.25% since March 2020.
To further curb exchange rate pressures and volatility, and to keep supporting foreign exchange rate liquidity, the US Federal Reserve extended its temporary US dollar swap lines, established in March 2020, up to December 2021 In the region, some bilateral currency swap arrangements were renewed in 2021, notably, between those of the PRC and Thailand; Canada and the PRC; as well as between the Republic of Korea and Switzerland (Cantú et al. 2021).
The global nature of the ongoing COVID-19 pandemic was reflected in the increase in the share of global shocks in the variation of Asian financial asset price returns.
The share of global shocks that explains the variation of equity returns in Asia increased from 19.8% at the onset of the COVID-19 pandemic in the first quarter of 2020 to around 20.4% during the pandemic from April 2020 to December 2021 (Figure 4.10). The share of regional shocks during the pandemic, likewise, grew from 7.2% at the onset of the pandemic to 9.2%. Across subregions, South Asia's equity markets witnessed a large increase in sensitivity to global and regional shocks between both periods. In contrast, responsiveness to global and regional shocks dropped noticeably in the East Asia subregion. Meanwhile, the share of domestic shocks explaining the variation of equity returns dropped from 73.0% in the COVID-19 onset period to around 70.4%

29
The temporary US dollar liquidity swap lines were extended up to 31 December 2021 (Board of Governors of the Federal Reserve System 2021a, 2021b). during the pandemic, suggesting the importance of global and regional shocks over domestic shocks during the pandemic period.
Compared with the 2008-2009 global financial crisis, the share of external shocks that account for the variations in equity returns was considerably lower during the COVID-19 pandemic, while the share of domestic shock was higher (Figure 4.10). In contrast, the proportions of global and regional shocks that account for the variations in bond returns were higher in the ongoing COVID-19 pandemic, while the proportion of domestic shocks was smaller (Figure 4.11). These imply the varying sensitivity of financial asset price returns to external shocks across these two different episodes.
While stable at the moment, Asia and the Pacific is not immune to risks of capital flow reversals, and financial vulnerabilities and uncertainties.
Diverging economic growth paths, due to uneven vaccine rollout, as well as differences in policy support and containment measures, could lead some large  Sources: ADB calculations using data from Bloomberg; CEIC Data Company; and methodology by Lee and Park (2011). Similarly, the proportion of global shocks that explain the variation of bond returns increased to 14.7% during the COVID-19 period (April 2020-December 2021) from 10.8% at the onset of COVID-19 period. Similarly, the proportion of regional shocks that explain the variation of bond returns also increased from 7.7% during the COVID-19 onset to 7.9% during the COVID-19 period (Figure 4.11). Across subregions, the increase in the share of global shocks between the COVID-19 onset and pandemic periods was highest for Southeast Asia, while the decrease in the proportion of regional shocks was largest for India. Similar to equity returns, the share of domestic shocks explaining the variation of bond returns dropped from 81.4% in the COVID-19 onset period to around 77.3% during the pandemic period.  Economies with slower inoculation and higher infections of new COVID-19 variants may turn to renewed and/or continued containment measures. This will slow economic recovery momentum by limiting economic reopening, resulting in weaker economic growth. Such weak recovery coupled with higher corporate debt levels may result in debt servicing difficulties. As governments and corporations borrowed to weather the pandemic, economies in the region reported increases in corporate debt ratios, between 2019 and the third quarter of 2021 ( Figure 4.12). For example, the changes in the corporate debt ratios of Hong Kong, China; Japan; and the Republic of Korea were greater than 20% of gross domestic product (GDP); while the increase in government debt ratios for the Philippines and Singapore were above 20% of GDP. Increasing interest rates, coupled with high debt levels, may lead to higher borrowing costs; and will make debt with variable interest rates more costly. Should that happen, debtors may face debt payment difficulties, particularly when growth remains fragile, and could result in higher debt premiums and lower credit ratings.
