Alfred Berg Global Alpha Engineered to perform in all markets
1
Presenters Ole Jakob Wold is Head of Global Quantitative Equity at Alfred Berg, and responsible for developing, testing and running all quantitative investment models for managing the portfolios. Ole Jakob has held this position since the inception of Alfred Berg’s quant investment centre in 1994. Ole Jakob earned a ”Siviløkonom“ (civil economics) degree from the Norwegian School of Economics and Business Administration in 1994. He majored in finance, with studies in empirical finance and a thesis on quantitative models used on the Oslo stock exchange. Ole Jakob is also a member of the Norway management team of Alfred Berg.
Nardin L. Baker is Senior Portfolio Manager in the Quant Equity team and actively involved in research, model development, and marketing. Mr. Baker is a well known expert in the field of multi-factor analysis. He researched and developed the theory behind the quantitative investment model in use at Alfred Berg together with Prof. Haugen. Mr. Baker previously directed the Global Asset Allocation group for Grantham, Mayo, Van Otterloo & Co. LLC (GMO) in Boston. He also led the Tax-sensitive Equities area at GMO. Before that, he directed the quantitative equity management group for National Investment Services of America (NISA) in Milwaukee, Wisconsin. He received his MBA and B.S. degree in mechanical engineering from the University of Illinois. He is a Chartered Financial Analyst (CFA).
Cristina Lugaro is the Investments Specialist of the Global Quant Equity team. Prior to joining Alfred Berg in 2008, she was a member of the Global Marketing team of Fortis Investments in Belgium. After graduating from university, she worked for The Bank of New York Mellon and Euroclear. Cristina received an Economics degree from Vesalius College in 2002. She also holds an MBA diploma from Solvay Business School (2005).
2
Table of Contents Section 1:
The Time for Global Equity
Portfolio Expectations
Experience and Track Record
Section 2:
Performance
Section 3:
Philosophy
Section 4:
Process
Conclusion Appendix
3
> The Time for Global Equity | Expectations | Track Record | Performance | Philosophy | Process
The Time for Global Equity
Why Global Equity?
Why Alfred Berg Global Alpha?
Why now?
Global Equity provides the broadest set of possible investments –
Largest number of country, industry and stock opportunities
–
More investment choices allows for more alpha
Global Alpha works around the world –
Same investment model is used in all countries
–
Systematic evaluation and comparison of stocks produces an accurate understanding of the opportunities and tradeoffs
–
Integrated stock selection and risk-control produces consistent performance
Stable performance in volatile and uncertain markets –
Lately, many managers have been revealed to be riskier than expected
–
Global equity exposure in a stable investment style anchors the allocation and allows investors to better manage all their assets
4
> The Time for Global Equity | Expectations | Track Record | Performance | Philosophy | Process
Significant Alpha Opportunity The median global equity manager outperforms their regional counterparts against their respective benchmarks. Investors seeking international exposure are best served by utilizing a global equity manager rather than a collection of “specialized” regional managers. Even on a risk-adjusted basis global equity managers have generated stronger results.
Alpha generated by the median manager over the 5YR period (ended 31 December 2008) 2.0% 1.5% 1.0% 0.5% 0.0% -0.5% -1.0% US Large US Small European Japanese Cap Equity Cap Equity Equity Equity
Asia ex Japan
Emerging Markets Equity
Global Equity
Information ratio generated by the median manager over the 3 & 5 YR periods (ended 31 December 2008) 0.4% 0.3% 0.2% 0.1%
Source: Mercer Based on trailing annualized performance of the peer group median relative to the comparable index as of 31 December 2008. All returns are shown gross of fees in local currency (Emerging, Asia ex Japan & Global Equity shown in USD). Median peer groups shown versus the Russell 1000 Index (US Large Cap Equity), Russell 2000 Index (US Small Cap Equity), MSCI Europe Index (European Equity), MSCI Japan Index (Japanese Equity), MSCI AC Far East ex Japan Index (Asia ex Japan), MSCI EMF Index (Emerging Markets Equity) and MSCI World (Global Equity).
