FACTOR INVESTING GIVES MORE INSIGHT INTO RISKS

CHAIRMAN Arjen Pasma, PGGM Investments PARTICIPANTS Foto's: Fotopersburo Dijkstra // ROUND TABLE PORTFOLIO CONSTRUCTION IN MULTI ASSET CLASS INVESTI...
Author: Stephanie Blake
2 downloads 0 Views 4MB Size
CHAIRMAN Arjen Pasma, PGGM Investments PARTICIPANTS

Foto's: Fotopersburo Dijkstra

// ROUND TABLE PORTFOLIO CONSTRUCTION IN MULTI ASSET CLASS INVESTING

Rob Brand, Blue Sky Group Wim van Hyfte, Candriam Maarten de Kok, MN, since September 1 ING Pension Fund Bill McCoy, FactSet Valentijn van Nieuwenhuijzen, ING Investment Management Bart van Poucke, BNP Paribas Investment Partners Jelle Ritzerveld, Kempen Capital Management

This Round Table has been established in collaboration with FactSet

FACTOR INVESTING GIVES MORE INSIGHT INTO RISKS By Hans Amesz

A risk factor based portfolio is not necessarily superior to a traditional one, but the use of risk factors is a useful addition to the traditional portfolio theory. Investing based on factor risks could give new insights in the risks in a portfolio. Financial Investigator organized a round table discussion about Portfolio Construction in Multi Asset Class Investing.

2

FINANCIAL INVESTIGATOR

5 / 2014

Are portfolios nowadays constructed in a different way than according to the traditional portfolio theory? Valentijn Van Nieuwenhuijzen: ‘Traditional portfolio theory cannot sufficiently capture the complexity of the financial markets today. We have to find alternative ways of investing. My view is that a sort of dynamic risk parity approach will produce more robust investment results. The goal is to reduce uncertainty.’ Bart van Poucke: ‘I don’t think we’re talking about a completely different way of investing. This is an evolution triggered by the bad investment experiences of the great financial crisis of 2007-2009.’

Wim Van Hyfte: ‘I would not say that risk factor based allocation is superior to the traditional asset allocation. You can’t compare apples to oranges. However, the use of risk factors is a really good addition to traditional asset allocation. Investors need to realize that allocation strategies based on risk factors - for example based on volatility, liquidity and credit – can lead to other kinds of systematic risks.’ Jelle Ritzerveld: ‘As a risk manager, I want to gain more insight, I want to get more information about the characteristics of a portfolio. Once you have obtained that, you still need to decide what to do with it. With both strategies, it is still necessary to make allocation decisions. Even in a world with risk factors one needs an investment framework.’

> Arjen Pasma is Chief Risk Officer Investments of PGGM. In his role he is responsible for the risk management of the entire investment process including ALM, Strategic Portfolio Construction and Fiduciary Management. From 2008-2012, he was the Investment Management and Pensions practice leader of Deloitte in the Netherlands and director in the Financial Risk Management team. From 1997-2008, Pasma fulfilled several roles within ABN AMRO Asset Management as senior portfolio manager and head of quantitative equity management, risk manager and part of the management team equity management. He studied Econometrics and is a CFA charter holder since 2002.

5 / 2014

FINANCIAL INVESTIGATOR

3

// ROUND TABLE PORTFOLIO CONSTRUCTION IN MULTI ASSET CLASS INVESTING

> Rob Brand is Senior Portfolio Manager TAA at Blue Sky Group. He is reponsible for the tactical asset allocation in clientportfolios. Before working at Blue Sky Group he was Head of Equity Discretionary Portfolio Management at ABN Amro Bank. His previous positions included Senior (Quantitative) Strategist and Senior Portfolio Risk Manager, both of which were for ABN AMRO Asset Management. Brand completed the Master’s programme in Business Administration of the Financial Sector at the Vrije Universiteit in Amsterdam.

Rob Brand: ‘With both strategies allocation decisions need to be made. It’s a matter of how you look at the investment world. I believe that factor investing gives more insight into the inherent risks in the portfolio. With risk parity I can ensure that risks are evenly distributed amongst the different asset classes within the portfolio. These are influenced by different factors and a distinction needs to be made between fundamental and behavioral factors.’