As economies in the region rely heavily on bank credit for corporate financing, this adds more reason to be concerned with looming rising interest rates (Figure 4.13). Should corporations be unable to make their debt payments on time, banks' debt quality could erode. As it is, some economies in the region have already experienced increased banking sector nonperforming loan (NPL) ratio in 2020 (Figure 4.14).  For instance, Papua New Guinea's NPL ratio increased from 3.8% in 2019 to 5.9% in June 2021; that of the Philippines increased from 2.0% in 2019 to 4.4% in June 2021; and the Kyrgyz Republic's NPL ratio increased from 7.7% in 2019 to 11.8% in October 2021. Uncertain recovery, higher interest rates, and rising NPL ratios may prompt banks to be more cautious in lending, which could undermine prospects of stronger recovery.
The Asia and Pacific region also faced potential cross-border financial risk spillovers emanating from Evergrande's debt woes and its impact on the PRC's property sector and the broader economy in the second half of 2021. The Evergrande group, with debts exceeding $300 billion, is the PRC's most indebted property developer. In Q4 2021, the company delayed its offshore bond payments amounting to $83.5 million, indicating its difficulties in repaying its debt obligations (Wilkins et al. 2021); its share price lost about 90% of its value and its market capitalization 89% of its value from early January 2021 to the end of 2021. Several property developers in the PRC, including China Properties Group, Fantasia Holdings Group, and Sinic Holdings Group also faced debt repayment difficulties in the latter part of 2021 (Nikkei Asia 2021, The Straits Times 2021). Rating agencies, including Moody's and Fitch, downgraded the credit ratings of several PRC property bond issuers (Toh 2021). As Evergrande and other property developers delayed their debt payments, concerns grew over their impact on the PRC's property sector, which accounts for almost 30% of its economy. Share prices of property developers plunged and the PRC's junk bond yields peaked in November 2021 (Figures 4.15 and 4.16).   Although the PRC's property sector debt problems remained contained within the sector as of final quarter of 2021, risks of wider impact on the economy and potential cross-border spillovers persisted (Magnus 2021, Tan 2021.
To mitigate potential risks in the finance sector, economies in the region need to continue to strengthen economic fundamentals. Speeding and scaling up the inoculation drive to contain the spread and emergence of new COVID-19 variants remains paramount in supporting economic recovery and reopening economies, particularly for emerging and developing economies. Sustained and stable growth momentum will help ease risks of rising corporate and government debt levels. Policy support must also be calibrated depending on domestic financial conditions and circumstances and the viability of recipients. Addressing rising NPLs will help improve debt quality and bank balance sheets.

Recent Trends in Asia's Cross-Border Financial Assets and Liabilities
In 2020, Asian investors continued to invest more outside the region than inside. 30 Asia's total cross-border financial asset holdings reached $25.4 trillion in 2020, up significantly from $16.3 trillion at the end of 2016 (Figure 4.17). 31 Most of the region's investment holdings in 2020 were FDI assets ($9.2 trillion), followed by portfolio equity ($6.5 trillion), portfolio debt ($5.5 trillion), and banking sector loan and deposit holdings ($4.2 trillion). Around two-thirds of Asia's asset holdings were placed in non-regional economies, and only one-third in regional economies. This proportion was roughly unchanged between 2016 and 2020, despite Asia's growing share of world output. The values reported for total cross-border assets, liabilities, and net position do not reflect total values in the International Investment Position. This is because reported values include only those with available bilateral breakdown to decompose regional and non-regional holdings and liabilities. Refer to Box 4.1 for discussion on the uses of the International Investment Position data set. Throughout this chapter, cross-border investment holdings include banking sector loan and deposit assets (claims) and liabilities, FDI, portfolio debt, and portfolio equity. . trillion ( . ) Asia s intraregional share: .

