good tactical buying

Multi-asset research

Uncommon truths

Uncommon truths Should we focus on a potential 20% S&P 500 hit or look to the future? A deepening of the Korean crisis would naturally

The model charts gold as a function of real 10-year

favour assets such as gold (in our opinion). Not

treasury yields (TIPS), 10-year inflation breakeven rates

knowing how this will unfold, we prefer to stick with a

and the trade weighted dollar index. Since 2007, gold

long-term asset allocation. Our latest work suggests

has tended to go up when any of those three

optimal portfolios should be focused on equities, real

variables goes down (prior to 2007, gold tended to go

estate and sovereign debt. If our optimism is not

up when inflation expectations increased). Gold

rewarded, history suggests a 20% decline in the S&P

moved to an unusual premium versus the post-2007

500 would represent a good buying opportunity.

model around the time of the US election in November 2016 and that premium had increased to

We have been uniquely blessed to witness great

12% by the end of that month. Today it stands at

athletes such as Usain Bolt and Mo Farah over the last

around 13% (or around $151 on a model-predicted

decade. What a shame we cannot say the same

level of $1136).

about the current crop of world leaders. The nuclear stand-off between Chairman Kim and President Trump

Gold is a natural refuge in a time of tension and crisis

has us all searching for historical comparisons.

and is up around 6% in the last month (the premium versus the model has not changed in recent months,

Such geo-political tension was always a possibility

suggesting gold is moving in line with bonds and the

when Donald Trump was elected. Though one set of

dollar). Another popular barometer of stress is the VIX

investors welcomed his presidency (driving stocks and

index, which has increased from around 9.5 a month

the dollar higher), another set was less optimistic:

ago, to a recent peak of 17.3. Interestingly, and

Figure 1 shows that gold developed a “Trump

despite the declines of recent days, the S&P 500 is

premium” during November 2016 and it still exists.

virtually unchanged over the last month.

Figure 1 – Price of gold versus model predicted values (USD per ounce) 2000 Gold

Post-2007 Model

1600 Pre-2007 Model 1200 800 400 0 2003

2005

2007

2009

2011

2013

2015

2017

Note: gold is modelled as a function of three variables: US TIPS 10-year yield; US 10-year inflation breakeven yield and trade weighted US dollar. The pre2007 model was fitted using data prior to 2007 (i.e. prior to the financial crisis). The post-2007 model uses data from 2007 onwards. Source: Datastream and Source Research. The past is no guide to future performance.

14 August 2017

For professional investors only

1-


Uncommon truths

What does history tell us about how far this market

this period of the year). If the Kim-Trump stand-off

reaction could go? The closest we came to fisticuffs

worsens, the above gives a template for what could

during the Cold War period was the Cuban Missile

happen: markets will act first and ask questions later

Crisis of October 1962. The VIX didn’t exist then and

and we would expect a sharp rise in gold and the

the price of gold was fixed, so the best market

VIX (and probably the yen and CHF), while we think

indicator was the S&P 500. Our total return index

stocks are likely to dive (especially in Asia).

suggests only a 3% loss during October 1962 but the tension had been building since late 1961 and the

However, history also suggests that the pain will be

year-to-date loss by end-October 1962 was 20% (at

relatively short-lived, especially if the protagonists

a time when the US economy was expanding and

pull back from the brink. We suspect that a 20%

the Fed was on-hold).

decline in the S&P 500 would represent an excellent buying opportunity for global stocks. Interestingly,

More recent was the tension that gripped the world

even upon the outbreak of War in 1939 using that

when Saddam Hussein’s troops invaded Kuwait on

20% rule would have paid handsomely (a 20% total-

August 2, 1990. At the time, Iraq was thought to

return loss from end-September 1939 would have

have the world’s fourth largest army and to own

been incurred by end-June 1940 and purchase at

chemical weapons (we feared a chemical attack in

that point would have been rewarded with an

London). Hence, it was anticipated that any

annualised return of 15.7% over the next five years).

conflict would be long and bloody. The VIX index increased from the 15-16 range in mid-July of that

