Credit Suisse Research Institute

February 2017

Global Investment Returns Yearbook

2017 - Slide Deck
Elroy Dimson, Paul Marsh and Mike Staunton | London Business School
Richard Kersley, Michael O'Sullivan | Credit Suisse
From past to present: the evolution of equity markets

100% 6
13 Others Chn Swi 2
5 Aut Can Aus 3
6 Rus
75%
Net 9
Jap
12 Fra 7
50% 13 Ger

25 UK USA 58
25%

15
0%
1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010

US UK Jap Ger Fra Can Aus Net Swi Rus Aut Chn Others

The continuing dominance of the USA is a striking feature

Source: Elroy Dimson, Paul Marsh, and Mike Staunton, Triumph of the Optimists, Princeton University Press, 2002, and subsequent research 3
From past to present: the USAs great transformation
Banks
Health Oil & gas
Banks
Retail
Rail
Other
industrial Other
Rail dominated
Other financial
USA in 1900 Iron, coal indus-
steel Insur-
trial
ance
Utilities Media
Tobacco Utilities
Telegraph Telecoms
Technology Travel & leisure
Other transport Drinks
Other Food Other

Start-1900 Start-2017
80% by weight in industries now small or non-existent
then small or non-existent 67% by weight in industries

Source: Elroy Dimson, Paul Marsh, and Mike Staunton, Triumph of the Optimists, Princeton University Press, 2002, and subsequent research 4
From past to present: the UKs great transformation
Banks Health Mines

Tobacco
Banks
Mines
Insurance

Rail dominated Textiles

Telecoms
Rail UK in 1900
Iron, coal
steel Oil & Utilities
gas
Drinks Media
Other industrial Other financial
Utilities Other Travel & leisure
Telegraph Retail
Other Insurance industrial Other Drinks

Start-1900 Start-2017
Disruptive technology has been familiar for two centuries
Investing in the new is not the obvious route to riches

Source: Elroy Dimson, Paul Marsh, and Mike Staunton, Triumph of the Optimists, Princeton University Press, 2002, and subsequent research 5
The past: Real returns and dividends, 19002016

10,000
United States 1,000
United Kingdom
513
1,402
1,000
Reinv- 100 Reinv-
ested ested
100
divi- divi-
10
dends
11.9
8.1
dends
10 9.8 3.3
2.7
2.6
1
1

0.10 0.10
1900 10 20 30 40 50 60 70 80 90 2000 10 1900 10 20 30 40 50 60 70 80 90 2000 10
Equities: return 6.4% p.a. Equities: capital gain 2.1% p.a. Equities: return 5.5% p.a. Equities: capital gain 0.8% p.a.
Bonds 2.0% p.a. Bills 0.8% p.a. Bonds 1.8% p.a. Bills 1.0% p.a.

We should focus on real return. Real return is largely dividends.

Source: Elroy Dimson, Paul Marsh, and Mike Staunton, Triumph of the Optimists, Princeton University Press, 2002, and subsequent research 6
Long-run asset returns for all countries, 19002016

Annualized real return (%)

6 6.4

5.1
4 4.3

0
Aut Ita Bel Fra Ger Prt Spa Jap Eur Nor WxU Ire Swi Net Wld Den Fin UK Can Swe NZ US Aus SAf
-2

-4
Equities Bonds Bills Historical equity risk Historical equity risk
premium vs. bonds = 3.2% premium vs. bills = 4.2%
-6

Prospectively, the world equity premium (versus bills) is 33%

Source: Elroy Dimson, Paul Marsh, and Mike Staunton, Triumph of the Optimists, Princeton University Press, 2002, and subsequent research 7
US equities were volatile with large real drawdowns
0%
-10%
-20%
-30%
-40%
-50%
-60%
Bonds
-70%
Equities
-80%

But bonds have also had large drawdowns in real terms

Diversification across asset classes helps to reduce risk

Source: Elroy Dimson, Paul Marsh, and Mike Staunton, Triumph of the Optimists, Princeton University Press, 2002, and subsequent research 8
The present: Yields on sovereign bills and bonds (%)
4
Since
In 2016
the 2016
rates hit
low,all-time
rates have
low risen
3