Bank:
. trillion ( . ) Asia s intraregional share: . holdings, much of its portfolio assets in 2020 were East Asian assets (68.7%). East Asia and Southeast Asia mostly held East Asian portfolio assets in 2020, at 67.1%, and 24.9%, respectively, suggesting the attractiveness of East Asian portfolio debt and equity assets.
Asia's portfolio debt holdings increased from $5.0 trillion in 2019 to $5.5 trillion in 2020, reporting a 11.1% increase and continuing a rising trend over the past 6 years. 32 Asian investors' portfolio equity holdings increased 19.4% from $5.4 trillion in 2019 to $6.4 trillion in 2020. Asia's crossborder loan and deposit claims, meanwhile, continued to rise in 2020, to $4.2 trillion from $3.9 trillion in 2019. For cross-border banking flows, loan and deposit asset flows decreased from $79.9 billion in 2019 to $51.4 billion in 2020, much of the decrease was attributable to the decline in banking flows to the rest of the world (Figure 4.18a).
The region's total external financial liabilities also inched higher, to $25.4 trillion in 2020, up from $18.0 trillion in 2016 (Figure 4.19). Much of the region's liabilities were FDI ($10.1 trillion), followed by portfolio equity ($6.8 trillion), banking sector loan and deposit liabilities ($4.9 trillion), then portfolio debt ($3.7 trillion). As in previous years, around two-thirds of the region's external investment liabilities were held by non-regional economies, and one-third by regional economies. Inward debt portfolio investment increased 16.4% to $3.7 trillion in 2020 from $3.2 trillion in 2019, while the value of inward equity portfolio investment rose 21.1% to $6.8 trillion in 2020 from $5.6 trillion in 2019. The intraregional share of inward portfolio debt edged down from 28.6% in 2019 to 28.5% in 2020; the intraregional share of inward portfolio equity increased from 19.1% to 20.3% in the same period. Asia's cross-border loan and deposit liabilities increased in 2020 to $4.9 trillion from $4.6 trillion in 2019.
For cross-border banking flows, Asia's loan and deposit inflows reversed from $28.7 billion in 2019 to -$32.6 billion in 2020 as the region's loan and deposit liabilities with the rest of world registered a large reversal, amounting to $90.1 billion, while liabilities with Asia increased from $23.1 billion in 2019 to $24.9 billion in 2020 (Figure 4.18b). . . -. -.
---Asia and the Pacific US Total EU UK ROW (excluding the EU UK and the US) -. -. . -.
-. Source: ADB calculations using data from the Bank for International Settlements. Locational Banking Statistics. https://www.bis.org/statistics/bankstats.htm (accessed December 2021). 32 The overall increase or decrease in stock portfolio holdings and liabilities is attributed to changes in flows and valuation changes of existing portfolio assets and liabilities.
As the region held more debt assets than debt liabilities, but more equity liabilities than equity assets, it retained its long debt, short equity position as of the end of 2020. The net debt positions in 2016 and 2020 were the same, at $1.2 trillion, while the net equity position improved from -$2.9 trillion in 2016 to -$1.2 trillion in 2020. As of the end of 2020, the largest share of its long debt and short equity positions were with non-regional economies, mirroring the regional breakdown of its international investment assets and liabilities.
The currency compositions of Asia's international investment assets and liabilities indicate the dominance of the US dollar. 33 Almost half of Asia's international asset holdings were denominated in US dollars as of the end of 2020, followed by other currencies, at 17% and then euros, at 11%. In contrast, around 64% of the region's external liabilities were dominated in local currencies, followed by US dollars, at 25% (Figure 4.20). Across types of international investments, equity-type assets, which include FDI and portfolio equity, were mostly denominated in other currencies, as it is assumed that the currency composition of these investments closely tracks geographic positions. Equity-type liabilities were denominated in local currency as FDI and portfolio equity ownerships were mostly denominated in the host economy's currency. The currency compositions of debt-type international investments indicated the dominance of the US dollar. For debt assets, which include portfolio debt, other investments, and official reserves, about 63% were denominated in the US dollar, followed by the euro (13%), and other currencies (9%). In contrast, half of debt liabilities, including portfolio debt and other investments, were denominated in US dollars, followed by local currency (28%), and other currencies (10%) (Figure 4.21). The dollar dominance in Asia's cross-border investment holdings can lead to several risks, including US monetary policy spillovers and their impact on global liquidity and the balance sheet and welfare effects of large exchange rate fluctuations between the US dollar and local currency. . trillion ( . ) Asia s intraregional share: .

