Emerging market debt: The dominant US Dollar by Paul McNamara, Investment Director, and Mike Biggs, Investment Manager April 2015

Summary • We expect global growth to strengthen, led by developed economies, and global headline inflation to fall on declining commodity prices.

Both trends should benefit local currency bonds. • The main risk to the asset class is further US dollar appreciation. However, we expect the dollar to remain stable from here. In any

case, it is a risk that can be hedged. • If the dollar stabilises, we expect returns of 5–10% in 2015 on the index level. We prefer exposure to the manufacturing and services-

producing economies with moderate to high inflation, but are still cautious on the outlook for commodity currencies.

Introduction The JPMorgan GBI-EM Global Diversified index of local currency debt declined 4.0% in the first quarter of 2015 (unhedged, in USD terms), with the positive returns from carry and rising bond prices more than offset by the 6.3% decline in emerging market (EM) currencies. In our view, the currency weakness was largely due to the US dollar’s strength – the USD rallied by 8.7% against a basket of developed market currencies 1. We expect global growth and inflation trends to remain supportive for the asset class. The euro area recovery has strengthened, and we expect US growth to rebound in the second quarter after a soft first quarter. Growth continues to weaken in China, but we do not expect the slowdown to be sufficient to derail the global expansion. Weak demand growth in China has, however, been sufficient to weigh on commodity prices and push China into deflation. We expect these trends to continue, and for global inflation to remain muted, despite stronger demand from the developed economies. Chart 1: Dominant US dollar -3.0 -2.5 -2.0 -1.5

JPMorgan ELMI CRB commodity price index EM Equities (MXEF) USD DXY (rhs)

% yoy

-1.0

4 3 2 1

-0.5 0.0

0

0.5

-1

1.0 1.5 2.0 2000 2001 2003 2005 2007 2008 2010 2012 2014

-2 -3

Source: Bloomberg, Haver; as of 16/04/2015

An environment of rising growth and modest inflation should be ideal for EM local currency rates; the biggest risk remains

1

This is based on the US Dollar Index DXY, an index that captures changes in the USD against a basket of developed market currencies (EUR, JPY, GBP, CAD, SEK and CHF).

.

further USD strength. Despite the events of the past three quarters, we believe that the role of the USD in asset price moves is still underappreciated. Chart 1 shows normalised changes in EM FX, EM equities, commodity prices and the USD. When the USD is strong, these asset classes tend to perform poorly in USD terms. In our view, there is a danger that investors start to look for fundamental indicators to explain EM local currency performance, when in reality the cause of the weakness is largely USD strength. These sharp moves in the USD have dwarfed the impact of other fundamental changes on EM assets. If the USD continues to appreciate, then returns for EM local currency bonds measured in USD terms will likely be poor. For those convinced of further USD strength, the risk can be hedged by being long USD against other developed market economies. The GBI-EM is up 6.2% in DXY terms year-to-date, after rising 6.3% in 2014. The outlook for the USD is obviously uncertain, and there has been little evidence so far to suggest that the USD rally is over. However, we believe there is a good chance that the USD stabilises at current levels. The move to date has been fast and has gone beyond what one might expect from interest rate moves. Furthermore, the growth outlook for the rest of the developed world (and in particular the euro area) has improved relative to the US. If we are correct, we expect stability in the USD to trigger improved performance of the asset class. We expect the GBI-EM to return 5% to (10% by the end of 2015. The outlook is by no means positive for the whole of EM. While the manufacturers and services producers should benefit from stronger DM demand growth (which translates into demand for EM exports) and lower commodity prices (which help contain inflation), EM commodity producers should continue to struggle. As a result, we are positive on the manufacturing and servicesproducing economies with moderate to high inflation, such as India, Mexico and the Philippines, but cautious on the currencies of the commodity producers, such as Colombia, Peru and South Africa. We expect growth for the commodity producers to remain weak, which could justify lower policy rates in future. As a result, we are positive on the bonds in Brazil and Russia. Outside these themes, we would draw attention to our short in the Turkish lira. The global macroeconomic trends should be supportive for Turkey, but we are concerned about the domestic

Emerging market debt: The dominant US Dollar

policy settings. Credit growth has picked up, demand has increased, and the current account balance has deteriorated as a result. The financing pressures on Turkey have grown, and a tightening in global liquidity conditions could put pressure on the currency.

