To: Viking Investors
From: O. Andreas Halvorsen
Our second quarter performance was a loss of 5.0% for VGE and a loss of 11.9% for VLF net of all fees on a composite basis
On an unlevered basis, VGE’s long portfolio was down 11.2% and the short portfolio was down 10.3%, yielding a long-short spread of negative 0.9% (see the attached Base Case Analysis). In the quarter, five long positions cost us 0.5% or more while no short contributed an equal amount.
We are greatly disappointed in Viking’s returns both on an absolute basis and relative to the indices. We do not blame factors outside of our control, but acknowledge that changes in the global macro-economic, political, and regulatory environment; a broad-based fall in stock prices around the world; and unusually high levels of correlation among these prices have increased the degree of difficulty in generating a profitable long-short spread. We are paid to deal with such challenges at all times and are in a business that requires hard work and consistent processes every day. When the market gives us a disappointing score for our efforts, we examine our results and our methods to ascertain whether we need to make adjustments – this is the Viking way. Rest assured that our objective remains to achieve maximum capital appreciation commensurate with reasonable risk, and we remain firmly dedicated to meet this goal. In light of this, we have engaged in a thorough examination of our results.
This review confirmed our strong belief in our core competency of creating a positive and sustainable long-short spread in VGE based on thorough investment research. We are stock pickers. We believe that our thoughtful analysis and disciplined valuation over time yield a diversified portfolio of longs and shorts whose stock price developments will deviate from each other and provide a profitable spread. Since inception, our average annual long-short spread has been 24%, although there is significant quarterly variability underlying this average ranging from a high of 24% to a low of -15%. While we are unhappy about our last two quarters of negative spread, we believe our basic investment process is sound – over the longer term, we identify longs that outperform the market and shorts that underperform – and we have a more experienced analytical staff than ever
conducting this investment process day-to-day.
We also confirmed our findings that the greatest alpha generation has been in our highest conviction ideas. As shown in the table below, of the 921 long and short investments we have made in VGE over the past five years that resulted in a profit or a loss exceeding ten basis points of performance on our total capital base, 59% were profitable. Of the 56 investments that resulted in a profit or a loss exceeding 100 basis points, 75% were profitable. Our top ten longs returned 28% per annum, outperforming the remaining longs in the portfolio by over 12 percentage points per annum on an unlevered basis. Over the same period, the MSCI World and S&P 500 indices returned 1.5% and 0.4%, respectively, per annum.
We conclude from these statistics that when we had strong conviction in an investment and scaled it, our success rate increased – in other words, our high conviction has been correlated with improved odds of being right. This is our profit engine. In addition to the historical data supporting this conclusion, the engagement of and collaboration among our senior portfolio managers – Tom Purcell, Dan Sundheim, Jim Parsons, and Dris Upitis – give me even greater confidence in our ability to identify compelling opportunities in the future. Collectively, these four, assisted by David Ott and me, assess the attractiveness of our best ideas across all portfolios. They are increasingly knowledgeable about opportunities in the incremental sectors they now cover. This is evident in the dialogue at our weekly portfolio meetings and in the many impromptu interactions throughout the week. I am very encouraged by this process and believe that it identifies which investments represent the most promising risk-adjusted return potential and highlights which positions require further investigation. As a result, we are gaining greater conviction in the order and relative sizing of our largest positions. A recent example of this process at work is our investment in American Tower, our most profitable long in the quarter and a new top ten position described later in the letter. Paul Enright initiated the investment in his portfolio early in the second quarter. Since then, Jim, Dan, and Tom have bought the stock and it is now the firm’s third largest position. While every portfolio manager has full discretion over the capital he manages, the intenseinteraction and debate among them result in a healthy comparison and challenging of ideas. The level of collaboration is impressive and the intellectual honesty employed in these deliberations is all-Viking.
To step back, I see concentration as a determinant of differentiated investment returns and think of it as a spectrum that spans two extremes. At one end is a market-weighted or index portfolio, and at the other end is a portfolio containing a single investment. If we position the fund at the former end and own a market-weighted portfolio, we will generate no alpha, which represents the foundation of our business model. Positioning the firm at the latter extreme and putting all capital in one investment would be entirely too risky as well. In spite of our attractive stock-picking statistics, we are frequently wrong, and combined with what would result in extremely poor portfolio liquidity, such concentration would be irresponsible.