Evolving Patterns of Capital Flows in Asia and the Pacific
An earlier monetary policy normalization in large and advanced economies, such as the US, has raised concerns about the impacts on emerging and developing economies. Specifically, higher interest rates in advanced economies are often associated with capital inflow decreases or reversals in emerging and developing economies (Bryne and Fiess 2016; Ghosh et al. 2014;Giordani et al. 2017;Li, de Haan, and Scholtens 2018;and Mercado 2018a). As the divergence of economic recovery becomes more apparent in 2021, assessing the evolution of nonresident capital flows is warranted to better understand the likely impacts of large capital flow reversals, in the light of uncertain pandemic outcomes. This subsection discusses the patterns of nonresident capital flows over the past 2 decades for Asia and the Pacific. It also considers policy tools used to address the adverse effects of volatile capital flows. The focus of analysis in this subsection is on nonresident capital inflows (gross capital inflows), instead of net nonresident capital inflows (net capital inflows) or resident capital outflows. Net capital inflows are usually more stable than nonresident capital inflows, which exhibit volatilities. Moreover, focusing on nonresident capital inflows will identify nonresident capital flow reversals, instead of net capital inflow reversals, which may either be attributed to residents or nonresidents. Nonresident capital flows include direct investment abroad, portfolio equity, portfolio debt, and other investments as defined by Balance of Payments Manual 6 (BPM6). Resident capital flows include foreign direct investment, portfolio equity, portfolio debt, other investment, and reserve assets as defined by BPM6. $0.6 trillion in 2011-2020. On the one hand, the doubling of gross flows to Asia reflects its attractiveness as a main destination of foreign investments. Specifically, compared with other emerging and developing economies, Asia and the Pacific received twice the FDIs in 2011-2020. But the doubling of gross capital inflows to the region implies greater potential adverse impact of capital flow reversals.

Nonresident capital inflows in
Nonetheless, as a percentage of GDP, the magnitude of gross capital inflows to the region remained roughly the same, at an average of 5.5% for both periods. In addition, resident capital outflows have grown, from an annual average of $0.8 trillion in 2001-2010 to $1.6 trillion in 2011 to 2020 (Figure 4.23). Hence, net resident capital outflows have mostly been positive in the last 2 decades, indicating that Asia and the Pacific had been a net capital exporter.
cross-border financial investments (Avdjiev, Chui, and Shin 2014). Nonfinancial multinational enterprises can provide within-company credit to their parent company or subsidiaries located elsewhere. This transaction appears as "FDI debt" in the balance of payments statistics. In Asia and the Pacific, FDI debt has more than doubled from an average annual value of $39 billion in 2001-2010 to $82 billion in 2011-2020. Nonfinancial multinational enterprises also provide trade credits and loans to other companies, and can make cross-border bank deposits. These partly explain the rise in cross-border currency and deposits and loans, which have increased from an average annual value of $84 billion and $60 billion in 2001-2010 to $190 billion and $170 billion in 2011-2020, respectively. These transactions may understate the true cross-border exposures of nonfinancial multinational enterprises that have borrowed abroad through their affiliates.
