SPECIAL EDITION 2013 BEYOND MODERN PORTFOLIO THEORY: ASSET ALLOCATION bY RISK FACTOR

FOREWORD
John Maynard Keynes wrote in the preface to his seminal work The General Theory of Employment, Interest, and Money, published in 1936, that the difficulty lies, not in the new ideas, but in escaping from the old ones. The systemic crisis shaking the global financial edifice to its foundations for over four years now has led economic, political and financial policymakers to explore, sometimes against their better judgement, completely new avenues as they look for solutions to problems that have ramifications stretching way beyond any that have had to be coped with since the Great Depression. New approaches to monetary policy, such as quantitative easing or nominal GDP targeting, are still in their early days. As classic Keynesian reflationary strategies have become inoperable as government have run up mountains of debt, chances may have to be taken with new ideas to revitalise flagging economies, boosting competitiveness by focusing more intently on innovation and entrepreneurship. Upheavals on financial markets, coupled with a structural tightening of correlations between asset classes, call for innovative concepts to improve asset allocation. With this purpose in mind, Pictet Wealth Management adopted a pragmatic and realistic approach, striving to pinpoint the most likely ways in which the changed economic and financial world of today will have to innovate to kick-start a fresh cycle of growth. Our approach has been geared to identifying financial tools best suited to delivering the most effective possible combination for generating returns and controlling risk in clients portfolios. As part of our ongoing process of blending the most progressive, state-of-the-art academic research with the best practice of recognised investment practitioners, we held the annual seminar of our Strategy Committee on 8 October on the shores of Lake Geneva. This brought together our in-house investment experts with outside specialists world renowned in their fields of expertise. In this, the second, annual issue of Perspectives Special Edition, we have presented the most relevant findings that emerged from this seminar. The document is divided into two sections. The first outlines how we have strengthened the process we use to allocate assets in portfolios by integrating the pioneering approach of making selections based on risk factors. In this respect, we have focused primarily here on presenting long-term implications for our strategic allocation, i.e. for an investment horizon of ten years. The second part has five articles on the main topics covered by our external contributors. The subjects discussed range from innovation in monetary policy to the latest developments in capturing risk. In addition, we have provided an insight into a topic experiencing rapid development: behavioural finance. We trust this second edition of our annual Perspectives Special Edition will provide much food for thought, maybe even answering some of the questions you might have about financial and macroeconomic conditions that are in such a state of flux at present, demanding ever more advanced and up-to-date know-how. To close, I would just like to take advantage of this opportunity to pass on to you all my best wishes for 2013. Philippe Bertherat Partner, Pictet & Cie

PERSpECTIVES - SPECIAL EDITION 2013

CONTENTS
23 editorial - sECULAR OUTLOOK THE DELEVERAGING CYCLE IS ADVANCING IN DIFFERENt WAYS Yves Bonzon, Chief Investment Ofcer, Pictet & Cie

46 79

tHe influence of tHe uniMaGinaBlE innoVation on returns

Christophe Donay, Head of Asset Allocation & Macro Research, Pictet & Cie INNoVAtIoN SHoCK: tHE UNImAGINAbLE IS poSSIbLE TACtICAL ASSEt ALLoCAtIoN oN tHE bASIS oF RISK FACtoRS

1013 EXPECTED RETURNS FROM ASSET CLASSES

central Banks

1416 MONETARY POLICY IS EVOLVING Frederic Mishkin, Former Member of the Board of Governors of the Federal Reserve System Alfred Lerner Professor of Banking and Financial Institutions Graduate School of Business, Columbia University

THe GloBal econoMY

2022 A JOURNEY INTO THE UNKNOWN Jim Reid, Managing Director Global Head of the Fundamental Credit Strategy Group, Deutsche Bank

BEHAVIOURAL FINANCE AND BEYOND

2426 PERCEPTIONS, BELIEFS AND ACTIONS Olivier Oullier, Professor of behavioural and brain sciences, Aix-Marseille University Strategic adviser to governments and private institutions

efficient asset allocation

2829 Viewpoints with Antti Ilmanen, Managing Director, AQR Capital Management (Europe) LLP

inVestMent ManaGeMent

3031 Viewpoints with Yves Choueifaty, President and Chief Investment Ofcer, TOBAM

Editorial - secular outlook

With the decision of the Federal Reserve to continue its policy of asset purchases (QE) as long as US employment remains depressed, we can say that ination targeting as a tool of monetary policy, introduced in the early 80s under Paul Volcker, has nally been buried.
THe deleVeraGinG cYcle is adVancinG in different waYs
Yves Bonzon Chief Investment Ofcer, Pictet & Cie

Four years after the collapse of Lehman Brothers, the likely economic and financial trends that we described for the first time in our Secular Outlook in late 2009 have begun to emerge more clearly. We said then that the bursting of a credit bubble gives rise to the structural risk of deflation in an economy, if excess debt is liquidated in a disorderly manner. Faced with such a possibility, governments respond more or less skilfully by employing one or more of three approaches: 1) transfers from creditors to debtors through financial repression; 2) debt restructuring (by rescheduling and writing off); and 3) printing money (monetisation). The first two approaches are deflationary and the third potentially inflationary. The challenge is to apply these three remedies in the right proportions, so as to offset deflationary forces without tipping into hyperinflation. In all cases, the state transfers wealth from creditors to debtors. A major consequence for investors of transferring wealth by these means is that safe assets are valued at levels implying negative returns after inflation, as Christophe Donay explains on page 13 of this publication. The negative real yield on US Treasury bonds on maturities up to 20 years is a symptom of depression and classic financial repression. Meanwhile, assets perceived as more risky, in most cases real assets, and in particular stocks, are valued at levels implying positive returns after inflation. However, macroeconomic risks will remain very high as long as the deleveraging cycle is incomplete. During this phase, the volatility of real assets will be very high, to the point where few investors are likely to have the resilience to absorb the ups and especially the downs that they encounter.

PERSpECTIVES - SPECIAL EDITION 2013

In the United States, the cycle of private-sector deleveraging is nearing its end and the coming years will see the public sector follow suit. In Europe, there is still a long way to go. Eventually, the mutualisation of debt seems highly probable. Although the Germans are aware that this outcome is inevitable, they are eager to delay it as long as possible until the public-sector reforms of their European partners are well entrenched. This transitional phase, under which the mechanisms and institutions to achieve real economic and political union are gradually put in place, means a continuing risk of deflation in Europe. It is a gentle deflation scenario, rather like the fate of Japan in the past 15 years. With the decision of the Federal Reserve to continue its policy of asset purchases (QE) as long as US employment remains depressed, we can say that inflation targeting as a tool of monetary policy, introduced in the early 80s under Paul Volcker, has finally been buried. Central banks are now moving towards a policy of targeting asset prices and other economic variables, primarily nominal GDP. The consequences of this monetarist revolution on asset price formation are difficult to assess. However, we cannot overemphasise the potential disruption to the correlation and volatility regimes to which investors have become accustomed. In such conditions, proven investment strategies may prove obsolete. More than ever, investors will need to be able to challenge and fight against preconceived ideas. Lastly, and fundamentally, it is to be hoped that the policy of quantitative easing (QE) does not last too long, because, ultimately, it could lead to a massive distortion in the allocation of capital. Given this background, we have decided to introduce an official allocation to unlisted assets in our portfolios from 2013. This is an area where Pictet & Cie has been active for more than 25 years. These markets offer the opportunity for experienced players to distinguish themselves by their access to proprietary deal flows and their ability to use more complete information than available on the heavily regulated public markets. These advantages allow us to increase the exposure of portfolios to equities and real estate, without proportionately increasing the volatility associated with short-term market fluctuations. This will certainly be a great advantage in withstanding the financial shocks that may lie ahead. We expect returns of between 6% and 8% for private-equity real estate and more than 10% for private-equity. As for gold, we remain positive until the middle of this decade, but we consider it to be in the third and final stage of the bull market that began in 2002. This phase will be more volatile and less profitable than the one we have been in for the past 10 years. In equity markets, quality defensive stocks form the backbone of our portfolios. Peripheral European markets contain value, but their potential is unlikely to materialise as long as these economies try to rebalance by internal deflation in prices and wages alone. The dramatic rise in industrial commodity prices, triggered by Chinas accession to the World Trade Organisation, seems complete. For this reason, we have decided to withdraw from the oil services sector, which, until now, was among our secular themes.

SECULAR TRENDS
1960 Bretton Woods ECONOMIC ENVIRONMENT 1970 Floating FX Oil shock Ination 1980 Disination Plaza Arbitrage 1990 Fall of Berlin Wall Globalisation Internet E-trading 2000 EMU Great global imbalance Chinas rise Structured credit 2010 Managed Western deleveraging Nominal GDP and asset price targeting EM discrimination EUR 2.0

Every decade is characterised by a different economic and investing environment

US Nifty Fifty stocks INVESTING ENVIRONMENT

Small caps Oil stocks Gold, CHF and JPY

Gvt. bonds Nikkei Hang Seng

Indexing Nasdaq SMI USD

Hedge funds EM equities Commodities EUR

Risk-factor-based TAA and SAA Developed quality blue chips Gold (until 2015) EM debt

tHe influence of tHe uniMaGinaBlE innoVation on returns

Christophe Donay
Christophe Donay, Head of Asset Allocation & Macro Research, Pictet & Cie

De la Grande Modration la Grande DiVerGence: facteur de Volatilit dans les MarcHs financiers

