The Coronation Global Emerging Markets Flexible fund returned 18.7% in the third quarter of 2017, which was 6.8% ahead of the benchmark’s return of 11.9%. The one and two-year performance figures are also strongly positive in both absolute and relative terms, recording 4.6% alpha over one year and 9.3% alpha p.a. over two years. While this short-term performance is pleasing - particularly after a tough period from late 2014 to late 2015 - it is the long-term performance that matters and is what we believe best illustrates the alpha generation ability of the fund. In this regard, the fund’s outperformance after fees since inception almost 10 years ago has been 2.8% p.a.

The biggest contributors to the fund’s outperformance over the last quarter were the two Brazilian education stocks, Kroton and Estácio. Their attempted merger was blocked just before the end of the previous quarter which resulted in a sharp share price decline for Estácio in particular. We believed that both Kroton and Estácio remained very attractive on a standalone basis and as such continued to hold close to 8.5% of fund (combined) at one point in the two businesses. During the quarter Estácio’s share price more than doubled in USD, while Kroton’s was up 35% in USD and collectively they contributed around 3.5% to alpha over the quarter. Estácio outperformed Kroton as their results for the second quarter were well ahead of expectations and investors reappraised the longer-term outlook for Estácio’s margins. Historically, Estácio’s operating margins have been less than half of Kroton’s level - mid-teen margins versus low 30s for Kroton. This large difference in profitability is partly why Estácio carried a lower rating than Kroton and one of the reasons why Kroton wanted to buy Estácio. This margin gap has already started to close and we believe Estácio can narrow the gap to Kroton even further, which brings down the multiple quickly as one looks further out; although, Kroton’s scale advantage will mean that Estácio’s operations are unlikely to ever reach Kroton’s margin on a sustainable level. We did reduce the Estácio position over the past few months but we believe it is still attractive. Today just over 7% of the fund is invested in the Brazilian education stocks with Kroton being a 5.0% position and Estacio a 2.2% position.

The Chinese internet stocks 58.com and Baidu were the next biggest contributors. The former was up 43% in USD (contributing over 1% to alpha), while the latter was up a more modest 38% and contributed 60bps to alpha. In Baidu’s case, the struggles of last year that came with cleaning up their advertiser base seem to have largely passed, while profitability has improved significantly, leading to earnings per share increasing by 70% year-on-year in the second quarter. The big concerns in the last 18 months with 58.com have centred around its property business as the company is the leading player in advertising existing homes sales. With the Chinese government having clamped down on the property market in a bid to reduce the potential of an asset bubble developing, investors sold out of 58.com, which created a buying opportunity for us. The results released for the second quarter saw the company beating their own previous guidance by almost 10%, with 33% revenue growth (a rate that they achieved for the first half of the year as a whole). More critically, there are signs of significantly improved profitability as they have gone from making losses in the first half of last year to now earning operating margins of 14%. The property business has held up well in spite of the issues in the broader market, which is not unexpected as it is more akin to a membership service for estate agents rather than an online estate agent that tries to earn commissions on sales – an inherently more cyclical business. Despite the sharp move in the share price, we still believe 58.com remains very undervalued: it has largely dominant positions and several years of high revenue growth ahead of it in all its key verticals (online blue collar jobs, online housing and general online classifieds), while the reduced investment needs and natural operating leverage that comes with higher revenue can see margins expand substantially from current levels.

The Russian stocks Sberbank (+37% in USD), Magnit (+20% in USD) and X5 Retail group (+29% in USD) all also came through strongly over the period. Brilliance China Automotive was up 32% in USD in the quarter, bringing the one-year return to 117%. We had been consistently reducing our position as it appreciated and sold out of the position during the quarter as it reached our fair value.