However, we can’t base our investment strategy on

year to a daily peak of 34 on October 11. During

the assumption that Armageddon is around the

that time, the S&P 500 declined by 20%. Gold had

corner (if they do not pull back from the brink, the

gained 18% by the end of August but gave most of

last thing we would be worried about is the value of

that back by the time the S&P 500 bottomed on

our pensions!). We recently published the first in a

October 11. However, matters are complicated by

series of articles about asset allocation (see A post-

the fact that the US economy was shrinking during

1914 perspective on asset allocation) and it made

1990 H2 and that the Fed was in the middle of a

the case for a preference for equities over the long-

loosening phase.

haul. Figure 2 is the result of a continuation of that work -- a shortening of the time horizon has allowed

So, here we are again facing a summer/autumn

the inclusion of more assets (HY, REITS and the S&P

geo-political crisis (we have noted on several

600 small-cap index). The 1987 start-date was

occasions the tendency for the VIX to peak during

dictated by data constraints.

Figure 2 – The efficient frontier for US assets based on CPI-adjusted calendar year total returns from 1987 to 2016 S&P 500 10

REITS

Return (%)

8 S&P SC 600 6

HY IG

4

Govt Max return/risk

2

CTY

Gold

Cash

0 0.0

5.0

10.0

15.0

20.0

25.0

Standard deviation (%) The efficient frontier shows the maximum return achievable for each level of standard deviation. The size of the bubbles is in proportion to the average pairwise correlation of the asset concerned with all other assets. Cash and gold have a negative correlation with other assets. That for government bonds is close to zero, which is why the “Govt” bubble is not visible. “Max return/risk” is the combination that gives the highest return per unit of risk. Source: BofA ML, GPR, JP Morgan, S&P GSCI, Datastream and Source Research. The past is no guide to future performance.

14 August 2017


2-


Uncommon truths

This is a classic risk-reward chart, with average

equity-like assets when it comes to correlation and

pairwise correlation also indicated by the size of the

therefore barely features among the optimal

bubbles (the correlations for cash and gold are

allocations. Don’t spend too much time searching

negative, while that for government debt is close to

for high-yield, it is zero-weighted across the board.

zero which is why it is hard to see). The efficient frontier shows the maximum return possible for each

How one uses that information is debatable. First, it

level of risk or standard deviation. The “Max

could be argued that the period covered is not

return/risk” point represents the highest return per

relevant to the future. We would respond that it is

unit of volatility (a simplified version of the Sharpe

hard to foresee the future and a longer historical

Ratio). The asset allocation at that point is: cash

template allows us to incorporate a broad range of

63%, government debt 18%, S&P 600 10%, Gold 7%....

cycles and scenarios (the period covered here includes the 1987 crash, the Iraqi invasion of Kuwait,

Cash, government debt, IG credit, HY credit, S&P SC

the 1990s Asian crisis, the Dotcom bubble and bust,

600 and S&P 500 are pretty much aligned, with each

the invasion of Afghanistan and Iraq and the

successive increase in risk bringing more reward.

financial, Greek and Eurozone crises).

REITS break that sequence by bringing more volatility but less reward than the S&P 500.

The second problem is knowing where to position

Interestingly, gold and commodities (CTY) could be

oneself on the risk spectrum. That is a matter of

replaced by other assets (or combinations thereof)

personal choice but one thing we can conclude

offering more reward for the same risk (this was also

from Figure 3 is that most risk appetites would have

the case when using data from 1914). The low

been served by a combination of large-cap stocks,

correlations of gold and commodities with other

REITS and government debt.

assets could give them an important diversifying role but Figure 3 suggests this is not the case (the chart

In conclusion, we believe a deepening of the

shows the asset allocation that would have

Korean crisis would favour so-called “safe-havens”

maximised returns for any given level of risk)

such as gold and sovereign debt. Such crises tend to be short-lived and a 20% drop in the S&P 500

Cash and government debt also exhibit low

would represent a good-entry point (in our opinion).

correlation with other assets and they do play an

However, we do not know what will happen and

important role in the optimal allocations (because

prefer to focus on our longer-term strategy. The

of their acceptable risk-reward characteristics). On

above analysis suggests this should be dominated

the other hand, IG credit, which is like government

by large-cap stocks, real estate and sovereign debt.

debt in terms of the risk-reward, is more akin to

Allocation

Figure 3 – Optimal allocations along the efficient frontier (based on US assets from 1987 to 2016) 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

Standard deviation of returns (%) S&P 500

REITS

Govt

S&P SC 600

CTY

Gold

IG

Cash

HY

This is the result of an unconstrained optimisation process based on CPI-adjusted calendar year US asset returns from 1987 to 2016. Source: BofA ML, GPR, JP Morgan, S&P GSCI, Datastream and Source Research. The past is no guide to future performance.

