Capital Flows To Emerging Markets APRIL 7, 2016

CAN THE EM REBOUND LAST? 













Since our January report, emerging markets have staged a striking turnaround as asset prices have bounced back and inflows revived. This recovery after a miserable half-year largely reflects a pickup of global risk appetite as concerns about downside risks to the global economy have receded, with added fuel from attractive EM valuations and underweight positions. Stabilization of expectations about the RMB together with a clear dovish bias among G3 central banks and a soft USD have helped to reassure investors.

This report is the result of the collective efforts of the IIF Economics and Capital Markets staff. All contributors are listed on the back cover. Questions and data requests can be submitted to [email protected].

Although tail risks are reduced, we do not see sustained improvement in the poor fundamentals in emerging market economies. Some countries are making headway in strengthening underlying policies but many are still struggling to adapt to a more challenging global environment.

Table of Contents

We continue to project substantial net outflows from EMs of about $500 billion in 2016, down from around $750 billion in 2015. The improvement is driven by a gradual recovery in non-resident capital inflows, which are projected to climb from $240 billion in 2015 to $560 billion this year (Chart 1), but heavy resident capital outflows will still keep overall flows negative.

Capital Flows Outlook: Climbing Back From Record Lows…...…………. 8

Net outflows will continue to be dominated by China. We project outflows to slow from $675 billion last year to $530 billion this year. This should allow reserve losses to diminish, thus helping to contain fears of a disorderly RMB depreciation. As discussed in our deep dive into China’s capital flows, we expect a modest recovery in foreign inflows to China but continued substantial resident outflows.

Another Roller Coaster Ride for EMs...……………. 2 Are Emerging Markets Still Cheap?.....………………... 5

Box: Banks Under Pressure: Implications for Emerging Markets……… 12 The Great Unwinding: Capital Outflows From China………………………. 13 Tables and Annex……….. 25 Contributors………………. 30

Key risks to a somewhat more benign outlook include renewed concerns about China’s economic trajectory and a more challenging global financial market environment as the Fed moves to raise policy rates further over the summer.

Chart 1 Capital Flows to Emerging Markets $ billion 1,500 EM ex. China Resident Outflows* China Resident Outflows* EM ex. China Non-Resident Inflows 1,000 China Non-Resident Inflows

IIF Forecast

Join the IIF for the launch of our new Global Seminar: Understanding Capital Flows to Emerging Markets

500

New York—May 12-13

0

-500 Total EM Net Capital Flows (Financial Account Balance) plus Errors & Omissions

-1,000 -1,500

Gain more insights

1995

2000

2005

*Resident Outflows exclude reserves, include errors & omissions. Source: IIF.

iif.com © Copyright 2016. The Institute of International Finance, Inc. All rights reserved.

2010

2015

Participants will explore key EM capital flows issues, focusing on the conceptual frameworks and practical tools needed for economic analysis. Class size will be kept small and the content will be tailored to fit participants’ interests.

CAPITAL FLOWS TO EMERGING MARKETS | APRIL 7, 2016

page 2

ANOTHER ROLLER COASTER RIDE FOR EMERGING MARKETS Since our January Capital Flows Report, emerging market asset prices and exchange rates have staged a striking turnaround, while portfolio inflows have resumed after an unprecedented seven months of outflows. Emerging market stocks are up some 20% from the troughs reached in late January, after having declined by over 25% since August 2015, while bond prices have risen sharply, especially for local currency bonds (Chart 2). Countries with the biggest gains in the recent rebound have in many cases been those that had larger losses in the downswing (Chart 3). It is worth noting that the scale of the swings has corresponded quite closely to the risk betas that we had calculated last year (see Which EMs Would Suffer from a Risk Shock?, April 2015), with the more risk sensitive countries showing the largest price swings (Chart 4). Correlations across EM asset prices have generally been quite high, both in the downturn and the recent upswing (Chart 5). These observations all suggest that this latest roller coaster ride for EM assets was mainly driven by external factors like shifts in global risk sentiment rather than EM country-specific developments. Chart 2 Change in Asset Prices August 3, 2015 - January 21, 2016 January 21, 2016 - March 31, 2016 Oil (Brent) MSCI EM Equity EM Sov. Bonds (LCY) Gold S&P 500 EM FX EM HY Corp. Bonds EMBI Total Return EM Corp. Bonds Commodities ex. Energy Nikkei 225 Euro Stoxx50 DXY

Chart 4 FX: Since Aug. 2015: Peak-to-Trough vs. Risk Betas percent, peak-to-trough -25

-20

-15

-10

COP BRL

ZAR MXN

0 -0.2

PLN IDR

ZAR

CLP MXN BRL RUB

Source: Bloomberg, IIF.

INR

CNY TRY

20 MYR

IDR

PLN

CLP

EM* CZK

0

-5

25

RUB

-50 -40 -30 -20 -10 0 10 20 30 40 50 percent change

Source: Bloomberg, IIF.

-30

Chart 3 FX: Since Aug. 2015: Peak-to-Trough vs. Return From Trough percent change, return from trough

-0.4

-30

-25

-20

15

KRW TRY

HUF

10 THB

INR PHP CNY

5

0 -15 -10 -5 0 percent change, peak-to-trough

Source: Bloomberg, IIF. LC vs. USD; *EM = JP Morgan EM FX Index.

Chart 6 Total Non-Resident Portfolio Inflows to Emerging Markets $ billion 50 Debt Equity 40 2010-14 Avg.: $22 bn

30 20

-0.6

10

-0.8

0

-1

-10

-1.2 -1.4 2012-2015 Risk Beta

iif.com © Copyright 2016. The Institute of International Finance, Inc. All rights reserved.

-20 -30 Mar 14

Total Flows Jul 14

Nov 14 Mar 15

Jul 15

Source: IIF Monthly Portfolio FlowsTracker Estimates.

Nov 15 Mar 16

CAPITAL FLOWS TO EMERGING MARKETS | APRIL 7, 2016

page 3

Chart 5 Correlation Across EM Asset Prices correlation coeficient, dashed lines show period averages, shaded areas represent nonJuly15-Jan.16 resident portfolio outflow episodes Taper Tantrum Downturn 0.9 US Rating Dec. 2014 downgrade/EU Sell-off Debt Crisis 0.8

0.7 0.6 0.5 0.4 Jan 10

Oct 10

Jul 11

Apr 12

Jan 13

Oct 13

Jul 14

Apr 15

Jan 16

Source: Bloomberg, IIF; This indicator is the simple average of asset-class correlations for stocks, bonds (both local and hard currency) and FX correlations. Within each asset class, we compute the 12-week rolling correlation in returns between the EM aggregate index and the country specific indices. We then take the average of correlations across all countries to compute the asset-class correlations.

Turning to the flow side, the March rebound was broad based, again pointing to the role of external “push” factors. Our latest tracking estimates indicate that EMs saw non-resident portfolio inflows of $37 billion in March, after $70 billion of outflows from July 2015 to January 2016 (Chart 6), with flows turning positive for nearly all EM countries for which we have data (Chart 7). Both EM equity and debt flows participated in the rebound. The main driver of the turnaround in appetite for EM assets appears to have been reassurance that downside risks facing the global economy and financial markets were less threatening than previously thought, reinforced by attractive EM valuations and underweight positions that encouraged bargain hunting by global investors (see page 5). The global downside risks that loomed large in January included the possibility of a disorderly devaluation of the RMB, global fallout from the slump in energy prices, and the increasing risk that the advanced economies

Chart 7 Monthly Portfolio Flows Diffusion Index diffusion index; above/below 0 = net share of EMs with net positive/negative non-resident portfolio inflows 100

March 2016

75

Chart 8 China: Net Capital Flows and Reserves Operations $ billion, an estimate of capital flows and reserve operations 100 Reserve Operations 50

50 0

25 0

-50

-25 -50 -75

GFC

Taper Tantrum U.S. Rating Aug. 2015 Downgrade/EU Debt Crisis

-100 2008 2009 2010 2011 2012 2013 2014 2015 2016 Source: IIF. Partial Data Through March 2016.

iif.com © Copyright 2016. The Institute of International Finance, Inc. All rights reserved.

Net Capital Flows

-100 -150 2010

Source: IIF.