-1
Swi Ger Net Swe Bel Fra Fin Aut Den Ire Spa
Jap Spa Ita Por UK US Can Nor Aus
Ita Jap

3-months 10 years 20 years

But long rates still low. So short rates likely to remain low
Real (after inflation) returns on long bonds still very low

Source: Elroy Dimson, Paul Marsh, and Mike Staunton, Triumph of the Optimists, Princeton University Press, 2002, and subsequent research 9
The race to zero and beyond? Is it over?
Real yield for representative 10-year index-linked bond

-1

-2
00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17

US UK Fra Ger Jap Can Swe Average

No need to extrapolate past returns. Just look at current yields

Real yields still negative or close to zero. Slight upturn recently

Source: Elroy Dimson, Paul Marsh, and Mike Staunton, Triumph of the Optimists, Princeton University Press, 2002, and subsequent research 10
Low real interest rates means a low return world
Real rate of return (%)
10 11.0
9.4 9.8
9.0
7.6
5 5.2 5.2 4.6 5.6
4.2 2.8 4.0
1.4 1.4 2.2
0 0.1
-3.4 -2.2
- 5.4
-5
-10.0
-10 -11

-15
Low 5% Next 15% Next 15% Next 15% Next 15% Next 15% Next 15% Top 5%
Percentiles of real interest rates across 2,317 country -years
Equities next 5 years % p.a. Bonds next 5 years % p.a. Real interest rate boundary %

Real interest rates impact subsequent real equity and bond returns

Source: Elroy Dimson, Paul Marsh, and Mike Staunton, Triumph of the Optimists, Princeton University Press, 2002, and subsequent research 11
Many investors are anticipating a return to normal

300 years of UK bond yields. What yield is normal?

20

15

10

5 4.5
1.9
0

Long-run average

The high bond returns since 1980 were not normal

Source: Elroy Dimson, Paul Marsh, and Mike Staunton, Triumph of the Optimists, Princeton University Press, 2002, and subsequent research 12
Many investors are anticipating a return to normal
Average annual % real interest rate (short-term interest rate minus inflation)

4
3.9
3.1 3.2
2
2.2

0.7 0.4 0.9 1.2 0.1 0.3

0 -0.7 -0.5
-0.9 -0.6
-1.6 -1.7
-2

19001980 19812008 20092016 All years

-4

USA UK Europe All Yearbook

What real rate of interest is normal?

Source: Elroy Dimson, Paul Marsh, and Mike Staunton, Triumph of the Optimists, Princeton University Press, 2002, and subsequent research 13
Inflation on the rise? Equities, bonds and inflation
Rate of return/inflation (%)
20 19
19.1
10 12.2 12.0 10.9 10.9 7.7
8.0 7.1 4.2
5.3 4.2 5.9
0 0.4
1.7 2.62.7 2.0
-3.5 -4.5
-11.6
-10

-20 -22.7

-30
Low 5% Next 15% Next 15% Next 15% Next 15% Next 15% Next 15% Top 5%
Percentiles of inflation across 2,453 country-years; bond and equity returns in same year
Real bond returns (%) Real equity returns (%) Inflation rate boundary (%)

Equities best during low inflation. Bonds better in deflation.

Higher inflation harms bonds but it also harms equities.

Source: Elroy Dimson, Paul Marsh, and Mike Staunton, Triumph of the Optimists, Princeton University Press, 2002, and subsequent research 14
Equities have been a poor inflation hedge
Real equity return

150%

100%

50%
Sensitivity to
0%
concurrent
inflation was
-50% 0.52

-100%
-30% 0% 30% 60% 90%year
Inflation in same

UK US Ger Jap Net Fra Ita Swi Aus Can Swe Den Spa Bel Ire SAf Nor NZ Fin

Dont confuse inflation beating with inflation hedging

Source: Elroy Dimson, Paul Marsh, and Mike Staunton, Triumph of the Optimists, Princeton University Press, 2002, and subsequent research 15
The future: What equity premium can we expect?
Annualised (% p.a.)

