For Professional Use Only

July 2015

FocusPoint

In-depth insights from NN Investment Partners

The combination of independent fundamental and quantitative research alpha and rules-based portfolio construction is a key distinctive factor of the Optimised Portfolio Strategies of NN Investment Partners.

Optimised Portfolio Strategies • NN IP’s Optimised Portfolio Strategies combine two independent alpha sources – fundamental and style • The aim is to create consistent alpha by constructing portfolios based on proprietary fundamental and quantitative research • Disciplined portfolio construction & risk management should lead to robust performance Jeff Meys – Head of Optimised Portfolio Strategies

www.nnip.com

FocusPoint

July 2015

Optimised Portfolio Strategies Capturing fundamental and style alpha Introduction

The optimised portfolios are actively managed by a dedicated team, consisting of highly experienced equity portfolio managers. The portfolio managers are responsible for the day-to-day ­management of the strategies by actively monitoring holdings, risk exposures, the Environmental, Social and Governance (ESG) ­profile and corporate actions. The objective of the portfolio ­managers is to efficiently realise consistent alpha by identifying unrecognised value based on fundamental and quantitative research. The portfolios are constructed in such manner that they optimally capture the fundamental research alpha whilst ­maintaining an intended style exposure and risk profile. On average, the portfolios have become more concentrated with a higher active share and a lower number of stocks. NN IP has made these enhancements in its core equity and sector strategies to align with client needs and to keep up with developments in capital markets.

The Investment Process The optimised portfolio strategies combine the fundamental research of our Global Equity Research and Quantitative Research & Strategy teams. As such, we combine two independent sources of alpha.

Fundamental Research The Global Equity Research team consists of 24 experienced ­analysts with diverse international backgrounds. The analyst team is NN IP’s global knowledge centre on company research. It is the key investment idea generator for individual stocks. A universe of about 2,500 stocks is continuously monitored with a focus on idea generation. The universe is split across the 24 analysts along ­sector lines. About 1,200 of the 2,500 stocks that the analysts track are covered by a proprietary company model and fair value estimate. Each analyst is responsible for identifying key trends and themes within their sector. The objective is to find the best

alpha opportunities within their respective universe. NN IP’s focus on global coverage provides the analysts with broad insights and enhances their alpha generation capacity. Analysts’ insights are supported by corporate access to the management teams of the companies that are covered as well as access to sell-side research and consultant insights.

Figure 1: Composition of the Global Equity Research team 24 equity analysts with 14 years average experience

ANALYSTS

NN Investment Partners (NN IP) has enhanced its Core Equity Strategies (Global, Europe and Euro Equity) and its ten Sector Equity Strategies. These strategies have been converted into Optimised Portfolio Strategies (OPS). They combine proprietary fundamental and quantitative research with rules-based portfolio construction and risk control. The Global Equity Research (GER) team, consisting of 24 analysts, is responsible for delivering ­fundamental research alpha. The Quantitative Research & Strategy (QRS) team, consisting of 13 strategists and analysts, is responsible for generating quantitative research alpha, risk ­control and portfolio optimisation. The combination of independent fundamental and quantitative research alpha and rules-based portfolio construction is a key distinctive factor of the optimised portfolio strategies.

Energy & Industrials 4 analysts

Consumer Goods 4 analysts

Financials & Real Estate 5 analysts

Credit Analysts & ESG Specialists

TMT 5 analysts

Materials 2 analysts

Health Care 2 analysts

Utilities 1 analyst

Source: NN IP

The analysts produce in-depth, bottom-up company analysis. This includes detailed financial models, valuations, ratings and scenario analyses. In addition, the mainstream analysts perform extensive ESG factor analysis to improve the overall investment analysis and conclusions. For the ESG analysis the team is ­supported by ESG specialists. When necessary, there is also close contact with NN IP’s credit analysts to review the liabilities. In addition, the fundamental research analysts create and manage their own model portfolio consisting of their highest-conviction ideas. The alpha of these portfolios is measured per analyst to ensure personal accountability and alignment with portfolio ­managers. In aggregate these model portfolios contain about 400 stocks globally and have generated an information ratio of 0.7 over the past 5 years. The high conviction model portfolios of the analysts are used as one of the key inputs for the Optimised Portfolio Strategies. The expected returns input of the fundamental research analysts is derived by ranking the active weights in the model portfolios. Besides generating investment ideas through their model port­ folios, the analysts also play a key role in performing a funda­ mental “sanity check” on the optimised model portfolios in ­collaboration with the portfolio managers. The aim of these sanity checks is to verify that the Optimised Portfolio Strategies are not exposed to issues that may be overlooked by a risk model, ­quantitative expected return models and optimisation algorithms. Examples include corporate actions, political and environmental risk, accounting issues, change of management and M&A.