2011

2012

2013

2014

2015

2016

CAPITAL FLOWS TO EMERGING MARKETS | APRIL 7, 2016

page 4

could tip into recession. These downside risks were clearly interrelated, and a key factor linking all three related to worries about economic prospects for China. However, in recent months, Chinese policymakers have managed to reduce concerns about an abrupt RMB depreciation through a combination of stepped-up communications emphasizing the role of the currency basket rather than the link to the dollar and added stimulus to the economy. A soft dollar also helped to stabilize the RMB’s path. After sizeable net capital outflows in December and January, flows pressures subsided somewhat in February and March and reserve declines have moderated (Chart 8). But, as we discuss in our deeper dive into the drivers of China’s capital account on pages 13-24, China is likely to continue to face substantial net capital outflows, and there remain concerns that the RMB may at some point be devalued more meaningfully in the face of persistent declining reserves. The recent rally was also supported by the recovery in oil prices, which regained over 40% between late January and late March, although crude oil has again fallen below $40/barrel following recent statements by Saudi Arabian leaders that they would only freeze production if Iran were to participate in the deal. Nonetheless, the Q1 oil price rally has brightened the outlook for the battered global energy sector and for a number of emerging markets that rely heavily on commodity exports, especially in the GCC and Latin America (Chart 9). We project Brent oil to rise a bit further over the next several months, averaging $40 per barrel for the year (Chart 10). Concerns that the mature economies could be on the verge of a recession have been reduced by improving data and dovish signals from the G3 central banks. In recent weeks, the balance of the data flow in mature economies has been positive, reflected in a pickup in economic surprise indices (Chart 11). This has provided some comfort that the underlying growth momentum in the mature economies remains intact despite weakness in GDP in 2015 Q4 and likely in 2016Q1. At the same time, the European Central Bank and the Bank of Japan have continued to ease policies while the Fed has scaled back its communications of future rate increases, with the median FOMC forecast being lowered from four to two rate hikes in 2016. As a result, market expectations for policy rates have been cut back across the board (Chart 12). Chart 10 Oil Prices and Forecast (Brent) $ per barrel 120

Chart 9 Emerging Market Currencies index, July 1, 2014=100; LC vis a vis $ 102 98

100

94 90

IIF Forecast (2016 avg: $40) (2017 avg: $45)

80

Oil Importer Aggregate

86 82

78

60 40

74

Oil Exporter Aggregate*

70

66 Jul 14

2016-2017 Path Implied by Market

Nov 14

Mar 15

Jul 15

Old Forecast (2016 avg: $35) (2017 avg: $41)

20

Nov 15

Mar 16

Source: Bloomberg, IIF. *Does not include pegged currencies.

iif.com © Copyright 2016. The Institute of International Finance, Inc. All rights reserved.

0 2014

2015

Source: Bloomberg, IIF.

2016

2017

CAPITAL FLOWS TO EMERGING MARKETS | APRIL 7, 2016

Chart 11 G3 Economic Surprise Indices index 120 100

Chart 12 G3 Policy Rate Expectations percent 1.8

Japan

80

Fed - Jan. 1, 2016

1.4

60

1.0

40

Fed - April 4, 2016

0.6

20

0

0.2

-20 U.S.

-40 -60 -80 Jan 15

page 5

Euro Area May 15

Sep 15

Jan 16

Source: Citi, Bloomberg, IIF.

-0.2 -0.6 16Q1

ECB - Jan. 1, 2016 BOJ - Jan. 1, 2016 BOJ - April 4, 2016

ECB - April 4, 2016 16Q3

Source: Bloomberg, IIF.

17Q1

17Q3

Economic news coming out on emerging markets has remained mixed in recent months. On the bright side the data flow has shown some signs of stabilization (see our EM Growth Tracker) and there have been positive policy developments in China and a number of other EM countries as well, including a reassuring budget in India and successful shift to flexible currency management by Argentina. On the other hand, little has been done to address deep-seated structural constraints across emerging markets that have held back growth performance in recent years, and serious questions remain about policy prospects in a number of vulnerable countries including Brazil, Turkey, South Africa and Venezuela.

ARE EMERGING MARKETS STILL CHEAP? In the context of a global revival of risk appetite since January, the EM rebound has been fueled by attractive valuations and underweight investor positions. As discussed in our January 2016 Capital Flows Report, valuations in emerging markets—across equities, bonds and currencies—had sunk to very low levels by mid-January 2016. Investor sentiment towards emerging markets, as reflected in the January BAML Fund Manager Survey, registered “record bearishness.” Moreover, mutual fund and ETF investor positioning had become increasingly underweight in both EM bonds and EM equities, as highlighted in our report Portfolio Allocation Trends, March 2016. These extremes in positioning and sentiment have been wound back somewhat over the past two months. For example, the net percentage of fund managers reporting overweight positions in EM equities rose from -35 in January to -10 in March. Given the rapid run-up in EM stock prices and the sharp narrowing in EM bond spreads since mid-January, how should valuations and positioning be assessed now? Looking first at fund allocations, investor positioning in both EM bonds and EM equities has shifted back towards a more neutral stance, but portfolio weights are still well below a three-year moving average, particularly for bonds (Charts 13 and 14). On this measure, there is arguably considerable iif.com © Copyright 2016. The Institute of International Finance, Inc. All rights reserved.

18Q1

18Q3

CAPITAL FLOWS TO EMERGING MARKETS | APRIL 7, 2016

page 6

Chart 13 Emerging Markets: Equity Fund Portfolio Weights % of global equity fund allocations, red=UW, green=OW 22

Chart 14 Emerging Markets: Bond Fund Portfolio Weights % of global bond fund allocations, red=UW, green=OW

20

12.5

18

11.5

16

10.5

14 12

Confidence band

Confidence band

9.5

Portfolio bond weight

Portfolio equity weight

8.5

3-year average

3-year average

10 2012

13.5

2013

2014

2015

2016

Source: EPFR, IIF; *Confidence band is centred at the 3-year rolling average of the weights. The width of the band is set at 3.0 times the stdev of weekly weights over the preceding 3-years.

7.5 2012

2013

2014

2015

2016

Source: EPFR, IIF; *Confidence band is centred at the 3-year rolling average of the weights. The width of the band is set at 3.0 times the stdev of weekly weights over the preceding 3-years.

scope for further correction in positioning, which could continue to support EM bond and equity flows in the near term. Turning to valuation measures, notwithstanding the recent repricing, EM sovereign bonds—both local and hard currency—still offer very attractive yields relative to mature market benchmarks. The global hard-currency sovereign spread (EMBIG), near 430bp, stands about 100bp over its 20052015 average, with a number of countries including Brazil, South Africa and Mexico offering a yield pickup of 200-300bp (Chart 15). Similarly, EM local currency bond spreads are about 115bp over 2005-2015 averages; for some countries, including Brazil, South Africa, and Russia, wider-thanaverage spreads likely reflect the rise in inflationary pressures in recent months (Chart 16). With futures markets pricing U.S. 10yr bond yields under 2% well into 2018, and Euro Area 10yr yields below 2% indefinitely, EM sovereign yields at these levels are likely to continue to attract inflows during periods of relative market calm, particularly for countries that are making credible progress towards addressing policy vulnerabilities.

Chart 15 Emerging Market Bond Spreads (EMBIG) bps 1000

900

2005-15 Average

800

Apr 2016

700

Jan 2016

600

Chart 16 Emerging Markets: Local Currency Bond Spreads percent, GBI yield to maturity less 10yr U.S. Treasury yield

18

2005-15 Average

16

Apr 2016

14

Jan 2016

12 8

300

6

200

4

100

2

0

0

UK NG EG AR EM BR ZA MX ID CO TR RU CL HU MY RO CN IN PH

400

Source: J.P. Morgan, Bloomberg, IIF.

iif.com © Copyright 2016. The Institute of International Finance, Inc. All rights reserved.

BR NG TR ZA RU ID CO IN PE EM MX PH CL MY CN RO PL HU TH KO CZ

10

500

Source: J.P. Morgan, Bloomberg, IIF.

CAPITAL FLOWS TO EMERGING MARKETS | APRIL 7, 2016

Chart 17 Corporate Bond Spread Differential: EM vs MM basis points EM vs MM High Yield

175 150

125 100

EM Cheaper

page 7

Chart 18 EM High-Yield Corp. Bonds: Credit Rating Downgrades number; includes credit watch; 2016 annualized 600 500

EM vs MM Investment Grade

400

75

300

50

S. Africa

Hong Kong

Mexico

India

China

Russia

Brazil

200

25 0

100

EM More Expensive

-25 -50 2012

Indonesia

2013

2014

2015

2016

Source: J.P. Morgan, Merrill Lynch, Bloomberg, IIF.

0

2010

2011

2012

2013

2014

Source: S&P, Bloomberg, IIF.