Second, in 2011-2020 nonresident portfolio debt inflows rose, coinciding with the rise in debt issuance in Asia and the Pacific. In the same period, total bond issuance in the region increased from $2.3 trillion to $7.2 trillion, representing a compounded annual growth rate of 13.2%. 35   their cross-border financial transactions. Such patterns may give rise to financial stability concerns if these flows lead to more financial operations rather than real economic activities (Avdjiev, Chui, and Shin 2014). The banking sector was the second-largest recipient of nonresident capital flows. But for some years, including 2020, the government sector received more nonresident capital flows than the banking sector, suggesting the rising importance of the public sector as borrower, as also noted by the Committee on Global Financial System report (   Fourth, nonresident capital inflows to Asia and Pacific economies mostly went to nonfinancial corporates (Figure 4.25). This pattern is unsurprising given that the region attracts a large share of global FDI and that nonfinancial multinational enterprises are increasing  -2020(CGFS 2021. The relative volatilities of gross capital inflows across types of investments remained consistent over the last 2 decades (Figure 4.26a). Other investment inflows, which include loans, currency and deposits, and trade credits, remained the most volatile capital flows, followed by portfolio flows and then FDI flows. The same patterns were previously noted by Mercado and Park (2011) for developing Asia economies from 1980 to 2009. Across the Asia and Pacific subregions, the PRC had the most volatile nonresident capital inflows in 2011-2020, followed by ASEAN4 and NIEs ( Figure  4.26b). Among sectors, public sector inflows (central bank and general government) as well as other financial corporate inflows were more volatile in both periods, compared with banks and nonfinancial corporates (Figure 4.26c).  Lepers and Mercado (2021); and national sources.

The changing patterns of foreign capital inflows into Asia and the Pacific reflect the varying relevance of global and domestic factors.
More recent studies show that global and domestic economic growth, investor risk appetite, domestic macroeconomic risks, trade and financial openness, quality of governance, and domestic financial depth are the relevant drivers of foreign capital inflows (Ahmed and Zlate 2014;Byrne and Fiess 2016;Fratzscher 2012;Ghosh et al. 2014;Giordani et al. 2017;Li, de Haan, and Scholtens 2018;Mercado 2018a;and Mercado and Park 2011). Other drivers have also been identified. CGFS (2021) highlighted the significance of the institutional infrastructure of the global financial system through which capital flows ultimately move, known as "pipes" as an important determinant of the magnitude of capital inflows; while Mercado (2018b and2020) found that gravity factors such as bilateral trade and distance drive bilateral capital flows. But the significance of these drivers change over time. For example, CGFS (2021) stressed that the changes in capital flow pipes have become the most important driver of capital flow patterns in the post-global financial crisis period.
Focusing on a sample of Asia and Pacific economies between 2001-2010 and 2011-2020, the conditional correlations between various types of gross capital inflows and global and domestic factors show that the significant negative correlation between portfolio equity flows and VIX have declined between the two periods ( Figure 4.27), while the positive conditional correlation between portfolio debt inflows and domestic GDP growth has increased and became significant in 2011-2020, compared with the previous period. This implies that foreign investors have become responsive to domestic economic growth in deciding whether to hold Asia and Pacific portfolio debt. These findings remain the same when additional domestic covariates are considered. In addition, the positive and significant correlation between domestic capital account openness and domestic financial depth with FDI inflows have increased in the second period; while the positive and significant correlation between domestic governance quality and other investment inflows likewise increased in 2011-2020, compared with 2001-2010. Again, these results suggest that foreign investors have become more responsive to domestic factors in Asia and the Pacific in deciding whether to invest in the region. Asia and Pacific economies experienced marked periods of large nonresident capital inflows and outflows over the last 2 decades. episodes of large nonresident capital inflows or "surges" and outflows (reversals) or "stops" are caused by various domestic and global factors, such as investor risk appetite, contagion effects, among others (Caballero 2016;Calderon and Kubota 2013;Calvo 1998;Calvo, Leiderman, andReinhart 1993 and1996;Calvo, Izquierdo, and Mejia 2008;Cavallo and Frankel 2008;Forbes andWarnock 2012a and2012b;Ghosh et al. 2014;Levchenko and Mauro 2007;Magud, Reinhart, and Vesperoni 2014;Mercado 2018a andMilesi-Ferretti and Tille 2011;Reinhart and Reinhart 2009;and Rothenberg and Warnock 2011).
