ASSET ALLOCATION IN A WORLD OF UNCERTAINTY VBA ALM-CONFERENTIE 2012
KNUT N. KJAER, HS ZEIST, 1 NOVEMBER 2012
[email protected]
OVERVIEW
Portfolio risk and performance is mostly driven by allocation and exposure to key risk factors. Expected risk premiums vary over time, thus the traditional portfolio allocation model, with relatively stable strategic portfolios and a split between the alpha and beta components, may not be appropriate in periods of structural change and through regime shifts. This presentation will address alternative models for asset allocation and risk management that takes into account uncertainty, regime shifts, governance structures and investor behavior.
Trient Asset Management
2
KEY ISSUES FOR THE PENSION FUND PRINCIPAL
Trient Asset Management
3
OUTLINE
1.
Regime shifts. High volatility. Real uncertainty
2.
Redefining the concept of risk
3.
The cardinal investor mistakes
4.
Risk ownership and fund governance
5.
Dealing with high volatility and real uncertainty: Building blocks of dynamic allocation of risk
6.
Hedging strategies
7.
Discussion
Trient Asset Management
4
THE NEW POTATO / GIFFEN GOOD
Trient Asset Management
5
LOW YIELDS ON FIXED INCOME TAKES DOWN INVESTORS CAPACITY TO HOLD RISKY ASSETS LOWER YIELDS CAN IMPLY HIGHER DEMAND
:
Proportion in risky assets
•
•
•
High vol and falling short rates makes investors using a drawdown approach piling out of risk from 2008 Extremely low short rates makes it difficult to get back on the «risk-train» Graph: Rebalancing between a risky asset and a risk free asset, with a shortfall constraint – less than 5% probability for negative return
Source: Trient Asset Management
Trient Asset Management
6
LOWEST RATES EVER
(would you hedge your interest rate risk now?)
Source: Robert Shiller and Goldman Sachs global ECS research
Trient Asset Management
7
FEEDBACK LOOPS AND REFLEXIVITY. HIGHER PRICE CAN IMPLY HIGHER DEMAND (HOUSING)
Source: “What Do Banks Do? What Should They Do? Adair Turner, CASS Business School, 17 March 2010
Trient Asset Management
8
1.
REGIME SHIFTS. HIGH VOLATILITY. REAL UNCERTAINTY
THE END OF “THE GREAT MODERATION”
G7 Real GDP Development vs. Long-term Trend
The Great Moderation The Great Clean Up
Source: Trient Asset Management, OECD Economic Outlook
Applying a 1970-2008 linear trend on real G7 GDP growth, the demand shortfall following 2008 is evident.
Trient Asset Management
10
The Great Moderation 1990 - 2007 A long period with high growth, basically driven by credit China enters WTO in 2001, some hundred million people in Asia joins the global export-oriented market The build-up of production capacity eases inflation pressure in the developed markets. Monetary policy can stay accommodative China becomes the lender of last resort and fuel the credit driven expansion in the West Demand created by the easy credit disguise the increasing structural challenges related to lack of competitiveness with the emerging market producers. The economic structure becomes unsustainable: Construction and housing booms in many countries
The Great Recession 2008 – 2009 (could have become the New Depression)
The inevitable adjustment of private sector balance sheets capsize the overleveraged financial sector. Self-reinforcing cycles of fall in asset prices, decline in the capital of financial institution and household wealth creates the largest fall in developed markets GDP since the 1930’s.
Trient Asset Management
The Great Clean-up 2010 - 2020 Public gross debt in advanced economies is above 100% of GDP, the highest in 130 years except for some years following the 2nd War. Private sector debt is record high in most of the advanced economies. US is the only one with significant decline in private debt. Debt levels must down and the process will reduce overall demand and growth potential. Government has little powder left to counterweight downturns. They can not afford fiscal policy stimulus and the monetary policy is near fully exploited. The Western welfare states are under treat from increased global competition and policy responses have so far been inadequate . We expect only moderate growth and quite high volatility over the next decade
11
FACTORS SHAPING THE OUTLOOK FOR THE GLOBAL MARKETS Globalization Hundreds of million of people have entered the export oriented global markets. Economic capacity and power are moved from West to East. Force changes in economic structure
Political and cultural factors
Debt deleveraging Public and private debt at unsustainable high levels Limits fiscal policy options, curbs public and private spending
The Great Clean Up
The euro straight jacket and moral hazard game
Reduced growth potential of advanced economies High volatility of the capital markets
The ideological cliff in the US politics and electorate Demographical headwind Higher life expectancy post retirement, low fertility rates, graying of the population , higher health care and pension costs
External events Oil-supply shocks caused by tensions in the Middle East New leadership in China
Disequilibrium maintained by heavy handed policy measures. The aim is to reduce short-term tail-risks, but the longterm consequences can be inflated capital markets and higher perceived risk
Trient Asset Management
12
MASSIVE BUILD UP OF CENTRAL BANK RESERVES DISTORTES MARKET PRICING
World Central Bank Reserves and QE Assets 25% Icl. QE CB reserves
% of World GDP
20% 15% 10% 5%
2014
2011
2008
2005
2002
1999
1996
1993
1990
1987
1984
1981
1978
1975
1972
1969
1966
1963
0%
Source: Trient Asset Management
The growth over the last decade in the World’s Central Banks’ asset is unprecedented. Central Banks typically enter the markets to manipulate currencies or to stimulate growth, and not to maximize profit. We would argue that this results in a distorted supply/demand balance in FX and interest rate markets where they operate.