5 0.5
5.1
0.5
4
4.1 0.8

3
3.3

0
Historical equity Real dividend Change in Average Treasury bill Expected equity
real return growth P/D ratio dividend yield return risk premium

We estimate the long-run equity premium vs bills to be 33%

Source: Elroy Dimson, Paul Marsh, and Mike Staunton, Triumph of the Optimists, Princeton University Press, 2002, and subsequent research 16
The high return world we grew up in

Annualized real returns on equities and bonds (%)

8

0
US Jap UK Can Aus Eur Wld US Jap UK Can Aus Eur Wld
Since 1950 Equities Bonds Since 1980
Baby boomers Generation X

Millennials? Past performance conditions our aspirations

Source: Elroy Dimson, Paul Marsh, and Mike Staunton, Triumph of the Optimists, Princeton University Press, 2002, and subsequent research 17
Looking forward from 2017

Annualized real returns on equities and bonds (%)

-2
World World World USA Japan UK Europe Emerging
since since markets
1950 1980 Equities Bonds
Historical high returns Prospective lower returns

Have expectations changed? We are still in a lower return world

Source: Elroy Dimson, Paul Marsh, and Mike Staunton, Triumph of the Optimists, Princeton University Press, 2002, and subsequent research 18
The future

Real interest rates are low, and bond yields are low
expected returns on all assets are lower than past returns
modest rises in inflation are unlikely to impact real returns

Long-run equity premium (vs bills) is around 33%

lower than the historical average
equities will stay volatile, but diversification lowers risk

What to do about the low-return world?

the Yearbook provides a long-term perspective
the current vogue is to seek returns from smart beta

Source: Elroy Dimson, Paul Marsh, and Mike Staunton, Triumph of the Optimists, Princeton University Press, 2002, and subsequent research 19
Investors ditching active management (USA, 200716)
Cumulati
ve asset
flow
(USD bn)

Sources: Investment Company Institute, Simfund, Credit Suisse 20

Rapid growth of factor-based ETFs
Smart beta assets under management USD bn
500

400

300

200

100

0
2012 13 14 15 16

Source: Financial Times, Morningstar Direct 21

Are investment institutions adopting smart beta?

75% now using or actively evaluating smart beta

Among users, two-thirds considering further allocation

20% now use 5 smart beta indexes (vs 2% in 2014)

22
The smart-beta zoo

Researchers have reported on 458 factors

few will work out-of-sample

Fama-French identify five factors

market, size, value, profitability, investment
others stress low-risk, and momentum

We study five factors

low-risk, momentum, size, value, income
over up to 117 years, and in up to 23 markets

Source: Elroy Dimson, Paul Marsh, and Mike Staunton, Triumph of the Optimists, Princeton University Press, 2002, and subsequent research 23
Low-risk investing

Classic strategy (published 1972)

low beta portfolio gives superior risk-adjusted return
now labelled as BAB (Bet Against Beta)

Recent approach uses idiosyncratic volatility

low-volatility portfolio gives superior return
compared to high-volatility portfolio

Many variants of these low-vol strategies

an approach that has become popular since the GFC

Source: Elroy Dimson, Paul Marsh, and Mike Staunton, Triumph of the Optimists, Princeton University Press, 2002, and subsequent research 24
Low risk factors in the USA, 19632016
1,000
247
216
212
100 122

10 9.7
8.8

0.10

Low specific risk 10.9% p.a. Low variance 10.6% p.a. Low beta 10.5% p.a.
High specific risk 4.1% p.a. High variance 4.3% p.a. High beta 9.4% p.a.

Weakest effect: Betting against beta

beta. Strongest: Specific risk

Source: Data from Professor Kenneth French, Dartmouth (website) 25

Specific risk factor in the USA
1,000
469
318
247
233
100

10 8.8

0.10

Lowest risk 10.9% p.a. Lower risk 10.7% p.a. Medium risk 12.2% p.a.
Higher risk 11.4% p.a. Highest risk 4.1% p.a.

Premium arises from underperformance of the highest-risk quintile

Based on daily returns. Source: Data from Professor Kenneth French, Dartmouth (website) 26
Specific risk factor in the UK
100

35.9
24.9

10

3.9

0.10

Lowest risk 11.6% p.a. Medium risk 10.3% p.a. Highest risk 4.2% p.a.

Same methodology using daily data: UK results similar to USA

Source: Elroy Dimson, Paul Marsh, and Mike Staunton, Triumph of the Optimists, Princeton University Press, 2002, and subsequent research 27
Longer-term (60-month) specific risk estimates in UK
1,000
623
571
282

100

10

Lowest risk 12.0% p.a. Medium risk 11.8% p.a. Highest risk 10.4% p.a.