The largest new buy during the quarter was a 3.3% position in Airbus. Airbus earns around 55% of revenue from emerging markets (EM), and the share of its order book (future confirmed orders by airlines) from EM airlines is 60% and rising. While Airbus’s earnings have historically been erratic, its capital intensity high and its return on capital low, we believe there have been some material changes in the way Airbus is owned and managed that make the past history a poor guide to its earnings power going forward. The most important of these changes is the reduced influence of the European state shareholders that historically drove the strategic decisions of Airbus. The company was originally formed by European nations to provide an alternative to US dominance of the industry, so the level of interference by European governments has always been a big detractor to the investment case. Since 2012, however, the aggregate shareholding of the major European governments (France, Germany and Spain) that ran Airbus reduced to below 30% and the company has been professionally managed with limited outside influence. This change makes it far less likely that Airbus will embark on value destructive developments like the A380 superjumbo
Today Airbus has an offering in all core segments of the market - from narrow-body single aisle aircraft (the A320 series and its derivatives) through to the A380 Superjumbo. Crucially, they have the A350 which is ramping up production, which will compete with Boeing’s 777 and 787 and was the one big gap in their portfolio. Having a full range of offerings is important because many airlines prefer to operate only Boeing or Airbus to achieve economies of scale on their purchases, maintenance as well as pilots who are typically certified to fly only one of the two manufacturers. Despite the cyclicality of air travel demand in individual countries and regions, the overall demand for air travel has grown almost uninterrupted since the early 1970s (growth of 2x global GDP over the last 40 years) when mass market commercial travel first became affordable. This period has spanned several oil and geopolitical crises, as well as the dotcom bubble and Global Financial Crisis. In addition, 80% of the world’s population have still never flown on an airplane. Importantly, Airbus’s order book has a 10-year backlog - meaning that at current rates of production it would take 10 years to fulfil all their existing confirmed orders.

Airbus has historically earned low single-digit margins as for most of the last two decades Airbus was effectively State owned and lots of capital was spent developing products with limited commercial appeal, such as the A380. Under professional management we do not believe this mistake will be repeated. The quirks of accounting also means that in the early years of developing an aircraft there are significant losses that depress earnings (as has been the case over the past few years), while once a new model enters into mass production (as is the case now) the margin increases quickly as these losses fall away. A look across Airbus’s product portfolio shows all their major designs are well beyond the development phase and, in particular as the A350 production ramps up, operating margins can reach double digits within the next few years. Besides the ramp up of the A350, higher pricing on the new A320neo model (the new version of the existing smaller short-haul model) will also assist margins, as will the unwinding of FX hedges. As a reference point, Boeing already generates double-digit operating margins and are targeting mid-teens margins. Importantly, the next several years will also see lower R&D and capex (Airbus is coming out of a heavy investment cycle) and hence higher free cash flow generation (c. 80% free cash flow conversion) and return on invested capital (19% ROIC from around 10% today). While corporate governance has improved at Airbus over the past few years, there are still risks in this regard and this remains a cyclical business that will be negatively impacted by extreme global events as well as currency movements. We do however factor this into our valuation (using a lower multiple to value the business than would otherwise be the case) and we still get substantial upside to fair value for what is in effect a global oligopoly (Airbus and Boeing together have 90% market share) in a structurally growing category (global air travel) with very visible earnings and free cash flow growth over the next five years. We continue to travel widely to meet companies we own or are interested in purchasing for the fund. During the quarter, the team undertook several trips, including meeting company management on trips to China and India. The big moves in share prices have seen us reduce, or sell out from, positions that are approaching fair value and re-invest in either new ideas (like Airbus), or buy additional exposure to some stocks that we believe are very attractive but have lagged in recent months (such as Magnit). The overall upside of the portfolio - our assessment of fair value versus current share prices - remains around 40%.

Portfolio managers
Gavin Joubert and Suhail Suleman as at 30 September 2017

The Coronation Global Emerging Markets Flexible fund returned 4.4% during the quarter, compared to the 3.5% of the benchmark MSCI Emerging Markets total return index, representing outperformance of 0.9% for the period. Year to date, the fund appreciated by 12.2%, marginally behind the 13.0% return of the index. Since inception nine and a half years ago, the fund has appreciated by 9.8% per annum, resulting in outperformance of the index of 2.2% per annum.

The biggest contributor to alpha for the period was JD.com; the share was up 26% in US dollar (JD.com is discussed in further detail below.) Yum China, which had previously been a detractor since being spun out from Yum Brands, was up 44% in US dollar in the quarter. Other notable contributors were 58.com (+25%), Naspers (+12%), Heineken (+15%) and Melco Resorts and Entertainment (+22%). The positive contribution from Naspers was more than offset by not owning Tencent directly, as the value of the Naspers stake in Tencent increased by even more, rising ever higher above the market value of Naspers standalone (the value of Naspers stake in Tencent is now 35% higher than Naspers market capitalisation). Besides Tencent, not owning Samsung Electronics detracted. Amongst the stocks the fund owned, notable detractors were Magnit (- 10.4%), Estácio (-10.5%) and Tata Motors (-7.3%).