Trient Asset Management
13
DEVELOPED ECONOMIES IN A SERIOUS DEBT TRAP
Historical Public Debt as % of GDP, Advanced Economies
Source: IMF WEO October 2012
There has been a steep rise in public debt following the 2008 financial crises, bringing advanced economies to a debt level above 100% of GDP. An empirical study by Reihart and Rogoff shows that economies with public debt level above 90% historically have seen a 1% drag annually on growth the following 10-20 years.
Trient Asset Management
14
IMF MORE CONCERNED ABOUT THE DOWN-SIDE RISK
Prospects for World GDP Growth, IMF WEO October 2012
Trient Asset Management
15
FOUR US CAPITAL ASSET PRICING REGIMES SINCE 1955
US Equities and Cash Returns 1955 - 2011
Period Performance Equities, Cash and Inflation 1400%
1 0.9
1200%
0.8 0.7
Equities
0.6
800%
0.5 600%
0.4 0.3
400%
0.2
Cash
200%
0.1
Inflation
01/08/2010
01/03/2008
01/10/2005
01/05/2003
01/12/2000
01/07/1998
01/02/1996
01/09/1993
01/04/1991
01/11/1988
01/06/1986
01/01/1984
01/08/1981
01/03/1979
01/10/1976
01/05/1974
01/12/1971
01/07/1969
01/02/1967
01/09/1964
01/04/1962
01/11/1959
0
01/06/1957
0%
01/01/1955
Index Value
1000%
Source: Trient Asset Management
Trient Asset Management
16
US RISK REWARD – A PATTERN WE HAVEN’T SEEN BEFORE
USA risk reward 1963 - 2012, start of each year Fed funds, 10y Govt and SP500 20.0% 18.0% 16.0% 1981
12.0% 10.0% 1988
8.0% 1968
6.0% 4.0%
2007
Sep 12
Sep-11 1963 1998
2009
2.0%
Risk free rate: Fed funds Monthly data Risk measure: Annual Std last 12 months Expected return Bonds: 10y Treasuries Expected return Equities: E/P (Source Earnings – Schiller)
0.0% 0% 2% 4% 6% 8% 10% 12% 14% 16% 18% 20% 22% 24% 26% 28% 30% 32% 34% 36% 38% 40% 42% 44%
Expected return
14.0%
Risk
Trient Asset Management
Trient Asset Management
17
DUALITY IN THE PRICING OF RISK – CAUSED BY GOVERNMENT MARKET INTERVENTIONS
Risk/Reward Duality Across Global Asset Classes
9% 8%
Expected Return
7% 6% 5% 4% 3%
Safe Haven
2%
Equities
1%
Credit
0% -1% 0%
5%
10%
15%
20%
25%
30%
35%
Risk Source: Trient Asset Management
Each point on the chart represents one financial asset, e.g. Bonds of different maturity and creditor, or equities divided among indices. The expected return is estimated from current yields (bonds) and from the historical relationship between risk premium, volatility and risk free rate for equities. Risk is measured as short term realized daily volatility. The chart highlights the dual pricing between typical safe have assets such as AAA-rated sovereign debt and riskier assets such as credit and equities.
Trient Asset Management
18
THE INVERSE RELATION BETWEEN VOLATILITY AND SHORTFALL RISK 1400
Vol last 12 mnd return next 12 mnth MSCI world 1975 - aug 11
Histogram
1200 1000
80%
800
60%
Normal 600
Low
400
High
Return
40% 20%
200 0
0% 0.0 %
10.0 %
20.0 %
30.0 %
40.0 %
-20% -40% -60%
Return Compensation ratios Odds Max Min Average Max/Min Draw down Low 10% 35% -23% 12.6 % 0.11 0.54 92.1 % Normal 80% 70% -52% 5.4 % 0.02 0.10 65.0 % High 10% 65% -27% 18.4 % 0.10 0.67 96.0 %
Risk
All
70%
50%
50%
50%
30%
30%
30%
10%
10%
10%
5%
10%
15%
20%
25%
30%
35%
40%
-10% 0%
5%
10%
15%
20%
25%
30%
35%
40%
Return
70%
-10% 0%
-10% 0%
-30%
-30%
-30%
-50%
-50%
-50%
-70%
-70%
Risk
-52%
7.4 %
0.03
0.14
High risk top 10%
The Normals (Mid 80%)
70%
Return
Return
Low risk, bottom 10%
70%
Risk
-70%
5%
10%
15%
20%
25%
30%
35%
40%
Risk
Trient Asset Management
Trient Asset Management
19
TIME VARIATIONS AT THE LARGEST IN PERIODS WITH LOW OVERALL RETURNS
Variations from trend seem largest and most frequent in periods with low overall returns
Deviations from trend has grown significantly
TVRP opportunities are best and its importance highest in though markets
Periode 1955-70 1970-85 1985-00 2000-12
Trend Return 7.0% 4.4% 12.8% 0.6%
Average Cycle Lenght 1.67 1.36 3.00 2.00
Average Cycle Dev. From Trend 14.9% 16.5% 20.4% 17.1%
Based on counting and measuring absolute deviations from trend larger than 8%. Trend computed using HP filter. Note that the current trend value will be affected by future index performance (not entirely backward looking).