Is smart beta a marketing concept, or does it actually mean something? Maarten De Kok: ‘Both the one as the other. There are many different sorts of investors with many different goals. With a focus on factors such as growth, value, momentum, low volatility, et cetera, investments can be reconciled to achieve what you want. Smart beta is an additional investment vehicle.’

What are the key risk issues when you employ factor-based investing in a multi asset class setting? Ritzerveld: ‘The most important question for me is what should be done if, for example, there is a large concentration of just one factor. You can’t know in advance if diversification is

always the best solution. It depends on your overall investment framework and that leads to discussions with the portfolio managers.’ Van Hyfte: ‘The focus on factor investing has greatly increased in the last two years, but the traditional asset allocation is still important. We all know that risk-based allocation has its limitations in terms of capacity. Moreover, we must always ask ourselves if the risk is real or if it is a result of economic behavior (behavioral) and will disappear.’ Brand: ‘It may be prudent to combine traditional asset allocation with factor investing. We need to realize though that it adds an extra layer of complexity and complexity does not always add value to the portfolio.’ Van Poucke: ‘It is quite possible that one or more factors don’t perform well over one or several years. It is difficult to explain to investors that they may need to think long term in factor investing. It may take a stretch of 5 to 7 years before they can achieve the most value.’

Does a strategic allocation to risk factors actually help pension fund board trustees to get a better grip on the portfolio and how should they explain? Bill McCoy: ‘It’s part of our fiduciary responsibility to continue to educate the clients about all developments that are of interest to them. When it comes to factor investing, the various risks are of course very important. In terms of strategic asset allocation, the financial crisis showed that it is too simplistic at times to divide the world neatly in stocks, bonds and alternatives, because a lot of the trouble came from assets that were classified as bonds but had significant equity components. I think a risk factor based approach could highlight where equity risk is hiding in a bond portfolio.’ Brand: ‘It will take a lot of communication before the concept of factor investing will be widely accepted by the board of trustees of pension funds. It will take some years for a board to even understand how factor investing works.’

4

FINANCIAL INVESTIGATOR

5 / 2014

De Kok: ‘Step by step we do see more and more boards becoming interested in smart beta, possibly because there is a marketing element to it. There is more attention for smart beta in all kind of publications such as Financial Investigator.’

Van Hyfte: ‘If investors are prepared to sit out the cycle, risk factor based allocation can be promising and give value. It should be well explained why this is so and what risks are involved. We need a better risk return profile.’

Van Nieuwenhuijzen: ‘When it comes to the investment horizon, I think a lot of the pension and insurance money managers say that they are long-term investors. However, in practice they don’t have a much longer horizons than mutual fund managers or a lot of the other players in the financial markets.’

De Kok: ‘Our clients are quite skeptical about factor investing. They not only want a rational explanation of the factors that affect the risk and return, but they also want consistent results throughout time. Otherwise they might as well buy the traditional index and save themselves a lot of trouble. I think it is almost impossible for a separate risk factor to always outperform, but you can’t have underperformance for too long either. Clients feel rather uncomfortable about that.’

What is the added value for pension funds not to panic when the going gets tough? McCoy: ‘The capture of the illiquidity premium, which is admittedly difficult. There are ways you can take away the impetus to liquidate the strategy when it’s at its worst and avoid liquidating assets at fire sale prices. This is not necessarily a market risk process but an operational risk practice. That needs to be added to the equation as well.’

Are many institutional investors not really focused on the long term?

Is this a reason for being a long-term investor?

McCoy: ‘It goes back to client education that investments are return-focused and we need to raise awareness of risk focus. By looking at the past cycles pain points can be demonstrated; for example when value or momentum underperformed.’

McCoy: ‘You should plan your liquidity needs in such a way that you don’t have to liquidate. The behavior of other investors also plays a role. You try to capture the premium.’ Van Nieuwenhuijzen: ‘I agree, but in reality, if the performance is disappointing at a certain moment, the regulator may step in and your own reputation is put at risk: you might lose your job. There are all kinds of pressure that can’t be ignored and you must take steps to protect yourself. The regulator can change its policy, how do you prepare for that?’