Year to date, the largest detractor from performance has been Magnit, which has declined by 22%. In this regard, we spent the last week of June in Russia, visiting Moscow (X5’s base), Krasnodar (home of Magnit) and Saint Petersburg (Lenta’s home town). We met with senior management of the three food retailers; Magnit (COO and CFO), X5 Retail (CEO) and Lenta (CEO). In summary, our positive view on the sector and these three individual companies (who together make up 6.7% of the fund) was confirmed. In particular, our conclusion on Magnit (which at 4.0% of fund is the largest position of the three) was that its current issues are of a short-term, cyclical nature as opposed to being structural. Over the past year or so, Magnit has gone from being a market darling to being almost universally disliked (with X5, the exact opposite is the case) due to negative like-for-like sales, and the usual extrapolation of current experience by short-term focused market participants. We believe that Magnit management is making the right adjustments to its value proposition as well as doing the required refurbishments to keep up with a newly invigorated X5, who is increasingly moving into its markets. These changes take several quarters to reflect in results, but should eventually see a return to positive same-store sales and contribute to maintaining its margins. In addition to this, the long-term opportunity still holds: large formal retailers will continue to take market share. In this regard, it was clear from our trip to Russia that just about every other large food retailer in Russia (Auchan, Metro, Dixy, O’Key) is struggling and collectively they are weak competitors compared to the ones we own. Accordingly, besides the informal sector, there is market share to be gained by Magnit, X5 and Lenta from the other formal operators in our view.

In terms of fund activity, we bought three new positions during the quarter: a 1.7% position in Naver, the leading search engine in South Korea; a 1.4% position in Alibaba (although the fund has effectively owned Alibaba since July 2014 through a position in Yahoo, with Alibaba representing 80% of its valuation) and a 0.7% position in Indiabulls Housing Finance, one of the leading housing finance companies in India. We also added to the fund’s positions in Naspers, JD.com and 58.com. In terms of disposals, we sold the fund’s position in the Brazilian financial holding company Itausa as it moved closer to fair value. We also sold out of the last portion of Old Mutual as we have grown more concerned about ongoing cost slippage, management changes and potential value destruction. We reduced a few of the positions in global consumer staples (notably Philip Morris and Nestle) after strong share price performance and resultant limited upside to fair value. In addition, we scaled down the C-Trip and Baidu positions: the former due to share price appreciation and resultant less upside, and the latter amid increasing potential risks to the search business. We still believe that all-in Baidu is very attractive from a valuation perspective, but the risks have increased in our view, which warrant a slightly smaller position on a risk-adjusted expected return basis.
Over the past few months, we continued to spend a lot of time on the Chinese internet sector (which makes up a large part of the fund) and several members of the team had separate interactions with management from JD.com (five meetings in total), Alibaba (we also attended its investor day in China), Tencent (both Tencent management and a Naspers board representative), Baidu and Ctrip.

JD.com reported excellent first-quarter results in April that beat consensus by some margin. It has historically been one of the most attractive stocks in our investment universe in terms of upside to fair value, but the evolution of its business has exceeded even our expectations. After repeated losses since its IPO (although it has always generated decent cash from suppliers funding its working capital, even when making accounting losses), JD.com has now become profitable due to an increase in gross margins. This has been driven by a mix effect and growth of its 3P business relative to its 1P business. Both 1P and 3P are very attractive businesses, but the 3P model has a higher margin - there is little cost involved in selling someone else's product on your established platform. As 3P becomes a larger proportion of the overall mix, JD.com’s margins will continue to rise. When we first assessed JD.com in 2014, we believed it could eventually earn (operating) margins of 4% to 5% due to gross profit margin expansion and achieving sufficient scale to dilute down the high fixed costs involved in rolling out a massive distribution and logistics network. As mentioned previously, the business was loss-making at the time and the lack of profits meant the market struggled to value JD.com appropriately. Investors focused on misleading short-term metrics, like quarterly gross merchandise value (the total value of all goods sold on their platform). The business has developed considerably since then and, together with the impact of the growth in its 3P business, management believe they can earn operating margins of 6% to 8%. It is this sort of profitability target that has driven the share price rally in recent weeks. Yet despite the move, we believe the share remains very undervalued - even if it only reaches the bottom end of this margin range in the long term.

As previously mentioned, the fund also bought a new position in Alibaba, although we have held an indirect stake in Alibaba for the past three years due to owning Yahoo. Alibaba operates a consumer-to-consumer marketplace (think of eBay) and a businessto- consumer marketplace for merchants. We have historically (and still do) favoured the JD.com business model over that of Alibaba, due to the competitive advantages of the fulfilment model that JD.com (and indeed Amazon) operates, that results in total control of the customer experience. With time, however, whilst we continue to believe that Alibaba will lose overall market share (and JD.com will gain share), we have gained increased conviction that Alibaba is a formidable business with a number of very promising assets besides its traditional e-commerce business. This includes its Cloud business and the payments business, Alipay. In fact, the e-commerce offerings from JD.com and Alibaba are largely complementary. As such, given the massive potential market in China, we believe that both can expand their relative businesses at high rates for the foreseeable future - without really taking each other on too directly. The fund also continues to hold Yahoo (now renamed Altaba after the sale of the core business to Verizon) where the upside to fair value is even higher than Alibaba if tax can be minimised. Taken together, Alibaba and Altaba (with Alibaba making up c. 80% of Altaba’s value) represent almost 3% of the fund.

