BUSTING THE MYTHS IN MID- AND SMALL CAPS

BUSTING THE MYTHS IN MID- AND SMALL CAPS WARREN JERVIS PORTFOLIO MANAGER ABOUT THE AUTHOR Warren is the portfolio manager of the Old Mutual Mid & Small-Cap Fund, as well as other client portfolios. He also manages the mid- and small cap components of Old Mutual Investment Group s institutional funds. KEY TAKEOUTS: MYTH-BUSTING THESE MISUNDERSTOOD SECTORS EXPERIENCE COUNTS WITH MID- AND SMALL CAPS OFFERING A PORTFOLIO TILT THAT ADDS RETURNS FOR LESS RISK 14

For more than two decades to December 2016, mid-cap companies have generated 2.66% and small caps 2.30% a year more than the Top 40 Index. This makes these two segments of the market hugely important contributors to the potential returns generated by pension funds and lifetime retirement savings, given the significant compounding effect of those superior returns over time. Mid- and small cap shares are also an excellent differentiated source of alpha for investors and therefore should comprise a permanent component of an investor s portfolio. In the South African market, the universe of potential investment opportunities is limited and therefore it makes sense to consider the full array of investment opportunities, especially when they have generated superior returns over time. In the Old Mutual Mid & Small-Cap Fund, we invest in quality companies, with excellent management striving to achieve superior compounding returns over time. MID-CAP, SMALL CAP AND TOP 40 INDICES SINCE 1996 Clearly, experience counts in a part of the market where successful stockpicking is the primary determinant of investment success. The understanding of the business models and key business drivers enables you to leverage your information advantage in an under-researched area of the market. To successfully invest in this sector, there are a number of midand small cap sector myths that you need to consider. MYTH 1: SECTOR RETURNS ARE EASILY EXPLAINED Over certain periods of time, different sectors have been the key drivers of the mid- and small cap performance. Two examples in the mid-cap sector have been property companies and the general retailers: Property company returns have been driven by mergers and acquisitions (R230bn of capital raised), the revaluation of assets and superior yields. Retailer returns were due to superior earnings growth and a strong rerating of the shares, largely driven by the closure of rating discounts to their global peers. 1 600 1 400 1 200 FTSE/JSE Africa Mid-Cap Index FTSE/JSE Africa Small Cap Index FTSE/JSE Africa Top 40 Index 1 299.54 These sizable sectors can be excellent drivers of three to four years of outperformance, but it remains a relatively short-term phenomenon. 1 000 800 600 400 1 004.97 864.89 All asset classes still move, more or less, in line with the overall business cycle. In 2016, as an example, the Retail Index derated and its relative performance was poor. 200 0-96 -97-98 -99-00 -01-02 -03-04 -05-06 -07-08 -09-10 -11-12 -13-14 -15-16 -17 Source: I-Net Graph, uary 2017 The mid- and small cap sectors consist of more than 120 stocks, roughly 60 stocks in each, that are very diverse in nature. Business models range from quarrying, poultry production, power generation, financial services and technology, to car sales and many others. Understanding their business models, management teams, competitive strengths and where they are in the business cycle enables better investment decision-making. Indices also change over time and many businesses either come to the market, restructure or are bought out. Consider some these names: Mvelaphanda, Gijima AST, JD Group, Curro, Alexander Forbes, Sibanye and Dis-Chem. All of them have had some impact on the constituents and weightings of the various indices. All indices change over time it is a given, and certain constituent changes can have a substantial short-term impact on any given index, but in the longer term these anomalies should normalise. Take the Gold Index in 2016 as an example. At the start of the year all of the major gold stocks (AngloGold, Sibanye and Gold Fields) were in the Mid-Cap Index. By the end of July 2016, 15