Once again, what is the added value to clients for going for risk-based investing, rather than just sticking to traditional allocation? Ritzerveld: ‘It is a matter of providing the right guidance to different customers. If a pension fund thinks value investing is the right way to go forward, you have to show the trustee board how it’s done.’

Van Hyfte: ‘They say they are, but they certainly won’t accept an underperformance for three years; possibly one or maybe two years.’

De Kok: ‘What is really missing are clear benchmarks for factor investing. There are a lot of inadequate benchmarks; especially for low volatility. A clear set of universally accepted benchmarks would be a tremendous support to factor-based investing.’ Van Poucke: ‘As regards benchmarks for factor investing there is a lot of confusion. It is very difficult to make the right choice.’ Van Nieuwenhuijzen: ‘Our ambition can’t be simply to inform and educate the client on, for example the right benchmarks. Ultimately we must deliver an overall investment solution. As I said earlier, the result should be more robust. Making unexpected choices can contribute to a better risk return ratio. However, we also have to recognize that there always remains large uncertainty.’

> Wim Van Hyfte is Senior Fund Manager – Quantitative Analyst at Candriam Investors Group. He is co-responsible for the quantitative management of institutional and retail equity portfolios. Van Hyfte conducts empirical research on asset pricing, the quantitative modeling of alpha/risk factors and portfolio optimization. As Visiting Professor at the Solvay School of Economics & Management, ULB, he teaches on ‘Asset Pricing in Practice’. He holds a PhD in financial Economics from Ghent University that focused on modeling security risk premia and the implications for portfolio allocation.

5 / 2014

FINANCIAL INVESTIGATOR

5

// ROUND TABLE PORTFOLIO CONSTRUCTION IN MULTI ASSET CLASS INVESTING

> Maarten de Kok was from 1998 until 2008 an equity PM at Robeco. First in the Emerging Markets team, later for the whole world in the Rolinco Global Growth team. With his switch to the GTAA team at APG he ended up in a multi-asset and multi-instrument environment where portfolio construction and risk management were central to the process. At MN de Kok worked from 2011 to 2014 on strategic issues in the Fiduciary Management department. They build and maintain a model portfolio, which is the source of idea’s to service pension clients. In September 2014 De Kok started as strategist at ING Pension Fund.

Ritzerveld: ‘The most difficult part is understanding the client’s wishes. Is he looking for short-term performance, or for a more or less stable return over a longer period? That should be the ultimate benchmark: performance against the wish of the client. You should let the customer explain what it is he really wants and then you try to accommodate that as best you can.’

In trying to reduce complexity you see a number of pension funds simplifying their investment policy. What is the role of a risk based investment strategy? McCoy: ‘A shift in investing away from complexity opens up more opportunities for those funds that can harvest the complexity premium and generate alpha.’

How do you sell the reality that certain factors do not always work and that a form of dynamic allocation strategy is needed? Van Nieuwenhuijzen: ‘Some clients, usually the larger ones, are open to this. Others less so. There are fewer opportunities for using factor analysis in smaller funds because more dynamics are required. If you move from a pension fund to a family office

or high net worth individuals, you shift into a sort of one-stop shop solution: a dynamic, all-in-one solution.’ Van Hyfte: ‘There is always a long-only (whether or not dynamic) allocation strategy needed to grow the portfolio. A dynamic risk factor allocation strategy provides additional diversification possibilities.’ Van Nieuwenhuijzen: ‘You have to decide how to construct a portfolio based on factor investing in order to get a better balance of risk than the traditional asset allocation balance.’ McCoy: ‘Due diligence is always of great importance.’ Van Nieuwenhuijzen: ‘A couple of months ago I spoke to a Harvard professor who does a lot of research on risk factors. It turns out that their development is very unstable over time. The result depends on which risk factors in which periods are performing or not.’ De Kok: ‘It is still very difficult to outperform a 60/40 benchmark with stocks and bonds. Around 90 percent of the ultimate result will be decided by having stocks and bonds at the right times. That leaves not more than 10 percent for decisions based on smart beta. You must spend most of your time on the most important decision; choosing the right benchmark. Then, if you still have time you can spend it on, for example, factor investing.’