InnoVation sHock: tHe uniMaGinaBle is possiBle

Innovation a word that is frequently bandied about and a phenomenon that is often underestimated and ultimately poorly understood. Mankind has the natural tendency to regard technical or economic hurdles placed in its path as unconquerable. Worse yet, such obstacles are viewed as only capable of leading to an inevitable downturn or even disaster. That is the pessimistic version of the history of mankind. The amazing progress made over the last few centuries should teach us that the boundaries and barriers we are confronted by today will be pushed further back in the future. Innovation plays a crucial role in helping us to reach an understanding of the long-term dynamics driving economies. In recent centuries, economies have expanded thanks to significant breakthroughs on the innovation front. These have forced back the bounds of the unimaginable. This limitation/innovation process generates much greater dynamism when the economic climate encourages risk-taking. Innovation is essential for understanding clearly the trajectories economies in the developed world might take over the next ten years. Innovation also has a magical quality in that it can conjure up growth drivers for economies where they do not appear to exist. This leads us to the conclusion that sustainably strong growth cannot be generated without innovations. The current

sluggish state of economic growth in the developed world could be largely attributable to the absence of any significant innovations at present. Innovation can take two main forms In macroeconomic analysis, innovation can take one of two forms: transient or radical. Transient innovation will tend to benefit one particular sector or its boost to growth will only be short-lived. For instance, the arrival of the iPhone or iPad has benefited Apple and its sub-contractors. By no stretch of the imagination has this particular innovation kick-started growth in the US or significantly brought the countrys jobless rate down. Whenever a new iPhone generation is launched, US retail sales experience a bounce and growth goes up a gear. The impact on economic growth is tangible, but it does not last long. The launch of the iPhone4 added half a percentage point to US GDP growth in one quarter before the effect evaporated in the subsequent quarter. In contrast, a radical innovation will spread through all layers of the economy. Its impact will be both unmistakable and permanent. It will emerge as a growth driver boosting the whole of the economy concerned. Lets take an historical example. The advent of the railways in the 19th century led to huge investment

PerspectiVes dition - SPECIAL annuelle EDITION 2012 2013

being pumped into the steel industry and the recruitment of tens of thousands of people to build railway infrastructure. By shortening journey times between two distant locations, the railways increased trading volumes and even created trading routes where none had existed before. Trains completely changed peoples movements: they began to travel further and for longer. When radical innovation shocks occur, productivity tends to improve by leaps and bounds, propelling economic growth up to noticeably faster rates. In this respect, there are some economists who do not hold with the view that the advent of the Internet can be regarded as a radical innovation as it is far from proven that productivity gains increased as a result of this technological advance. By very definition, radical innovations will arrive on the scene haphazardly. They more or less, however, follow the Kondratiev Cycle, with an innovation coming once every sixty years on average over the past three centuries. However, innovation shocks should not be viewed solely from the perspective of Kondratiev long-wave theory*. Apart from technological cycles, we have also pinpointed two other types of major innovation shocks. The first relates to shocks of a cognitive nature. These play a crucial role in how economies function so, by osmosis, they have an influence on the long-term growth and inflation regimes. Such shocks are often underestimated or even overlooked altogether. The second type of shock relates to statutory regulation. Liberalisation and growing entrepreneurship in emerging economies, especially in Brazil and China, are thoroughly changing growth regimes in these economies. Two main types of innovation shock Two main types of innovation shock seem the most likely among the five major possible kinds we pinpointed and discussed in last years Special Edition 2012 (see page 8: shocks of a demographic nature; currency shocks; legal and institutional shocks; shocks down to new economic knowledge/thinking; shocks caused by radical innovations and breakthroughs in technology). Technological inventions are the preludes to such technological innovations. We have defined an innovation as an invention that has already progressed into its industrialisation phase. In the coming decades, technological innovation could come from six distinct areas: Nanotechnologies: a market potentially worth USD3,500bn in ten years time. Their arrival on the scene could, for instance, thoroughly change the quality of fabrics and materials used to make clothing or medical treatments. Lets take one feasible innovation: the quality of fabric in the same outfit could be enhanced so that the item of clothing could be worn all year round in the height of summer to the depths of winter. Neurosciences. Biotechnology. Ongoing development in new IT/communication technologies: such as development of the quantum computer, artificial intelligence making it possible to simulate how the human brain works more effectively, 3D printing, etc. Such technology breakthroughs would radically alter the capital/employment equation in production. Shale oil and gas: in an optimistic scenario, these two new energy sources could completely reverse the tide of recent history, turning the US into a net exporter of energy and, above all, furnishing the world with reserves of energy for several hundreds of years, dispelling the threat of peak oil production and the damaging repercussions of an ensuing downtrend in economic development. Production of electricity (solar, geothermal, etc.), power storage and distribution (smart grids) and individual energy transmission. As for cognitive shocks, changes in approaches to monetary policy, fiscal/budget policy and financial innovation would play a key role. Whenever there is a shift in such influential aspects, economic policies have altered how the economy functions and behaves, significantly modifying the growth and inflation regimes. The table on page 6 summarises both shifts in economic-policy styles, when they have occurred, and the resultant shifts in growth and inflation regimes. We would make two rather surprising observations: Firstly, the decades have closed with growth and inflation regimes different from those at the outset of that same decade. Secondly, regimes shift into reverse between the start and end of the decade. For instance, the high inflation of the early 1980s was followed, by the end of the decade, by a spell of controlled low inflation. The sluggish economic growth of the early 1990s developed by the end of the decade into rapid expansion fuelled by overinvestment in new IT/communications technologies. These twin observations have prompted us to devise a simple rule of thumb for long-term forecasting: establishing where an economy stands at a given time, then projecting where it will go on the basis of the dynamic forces that underpinned its position at the time by considering that these will continue to hold sway over the ten years to come, would imply completely ignoring what the evidence of historical observations has demonstrated. In any event, if this approach had been adopted over the past forty years, it would clearly have delivered some huge forecasting mistakes. If we now apply the simple empirically derived rule of thumb to the decade of the 2010s, we will need to make allowance for a possible major reversal in economic dynamics. This could affect either growth or inflation or both. Applying the rule of alternation from better to worse and from worse to better opens the door to a possible shift in economic regime. The trend-

tHe influence of tHe uniMaGinaBlE innoVation on returns

GrowtH, inflation and fiscal/MonetarY-policY sHocks oVer tHe last four decades
Ination Decade Growth Fiscal Policy Monetary Policy End Beginning End Beginning End Beginning End Beginning

1970s Moderate High High Low Keynesianism Keynesianism Low interest Low interest rates rates 1980s Very Low Low High Keynesianism Supply side Low interest High rates Ination targeting

Endogenous Ination Ination 1990s Low Low Low High Endogenous growth growth targeting targeting 2000s Low Low Low Deep Endogenous Keynesianism Ination Ination Recession growth targeting targeting 2010s Low Moderate? Moderate High? Keynesianism Supply side? Ination targeting
Source: Pictet & Cie, AA&MR

Nominal GDP and income targeting?

break scenario we favour would see growth quickening rapidly over the next ten years accompanied by a more gentle acceleration in the rate of inflation. Forthcoming shifts in monetary/fiscal policy Since the early 1980s, monetary policies throughout major developed nations had been focusing on inflation, evolving in the early 1990s into specific inflation targeting, but the sub-prime crisis in late 2008 prompted central banks to take pragmatic emergency steps. Inflation targeting gave way to quantitative easing. This particular approach to monetary policy cannot be regarded as a transient phenomenon though: after all, its primary objective has, purely and simply, been to prevent the whole financial edifice from crumbling. We will see a shift away from quantitative easing over the coming years. But what will the new monetary policy look like? The answer to this question is self-evidently unclear. However, the persistence of below-potential economic growth, high unemployment and high debt levels simultaneously in the developed world do seem to point to two possible paths: firstly, a return to virtuous growth would help to mop up unemployment and provide a virtuous route to paying down all the accumulated debt; secondly, quickening inflation, ideally to a limited extent, would also help to lighten the debt burden. Taking those two factors into consideration, monetary-policy stances geared to nominal GDP targeting or nominal income targeting must be regarded as serious candidates. The implications for the major asset classes i.e. accelerating growth and/or inflation to a limited degree, i.e. not exceeding 4%-5% p.a. would be positive for equities, but very negative for sovereign bonds regarded as the safest of havens (Bunds and US Treasuries) and positive for peripheral eurozone government bonds (Spain, Portugal, etc.).

Since 2008, several, if not all, developed countries have been pursuing a Keynesian fiscal policy mix. This has betrayed its limitations. To start with, governments financing capabilities are close to being exhausted. Secondly, the spending spree by governments has failed to deliver virtuous economic growth, as stubbornly high jobless rates show. Worse still, once government spending is reined back, growth immediately slows and drops permanently below potential. In conclusion, innovation shocks are frequent enough to be a prospect when the forecasting time-frame is long. The time-frame chosen for estimating expected returns from asset classes is ten years. We have demonstrated the chances of a shock equating to a shift in the style of monetary and/or fiscal policy, or a shift in the growth and/or inflation regime, are high. We believe that, among all the errors and approximations that forecasting over the long run involves, one particular one must be avoided: the mistake of assuming that the environment of tomorrow along with its underlying economic mechanisms will be the same as those of today. This has informed our efforts to integrate the notion of a trend-break and regime change into our approach to allocating assets. *A Kondratiev Cycle, also known as the long-wave cycle, is a cycle lasting 40 to 60 years. A purely nancial explanation is given for this: during the upcycle when the economy is expanding, surplus investment pushes interest rates up; the increased cost of borrowing, in turn, triggers an economic downswing, eventually leading to the cycle being rebooted and starting all over again.