Global macroeconomic environment Our global growth outlook is unchanged – we expect stronger growth in DM driven by the continued expansion in the euro area and a rebound after softer data in the US and Japan. Growth in China is expected to slow further, and weak demand growth could weigh on commodity prices and global inflation. Global growth We have discussed these views at length in previous publications, and as a result we will be brief. In the US, a mildly positive credit impulse and the end to fiscal austerity should support demand growth. USD strength, unfavourable weather and the uncertain impact of lower oil prices might have contributed to weak growth in the first quarter, but we expect a rebound in activity in the second quarter. The main downside risk to our view would be an unanticipated tightening in credit conditions, and we will look to the second-quarter Senior Loan Officers Survey (due in early May) for evidence of weakness we might have missed. In our January update, we argued that US growth numbers were likely to disappoint – given the shift in expectations, we now expect them to surprise positively.

% qoq

Euro area

1.2

0.5

0.6

0.0

0.0

-2.0 2001

Chart 3: China moves into deflation Real GDP

30

China, % yoy

Nominal GDP

25

GDP Deflator

20 15 10

-5 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 Source: Haver; as of 16/04/2015

-0.5

-1.5

Global inflation In our view, China today is not dissimilar to Japan in the 1990s. As in Japan, falling credit growth has put downward pressure on demand, which in turn has put downward pressure on prices. In the first quarter 2015, GDP inflation in China turned negative for the first time since 2009 (chart 3).

0

% qoq

1.0

-1.0

Chinese policymakers are faced with the choice between 1) a slowdown in credit growth and the consequent negative credit impulse on the one hand, and 2) an aggressive stimulus and the consequent surge in the debt ratio on the other. They seem to have chosen the former, which seems prudent, but means that domestic demand growth will remain weak. While we do not expect growth in China to be soft enough to derail the global recovery, it is likely to keep inflation pressures subdued.

5

Chart 2: Euro area consumer demand surging 1.5

China needs credit growth to rise, but this would cause the debt ratio to rise even faster.

-0.6

Retail sales (lhs) Consumption (rhs)

-1.2

Weak growth in China is also putting downward pressure on commodity prices. Changes in global metals prices have been well correlated with Chinese industrial production (IP) growth, and IP growth below 10% has generally been associated with falling prices (chart 4).

-1.8 2004

2007

2010

2013

Source: Haver; as of 16/04/2015

The euro area remains our favourite structural bull case – credit growth is rising from negative levels, and the positive credit impulse is boosting demand growth. There is also abundant evidence that lower oil prices are boosting consumer spending. As chart 2 shows, real consumer spending growth could be close to 1.0% quarter-on-quarter in the first quarter. The main headwind for the euro area is that expectations for eurozone growth have been revised up sharply, and it will be difficult for the region to outperform expectations by the extent it did in the first quarter. In China, credit growth continues to slow, and the negative impulse is proving a drag on domestic demand growth. Real GDP growth fell to 7% in the first quarter, but nominal GDP growth fell to 6.2% from levels of nearly 20% in 2011 (chart 3). The problem for China is that even though credit growth has declined since 2010, it is still above nominal GDP growth and the debt ratio continues to rise. To reverse the growth slowdown, Past performance is not indicative of future performance.

Chart 4: China industrial production and commodity prices 35

200

IP, 3m/3m

30

150

Global metal prices

25

100

20 15

50

10

0

5

-50

0

-100

-5 -10 2008

China IP of 10% = flat metal prices 2009

2010

2011

2012

2013

2014

-150 2015

Source: Haver; as of 16/04/2015

This is important for global inflation. As chart 5 shows, commodity price moves are closely correlated with CPI inflation even in the OECD economies.

Emerging market debt: The dominant US Dollar

Chart 5: Commodity prices and OECD CPI inflation 70

% yoy

% yoy

Chart 7: Latin America vs. Asia – GDP growth 5

12

4

50 30

4 2

-10

-50

1

OECD CPI (rhs)

-70 2004

2006

0 -1

2008

2010

2012

Latam

6

2 Commodity prices (lhs)

Asia

8

3

10

-30

GDP growth, % yoy

10

0 -2 -4 2004

2006

2008

2010

2012

2014

2014

Source: Haver; as of 16/04/2015

Source: Haver; as of 16/04/2015

In short, weak demand growth in China is putting downward pressure on global inflation. The pick-up in demand in places like the euro area might put some upward pressure on prices and lower the risks of outright deflation, but in general we expect inflation pressures to remain muted in the coming quarters.