Consequently, identifying episodes of nonresident capital flow surges and stops is important in undertaking macrofinancial surveillance. Knowing "how large" nonresident capital inflows and outflows should be needs consideration before assessing what policy tools or combination thereof would be best in managing capital flow surges and stops. Annex 4b discusses commonly used methods in identifying capital flow stops and surges.
Applying the capital flow surge and stop definition of Forbes andWarnock (2012a and2021) to selected Asia and Pacific economies from 2000 to 2020 reveals two noteworthy observations (Figure 4.28). First, surges and stops are rare occurrences. On average, around 10% of the Asia and Pacific sample experience extreme episodes per quarter. Second, stops and surges may occur in ripples or waves. More than a third of the sample experienced surges in 2007 and stops in 2008 and 2009, whereas occurrences of these two extreme episodes were significantly less for other periods.
Periods of large nonresident capital inflows and outflows tend to coincide with improving or deteriorating domestic macroeconomic and financial indicators, suggesting their policy relevance. Figures 4.29a to 4.29f trace the patterns of several macrofinancial indicators before, during, and after years of large nonresident capital inflows (surges) and outflows or reversals (stops) for selected Asia and Pacific economies from 2000 to 2020. Using annual capital flows data sourced from the Balance of Payments data set of the International Monetary Fund (IMF) and national sources, large nonresident capital inflows or surges are defined as the largest positive nonresident capital inflow reported by each economy in the sample from 2000 to 2020. In contrast, large nonresident capital outflows or stops or reversals are distinguished as the largest negative nonresident capital flows reported by each economy in the sample from 2000 to 2020.  The years with the largest nonresident inflows and outflows are noted as time (t). Then, the median values of macrofinancial variables are taken, including GDP growth, current account balance, equity price, among others, across the sample of Asia and Pacific economies at time t as well as those 3 years before and 3 years after the identified large episode of nonresident capital flows at time t, following Reinhart and Reinhart (2009).
Several key observations are noted. First, GDP growth declines during and after large nonresident capital outflows (stops or reversals), before recovering 2 years following the stop episode (red line in Figure 4.29a). Output growth is often weaker during foreign capital flow reversals as they are often associated with economic slowdowns or output drops.
GDP growth also appears weaker following episodes of large nonresident capital inflows or surges (blue line in Figure  4.29a). Second, the current account balance of selected Asia and Pacific economies tends to deteriorate before and during large nonresident capital inflows (blue line in Figure  4.29b). In contrast, the current account balance improves when large nonresident capital outflows (stops) occur, due to weaker domestic demand (red line in Figure 4.29b). Third, the Asia and Pacific region accumulates official reserves during surges and decumulates reserves during stops ( Figure  4.29c). Fourth, real exchange rate usually depreciates during and 1 year after large nonresident capital flow reversals (Figure 4.28d). Fifth, fiscal balance worsens during surges but slightly improves during stops (Figure 4.29e). Last, equity prices usually rise before and during surges but decrease after. They decrease during large nonresident capital flow reversals and remains depressed 1 year after (Figure 4.29f). 37 The Asia and Pacific economies used various policy measures to address the adverse impacts of large and volatile capital flows.
Although capital inflows provide substantial direct and indirect benefits to emerging and developing economies, they also carry risks and pose a challenge to policy makers in the region. Specifically, the changing nature and varying significance of domestic and global factors require deeper understanding of the dynamics and evolution of nonresident capital flows. Moreover, large capital inflows and large capital flow reversals are often associated with either improving or deteriorating macroeconomic and financial conditions, thereby warranting appropriate policy responses, as shown in Figure 4.29.