2

July 2015

The Quantitative Research & Strategy team consists of 13 ­investment strategists, analysts and quantitative developers. The activities of the QRS team are integral to a multitude of investment processes within NN IP. On the equity side of NN IP, the team is responsible for developing and implementing quantitative alpha factors that lead to stock selection, factor positioning and investable universe screening. In addition, the quantitative equity team provides support to all fundamental equity teams in the form of portfolio construction, tooling, risk and drawdown analysis. For the OPS strategies, the QRS team is responsible for modelling expected returns using a multi-factor alpha model and for ­combining the fundamental and quantitative alphas, portfolio ­construction and risk control.

Quantitative Equity Factors In our multi-factor alpha model we use stock characteristics as factors to predict which stocks will outperform and which stocks will underperform. For example, a well-known stock characteristic is the dividend yield, which differs across stocks. Keim (1985) finds that for NYSE-listed dividend paying stocks, the highest dividend yielding stocks outperform the lowest yielding stocks by 3.5% per year. Hence, the dividend yield is a useful Value factor to include in a multi-factor model since it contains useful information for modelling expected returns. The Quant Equity Multi-Factor model consists of several hierarchical layers with factor groupings at the highest level being defined by Value, Momentum, Quality and Profitability. Each of these factors is subsequently subdivided into factor composites that make up the next layer in the model. For instance, within momentum we use price momentum, earnings revisions and short-term reversals as the three second-layer factor composites. The third and final layer of the model contains individual factors – for example, the price momentum composite includes 6-month, 9-month and 1-year price momentum. The first layer (factor families) and the second layer (factor composites) of the model are weighted to maximise the predictive power and stability of the model, whereas the factors within the composites are equal-weighted. The factors in the composites are selected based on their economic rationale, academic evidence and replicable in-house back-test results. Below we provide further background on the highest level factor definitions. Value Value factors are measures of cheapness and have an established history among academics and practitioners. Stocks that are cheap relative to their dividends (Keim, 1985), earnings (Basu, 1977), sales, book value and cash flow (Jensen et al., 1988), on average tend to outperform stocks that are expensively valued on these metrics. Our value factor incorporates many measures of cheapness since this provides a more complete picture. In general, the use of more than just one factor increases alpha and reduces risk since using only one factor might cause the investor to overlook other aspects that are related to its future risk-reward profile. An example would be a stock that has a very high dividend yield but very low or negative earnings and cash flows. Looking only at

the dividend, the stock seems attractive; however, its poor earnings and cash flow illustrate that the stock might not be cheap at all and that the dividend might not be sustainable.

Figure 2: Cumulative Value (price-to-book) performance (1964 - 2014) 10.000

100 Logarithmic

Quantitative Research

1

0

1964

1974

1984

1994

2004

2014

Source: Kenneth R. French (http://mba.tuck.dartmouth.edu/pages/faculty/ ken.french/data_library.html)

Momentum Momentum factors are technical and sentimental in nature. Jegadeesh and Titman (1993) find that buying the 10% best performing stocks in the past six months and selling the 10% worst performing stocks and holding this portfolio for six months generates excess returns of 12% per year. Earnings surprises and revisions in analysts’ estimates of future stock earnings are closely related to momentum strategies. Givoly and Lakonishok (1979) find that stocks that experience upward (downward) revisions in analysts’ earnings estimates produce significant positive (negative) abnormal returns in the months surrounding the revision. Price momentum and earnings revisions are often explained by the underreaction of market participants to recent information. Hence, exceptionally good (bad) news for specific stocks tends to be followed by more good (bad) news and these trends tend to last for several months. In contrast to the medium-term horizon mispricing caused by underreaction, Lehmann (1988) finds strong evidence of overreaction to very recent news that lasts for several weeks. This short-lived overreaction causes price reversals that can last up to a month. For this reason most price momentum strategies (including Jegadeesh and Titman, 1993) exclude the most recent month when forming winners minus losers portfolios. Price momentum, earnings revisions and short-term price reversals are second-layer factor composites in our momentum factor.