The quarter also saw the completion of the antitrust assessment process of the proposed merger between the two Brazilian education holdings Estácio and Kroton. Despite the best efforts of Kroton to propose remedies that would allay antitrust fears, the board of the antitrust authority voted against approving the merger. We had expected that, on a balance of probabilities, the merger would be approved - but had always positioned the fund for the risk of a rejection. Estácio never quite traded at the final agreed swap ratio of approximately 1.28 Kroton shares per Estácio share, so investors with 100% certainty that the merger would pass would have held only Estácio shares to benefit from the unwind of the swap ratio discount. We maintained the fund's overall Kroton/Estácio relative exposure at similar levels (2-2.5x Kroton weight vs Estácio weight) to when the deal was first announced, reflecting our view that in the event the merger was vetoed or approved with too many conditions, we would want to own both companies just as we had for the 18 months prior to the merger developments, but would want a bigger position in Kroton. Now that the deal has been rejected, the share prices had, by quarter-end, pretty much converged back to parity - which is where they were a year ago when this whole process started. The unwinding of Estácio's merger premium is the reason why Estácio shows up as a detractor in the quarter, but Kroton contributed positively to alpha. We believe both businesses remain extremely attractive on a standalone basis, with over 60% upside to fair value for both. Kroton trades on less than 11x earnings and is a 5.0% position in the fund, while Estácio trades on less than 10x and represents 2.0% of the fund. They remain the largest and second largest players in the industry with massive scale advantages over their smaller peers, who generally only trade at marginal profitability. We also believe that Estácio is very likely to be a participant in further industry consolidation.
Overall this quarter proved to be a busy one, with team members spending a lot of time travelling to meet the management of portfolio holdings, competitors or looking for new ideas in Brazil, Russia, Hong Kong, China, Singapore, Macau and Indonesia. The weighted average upside to fair value of the fund at the end of June was around 50%, which is broadly in line with the historical five-year average.

Portfolio managers
Gavin Joubert and Suhail Suleman as at 30 June 2017

The Coronation Global Emerging Markets Fund returned 9.8% in the first quarter of 2017, which was 1.6% behind the index's return of 11.5%. The biggest positive contributors over the quarter were JD.com (+22.2%, 0.45% contribution), YES Bank (+39.6%, 0.37% contribution) and Naspers (+17.2%, 0.3% contribution). Three stocks detracted by 0.50% or more: Magnit (which declined by 13% and detracted by 0.9%), Kroton (which appreciated by 2%, but this was well below the index's return resulting in a -0.5% attribution) and not owning Samsung Electronics (which detracted by 0.5%). The fund is now approaching its nine-year track record and over this period has outperformed the market by 4.6% p.a. Over the past eight and five-year periods, the fund has outperformed the market by 2.6% p.a. and 0.9% p.a. respectively.

In terms of portfolio activity over the quarter there was only one new buy and we sold out of five smaller (all less than 1%) positions. We sold the fund's entire holding in Richemont, Arcos Dorados and Netease as all three reached our estimation of fair value for the respective stocks, with both Arcos Dorados and Netease having appreciated by 100%+ over the past year. Sohu.com was sold as a result of our increasing concern about the long-term prospects for the video business, where they compete against three formidable players (Alibaba, Baidu and Tencent). Lastly, while still undervalued in our view, we sold out of Pao de Acucar as a result of a reduced margin of safety given a reduction in our fair value due to slower long-term top-line growth and lower long-term normal margins. Other selling activity of note include our halving of the fund's position in Brilliance China (from 3% to 1.6% of fund) as the share moved closer to our fair value, having appreciated by 64% over the past year.
The only new buy during the quarter was a 1% position in Norilsk Nickel. Norilsk is the no.1 nickel producer in the world (35% of revenue), the no.1 palladium producer in the world (palladium/platinum make up just over 30% of revenue) and a top 10 global copper producer (25% of revenue). In our view, Norilsk's ore body in Siberia is one of the best geological assets in the world. The Nickel grades of Norilsk are for example 40% higher than the industry average, copper 25% higher and palladium/platinum above the South African platinum miners' average. This means that the revenue per ounce that Norilsk generates is far higher than most peers and as a result margins are industry-leading (EBITDA margins have averaged 50% over the past 10 years) as is the company's free cash flow generation (Norilsk have converted 100% of earnings into free cash flow over the past 10 years). Norilsk trades on 10x this year's earnings with a 9% dividend yield. In addition, the prices of all three of its core commodities (nickel, palladium and copper) are currently trading below normal (marginal cost of production) levels and earnings are therefore below normal in our view. We have valued Norilsk on a low multiple/high discount rate to take into account the risks (a cyclical asset, based in Russia) and this valuation still gives substantial upside, making the share attractive on a risk-adjusted expected return basis in our view.