PerspectiVes - SPECIAL EDITION 2013

Tactical asset allocation on tHe Basis of risk factors

The purpose behind this article is to present the main phases of research and analysis that underpinned formulation of our proprietary model for allocating assets on the basis of risk factors. We have not, however, presented the quantitative aspects in this modelling. Our approach to asset allocation by risk factor takes account of a trio of risk factors: (1) macroeconomic; (2) market; (3) investment. For reasons of simplicity, we have restricted ourselves here to a discussion of macroeconomic risk factors. Since 2008, the spate of systemic crises, first in the US, then in the eurozone, has thoroughly shaken up the economic mechanisms influencing financial markets. Correlations in returns from risk assets, for example between equities and commodities, are running close to 1. As a result, investors have fallen foul of Murphys Law: portfolios can no longer be effectively diversified just at the very moment when investors most need them to be. One of the best precepts in investment for investors seeking to protect themselves from the ravages of systemic shocks to the system is to ensure that portfolio allocations are broadly diversified using assets whose behaviours are not correlated. In this article, we present our method of asset allocation on the basis of risk factors. Asset allocation by risk factors: solution at the margins of conventional portfolio diversification Economies in the Western world have moved from an economic regime described by economists as the Great Moderation to ones we have dubbed the Great Divergence for Europe and the Great Deleveraging for the US. The stubbornly divergent trajectories being taken by debt (trending upwards) and economic growth (trending downwards) have pushed governments into the frightening sequence of overindebtedness, followed by liquidity/solvency crisis, culminating in deleveraging. Rogoff and Reinhart have shown, from their analytical observations of thirty or so postwar economies, that a complete deleveraging cycle involving both the private and public sectors lasts on average eight years. The deleveraging cycle for US and European governments has not yet started. For investors, Murphys Law and its unpleasant ramifications for portfolio management might well linger for some time to come. Given this predicament, a new way of constructing a portfolio of investments has gradually been emerging: asset allocation based on risk factors. The aim is to inject some diversification into investment portfolios in a setting where correlations are tight. This approach to allocating assets involves diversifying underlying risk factors in the portfolio. However, opting for asset allocation based on risk factors is far from straightforward as the approach lacks a sturdy theoretical framework enabling practitioners to define the style and implement it easily. Academic research on this subject is really still only in its infancy and has not advanced far enough to construct as robust a framework as is the case for modern portfolio theory, with its classic Capital Asset Pricing Model (CAPM) or Arbitrage Pricing Theory (APT). Not surprisingly, there are as many ways of approaching asset allocation based on risk factors as there are portfolio managers who have recently adopted this style. Even defining a risk factor can be rather vague. Faced with this state of affairs, we have endeavoured to come up with a definition that it is the most overarching, consistent and, above all, the simplest possible. The most effective definition of a risk factor, to our way of thinking, is: a fundamental macroeconomic variable that has a decisive influence on movements in the pricing of one or more assets. Unlike other definitions, we have ruled out initially the possibility that one particular asset class can be a risk factor affecting other asset classes. Although there is a link between shares and sovereign bonds, that does not endow equities with the property of being a risk factor for corporate bonds. The link in the latter case has more to do with the regime of correlations between asset classes. Our proposed definition does possess three notable advantages: firstly, risk factors make it feasible to connect together fundamental economic scenarios upstream in our top-down approach with scenarios for asset classes downstream; secondly, by stripping out risk factors from the overall macroeconomic setting, the connection can be consistently made; thirdly, the risk factors chosen are homogenous in that all of them are fundamental economic variables, with no danger of any distortions as a result of mixing them with market variables, such as volatility or equity risk. Identifying features of macroeconomic risk factors Although macroeconomic risk factors to be selected do depend on the style of asset allocation, they also hinge on the investment time-frame, the prevailing economic regime, the phases of the economic and business cycle, and the relevant asset and country. Investment time-frame. For example, the next decision made by the US Federal Reserve is unlikely to have much impact on returns from bonds over the next five years. In contrast, its effect in the more immediate term is likely to be much greater. We have described in detail those risk factors influential for long-term behaviour of asset classes in the article on

tHe influence of tHe uniMaGinaBlE innoVation on returns

the method for calculating expected returns from asset classes (see page 10). In order to allocate assets based on risk factors, we need to distinguish between fundamental risk factors that are strategic (long-term) and tactical (short-term), detailed below. Prevailing economic regime: The economic regime undergoes significant and often unsuspected changes over the long run. For instance, a major technology breakthrough can thoroughly change underlying economic mechanisms (see page 4 for more explanations). By way of example, the arrival of petrochemicals or the advent of new IT and communication technologies gave lasting boosts to economic growth rates in the developed world. By definition, the prevailing economic regime does not vary over the short term. It will experience no short-run changes during the lengthy period during which the shock factor delivers its effects. Cycle phases. Tactical fundamental risk factors do not all exert the same influence over asset classes at the same stage in the economic cycle. The one-trackmind tendency among financial-market operators results in them treating just one or a few risk factors as crucial influences. As a result, we have sought to sort out active risk factors from those lying dormant. The active ones are those at the centre of the markets attention. Markets shift focus though so dormant factors at some point will become active and the active factors fall dormant. The skill in allocating assets by risk factor lies in not just correctly pinpointing factors, but also selecting dynamically those having a decisive influence. Assets and countries. Risk factors do not necessarily influence different asset classes in the same ways. Clearly, a given risk factor may work more to the advantage of one asset class over another. In 2012, for example, price movements in gold were highly sensitive to the extremely accommodating policies being pursued by the worlds central banks. These encouraged investors to flock to gold as a safe haven to protect themselves against the devaluation of currencies caused by the wholesale printing of money. In contrast, commodities as such were not that influenced by these policies. One final complication: the influence exerted by an active risk factor might be greater for a given asset in one country compared to another. The six tactical fundamental economic risk factors and their current influence on asset classes In the second half of this article, we have applied our approach to the task of tactical asset allocation. We have pinpointed six tactical fundamental risk factors for all asset classes: business cycle; corporate earnings cycle; inflation; fiscal policy; monetary policy; systemic liquidity/solvency risk in Europe. At the time of writing (December 2012), using these factors for arriving at a tactical allocation of assets by risk factor making allowance for the distinction between the US and Europe does involve taking account of two different groups of active factors: In the US, forthcoming economic policy as shaped by the twin risk factors of monetary and fiscal policy will play a crucial role in influencing how assets will behave. In particular, GDP growth prospects for 2013 will hinge heavily on the tenor of postfiscal-cliff fiscal policy. As for Europe, the systemic crisis and monetary policy factors are crucial. The following chart illustrates the influences likely to be exerted on the main asset classes by tactical fundamental risk factors for Europe. In the concluding phase, finalising the asset allocation involves diversifying the portfolio on the basis of active risk factors. One simple way of doing this is to se-

SensitiVitY of asset classes to actiVe and dorMant risk factors*

EU and Currency European EU periph. US core US corp. Risk factors Dynamics EUR/USD equities US equities sov. bonds sov. bonds bonds (IG) Business cycle Systemic risk Monetary policy Fiscal policy Ination cycle Acceleration in economic growth Less liquidity risk, scal cliff: -1% on US GDP growth EU solvency risk modestly addressed Aggressive and QE-like Austerity Core-ination stable around 2% US corp. EM corp. bonds (HY) Gold bonds

Earnings cycle Double-digit growth

*Framed risk factors are active risk factors

Sources: Pictet & Cie, AA&MR

PERSpECTIVES - SPECIAL EDITION 2013 PerspectiVes

lect those assets that should provide the portfolio with favourable exposure to active risk factors and avoid any damaging exposure to the same factors. Looking back at the chart, we can identify two groups of assets likely to be either winners or losers. Group of assets providing favourable exposure. Those assets likely to benefit from forthcoming political decisions paving the way towards resolution of the eurozone crisis and/or from fresh monetarypolicy measures by the ECB are: equities; gold; investment-grade corporate bonds; sovereign and corporate bonds from eurozone countries currently in financial distress. Group of assets liable to cause damaging exposure. Prices of German Bunds might well be affected if Germanys solid finances are deployed to back and underpin those of financially weaker countries like Spain. The role of the European Stability Mechanism (ESM) is to transfer funds towards troubled countries. The ESM might result in the quality of Germanys finances being undermined. As yields on Bunds are running well below 2%, the threat of deterioration in Germanys financial position is patently not being priced in. The risk of yields climbing back up on German Bunds and, by extension, making their valuation less attractive in general has prompted us to underweight sovereign bonds in our asset allocation. The tactical risk factors we have defined are economic in origin. With a top-down approach, they ensure there is consistency with the chosen economic scenario. For the sake of consistency, we took the decision to set to one side other risk factors such as liquidity, volatility or even the value investment style. Such factors do, of course, play a role that cannot be overlooked. Nevertheless, for reasons of simplicity, the approach outlined in this article is geared to macroeconomic aspects. Other specific factors also apply at two other levels: markets and investment. For instance, volatility is a risk factor inherently characteristic of markets whereas the value approach relates to investment.