0.0% -0.1% -0.2% -0.3% -0.4% -0.5% -0.6% -0.7% US Large US Small European Japanese Asia ex Cap Cap Equity Equity Japan Equity Equity 3 year
5 year
Emerging Markets Equity
Global Equity
5
| The Time for Global Equity > Expectations | Track Record | Performance | Philosophy | Process
Portfolio Expectations Performance
Risk Control
Consistency
Engineered to outperform –
Systematic process to evaluate and rank all stocks in the developed markets
–
Alpha: 2.5% to 5.0% per year
Diversified exposure –
We invest in all major geographical regions
–
We diversify across all sectors and industries
–
We hold 100 to 200 names in the portfolio
–
Tracking error: 3.0% to 5.0% per year
Consistent performance relative to the MSCI World benchmark over 5-year periods –
Higher return 80% of the time
–
Lower risk 60% of the time
–
Higher return-to-risk ratio 75% of the time
6
| The Time for Global Equity | Expectations > Track Record | Performance | Philosophy | Process
Experience and Track Record Experience
15 year record outperforming MSCI World Index
Stable investment process and team
Factor model development and published research
Performance
Excellent returns in growth, value, large cap, small cap, bull and bear markets
Award* winning performance during “07-08” turmoil 5-year Return vs Risk (March 2009)
Risk Management 3%
Volatility lower than benchmark 60% of the time**
Tracking Error 4%
Consistency
Positive alpha 83% of the time since inception**
Higher return-to-risk 73% of the time**
AB Global Alpha 2% Alpha 1%
MSCI World
0%
-1 % 10 %
11 %
12 %
13 %
14 %
15 %
Risk
* Morningstar Award, Large Cap Equity 2009 ** Fund Inception in October 1994, measured on 5 yr rolling periods
7
| The Time for Global Equity | Expectations | Track Record > Performance | Philosophy | Process
Consistent Performance Global
MSCI World
Alpha
Tracking Error
-6.48% -25.70% -13.59% -2.40%
-7.78% -31.47% -16.31% -4.93%
1.31% 5.77% 2.72% 2.54%
3.1% 2.7% 2.5%
1.9 1.0 1.0
Since Inception Return Volatility Return/Risk
4.6% 15.5% 0.30
2.9% 15.8% 0.18
1.8%
3.7%
0.5
Characteristics Beta: 5 years Intercept (alpha): 5 years R-squared: 5 years
0.95 2.48% 0.97
Year to date 1 Year 3 Years 5 Years
Information Ratio
These results are based on the track record of the existing Alfred Berg Global Quant fund, which is domiciled in Norway (NO0010089501) and representative for the global alpha strategy. Alfred Berg Global Alpha is a clone of this fund and has identical holdings. Domiciled in Luxembourg, it is a UCITS III-compliant SICAV and is available for sale to investors worldwide. Returns are expressed in EUR and are gross of fees as at the end of March 2009. Benchmark is the MSCI World Net Index. Periods greater than one year are annualised. Launch date: October 1994.
