Multiple Fears Converge

This article is excerpted from a letter by MOI Global instructor Patrick Brennan, founder and portfolio manager of Brennan Asset Management, based in Napa, California

Many of the various macro concerns discussed in previous letters converged during the fourth quarter and created a powerful vortex of selling pressure that pummeled most asset classes. Year-end tax loss selling further fueled the storm’s strength and the S&P 500 declined -13.5% during the fourth quarter, including a -9% decline during December. Trade war and interest rate concerns, first concerns about higher rates…then fears of yield curve inversion, account for some of the anxiety. That said, we think the biggest worries centered around global economic weakening and whether a recession was about to emerge in China or Europe and whether this recession might quickly spread to the United States. Despite concerns about a more fragile US growth outlook, Federal Reserve commentary was more hawkish than some would have liked and created another round of selling. With risk-free cash finally earning above 2 percent, many concluded that a long economic expansion/stock market rally was ending. Copying the behavior of sports fans who empty stadiums to avoid traffic midway through the fourth quarter (or halftime in the case of Notre Dame fans watching the Cotton Bowl carnage), investors decided to exit before feared negative economic prints became visible.

So, what’s next? Will stocks keep going down or will the market bounce back? Are we about to enter a recession? In past letters we’ve rhetorically repeated these common questions and essentially answered: “We don’t know.” Unfortunately, we still haven’t found the magic crystal ball which will provide answers. Of course, we remain suspicious that one even exists. In our Q1 2016 letter, a time when similar recession worries surfaced and a time when many prognosticators noted the unsustainable length of the current expansion, we described a study by two IMF researchers^[1] who found that professional forecasters had picked only 2 of 60 recessions one year in advance during the 1990’s. The same group was essentially perfect in its inability to predict the 2008/2009 recessions one year in advance. If this group — likely armed with sophisticated econometric models and closely monitoring every macro data point — struggles, we find it difficult to believe that we (or most other investors) would do much better. Despite this, we find this exact topic is among the most frequently discussed by many investment professionals. The conversation vaguely resembles discussions among a group of weekend warriors where the “glory day” amateurs describe the best way to guard basketball star James Harden in a game of one-one-one…after watching him torch fellow professional Klay Thompson for 50 points. Economic expansions end at some point, and there will be another recession. But, we still think it is impossible to know the timing.

Cheap Gets Cheaper

Moving away from sports analogies and moving from the macro to micro, our portfolio struggled during 2018. We own a small number of names that we viewed as statistically cheap prior to the fourth quarter rout. While several names initially held up better than the overall market, this resilience quickly faded and these inexpensive and illiquid names often declined more than the overall market. Towards the end of December, tax loss selling and/or fund liquidations caused even sharper declines and several holdings dramatically diverged from any reasonable estimate of fair value.

We take a longer-term view on any investment, generally looking to own names for 3-5-year periods and often longer. Normally, it takes time for price and fair value to converge and often the gap widens before narrowing. Having invested throughout 2008/2009, we certainly appreciate the fact that cheap often becomes far cheaper than expected. We also understand the frustration that comes from the recent declines and near-term underperformance. That said, we do think there is substantial upside in our portfolio, although we acknowledge it is difficult to predict the exact time when this value will be recognized. Certainly, not all of our holdings will play out as anticipated. We have made errors in the past and will do so again in the future. That said, we think at least a couple of our holdings can generate substantial returns which carry the entire portfolio, given our higher position weightings.

In past letters, we have provided a more in-depth profile of a holding. As we’ve continued buying many of the same names discussed in past letters, we thought it would be helpful to update several of our largest holdings and to quickly summarize our investment rationale. For those who want the simplest summary, we would simply say that little has changed from past letters, that the declines have widened the price to intrinsic value gap and this recent underperformance represents a great time to add money, notwithstanding the likely continued volatility in the months ahead.

We will divide the summary into 3 categories: cable, financial service and consumer discretionary names. While there are nuances to these investments, we would note some broadly common themes:

• Recurring customers/cash flow
• Complicated (and often leveraged) capital structures
• Strong management teams
• Large capital returns (generally buybacks)