Let’s assume that the average pension fund needs a nominal return of approximately 5.5 percent in order to meet its objectives. A number of risk parity portfolios have so much allocation to fixed income and commodities that they are not reaching that level. What can be done? Van Hyfte: ‘If the interest rate, for whatever reason abruptly increases, risk parity funds will be confronted with huge losses because they have an overexposure to bonds arising from the current low volatility of fixed income. Investors should be aware of that.’ 6

FINANCIAL INVESTIGATOR

5 / 2014

Van Nieuwenhuijzen: ‘The one and only asset class in which volatility has increased over the last two or three years, is fixed income. In all other asset classes volatility has decreased. Everybody talks now about what will happen if interest rates rise. I would remind you that some 25 years ago everybody looked at Japan in that way and interest rates continued to fall. This is also possible in Europe: ten-year yields of 50 basis points.’ McCoy: ‘Fixed income versus a liability stream also depends upon the direction of the hedging asset; you can protect against inflation risk by buying inflation linked bonds or by buying equities or real estate.’ Van Poucke: ‘We are in a very low yield environment, which is much more volatile than it was in the past. We are now seeing daily movements that were unthinkable five or six years ago. The future duration must be very well monitored and, if necessary, adjusted. The right timing is of the utmost importance to a portfolio. The wrong timing can lead to huge losses.’

What is the role of leverage, shortage and derivatives in conjunction with risk based strategies? Brand: ‘If you really want to achieve the goal of the strategy, using derivatives is probably very important one should enter the long-short world. Another important factor is rebalancing: to what extent and how frequently are you willing to rebalance the portfolio or do you stick to a long term horizon?’

What kinds of factors are in each of the asset classes and how can you minimize the exposure to a certain factor within the longonly space? Ritzerveld: ‘If you know how much of each risk factor is present in a certain asset class, you can redistribute the asset classes in a smart way and get closer to the desired exposure, also in the long-only space.’

you measure liquidity risk in fixed income. For government bonds there is an active market, but there is still a premium for on-therun versus off-the-run securities. We are investigating where that premium comes from and look at accidental indicators that occur in illiquid markets.’ Van Nieuwenhuijzen: ‘I think we possess too limited data and have too few observations of crisis periods to correctly estimate how this liquidity premium will evolve in the next crisis.’ Van Poucke: ‘Often, developments with regard to liquidity are not gradual but sudden: All at once the market can completely dry up.’

It may be prudent to combine traditional asset allocation with factor investing.

McCoy: ‘One of the challenges we are actively exploring is how

> Bill McCoy is a Senior Product Manager in the Fixed Income group at FactSet. In this role, he actively works in research, client support and sales to help the firm enhance its position as a leading provider for comprehensive analytics for fixed income securities and the derivatives used to hedge them. Prior to FactSet, McCoy worked for other fixed income software vendors as well as in fixed income portfolio management. He has written and spoken extensively on fixed income hedging and return attribution. McCoy has a Master’s degree in Operations Research from the University of North Carolina, and is a Chartered Financial Analyst.

5 / 2014

FINANCIAL INVESTIGATOR

7

// ROUND TABLE PORTFOLIO CONSTRUCTION IN MULTI ASSET CLASS INVESTING

> Valentijn van Nieuwenhuijzen is Head of Multi Asset Boutique. He is responsible for

the macroeconomic outlook, investment strategy research and design, tactical asset allocation (TAA) decisions and portfolio management relating to the multi-asset funds and mandates managed by ING Investment Management. Van Nieuwenhuijzen has 15 years of experience working in financial markets. He appears frequently in international broadcast media like CNN, CNBC, Bloomberg, Reuters and Euronews and journals like the FT, IPE, and other newswires. Moreover, he appears regularly on the major Dutch financial and national news programmes and journals (RTL Z, NOS, BNR, Radio 1, FD)) Also he contributes on a regular basis to international publications such as Financial Times, Wall Street Journal en FD. In 1999 Van Nieuwenhuijzen started his career within ING Investment Management, after graduating in Economics at the University of Amsterdam (UvA).

Van Hyfte: ‘The question is do you hedge liquidity risk or not.’ De Kok: ‘It is common practice to hedge most foreign security risk, especially the dollar one. That provides an alternative to manage the liquidity risk. Because the dollar has a safe haven status, it becomes stronger in times of extreme stress. However, a lower dollar hedge reduces the liquidity needs and then it is not so horrible to be in a somewhat less liquid market.’