14 August 2017


3-


Uncommon truths

Figure 4 – Asset class total returns Data as at 11/08/2017

Current Level/RY

1w

Equities World Emerging Markets US Europe Europe ex-UK UK Japan Government Bonds World Emerging Markets US (10y) Europe Europe ex-UK (EMU, 10y) UK (10y) Japan (10y) IG Corporate Bonds Global US Europe UK Japan HY Corporate Bonds Global US Europe Cash (Overnight LIBOR) US Euro Area UK Japan Real Estate (REITs) Global Emerging Markets US Europe ex-UK UK Japan Commodities All Energy Industrial Metals Precious Metals Agricultural Goods Currencies (vs USD)* EUR JPY GBP CHF CNY

471 1043 2324 1687 1999 1149 3163

-1.5 -2.2 -1.4 -2.2 -2.1 -2.6 0.6

1.5 2.8 0.8 1.6 1.6 1.6 3.9

1.6 3.8 0.9 1.7 2.1 0.7 3.0

BofA-ML JPM Datastream Bofa-ML Datastream Datastream Datastream

0.95 6.53 2.19 0.64 0.31 1.11 0.06

0.9 -0.5 0.7 0.7 1.3 0.6 1.7

3.6 3.0 1.9 4.8 5.6 3.3 5.3

BofA-ML BofA-ML BofA-ML BofA-ML BofA-ML

2.51 3.15 0.84 2.61 0.28

0.2 0.0 0.5 0.0 1.6

BofA-ML BofA-ML BofA-ML

5.76 6.17 3.20

Total Return (USD, %) 1m QTD YTD

12m

Total Return (Local Currency, %) 1w 1m QTD YTD 12m

13.6 23.1 10.5 17.9 20.8 10.8 13.4

14.8 18.0 13.9 17.1 20.0 9.9 14.5

-1.6 -1.7 -1.4 -2.4 -2.5 -2.1 -0.9

0.4 2.0 0.8 -0.6 -1.1 0.6 -0.9

0.8 3.5 0.9 -0.2 -0.6 0.8 0.0

10.2 19.0 10.5 8.3 9.3 5.5 6.1

14.6 18.2 13.9 13.4 14.9 9.8 23.4

2.6 2.6 1.4 4.0 5.0 2.1 3.2

6.7 13.4 3.2 11.0 11.6 8.5 7.0

-3.0 6.4 -3.1 0.8 2.8 -1.9 -8.1

0.4 0.0 0.7 0.4 1.0 1.0 0.1

1.2 1.3 1.9 1.4 2.3 2.3 0.4

0.8 1.0 1.4 0.6 1.6 2.2 0.3

1.2 7.4 3.2 -0.8 -0.2 3.3 0.2

-2.3 7.3 -3.1 -4.4 -2.5 -2.0 -1.0

2.1 1.1 4.6 0.0 5.2