The picture is less clear for EM corporate bonds. Relative to its 2005-2015 average, the hard-currency EM corporate bond (CEMBI) spread is still about 100bp higher, even after narrowing quite a bit in recent months. But spreads relative to the high yield sector are much less attractive, as EM corporates are no longer paying a premium over sub-investment grade mature market corporates (Chart 17). Considering the rapidly growing indebtedness of the EM non-financial corporate sector, now over 100% of GDP vs. 87% in mature markets (see our Emerging Markets Debt Monitor, March 2016) and the sharp increase in ratings downgrades for this sector (Chart 18), the hunt for yield has left EM high-yield corporate bond spreads at very low levels. Similar concerns apply to the equity sector. Certainly on a cyclically-adjusted basis (CAPE), most EM equities look very undervalued (Chart 19), particularly in comparison to mature markets. For example, the CAPE ratio for the U.S. is now over 26, vs. a mean of about 16.5 in EMs. However, forward price-earnings ratios have risen very quickly for many EM equity markets over the past few months, due to a combination of rising equity prices and a steady stream of downgrades to EM corporate earnings forecasts, which have come on top of falling corporate profits since mid-2014. This has left forward P/E ratios in many cases well above their average levels over the past decade (Chart 20). And while equity market returns adjusted for volatility were very strong in Q1 2016 (Sharpe ratio over 1.0), their track record over the past 15 years has been poor (Sharpe ratio of less than 0.5, vs. about 0.8 for U.S. equities). Generally speaking, given the poor fundamentals prevalent in EMs, the expansion of EM multiples in the past two months is unlikely to be sustainable. Finally, looking at EM currencies, real effective exchange rates still tend to be modestly-tosignificantly undervalued relative to trend across EMs even after the rise of almost 9% in EM currencies from Q1 lows. However, there are considerable cross-country variations (Chart 21). In China and Saudi Arabia, currency values are 10-20% higher than their 2005-2015 averages due to the USD linkage. By contrast, a swathe of large EM commodity exporting countries with more flexible exchange rate management now have currency values substantially lower than average, including Mexico, Brazil, South Africa, Russia and Argentina. Factoring in still-high EM short rate differentials with mature markets and the decline in EM FX volatility in recent months, the carry-tovolatility ratio is now quite attractive for a number of EM currencies, particularly in Asia (Chart 22). iif.com © Copyright 2016. The Institute of International Finance, Inc. All rights reserved.

2014

2016A

CAPITAL FLOWS TO EMERGING MARKETS | APRIL 7, 2016

Chart 19 Emerging Market Equities: Cyclically-Adjusted P/E Ratio ratio 40

2005-2015 Average*

35

Mar 2016

30

page 8

Chart 20 Forward Price-to-Earnings Ratios ratio, current price to 12m fwd earnings 18

16 14

25

12

20

10

15

8

10

*Mature markets indicated by light blue. Source: Bloomberg, IIF.

RU

TR

CH

PL

BR

EM

JP

CL

EUR

ID

ZA

MY

IN

US

PH

PH JP US MM ID IN ZA DE MY FR MX GB KR HU CN TH TR EM AR BR RU

4

MX

6

5 0

2006-Present* Jan 2016 Apr 2016

*Mature markets indicated by light blue. Source: Bloomberg, IIF.

CAPITAL FLOWS OUTLOOK: CLIMBING BACK FROM RECORD LOWS Looking forward, we continue to project substantial net outflows from EMs in 2016, albeit at a more moderate pace than last year (see Chart 1). Despite reduced tail risks, we do not see the improvements in overall fundamentals in emerging market economies that would be needed to underpin a more robust revival of EM flows. Some countries are making headway to strengthen underlying policies but many are still struggling to adapt to headwinds from a more challenging global environment, including low commodity prices and a secular shift away from manufacturing to services. Internal constraints include excessive indebtedness, declining productivity growth, aging populations, and populist pressures on policies. Moreover, as we argued above, while the current revival in appetite for EM assets could continue to carry valuations higher and raise portfolio allocations, valuations are now less compelling after the strong Q1 rally and underweight positions less extreme.

Chart 21 EM Currencies: Real Effective Exchange Rates index 2005-Present 130 April 2016 120 Jan. 20, 2016 110 100

Chart 22 Carry-to-Volatility ratio of 3-month rate diff. vs over implied FX volatility 1.0 2010-2015 0.8

2016Q1

0.6

90 0.4

80 70

0.2

60 50

Source: J.P. Morgan, Bloomberg, IIF.

iif.com © Copyright 2016. The Institute of International Finance, Inc. All rights reserved.

Source: Bloomberg, IIF. PLN, CZK, RON and RUB computed vs. EUR and others vs. USD.

CZ

RO

PL

KO

HU

MX

TH

CO

CL

ZA

RU

PH

BR

ID

CH

IN

-0.2

CN SA KO PH BG TH PE RO PL CZ MY HU CL NG MX ID IN EG CO TR BR UK ZA RU AR

30

TR

0.0

40

CAPITAL FLOWS TO EMERGING MARKETS | APRIL 7, 2016

page 9

Nevertheless, as long as global activity continues at a moderate pace and major market stress is avoided, conditions should be in place for a continuing recovery of non-resident inflows to EMs, and an easing of resident outflows. 

A key assumption is that the Fed remains cautious in coming quarters. Consistent with the FOMC’s own projections, we project two rate hikes this year, in June and December, given growing evidence of tightening labor markets and incipient wage and price pressures. At the same time, we think the Fed would be ready to wait longer in the event of heightened market stress.



The second central assumption is that China maintains its commitment to avoid a disorderly movement in the RMB, doing what is needed to sustain growth in line with its 2016 growth target, even at the cost of slowing down structural reform and allowing already high debt levels to continue to rise.



Overall, we have made net downward revisions to our outlook for the global economy since our January report, but stepped-up policy support in both mature and emerging economies has reduced concerns about imminent global recession (Chart 23).

On this basis, we project private non-resident private capital flows to emerging markets to rise gradually during 2016, reaching $560 billion for the year as a whole, up from a 13-year low of $240 billion in 2015 (Chart 24 and Table 1). About half of this improvement relates to China, where nonresident inflows should turn positive again this year, driven by reduced repayments of external debt and a recovery in portfolio flows (see pages 22-24). We also project a substantial turnaround in flows to the rest of EM Asia, led by India, Korea, and Malaysia. The other EM regions should see a more modest recovery in capital inflows this year, with flows to EM Europe held back by continued flows out of Russia of around $42 billion. Looking at the composition of inflows, reduced global risk aversion and lower G3 interest rates should support continued portfolio inflows in most EMs in coming quarters, after exceptionally weak flows in 2015 (Chart 25). However, we expect FDI flows to soften in the context of continued Chart 23 Forecast Revisions to GDP Growth Relative to Jan. 2016 CFR percentage points 0.2 0.0 -0.2 -0.4

800

-0.6

2015

-0.8

Total EM

600

2016

Source: IIF.

AFME

EM Europe

EM Asia

LatAm

Japan

Euro Area

U.S.

EMs

Mature Econ

400

World

-1.0

Chart 24 Private Non-Resident Capital Flows to Emerging Markets IIF $ billion EM Europe Forecast 1400 Latin America Middle East/Africa 1200 EM Asia ex. China 1000 China

iif.com © Copyright 2016. The Institute of International Finance, Inc. All rights reserved.

200 0

-200

2000 2002 2004 2006 2008 2010 2012 2014 2016

Source: IIF.

CAPITAL FLOWS TO EMERGING MARKETS | APRIL 7, 2016

page 10

Table 1 Emerging Market Economies: Capital Flows $ billion 2013

2014

2015 Est.

2016 Proj.

1337

1098

292

602

1302

1058

240

560

667 584

654 557

533 516

532 474

84

97

17

58

Commercial Banks

635 190

404 193

-293 -216

27 -34

Nonbanks

445

210

-77

62

Official Inflows

35 3

41 16

52 21

42 26

Non-Resident Capital Inflows Private Inflows Equity Investment Direct Investment Portfolio Investment Private Creditors

International Financial Institutions Bilateral Creditors Resident Capital Outflows Private Outflows

32

25

31

16

-1342

-1228

-372

-694

-808

-1101

-862

-977

Equity Investment Abroad

-367

-393

-423

-448

Resident Lending/Other

-441

-708

-439

-529

-533

-126

490

283

528

-3

-570

-375

323

-175

-755

-501

-206

-172

-185

-126

210

302

265

218

Reserves (- = Increase) Net Capital Flows excl. Reserves (Financial Account Balance) Net Capital Flows plus Errors and Omissions Memo: Net Errors and Omissions Current Account Balance Source: IIF. See page 29 for guidance on how to interpret these data.

concerns about growth prospects, while banking flows to EMs would remain negative, reflecting both continued unwinding of borrowing by Chinese corporates and more general constraints on lending by global banks facing a more challenging regulatory environment and generally weak profitability (see Box 1 on page 12). Relative to our January Capital Flows Report, we have revised upwards our projections for nonresident inflows in 2016 by $78 billion for 2016, entirely reflecting an improved outlook for debt flows, both bank lending and bond flows. On a country basis, most of the upward revisions reflect somewhat reduced debt related outflows from China. By contrast, we have lowered our projections for capital flows to Russia, Turkey, and Ukraine by about $10 billion each. Net capital flows to emerging markets are projected to be negative for a second consecutive year in 2016, albeit at a somewhat reduced pace than in 2015. We project net outflows (including unrecorded capital outflows captured by net errors and omissions) of $501 billion, down from $755 billion in 2015 (Chart 26). Again, this improvement is mainly due to a projected easing in outflows from China (especially on the non-resident side), after a very weak outcome in 2015. Taking account of the aggregate EM current account surplus for EMs as a group, we project that EM reserves would continue to decline, but at a diminished pace. China and Middle Eastern oil

iif.com © Copyright 2016. The Institute of International Finance, Inc. All rights reserved.