Trient Asset Management
20
2.
REDEFINING THE CONCEPT OF RISK
VOLATILITY IS A DUBIOUS MEASURE OF RISK
Short horizon: Volatility is the normal indicator. High volatility signals high risk
Long horizon: The relevance of volatility goes often the other way around. Volatility may indicate high expected risk premiums and good buying opportunities
Low volatility may signal elevated valuations and low expected risk premiums
Key risk management challenges for long-horizon investors: Mastering volatility. Avoiding procyclical investment behavior Build alternative long-term indicators of risk Ensure alignment of interest between asset owner and manager
Trient Asset Management
22
RISK AND UNCERTAINTY
One definition of risk management is controlling left-hand tails. Risk is the probability of tail events given that the distributions are known. The distributions are certainly non-normal and change over time, but they are known.
Uncertainty is that the distributions are unknown and perhaps the full range of distributions cannot be known, at least in finite samples.
Risk managers have usually been pretty good at risk, but a lot of the financial crisis has stemmed from Knightian uncertainty – or unknowable distributions ex ante. Many risk managers have never used normal distributions, at least not in the last 10+ years.
A central feature of the financial crisis is impact of the network effects; factors that are not being measured and the full range of distributions not being known.
[Any given] “instance” … is so entirely unique that there are no others or not a sufficient number to make it possible to tabulate enough like it to form a basis for any inference of value about any real probability in the case we are interested in. The same obviously applies to the most of conduct and not to business decisions alone. Frank Knight, 1921, quoted from Bernstein, 1996
By “uncertain” knowledge … I do not mean merely to distinguish what is known for certain from what is only probable. The game of roulette is not subject, in this sense, to uncertainty ... The sense in which I am using the term is that in which the prospect of a European war is uncertain, or the price of copper and the rate of interest twenty years hence, or the obsolescence of a new invention … About these matters there is no scientific basis on which to form any calculable probability whatever. We simply do not know! John M. Keynes, 1937, response to criticism of The General Theory, quoted from Bernstein, 1996
Even with more knowledge on the nature of financial markets – and improvements in how we monitor and assess risk, investors will continue to face and live with uncertainty
Trient Asset Management
23
THE FALLACY OF VOLATILITY AS MEASURE OF RISK
“Risk appears to be at its greatest when measures of it are at their lowest” Quote from Mark Carney, Governor of the Bank of Canada, at the 2009 Jackson Hole conference
“The received wisdom is that risk increases in the recessions and falls in booms. In contrast, it may be more helpful to think of risk as increasing during upswings, as financial imbalances build up, and materializing in recessions.” By Andrew Crockett in 2000, quoted from Hyun Song Shin, Risk and Liquidity, 2009
CONVERGENCE OF LONG-HORISION RISK MEASURES AND VALUE INVESTING PRINCIPLES
Graham Risk Measures: Inspired by Benjamin Graham’s definition of investment risk as the difference between the intrinsic value of an asset and the price paid for it (“The Intelligent Investor”, 1949) *)
Seth Klarman: “Risk is a function of price paid.” (value investor creed; you get more “margin of safety” when you buy cheaply)
This brings risk management and asset management strategy together
Lowering risk by buying assets with a safety of margin to long-term fair value equals risk management strategy with value investment strategy
But we shouldn’t throw away volatility as a meaningful measure of noise and a possible indicator of regime shifts *) Graham Risk examples are given at www.grahamrisk.com
Trient Asset Management
24
RISK MEASURED BY VOLATILITY AND VALUE DOES OFTEN NOT COINCIDE
Z-score 'Volatility Stress' and US Equity Value Model Volatility Stress: 1972-1999 in sample , 1999-> out of sample; US Value Models: 1911-2011 4 Volatility Stress
US Value Model
3 2
High perceived risk (realized volatility) and low/reasonable valuation risk.
1 0
Low perceived risk (realized volatility) and high valuation risk.