Gold price
1,800 QE 1 QE 2 QE 3

1,600

1,400

Gold (USD/oz)

1,200

1,000

800

600 2008 2009 2010 2011 2012 2013 2014

Sources: Datastream, AA&MR

tHe influence of tHe uniMaGinaBlE innoVation on returns

EXpected returns froM asset classes

In this article, we have outlined the key steps leading to the results we have presented for expected returns from asset classes. On account of their proprietary nature, we have not provided specific details about the models we have developed for this purpose or about the specific calculations required. When it comes to wealth management, calculating likely long-term returns to be expected from the various asset classes is essential in order to construct asset allocations for portfolios. The need to establish a strategic asset allocation is a common one when seeking to manage considerable wealth as this type of allocation serves twin purposes: safeguarding the standard of living and lifestyles of the beneficiaries while simultaneously ensuring that the wealth in its entirety can be passed on to the next generation. Three types of models used to calculate potential returns We have sought to estimate returns and, in this article, we have described the method formulated to arrive at our estimates. To conclude our analysis in this article, we have presented ten-year returns for the various asset classes estimated according to our approach. In order to calculate returns, several methods could have been adopted, the three main ones being outlined below. The first is the most widely employed. It is founded on the method of mean reversion. This presupposes that, over a long enough period, returns from asset classes will not deviate from their very long-term performance trend. For instance, a period of negative returns lasting several years will be followed by a rally which would see the long-run return reverting to a level close to its historical average measured over a very long period. According to this particular method, after the lost period of the first decade of this millennium, equities in developed markets could record a nominal compound average rise of roughly 6% p.a. over the coming decade. The second method involves forecasting returns on the basis of statistical analysis of chronological series. Making use of more or less sophisticated methods, such as time-series analysis, spectral analysis, wavelet decomposition or analysis of complex systems, one can arrive at a confidence interval for future values the series might show. The third method is founded on an approach based on risk factors. This is the method we have employed. In the article on page 7, we have outlined the method we have formulated for constructing our tactical asset allocation for portfolios. This option offers three advantages. Firstly, it makes it feasible to preserve a thread of consistency between the two categories of asset allocations for investment the short-term tactical and the longer-term strategic. Secondly, it integrates the main assumptions underpinning the macroeconomic scenario into calculations of expected returns. Thirdly, the calculations of expected returns take due account of innovation shocks (see the accompanying article on page 4). In stark contrast, the first two approaches, by their very construction, drastically restrict the possibilities of factoring in any shake-up caused by a shock event liable to disrupt and alter the behaviour of economic fundamentals. Two essential and simple risk factors Having opted for our approach, the risk factors to be incorporated into the model need to be identified. There is no compelling reason why those risk factors pinpointed for the purposes of establishing the tactical allocation (see the article on page 7) will be the same as those taken into account to construct the very long-term strategic allocation. Following extensive research and analysis, the intuition we had the outset of the process eventually proved to be correct: the two cornerstone risk factors are real economic growth and inflation. This finding is very appealing as it endows this particular approach with the beauty of simplicity. Other risk factors also emerged as key discriminating influences. These include such variables as productivity gains or per capita GDP growth. Trying to forecast productivity, however, is a particularly awkward exercise, especially considering that economists struggle to agree about actual productivity levels in the past. Productivity is affected by economic growth, so forecasting growth, by very definition, does encapsulate the notion of productivity: any rapid quickening in economic growth will be matched by significant productivity gains. The second welcome discovery from our research is that the two predominant risk factors of real economic growth and inflation employed for calculating expected returns are applicable across the board for all major asset classes: cash, equities, sovereign bonds, corporate bonds on both developed and emerging markets in all three cases and, to some extent, commodities. Trends on asset classes, on one side, and the twin risk factors, on the other, are statistically cointegrated: over a long period, their two respective trends do not diverge much from each other statistically speaking, they share a common stochastic drift. Challenge of imagining potential shocks is vital when it comes to projecting future returns The exercise of forecasting long-term expected returns would be pointless and misguided if no attempt is made to factor in potential shock events that might modify the underlying trend being

10

PerspectiVes PERSpECTIVES dition - SPECIAL annuelle EDITION 2012 2013

tracked by asset classes. We have investigated this particular aspect in the article on page 4 (Innovation shock: the unimaginable is possible). We have also sought to explain why this aspect has been a source of error by highlighting those structural changes already witnessed over previous decades that affected growth and/or inflation. Expected returns from asset classes are dictated by the scenarios formulated for our two essential risk factors. This approach enables us to integrate coherently macroeconomic analysis and relevant structural changes into the method for calculating expected returns. From the mind-boggling infinite to a more reassuring and manageable discretised number of regimes It is all very well identifying the two crucial risk factors; they now need to be estimated. At this juncture, we are now confronted by the daunting challenge of the infinite. Clearly, rates of economic growth and inflation can show a practically infinite number of permutations and values. To overcome this, our approach is quite straightforward. Infinity needs to be broken down into manageable chunks. In economic and market finance parlance, these tranches of infinity are referred to as regimes. We have transformed a continuous and infinite series into discrete tranches. Through this discretisation, we have duly reduced the number of possible values that each of the two risk factors can have. In more detail, we have pinpointed three main regimes for growth and three for inflation. The intersection of the regimes for the two factors describes the economic environment. The schematic diagram below sums this up far more succinctly than any lengthy and wordy explanation could. Values for these thresholds we have defined depend on the economy under consideration. Threshold levels that influence shifts from one regime to another are not the same for the US, Europe or emerging nations. For our purposes here, we are focusing on the case of the US. Standard regime: Long-term potential growth for the US economy is estimated at between 2.5% and 3%. At that tempo, the economy is considered to be in its standard growth regime. The US economy expanded at that pace between 2003 and 2007. Deflationary regime: When growth is sluggish and well short of the potential rate, it will be because deflationary pressures are depressing investment and consumer spending. The US economy has been languishing in this regime since 2008. This is also the regime in which Japans economy has been trapped since the early 1990s. Shock regime: When an innovation shock event occurs, economic growth will quicken, often to a rate well above that common during the standard growth regime. This was the case in the US in the early 1980s in response to a shift in fiscal policy away from Keynesian principles to supply-side economics favouring entrepreneurship, investment and, by extension, jobs. At that time, as in the latter half of the 1990s, growth was running at almost 4% p.a. We have pinpointed four regimes for inflation, three illustrated in the chart below and the fourth regime being galloping inflation or even hyperinflation. Standard inflation regime: Since the advent of monetary policy geared to inflation targeting, inflation, as measured using the core consumer price index excluding food and energy, is considered to be at equilibrium when close to the 2%-mark. Economies in the developed world were in this regime during the 1990s and 2000s. Disinflationary regime: When inflation falls noticeably below the 2% threshold rate, deflationary pressures are spreading through the economy. This state of affairs is often linked to the regime categorised above as deflationary. Japan has unmistakably been subject to this regime as its annual average inflation has been running at -0.4% since 1995. Inflationary regime: When inflation climbs above that 2% threshold, the economy will be moving into a regime of sustained high inflation. Galloping inflation or hyperinflation: Once inflation soars above 6%, economies are entering the territory of the fourth and final inflation regime. Galloping inflation is not such a rare phenomenon. This regime held sway in the 1960s and 1970s. Hyperinflation is a rarer beast, but it wreaks devastating damage on an economy. Usually, it will involve a major shock, often relating to currencies, such

THree Main reGiMes for inflation and tHree reGiMes for econoMic GrowtH
Sluggish growth Standard growth Innovation shock Gr=1% Gr= 2.5% Gr= 4% Disination = 0.5% Standard ination = 2% Deationary Unlikely environment Unrealistic

New normal

Goldilocks

Golden growth

Ination Stagation Inationary Overheating = 4% environment environment

Source: Pictet & Cie, AA&MR

11

tHe influence of tHe uniMaGinaBlE innoVation on returns

as the one that fuelled hyperinflation in Germany under the Weimar Republic in the early 1920s. The hyperinflation regime needs to be dealt with separately though as it deforms mechanisms of both economic growth and policy. The interplay of the three growth and three mainstream inflation regimes obviously gives us nine possible economic environments. Of these nine, two are fairly implausible: for example, we are highly unlikely to experience a combination of robust economic growth of 4% and a disinflationary regime at the same time. Our discretised approach has managed to scale infinity down to seven feasible economic regimes. A starting regime and an ending regime: two scenarios proposed One final stage remains before we can proceed with calculating expected returns: selecting the regime to input. As we are introducing into the equation the possibility of shock events, a shift in regime could happen at some point over the next ten years. This means we need to decide upon the regime prevailing in the economy at our point of departure and the regime at our destination. Although pinpointing the starting regime is ascertained via economic analysis, selecting the expected regime at the end-point will be influenced by assumptions made about possible shock developments. We have conducted our calculations using two main scenarios. Scenario I: an innovation shock. In our first scenario, we have worked on the assumption that, at some juncture over the coming 10 years (2013-2022), an innovation shock will occur. According to our analytical approach, this would significantly alter dynamics for our two chosen risk factors. Real economic growth would shift from a deflationary regime to one of robust growth. Inflation might then also shift from the disinflationary to standard inflation regime. As we highlighted earlier, the shock effect could come from either a technological breakthrough or from a sea-change in monetarypolicy style. Central banks could abandon inflation targeting to switch to quantitative easing, then onto price-level targeting or even asset-price targeting. We have factored in two hypotheses. Scenario I: timetable for a shock to materialise. 2013-2015: Firstly, the innovation shock is presumed to occur around 2016. Until then, economies will continue to stutter along in the current regime. 2016-2018: Secondly, the knock-on effects from the shock would filter through over three years before having their full impact on growth and inflation in the latter part of the decade. 2019-2023: The economy would operate in its new regime from 2019 and for the rest of our ten-year period.