the Gold Index was up 168%, and shortly thereafter all three stocks were then included in the Top 40 Index. AngloGold moved back into the Top 40 Index just prior to the other two. Post July 2016, after being included in the Top 40 Index, the Gold Index was subsequently down over 58%. Constituent changes and extreme volatility clearly had an impact on the various indices. MINING INDEX VERSUS TOP 40 INDEX 1.8 1.6 1.4 1.2 1.0 0.8 0.6 GOLD INDEX (J150) 0.4 0.2 2 900 2 800 2 700 J150 (1585.010) 2 900 2 800 2 700 0.0-96 -97-98 -99-00 -01-02 -03-04 -05-06 -07-08 -09-10 -11-12 -13-14 -15-16 -17 2 600 2 500 2 400 2 300 2 600 2 500 2 400 2 300 Source: I-Net Graph, uary 2017 2 200 2 200 2 100 2 100 2 000 2 000 1 900 1 900 1 800 1 700 1 600 1 500 1 400 1 300 1 200 1 100 1 000 900 O N D J F M A M J J A S O N D J 2015 2016 2017 1 800 1 700 1 600 1 500 1 400 1 300 1 200 1 100 1 000 900 The mining sector has clearly impacted the Top 40 Index returns over time, but so has Naspers with a massive >600% performance over an eight-year period alone. Sizable constituents also play a huge role in the overall performance of indices. Source: I-Net Graph, uary 2017 There are also individual anomalies, such as AngloGold, which entered the Mid-Cap Index in November 2015 with a 4.5% weighting, and within three months had doubled in market cap and moved back into the Top 40 Index. Three months later it began to decline in line with Sibanye and Gold Fields. The other major sector that had a significant impact on the performance of the mid-caps last year was the mining sector in general, which strongly recovered off a very low base in 2015. Overall, however, the Mining Index was a major detractor from the Top 40 Index performance over the previous five years. If you look over the longer term, the Mining Index had two very strong periods of outperformance lasting more than five years each. The short, sharp corrections post these bull markets were also fairly brutal. Over a long period of time, 21 years, it is not simply a case of saying that the Mining Index hindered the overall Top 40 Index performance. Again, sectors can have a major impact over the short to medium term, but over the longer term they have much less influence. From an attribution point of view, we looked at the Old Mutual Mid & Small-Cap Fund against the Mid- and Small Cap Index dating back to 2004 (as far back as we could go with reliable data). The average resources weighting in the index was 13.0%, with the Property Index weighting at 14.8%. The unit trust had a long-term resources underweight position and this added alpha over the period. Resources as a whole has been a negative contributor to all the main indices, such as Top 40, Mid- and Small Caps. Interestingly, despite the higher 14.8% property weighting in the index and the unit trust s low exposure to the asset class, the performance foregone as a result of being underweight was far less than expected. Despite strong Property Index performance over the last 10 years, the portfolio exposure to performing industrial shares limited the damage of a property counter underweight. Regardless of what the sectoral drivers are in the short term, or what the various constituents of the index are, the fact remains that long-term assets should still be suitably exposed to the midand small cap area of the market. It is a sizable opportunity set of potential investment opportunities that should not be ignored. 16

MYTH 2: SUPERIOR ALPHA GENERATION IS IN THE TOP 40 STOCKS The average annual performances for the three sectors over 21 years are: Top 40 Mid-Cap Small Cap 15.78% 18.43% 18.08% RETURNS PER UNIT OF RISK Top 40 Mid-Cap Small Cap Std Deviation 23.3% 20.2% 22.4% Maximum 79.17% 47.53% 55.72% Minimum -23.58% -27.40% -31.20% Median 13.31% 21.39% 20.60% Return/Risk Unit 67.8% 91.2% 80.8% Source: Old Mutual Investment Group, uary 2017 Source: Old Mutual Investment Group, uary 2017 Mid-caps have outperformed the Top 40 by an average of 2.66% per annum and small caps have outperformed the Top 40 by an average of 2.30% per annum. The compounding effect of these superior annual returns is astounding. When looking at the compounding of superior returns over a longer time period, we did the following simple exercise. Investing R1 million into each of the three segments over 21 years would yield a 70.6% better result in the mid-cap sector (growing to R25 117 783), and a 50.3% better result in the small cap sector (R22 137 801) than the Top 40 sector (R14 725 051). This is a significant difference generated by better annual results compounding over a 21-year period. The young are encouraged to start saving early as the compounding effect, over long periods of time, of returns on their retirement savings is quite staggering. Adding exposure to the mid- and small cap sectors of the market would further enhance portfolio returns. MYTH 3: MID- AND SMALL CAPS ARE RISKY The level of return per unit of risk (as measured by the standard deviation) of small to mid-cap stocks versus the Top 40 Index has also been superior. Looking at the average return divided by the standard deviation gives return per unit of risk figures of 67.8% for the Top 40 Index, 91.2% for mid-caps and 80.8% for small caps. This is not a definitive risk answer, as the definition of risk is way more complex than that. However, it does help in understanding whether mid- and small cap shares are too risky for investors or not. One of the key parameters of risk management for me as a portfolio manager is the protection of clients capital. There are many types of risks out there and at all times a portfolio manager needs to consider an appropriate risk return tradeoff when making investment decisions. Avoiding many of the share calamities is a key component of mid- and small cap investing, and that is where the experience of the portfolio manager is vital. I am often told that mid- and small caps are risky regardless of what the return per unit of risk numbers say. So I decided to look at the unit trusts of the mid- and small cap peer group versus the general equity unit trusts in the same asset management house over the last five years. Quite simply, you have returns on one axis and standard deviation (risk) on the other axis. Although standard deviation is not a comprehensive risk measure, as mentioned, it is nevertheless appropriate for graphical purposes. STANDARD DEVIATION AND RETURNS OF MID- AND SMALL CAP VERSUS GENERAL EQUITY UNIT TRUSTS Annualised 5-Year Total Return (%) 16 15 14 13 12 11 10 9 Risk/Return: 10 years ending 31 October 2016: Mid- & Small Cap and Core Equity Funds Mid- & Small Cap UT Peers Core and specialist equity funds SWIX Nedgroup Inv Rainmaker R Coronation Smaller Companies Nedgroup Inv Entrepreneur R SIM General Equity R Old Mutual Mid & Small-Cap R SIM Small Cap R Investec Equity R SWIX Old Mutual Investors R Momentum Small/Mid-Cap A Momentum Equity A Coronation Top 20 A Investec Emerging Companies R 8 12.0 12.5 13.0 13.5 14.0 14.5 15.0 15.5 Risk (Annualised 5-Year Standard Deviation) Source: Morningstar Source: Morningstar 17