The divergence is also evident in the trade and inflation numbers. Trade balances have on average been deteriorating in Latam since 2012, but they have been improving in Asia since mid-2013 (chart 8). In Latam these deteriorating trade balances have caused currencies to weaken and inflation to increase, whereas in emerging Asia (excluding China) the softening in commodity prices has caused headline inflation to fall (chart 9).

EM macro outlook: manufacturers vs. commodity producers As we have argued many times before, the global recovery is good for EM. EM exports are as correlated with DM demand as they have ever been, and if DM demand strengthens as we anticipate, we should see EM export volume growth rise in 2015.

Volumes, % yoy

OECD forecast

5

25 20

15

3

10

EM monthly trade balances, USD bn

1

-1

-2

Latam

-3

Asia

5

1

0 -1 -3

2

0

Chart 6: DM demand and EM exports 7

Chart 8: Latin America vs. Asia – trade balances

-5 OECD domestic demand

EM exports (rhs) -5 1998 2000 2002 2004 2006 2008 2010 2012 2014

-10

-4 2001

2003

2005

2007

2009

2011

2013

2015

Source: Haver; as of 16/04/2015

-15 -20

Chart 9: Latin Amercia vs. Asia – inflation 10

Source: Haver; as of 16/04/2015

At the same time, not all areas of EM are likely to benefit equally. We expect manufacturers and services producers that export to DM to benefit from these macro trends, whereas commodity exporters are likely to struggle. In our view, this has been happening for some time, and is most evident if we compare developments in Latin America (Latam), made up mostly of commodity exporters, with Asia, made up mostly of manufacturers. Growth in Latam and Asia has been roughly similar for most of the past decade, but in 2014 GDP growth in our sample of Latam economies fell below 2%, whereas growth in our sample of Asian economies strengthened to above 5%. (The spikes in Asia in 2011 and 2012 were due to the floods in Thailand.)

% yoy

EM average inflation

9 8 7

Asia

6

Latam

5 4

3 2 1 0 2007

2009

2011

2013

2015

Source: Haver; as of 16/04/2015

These developments are also evident in asset prices. Local currency bonds in the GBI-EM have declined 13.8% in Latam over the past 12 months, but only lost 0.6% in Asia. Our macro numbers in the charts above include India, whereas these Past performance is not indicative of future performance.

Emerging market debt: The dominant US Dollar

return numbers are from the GBI-EM and do not. India gained 12.5% over the past year, which pushes the Asia returns well into positive territory. The outlook for local currency rates – USD dominance A macroeconomic environment of strengthening global growth and falling inflation should be good for risky assets in general and EM local currency bonds in particular. But despite this, the JPMorgan GBI-EM Global Diversified index is down year-to-date, and has fallen more than 10% since the middle of last year. Chart 10: EM FX and the USD -20 -15 -10

ELMI

% yoy

25 20

DXY (rhs)

15

If one is convinced of further US economic strength, then pressures from the USD can be hedged by being short EUR/USD. In EUR terms, the GBI-EM returned 7.3% in 2014 and is up just less than 10% year-to-date. Chart 12: EUR/USD and interest rates 1.5

US EUR 2y real rate differential

1.05

1.0

EURUSD

1.10 1.15

0.5

1.20

0.0

1.25

-0.5

1.30 1.35

-1.0

1.40

-5

10

-1.5

0

5

5

0

-2.0 2010

10

-5

Source: Bloomberg, Haver; as of 16/04/2015

15

-10

20

-15

25 2001 2002 2004 2005 2007 2008 2010 2012 2013

-20

Source: Bloomberg, Haver; as of 16/04/2015

In our view, this is largely due to movements in the USD. To demonstrate how developments in the USD dominate asset prices, chart 10 shows the correlation between the USD basket DXY and the JPMorgan ELMI index. Chart 11: Commodity prices and the USD -60 CRB Index -40

% yoy

DXY (rhs)

-20

0 20 40

60 2001 2002 2004 2005 2007 2008 2010 2012 2013

25 20

15 10 5

Past performance is not indicative of future performance.

2013

2014

2015

Chart 13: Economic surprises in the US and the euro area 100

Economic surpise index - euro area

80

-5

40

-10

20

-15

0

-20

-20

A range of other asset classes are heavily affected by movements in the USD, and the sharpness of the USD moves has dwarfed other changes in fundamentals. In our view, EM yields look attractive at current levels, and the fundamentals for many EM economies have improved. The weakness in the asset class is in large part due to the strength in the USD, and if the USD appreciates further from here, then returns to the asset class measured in USD terms will be under pressure once again.