Effect driven entirely by highest risk stocks during dot-com crash

Source: Elroy Dimson, Paul Marsh, and Mike Staunton, Triumph of the Optimists, Princeton University Press, 2002, and subsequent research 28
Momentum

Sort stocks by return over past 6 or 12 months

target the top and bottom quintile (say)

Wait a month
then buy past winners (and short past losers)

Rebalance periodically
typically after 6, 3 or 1 month(s)

Measure premium: WML (Winner Minus Loser)

Source: Elroy Dimson, Paul Marsh, and Mike Staunton, Triumph of the Optimists, Princeton University Press, 2002, and subsequent research 29
The momentum premium, 1900*2016

United States United Kingdom

10,000,000
10,000,000 5,043,966
2,131,993
1,000,000
1,000,000

100,000
100,000

10,000
10,000
3,634

1,000 1,000

100 100
61

10 Cumulative difference 10
between winners and
losers 7.4%
1 1 Cumulative difference
between winners and
losers 10.4%
0.10 0.10
1900 10 20 30 40 50 60 70 80 90 2000 10

Winner 17.5% per year Loser 9.5% per year Winner 14.1% per year Loser 3.6% per year

Large long-run, pre-cost returns; but volatile with high turnover

* from 1926 in the US Based on a 6/1/6 momentum strategy Based on a 12/1/1 momentum strategy

Source: Elroy Dimson, Paul Marsh, and Mike Staunton, Triumph of the Optimists, Princeton University Press, 2002, and subsequent research 30
Momentum returns around the world

Winner minus loser returns, % per month

1.5

1.0
.79
.71

0.5

0.0

-0.5
Jap Chi Rus Spa US Can Aus Swe Prt Fra Aut Avg Ger Neth Ita Bel Swi Fin Ire UK Nor Den NZ SAf

Griffin, Ji & Martin: to end-2000 Full period to end-2016 (updated by Dimson, Marsh & Staunton)

Unusually, the factor premium was larger after the original study

Source: Elroy Dimson, Paul Marsh, and Mike Staunton, Triumph of the Optimists, Princeton University Press, 2002, and subsequent research 31
Other factors: the size premium, 1926*2015

United States 100,000 United Kingdom

100,000
53,263
33,879 27,256

10,000 10,000
6,861
4,690
3,220
1,000
1,000 1,087

100

100
10

10
1

0.10 1
26 30 35 40 45 50 55 60 65 70 75 80 85 90 95 00 05 10 15 1955 60 65 70 75 80 85 90 95 2000 05 10 15

Micro-caps 12.7% per year Small-caps 12.1% per year Micro-caps 17.9% per year Small-caps 15.3% per year
Larger-caps 9.7% per year Mid-caps 13.9% per year Large-caps 12.0% per year

The smallest firms have performed the best, but not consistently
* from 1955 in the UK

Sources: US CRSP capitalization deciles are from Morningstar; UK Small and Mid-caps are Numis indices 32
The size effect around the world

% per month (small minus large)

0.8

.45
0.4 .32

0.0

-0.4
Nor Net Fin NZ Ita Den SAf Spa Chi UK Bel Swe Ger US Avg Swi Can Aus Rus Jap Por Ire Aut Fra

Longer term 2000 to 2016

A global phenomenon; large premium since 2000

Source: Elroy Dimson, Paul Marsh, and Mike Staunton, Triumph of the Optimists, Princeton University Press, 2002, and subsequent research 33
Other factors: the value premium, 1926*2016

United States United Kingdom

100,000 10,000
9,173
56,247

10,000
Value Value
4,274 1,000 1,146
3,061
1,000 419

100 100

10
Growth
10
Growth
1

0.10
1
26 30 35 40 45 50 55 60 65 70 75 80 85 90 95 00 05 10 15 1955 60 65 70 75 80 85 90 95 2000 05 10 15

High book-to-market 12.9% p.a. US market 9.8% p.a. High book-to-market 16.0% p.a. UK market 12.1% p.a.
Low book-to-market 9.3% p.a. Low book-to-market 10.3% p.a.