In terms of other buying activity during the quarter, we added to the fund's Naspers position (the fund's largest position, now 7.7% of fund). Naspers' core asset is its 33% stake in Tencent and today Naspers trades at a 15% discount to the market value of this Tencent stake. In addition, Naspers owns considerable other assets, of which the two most notable include its online classified businesses in numerous emerging markets and its PayTV assets in South Africa and c. 40 other African countries. In our view, Tencent (the leading gaming/social network internet company in China) is a great asset that we would happily own. Naspers, however, effectively enables us to buy Tencent at a discount to Tencent's current market value, plus we get substantial other assets, including the online classified ads and PayTV, for free. In addition, we have high regard for the management team of Naspers and believe that over time they will create value in the assets excluding Tencent. The upside to our fair value for Naspers is around 70%, making it extremely attractive in our view.
At the end of the quarter, the fund had 5.5% combined exposure to the Indian financial services sector spread across the two private sector banks, Axis Bank and YES Bank, as well as the country's leading mortgage provider, the Housing Development Finance Corporation (HDFC). We have written before about why we like the private sector financials - in short, they are taking market share from their public sector counterparts in a market that is growing strongly due to low financial services penetration. We had been reducing the fund's exposure to each of the names during the course of 2016 as their share prices appreciated and their associated margin of safety (to our fair values) declined. In November 2016, the Indian government announced an immediate end to the acceptance of all existing Rs500 and Rs1000 notes as legal tender. The outstanding notes had to be deposited by year end or expire worthless. This process was aimed at catching tax evaders and bringing more of the economy into the formal market, partly in preparation for a national sales tax planned for introduction in April 2017, but also because so much money had left the banking system after gold import restrictions were introduced as a foreign currency saving measure during the mini-crisis of 2013. Since these notes represented approximately 98% of monetary value outstanding, and there was no transition period, the disruption to the economy was immense, with massive queues of people waiting to deposit their money into bank accounts and a virtual standstill in large parts of the economy. The effect was compounded by insufficient availability of the new replacement notes, and with India being a predominantly cash economy there was a big crunch on consumer spending in the final quarter of last year. The combined effect of this hit sentiment toward the banking segment and most of the listed banks fell. In our view, this represented a buying opportunity as the long-term impact on the banking sector (and hence the long-term fair values of the individual banks) is limited, so we increased the fund's positions accordingly.

Our view has largely been borne out by subsequent developments this year. In our recent interactions with the management teams of all three businesses it has become apparent that their franchises have actually been enhanced by the demonetisation events, even if there has been short-term pressure on the economy and hence lending demand. In the case of Axis and YES Bank, the investment in building out retail branch infrastructure (and their ATM networks) in recent years has paid dividends sooner than would otherwise have been the case. Both saw a significant inflow of deposits, which is a lower cost source of funding, allowing them to increase their current and saving accounts (CASA) ratio (as a percentage of total liabilities) at a faster rate than they have been able to historically. This is positive for long-term net interest margins and if maintained, should result in strong profit growth once lending volumes normalise. The private sector banks are also better placed to respond to the pickup in lending demand as their lending books have been less affected by the slowdown of the fourth quarter of 2016. For HDFC, the inability of all industry lenders to accept cash for mortgage payments has hurt the competition only because HDFC have not historically accepted cash. Other developments in the housing finance industry underpin HDFC's long-term prospects - in particular, the expansion of tax incentives for low income housing, the easing of regulations related to raising of funding as well as better and more uniform regulation of housing developers themselves.

The weighted average upside to the portfolio at the end of March was just above 50%, which is in line with the long-term average. We continue to come across a number of potential new buys and the bigger challenge is deciding which positions to reduce or sell to make room for these potential new holdings. During the first quarter of 2017, we went on two research trips to Brazil as well as trips to India and Hong Kong. In the coming months members of the team will be going to China, Russia, Indonesia and Singapore.