2012

Second, economic surprises have moved against the USD. In 2014, positive economic surprises in the US and negative surprises in the euro area justified stronger capital inflows into the US and a stronger USD. In 2015, however, the growth surprises have reversed (chart 13).

60

USD moves have had an equally strong impact on commodity prices (chart 11), which in turn have been strongly correlated with EM equities.

1.50 2011

If the USD stabilises, however, better USD returns for the asset class could be just around the corner. And we think there are some excellent arguments in favour of stabilisation. First, movements in EUR/USD were closely correlated with the real 2year interest rate differential between the euro area and the US. As chart 12 shows, however, the exchange rate has moved to levels well beyond those suggested by the interest rate differential. The recent move in DXY and EUR/USD was the sharpest move over three quarters we have seen this century.

0

Source: Bloomberg, Haver; as of 16/04/2015

1.45

Economic surpise index - US

-40 -60 -80 -100 01/13

05/13

09/13

01/14

05/14

09/14

01/15

Source: Bloomberg, Haver; as of 16/04/2015

Over time the difference in economic surprises between the regions has been reasonably well correlated with changes in the EUR/USD exchange rate. In 2015, however, relative economic surprises have favoured the EUR, but it is the USD that has appreciated.

Emerging market debt: The dominant US Dollar

Chart 14: Economic surprises and EUR/USD 200

20

150

15

100

10

50

5

0

0

-50

-5

-100

-10

-150

-15

-200

Economic surprises (Euro - US)

-20

-250

EUR, % m/m-3

-25

-300 2008

-30 2009 2010 2011 2012

2013

2014 2015

Source: Bloomberg, Haver; as of 16/04/2015

In our view, this breakdown in the relationship is similar to what happened in the first half of 2010, when the situation in Greece started to deteriorate. Drivers of the current breakdown could be Greece once again, the introduction of QE, or, in January, the SNB’s decision to abandon the currency floor of the Swiss franc. These factors can continue to play a role, and given the uncertainty surrounding Greece in particular, we know there are risks to our EUR/USD forecast. In our view, however, the relationship shown in chart 14 could re-assert itself, and we believe EUR/USD could easily stabilise at current levels. If it were to do so, we believe this would trigger stronger returns for local currency rates. Opportunities within EM Given our view of strengthening global growth, falling global inflation and a stable USD, we are positive on the outlook for local currency rates. If the USD stabilises at current rates and US yields rise by 50 bps, we expect the asset class to achieve returns of 5–10%. The best opportunities, however, might be within the asset class, and could stem from the divergence between the manufacturers and the commodity producers outlined earlier. To summarise these themes and how we are positioned to benefit from them: 1.

We are long the currencies and rates of the manufacturing and services producers with moderate to high inflation. We are long Mexican rates, and long the Indian rupee, Philippine peso and Polish zloty.

2.

We are long the bonds of high-yielding commodity producers with favourable valuations. From this perspective we are long Brazilian bonds and internationally settled rouble-denominated bonds.

3.

4.

We are short the currencies of commodity producers. We are short the Brazilian real, Colombian peso, South African rand and Peruvian nuevo sol. We are long the hard currency debt of the CEE3 economies. The central European manufacturers should benefit from the DM rebound, but deflation poses a risk to their currencies. As a result, we choose to be long the hard currency debt of Hungary, Poland and Romania.

It is worth spending a moment on one position that does not fit with our themes – namely a substantial short on the Turkish lira. As a high-inflation manufacturer with the euro area as a major trading partner, Turkey should be benefiting from the DM recovery, the fall in the oil price and the decline in global Past performance is not indicative of future performance.

inflation. However, despite the favourable macro environment, we are concerned about the domestic policy setting and the build-up in foreign debt. Starting with policy, the Central Bank of the Republic of Turkey (CBRT) tightened interest rates aggressively in January 2014. Credit growth declined (to 10% annualised using the CBRT’s preferred 13-week growth rate), the credit impulse turned negative, demand growth slowed and the current account deficit narrowed. Chart 15: Turkey – new borrowing and the current account 25 % % of GDP