CAPITAL FLOWS TO EMERGING MARKETS | APRIL 7, 2016

page 11

Chart 25 Private Non-Resident Capital Inflows by Component Other $ billion 450 Portfolio Debt 400 Portfolio Equity FDI 350 IIF Forecast 300 250 200 150 100 50 0 -50 -100 -150 2012Q1 2013Q1 2014Q1 2015Q1 2016Q1

Chart 26 Emerging Markets: Net Capital Flows, C/A Balance, Reserve Accumulation IIF $ billion Forecast Reserve 1,200 C/A Balance Accumulation 1,000 800 600 400 200 0 -200 -400 -600 Net Capital Flows * -800 -1,000 1995 2000 2005 2010 2015

Source: IIF.

Source: IIF.

*Calculated as the financial account balance ex. reserves, plus errors & omissions.

exporters would continue to lose reserves, but at a more manageable pace than in 2015, while other countries would return to reserve accumulation (Chart 27). While our overall outlook for EM capital flows shows a moderate recovery, downside risks remain considerable. If U.S. inflation were to continue to accelerate at the pace observed in recent months, there remains a clear risk that the Fed will have to move more rapidly than priced in by the market, which could lead to a sharp upward shift in rates and renewed dollar appreciation (Chart 28). Such developments could be quite disruptive for EM markets and most importantly could substantially complicate the PBOC’s efforts to stabilize the RMB, by exacerbating concerns about RMB depreciation.

Chart 27 EM FX Reserve Accumulation by Region $ billion EM Europe Latin America 1400 Middle East/Africa 1200 EM Asia ex. China 1000 China 800

IIF Forecast

Chart 28 March FOMC Dot Plot, Market Expectations, IIF Forecast percent 4 FOMC Dot Plot as of 3/2016 IIF Forecast 3

Total EM

600 400

2

Market Exp. as of Jan. 1, 2016

200 0

1

-200

Market Exp. as of April 4, 2016

-400 -600

2000 2002 2004 2006 2008 2010 2012 2014 2016

Source: IIF.

iif.com © Copyright 2016. The Institute of International Finance, Inc. All rights reserved.

0

2016

2017

Source: U.S. Federal Reserve, Bloomberg, IIF.

2018

CAPITAL FLOWS TO EMERGING MARKETS | APRIL 7, 2016

page 12

BOX 1: BANKS UNDER PRESSURE: IMPLICATIONS FOR EMERGING MARKETS As noted in our April 2016 Capital Markets Monitor, faced with a challenging economic and regulatory environment, profitability in the mature market banking sector has been under pressure—particularly for Euro Area banks. The combination of weak loan demand, low/negative policy rates, and flat yield curves has weighed on margins, as have increased regulatory requirements and legacy costs of cleaning up balance sheets. While capital and liquidity ratios are much improved, returns on equity have weakened across the board and for many Euro Area banks is well below the cost of capital. Against this backdrop, pressure to retrench from riskier business lines, including lending in emerging markets, has continued. At the same time, weak growth prospects across emerging markets do not offer much incentive for foreign banks to lend. The impact of these factors on banking flows to emerging markets is increasingly clear. After recovering from the contraction in the wake of the global financial crisis, mature market banks have recently been cutting back

Chart 31 International Claims on EMs

or not increasing exposure to emerging market borrowers, apart from Japan, domestic market. As a result, international claims of G4 banks on emerging markets have declined by over 10% from their recent peak in mid-2014 through Q3 2015 (the last available data point from the BIS statistics) (Chart 29). Weak syndicated loan issuance in emerging markets during Q1 2016—down over 50% from the same period in 2015—suggests that this trend has continued (Chart 30). This was in line with the findings of our EM Bank Lending Survey, which shows that international funding conditions for EM banks has increasingly tightened since 2014Q3. Part of the retrenchment reflects an unwinding of bank lending to Chinese corporates eager to reduce U.S. dollar

Thousands

where banks have been looking to diversify away from ultra-low returns in their

$ trillion, BIS-reporting banks 4.0 China 3.5 Other EMs 3.0 2.5 2.0 1.5 1.0 0.5 0.0 2000 2005 2010 2015 Source: BIS, IIF.

exposure as the RMB has weakened, but international claims have also declined to EMs ex-China (Chart 31). Chart 29 BIS-Reporting G4 Banks: International Claims on EMs $ billion, immediate counteryparty basis 2500 EA

UK

Japan

Chart 30 EM25 Loan Issuance $ billion, includes sovereigns and corporates 100 90

U.S.

2000

80 70

1500

China Others

60 50

1000

40 30 20

500

10 0 2000

Source: BIS, IIF.

2005

2010

2015

iif.com © Copyright 2016. The Institute of International Finance, Inc. All rights reserved.

0 Jan 13

Oct 13

Source: Thomson One, IIF.

Jul 14

Apr 15

Jan 16

CAPITAL FLOWS TO EMERGING MARKETS | APRIL 7, 2016

page 13

THE GREAT UNWINDING: CAPITAL OUTFLOWS FROM CHINA Following years of large foreign inflows and a record build-up of international reserves, China experienced massive net capital outflows of close to $675 billion in 2015 (including unrecorded capital outflows captured by net errors and omissions; Table 2). These net outflows reflected both reductions in non-resident claims on China and increased efforts by Chinese residents to move funds overseas. Net flows have in fact been negative since 2014Q2, but the retrenchment was particularly pronounced in the second half of last year when net outflows accelerated to around $433 billion (Chart 32). FX debt repayments by non-financial corporates and the decline in offshore RMB deposits have been key contributors to the large non-resident outflows, propelled by a shift in expectations about the RMB’s path from appreciation to depreciation (Chart 33). However, it is striking that the channels for outflows have broadened in recent quarters, expanding to pullbacks in nonresident portfolio equity and debt investment, as well as rising resident outflows and error and omissions. Given the scale of these capital outflows from China and their impact on the broader financial market environment affecting EM capital flows, this section provides a deeper, more granular analysis of the channels for capital outflows over the past years to help provide guidance about prospects for future flows. For example, some foreign investments are easier to unwind than others, and for some components future outflows will be limited by the outstanding stock of liabilities, e.g. foreign holdings of domestic debt. For this analysis, we distinguish three components of capital flows: (1) non-resident capital inflows, recorded on the liability side of China’s financial account, (2) resident capital

Table 2 China: Capital Flows $ billion 2013

2014

2015 2016 Proj.

283

-160

-674

-532

343 563 291 33

-51 411 268 52

-486 -94 250 15

-405 60 210 20

Portfolio Debt Inflows Other Inflows

26 214

41 50

-8 -350

26 -196

Resident Capital Outflows ODI Portfolio Equity Outflows Portfolio Debt Outflows Other Outflows

-220 -73 -3 -3 -142 -431 -60 148

-463 -123 -1 -9 -329 -118 -108 277

-392 -188 -40 -34 -131 343 -188 331

-465 -213 -36 -34 -182 243 -127 289

Net Capital Flows plus Errors and Omissions Net Capital Flows excl. Reserves (Financial Account Balance) Non-Resident Capital Inflows FDI Portfolio Equity Inflows

Reserves (- = Increase) Net Errors and Omissions Current Account Balance

Source: SAFE, IIF. See page 29 for guidance on how to interpret these data.

iif.com © Copyright 2016. The Institute of International Finance, Inc. All rights reserved.

CAPITAL FLOWS TO EMERGING MARKETS | APRIL 7, 2016

Chart 32 China Quarterly Capital Flows $ billion 250

Net Errors & Omissions Resident Outflows Non-Resident Inflows

175 100

25 -50 -125 -200 -275

Net Capital Flows Plus Errors & Omissions 12Q1

12Q4

Source: SAFE, Haver, IIF.

13Q3

14Q2

15Q1

15Q4

page 14

Chart 33 China: Net Capital Flows vs CNYUSD $ billion % change over a year ago 140 7 CNYUSD, % oya 120 6 100 5 80 4 60 3 40 2 20 1 0 0 -20 -1 -40 -2 -60 -3 Net Capital Outflows -80 -4 -100 -5 -120 -6 -140 -7 2010 2011 2012 2013 2014 2015 2016

Source: SAFE, Bloomberg, IIF.

outflows, recorded on the asset side of China’s financial account, and (3) incorrectly recorded or unrecorded financial account transactions (captured by net errors and omissions). Which of these three components has been most important in driving net capital outflows from China? A simple first step to answer this question is to look at cumulative flows from the time when net outflows began in 2014Q2 through 2015Q4 (Chart 34). This simple metric indicates that just over half of all gross outflows of around $1.5 trillion were due to residents increasing their foreign assets, while roughly a quarter were due to non-residents reducing their holdings of Chinese assets (and the remaining quarter were related to net errors and omissions). Foreign direct investment into China remained a substantial net source of capital inflows over this period. However, this metric does not take into account that net outflows of resident capital are normal and to be expected, while outflows of non-resident capital are not. In other words, Chinese residents generally raise their holdings of foreign assets even in tranquil times, Chart 34 China Capital Flow Components $ billion; cumulative since 2014Q2 600 400 200 0 -200 -400 -600 Errors & Omissions -800 Other Outflows -1000 Portfolio Debt Outflows -1200 Portfolio Equity Outflows -1400 ODI Net Capital -1600 Other Inflows Flows plus E&O -1800 Portfolio Debt Inflows -2000 Portfolio Equity Inflows -2200 FDI 14Q2 14Q3 14Q4 15Q1 15Q2 15Q3 Source: SAFE, Haver, IIF.