Should we try to create new approaches? Van Hyfte: ‘The impacts of illiquidity are so drastic that you cannot ignore them, and they have been ignored for a long time. The years 2007, 2008, 2009 have shown what happens if your reaction is too relaxed and complacent. We don’t have tools to predict these sorts of shifts, regime changes, perfectly. But the possibility that they will occur should be integrated into allocation decisions by hedging or dynamic allocation.’ Ritzerveld: ‘As for building a model to predict these sorts of regime shifts I see no holy grail. I am, on the other hand, a strong supporter of modeling, in the sense of interpolation, trying to have a better understanding why something has

happened in the past. This way we can better understand in what way things are interconnected.’ McCoy: ‘Liquidity risk does not always involve the entire market. Different sectors of the market can face illiquidity at different time periods. You won’t find the answer to illiquidity by looking at market risk factors. It is influenced by other factors, such as the connections between different dealers. They can force a sale of the asset if the funding disappears. One way to reduce illiquidity risk is a derivative-buying-strategy. You’ll continually lose little bits of money over long periods, but it will work out very well during a crisis.’ Brand: ‘Unfortunately we will never be able to build a perfect model. It is a matter of sensitivity: can we cope with so and so much volatility in the portfolio? Is this a stress situation we can handle or should the risk in the portfolio be reduced?’

Another problem with many of these illiquid alternative asset classes is a lack of data. How should this be dealt with? Van Hyfte: ‘I think alternative and traditional assets are, in terms of risk profile, quite similar to each other. In the long run you end up with the same kind of exposure to the economy.’ Van Nieuwenhuijzen: ‘Certainly in terms of the exposure to fundamental risk factors that drives the performance, the risk profiles of illiquid alternative and traditional assets are quite similar to each other. Alternative assets are less transparent and this can have a great influence on their performance. ‘They may also have the advantage - not so much hedge funds, but the private equity firms - that they connect to a different stage of the economic cycle. This means that they can add to the overall growth exposure in the portfolio.’

Does it make a difference if certain ESG (environment, social, governance) profiles are built-in to a traditional or a risk based allocation strategy? Van Nieuwenhuijzen: ‘You can make ESG the dominant driver of 8

FINANCIAL INVESTIGATOR

5 / 2014

your investment decisions or you can make it one of the sources of return in your overall analysis. I would not favor making ESG a separate factor for asset allocation.’ De Kok: ‘We have integrated ESG completely in our portfolios. It is still a pretty soft concept. ESG is not going to make or break the performance in the long run. It is your asset allocation that’s going to make the performance, possibly with some ESGcriteria.’ McCoy: ‘We created portfolios with an ESG tilt to them and noticed some modest outperformance. The reason we use ESGcompanies is that they are less exposed to blow-up, such as environmental disasters or labour disputes.’ De Kok: ‘At ESG there are three different parts that you have to distinguish. Governance is the easiest to link to factors. Policy in the areas of the other criteria is now more focused on exclusions.’

will be a greater standardization of the appropriate factors to consider for alternative investments.’

Van Hyfte: ‘Investors should realize that ESG might impact returns in the long run because there are investment limitations imposed. For example you have to adhere to ethical rules which excludes certain emerging countries.’

Van Hyfte: ‘We will still be in the same kind of strategic asset allocation, but I agree that it is gotten more complex. On the other hand there is a lot more standardization, because of which the possibilities for multi asset investing will increase.’

Ritzerveld: ‘Maybe in ten years time when we have more information about the way these SRI (social responsible investing) funds are performing over the long run, maybe then we can see what the performance and risk characteristics are for a group of key funds. I don’t think there will be one ESG-factor, it will be a combination of all kinds of already existent factors.’

Van Poucke: ‘We have to acknowledge that the developments in the financial markets for the last two or three years have been determined by the policy of the central banks and not by the fundamental economic developments. Therefore, for the future of multi asset investing it is of great importance whether and how that policy is going to change in the coming years. The question is when will macro-economic developments become more important again in the construction of portfolios.’