8
Product Characteristics Comparison with the Mercer Global Equity Universe Risk and Return Characteristics (calculated quarterly) versus MSWF for the period from mar 2004 to des 2008 7
22
0,4
11,2
1,1
4
19
0,2
8,8
0,6
1
16
0,0
6,4
0,1
-2
13
-0,2
4,0
-0,4
-5
10
-0,4
1,6
-0,9
Alfred Berg FIMGEQ5 Global Alpha 95th Percentile Upper Quartile Median Lower Quartile 5th Percentile Number of Funds
Return (% pa)
Std Deviation (% pa)
Reward to Risk
2,2 (75)
15,6 (179)
0,1 (67)
7,6 3,3 1,4 -0,3 -2,9 218
22,9 19,0 17,3 16,1 13,6 218
0,4 0,2 0,1 0,0 -0,1 218
Tracking Error (% pa) 2,4 (194) 11,3 6,7 4,4 3,2 2,0 218
Information Ratio 0,9 (29) 1,1 0,6 0,3 -0,1 -0,6 218
9
Significance of Alpha (T-stat) Comparison with the Mercer Global Equity Universe Confidence of Value Added for periods ended desember 2008 Confidence (%) 100
75
50
25
0
Alfred Berg FIMGEQ5 Global Alpha 95th Percentile Upper Quartile Median Lower Quartile 5th Percentile Number of Funds
3 Years 95,9 96,7 86,7 67,0 40,9 8,3 292
5 Years (29)
97,2 99,0 91,7 78,0 47,7 11,6 218
7 Years (31)
89,6 99,7 96,1 85,2 54,4 20,0 169
(75)
10
Top Ranking from Mercer Mercer Ranking on Information Ratio, December 2008 (Universe core global equity covering 173 managers)
Percentiles
Ranks
Top 10% on last 2 and 3 years Highest 15% in all horizons
7th best strategy last 3 years 11th best last 2 years 15th best lat 5 years 23rd best last 1 year
11
| The Time for Global Equity | Expectations | Track Record | Performance > Philosophy | Process
Philosophy Dynamic Markets
As the economy changes, investment opportunities change –
Macro-economic environments evolve
–
Stock markets react to changes in the economy by re-pricing fundamental characteristics
–
Our stock rankings change as the payoffs to fundamentals are repriced
–
Paraphrasing John Maynard Keynes:
“When the market changes, we change our stock opinions. What do others managers do?”
Systematic Methods
Engineered for consistency –
Consistency comes from systematic portfolio construction
–
The same stock ranking model is applied across all developed stock markets, with different weights in each
–
Risk control is applied to manage exposures: country, sector, stock, volatility, tracking error, liquidity and others
12
Market Profiling Systematic fundamental analysis Market Profiling
Forecasting
Stock Ranking
Portfolio Construction
Fundamental data for over 23,000 companies Multi-factor regression analysis using 60 factors We measure market sentiment and market payoffs
13
| The Time for Global Equity | Expectations | Track Record | Performance | Philosophy > Process
Fundamental Drivers Which factors affect stock returns? Value
• Book/Price • Earnings/Price • Sales/Price
Growth
• Earnings Growth • Sales Growth
Size
• Market Capitalization • Earnings Stability
Profitability
• Return on Assets • Return on Equity
Risk
• Volatility • Trading Range • Debt/Equity
Momentum
• 1 Month • 12 Month
GICS Sector • Financials, IT, Energy… Macro
• Interest Rates, Inflation..
14
Why Use Many Factors? The red area indicates market focus in a value environment The black area indicates market focus in a growth environment For long periods, these characteristics may have little or no effect, and models with few factors may fail The flash light highlights the unusual 2007-2008 market environment Limiting the number of factors can creates style bias and inconsistent performance
15
Forecasting Systematic fundamental analysis Market Profiling
Forecasting
Stock Ranking
Portfolio Construction
The time series of the estimated factor pay-offs are used to predict the following month's factor returns.
16
| The Time for Global Equity | Expectations | Track Record | Performance | Philosophy > Process
Regression Analysis The slope of the line is the payoff to market cap Relative performance, next month
Positive Zero regression regression payoff: payoff: Negative regression payoff:
No market cap effect Small-Cap Large Cap Market Market cap
17
Factor Payoffs Through Time The gold line is our estimate of the market payoff to size
Source: Alfred Berg
18
Our Analysis Reveals… …the strong value market in the USA from 2001 to 2007
Long lasting value market in USA Source: Alfred Berg
19
Stock Ranking Systematic fundamental analysis Market Profiling
Forecasting
Stock Ranking
Portfolio Construction
Stocks are ranked based on their ability to deliver performance in the following month. We determine this expected performance by matching company fundamentals with the pay-off forecasts.