EXpected returns froM asset classes witH an innoVation sHock (Scenario I)

Standard growth Starting regime Ending regime Annualised Expected Return and standard ination 2013 2015 2016 2022 2013 2022 Cash Government Bonds Corporate Bonds IG Corporate Bonds HY Equities US 2.7% 12.9% 9.7% EU 3.3% 14.4% 11.0% Asia ex-Japan 3.0% 10.2% 8.0% - - 8.0% - - 15.0%
Sources: Pictet & Cie, AA&MR

US US Germany

0.0% 4.5% 3.1% 1.7% 0.5% 0.8% -2.5% 1.0% 0.3%

US 2.8% 2.8% 2.8% EU 2.2% 2.8% 2.7% US 7.0% 7.7% 7.5% EU 7.3% 8.0% 7.8%

Private Equity Real Estate (Monte Rosa) Private Equity (Monte Rosa) *
* Monte Rosa 2011 expected annualised net IRR 15%

12

PerspectiVes PERSpECTIVES dition - SPECIAL annuelle EDITION 2012 2013

EXpected returns froM asset classes witH a return to standard GrowtH (Scenario II)
Central scenario with Starting regime Ending regime Annualised Expected Return innovation shock on growth 2013 2015 2016 2022 2013 2022 Cash US 0.0% 3.8% 2.6% Government Bonds Corporate Bonds IG Corporate Bonds HY Equities US 2.7% 8.1% 6.5% EU 3.3% 9.2% 7.4% Asia ex-Japan 3.0% 7.4% 6.1% - - 8.0% - - 15.0%
Sources: Pictet & Cie, AA&MR

US Germany

1.7% 0.8% 1.0% -2.5% 1.6% 0.8%

US 2.8% 2.8% 2.8% EU 2.2% 2.6% 2.5% US 7.0% 7.4% 7.3% EU 7.3% 7.7% 7.6%

Private Equity Real Estate (Monte Rosa) Private Equity (Monte Rosa) *
* Monte Rosa 2011 expected annualised net IRR 15%

Scenario II: no shock, but a process of normalisation Under this scenario, we have worked on the assumption that, gradually, US economic growth would revert to its potential rate. Such a scenario is very close to the mean-reversion scenario cited in the introduction to this article. We have not gone into detail about how our estimates of expected returns have been calculated as this is not our aim here. We have adopted the premise that the trajectory of expected returns on asset classes takes on the shape of the economic cycle. We have summarised the findings in the tables. Returns on asset classes: risk assets hold the better hand We can summarise the main conclusions concerning our estimated expected returns on asset classes in four points: In both the scenarios, risk-based assets look to offer the best return prospects. Expected returns from equities in developed markets would be above their historical average if an innovation shock occurs: 10% on average vs. 7%. If there were an innovation shock, equities in the developed world would clearly outperform emergingmarket equities even though growth in the emerging world would be faster. There are two reasons for this. Firstly, the conversion of economic growth into

corporate earnings is still less efficient for companies in the emerging world than in developed nations: nominal GDP growth/earnings growth elasticity is 1.4X for businesses in the developed world compared to 0.8X for those in emerging countries. Secondly, companies in the developed world tend to have increasingly broad international exposure this allows them to lock on not just to economic growth being generated in their own backyards, but also to global economic growth to boost turnover and profitability. Sovereign bonds issued by countries perceived as being safe havens since 2008 (Germany and the US) today offer investors pitifully low interest rates. After generating an annualised total return of 8.2% since 1974, annual returns on US Treasury bonds would work out at between just 0.8% and 1.0% depending on the regime. Sovereign bonds would be likely to underperform equities. The same goes for corporate bonds. We, therefore, see a likely relative reversal of fortunes for asset classes. Corporate bonds on developed markets would struggle on account of both very low interest rates and historically narrow credit spreads. Expected annualised returns from investment-grade corporates are estimated at 2.8% for both the innovation shock and normalisation scenarios.

13

CENTRAL BANKS

MONETARY POLICY IS EVOLVING

The nancial crisis has challenged many of the traditional assumptions behind central banking pratice, providing new challenges for monetary policy. Frederic Mishkin Former Member of the Board of Governors of the Federal Reserve System Alfred Lerner Professor of Banking and Financial Institutions Graduate School of Business, Columbia University

The basic paradigm of central banking has been that, in the words of Milton Friedman, inflation is a monetary phenomenon and that central bankers could control inflation. While they worried about fiscal issues, they always felt that those would be taken care of by governments in advanced economies and it was really only a problem in emerging markets where governments could not get their act together. The role of central banks was to focus on controlling inflation and not to worry about unsustainable debtto-GDP levels what is called fiscal dominance. A second central banking tenet was that monetary policy did not have to think about financial stability issues. It was concerned about financial stability, because if things blew up, that could be costly in terms of government bailouts or a big hit to the economy. But financial stability was the responsibility of the regulators. The objective of central bankers was to use monetary policy to control inflation, only secondarily worrying about business cycles and output.

These two ways of thinking led to the framework of flexible inflation targeting used by central bankers in advanced countries. The key is to ground inflation expectations over the long run with numbers say, 2% per annum. Then flexibility may be applied in the short run to deal with business cycles, being careful not to undermine low inflation expectations. However, the financial crisis has shown that these underlying assumptions of central banking practice are no longer valid, requiring a rethink to address new challenges. The first challenge is the issue of what we call the zero lower band. It is clear that nominal interest rates cannot fall much below zero, because people can just stuff their cash under their mattresses or in safe-deposit boxes. When the floor of zero nominal interest rates is reached, central banks must turn to non-conventional policy measures. The second challenge is that financial crises impose very high costs of cleaning up after them and restoring

14

PERSpECTIVES - SPECIAL EDITION 2013

financial stability can undermine what central bankers are trying to do with monetary policy. Furthermore, central banks may contribute to financial crises by creating an environment of price and output stability, as in the Great Moderation before the financial crisis. The stability at the time led to risk-taking because people underestimated the degree of risk. Worse, when interest rates are very low, people take excessive risks in the search for yield. And the third challenge is that monetary policy may end up being inflationary when central banks find themselves operating in a framework of fiscal dominance. If the fiscal authorities fail to reduce debt levels to a sustainable level, the central bank faces only two policy options, neither of which is palatable: They can bail out the government by monetising (purchasing) the debt which creates a highly inflationary environment. They can stand tough and refuse to monetise the debt. However, that would increase the likeli hood of default which could push up interest rates to fund the debt making it harder to reduce the deficit. So what does all this mean for central banking? The first issue is to deal with the zero lower bound, which means that the only way to pursue expansionary monetary policy now is through non-conventional policy measures. These can involve lending programmes providing credit in chosen markets as was done during the crisis. Or it can involve purchases of assets, frequently called quantitative easing (QE). Or there is management of expectations making a commitment to keep interest rates low for a very long time.

In the US, for example, inflation expectations have been very solidly grounded by the Federal Reserve, but core inflation is actually below the targeted 2% level because there is so much slack in the economy. So while inflation is low, the other element of the so-called dual mandate unemployment is not. Unemployment is very high and notwithstanding the recent drop, the labour markets do not look very strong. The Fed needed to do something more expansionary, hence three episodes of QE. The first was during the crisis, which was effectively fire-fighting rightly doing whatever it took to put out the fire. Once the crisis started to dissipate, the second episode of QE was still fire-fighting by purchasing long-term bonds, but there was no long-term strategy behind it. The Federal Reserve has also been managing expectations by saying it expected to keep the Federal Funds rate at zero until a specified date which has now been extended to mid-2015. The problem with that promise is that it depends on the state of the economy if it recovers more strongly than expected, rates may have to rise earlier. So the latest round of QE is open-ended: the Fed said it would keep buying assets until labour markets improved. The Fed may be moving towards nominal GDP targeting. This means committing to keep interest rates low until nominal GDP is growing adequately, even if for a short period of time inflation rises above 2%. The problem is, of course, to avoid undermining long-term inflation expectations and there will be a lot of discussion at the Fed about how to articulate this. The second issue for central banking is financial stability. Since a good monetary environment and potentially low interest rates can stimulate credit booms and credit bubbles, central bankers now recognise their re-

Central banks may contribute to nancial crises by creating an environment of price and output stability, as in the Great Moderation before the nancial crisis. The stability at the time led to risk-taking because people underestimated the degree of risk. Worse, when interest rates are very low, people take excessive risks in the search for yield.

15

CENTRAL BANKS
sponsibility for stability. Indeed, some people are now saying that the Feds interest rates are too low and will lead to a bubble. In practice, there is no expansion in credit at the moment because the economy is still deleveraging, but the Fed needs to monitor this while it is keeping interest rates at zero. That is another seachange in the way central banks have to operate and what they have to think about in terms of policy. The third issue is fiscal dominance. In the US, the politicians have yet to come to grips with the debt-to-GDP level, and that could lead to a rise in bond rates. Since the Fed recognises that a sharp rise could create a deep recession, it might decide to buy bonds to hold rates down. That would pose a very serious inflationary risk for the US if the fiscal authorities do not get their act together. Despite these fears, I am a typical American optimist. In previous periods of acute fiscal deficits, the Congress has eventually decided to carry out fundamental tax reforms and fix the problem. Despite the political divisions now, the problem is still solvable and I believe that the US political process will fix it. How long that takes is another issue, but I do not think fiscal dominance is likely to be a problem in the US. But turning to the eurozone and Europe, the fiscal dominance issue is much more scary. Everyone knows that one of the major flaws in the creation of the eurozone was the failure to think hard enough about the fiscal union and what to do about member states overspending. The no bailout rule is dead, but new fiscal rules are not going to solve the problem because it is easy to get around them, as with the old Growth and Stability Pact. But a second flaw in creating the euro was the absence of a banking union unlike in the US where states are involved in regulation but bailouts are done by the centre. When Greece goes down, there is no way that it can repay its debt and it will have to default. The issue then is how to stop the contagion spreading in Europe. The key is to make sure that the banks are healthy, that there is a resolution authority when banks collapse, that there is deposit insurance and that the governance problems are sorted out. The classic procedure for dealing with banking crises is to recapitalise the banks after establishing the true state of their balance sheets. That was done in the US, but the European stress tests were inadequate and the authorities avoided dealing with the issues. Instead, they focused on boosting capital ratios, but that deals with individual banks rather than the systemic problems. And since it is costly to raise capital, the banks will shrink their balance sheets by selling assets, making things worse for other banks and reducing lending which hits economic growth. It has been remarkable to watch this slow-motion train wreck for the past couple of years, and the contagion problem could be a nightmare if Greece goes under. In practice, I think that Europe will muddle through because the European Central Bank will do whatever it takes to deal with the crisis. The ECB has said it is willing to buy sovereign debt in countries facing debt problems through what are called Outright Monetary Transactions (OMT). The pressure on the ECB to monetise the debt will continue, with consequences for inflation. There is one hope, however, which is that they actually get their act together and create a proper fiscal union something that Brazil achieved when it reined in the spending of the states. But we have not seen a lot of progress along these lines yet in Europe, so I cannot say that there is a very high probability of this happening.

The Fed may be moving towards nominal GDP targeting. This means committing to keep interest rates low until nominal GDP is growing adequately, even if for a short period of time ination rises above 2%. The problem is, of course, to avoid undermining long-term ination expectations.