2.0 1.0 4.7 0.0 3.1

7.4 4.9 14.0 8.1 7.3

2.9 2.2 6.1 -1.2 -7.0

0.1 0.0 0.2 0.0 0.1

1.1 1.1 1.3 0.0 0.3

1.0 1.0 1.3 0.0 0.2

4.0 4.9 1.9 2.8 0.4

1.7 2.2 0.6 -1.4 0.2

-0.5 -0.8 0.0

1.3 0.6 4.5

1.2 0.4 4.4

7.6 5.3 17.5

9.7 9.0 13.3

-0.6 -0.8 -0.4

0.8 0.6 1.2

0.6 0.4 1.0

5.5 5.3 5.1

8.8 9.0 7.4

1.18 -0.43 0.22 -0.04

0.0 0.4 -0.2 1.4

0.1 3.0 1.3 4.4

0.1 3.4 -0.1 2.9

0.6 12.1 5.6 7.0

0.7 5.7 0.7 -6.6

0.0 0.0 0.0 0.0

0.1 0.0 0.0 0.0

0.1 0.0 0.0 0.0

0.6 -0.3 0.1 0.0

0.7 -0.4 0.2 0.0

FTSE FTSE FTSE FTSE FTSE FTSE

1800 2125 2871 3345 1395 2507

-1.8 -2.6 -2.2 -1.9 -2.2 0.6

2.8 4.1 1.2 4.2 2.3 4.7

0.8 5.3 -1.5 3.0 -0.2 0.3

8.1 33.1 0.0 20.1 9.2 -0.4

-0.1 17.7 -5.9 4.2 5.5 -6.2

-2.1 -2.9 -2.2 -2.3 -1.8 -1.0

-0.4 0.8 1.2 0.9 1.3 -0.2

-2.5 1.8 -1.5 -0.3 -0.1 -2.6

-3.3 19.1 0.0 7.4 4.0 -6.8

-5.3 11.6 -5.9 -1.2 5.4 1.0

GSCI GSCI GSCI GSCI GSCI

2233 375 1293 1559 396

-0.7 -0.8 2.9 2.6 -2.5

3.4 8.0 7.4 6.3 -9.9

2.9 6.3 6.6 3.6 -6.0

-7.6 -13.7 15.2 10.7 -8.1

1.3 1.5 24.2 -6.5 -11.0

-

-

-

-

-

1.18 109.19 1.30 1.04 6.66

0.4 1.4 -0.5 1.1 1.0

3.1 4.4 1.1 0.2 2.1

3.5 2.9 -0.1 -0.4 1.7

12.4 7.1 5.0 5.9 4.2

6.1 -6.6 0.1 1.4 -0.6

-

-

-

-

-

Index MSCI MSCI MSCI MSCI MSCI MSCI MSCI

Source: MSCI, Datastream and Source Research. Notes: *The currency section is organised so that in all cases the numbers show the movement in the mentioned currency versus USD (+ve indicates appreciation, -ve indicates depreciation).

14 August 2017


4-


Uncommon truths

Figure 5 – Equity sector total returns relative to local market (%) Data as at 11/08/2017