3

FocusPoint

July 2015

Figure 3: Cumulative Momentum performance (1964 - 2014)

Novy-Marx (2013) finds that buying the most profitable stocks and selling the least profitable stocks yields a highly significant return. In addition, Novy-Marx detects that such strategy has a negative exposure to Value. This result supports the positive relation between profitability and growth. While such growth stocks historically have shown to underperform, when adjusting for profitability the reverse is true, illustrating that growth stocks with strong profitability outperform. For this reason we combine growth in earnings, dividends and margins with profitability factors. Due to their successful interplay, profit and growth make up the second layer factor composites in our profitability factor.

10.000

Logarithmic

100

1

Back-testing the four factor families

Profitability Profitability and growth are closely related since highly profitable firms grow their earnings and dividends faster. Haugen and Baker (1996), Cohen et al. (2002), Fama and French (2006) and Ball et al. (2014), find that future returns are positively related to current measures of profitability.

200%

150%

100%

50%

2014

2013

2012

0% 2011

We model these characteristics by using stability, efficiency and balance sheet composites. Stability includes factors that are closely linked to low-volatility investing. Low-risk investing is based on the assumption that high-volatility stocks exhibit lower expected returns than low-volatility stocks. Haugen and Heins (1975) find a negative relation between risk and return, and Jagannathan and Ma (2003) document that minimum-variance portfolios outperform cap-weighted benchmarks. Frazzini and Pedersen (2014) find that a portfolio that shorts high-beta stocks and buys low-beta stocks produces significant risk-adjusted alphas. Efficiency factors include measures related to accruals, capital expense, inventory build-up, cash-flow and sales relative to assets or employees. Accruals are a key indicator of earnings quality. Earnings components that are stable over time are more valuable to investors than non-recurring income components. Ou and Penman (1989) and Sloan (1996) find that stocks with ­relatively low accruals earn higher returns and have less volatile earnings and returns than stocks with high accruals. Titman et al. (2004) show that the relation between capital expenditure and stock returns is negative. Within the balance sheet composite we mostly focus on factors that improve the financial health of companies.

Figure 4: Back-test result of the four factor families used in our multi-factor alpha model (2002 - 2014)

2010

Quality Quality factors are used to obtain exposure to firms that are stable, have a good reputation, a competitive advantage, are financially sound and efficiently managed. Typically these stocks exhibit a better-than-average ability to deliver sustainable returns whilst having a lower risk profile.

2009

(http://mba.tuck.dartmouth.edu/pages/faculty/ken.french/data_library.html)

2008

Source: Kenneth R. French

2007

2014

2006

2004

2005

1994

2004

1984

2003

1974

2002

1964

2001

0

A back-test of the four factor families that we use is presented in Figure 4 for the 2002-2014 period. At the end of each month the 2,500 largest stocks at that time are scored using our value, momentum, quality and profitability factors. Returns are generated by ranking stocks on each of the factors from most to least attractive. To obtain factor composite rankings we calculate the average ranking, using all the individual factor rankings within a composite. Subsequently the value, momentum, quality and profitability rankings are constructed by taking the weighted-average ranking of all factor composites. The total model score is then a weighted-average of these four rankings.

-50% Value

Momentum

Quality

Profitability

Total Model

Source: NN IP

For back-testing purposes the 2,500 largest developed global stocks are ranked at the end of each month. The factor/model returns are generated by buying the equally-weighted top 20% ranked stocks and selling the bottom 20% stocks at month-end and holding these positions through the next month. At the end of the next month we then record the strategy return for that month and repeat this process with the 2,500 largest stocks at this new month-end.

4

July 2015

As exemplified by the back-tests, each of the four factor groups generates positive alpha over the sample period. The performance of each of the factors, however, varies over time. The correlations between factor returns are dynamic and vary across the different phases of the business cycle. For this reason we diversify risk over different factors groupings. Doing so increases alpha and reduces risk. Because the value and momentum factors are negatively correlated, they make up a larger part of the total model.