16

PERSpECTIVES - SPECIAL EDITION 2013

17

18

19

THE GLOBAL ECONOMY

A JOURNEY INTO THE UNKNOWN

The study of economic history reveals that the world economy is in uncharted territory. The outlook is uncertain, with two conicting trends creating an extended period of violent risk-on/risk-off volatility.

Jim Reid Managing Director Global Head of the Fundamental Credit Strategy Group Deutsche Bank

In this presentation, I want to discuss the latest edition of a report that I write every year which looks back at the history of financial markets in an attempt to understand what we can learn about them and asset price performance that will help us in the future. This years report was called A Journey into the Unknown, reflecting the fact that while economic history has been invaluable in this crisis, several of the key economic and financial variables are at levels never seen before. This is a giant experiment, effectively: we are all lab rats in todays global economy. First, interest rates. The longest time series of shortterm interest rates I can find is for UK base rate, which goes back to 1694 when the Bank of England was created. The UK has been through a lot in that 318-year history, with the industrial revolution, manias, panics, bubbles, depressions and wars. Yet base rate had never fallen below 2% until the recent financial crisis, and it has now been at 0.5% for more than three years.

Something similar is true of other advanced economies: over the last 150 years, base rates for the G7 countries, Switzerland and Spain have not been much below 3% until recently. Now, rates across those nine big economies have been hovering at around 0.5% for the last three years or so. Never before in history have so many countries had base rates so low for so many years with no sign that they are going to rise soon. Moving further up the yield curve, something similar has happened with Dutch 10-year government bonds. With nearly 500 years of data available, we can see that yields reached an all-time low last summer. The same is true for US 10-year Treasuries, which have reached record lows in the 220-year history that I have data for. Yet government debt is pushing to levels where history tells us that sovereign crises often occur. Carmen Reinhart and Kenneth Rogoffs seminal study three years ago suggested that, once a governments debt

20

PERSpECTIVES - SPECIAL EDITION 2013

GOLD PRICE IN USD AND GBp SINCE 1790 (LOG SCALE)

10'000 Gold price in GBP/oz (log-prices) Gold price in USD/oz (log-prices)

1'000

100

10

1 1810 1830 1850 1870 1890 1910 1930 1950 1970 1990 2010
Source: Deutsche Bank, GFD

rises above 90% of gross domestic product, the economy tends not to grow out of it without financial repression, inflation, devaluation or default. Today, although many developed economies have debt levels above the 90% of GDP which history says often creates a problem, their bond yields are mostly at all-time lows. Second, fiscal deficits. Throughout US history, balanced budgets were the normal peacetime phenomenon until around 40 years ago. While big deficits accumulated during wars, fiscal balances very quickly moved back into surplus when peace returned. There is also evidence from other countries indicating that balanced budgets were the fundamental cornerstone of economic policy until the last 40 years. Since then, however, deficits have become the cornerstone of most countries economic policy. In the last 44 years, there have been only four surplus years in the US, which all occurred around 2000 when the Internet boom produced artificially high tax revenues. Apart from those four years, the US has been in deficit every year since 1968. Something similar has been true for other leading economies surpluses have been extraordinarily rare over the last 40 years. Now the sovereign debt crisis has hit and many countries are under pressure to cut their deficits. But most of their economies have been based on running deficits, and they will find it very difficult to restore balanced budgets. Every country that has cut its deficit so far has seen disappointing growth. Why did we move from balanced budgets to deficits? The turning point for me was in 1971 when the ties that bound currency systems to precious metals were broken. All through modern history, most currencies were backed by precious metals through a gold standard or, later, the Bretton Woods system. Before 1971, if

you had run a deficit for more than a couple of years, the likelihood was that the economy would be hit by outflows of gold reserves. Countries were disciplined into running cautious budgetary stances by the gold standard. When President Nixon suspended dollar convertibility to gold in 1971, we entered an era of freer currencies with nothing backing money. In a paper money system, running a budget deficit might hit the currency, but there is no loss of reserves. And that started a slow growth of beggar-thy-neighbour deficit accumulation, and beggar-thy-neighbour increases in the money supply. One interesting consequence was that the price of gold, more or less fixed until the 20th century, began to explode (see above chart). The biggest impact of this on financial markets was higher inflation. In the 700 years before the start of the 20th century, global prices rose very gradually, with as many down years for inflation as up years. Although there was a small upward long-term trend of inflation, deflation was as commonplace as inflation. But in the 20th century, there has not been a single negative year for global inflation since 1933. Again, I think that this is down to the gradual weakening of ties to precious metals to the point where they were completely abandoned in 1971. It is the conventional wisdom that inflation has been well controlled over the last 20 or 30 years, and that central banks have done a good job. But an analysis of 40 developed and emerging economies around the globe shows that very few countries have kept their average annual inflation rate below 5% since 1971. When most countries have had to deal with a problem such as a currency crisis, debt crisis or growth problem, invariably the inflation rate has been pushed higher in a freer currency world. So my realistic, pragmatic conclusion

Over the last 150 years, base rates for the G7 countries, Switzerland and Spain have not been much below 3% until recently. Now, rates across those nine big economies have been hovering at around 0.5% for the last three years or so. Never before in history have so many countries had base rates so low for so many years.

21

THE GLOBAL ECONOMY

is that, in a world facing huge fundamental problems, it is inevitable that countries will print money until politics gets in the way or, more likely, inflation surges. Next, asset prices. The first thing to note is that periods of expansion in the US have lengthened in recent years, with the last three completed cycles much longer than those before. The main reason was that policymakers had an unparalleled level of flexibility: if the economy weakened, deficits could be allowed to rise to cushion the impact. Central banks could also loosen monetary policy without necessarily unleashing inflation, because the reintegration of China into the global economy started a 25-30-year trend of falling inflation by pushing down the price of traded goods and labour. In those longer business cycles, asset prices performed much better because the confidence of investors was higher. But countries are now under pressure to return to more balanced budgets and monetary policy flexibility has ebbed China is no longer depressing global prices to the same extent and interest rates are already close to zero. Quantitative easing (QE) has kept the system afloat and avoided a depression, but it does not necessarily stimulate growth. With both fiscal and monetary policy more constrained than over the last three long business cycles, I believe we are returning to shorter business cycles. And since asset prices have tracked business cycles fairly well throughout history, I foresee violent riskon/risk-off volatility over the next few years. QE and money printing should avoid risk-off swings triggering a depression, but much greater volatility in asset prices will create serious difficulties for attempts to reduce the severe debt overhang. In fact, the reality is that virtually no country is deleveraging at aggregate level (government plus private-sector debt). Figures for 15 leading economies show that every single country has a higher level of aggregate debt to GDP than in 2007, apart from the US which had a small decline. Even the countries that have slightly deleveraged from peaks reached after 2007 have not made significant reductions and many are still stuck at their peak debt-to-GDP level. In five years of crisis fighting, the system has been propped up by issuing more debt and printing money to finance it. Part of the problem is the absence of significant growth. Most of the 15 countries are still below their 2007-08 GDP in real terms. It is very difficult to reduce the debt-to-GDP percentage when growth is still very weak, and it is difficult to grow when the debt burden is so high. Propping up an old, failed system is a recipe for very low growth and shorter business cycles though it is arguably better than letting it all collapse which would be a disaster for markets. To sum up, I think we are in a world of violent risk-off/ risk-on swings. Risk-off periods will reflect perceptions that fundamentals are still very bad and the debt overhang is showing no signs of shrinking indeed, it could be argued that economies are in a worse state than four years ago. However, risk-on periods will be encouraged by the growth in money printing and the increasing response from the authorities, albeit jagged and maybe not continuous. As those two forces battle it out massively over the next few years, this will continue to fuel volatility. I believe that money printing will slowly start to win and the system will be slowly healed. But it is going to be a long, hard slog.

With a currency crisis, debt crisis or growth problem, invariably the ination rate has been pushed higher in a freer currency world. So my realistic, pragmatic conclusion is that in a world facing huge fundamental problems, it is inevitable that countries will print money until politics gets in the way or, more likely, ination surges.

22

PERSpECTIVES - SPECIAL EDITION 2013

23

BEHAVIOURAL FINANCE AND BEYOND

PERCEPTIONS, BELIEFS AND ACTIONS

Olivier Oullier, a renowned expert in behaviour change and neuroeconomics, explains how research developments in brain sciences and behavioural nance help us comprehend the biases that distort nancial and economic decisions and how investors and traders can better understand and cope with such problems.