In Among capital flow impacts, policy makers were mostly concerned with their impact on exchange rate followed by financial stability. Among policy measures, most used greater exchange rate flexibility, while others also used foreign exchange intervention and macroprudential measures. Empirical evidence on the effectiveness of these policy measures in addressing capital flow volatilities, surges, and stops has shown their usefulness under specific conditions (Eller et al. 2021;Lepers and Mehigan 2019;Frost, Ito, and Stralen 2020;Lepers and Mercado 2021;and Carvalho, Lepers, and Mercado 2021). 37 Most of these patterns hold if the identified episodes of large nonresident inflows and outflows are restricted from 2003 to 2017 to completely capture patterns 3 years before and after the identified episode at time t.
Trend during largest positive nonresident capital inflows Trend during largest negative nonresident capital flows (reversals) Given the volatile nature of capital flows and associated risks, several considerations are warranted.
First, the pattern and composition of capital flows need to be carefully monitored, as the US is edging toward policy normalization, while emerging and developing economies are still addressing the ongoing COVID-19 pandemic. In particular, understanding cross-border financial flows along sectoral lines is needed as investment flows of nonfinancial corporations have become more complex, as emphasized in this chapter; and other financial corporates or nonbank financial institutions are now the main source of capital flows from advanced economies . In addition, assessing the importance of domestic and global drivers, and more recently, "pipes" is required as changes in these factors will eventually determine the patterns and compositions of capital flows.
Second, large nonresident capital inflows and outflows could lead to deteriorating macroeconomic and financial conditions, and hence, can amplify risks and vulnerabilities. Moreover, earlier studies note that capital flow episodes transition from one to another, such as "surges" that are followed by "stops" (Efremidze et al. 2017;Mercado 2018a andand Sula 2010). Consequently, identifying these episodes is vital in deciding whether and when to use policy measures to help address these episodes of volatile capital flows (The SEACEN Centre 2019).
Third, the use of policy tools should be aligned with domestic situations and conditions. Yet, policy frameworks are a useful guide in deciding the appropriateness of policy tools. 38 Fourth, as the patterns, compositions, and drivers of capital flows constantly evolve, the sharing of information and experiences among regional economies is helpful, specifically in identifying emerging trends as well as in the appropriateness and effectiveness of policy measures. In this regard, regional cooperation can offer a venue for sharing information and experiences in managing capital flows. The IMF published its Institutional View on capital flows in 2012 and, subsequently, the Integrated Policy Framework in 2020 as guide on the appropriate use of various policy measures in addressing capital flow surges and sudden stops (IMF 2012). The Committee on Global Financial System in its 2021 report concluded that there is no "one size fits all" on how these policy measures are best combined, as it will depend on economy conditions and contexts (CGFS 2021). Figures 4.17 and 4.19 show the region's holdings of international investment assets and liabilities. The stylized facts drawn from these figures are based on underlying bilateral holdings data, where regional values are derived. Although the figures are informative and useful in understanding the proportion of external investment assets and liabilities held by regional versus non-regional economies, they do not provide the complete information as to the region's total external assets and liabilities, as bilateral source data are limited. To understand the region's external investment position, the International Investment Position is a useful statistic in tracking external adjustments and holdings. The compilation of the International Investment Position has improved over the last 2 decades, allowing policy makers more information on external debt assets and liabilities, as well as external equity investments. But information prior to 2000 is limited. Hence, longterm view of external adjustments and net international investment positions are constrained. This data gap has been addressed by Lane andMilesi-Ferretti (2007 and in their External Wealth of Nations Database.

box 4�1: International Investment position
The International Investment Position data are useful in understanding global imbalance and external adjustments (Lane andMilesi-Ferretti 2012 and; as well as tracking de facto financial integration measure (Park 2013  instance, Park (2013) noted that emerging Asia showed a steady uptrend in de facto financial integration from 1970 to 2010 despite the declines in de jure financial integration measure using the Chinn-Ito database (Chinn and Ito 2008), highlighting the substantial divergence between de facto and de jure measures of financial openness and integration. These studies show the importance and usefulness of International Investment Position data in understanding external positions and financial integration trends.