-0.2. In addition, value and quality are typically negatively ­correlated, as excellence at a cheap price is hard to find in equity markets. The same holds for profitability and value as investors will quickly bid up the prices of cheap but highly profitable firms. The correlation between momentum and profitability and quality typically varies over the business cycle. For this reason, combining different ­factor exposures improves the risk/reward profile of a portfolio.

Combining Fundamental and Quantitative Research Alphas

Portfolio Construction

The Optimised Portfolio Strategies maximise the combined ­fundamental and quantitative research alphas subject to several constraints. Sophisticated algorithms are used to balance the return and risk characteristics of the portfolio while taking into account practical implementation guidelines regarding ­compliance, portfolio t­ urnover, ESG and the liquidity of the fund.

Value added through diversification of alpha signals The Fundamental and Quantitative Research alphas generated positive returns and are negatively correlated. Combining these information sources results in less volatile returns and smaller drawdowns. Risk-adjusted returns increase with the ability to add value through diversification across alpha signals. In the multi-­ factor model we also combine negatively correlated factors. It is well-known, for example, that value and momentum are negatively correlated. Over the 1964-2014 sample we find a correlation of

The next step in the investment process is to translate the ­combined alpha signals into the actual model portfolio. The QRS team determines the optimal portfolio that maximises alpha transfer while obeying numerous constraints and restrictions related to risk, concentration and diversification. In our portfolio construction process we use two different types of restrictions, “hard” and “soft” ones. Hard restrictions are satisfied at each rebalancing whereas in the optimisation the soft restrictions are targeted but not necessarily satisfied at every rebalancing. Instead, unwanted deviations from soft restrictions are penalised in the objective function, resulting in minimal deviations. The use of soft restrictions ensures feasibility of the optimisation problem and avoids unnecessary small trades that would be necessary to enforce these restrictions if they were rigorously imposed.

5

FocusPoint

July 2015

OPS Portfolio Management team

Conclusion

The OPS team, consisting of experienced portfolio managers, facilitates an efficient translation of the model portfolio into the actual portfolio and continuously monitors the performance and risk. The portfolios are rebalanced on a monthly or bi-monthly basis. In exceptional cases such as financial market turmoil or a sell recommendation from an analyst, the portfolio manager takes appropriate action and is supported by the QRS team to make the necessary adjustments. The team also takes on the role of Client Portfolio Manager and as such forms the link between the client and the portfolio. In the implementation phase the OPS team starts with a pre-trade analysis. Together with the GER team they perform sanity checks on the stocks in the model portfolio. Key aspects include issues that cannot be ­captured through quantitative data and risk models. These include potential ESG issues, M&A, political risk, accounting issues, m ­ anagement change and corporate actions. The Portfolio Management team also checks that the portfolio is aligned with client and regulatory guidelines.

The Optimised Portfolio Strategies aim to create consistent alpha by constructing portfolios based on a unique combination of ­proprietary fundamental and quantitative research. The portfolios combine unrecognised value identified through fundamental ­analysis, intended style exposures and rigorous portfolio ­construction and risk control.

The final step is the actual implementation of the model portfolio. The OPS team communicates the orders to our Global Trading team that executes the orders. After the implementation a posttrade analysis is performed. In case of extreme external circumstances or a situation where there is a significant impact on a ­specific stock that needs immediate attention, the OPS team will trigger a re-run of the model portfolio. This ensures that the most recent data is reflected and stock-specific information can be implemented immediately if necessary.

The investment process of the Optimised Portfolio Strategies is unique as it c ­ ombines two independent alpha sources – fundamental and style – which ultimately results in a concentrated ­portfolio with a higher active share and higher active specific risk.

Figure 5: Capturing fundamental and style alpha into an optimised portfolio Alpha Generation Fundamental research • Large pool of fundamental analysts • Proven track record

Portfolio Construction

Portfolio Management

Portfolio optimisation • Optimal transfer of multiple research alpha sources • Effective diversification • Risk control

Day-to-day portfolio management • Pre-trade analysis • Portfolio implementation • Risk management • Performance monitoring

Optimised model portfolio

Efficient translation to actual portfolio

Fundamental research by a large pool of experienced equity ­analysts with a proven track record is combined with extensively back-tested quantitative research of well-established styles such as value, momentum, quality and profitability. This results in multiple strong and independent alpha sources. Through diversification across alpha sources, risk-adjusted returns increase. Sophisticated algorithms are used in the portfolio optimisation process. The portfolio management team then facilitates an ­efficient translation of the model portfolio into the actual portfolio while continuously monitoring performance and risk.