Olivier Oullier Professor of behavioural and brain sciences, Aix-Marseille University Strategic adviser to governments and private institutions

At first sight, my profile may seem a bit unusual with respect to finance: I have a background in complex systems and then moved into psychology and neuroscience, with a particular emphasis on how we make decisions and what is going on in our brains. I work as an academic professor/researcher and I advise public and private institutions on how to develop prospective or crisis management strategies based on what we know about the human brain and behaviour dynamics. Many of the decisions we make every day in our jobs are related to things that we are used to dealing with. But even if we have no idea about something, no background or training on a topic, we are very likely to have an opinion about it and sometimes to share it. As human beings, we always make calls. These decisions are biased for most of them and, at times, not appropriate for the context in which we are evolving. The investors chief problem and his worst enemy is likely to be himself, yet we have a strong tendency to blame our mistakes on information or others. A strong belief, rooted in standard economics and its models of rational decision-making, is that a person will use information to make the right choices after making an informed judgement, balancing rights and wrongs, pros and cons, costs and benefits, etc. But people do not behave in this way. Real human beings behave very differently from the artificial economic agent that is still at the heart of most decision-making models today. Behavioural finance draws on cognitive and social psychology to analyse dozens of psychological biases that influence how we process and use information to make choices. But we can now go further by using neuroscience to analyse how our brains function in the decision-making process. I am not here to sell you the perfect solution; anyone who would claim that brain sciences alone can provide the perfect strategy, regardless of the field (economics, finance, marketing, management), would be at best misinformed, if not lying. The main advantage of learning about what we know about behaviour and the brain in finance is to avoid mistakes, to avoid repeating mistakes and perhaps to develop insights that complement and, sometimes, supplement traditional financial theories and tools in understanding what is at play when making decisions. For example, we are all overconfident. Many of the decisions we make every day in our jobs are related to things that we are used to dealing with. But even if we have no idea about something, no background or train-

ing on a topic, we are very likely to have an opinion about it and sometimes to share it. As human beings, we always make calls. These decisions are biased for most of them and, at times, not appropriate for the context in which we are evolving. For instance, we have a powerful aversion to loss: if you ask people whether it is more painful to lose money or to earn money, of course people prefer to earn money and yet loss-aversion can make them reluctant to cut their losses when it would otherwise make sense to do so. When we look at brain data, it appears there are two distinct dynamics that are employed when we compute what we can gain and what we can lose. Besides, when people are in loss-aversion mode they rely more on the brain networks that deal with fear. There is also herding, our tendency to mimic or follow what other people do. If you want your employees to do something, telling them that almost everyone they know does it is much more effective than telling them it is the right thing to do. In the UK, Her Majestys Revenues & Customs (HMR&C) did an experiment with letters sent to citizens informing them of the percentage of people who pay their taxes on time compared to traditional reminders. This strategy increased significantly the number of people who paid their taxes on time. Our tendency to herd might come from our strong tendency to mirror others. We have brain cells called mirror neurons that fire whether one is performing an action or seeing someone else perform the same action. And if someone is performing an action that you are thinking about, you somehow start to resonate with the person and might even mimic that action spontaneously; this is what is at the heart of herding behaviour. Some of the most common biases are listed on page 26. There are many others. Any or all of them may distort how we process information and reach decisions; we are living in a jungle of biases that we have to navigate and endure.

24

PERSpECTIVES - SPECIAL EDITION 2013

Over the past 20 years, incredible machines have been developed that allow us to look into the brain and see how it processes information. This does not stop us from being fooled, and it overlooks the role of knowledge in making decisions. If I flip a coin, plot the results and ask whether there is a pattern, most people from outside finance would detect one from the visual representation of the distribution between heads and tails. When I ask people working in finance or economics, knowledge manages to control information and they rightly say there is no pattern. Our certainties can fool us. Being informed is necessary, but usually not enough to counter biases. Also, looking at the brain on its own is not really useful. For cognitive neuroscience to be useful in finance, one needs to pay equal attention to the brain and to the information exchange in and outside of it. This exchange occurs within the brain, between the brain and the body (thanks to the nervous and hormonal systems) and the brain/body and the physical and social environments. Add on top of this our changing emotions, our history and experiences, as well as our future which, at least temporarily, is represented by our intentions and we are dealing with the most complex, adaptive and information-sensitive system ever. In order to make a decision you may want to make a representation of what is going on in your brain and the environment. You select an action and you act, and, after, you can evaluate and learn. Of course we do learn, but when you look at most of the literature in economics and finance, the word learning is not so common. It is barely considered. Looking deeper inside the brain, we see a reward system is at play, as is a kind of neural currency: a neurotransmitter called dopamine that plays a key role in the functioning of this system. The reward system is what makes us feel good, whether we are expecting to reach a goal, waiting for it or achieving it. Behaviour can also depend on the level of hormones in the body a traders morning testosterone level can predict his profitability for the day. A commonly held belief is that more testosterone leads to less rational behaviour, leading some to suggest that hiring only female traders would mean fewer problems. However, as volatility in markets, stress, reward and many more behavioural and biological factors are at play, this is unfortunately not that simple. A final example of what we can learn from brain research relates to trust. Experimental settings have revealed that, in contexts where people exchange money and trust evolves throughout these exchanges, the brain networks in charge when making these decisions can see their peak activity evolve with time. The more Brain sciences have provided unprecedented insights into how judgement and decision-making can be processed differently in the brain. Such ndings hint at the reasons why people very often make poor decisions in spite of having a clear judgment of a situation.

25

BEHAVIOURAL FINANCE AND BEYOND

we know people, the more activity in these areas of the brain peaks. Thanks to neuroimaging there is evidence that when people first interact, the evaluation of the decision to invest money occurs after another individual invested. As people interact with each other, the brain switches from reactive mode to anticipation mode, as we can see the brain network supporting the decision is active before knowing how much the other one invests. This casts light on how people have a tendency to trust the market too much and make decisions even before they have all the information. Anticipation can be a great asset, but it can be a disaster as well. So what are the challenges for researchers and people in finance? We understand a lot more about human behaviour, our biases, how our beliefs can mess up our judgement and how this judgement can influence decisions. But the true dynamics between the individual and the collective level remain very hard to evaluate, and we do not yet have efficient models to connect them. Yet, brain sciences have provided unprecedented insights into how judgement and decision-making can be processed differently in the brain. Such findings hint at the reasons why people very often make poor decisions in spite of having a clear judgment of a situation. On top of that, the intrinsic dynamics and emerging self-organised technical patterns of a market can disturb or stabilise behaviour as much as external information can, contrary to efficient market theory. Another challenge is that most of the experiments are conducted in labs, on Western-educated students on campuses in democratic, industrialised countries. As a result, they are not universally representative, and the context is not very realistic. The insights could be totally wrong if applied to a global population. This is why I do a lot of work with subjects who work in financial institutions rather than students on my campus. We have also developed new techniques of data mining and crowd sourcing for large-scale analyses and new portable neuroscientific devices that can be

used in workplaces unobtrusively so we can establish more sound general insights. Thanks to our growing knowledge of cognitive biases and how the brain works, we know that we will not deliver perfect solutions. But if we can train people to identify their biases, they could identify very complicated days when the stakes are really high and they could assess whether or not they are ready to deal with them. People will have to learn how to say that, for example, someone is dying in a persons family and the people in charge will have to learn to accept that such occasions can totally flip that persons ability to deal with major crises. Anchoring: We rely too much on previous information, regardless of its accuracy. The effect can be multiplied if we hear the information from someone we respect a lot or someone we totally despise. Negativity bias: When weighing options we put more weight on nega- tive factors that are painful, not only financially but in terms of ego. Illusion of control: We overestimate our ability to influence events. If we miss a day at work and some stocks dive, we tell ourselves that, if we had been at work, it would have been different. Priming:The way that we process and use information depends on who gives it to us. Observer expectancy effects: If we are convinced that something is right, that will influence the way we process information. In the models we use, the way we interpret information and the people we rely upon, our choices reflect our strong beliefs. Endowment: Because we bought something, we value it more. We will be less reluctant to sell something that we did not buy. Backfire effects: We evaluate people according to whether they say what we want to hear. We downplay them if they go against our views and it may even strengthen those views. Gamblers fallacy: We believe that previous events have a strong influence on what is going to happen, even if they are totally unrelated. Hyperbolic discounting: We prefer a gain that arrives sooner rather than later.

26

PERSpECTIVES - SPECIAL EDITION 2013

27

EFFICIENT ASSET ALLOCATION

Our answer to a low-return environment in which most portfolios are highly correlated with equities is cost-effectively investing in many return sources to reduce equityrisk concentration without sacricing expected returns
Investment perspectives from Antti Ilmanen Managing Director, AQR Capital Management (Europe) LLP

Q: How do you see the current investment environment?

Virtually every investor faces two big challenges. The first is that we are in a low-return environment: the real yields of both stocks and bonds have come down over the last 30 years, and over the last 15 years, real returns have been at all-time lows of around 2%. I believe this low-yield environment will persist for some time. Second, portfolio asset allocations are increasingly correlated, and alternative investments are not behaving differently. Global 40/60 portfolios (40% stocks, 60% bonds), 60/40 portfolios and endowment portfolios all have unusually high correlations to the MSCI World index, on the order of 0.97 in recent years. Hedge funds have also been very highly correlated they are not quite absolute return providers. Even with a portfolio of hedge funds, private equity, real estate and commodities, the correlation with a global 40/60 portfolio has exceed 0.8 over the last 10 years.

Q: What should investors do in such an environment?

AQRs answer to a low-return highly correlated environment is wide harvesting cost-effectively investing in many return sources to reduce equity-risk concentration without sacrificing expected returns. Our firm has been using the pyramid in the chart shown on page 29 to depict how we approach different sources of returns. It also describes our favoured portfolio allocation methodology. At the base are long-only investments capturing market-

28

PERSpECTIVES - SPECIAL EDITION 2013 PerspectiVes

Q: What are style-factor premia?
There are four big styles that are relevant: Highest Cost (2 and 20) Lowest Capacity
Value: overweight assets with good valuations

AQR PYRAMID OF RETURN SOURCES

Illiq s uid

Alpha

Carry: overweight assets with high income Momentum/trend: overweight assets with high recent returns Defensive: overweight assets with low beta or high quality.

Alternative Beta Premia (Hedge fund premia and style premia)

Moderate Cost Medium Capacity

Illiq ui d s
uid Illiq

Market Risk Premia

Lowest Cost (Index Funds) Highest Capacity

These styles travel well: value-investing, for example, was pioneered in the 1930s by Benjamin Graham in equities and has been found to work in other asset classes. The same is true for carry, momentum/trend and defensive styles. In stock selection or in more macro trading, all these styles have done well over the last 35 years. Risk-reducing styles such as trend-following and defensive equities can be especially interesting portfolio allocation strategies on their own, but I believe there are advantages in multi-strategy portfolios. One is obviously that they give better diversification, potentially reducing costs through netting. But perhaps more important, a multi-strategy approach induces less return-chasing. Investors like to do some market-timing and style-timing but the right amount is pretty small you need to be very humble about any timing you try to do. The most important thing is to avoid the common mistake of chasing strategies that have performed well in the past two or three years. This is usually a trap in momentum/trend-following. Even people who know that they make this mistake find it difficult to avoid, so a multi-strategy portfolio ties you to the mast and makes you more likely to behave like a long-term investor. An example of the success that can be achieved in this way is provided by Warren Buffett, an investor with an extraordinary track record. Everybody is interested in knowing the secret of his success, and although he is Mr Value Investor, there is much about his record that cannot be explained by looking at value signals. But you can explain virtually everything by putting in factors such as high quality and low beta. He has always said that he buys value and quality, but he also levers up low-beta opportunities.

risk premia such as equity premia and duration premia. At the top are alpha investments, which we think should be a small part because they are elusive and costly. We believe investors can capture some of the returns that are categorised as alpha better by looking at alternative-beta premia.