Value stocks have outperformed, but not always

* from 1955 in the UK

Source: US data from Professor Kenneth French, Dartmouth (website) 34

The value effect around the world
% per year (value minus growth)
10

5
2.1 2.5

-5

-10
Ire Fin Den Swi Ita Por NZ US Bel Ger UK Spa Wld Fra SAf Net Swe Can Aus Aut Nor Jap Chi Rus

Longer term Since 2000

Value beat growth in most countries over the long-run

Source: MSCI value and growth indexes 35

Other factors: the income premium, 1900*2016

United States United Kingdom

100,000 1,000,000

13,991 158,727
10,000 100,000
7,156
35,966
2,312
1,000 1,451 10,000
6,810

100 1,000

10 100

1 10

0.10 1
End 1900 10 20 30 40 50 60 70 80 90 2000 10

High yield 11.3% p.a. Medium yield 10.4% p.a. High yield 10.8% p.a. UK market 9.4% p.a.
Low yield 9.0% p.a. Zero yield 8.5% p.a. Low yield 7.8% p.a.

High yielders have beaten low yielders; a variant of the value effect
* from 1927 in the USA

Source: US data from Professor Kenneth French, Dartmouth (website) 36

The income premium around the world
Yield premium (% per year)

15

10
6.2
5 3.9

-5

-10
NZ Ire Bel Swi US Ger UK Ita Den Fin Avg Spa Sin Can HK Net Aus Fra Nor Jap Swe Aut
Longer term Since 2000

Positive premia in most countries; large premia since 2000

Source: All data except UK data is from Professor Kenneth French, Dartmouth (website) 37
Factor performance since the financial crisis

USA 2008 2009 2010 2011 2012 2013 2014 2015 2016 200816
Low vol Size Size Low vol Value Size Low vol Momentum Value Low vol
Highest 90.3 28.5 13.6 40.5 11.5 5.3 11.3 42.3 17.2 6.0
Income Value Momentum Income Size Value Income Low vol Income Size
20.7 8.0 8.5 29.5 7.8 4.7 1.6 13.9 14.8 4.0
Momentum Income Income Momentum Momentum Momentum Value Income Size Income
2.4 17.2 7.1 1.4 0.9 4.5 2.2 2.4 9.6 3.8
Size Low vol Value Size Low vol Income Momentum Size Low vol Value
4.3 33.0 4.5 3.7 1.5 8.2 5.3 9.3 1.8 1.8
Value Momentum Low vol Value Income Low vol Size Value Momentum Momentum
Lowest 6.0 50.6 15.2 12.8 7.6 9.3 6.7 12.0 22.4 6.0

UK
Low vol Size Size Low vol Size Momentum Momentum Low vol Value Momentum
Highest 127.0 24.9 12.4 35.0 17.0 32.4 42.8 23.7 20.2 12.8
Momentum Income Value Income Value Size Size Momentum Income Size
78.8 1.1 3.2 28.3 14.8 15.5 12.1 20.1 15.3 6.5
Income Value Momentum Momentum Momentum Low vol Income Size Size Low vol
15.7 6.9 0.7 20.6 1.7 11.5 1.3 11.1 4.9 5.5
Value Low vol Income Size Income Income Low vol Income Momentum Income
11.8 20.1 13.7 4.9 8.1 0.0 6.2 11.2 18.3 2.1
Size Momentum Low vol Value Low vol Value Value Value Low vol Value
Lowest 17.5 25.4 22.9 10.7 15.7 0.0 10.0 20.9 21.2 3.2

Differences over time and across markets; scope to diversify

Source: Elroy Dimson, Paul Marsh, and Mike Staunton, Triumph of the Optimists, Princeton University Press, 2002, and subsequent research 38
Smart beta

Distinguish factors vs premiums

factor: an influence on security returns
premium: a superior expected return

Premiums may be evident over the long run

some (size, value) may be harvested passively
some (low-vol, momentum) require portfolio churning

Factor exposures can have a large performance impact

investors can unwittingly take large positions

Factors can become too expensive

popularity can make them an over-crowded trade

Source: Elroy Dimson, Paul Marsh, and Mike Staunton, Triumph of the Optimists, Princeton University Press, 2002, and subsequent research 39