Of the six asset managers that were chosen, and assuming the same overall house view assumptions applied, there were five mid- and small cap unit trusts with lower risk profiles than the asset managers general equity funds. The only outstanding fund was a fund that was recently decimated by a large holding in African Bank during the assessment period. So over the last five years, the mid- and small cap funds have not been more risky than the general equity unit trusts in the same asset managers. MYTH 4: TOO ILLIQUID TO INVEST IN I hear too often from general equity portfolio managers that the sector is too illiquid for large asset managers. I agree. However, where does the management of liquidity risk fit into the overall investment process and what are the parameters being used to determine liquidity? Quite simply, liquidity is a function of the assets under management in which you are trying to invest. The more funds you manage, the harder it will be to participate in the small cap sector and a natural shift towards more mid-cap stocks will occur. In our mid- and small cap investment process, liquidity is the first starting point of the process and is a key risk to analyse. It is vital to understand how long it will take to move stock into and out of the portfolio, and what the overall equity holding will be in the company once the targeted portfolio weighting has been reached. Many stocks such as Bell, Basil Read, Hulamin, Caxton, Mustek and Rainbow are too illiquid to consider for a fund that has even assets under management of R2 billion. In our mid- and small cap universe, 24 stocks out of 123 stocks available are currently excluded based on liquidity issues. So slightly less than one-fifth of the universe is currently excluded, and this number of exclusions will slowly increase as assets are accumulated. Given liquidity concerns, actual practical experience still provides more insight than a detailed spreadsheet, and in the recent past we have also been able to transition over R3bn within a three-month period in the sector. So provided you have a quality portfolio of stocks that are appropriately priced, you will always find willing buyers or sellers in the market. However, trying to exit a 20% stake in a very poor business is not an easy task when things go wrong in the company. Liquidity is a constraint, but it is still not enough of a reason to completely exclude the sector purely on liquidity concerns. It is a vital part of the investment process and needs to be managed carefully. 18

SOLUTION: A MID-CAP AND SMALL CAP TILT IN THE PORTFOLIO Small and mid-cap shares offer an uncorrelated additional source of alpha within an overall portfolio and thus, as a large asset manager, it is appropriate to have some exposure to these companies. The extent of this portfolio tilt depends on the level of knowledge of these companies and the proportion of the portfolio that is able to be invested in equities in general. Mid- and small caps are never meant to be a substitute for Top 40 exposure. They should be considered as a portfolio tilt that adds additional returns for less risk. Consider a 100% Top 40 portfolio versus three portfolios with tilts of 10%, 20% and 30% respectively towards mid-caps only. Over 21 years, the additional 10% tilt to mid-caps would have yielded an additional 7.1% return; a 20% mid-cap tilt a 14.1% return, and the 30% mid-cap tilt would have yielded a 21.2% additional return. R100 initial investment OLD MUTUAL MID AND SMALL-CAP UNIT TRUST VERSUS THE PEER GROUP AVERAGE SINCE SEPTEMBER 2009 R300 R250 R200 R150 R100 R50 Aug-09 Dec-09 Cumulative Excess Value (RHS) Old Mutual Mid & Small-Cap Fund (ASISA) South African EQ Mid-/Small Cap Peer Group Average Apr-10 Aug-10 Dec-10 Apr-11 Aug-11 Dec-11 Apr-12 Aug-12 Dec-12 Apr-13 Aug-13 Dec-13 Apr-14 Aug-14 Dec-14 Apr-15 Aug-15 Dec-15 Apr-16 Aug-16 Dec-16 R80 R60 R40 R20 R0 - R20 Source: Old Mutual Investment Group (Pty) Ltd, uary 2017 Cumulative Excess Value (Rand) As a small and mid-cap manager, we strive to generate alpha for our clients and we believe we have a competitive advantage in the sector. We have achieved superior performance against the peer group average over 1-, 2-, 3-, 5- and 10-year periods. Given the superior risk-adjusted returns delivered by the midand small cap segments of the market, can you ignore the value they could potentially add to your retirement savings portfolio over time? We think not. Note: The performance quoted is for a lump sum investment and in respect of the Old Mutual Mid & Small-Cap Fund. The actual highest, average and lowest 12-month return figures since inception to 31 uary 2017 are 108.3% (highest), 17.1% (average) and -40.3% (lowest). The fund was launched on 30 April 1997. Performances are in ZAR and as at 31 uary 2017. 19