-1 -2
2012
2011
2009
2008
2006
2005
2003
2002
2000
1999
1997
1996
1994
1993
1991
1990
1988
1987
1985
1984
1982
1981
1979
1978
1976
1975
1973
1972
-3
US Equity Value Model: Combination of Shiller cyclically adjusted PE, Tobin Q and Price/book
Source: Trient Asset Management Trient Asset Management
25
MULTIDIMENSIONAL RISK DASHBOARD (INDICATIVE) October-12
Overall
US
Europe
Norway
Risk premiums Growth risk premium Risk free rate (official) Term Premium Credit premium Fundamental Values Equities Offical real rates Debt ratio governments Debt ratio housholds Debt ratio corporates Macro Environment Leading Indicators Corporate earnings share of GDP House Prices Private Savings Inflation Risk Deflation Risk Imbalances Expected Return Equities Bonds Credit
Trient Asset Management
Trient Asset Management
26
3.
THE CARDINAL INVESTOR MISTAKES
THE TWO CARDINAL INVESTMENT MISTAKES
The biggest mistake in investments is procyclicality:
Buy when risk feels low, at high prices
Sell when the market panics, at low prices
Even large institutional investors all over the world made this cardinal mistake in 2008-2009
The second most important mistake is letting agents control the principals (risk owner)
Paper co-written with professor Andrew Ang at Columbia: “Investing for the long run” http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1976310
Trient Asset Management
28
PROCYCLICALITY IN HIRING AND FIRING OF EXTERNAL MANAGERS
Source: Charles D. Ellis, Murder on the Orient Express: The Mystery and the Underperformance, Financial Analyst Journal, Volume 68, Number 4, CFA Institute, 2012
Trient Asset Management
29
SHORT-TERMISM: DOING THE SAME THINGS, TAKING UP CORRELATIONS AND BETA EXPOSURES
Bottom up strategies and short-sighted alpha vehicles drives normally only 5 – 20 percent of a portfolio expected risk and return Short-term alpha space seems overcrowded. Correlation between hedge fund return is increasing, and so is also the correlation to beta factors Typical institutional investors spend 80-90 percent of their time on implementation issues, building complexity, rather than on strategic and ALM issues that normally determine 80–90 percent of expected risk and return – a paradox The graph is from “Third Generation Asset Allocation”, Deutsche Bank Asia Investor Letter 21 Nov 2011, by Brad Jones. A good report!
Trient Asset Management
30
MAYBE MISSING THE BIG PICTURE?
Investors typically build complex portfolios to diversify risk. Huge allocations to “alternatives” and advanced instruments have blurred the true risk characteristics of portfolios and imposed unnecessary high risk and costs. Diversification didn’t work because complexity hid the underlying drivers of returns – which turned out to be the same
Trient Asset Management
31
OVERPAYING FOR DIVERSIFICATION?
New York Times 12 October 2012 University Endowments Face a Hard Landing
Trient Asset Management
32
4. RISK OWNERSHIP AND FUND GOVERNANCE
PRIORITIES FOR LONG-TERM INVESTORS
Importance for expected risk and return
Harvest risk premiums
Exploit time variations in risk premiums
Pursue fundamentally based active management (idiosyncratic risk)
Trient Asset Management
34
PRIORITIES FOR LONG-TERM INVESTORS
Time and resources spent
Harvest risk premiums
Exploit time variations in risk premiums
Pursue fundamentally based active management (idiosyncratic risk)
Trient Asset Management
35
THE QUALITY OF THE GOVERNANCE STRUCTURE SETS THE APPROPRIATE AMBITION LEVEL
Governance How well is the risk ownership defined and risk taking anchored? What ability and willingness has the risk owner to build and encourage investment professionalism? How well are the principal and agent aligned?
Accountability Precondition for confidence and risk taking, and for building professionalism
Trient Asset Management
36
BUILDING THE GOVERNANCE STRUCTURE The risk ownership Creating a framework for delegating the operative management to professionals:
Expressing the risk appetite Setting measurement standards and performance requirements Building a system for oversight, risk control and active communication
In practice – defining an asset class based benchmark (60/40) or policy portfolio
That should represent a low cost way of achieving the Fund’s objectives and should be investable (in liquid assets)
It should ideally be efficient, but that concern must be balanced against the need for accountability (not high degree of complexity)
It must be owned (and understood) by the risk owner (the entity that may impose a change in investment appetite and in the strategy)
Key issues:
How to deal with illiquid assets?
How to exploit the time varying nature of expected risk premiums?
Trient Asset Management
37
5.
DEALING WITH HIGH VOLATILITY AND REAL UNCERTAINTY: BUILDING BLOCKS OF DYNAMIC ALLOCATION OF RISK
INGREDIENTS OF A DYNAMIC STRATEGY
Two main components
Governance structure Well defined risk ownership, clarity in roles and responsibilities Decision discipline Institutionalizing of contrarian behavior: Rebalancing Increasing principal – agent alignment, reducing agency costs
Active strategies for assessing current and future pricing of risk Tail risk monitoring, scenario-building, early warning indicators Identifying of possible changes in regimes Monitoring of current asset pricing against their long-term fair values Building a multidimensional risk dashboard (more than volatility measures!)