US

Europe

1w

1m

QTD

YTD

12m

1w

1m

QTD

YTD

12m

Oil & Gas

-1.2

-1.7

-2.3

-21.9

-16.2

0.3

2.8

2.0

-13.3

-3.9

Materials

-0.7

-2.4

-1.5

-1.6

-2.1

0.3

0.2

1.2

-1.8

5.0

-0.5

4.3

6.7

-7.4

-14.6

0.5

3.0

6.8

-1.1

14.8

Basic Resources Chemicals

-0.6

-2.5

-1.6

0.4

0.8

0.3

-1.5

-2.2

-2.4

-0.6

-0.2

-2.1

-1.5

-1.3

2.1

0.2

-1.7

-1.3

2.1

2.2

-1.6

-7.7

-7.9

-10.4

-14.4

0.1

-0.7

-1.0

0.5

0.4

0.0

-1.4

-0.7

-0.6

1.6

0.3

-2.0

-1.3

2.6

2.7

Consumer Discretionary

0.1

0.0

-0.7

0.8

-2.3

0.2

1.0

0.6

-2.6

-2.3

Automobiles & Parts

Industrials Construction & Materials Industrial Goods & Services

-0.1

-3.9

-2.1

-6.8

-5.0

1.2

-2.9

0.1

-5.1

-0.7

Media

1.3

2.2

1.5

-0.6

2.3

-0.2

0.4

-2.6

-7.8

-10.8

Retail

-0.1

1.4

-0.3

-1.4

-7.4

0.2

2.0

0.7

-7.8

-9.9

Travel & Leisure

-0.4

-5.0

-4.7

2.9

11.8

-0.7

-0.5

-3.3

0.1

-1.1

Consumer Staples

1.5

1.1

-0.8

-2.0

-9.1

0.3

0.3

-1.2

-0.1

-9.2 -8.2

Food & Beverages

1.7

2.4

0.4

-3.5

-10.5

0.6

2.7

1.0

2.8

Personal & Household Goods

1.0

-1.6

-2.8

1.9

-7.0

-0.3

-0.2

-1.1

4.0

-2.0

Healthcare

0.2

-1.2

-1.7

4.3

-6.4

1.1

-2.0

-4.1

-2.2

-11.4

Financials

-1.3

-1.3

-0.7

-3.0

13.6

-1.1

-0.1

1.9

3.6

15.5

Banks

-1.9

-2.9

-2.1

-5.4

21.9

-1.4

-0.7

2.0

4.8

24.2

Financial Services

-0.8

-0.1

0.0

-3.8

8.8

-0.7

-0.1

0.0

8.3

8.3

Insurance

-0.8

0.3

0.7

0.5

9.6

-0.9

0.8

3.0

1.8

12.1

Real Estate

-13.3

-0.6

1.0

-1.7

-4.3

-13.7

0.2

1.6

-0.3

-3.0

Technology

0.5

1.8

3.4

10.9

11.5

-0.4

0.2

1.0

7.0

3.8

Telecommunications

0.0

5.8

3.4

-15.6

-18.9

-0.6

3.1

2.0

-3.3

-10.7

Utilities

1.2

3.7

2.4

1.8

-4.1

1.7

4.1

3.6

4.3

-5.9

We use a sector classification created by merging the two main systems used by Standard & Poors (S&P) for the US and Stoxx for Europe. We have decided to classify our 10 top level industries using categories that most closely resemble the Global Industry Classification Standard (GICS) and at the level below that (super sectors) we are using the Industry Classification Benchmark (ICB). The former is used for the S&P 500 index and the latter for the Stoxx 600, our benchmark indices. The two systems overlap in most cases and the only material difference seems to be in the consumer sectors. Therefore, we define consumer staples as the aggregate of personal & household goods and food & beverage, while consumer disc retionary includes automobiles & parts, media, retail and travel & leisure. For the rest, we assume 100% overlap for the corresponding top level sectors. Source: Datastream and Source Research

14 August 2017


5-


Uncommon truths

Figure 6 – Source Multi-Asset Portfolio

Cash Cash Gold Bonds Government US Europe ex-UK (Eurozone) UK Japan Emerging Markets Corporate IG US Dollar Euro Sterling Japanese Yen Corporate HY US Dollar Euro Equities US Europe ex-UK UK Japan Emerging Markets Real Estate US Europe ex-UK UK Japan Emerging Markets Commodities Energy Industrial Metals Precious Metals Agriculture Total

Neutral Policy Range 5% 0-10% 2.5% 2.5% 45% 10-80% 30% 10-50% 10% 8% 2% 8% 2% 10% 0-20% 5% 3% 1% 1% 5% 0-10% 4% 1% 45% 20-70% 25% 7% 4% 4% 5% 3% 0-6% 1% 1% 0.5% 0.5% 0% 2% 0-4% 1% 0.3% 0.3% 0.3% 100%

Currency Exposure (including effect of hedging) USD 49% EUR 21% GBP 8% JPY 14% EM 7% Total 100%

Allocation Position vs Neutral 10% 10% 0% ↓ 39% ↓ 19% ↓ 12% 2% 2% 0% ↓ 3% 20% 10% 6% 2% 2% ↓ 0% ↓ 0% ↓ 0% ↑ 45% ↑ 10% ↑ 14% 6% ↑ 8% 7% 6% 2% 2% 0% 0% 2% 0% 0% 0% 0% 0% 100%

Hedged Currency

↓ ↑

38% 27% 11% ↑ 11% ↓ 13% 100%

Source: Source Research Note: This is a simulated portfolio (See the latest The Big Picture for more details). Arrows indicate latest changes in allocation.