% of GDP

Turkey

-15

20 -10

15 10

-5 5

0

0 New borrowing

-5

CA balance (inverted, rhs) -10 2000 2002 2004 2006 2008 2010 2012 2014

5

Source: Haver; as of 16/04/2015

Thereafter policy rates started to ease, but in our opinion they started to ease too early and by too much. Credit growth has increased gradually, and by our estimates now stands at an annualised rate of 27% using the 13-week measure. As new borrowing has increased, the credit impulse has turned positive, real private sector demand growth has recovered (to 3.0% yearon-year in the fourth quarter) and the current account deficit has widened (despite the fall in the oil price). The ability of the CBRT to keep rates tight in this environment appears to be compromised by political interference. In previous research the CBRT estimated that credit growth of 15% might be appropriate for Turkey. If its estimate is correct, current rates of growth are excessive and should put further pressure on the balance of payments. In the end, this is likely to lead to further weakness in the Turkish lira. Turkey’s ability to tighten rates in the face of these pressures appears compromised. Chart 16: Rising external debt 60 55 50 45

40 35

30 Turkey: external debt as a % of GDP

25

20 1992

1995

1998

2001

2004

2007

2010

2013

Source: Haver; as of 16/04/2015

The problem is compounded by the debt inflows used to finance Turkey’s borrowing. Turkey’s external debt has increased from

Emerging market debt: The dominant US Dollar

39% at the end of 2011 to 50% by the end of 2014, and the bulk of this debt is foreign currency. Nearly 60% of it is in USD. If demand in Turkey remains strong, the current account deficit will stay wide and external borrowing needs will be high. If demand collapses, the borrowing needs will fall, but Turkey will be a less attractive lending destination. Either way, the currency could remain under pressure.

Conclusions At present, the portfolio has a modest amount of risk, is moderately long EM rates but nearly flat EM currencies. The long rates position reflects our view that global inflation pressures are subdued. Gross EM FX exposures are at average levels, and the net flat FX position reflects our preference for manufacturers and our concerns about commodity producers. This position also seems appropriate given the uncertainty around the USD. If the USD does stabilise at these levels, then commodity prices could firm as well. Under these circumstances we would be looking to cut our EM shorts (for example our Brazilian real hedge) and become more aggressively positioned in aggregate. In the event that the USD seems more likely to strengthen further, we might hedge these risks by being short EUR.

Source: GAM unless otherwise stated. The views are those of the Manager as of publication date and may not reflect the views anytime thereafter. These views are aimed to help readers in understanding the Manager's investment process and should not be construed as investment advice. Holdings and allocations are subject to change. Past performance is not indicative of future performance. There is no guarantee that market forecasts will be realised. This material is for information only and has been produced solely for the use of the person to whom it is given or sent. It may not be used for any other purpose and may not be reproduced, copied, given, distributed or disclosed, in whole or in part, to any other person. It is aimed at sophisticated, professional, eligible, institutional and/or qualified investors who have the knowledge and financial sophistication to understand and bear the risks associated with the investments described herein. It is not an invitation to subscribe to any GAM product or service and subscriptions for shares or units will only be received and shares or units will only be issued on the basis of the current offering document for the relevant GAM product. This document is not available for distribution in any jurisdiction where such distribution would be prohibited. Nothing contained herein constitutes investment, legal, tax or other advice nor is to be relied on in making an investment or other decision. Nothing in this document should be construed as a solicitation, offer or recommendation to acquire or dispose of any investment or to engage in any other transaction. Some of the statements and data included in the document may have been sourced from or based on information provided by external parties. Whilst the managers believe it to be accurate at the time of writing, GAM has not independently verified such information and no assurance can be given as to whether such information is accurate, true or complete and GAM makes no warranty, expressed or implied, regarding such information. In the United Kingdom, this document has been issued and approved by GAM London Limited, 20 King Street, London, SW1Y 6QY, authorised and regulated by the Financial Conduct Authority. In Japan, this document is restricted to professional, institutional and/or qualified investors. In Hong Kong, this document is restricted to professional investors (as defined in the Securities and Futures Ordinance (Cap 571)) only. In Singapore, this material is limited to institutional investors (as defined in the Securities and Futures Act (Cap.289)) (‘SFA’) only and does not constitute to an offer to subscribe for shares in any of the funds mentioned herein. In other countries in Asia Pacific, this document should only be distributed in accordance with the applicable laws in the relevant jurisdiction. This document may not be used as sales literature for the public in the US. Please note GAM London Limited is not licensed in Israel under the Investment Advice Law and this material is only made available to Eligible Clients. Please consult your financial professional for more information on available GAM products. Reference to a specific security does not constitute a recommendation to buy or sell that security.

Past performance is not indicative of future performance.