Authors Jeff Meys – Head of Optimised Portfolio Strategies Karim Bannouh – Quantitative Research & Strategy Jeroen Bos – Head of Equity Specialties Bas Peeters – Head of Quantitative Research & Strategy Marc van Loo – Head of Global Equity Research

Quantitative research • Extensively back tested • Well-established styles

Multiple strong & independent alpha sources Source: NN IP

6

July 2015 References: Basu, S., 1977, Investment performance of common stocks in relation to their Price/Earnings ratios: a test of the Efficient Market Hypothesis, Journal of Finance, pp. 663-682. Cohen, R., P. Gompers, and T. Vuolteenaho, 2002, Who under­ reacts to cash-flow news? evidence from trading between individuals and institutions, Journal of financial Economics, pp. 409-462. Fama, E. and K. French, 2006, Profitability, investment and ­average returns, Journal of Financial Economics, pp. 491-518. Frazzini, A. and L. Pedersen, 2014, Betting against beta, Journal of Financial Economics, pp. 1-25. Givoly, D. and J. Lakonishok, 1979, The information content of financial analysts’ forecasts of earnings: some evidence on ­semi-strong inefficiency, Journal of Accounting and Economics, pp. 165-185. Haugen, R. and A. Heins, 1975, Risk and the rate of return on ­financial assets: some old wine in new bottles, Journal of Financial and Quantitative Analysis, pp. 775-784. Jegadeesh, N. and S. Titman, 1993, Returns to buying winners and selling losers: impmlications for stock market efficiency, Journal of Finance, pp. 65-91. Jensen, G., Johnson, R. and J. Mercer, 1998, The inconsistency of small-firm and value stock premiums, The Journal of Portfolio Management, pp. 27-36. Keim, D., 1985, Dividend yields and stock returns: implications of abnormal January returns, Journal of Financial Economics, pp. 473-489. Lehmann, B., 1990, Fads, martingales and market efficiency, Quarterly Journal of Economics, pp. 1-28. Novy-Marx, R., 2013, The other side of value: the gross profitability premium, Journal of Financial Economics, pp. 1-28. Ou, J. and S. Penman, 1989, Financial statement analysis and the prediction of stock returns, Journal of Accounting and Economics, pp. 295-330. Sloan, R., 1996, Do stock prices fully reflect information in accruals and cash flow about future earnings?, The Accounting Review, pp. 289-315. Titman, S., Wei, K. and F. Xie, 2004, Capital investments and stock returns, Journal of Financial and Quantitative Analysis, pp. 677-700.

7

FocusPoint

Disclaimer The elements contained in this document have been prepared solely for the purpose of information and do not constitute an offer, in particular a prospectus or any invitation to treat, buy or sell any security or to participate in any trading strategy. This document is intended only for MiFID professional investors. While particular attention has been paid to the contents of this document, no guarantee, warranty or representation, express or implied, is given to the accuracy, correctness or completeness thereof. Any information given in this document may be subject to change or update without notice. Neither NN Investment Partners B.V., NN Investment Partners Holdings N.V. nor any other company or unit belonging to the NN Group, nor any of its officers, directors or employees can be held direct or indirect liable or responsible with respect to the information and/or recommen­ dations of any kind expressed herein. The information contained in this document

July 2015

cannot be understood as provision of investment services. If you wish to obtain investment services please contact our office for advice. Use of the information contained in this document is solely at your risk. Investment sustains risk. Please note that the value of your investment may rise or fall and also that past performance is not indicative of future results and shall in no event be deemed as such. This document and information contained herein must not be copied, reproduced, distributed or passed to any person at any time without our prior written consent. This document is not intended and may not be used to solicit sales of investments or subscription of securities in countries where this is prohibited by the relevant authorities or legislation. Any claims arising out of or in connection with the terms and conditions of this disclaimer are governed by Dutch law.

8