In addition, the statistic is valuable in understanding the improvement or deterioration of the net foreign asset position at the outset of the COVID-19 pandemic in 2020.
Comparing the change in net foreign asset position between 2019 and 2020 for selected Asia and Pacific economies, the net foreign asset position of several economies including Armenia, Australia, Fiji, Georgia, Kazakhstan, Mongolia, and New Zealand further declined in 2020 as these economies have negative net foreign asset position in 2019, as shown in the figure. In contrast, the net foreign asset position of Bhutan, India, Indonesia, Malaysia, the Philippines, and Solomon Islands improved in 2020 despite these economies also having a negative net foreign asset position in 2019. In fact, for the latter group of economies, the improvement in net foreign asset position in 2020 coincided with the improvement of the current account balance.   Note: ASEAN4 includes Indonesia, Malaysia, the Philippines, and Thailand. Other emerging and developing Asia includes Armenia, Georgia, Kazakhstan, Mongolia, and Pakistan.

Annex 4a: Sectoral and Subregional Decomposition of Capital Flows
Source: ADB calculations using data from Lepers and Mercado (2021).

Annex 4b: Identifying Capital Flow Surges and Stops
Various methods are used in the literature to identify capital inflow surges and stops. For surges, they are usually defined to imply more than the usual increase in capital inflows. However, there are various approaches in measuring "more than usual." For instance, more than usual could refer to one or two standard deviations from historic mean, filtered trend, or relative size of capital inflows. For example, Forbes and Warnock (2012a and 2021) and Mercado (2018a and 2019) used two standard deviations from historic mean. In addition, surges can also be identified based on some threshold percentile. For instance, Reinhart and Reinhart (2009) used the top 20th percentile as threshold, while Ghosh et al. (2014) used the top 30th percentile as threshold.
For "stops, " Calvo, Izquierdo, and Mejia (2008) defined "sudden stops" as a sharp fall in net capital inflows. A "sharp fall" pertains to a one standard deviation drop of the year-on-year change of the 12-month moving sum of net capital inflows relative to its historic mean, provided it drops two standard deviations within the episode. In contrast, Forbes and Warnock (2012a and 2021) defined "stops" as a sharp decline in nonresident capital flows, instead of net capital inflows as used by Calvo, Izquierdo, and Mejia (2008). A sharp decline pertains to a one standard deviation drop of the year-on-year change of the 12-month moving sum of gross capital inflows relative to its 5-year rolling historic mean, provided it drops two standard deviations at some point within the episode.
In this chapter, surges and stops are derived using quarterly nonresident capital inflows sourced from the International Monetary Fund's Balance of Payments and International Investment Position Statistics. To state, "surge" is defined as an episode where nonresident capital inflows increase more than one standard deviation above its historic mean provided that (i) the entire episode lasts more than one-quarter; (ii) there are at least 5 years of data to calculate the historic mean; and (iii) it reaches at least two standard deviations above at some point within that episode. Specifically, we let C t be the four-quarter moving sum of gross capital inflows (GINFLOW) and derive annual year-on-year changes in C t : Rolling average and standard deviations of ∆C t are computed over the last 20 quarters. A "surge" episode is defined to start at the first month t when ∆C t increases more than one standard deviation above the rolling mean. But in order for an entire episode to qualify as "surge" there must be at least one quarter t when ∆C t increases at least two standard deviations above its mean. A "stop" episode is defined using the same approach but pertains to the opposite direction, i.e., a large decrease in nonresident capital flows. "Normal" episodes are defined as the absence of either surges or stops for a given quarter. Annex Figure 4b.1 provides an illustrative example in defining surges and stops using quarterly data for the Philippines. The figure shows that the Philippines had a surge and then stop episode before and during the global financial crisis of 2008, and a stop episode in the first quarter of 2020 at the start of the COVID-19 pandemic.