Trient Asset Management
39
DYNAMIC ALLOCATION OF RISK IS IN THE GREY AREA BETWEEN PRINSIPAL AND AGENT
Importance for expected risk and return
Harvest risk premiums
Principal
Exploit time variations in risk premiums
Pursue fundamentally based active management (idiosyncratic risk)
Trient Asset Management
Grey area
Agent
40
“GREY AREA” WARNINGS
Concentrated and few bets, potentially high impact (you can fail spectacularly)
The investment process must be very disciplined. Risk taking must be anchored by the principal / risk owner
Requires a medium to long time horizon: Can the ultimate risk owner stay consistent? Risk of procyclical investment behavior
Potential agency conflicts (manager versus risk owner)
Decisions in the “grey zone” between manager and board: Lack of accountability? The division of responsibilities must be crystal clear Group think? Is decision-power matched with skills? The principal must refrain from taking investment decisions
Many investors will be better off with a mechanical rebalancing system and no leeway for active attempts to exploit the time variations of expected risk premiums
Trient Asset Management
41
TAIL RISK MONITORING
“Fiscal cliff” in the US Automatic contraction of the budget equivalent to 4 – 5 percent of GDP
The euro crisis
China hard landing?
Oil market supply side chock
Heavy handed government and central bank interventions in the market place
Geopolitical risk
Trient Asset Management
42
BUILDING GLOBAL MACRO SCENARIOS World GDP Growth; PPP Weights, Volume
Global GDP Growth, Y/Y %, Volume, PPP Weights
6 5 4 3 2 1 0 -1
IMF Trient China hard landing Implied from Realrate market
Trient mean Trient US led growth Implied from Equity market
Trient Euro bust Trient Iran Nuclear
Source: Trient Asset Management, IMF We estimate a lower growth path than the IMF, but higher than the growth we estimate implied by equity and real rate markets. Note that none of our scenarios matches the low growth implied by the real rate market.
Trient Asset Management
43
THE RE-PRICING OF RISK
Market Implied Pricing and Trient Forecast of Global GDP Growth 2013
Trient Growth Forecast 2013 Equities Implied growth Rates Implied growth
6
Global GDP Growt %
5 4 3 2 1 0 -1 -2 Sep'11
Jan'12
April'12
July'12
Now
Source: Trient Asset Management Trient’s 2013 world GDP growth expectation has during 2012 seen a marginal decline, however at high levels of 3.5% -4.0%. From both fixed income and equity markets we do however estimate significantly lower implied growth. For equities the implied growth estimate has increased significantly over the year, driven by increased prices and reduced earnings forecasts.
Trient Asset Management
44
LONG-TERM FAIR VALUE ANALYSIS, US EQUITIES EXAMPLE
Trient Asset Management
45
THE LINK BETWEEN RETURN REGIMES AND LONG-TERM FAIR VALUE & RISK MEASURES US Period Regimes Performance
S&P 500: Long Term Value Indicators
Inflation, Cash, Earnings S&P 500 and SP 500 Total Return, 15 yr periods starting 1955
Z-Score on ln values of Price/Cyclically adjusted Earnings, Price/Book and Tobin's Q
1400%
3
SP 500 tr 10Y Gov. Bonds CPI Earnings
1200% 1000% Index Change
1
800% 600% 400%
0
200%
-2
Jun-14
Apr-09
Nov-11
Sep-06
Jul-01
Feb-04
Dec-98
Oct-93
May-96
Mar-91
Aug-88
Jan-86
Jun-83
Nov-80
Apr-78
Sep-75
Feb-73
Jul-70
Dec-67
May-65
Oct-62
Mar-60
Aug-57
Jan-55
-3
The differences in medium term equity return seem to be more related to initial valuation than underlying fundamentals (earnings and GDP growth) Source: Trient Asset Management Trient Asset Management
Jan-10
Jan-05
Jan-00
Jan-95
Jan-90
Jan-85
Jan-80
Jan-75
Jan-70
Jan-65
-1
Jan-60
0%
Jan-55
Standard deviation from mean
2
Our indicator for long-term fair value (combined Shiller PE, price-book and Tobins Q), that we also call an indicator of risk for long-term equity investors, seems to predict the medium term equity returns quite well. When we start a period with valuations significantly below fair value (we call it low risk), as in 1980, the equity return over the following period is very good. Not because of higher growth in cash-flows, but because the capital market price the cash-flows differently. The various regimes of equity return are related to differences in valuation, and that is showing a cyclical pattern. This is just an example and don’t prove anything. But we think it illustrates that the starting valuation can be a good indicator for future return of equities. However, this information is only helpful if the investor can apply a medium to long-term horizon and stay consistent over time. Extreme valuations have obviously more information value than moderate deviations from normal
46
CREDIT SPREAD TIME VARIATIONS
Source: “Re-thinking Asset Allocation – The Role of Risk Factor Diversification. Bradley A. Jones, Deutsche Bank, September 2011
Trient Asset Management
47
ASSESSING CURRENT INTEREST RATES AGAINST “TAYLOR BANDS”
Rates relative to Taylor Range Reveal Current Monetary Policies
8%
8%
7% 6%
6% 5%
4%
4% 3%
2%
2% 1% 0%
0% NOK
USA
SEK
GBP
AUD
CAD
KRW
MXN
BRL
CHL
EUR
Source: Trient Asset Management
The lines indicate Taylor bands. Central Bank rates (blue dots) outside the Taylor range indicate a monetary policy that differs from the economic environment.