How will multi asset class investing look like, let’s say in 2020? What are the trends? McCoy: ‘There’s been continual improvement in computing power and more complex problems can be solved. But by the same token, there is an increase in the complexity of the models which will suck up this available computing power. So we are still going to be left without the perfect model. I think there will be more inclusion of multi asset class investing, and there

Brand: ‘Alternative asset classes will be sought not only in the traditional way - real estate, hedge funds, and so on - but also in established asset classes as fixed income, such as the focus on government bonds transferring to leveraged loans. The search for yield will continue and that search depends very much on market developments. Will the financial markets go to a new normal or will they stay like this for a couple of years? Investors

> Bart van Poucke became Head of Core Portfolios & Funds Benelux within the Multi Asset Solutions team of BNP Paribas Investment Partners in 2007. Prior to this, from 2001 to 2006 he worked as GIPS & Performance Measurement Specialist and as Portfolio Manager Asset Allocation (from 2006) at Fortis Investments. Van Poucke holds a Master's degree in Banking & Finance and a degree in Applied Economics.

5 / 2014

FINANCIAL INVESTIGATOR

9

// ROUND TABLE PORTFOLIO CONSTRUCTION IN MULTI ASSET CLASS INVESTING

> Jelle Ritzerveld is Director of Risk Management at Kempen Capital Management. He is responsible for financial risk management, operational risk management and performance analysis. Before he joined Kempen Capital Management in 2012, he worked at LeasePlan and ABN AMRO, also as Risk Manager. Ritzerveld obtained his Master degree in physics and astronomy, and his PhD in theoretical astrophysics from Leiden University.

will still be faced with certain return requirements so these developments will determine their investment decisions.’

hope that we will get more standardized benchmarks for factors, because factor investing will only get bigger.’

Van Nieuwenhuijzen: ‘In 2020 we will experience another shock that no one expects, so not an inflation or central bank shock. Increasing computing power will make the world more complicated; creating models remains difficult. I think the majority of the investments continues to orientate on the various asset classes, but I also think that investing will become less traditional because it will be focused more on the riskreturn ratio.’

Is it wise to combine different ways of allocation?

Ritzerveld: ‘The regular asset class allocation is here to stay. I hope that, with the additional insights of risk factors, there will be instruments by which pension funds can more easily understand and control their risks and make it comprehensible for their participants.’ De Kok: ‘In 2020 we will know more about the impact of demographics on both the liabilities of pension funds and on asset classes. Japan is an interesting case in point: it already is experiencing a lot of what we will also see. Furthermore I

Van Poucke: ‘This is what we actually do. We consider ourselves not only as asset allocators but also as risk allocators.’ Van Hyfte: ‘All kinds of allocation are interrelated because it is always about taking risks. As already expressed, a strategic asset allocation is also a risk allocation. Each investor must, depending on his horizon, estimate the risk that he is willing to take. So there is a risk based allocation.’ Van Poucke: ‘You should combine traditional asset allocation, using a 60/40 allocation, with a different allocation, like 80/20 or whatever, on top of that. Another approach is that of large hedge funds like Bridgewater, which are using leverage to boost the return of what they call their all-web portfolio to the level of return they want. The supervisor does not allow Dutch pension funds to do that.’

SUMMARY The traditional portfolio theory is not able to capture the

of the ultimate result will be decided by having the right

complexity of the financial markets. In terms of strategic

stocks and bonds at the right times. That leaves no more than

asset allocation, the financial crisis indicated that it could

10 percent for decisions based on smart beta. How will multi

sometimes be too simplistic to divide the world neatly in stocks,

asset class investing look like around 2020? There will be more

bonds and alternatives. A risk factor based approach could

use of risk factors in the allocation process and there will be a

highlight the risks, for example where equity risk is hiding

greater standardization of what will be the appropriate factors

in a bond portfolio. If investors are prepared to weather the

to consider for alternative investments. Investing will be

storm, risk-factor based allocation can be promising and add

less traditional because it is more risk return orientated. The

value. Research on factors indicates that their development

hope is that more standardized benchmarks for factors will be

is very unstable over time. The result depends on which risk

developed, because factor investing will get bigger.

factors are performing or not performing. Around 90 percent

10

FINANCIAL INVESTIGATOR

5 / 2014