Therefore we must find the appropriate position somewhere between the two extremes. Unfortunately, there is no precise formula that computes optimal concentration. We cannot afford to be too concentrated, yet we cannot afford to be too diversified either. When determining the appropriate level of concentration, we have to overcome an inherent bias in our business model. We have built the firm on a strategy that hinges on
our ability to retain talented analysts by training them in accordance with their aspiration of becoming portfolio managers and allowing them to utilize those skills at Viking rather than having to go elsewhere. This model leads to a growing number of portfolio managers over time – we currently have nine. Left unchecked, this will result in an increasingly diversified portfolio in the aggregate and take us too far towards the “market-weighted portfolio” end of the concentration spectrum. To counterbalance this tendency, and thereby increase concentration, we limit the number of positions in five of the portfolios, encourage and reward analysts for building positions of size, and mandate our four senior portfolio managers with investing in each other’s best ideas.
In light of our stock-picking statistics – indicating that our highest conviction ideas are more likely to yield profits – we have increased concentration in our favorite, liquid longs to raise the impact these positions will have on our results. Over the past week, the top ten positions within VGE and VLF have been boosted by 6.9% and 1.4% of capital to 42.5% and 43.8%, respectively. We believe this trend will improve performance in both funds over time and we may choose to increase concentration further in the near future. Incidentally, the current concentration level is nothing new at Viking – since inception, the average amount of capital in our top ten longs in VGE has been 39% and, at times, concentration has been significantly higher.
We were comfortable increasing long exposure in VGE because we simultaneously increased short exposure. We had for some time been aware of how the opportunity on the short side has evolved and in response to this, recently made changes to improve our sourcing of short ideas. I believe that the declining idea flow on the short side may have been due to two factors:
(i) The practice of shorting stocks has changed since we started honing our skills as short sellers. Twenty years ago, the search for shorts could be limited to truly broken business models as the amount of capital dedicated to short selling was limited and the practitioners in the field were few. Now, the search has to be broadened to include companies expected to merely underperform market averages as well as terminal shorts. Generally, there are as many underperformers as there are outperformers, so the pool of candidates is vast. However, the process of identifying underperformers is different from that of searching for companies expected to fail. Though not a novel realization, it is one that requires a modification in behavior. We
believe this applies to a long-short fund such as VGE regardless of size.
(ii) Structurally, analysts are incentivized to look for longs over shorts. Researching apotential short idea is every bit as time-consuming as identifying a potential long. Since we typically make larger long investments than short sales, the potential reward in dollars is larger from being right on the long side than on the short side. An analyst who wants to make an impact on fund profitability will endeavor to engage as much capital as possible behind his ideas and is therefore inclined to focus his search on longs over shorts, everything else equal.
Recognizing that generating incremental short ideas enables us to scale our highest-conviction longs and thereby amplify expected returns, we chose to address these two challenges by explicitly charging our analysts with generating additional short ideas for VGE. Each analyst now provides on a weekly basis the one or two companies they like least well among their most liquid names. We have instituted this practice because we believe it is inappropriate to further concentrate our most attractive short positions due to
the asymmetric risk-reward profile associated with shorts – limited upside and unlimited downside – and the firm requirement that our shorts remain appropriately liquid at all times. To encourage an intensified search for short ideas, we also modified our compensation system to reward this activity and to be more closely aligned with firm objectives. Since implementing these initiatives in the past week, we have increased our
short exposure by 14%, primarily in new names. We believe an optimally constructed portfolio maintains balance between longs and shorts, resulting in lower-than-market volatility. Combined with liquid investments and a fairly stable capital base, this gives us the confidence to employ modest leverage in VGE.
What we consider to be appropriate leverage depends on the balance between longs and shorts in the portfolio – if our longs and shorts are distributed relatively evenly across industry groups and sectors, we may choose to run the hedge fund with higher leverage than if they are not. By recently having introduced somewhat higher leverage, with the expectation that volatility will increase accordingly, it remains our goal to maintain VGE’s return volatility below that of the broad indices.