14 August 2017


6-


Uncommon truths

Figure 7 – Source sector strategy US

Europe

Neutral Source

Neutral Source

Region

Oil & Gas

5.8% Neutral

5.3% Neutral

Europe

Materials

2.2% Neutral

7.0% Underweight

US

0.4% Underweight

2.8% Underweight

US

4.2% Underweight

US

Basic Resources Chemicals Industrials Construction & Materials Industrial Goods & Services

↑

Preferred

1.8% Neutral 11.8% Overweight

14.2% Neutral

0.5% Overweight

US

2.8% Overweight

US

11.3% Overweight

11.4% Neutral

15.0% Neutral

11.0% Overweight

↑

Europe

Automobiles & Parts

0.8% Neutral

3.2% Overweight

↑

Europe

Media

3.2% Neutral

2.4% Underweight

↓

US

Retail

8.1% Neutral

3.5% Overweight

Travel & Leisure

3.0% Overweight

1.9% Overweight

Consumer Discretionary

Consumer Staples

9.6% Neutral

US

Europe ↑

US

16.8% Underweight

US

Food & Beverage

4.1% Overweight

7.3% Overweight

US

Personal & Household Goods

5.5% Neutral

9.5% Underweight

US

Healthcare

13.3% Overweight

11.0% Neutral

US

Financials

18.5% Neutral

21.9% Neutral

Europe

Banks

6.4% Neutral

12.7% Neutral

Europe

Financial Services

5.5% Underweight

Insurance

3.7% Underweight

↓

5.5% Overweight

↑

Europe

Real Estate

2.9% Overweight

↑

1.7% Neutral

↑

US

Technology

18.6% Neutral

2.0% Overweight

Europe

3.9% Underweight

Telecommunications

2.1% Overweight

4.3% Underweight

Utilities

3.1% Underweight

4.5% Underweight

US ↓

US US

Source: Datastream and Source Research. Notes: See the latest Source Sector Selector for more details. Arrows indicate latest changes in recommendations.

14 August 2017


7-


Uncommon truths

Authors Paul Jackson Head of Research T. +44 (0)20 3370 1172 E. [email protected] András Vig Director T. +44 (0)20 3370 1152 E. [email protected]

Important information Investors in Source products should note that the value of your investment may go down as well as up. As a result you may not get back the amount of capital you invest. This document is for discussion purposes only and is intended for professional investors pursuant to Directive 2004/39/EC (MIFID) Annex II Section I. Without limitation this document does not constitute an offer or a recommendation to enter into any transaction. The calculations and charts set out herein are indicative only, make certain assumptions and no guarantee is given that future performance or results will reflect the information herein. Past performance is not a guarantee of future performance. Simulated performance is not necessarily indicative of future performance. Simulated performance may have many inherent limitations. Performance may be volatile, and an investor could lose all or a substantial portion of his or her investment. When making an investment decision, you should rely solely on the final documentation and any prospectus relating to the transaction and not this information document. Investment strategies involve numerous risks. The directors of Source UK Services Limited and Source Investment Management Limited (collectively and separately “Source”) do not guarantee the accuracy and/or the completeness of any data included herein and Source shall have no liability for any errors, omissions, or interruptions herein. Source makes no warranty, express or implied, as to the information described herein. All data and performance shown is historical unless otherwise indicated. Investors should consult their own business, tax, legal and accounting advisors with respect to this proposed transaction and they should refrain from entering into a transaction unless they have fully u nderstood the associated risks and have independently determined that the transaction is appropriate for them. In no way should Source be deemed to be holding itself out as a financial adviser or a fiduciary of the recipient hereof and this document is not intended to be "investment research" as defined in the Handbook of the UK Financial Conduct Authority. Source, Source’s shareholders, or employees of Source or its shareholders may from time to time have long or short positions in securities, warrants, futures, options, derivatives or financial instruments referred to in this material. As a result, investors should be aware that Source may have a conflict of interest that could affect the objectivity of this report. Investors should consider this report as only a single factor in making their investment decision. This document is only intended for and will be only distributed to persons resident in jurisdictions where such distribution or availability would not be contrary to local laws or regulations. This document is provided by Source UK Services Limited, 110 Cannon Street, London EC4N 6EU, authorised and regulated by the Financial Conduct Authority. © 2017 Source UK Services Limited. All rights reserved.

14 August 2017


8-