Chart 35 China Cumulative Flows During the Retrenchment Deviation from 2011-2013 Average $ billion, cumulative since 2014Q2 200 0

-200 -400 -600 -800 -1000 -1200 -1400

-1600 15Q4

-1800

Errors & Omissions Other Outflows Portfolio Debt Outflows Portfolio Equity Outflows ODI Other Inflows Net Capital Portfolio Debt inflows Flows plus E&O Portfolio Equity inflows FDI 14Q2 14Q3 14Q4 15Q1 15Q2 15Q3

Source: SAFE, Haver, IIF.

Note: Resident outflows components shown in shades of blue, non-resident inflows components in shades of red/orange. iif.com © Copyright 2016. The Institute of International Finance, Inc. All rights reserved.

15Q4

CAPITAL FLOWS TO EMERGING MARKETS | APRIL 7, 2016

Chart 37 International Investment Position "Other" Liability Components $ billion 1600 Trade Credit & Advances 1400 Currency & Deposits

Chart 36 Capital Inflow Components and IIP Liabilities $ billion; flows cumulative since 2000Q1 5000 4500

Other Inflows

page 15

Total IIP Liabilities

4000

Portfolio Debt Inflows

3500

Portfolio Equity Inflows

1200

Loans

3000

FDI

1000

Other

2500 1500

Total

800

2000

600

Total Inflows

400

1000

200

500 0 2000

2005

2010

2015

Source: SAFE, Haver, IIF.

0 2004

2006

2008

Source: SAFE, Haver, IIF.

while non-residents normally do not decrease their holdings of domestic assets (as has recently been the case). A more holistic way to analyze recent net outflows is to look at how the three components have changed relative to previous trend. This is done in Chart 35, which illustrates how flows have shifted relative to average quarterly flows in the 2011-2013 period, showing the cumulative difference for each component since net capital outflows began in 2014Q2. The chart highlights the huge swing in non-resident capital flows, which shifted from large inflows to sizeable outflows, accounting for over half of the total change from the previous trend. Resident capital outflows and capital flight captured by net errors and omissions have also experienced a substantial increase, but this has been more limited in magnitude compared to non-resident capital flows. FDI inflows have continued broadly in line with their earlier trend. Non-Resident Capital Flows: The Great Reversal Until recently, China was a large recipient of foreign capital inflows, accounting for almost 30% of total inflows to EMs in the period from 2000 to 2014. Adding up the inflows since 2000, cumulative foreign investments in China reached $3.6 trillion in 2014Q4, before nonresident inflows turned negative (Chart 36). Stock data on China’s international investment position, which take into account changes in the valuation of past investments, show that the market value of those investments was even higher, reaching $4.6 trillion as of 2014Q4. When compared to the large magnitude of prior inflows, the recent retrenchment in foreign capital does not seem particularly large, totaling $130 billion in 2015H2 for all nonresident inflows components taken together (on a flow basis). The dominant source of foreign capital for China has been foreign direct investment (FDI), which accounts for 61% of the liabilities in China’s international investment position at end2015. FDI is typically long-term in nature, depends on underlying business strategies and cannot be unwound quickly. In fact, FDI inflows have slowed only moderately in recent quarters, remaining positive at $153 billion in 2015H2. Meanwhile, portfolio debt and equity flows were negative in 2015H2, but the scale of these flows is much smaller than FDI, in part due to restrictions on non-resident portfolio investments in China. In 2015H2, global

iif.com © Copyright 2016. The Institute of International Finance, Inc. All rights reserved.

2010

2012

2014

CAPITAL FLOWS TO EMERGING MARKETS | APRIL 7, 2016

Chart 39 Offshore Renminbi Deposits $ billion 300 Korea 250 Macao SAR

Chart 38 International Bank Claims on China $ billion Other 1250

Euro Area 1000

UK

200

Foreign Subsidaries of ChinaOwned Banks

750 500

250 0

page 16

Singapore

150

Chinese Taipei

100

Hong Kong

50

2007 2008 2009 2010 2011 2012 2013 2014 2015

Source: BIS, IIF.

0 2008

2010

2012

2014

2016

Source: National central banks, Haver, Bloomberg, IIF.

investors were net sellers of $26 billion of portfolio equity and debt securities, or 7% of cumulative portfolio inflows since 2000. By contrast, the retrenchment has been intense in the “other investment” component of foreign capital inflows, which saw accelerated outflows of $223 billion in 2015H2 on a flow basis (Chart 37). This category broadly encompasses bank-related flows including foreign bank loans to Chinese residents, trade credit, and currency and deposits. Among these components, the unwinding of flows in 2015H2 was particularly rapid in cross-border loans ($91 billion) and currency and deposits ($105 billion). There have been two main drivers of this sharp retrenchment in “other inflows,” namely (1) Chinese corporates’ accelerated repayments of foreign currency debt and (2) reduced interest by international investors in holding offshore RMB deposits. The first driver largely explains the sharp decline in bank lending to China, while the second driver primarily explains the reduction in non-resident holdings of currency and deposits.1 Chart 41 China: Non-Financial Corporate Foreign Currency Debt $ billion Decline in USD debt 1200

Chart 40 China: External Debt $ billion

2000 1800 1600 1400 1200 1000 800 600 400 200 0

RMB (2015Q1-Present) RMB-adjusted (2005-2014) Foreign Currency

Decline in JPY and CHF Debt

1000 800

Total

600

Other Foreign Currencies EUR USD

400 2005

2007

2009

2011

2013

2015

Source: SAFE, IIF; SAFE changed its definition of external debt in 2015Q1 to include external debt denominated in RMB, leading a jump in debt from $895bn in 2014Q4 to 1.6trn in 2015Q1. We adjust historical data by assuming that the currency decomposition of debt between 2005 and 2014 was the same as in 2015Q1. 1To

200 0 2005

2007

2009

2011

2013

Source: BIS, IMF, National sources, IIF EM Debt Monitor.

2015

some extent, reduced offshore RMB deposits may lead to outflows in BoP components other than non-resident holdings of currency and deposits. For example, this would occur if offshore banks reduce their loans to onshore banks when they experience withdrawals of RMB deposits. iif.com © Copyright 2016. The Institute of International Finance, Inc. All rights reserved.

CAPITAL FLOWS TO EMERGING MARKETS | APRIL 7, 2016

page 17

Further insights can be gained by looking at BIS consolidated banking statistics. BIS data through 2015Q3 show that around 40% of the total decline in international banks’ claims on China was attributable to foreign subsidiaries of Chinese banks, mainly located in Asian hubs (Chart 38). Much of this decline seems related to the sharp drop in offshore RMB deposits observed in Asian hubs since June 2015 (down by $65 billion). A significant portion of the retrenchment in cross-border bank claims was thus denominated in RMB as offshore RMB deposits—which had been built up in part due to the expectations of RMB appreciation—were reduced due to concerns about RMB depreciation. Banks outside China in turn reduced their cross-border renminbi deposits with banks domiciled in China (Chart 39). In line with this, SAFE’s external debt statistics also show a substantial decline in RMB‑denominated external debt, exceeding the decline in foreign currency denominated external debt (Chart 40). Focusing on non-financial corporates, our March 2016 EM Debt Monitor estimates that they reduced their foreign currency debt by around $70 billion to $0.9 trillion in 2015H2 (Chart 41).2 The reduction in foreign currency debt of non-financial firms predominantly reflected reduced FX lending provided by onshore banks, and to a lesser extent repayments of loans from non-residents (Chart 42). However, the decline in firms’ foreign currency onshore loans appears to have induced domestic banks to reduce their international liabilities at the same time, registering as reduced interbank liabilities. Despite the decline in China’s foreign liabilities last year, the stock of external debt is still substantial and remains a source of potential pressure on the balance of payments. Overall, the retrenchment in debt flows topped $232 billion in 2015H2, bringing the stock of foreign liabilities (excluding equity), measured by IIP data, to around $1.4 trillion at end2015 (Chart 43). This is broadly consistent with SAFE external debt statistics, which show a decline of $264 billion in 2015H2 to around $1.4 trillion (Chart 40), with short-term and Chart 42 BIS-reporting Banks' Cross-border Claims on Chinese Banks $ billion 800 700

Bank (Gross)

600

Bank (Net)

Other (Mainly Intercompany Lending) Portfolio Debt Securities Trade Credit Loans Currency and Deposits

2000

500

1500

400 300

1000

200 100 0

Chart 43 International Liabilities, Excluding Equity $ billion, IIP data 2500

500 2010

2011

2012

2013

2014

2015

Source: BIS, IIF. Gross claims represent Chinese bank's liabilities to BIS-reporting banks. Net claims shows the difference between Chinese banks liabilities to BISreporting banks and BIS-reporting banks' liabilities to Chinese banks.