Our leverage has increased by way of adding shorts selected from the most liquid stocks we cover and increasing exposure to our highest-conviction longs. As a result, I believe we will act with more urgency as trade-offs have to be made continuously between new ideas and exiting positions. On July 12 gross exposure stood at 165% in VGE. We currently operate at what we consider to be an appropriate level of leverage with the wonderful result that our investment staff feels a healthy push to liquidate less attractive positions in order to get new ideas into the portfolio. I believe this is a very favorable development that will improve the quality of stock picking going forward. We will continue our active risk management practices for which I remain responsible. If market, economic, political or regulatory factors dictate, we will reduce leverage as we see appropriate, yet we will strive to retain significant concentration in our best ideas. This can be accomplished easily by selecting investments to be deemphasized or liquidated, as we have done in the past, rather than simply scaling back positions across the board.
To conclude this discussion, I should emphasize that our firm goals are embraced by all analysts and portfolio managers. Our most important job day-to-day is to protect and take advantage of our stock-picking statistics: It is the responsibility of our analysts to continue the thorough research and disciplined valuation analysis that we expect will lead to consistent and high quality investments, and it is the responsibility of our portfolio managers to exploit the resulting odds and capitalize on them by carefully constructing a diversified portfolio of longs and shorts. In order to accomplish this, we stick to what we know, organize the investment team by industries, sectors, and occasionally geographies of expertise, and give our analysts and portfolio managers the best resources possible.
There are two important determinants of the firm’s performance that we have complete control over: the level of concentration we allocate to our best ideas and the amount of leverage we employ. Analysts must go to bat for their best ideas and spend more time sourcing liquid shorts for VGE; portfolio managers have significantly increased their efforts to compare and contrast our largest positions and make certain that our favorite ideas are backed by the largest amounts of capital. I will remain fully engaged on the investment side and will continue to ensure that incentives and organizational structures remain aligned with firm goals. We continue to have confidence in the basic premise that, through thorough research of company fundamentals and disciplined valuation analysis, we will uncover companies whose share prices over time will outperform, or alternatively underperform, market averages.
Gross exposure in VGE increased to 146% at June 30 from 144% at the end of the first quarter. Net exposure decreased 6.0% and ended the quarter at 32.9%. (On July 12, gross exposure was 165% and net exposure was 34.3%.) For both VGE and VLF, the Telecommunication Services sector was the largest profit contributor in the second quarter and Financials was the worst performing sector. Within Financials, the Diversified Financials industry group (which includes asset management and consumer finance companies as well as large diversified bank holding companies) cost us 1.5% in VGE and 2.6% in VLF. Attached to this letter, you will find a breakdown of VGE and VLF exposures by sector and industry group as defined by the
Global Industry Classification Standard (“GICS”).
At the end of the second quarter, Invesco was our largest position in both VGE and VLF at 5.0% and 5.8% of capital, respectively. The largest individual short position in VGE represented 2.2% of capital. VGE’s ten largest longs comprised 35.4% of capital and the ten largest shorts accounted for 12.9% of capital on June 30. VLF’s ten largest positions comprised 40.6% of capital. The following were our long positions on June 30 in order of size for both VGE and VLF:
Invesco Limited (IVZ.N)
Unilever NV (UNc.AS)
American Tower Corp-CL A (AMT.N)
Oracle Corp. (ORCL.O)
Comcast Corp-CL A (CMCSA.O)
News Corp. (NWSA.O)
Tyco International Limited (TYC.N)
The Sherwin-Williams Company (SHW.N)
Goodrich Corp. (GR.N)
Adobe Systems Inc. (ADBE.O)
Six of the top ten long positions in both VGE and VLF were new to, or reentered, the list this quarter: Adobe, American Tower, Comcast, Goodrich, Oracle, and Sherwin-Williams. A financial short was our largest winner for the second quarter in VGE, contributing 0.4%. Our largest long winner was American Tower, contributing 0.2% to VGE’s performance and 0.2% to VLF’s performance. American Tower is the largest telecommunications and broadcast tower operator in the U.S. with overseas operations in India, Brazil, Mexico, and Chile. Its business model is among the best we have encountered due to high barriers to entry, pricing power, and strong secular growth. We have owned American Tower in the past and we re-initiated a
position this quarter because we believe the market has taken many of these characteristics for granted and is underestimating future growth opportunities both domestically and internationally. Additionally, we believe that American Tower's shareholder remuneration will accelerate over the next several quarters and that, in light of certain tax incentives, the company may convert to a REIT. We find American Tower to have a superior business model relative to most traditional REITs, yet it trades at a discount to the REIT-average. We believe the combination of predictable growth, accelerating shareholder returns, and pending REIT status will generate greater shareholder interest over the next several quarters causing the stock to trade closer to our price target over time. As of June 30, American Tower was our third largest long position at 4.3% of VGE capital and 4.9% of VLF capital.