Trient Asset Management
48
THE EURO RATE DOESN’T FIT ALL COUNTRIES Examining euro-zone monetary policy stance, taking into account Spanish and Italian domestic rates 5%
5%
4%
4%
3%
3%
2%
2%
1%
1%
0%
0%
-1%
-1% -2%
-2% EUR
FRA
ITA dom
SPA dom
GER
Source: Trient Asset Management
Spain and Italy needs much more stimulus, Germany if anything needs less stimulus
Trient Asset Management
49
GERMAN RATES AN OUTLIER
Examining Interest Rate Policy Stance With a Taylor Approach Exogenous changes:
16
Too weak currencies
Exogenous changes
14
- USD in ’73 after the Vietnam War and Bretton Woods collapse, - USD in ’81 when Fed (Volcker) started fighting inflation - DEM ’90 - uncertainty after the reunion
Too weak currencies: NOK in ’86 and ’92, SEK ’92, UK’92
12 Taylorrange
Actual rate
10
Too strong currencies: Japan in ’85 and ’95
Rate higher than range
8 6
Rate lower than range
4
Too strong currencies 2
Germany in EUR 2012
0
Extraordinary 2012
-2 0
2
4
6
8
10
12
Taylor rate
Source: Trient Asset Management
Trient Asset Management
50
PRICE FILTERING – IDENTIFYING EXTREME VALUATIONS
World US Eurozone Japan UK Sweden Canada Australia Norway Korea Mexico Brazil Germany US BBB US HY Euro HY Scandi HY USA Euro Japan UK Switzerland Sweden Norway Australia Canada China Korea Mexico Brazil Chile USA 5 USA 10 USA 30 Euro 5 Euro 10 Euro 30 UK 5 UK 10 UK 30 SW 5 SW 10 SW 30 NO 5 NO 10 CA 5 CA 10 CA 30 AU 5 AU 10 AU 30
BUY
z score
SELL
Global Position Price Filtering October 2012
Equities
Credit
FX
Rates
Source: Trient Asset Management The price map indicates that equities are attractively priced whereas interest rates seem expensive. The blue dots mark the difference between the observed price and the estimated fair value.
Trient Asset Management
51
ADAPTING TO THE VOLATILE WORLD, PRINCIPAL – AGENT ALIGNMENT
Short time horizon Short-term alpha, A crowded space
Medium time horizon The grey area
Long horizon Long-term strategic weights
Agent / Asset manager(s)
Principal / Asset owner
Investment competence
Risk ownership
Protects own business risk
Too often ad hoc decisions Lacks closeness to markets and skills to assess current pricing of risk
Comfort zone: Measurement against benchmarks
Often lacks investment culture and discipline
Trient Asset Management
52
ADAPTING TO THE VOLATILE WORLD, PRINCIPAL – AGENT ALIGNMENT
Short time horizon Short-term alpha, A crowded space
Medium time horizon The grey area
Long horizon Long-term strategic weights
Agent / Asset manager(s)
Principal / Asset owner
Investment competence
Risk ownership
Keeping manager accountable Delegation of investment decisions
Disciplined investment process Totally honest communication to the principal Proactively sharing of all information
Shared investment beliefs Formalized and disciplined interaction Contingency plans, fire drills
Trient Asset Management
Provides trust and confidence Avoids interfering with the investment decisions
53
HEDGING STRATEGIES
54
TWO TYPES OF HEDGING STRATEGIES
Strategic hedging ALM framework. Interest rate risk, inflation risk, currency risk
Tactical hedging Tail risks
Trient Asset Management
STRATEGIC HEDGING – INTEREST RATE RISK
Hedging the coverage ratio against further decline in the 30y swap rate…
Hedging a significant part of the interest rate risk is normally prudent ALM
This is about hedging future pension payments by buying long-term safe cash-flows from government bonds
But is this the case today? Very low cash-low How safe are government bonds?
Strategic hedging is only meaningful in an equilibrium, non bubble environment
(We should also take into account the operational and liquidity aspects of the current derivative transactions to hedge against decline in the 30y swap rate)
Trient Asset Management
56
GERMAN 10Y AND 30Y YIELD AND EURO 30 YEARS SWAP RATE
Macrobond, Trient Asset Management
Trient Asset Management
GERMAN EARNINGS YIELD VS GOVERNMENT BOND YIELD, REAL TERMS. WHERE IS THE CASH FLOW?