20
Stock Valuation Calculating expected returns Easily available
Model output
Expected return
Forecast Forecast
Company Exposure
factor payoff
Alpha Contribution
2.38 σ
-0.25%
-0.63%
2 Return on Assets 3 Book to Price
1.81 σ -1.60 σ
0.15% 0.11%
0.26% -0.20%
4 Volatility, 24 Mo
-1.81 σ
-0.16%
0.19%
……
…
…
…
……
…
…
…
-0.41 σ
0.29%
-0.12%
Factors 1 Market Capitalization
60 Relative Strenght (12 Mon)
x
Expected Relative Return
=
Σ
0.88%
Company exposure is measured in standard deviations relative to the mean.
21
Stock Example 1.13
22
Portfolio Construction Systematic fundamental analysis Market Profiling
Forecasting
Stock Ranking
Portfolio Construction We use optimization technology to construct a reliable portfolio. We diversify by sector, country, and factors. We minimize portfolio risk and trading costs.
23
Portfolio Construction Objectives
Constraints
Implementation
Portfolio Optimization –
Maximize expected return
–
Minimize portfolio risk
–
Minimize trading costs
Relative to Benchmark –
Country exposures
± 5.0%
–
Sector exposures
± 5.0%
–
Stock exposures
± 2.0%
Efficient Trading –
Liquidity analysis
–
Managing market impact
–
Trading with low commissions
24
| The Time for Global Equity | Expectations | Track Record | Performance | Philosophy > Process
xProfile – Style Skyline In 2004 we owned a pure value portfolio
25
Profile – Style Skyline In 2008 we saw a flight to quality
26
| The Time for Global Equity | Expectations | Track Record | Performance | Philosophy > Process
Investment Product Matrix
Flexible investment process is applied to global markets, regions, and countries Customization achieved through selecting the investment universe and benchmark Style universes can be selected such as value and growth or large and mid-cap Same stock ranking and portfolio construction methods are applied across the matrix 27
> Conclusion
Conclusion Our global strategy anchors the equity allocation A product for all seasons, delivering consistent performance over time and through different market environments Experienced investment team with strong ties to the academic and research communities A systematic approach using 60 fundamental factors in all countries Engineered for consistency: wins 80% of the time with lower risk 60% of the time
28
APPENDIX
MAY 2009
29
> Appendix
GLOBAL ALPHA TEAM
High Level of Competence… …across the whole process: data, modeling, analysis and execution Portfolio management
Research
Ole Jakob Wold (NO)
Nardin L. Baker (US)
Stig Arild Syrdal (NO) Robert A. Haugen (US)
Overall responsibility for performance
Quantitative stock selection analysis
Portfolio optimisation
Technology advisor
Risk management
Accountable for model portfolio and decision making
Alpha modelling
Research and model development Marketing events
Development and testing of new strategies
Model development and testing
Data management Tom Erik Brurok (NO) Quantitative data analysis
Portfolio construction
Portfolio implementation
Sales/Marketing
Roelof Koopmans (NO)
Jonathan Moor (FRA)
Cristina Lugaro (SE)
Portfolio Constructor
Global Dealing Desk
Investment Specialist
Support functions
Support functions
Support functions
Kristof Holsters (BE)
Stephane Jouveaux (FRA)
Tom Bagguley (NL)
Reconciliations & settlements
Equities dealer
RFP Specialist
Nicholas watts (UK)
Frederic Olanie (FRA)
Bettina Sökare (SE)
Portfolio administration
Equities dealer
PR & Marketing support
Jean Hoflack (FRA)
Björn Hellström (SE)
Equities dealer
Campaign Manager
30
> Appendix
Biographies Global Alpha Team Ole Jakob Wold is Head of Global Quantitative Equity at Alfred Berg, and responsible for developing, testing and running all quantitative investment models for managing the portfolios. Ole Jakob has held this position since the inception of Alfred Berg’s quant investment centre in 1994. Ole Jakob earned a ”Siviløkonom“ (civil economics) degree from the Norwegian School of Economics and Business Administration in 1994. He majored in finance, with studies in empirical finance and a thesis on quantitative models used on the Oslo stock exchange. Ole Jakob is also a member of the Norway management team of Alfred Berg.