2 Information

0

2011

Source: SAFE, IIF.

2012

2013

2014

2015

on the contribution of non-financial firms’ FX debt repayments on non-resident capital outflows can be obtained from IIF EM Debt Monitor, which provides a comprehensive foreign currency debt database based on national sources on domestic bank lending, the IMF’s Financial Soundness Indicators and the BIS’s debt and credit statistics.

iif.com © Copyright 2016. The Institute of International Finance, Inc. All rights reserved.

CAPITAL FLOWS TO EMERGING MARKETS | APRIL 7, 2016

page 18

foreign currency liabilities making up around 65% and 54% of the total stock of external debt, respectively. Assuming that a further $68 billion debt was repaid in 2016Q1, consistent with our estimated non-resident capital outflows during this period, this would imply that China’s external debt stock remains at just under $1.35 trillion.3 However, of this total the combined stock of loans and currency and deposits has now been reduced to only around $656 billion, down from $1.075 trillion at the end of 2014. The rest is accounted for by trade credits and other more sticky forms of debt. This calculation suggests that the unwinding process has already gone quite a long way but could potentially continue to feed significant outflows. Over the medium term, China should benefit from structural factors supporting nonresident portfolio debt and equity inflows. Following the IMF’s decision to include the RMB in the SDR basket, SAFE relaxed its quota restrictions on QFIIs regime to encourage foreign institutional investors to buy RMB-denominated bonds. This move has already induced major providers of emerging market indices (both bond and stock market indices) to consider inclusion of China’s domestic securities in their benchmark indices. As and when this happens, potential non-resident portfolio inflows could be substantial, although restrictions on capital mobility and concerns about currency and liquidity risks will likely continue to weigh on institutional demand for some time to come. Resident Outward Capital Flows: Expanding Horizons In contrast to non-resident capital inflows, resident outward investment is not inherently limited in scale, although Chinese investors’ purchases of foreign assets are restricted (notwithstanding considerable leaks, as discussed in the next section). Since the global financial crisis, resident outward investment has grown rapidly in the context of gradual relaxation of restrictions on outward investment (Chart 44). The increase in Chinese resident accumulation of foreign assets has in large part reflected China’s growing integration with the rest of the world economy, as well as a desire for portfolio Chart 44 Resident Capital Outflow Components $ billion; cumulative since 2000Q1

Chart 45 Other Investment Outflow Components $ billion; cumulative since 2000Q1

500

200

0

0 -200

-500

-1000

-400

Total Outflows

-600 -800

-1500 -2000 -2500 -3000 2000

Other Outflows

-1000

Portfolio Debt Outflows

-1200

Portfolio Equity Outflows ODI

Source: SAFE, Haver, IIF.

2005

2010

2015

Total Other Currency & Deposits

-1400

Trade Credit

-1600 2000

Loans

Source: SAFE, Haver, IIF.

2005

2010

Although SAFE’s new external debt statistics start from 2015Q1, data from the international investment position can be used to approximate historical data on external debt. Indeed, China’s IIP data suggest that the stock of external liabilities declined by around $362 billion in 2015 to around $1.4 trillion at end-2015 (excluding equity). 3

iif.com © Copyright 2016. The Institute of International Finance, Inc. All rights reserved.

2015

CAPITAL FLOWS TO EMERGING MARKETS | APRIL 7, 2016

Chart 46 Equity Home Bias x-axis: GDP per capita ($ thousand) y-axis: home bias*; 1=perfect home bias (no foreign holdings) 1.1

India

1.0 0.9

Russia

China

Korea

Brazil

0.8 0.7

page 19

Chart 47 China's International Investment Position, 2004-2015 $ trillion Reserves 8 Other Investment Portfolio Investment 6 Direct Investment NIIP 4 2

Japan

0.6

0

UK

U.S.

0.5

-2

-4

0.4 0

10

20

30

40

50

60

Source: National sources, Haver, IIF. *Home bias defined as a ratio, with 1 = perfect home bias.

-6

2005

2007

2009

Source: SAFE, IIF.

diversification. Sustained overseas expansion of Chinese bank lending—usually in support of foreign direct investment—has been an important part of this process (Chart 45). Since mid-2014, however, outward investment has accelerated, driven by rising concerns about RMB depreciation and the prospect of slowing domestic growth. Decomposition of the components of foreign asset accumulation shows a marked increase in resident holdings of bank-related foreign assets, particularly of currency and deposits. These investments have presumably been facilitated by liberalization initiatives since September 2006, permitting Chinese residents to make overseas investments up to $50,000 per year. This process of international portfolio diversification has only just begun. At present, the home bias of Chinese investors is much higher than in mature economies (Chart 46), and this should gradually decline as income levels continue to rise and China continues to liberalize its financial account. Portfolio diversification should thus support private demand for foreign equities and debt securities in coming years. Foreign asset holdings of the private sector still account for less than half of China’s total international assets, but this share should rise over time while the share held in official international reserves should come down (Chart 47). Errors and Omissions: Behind the Scenes The final element for tracking China’s capital flows is to analyze BoP errors and omissions, which have been persistently and increasingly negative in recent years. In our view, errors and omissions largely reflect unreported capital transactions, an assessment supported by the observation that they have moved closely in line with shifts in net capital flows. A close look at China’s customs trade data and banks’ transactions for foreign trade suggests that over-reporting of imports has been a key factor behind the sharp increase in errors and omissions. In particular, bank transaction data show that importers in China paid around $2.5 trillion for goods imported since 2014, but customs good data only shows $1.8 trillion of imports (Charts 48). The discrepancy between the corresponding data on goods

iif.com © Copyright 2016. The Institute of International Finance, Inc. All rights reserved.

2011

2013

2015

CAPITAL FLOWS TO EMERGING MARKETS | APRIL 7, 2016

Chart 48 China: Import Discrepancy $ billion (both scales) Bank Payments for Goods 230 210

80 60

Gap (rhs)

190 170 150 130

Chart 49 China: Export Discrepancy $ billion (both scales) Bank Receipts 230 for Exports 210

70 60 50

190

40

40

170

30

20

150

20

130

10

0

0

110

110 -20

Goods Imports

90 70 2010

page 20

2011

2012

2013

2014

2015

2016

-40

Source: SAFE, IIF.

90

70 2010

Gap (rhs) 2011

2012

-10

Goods Exports 2013

2014

-20 2015

2016

-30

Source: SAFE, IIF.

exports has been more modest (Chart 49). In principle, the difference between BoP trade balance and banks’ net transactions should correspond to net trade credits/advances in the balance of payments. In practice, however, the discrepancy between these two series has been much higher than the trade credit & advances recorded in China’s balance of payments (Chart 50). This suggests that overvaluation of imports and the associated unrecorded capital outflow has been the main driver behind the large error and omissions in recent years (Chart 51). Capital Outflows, the RMB Regime, and Capital Controls Prospects for continuing capital outflows from China are crucial for the outlook for EM capital flows, for two main reasons. First, China simply accounts for a large part of overall EM capital flows. Second, sustained outflows could lead to further heavy loss of reserves that could force a shift in the approach to exchange rate management and could result in a disorderly depreciation that would have heavy spillovers to other EMs (Chart 52). A sharp drop in the RMB would likely spark a renewed sell-off of global risk assets and trigger a flight of portfolio capital from EMs. Moreover, a sharp depreciation of the RMB could Chart 50 Trade Credit vs Gap btw Bank Transactions and Goods Balance $ billion 140 Difference between Net Bank 120 Transactions and BOP Good Balance 100

Chart 51 Trade Credit vs Gap btw Bank Transactions and Goods Balance $ billion 140 Difference between Net Bank 120 Transactions and BOP Good Balance 100

80

80

60

60

40

40

20

20

0

0

-20

-20

Trade Credit/Advances Reported Under BOP

-40 -60 2010

2011

2012

2013

2014

2015

Source: SAFE, IIF.

iif.com © Copyright 2016. The Institute of International Finance, Inc. All rights reserved.

Trade Credit/Advances Reported Under BOP

-40 -60 2010 Source: SAFE, IIF.