Trient Asset Management
58
TACTICAL HEDGING
Long-term investors are natural sellers of insurance
Option strategies are normally very costly
There are cases where buying tail risk insurance can make sense
Such strategies should be decided within the dynamic risk allocation framework outlined in this presentation
An alternative to options for creating an asymmetric return distribution is to apply an “extreme value” approach: Minimizing downside risk by always having in mind current valuations against long-term fair value
Trient Asset Management
59
APPENDIX BUILDING A GOVERNANCE STRUCTURE FOR ADAPTING TO THE VOLATILE ENVIRONMENT
60
CHOICE OF SYSTEM FOR CONTRARIAN INVESTMENTS
(A) is the natural starting point for long-term investors: Get the best possible risk factor based diversification, state clear objectives and investment beliefs, define risk ownership, build formal decision rules including rebalancing regime (B) is the next step. It may form an important reference for investors that have ambitions to pursue discretionary dynamic decisions (C) Valuation metrics are vulnerable to structural shifts However, those shifts do not happen often, and it is very hard to get the timing right Exercising (C) should be within the same structure of discipline as recommended for (A) and (B) The average investor holds the market weights. Only long-term investors that pass suitability tests should have ambitions to follow contrarian strategies (as above). (B) and (C) are more contrarian and demanding than (A)
Trient Asset Management
61
DYNAMIC STRATEGIC ALLOCATION OF RISK
We are in the “age of time varying expected risk premiums”
An investor can probably add significant value by applying a more dynamic allocation of risk than just pursuing mechanical rebalancing strategies based on a fixed policy portfolio
It is a paradox that even asset prices seem to be more predictable over a medium horizon, many long-term investors continue to prioritize hunting alpha in the crowded short term space rather than focus on the top-down risk factor management of the portfolio, that normally contributes to 80-90 percent of expected risk and return
This is probably due to agency issues, but also bad experience with TAA type of decisions, taken on ad hoc basis without the guidance of discipline, structure and professionalism
The agency problem with dynamic strategic allocation of risk:
The Board must own the medium and long horizon risk taking. Often boards seems to be involved in investment decisions. But they lack the professionalism and acts like a committee, that often means lagging the market and invest with and not against the cycle
The manager likes to be in the comfort zone, close to the policy portfolio, and may be happy to put investment decisions on the Board’s table
The outcome may be blurring of responsibilities and again – procyclical investment behavior
Trient Asset Management
62
STEPS TO BUILD A STRUCTURED APPROACH TO MEDIUM-HORIZON ACTIVE MANAGEMENT Principal and agent alignment (Board/ risk owner and internal management company)
Shared sense of mission and purpose Shared investment beliefs; why risk taking is rewarded, the investor’s competitive edge and how to exploit that, etc
Clarity in roles and responsibility Well defined risk limits, measurement criteria and time horizon for measurement, supported by an appropriate reward system Accountability
Manager skills and business culture
Skills and structural competence for medium term position taking: Scenario building, regime shift analysis, early warning indicators, long-term fair value indicators, building of multidimensional risk dashboards
Ongoing exposure to market movements, intuitive and practical understanding of how market price risk, Deep knowledge about financial instruments and optionality
Ability to express market views while controlling downside risk
Business culture, willingness to take on risk. Talent based leadership. Meritocracy
Information privileges, ability to tap into networks, e.g. applying external manager skills
Balancing empowerment of individual talents with team approach – avoiding group think
Trient Asset Management
63
STEPS TO BUILD A STRUCTURED APPROACH TO MEDIUM-HORIZON ACTIVE MANAGEMENT (2)
Board
Sets risk appetite and revisit frequently. Oversees, evaluates and controls the Manager
Set return expectation and risk limits, defines measurement horizon for the various active strategies
Articulates investment beliefs, shares with the manager
Must have investment competence, willingness to take risk and stay consistent over time
Ability to delegate to the manager and not interfere with the manager’s investment decision
Competence and resources to assess the quality of the manager
Trient Asset Management
64
STEPS TO BUILD A STRUCTURED APPROACH TO MEDIUM-HORIZON ACTIVE MANAGEMENT (3)
Board and manager interaction
Open and honest communication from the manager to the board on investment philosophy, key risk drivers of the portfolio, performance and risk attribution, status of the talent pool, etc
The manager should have discretion to run an overlay portfolio with significant risk (the board can eventually set a minimum risk level to take the manager out of the “comfort zone”). The Board can then, based on ongoing communication with the Manager and understanding of the reasons for the position taking, adjust the risk mandate to the Manager or make changes directly in the strategic portfolio
This model will inspire the manager to build the senior team and procedures needed to exploit medium term opportunities. The board will get valuable information by seeing what the manager actual is doing.