Q: What about alternative-beta premia?

We use dynamic long-short strategies to pursue alternative returns, which are inherently more complex to capture and in shorter supply than market risk premia.

Q: How do you capture market-risk premia?

The traditional way of capturing marketrisk premia is through capital allocation to stocks, bonds and other long-only sources of returns. An alternative is to use a risk-balanced approach such as risk parity. In a 60/40 portfolio, for example, at least 90% of risk typically comes from equities because they are more volatile. Risk-balancing aims for a better balance among traditional returns sources such as growth, interest-rate risk, inflation risk and credit risk usually through a combination of stocks, bonds and commodities. At AQR, we are big believers in targeting risk, not just balancing across investments but also through time. This may be just as important as choosing which asset classes to balance risk across, or whether you think about balancing between economic environments. This is not something which every investor can do, but we believe that those who can will reap rewards with better quality portfolios.

Q: What are hedge fund risk premia?

They are the portion of a hedge fund portfolios returns that once were defined as alpha, or extraordinary returns, but more recently have been explained by a new understanding of more common risk-factor exposures, or betas. Another way to explain it is to say that hedge fund risk premia are the common risk exposures shared by hedge fund managers pursuing similar strategies. Some of my colleagues at AQR wrote a paper on this, called Is Alpha Just Beta Waiting to Be Discovered?. In it, they concluded that the correlation between hedge funds and the equity market suggests that maybe not all of their return is alpha. For example, if a fund is on average long on equities, duration and credits, there are probably some beta gains there. The same is true with long-short strategies. The key is that it is possible to capture the returns to these longshort strategies without getting charged 2-and-20 (2% management fee plus 20% of profits)

Source: AQR

29

INVESTMENT MANAGEMENT

We believe that we are the only portfolio manager in the world which can say to its clients that we will never bet with their money.
Yves Choueifaty President and Chief Investment Ofcer, TOBAM

Q: What is the Anti-Benchmark?

The Anti-Benchmark is a portfolio which is extremely easy to define in any given market, whether it is UK, Japanese, emerging market, Swiss or Swedish equities. It is simply the most diversified portfolio you can build, being long-running and fully invested without leverage. The only focus of the Anti-Benchmark is diversification. What does it mean to be diversified? It is simply the opposite of being biased. If you have a very strong bias towards US stocks or bank stocks or value stocks, you will not be able to say that you are diversified because you are exposed to a limited number of sources of risk. So a good way to summarise a diversified portfolio is one where you are trying to build a portfolio whose volatility is evenly provided by all the possible sources of risk. If I build such a portfolio, by definition you will not be able to tell me that I am biased; and if I am not biased that means that I am diversified. Now diversification is not a well-defined concept. We all know how to measure volatility, tracking error, alpha, beta, gamma, delta, etc., but there is no measure for diversification. So we have introduced a measure of diversification for any given portfolio which has been patented in the US and Australia and will be

30

PerspectiVes - special edition 2013

patented very soon in Canada, Japan and some European countries. With this tool we can build a portfolio that maximises its diversification.

THE ANTI-BENCHMARK DEFINES AND MAXIMIZES DIVERSIFICATION

Return The Anti-Benchmark delivers broad equity market exposure that provides superior performance with lower risk

Q: What are the advantages of maximum diversification?

The chart shows the investable universe on a risk/return basis with the efficient frontier being the upper side of the red line. It is a well-established fact that whatever geography or time period you consider, the market-cap weighted benchmark is always far from the efficient frontier after two or three years. The first consequence of diversification will be risk reduction the portfolio will move to the left on the chart as risk falls. What we also very strongly believe is that diversification will systematically increase the return, for two intuitive reasons. First, the most diversified portfolio is one with real neutral risk allocation we try to allocate the portfolios volatility evenly to all the drivers of risk. It is very important to realise that if you describe your aim as capturing a risk premium, your job is about diversifying. Let me give you two real life examples. When you go to a casino, you will never be able to say that you have captured the risk premium from the casino even if you have made a winning bet. Only one person captures the risk premium from a casino: the owner, whose strategy is diversification. If you launch a casino with only one slot machine, you will go bankrupt. Another example is provided by insurance companies: when I trained as an actuary, the first lesson came when the professor told us that the insurance companys job is never about trying to forecast. An insurance companys job is about pricing and diversifying the risks it covers. It is the same in stock markets. If you build a portfolio which is different from the most diversified portfolio, you have made some bets, whether explicit or implicit, and as soon as you have made bets you have destroyed some risk premium. Depending on your success in forecasting, you will create wealth or destroy wealth but in both cases you have paid back risk premium instead of receiving all the risk premium.

Most Diversified Portfolio Risk Premium Benchmark

Backed up by empirical evidence and supported by academic theory

Anti-Benchmark captures the full equity market risk premium

Risk (Volatility)

We believe the higher returns result from better capturing of the risk premia
Source: TOBAM

The second intuitive reason why diversification will increase the return can be seen in the weighting of the sectors of the S&P 500. As stocks appreciate, the greater their weighting becomes in the index, and vice versa. So, for example, after the first oil shock, energy stocks rose and if you were following a market-cap weighted index, you would increase your allocation to energy stocks until the peak was reached when the oil counter-shock happened, which was the worst day to be holding them. A market-cap weighted benchmark is not a neutral risk allocator; it is a biased risk allocator. And not only is it biased, those biases change over time it is a dynamic risk allocator.

the Anti-Benchmark outperformed or underperformed. They start asking whether the benchmark outperformed the Anti-Benchmark or underperformed it, because they realise that we do not make bets. Ours is a diversified portfolio, the most boring portfolio. The nonboring portfolios simply make enormous bets.

Q: When does the Anti-Benchmark underperform?

We underperform when the Benchmark outperforms us, because of the way the dynamic risk allocation of the benchmark is managed! The adverse scenario of the Anti-Benchmark is when the benchmark outperforms by winning the bets it is making and if that happens, and since the Benchmark is buy and hold, the bets will simply grow. At the end, the adverse scenario for the Anti-Benchmark is full concentration of the benchmark.

Q: So what is your investment strategy?

In a way, the Anti-Benchmark is not an investment strategy, but the absence of strategy. It is the agnostic portfolio, the neutral risk allocator, the only portfolio you should buy if you are not able to forecast. If you cannot forecast, you are left with no other decision than to diversify. We believe that we are the only portfolio manager in the world which can say to its clients that we are never going to deliver any alpha. We cannot deliver any alpha because we do not even have a strategy. After six or nine months of investing with us, our clients stop asking whether

31

contriButors
EDITORIAL TEAM PICTET & CIE Kalina Moore Wilhelm Sissener DesiGn and editorial consultancY Stphane Bob Production Multimedia Winkreative REPORTER John Willman ENGLISH TRANSLATION/REVISION Stephen Barber Daniel Steffen Keith Watson pHotoGrapHY Mara Truog

32

PERSpECTIVES - SPECIAL EDITION 2013

Disclaimer This report is issued and distributed by Pictet Group. It is not aimed at or intended for distribution to or use by any person or entity who is a citizen or resident of, or located in, any locality, state, country or other jurisdiction where such distribution, publication, availability or use would be contrary to law or regulation. The information and material presented in this report are provided for information purposes only and are not to be used or considered as an offer or invitation to buy, sell or subscribe to any securities or other financial instruments. Furthermore, the information, opinions and estimates expressed herein reflect a judgment as at the original date of publication and are subject to change without notice. The value and income of any of the securities or financial instruments mentioned in this document can go up as well as down. The market value may be affected by changes in economic, financial or political factors, time to maturity, market conditions and volatility, or the credit quality of any issuer or reference issuer. Moreover, currency exchange rates may have a positive or adverse effect on the value, price or income of any security or related investment mentioned in this report. Past performance should not be taken as an indication or guarantee of future performance, and no representation or warranty, expressed or implied, is made by Pictet Group regarding future performance. Instructions stipulated by the client as regards trading, transactions and investment constraints shall take precedence over, and may diverge from, the Banks overall investment strategy and recommendations. Portfolio managers are granted a degree of flexibility so as to accommodate the individual wishes and particular circumstances of clients. As such, the asset allocations specified in this report do not have to be strictly adhered to. Actual allocations to alternative, non-traditional investments (e.g. hedge funds) may exceed those mentioned in the grids herein provided that positions in traditional equities are adjusted accordingly. This publication and its contents may be quoted provided that the source is indicated. All rights reserved. Copyright 2012 Pictet Group

AG

Are you on Facebook? Add Pictet to your daily flow of information by liking our page. Search for Pictet Wealth Management on Facebook and join the conversation.

Follow us on twitter for fast-paced and opinionated updates from our investment specialists and analysts.

facebook.com/pictetwealthmanagement

twitter.com/pictetwm

Subscribe to our YouTube channel to receive the latest interviews with Pictets specialists discussing investment strategies and macroeconomics.

Our video interviews are also available as a podcast on iTunes. Subscribe free of charge, and take Pictets videos with you on your iPad or iPod.

youtube.com/pictetwm

bitly.com/pictet-itunes

perspectives.pictet.com