2011

2012

2013

2014

2015

CAPITAL FLOWS TO EMERGING MARKETS | APRIL 7, 2016

Chart 52 China's Reserve Operations vs. EM Asia Portfolio Flows monthly change, $billion $ billion EM Asia Non-Resident 20 25 Portfolio Flows 0

20

-20

15

-40

10

-60

5 Reserve Operations

-80 -100 -120

Jan 15

Apr 15

0 -5 Jul 15

Oct 15

Jan 16

-10

Source: PBOC, Haver, IIF.

page 21

Chart 53 Chinese Renminbi and the Dollar index, end-2014=100 112

DXY

110

108 106

104

CNY vs. CFETS Basket

102 100 98 96 94 92 Jan 15

CNYUSD

Apr 15

Jul 15

Source: Bloomberg, CFETS, IIF.

lead to a round of competitive devaluation in other emerging markets, particularly in those with close trade linkages to China. Stabilizing market expectations of the RMB’s path will be crucial in determining whether a disorderly devaluation of the RMB can be avoided. The greater investors’ expectations of RMB depreciation become, the greater their incentive to reduce RMB assets and foreign currency liabilities, reinforcing the pressure from capital outflows on the currency. In recent months, the authorities have made some progress in re-anchoring investor expectations of the exchange rate by increasing communications that have emphasized an increased focus on the RMB’s valuation vis-à-vis a basket of currencies, rather than just vis-à-vis the U.S. dollar. Stabilization of the RMB has also been helped by a relatively soft performance of the USD against other major currencies (Chart 53). Pressure on the RMB could rise if the dollar starts to strengthen again but managing against a basket should reduce the need for the PBOC to support the RMB by allowing for a moderate degree of depreciation against the dollar. However, a sustained depreciation against the dollar and a continued loss of reserves could call into doubt the authorities’ commitment to maintaining RMB stability. In this scenario, there could be a self-fulfilling dynamic whereby investors rush to the exit, intensifying pressure on the PBOC to devalue. One important unknown is the threshold below which the PBOC would start to worry that Chinese reserves were falling below an adequate level and feel a need to either curtail interventions and allow greater flexibility, or markedly tighten capital controls. Reserves have already fallen from $4 trillion in June 2014 to around $3.2 trillion in February 2016—but still look high relative to other countries on some standard metrics like months of import cover (16 months). However, reserves look less impressive by other measures, for example as a ratio to M2 (a measure of vulnerability to capital flight by domestic residents). To balance different factors, the IMF has developed a composite methodology for assessing the adequacy of foreign reserve holdings, which takes into account a set of country-

iif.com © Copyright 2016. The Institute of International Finance, Inc. All rights reserved.

Oct 15

Jan 16

CAPITAL FLOWS TO EMERGING MARKETS | APRIL 7, 2016

Chart 54 Adequate Levels of Reserves under Different Assumptions $ billion 4.5 Actual Stock of Reserves 4.0 3.5

Chart 55 Reserve Adequacy Ratio percent of risk weighted metric With Capital Control 400 350

Without Capital Control or Ineffective Capital Controls

300

3.0

Ineffective Capital Controls

2.5

250

200

2.0

100-150% of the ARA metric is typically considered adequate for precautionary purposes.

150 Effective Capital Controls

1.0 0.5

100 50 0

2008

2010

2012

2014

2016

Source: National sources, IIF.

UA TR EG CL HU PL ZA MY ID CZ MX CN CO KR RU IN BR TH PH

1.5

0.0 2006

page 22

Source: National sources, IIF.

specific factors ranging from external liabilities to export revenues.4 The IMF then develops a range for reserve adequacy, depending on the extent and effectiveness of capital controls. Our calculations based on this methodology suggest that an adequate level of reserves for China falls in the range of $1.5 trillion (assuming controls are fully effective) and $2.8 trillion (assuming that capital controls are ineffective). The current level of FX reserves in China is around 15% above the high end of this range, a cushion that has been reduced from 50% in mid-2014, and quite low compared to other EM Asian countries (Charts 54 and 55). From this perspective, continued large capital outflows could lower the country’s official reserves to a level that is regarded as inadequate without a serious tightening of capital controls. The sustained depletion of reserves amidst large capital outflows has already prompted Chinese authorities to tighten the restrictions on cross-border flows. These measures have included restrictions on the capacity of foreign firms to repatriate earnings, interventions to offshore RMB markets to tighten CNH liquidity in an attempt to reduce speculative short positions against the currency, limitations on RMB-based funds under the QDII quota scheme, and constraints on domestic banks’ foreign currency transactions. Outlook for Capital Outflows and the RMB In our baseline projections, capital outflows and pressure on the RMB persist through 2016, but diminish in intensity as policymakers persuade investors that they will be successful in stabilizing the RMB’s value. Greater confidence that the economy is on track to meet the growth target of 6.5-7 percent would also help (Chart 56). Most of the improvement is projected to be in a turnaround in non-resident capital flows, while resident outflows and errors and omissions are forecast to moderate more gradually.

4 In

particular, the IMF proposes to calculate reserve adequacy based on four risk variables: (i) short-term external debt, (ii) other external liabilities, (iii) broad money (M2) and (iv) exports. As the reserve adequacy estimates are subject to change depending on the choice of the variable weights, and the assumptions on the degree/effectiveness of capital controls, they should be used with a caution. For instance, if we assume capital controls are effective in China, this will imply that current level of reserves are around 90% above the adequate level of reserves. iif.com © Copyright 2016. The Institute of International Finance, Inc. All rights reserved.

CAPITAL FLOWS TO EMERGING MARKETS | APRIL 7, 2016

Net Errors & Omissions Chart 56 China Quarterly External Accounts Resident Outflows $ billion Non-Resident Inflows 250 Reserves IIF C/A Balance (- = increase) 175 Forecast

Chart 57 China: FDI Capital Utilized percent ot total

80 70 60

100

Manufacturing

50

25

40

-50

Services

30

-125

20

-200 -275

page 23

12Q1

Net Capital Flows Plus Errors & Omissions 12Q4 13Q3 14Q2 15Q1

Real Estate

10 15Q4

0

16Q3

2001

2003

Source: CEIC, IIF.

Source: SAFE, Haver, IIF.

2005

2007

In particular, we project non-resident capital flows to turn positive beginning in 2016Q2 and to pick up further towards the end of the year, helped by a slowing pace of loan repayments and a more constructive backdrop to portfolio flows (Chart 58). As has been the pattern in recent years, FDI inflows should be supported by growing investment in China’s services sector, even as manufacturing FDI declines as a percentage of total FDI (Chart 57). Meanwhile, resident capital outflows are projected to stabilize at around $120 billion per quarter, as portfolio diversification remains a driver for outflows in the context of more stable RMB expectations. Similarly, net errors and omissions should level off around -$25 billion (indicating continued unrecorded capital outflows)—significant but well below levels observed in 2015H2. In this scenario, reserve depletion would slow from $276 billion in 2015H2 to $165 billion in 2016H1 and $72 billion in 2016H2. Leaving aside possible valuation effects from the dollar Errors & Omissions Other Outflows Portfolio Debt Outflows Portfolio Equity Outflows ODI Other Inflows Portfolio Debt Inflows Portfolio Equity Inflows FDI

Chart 58 China: Capital Flows $ billion 200

150 100

50 0 -50

-100 -150 -200 -250 -300

IIF Forecast 14Q1

14Q3

15Q1

15Q3

16Q1

16Q3

Note: Resident Outflows components shown in shades of blue, non-resident inflows components in shades of red/orange. Source: SAFE, IIF.

iif.com © Copyright 2016. The Institute of International Finance, Inc. All rights reserved.

2009

2011

2013

2015

CAPITAL FLOWS TO EMERGING MARKETS | APRIL 7, 2016

page 24

moving against third currencies, this would leave the 2016 year-end level of reserves still close $3 trillion, down 25% from their peak in June 2014, but still at a comfortable level relative to the IMF yardstick. Recent market developments provide some support for the view that this relatively benign scenario is playing out. On the back of improvements in global investor sentiment, emerging markets in the aggregate experienced sizeable portfolio inflows in March. Available data on fund flows and Shanghai-Hong Kong stock connect suggest that China is likely to have benefitted from these inflows since February (Charts 59 and 60). Nonetheless, China remains vulnerable to a renewed intensification of capital outflows, particularly if doubts about the stability of the RMB were to come to the fore again. In particular, stress could rise through a combination of a stronger USD—particularly if the market comes to believe that the Fed will tighten more quickly than currently priced in— and further disappointment about China’s growth momentum. Our analysis highlights that notwithstanding considerable paydown of dollar debt to nonresidents, there remains much scope for further capital outflows in such a situation, both in terms of non-residents unwinding positions in China and in terms of residents taking their funds out of the country.

Chart 59 China: Cumulative Fund Flows $ billion Offshore Flows 10

Chart 60 China: Cumulative Portfolio Equity Flows via Shanghai-Hong Stock Connect $ billion, cumulative flows 45

5

40

0 -5

30

-10

25

-15

20

-20

15

Onshore Flows

-25

10

Northbound (Flows to China)

5

-30 -35 2013

Southbound (Flows to Hong Kong)

35

2014

2015

2016

Source: EPFR, IIF.

iif.com © Copyright 2016. The Institute of International Finance, Inc. All rights reserved.

0 Jan 15

Apr 15

Source: Bloomberg, IIF.