Our largest loss in the quarter was in Invesco which cost us 1.3% in VGE and 1.4% in VLF. Invesco has been in our top ten list since we initiated the position in the fourth quarter of 2007 and was our second most profitable investment in 2009. During the second quarter, Invesco sold off along with other asset managers despite reporting better than consensus first quarter earnings and higher synergy estimates from the Van Kampen acquisition. Encouraged by the fundamental strength of the company and financial and strategic benefits from the Van Kampen acquisition, our core thesis has not changed and we continue to believe that Invesco will outperform its competitors. Viking is currently net long 2.4% in the Asset Management and Custody Banks sub-industry group, which includes the Invesco long position and short positions in asset
managers that we believe will experience deteriorating fundamentals and are more levered towards a declining market. The net loss attributable to this industry group for VGE during the quarter was 0.6%.
In our last letter we wrote about the opportunity we see in Asia and our intention to increase Viking’s presence in the region. We are excited to report that we are opening an office in Hong Kong and that Ning Jin will head the office beginning in the third quarter. Ning joined Viking in 2007. He speaks Mandarin fluently and is an experienced analyst and recent portfolio manager. He has been responsible for identifying and researching investments in the internet, industrials, transport, aerospace, and defense sectors. We expect to hire one or two analysts to assist Ning in his coverage of the non-Japan Asian markets. Three factors contribute to our excitement about the opportunities in the region: One, the investment universe there is vast; in China alone, there are 380 companies that trade more than $30 million a day, nearly a third the size of the combined rest-of-world universe of 1,200 such companies. Two, the range of opinions on these companies is large, providing an ideal environment for an actively managed equity fund. Three, under Ning’s leadership, we will bring our well-established investment process to markets that are not covered as thoroughly as elsewhere in the world. Ning will work closely with our industry specialists based in New York and London.
We are delighted to announce that Nick Lagaros joined us as Chief Technology Officer at the beginning of June. Prior to joining Viking, Nick spent fifteen years at Moore Capital Management, most recently serving as Co-Chief Information Officer. His initial charge is to continue the redesign of our technology platform and build a scalable hardware and core application infrastructure. He will work to improve information security and implement superior levels of business continuity and disaster recovery. Nick will also lead the development of a new application interface that provides leading-edge trading and analytics tools. We are excited about the experience and enthusiasm Nick brings to Viking and look forward to providing updates on technology-related improvements.
Following David Ott's decision to step down from his role as CIO, we offered all investors a one-time liquidity option on August 1 with notification due by June 15. Gross external redemptions for August 1 amount to approximately 10.5% of capital in VGE and approximately 13% of capital in VLF. We will honor all such redemption requests effective August 1. We intend to replace all redemptions in VGE in accordance with our ongoing practice of keeping our overall capital flat with respect to fund flows. To date, we have indications of interest for over 50% of the redeemed amount which we anticipate will be invested over the next several months. We look to replace the remaining redemptions throughout the second half of the year. If you have
interest in increasing your investment, please contact any member of the investor relations team.
With respect to VLF, we will continue our efforts to grow the fund in aggregate.
As a reminder, our 2010 Annual Meeting will be held on Wednesday, October 13 beginning at 2:00pm at Cipriani, 110 East 42nd Street in New York City. We hope to see you there.
We enjoyed meeting with many of you in April and May and thank you for your continued trust in the Viking team as we focus on improving our performance until we again deliver the returns you deserve on the capital you have placed under our management. Having shifted my attention in large part to the investment side of the firm, I really enjoy my expanded engagement with the portfolio managers and see a great commitment to improved performance in all my fellow Vikings. To reiterate what I have said previously: I believe my job is the best in the world. I work with 93 dedicated Vikings who each care deeply about the firm and share a common goal of providing attractive returns to investors. We will continue our practice of keeping you
informed of our progress and wish you a great summer.