A precondition for risk taking is mutual trust between risk owner and manager. The board must be able to delegate, not interfere in investment decisions, stay consistent over time, trust and back the manager. The manager must actively inspire trust with the board by communicating honestly and beyond expectations
Trient Asset Management
65
STEPS TO BUILD A STRUCTURED APPROACH TO MEDIUM-HORIZON ACTIVE MANAGEMENT (4) Structure and Discipline
Formalized investment beliefs
Managers investment philosophy and framework for taking medium term active decisions as well as risk management procedures is articulated in writing and shared with the board. Updated frequently
Predefined agenda for board meetings. Use of standard, multidimensional dashboards to asses portfolio status against predefined expectations and risk parameters
Frequent updates on scenarios, supplemented by status on early warning indicators and regime shift analysis
“Fire drills” on reaction to high impact events
Formal rebalancing procedures
In case of deviations from the formal procedures – always establishing decision rules on how to act when market situation
Trient Asset Management
66
TRIENT ASSET MANAGEMENT AND TRIENT GLOBAL MACRO FUND ◆
Trient Asset Management AS is a newly established company based in Oslo, Norway. Trient is a partnership capitalized by the senior partners who have a significant portion of their net worth invested in the company and the Fund.
◆
Trient Global Macro Fund (“TGM” or “Fund”) is a discretionary global macro fund utilizing fundamental research. The Fund targets opportunities where asset prices deviate materially from their long-term fair values and tactically exploits the time-varying nature of expected risk premiums.
◆
Liquid market instruments are used in equities, fixed income, credit and foreign exchange.
◆
TGM is managed by Dag Loetveit, who has 20+ years of macro investing experience. Previously, Mr. Loetveit served as Chief Investment Officer of Fixed Income, and subsequently, as Global Head of Allocation Strategies at Norges Bank Investment Management (“NBIM”). At NBIM, Mr. Loetveit managed a long/short macro fund from 2008-2011 using the same methodology that will be employed at TGM.
◆
Mr. Loetveit is supported by a seasoned team who have shared professional experience, including Knut Kjaer, founding CEO of NBIM, and Tore Andre Lysebo, Head of Global Value Strategies for Fixed Income and Foreign Exchange at NBIM.
◆
The portfolio will be transparent and TGM will share its analytics and research with investors.
◆
TAM has formed a strategic partnership with Tiger Management L.L.C. and Julian Robertson to assist with establishing a world class investment management platform.
The Trient Plateau Photo by Erik Syvertsen, March 2010
Trient Asset Management
67
KNUT N. KJAER, FOUNDING PARTNER AND CHAIRMAN TRIENT ASSET MANAGEMENT
Mr. Kjaer, co-founding partner and Chairman of Trient Asset Management in Oslo, has many years of experience in the asset management business. Prior to establishing Trient, Mr. Kjaer was President of RiskMetrics Group in New York, with direct responsibility for the firm’s global risk management and corporate governance advisory businesses. In 1998, Mr. Kjaer became founding Chief Executive Officer of NBIM, which was established by the Norwegian government to manage its sovereign wealth fund and foreign exchange reserves. Over his 10 year career at NBIM (1998-2008), Mr. Kjaer implemented a successful investment program and built an asset management team that grew NBIM’s reserves from $25 billion to approximately $400 billion USD.
Mr. Kjaer is a member of ABP’s investment committee, which manages public and private Dutch pension funds, and is the largest pension fund in Europe; serves on the International Advisory Board of China Investment Corporation; and is member of the Investment and Risk Advisory Panel of the Monetary Authority of Singapore. He is also a member of the commission that manages the Irish National Pensions Reserve Fund and serves as Chairman of the Nordic private equity company FSN Capital Partners. In 2008, Mr. Kjaer was appointed a Knight Class 1 of the Royal Order of St. Olav by His Majesty King Harald of Norway for his management of the Norwegian sovereign wealth fund and his contribution to the Norwegian economy and the future financing of its pensions.
Mr. Kjaer holds a master’s degree in Economics and a degree in political science from the University of Oslo (1982). He attended the Advanced Management Program at Harvard Business School (2003).
Trient Asset Management
68
DISCLAIMER This document is intended for use only by those professional investors to whom it is made available by Trient Asset Management AS and no part of this report may be reproduced in any manner, or used other than as intended, without the prior written permission of Trient Asset Management AS. The information contained in this document has been taken from sources deemed to be reliable. We do not represent that such information is accurate or complete and it should not be relied on as such. Any opinions expresses herein reflect our judgement at this date and are subject to change. Trient Asset Management AS accept no liability whatsoever for any direct, indirect or consequential loss rising from the use of this document or its contents. Trient Asset Management and/or their employees may hold shares, options or other securities of any issuer referred to in this report and may, as principal or agent, buy or sell such securities. This document does not constitute or form part of any offer for sale or subscription of or solicitation or invitation of any offer to buy or subscribe for any securities nor shall it or any part of it form the basis of or be relied on in connection with any contract or commitment whatsoever.
Knut N Kjaer
[email protected] TRIENT ASSET MANAGEMENT AS
Kanalen 5,Tjuvholmen N-0252 Oslo, Norway
Phone: +47 22 39 88 80 Fax: +47 22 39 88 81
Org.no: NO 996 906 191 69