Jul 15

Oct 15

Jan 16

Apr 16

CAPITAL FLOWS TO EMERGING MARKETS | APRIL 7, 2016

page 25

Table 3 Emerging Asia: Capital Flows $ billion Non-Resident Capital Inflows Private Inflows Equity Investment Direct Investment Portfolio Investment Private Creditors Commercial Banks Nonbanks Official Inflows International Financial Institutions Bilateral Creditors Resident Capital Outflows Private Outflows Equity Investment Abroad Resident Lending/Other Reserves (- = Increase)

2013

2014

2015 Est. 2016 Proj.

715 707 415 369 47 292 137 155 8 3 5

638 633 417 347 70 216 109 107 6 4 2

-32 -38 325 331 -6 -363 -238 -125 7 4 2

247 239 329 301 29 -90 -98 8 9 4 4

-788 -332 -156 -175 -456

-868 -671 -202 -469 -197

-227 -565 -302 -264 338

-542 -742 -325 -417 199

Net Capital Flows excl. Reserves (Financial Account Balance)

384

-33

-597

-494

Net Capital Flows plus Errors and Omissions

271

-151

-791

-620

-113 185

-118 347

-194 453

-125 420

Memo: Net Errors and Omissions Current Account Balance Source: IIF. See page 29 for guidance on how to interpret these data.

iif.com © Copyright 2016. The Institute of International Finance, Inc. All rights reserved.

CAPITAL FLOWS TO EMERGING MARKETS | APRIL 7, 2016

page 26

Table 4 Emerging Europe: Capital Flows $ billion 2013

2014

230 228 60 63 -3 168 29 139 2 -7 9

60 52 58 66 -8 -6 22 -28 8 7 1

7 -6 39 45 -6 -46 -9 -36 14 10 4

12 3 42 40 2 -39 4 -43 9 11 -3

-172 -175 -100 -75 4

-55 -163 -59 -104 108

-77 -77 -34 -43 0

-58 -43 -31 -12 -15

Net Capital Flows excl. Reserves (Financial Account Balance)

55

-103

-69

-31

Net Capital Flows plus Errors and Omissions

27

-116

-59

-32

-28 -31

-14 8

11 59

-1 47

Non-Resident Capital Inflows Private Inflows Equity Investment Direct Investment Portfolio Investment Private Creditors Commercial Banks Nonbanks Official Inflows International Financial Institutions Bilateral Creditors Resident Capital Outflows Private Outflows Equity Investment Abroad Resident Lending/Other Reserves (- = Increase)

Memo: Net Errors and Omissions Current Account Balance Source: IIF. See page 29 for guidance on how to interpret these data.

iif.com © Copyright 2016. The Institute of International Finance, Inc. All rights reserved.

2015 Est. 2016 Proj.

CAPITAL FLOWS TO EMERGING MARKETS | APRIL 7, 2016

page 27

Table 5 Latin America: Capital Flows $ billion 2013

2014

303 285 137 118 19 148 24 124 18 5 14

309 285 132 110 22 153 34 119 23 4 19

227 203 124 109 16 79 7 71 24 4 20

227 214 117 99 17 97 19 78 13 4 8

-145 -154 -60 -93 8

-139 -114 -59 -55 -25

-64 -101 -34 -66 36

-130 -138 -38 -100 9

Net Capital Flows excl. Reserves (Financial Account Balance)

150

194

127

88

Net Capital Flows plus Errors and Omissions

127

181

113

88

-22 -135

-13 -156

-14 -149

0 -97

Non-Resident Capital Inflows Private Inflows Equity Investment Direct Investment Portfolio Investment Private Creditors Commercial Banks Nonbanks Official Inflows International Financial Institutions Bilateral Creditors Resident Capital Outflows Private Outflows Equity Investment Abroad Resident Lending/Other Reserves (- = Increase)

Memo: Net Errors and Omissions Current Account Balance Source: IIF. See page 29 for guidance on how to interpret these data.

iif.com © Copyright 2016. The Institute of International Finance, Inc. All rights reserved.

2015 Est. 2016 Proj.

CAPITAL FLOWS TO EMERGING MARKETS | APRIL 7, 2016

page 28

Table 6 Africa/Middle East: Capital Flows $ billion Non-Resident Capital Inflows Private Inflows Equity Investment Direct Investment Portfolio Investment Private Creditors Commercial Banks Nonbanks Official Inflows International Financial Institutions Bilateral Creditors Resident Capital Outflows Private Outflows Equity Investment Abroad Resident Lending/Other Reserves (- = Increase) Net Capital Flows excl. Reserves (Financial Account Balance) Net Capital Flows plus Errors and Omissions

Memo: Net Errors and Omissions Current Account Balance Source: IIF. See page 29 for guidance on how to interpret these data.

iif.com © Copyright 2016. The Institute of International Finance, Inc. All rights reserved.

2013

2014

2015 Est. 2016 Proj.

89 82 55 34 21 27 0 27 7 2 5

91 87 46 34 12 41 28 13 3 1 3

89 82 45 32 13 37 24 13 7 3 5

116 104 45 35 10 60 40 19 12 6 6

-237 -148 -50 -98 -89

-166 -153 -73 -80 -13

-3 -119 -54 -65 116

36 -54 -55 1 90

-59

-62

-30

62

-103

-90

-19

62

-43 191

-27 102

11 -97

0 -152

CAPITAL FLOWS TO EMERGING MARKETS | APRIL 7, 2016

page 29

ANNEX 1: IIF CAPITAL FLOWS DATA – A LAYMAN’S GUIDE Capital flows arise through the transfer of ownership of assets from one country to another. When analyzing capital flows, we care about who buys an asset and who sells it. If a foreign investor buys an emerging market asset, we typically refer to this as a non-resident capital flow (or inflow) in our terminology. We report capital flows on a net basis. For example, if foreign investors buy $10 billion of assets in a particular country and sell $2 billion of that country’s assets during the same period, we show this as a (net) capital inflow of $8 billion. Note that non-resident capital flows can be negative, namely if foreign investors sell more assets of a country than they buy in a given period. Our “non-resident private capital flows to emerging markets” measure is the sum of all net purchases of EM assets by private foreign investors. Correspondingly, if an investor from an emerging market country buys a foreign asset, we call this a resident capital flow (or outflow). Resident capital flows can also be positive or negative. Following standard balance of payments conventions, we show a net increase in the assets of EM residents with a negative sign. For further details regarding terminology, concepts and compilation of our data, please consult our User Guide located on our website at www.iif.com/emr/global/ capflows.

IIF CAPITAL FLOWS REPORT COUNTRY SAMPLE (25)* Emerging Europe

Czech Republic

Latin America

Argentina

(6)

Hungary

(6)

Brazil

Poland

Chile

Russian Federation

Colombia

Turkey

Mexico

Ukraine

Venezuela

Emerging Asia

China

Africa/Middle East

Egypt

(7)

India

(6)

Lebanon

Indonesia

Nigeria

Malaysia

Saudi Arabia

Philippines

South Africa

South Korea

UAE

Thailand *With this issue of the IIF Capital Flows Report, we have reduced our group of EMs covered from 30 to 25 countries, focusing on the larger players in EM capital flows. Previously, IIF coverage also included Ecuador, Peru, Romania, Bulgaria, and Morocco. The change in country coverage has only a very limited impact on our EM capital flows aggregates, which have been recalculated to reflect the new coverage. iif.com © Copyright 2016. The Institute of International Finance, Inc. All rights reserved.

CAPITAL FLOWS TO EMERGING MARKETS | APRIL 7, 2016

page 30

GLOBAL MACROECONOMIC ANALYSIS

GLOBAL CAPITAL MARKETS

Charles Collyns, Managing Director & Chief Economist

Hung Tran, Executive Managing Director

Ulrik Bie, Chief Economist

Sonja Gibbs, Senior Director, GCM Department

Robin Koepke, Senior Economist

Emre Tiftik, Deputy Director

Scott Farnham, Research Analyst

Fiona Nguyen, Senior Research Analyst

Brent Harrison, Research Analyst

Khadija Mahmood, Research Analyst

Alli Schultz, Program Associate

Yufei Chai, Research Intern

Dimitrije Tasic, Research Intern

Husnu Oney, Research Intern

AFRICA/MIDDLE EAST DEPARTMENT

ASIA/PACIFIC DEPARTMENT

David Hedley, Chief Economist, Sub-Saharan Africa

J. C. Sambor, Director

Garbis Iradian, Chief Economist, MENA

Bejoy Das Gupta, Chief Economist

Giyas Gökkent, Senior Economist

Feng Guo, Chief Representative, Beijing

Amanda Preston, Research Analyst

Zhou Wang, Senior Economist Kevin Sanker, Associate Economist

EUROPEAN DEPARTMENT

LATIN AMERICA DEPARTMENT

Clay Berry, Chief Economist

Ramon Aracena, Chief Economist

Ondrej Schneider, Senior Economist

Martin Castellano, Senior Economist

Ugras Ulku, Senior Economist

Maria Paola Figueroa, Economist

Amy Kuhn, Research Analyst

Julia Smearman, Associate Economist

iif.com © Copyright 2016. The Institute of International Finance, Inc. All rights reserved.