Nardin L. Baker is Senior Portfolio Manager in the Quant Equity team and actively involved in research, model development, and marketing. Mr. Baker is a well known expert in the field of multi-factor analysis. He researched and developed the theory behind the quantitative investment model in use at Alfred Berg together with Prof. Haugen. Nardin previously directed the Global Asset Allocation group for Grantham, Mayo, Van Otterloo & Co. LLC (GMO) in Boston. He also led the Tax-sensitive Equities area at GMO. Before that, he directed the quantitative equity management group for National Investment Services of America (NISA) in Milwaukee, Wisconsin. He received his MBA and B.S. degree in mechanical engineering from the University of Illinois. He is a Chartered Financial Analyst (CFA).
Robert A. Haugen is a long-term associate of Alfred Berg, and has supported our business and quant team since 1994. He has no day-to-day role in the business, but acts as general advisor to our model development and clients. Having published the first academic article documenting the nature and power of the expected return factor model in The Journal of Financial Economics in 1996, Haugen is considered to be the father of this important, quantitative investment tool. Prof. Haugen has written fourteen books, published in seven languages, including Modern Investment Theory (widely used in M.B.A. courses throughout the world), The Incredible January Effect and The New Finance (required reading on the Chartered Financial Analysts’ examination for the last five years). Haugen has held chairs at the University of Wisconsin, University of Illinois, and the University of California. Prof. Haugen is currently President of Haugen Custom Financial Systems, providing estimates of the expected returns for 8,000 stocks world-wide to institutional investors.
31
> Appendix
Biographies Global Alpha Team Stig Arild Syrdal is a member of the Portfolio Management team. His role focuses on model development, portfolio optimization and analysis. Stig Arild’s educational background includes a M.Sc degree in naval architecture and offshore engineering from the Norwegian University of Science and Technology in Trondheim. Part of the degree was completed at University of California, Berkeley. Stig Arild also holds a Cand.Merc degree (equivalent to a Master’sdegree) in Economics and Business Administration from the Norwegian School of Economics and Business Administration in Bergen.
Roelof Koopmans is a Portfolio Constructor in Alfred Berg’s Global Quant team. He joined the company in 2000 as Settlement specialist, joining the team in Oslo in 2007 and assuming his current responsibilities of cash management and portfolio implementation in August 2008. Roelof Koopmans received the “Businessøkonom" Degree in 1997 at the Amsterdam School of Economics and Business Administration. His major subject was Treasury/Controlling.
Jonathan Moor is Head of European Equity Dealing in Paris for Fortis Investments. He came to Fortis in 2007 from Merrill Lynch in London where he was Vice President, trading portfolios, futures, SWAPs and ETFs. Prior to working at Merrill Lynch, Jonathan was a Director at UBS in London and before that HK, having started the Portfolio Trading desk there for UBS Warburg. Before a career in finance, Jonathan earned a double degree in Developmental Anthropology and Chinese from the University of Montana and Hangzhou University in Zhejiang, People's Republic of China in 1995.
Cristina Lugaro is the Investments Specialist of the Global Quant Equity team. Prior to joining Alfred Berg in 2008, she was a member of the Global Marketing team of Fortis Investments in Belgium. After graduating from university, she worked for The Bank of New York Mellon and Euroclear. Cristina received an Economics degree from Vesalius College in 2002. She also holds an MBA diploma from Solvay Business School (2005).
32
> Appendix
Global Alpha Characteristics Name: Fund category: Legal structure: Domicile: Denomination: Expected launch date: NAV Calculation Agent: Fund manager: Benchmark: Investment Style: Investment Approach: Tracking error target: Tracking error Limit: Management fee: Classic shares: Institutional shares: Share class: ISIN code (C-share): (I-share):
Alfred Berg Global Alpha Quantitative Global Equities SICAV with UCITS III status Luxembourg EUR 14 May 2009 Fastnet Lux SA Ole Jakob Wold MSCI World Index (NR) Neutral, core Active Quant 4% 5%
Investment guidelines Benchmark:
MSCI World Index (NR)
Risk tolerances: Target ex-post tracking error: Target Information ratio:
4% (max. 5%) > 0.5
Internal guidelines: # of holdings: Maximum position size*: Maximum sector exposure*: Maximum country exposure*: Annual turnover:
150-200 +/- 2% +/- 5% +/- 5% 100%
1.5% 0.60% Classic (Cap & Dis), I (Cap) LU0410864085 LU0410865132
Global Alpha is attractive to clients who… …are interested in gaining active exposure to global equities with a low volatility …appreciate the disciplined nature of quantitative investing …want a core equity product which adjusts to market cycles but is style neutral over time 33
> Appendix
Risk Modeling and Volatility Risk is modelled using the full covariance matrix of historic returns – Returns are smoothed and normalized using similar techniques as in the alpha modelling – The Var-Covar calculations are weighted, using monthly return observations – Various factor based models are being tested, but not implemented
The optimizer routine is coded in Matlab, with state of the art solver technology1) from Stanford Systems Optimization Laboratory, and Tomlab as interface. Our portfolio optimizer routine is fully integrated with our MySQL based holdings- and alpha-analysis databases, used for back testing.
1) The
routine follows P.E. Gill and W. Murray; Numerically stable methods for quadratic programming. Mathematical Programming, 14, 349-372, 1978)
34
> Appendix
Risk Modeling and Relative Volatility Tracking Error is controlled by constraining exposure versus benchmark weights on: –
GICS Sectors
–
Country
–
Security
Ex-ante Tracking Error is monitored daily: –
Barra Aegis analysis done by an external reporting group
–
Analysis in Algorithmics tools from Bloomberg
1) The
routine follows P.E. Gill and W. Murray; Numerically stable methods for quadratic programming. Mathematical Programming, 14, 349-372, 1978)
35
> Appendix
Risk Monitoring and Analysis We upload our portfolios to Bloomberg daily. This allows: – Real-time performance analysis with breakdown on sector, industry, country, currency, region as examples – VaR analysis – Scenario analysis of simulated sensitivities to historical events – Ex-ante Tracking Error – Portfolio attribution – using routines provided by Algoritmics – Portfolio alerts – news collection and corporate action calendars
36
> Appendix
Understanding and Modeling Stock Returns Stock returns are sensitive to country, sector, and fundamental factors
Ri = Rcountry + Rsector + Rfundamentals + Rresidual Country returns are volatile, are effected by currency returns, and are difficult to forecast Sector returns are also volatile and difficult to forecast Fundamental returns are more persistent, have strong autocorrelation and can be forecast
37
> Appendix
Fundamental Returns are Local Fundamental returns are country specific
Rfundamentals = Rlocal macro + Rlocal preferences Comparing fundamentals within each country provides consistent treatment of accounting, reporting, currency and macro-economic effects Fundamental returns are driven by the momentum and inertia of the local economic environment Relative stock ranks can be forecast within each country due to the consistency of the treatment of fundamentals
38
APPENDIX
Disclaimer
This document is for information purposes only. We have taken all reasonable care to ensure that the information contained herein is reliable. This material is intended for the use of intermediate customers and market counterparties only, and is not for distribution to private customers as defined by the Financial Services Authority. Please note that past performance is not an indication of future performance. The value of investments can go down as well as up, and you may not get back the full amount invested. Changes in the rates of foreign exchange may cause the value of investments to fluctuate. The returns will be reduced by the management fees and other expenses incurred in the management of a customer’s account. The performance numbers shown are for illustrative purposes only to indicate possible returns. Please note that returns are unlikely to match exactly those of the funds shown as charges and tax treatment may differ. The companies within the Alfred Berg Group may make markets in the securities mentioned in this material and may, at any time, have long or short positions in any such security. They may also have provided within the previous 12 months significant advice or investment services in relation to these securities.
39