Certainly, not all these characterizations apply to every holding. Additionally, several of the largest holdings are either based or operate outside the United States and are exposed to FX movements. When the dollar is strong (as has been the case over the past couple of years), this has been a headwind as overseas markets have generally underperformed US markets. Furthermore, multiple names (LBTYA, KW, TSB) have sizeable exposure to UK/Ireland and are therefore viewed as more susceptible to Brexit uncertainty. As we’ve noted in past letters, the Liberty names and private equity names are often difficult for multiple classes of investors to own. Several of our holdings would be considered smaller or mid-cap companies, but their effective tradable float is even smaller, given larger insider ownership and often concentrated ownership among a small group of institutions. This smaller float makes these names more exposed to the types of extreme movements experienced during the fourth quarter. Finally, but we believe most importantly, the names below appear inexpensive on a relative and absolute basis. So, to summarize, our names are good businesses, with many operating outside the US, but are often less liquid and often viewed as “risk off” stocks – i.e., the exact opposite of what investors wanted to own in Q4 2018. We see upside as described below:

Recession-Resistant Cable Names

If an investor were convinced that a recession was approaching and he or she had to own stocks, utilities and consumer staple names have traditionally been considered safe havens (rightly or wrongly). We’ve long discussed how cable looks like something closer to an unregulated utility. If we simply plotted cable cash flow over the past 15 years and provided no time markers, it would be difficult to pinpoint when a recession begins or ends based only on cable’s operating results. While we would concede that some consumers may consider downgrading internet speeds if times really get tough, we still suspect the internet connection ranks close to electricity/water/brand-name diapers in the pecking order of last items standing during a recession. While video cord cutting may accelerate, we think consumers will continue to need broadband as they consume more data. We own US, European and Latin American cable through four names.

GCI Liberty (GLIBA)/Liberty Broadband (LRBDK): We think Charter (CHTR) has the strongest operating/capital management team and we see a path for the roughly $290 stock to eventually produce $30+ in free cash flow per share (FCPS). We own the name through two Liberty vehicles trading at 10-20 percent discounts. Compared to international peers, US cable companies face weaker competition for broadband subscribers, and therefore the visibility on cash flow is probably highest relative to our other cable names. While CHTR does trade at a premium to other cable names, we receive a roughly 15 percent discount through our LBRDK/GLIBA holdings. As discussed in past letters, we think the 5G risk is very manageable. We acknowledge that there is regulatory risk and certainly concede President Elizabeth Warren (…gulp) would not be good for cable stocks. But, we also believe that investors are underestimating cable’s mobile opportunity. Furthermore, CHTR will likely repurchase a large portion of its share base over the coming years. To the extent that the price doesn’t move, the per share economics get even more attractive, especially considering the large amount of cheap debt financing fixed for the next ~10 years. It is highly likely that we will ultimately receive CHTR shares for our stake in LBRDK/GLIBA and believe a combination of LBRDK/GLIBA is possible within the next couple of years.

Liberty Latin America: As a reminder, we made LILAK a large holding in late 2016/early 2017 after sharp declines related to a botched integration of an expensive deal for Cable & Wireless (CWC). CWC results showed signs of improvement in 2018, and Chile continued its steady performance. While management should be credited for its Puerto Rico^[2] rebuild efforts, Puerto Rican operating cash flow was still down substantially year over year and prevented stock buybacks. A large institutional holder liquidated its LILAK investment during December, and the year-end declines therefore were more a function of stock supply/demand realities versus underlying business issues. LILAK currently trades at a 30+ percent discount to where Dr. Malone bought $37 million in 2017. We believe that free cash flow will improve materially in 2019 and we still think there is substantial room for future deals. Speaking of which, LILAK recently disclosed that it was in merger discussions with Millicom International Cellular SA, but then subsequently revealed that the talks had ended without a transaction. While there is consternation about issuing (versus repurchasing) cheap shares to support a large transaction, we believe a well-structured deal could create substantial value, given the importance of scale in the cable business. We trust the current team to consummate a deal only if it is in the best economic interest of LILAK shareholders. We suspect merger talks could resurface and we will discuss any transaction in more detail in future letters.

Liberty Global: Of all the cable names, LGI has been the biggest detriment to performance. Having owned it for years from its original 2004 spin, we sold too soon (with hindsight…more on this later) and then reentered the name too early. We were certainly aware that markets are more competitive in Europe, but we thought execution would be sharper (the biggest understatement in 2019) and did not anticipate the degree of negative regulatory action — nor did we bank on the rolling Brexit drama. LGI might be the most unpopular Liberty name we’ve ever owned. While the management team can be rightly criticized for operational execution, we would note that they have actively tried to arbitrage the difference between private and public market values (Netherlands, Austria, Germany, etc.). LGI faces two binary risks: hard Brexit and EU approval for Germany/Eastern European sales. On the former, other asset prices suggest a Brexit compromise, delay or second referendum are the more likely scenarios. On the latter, deal approval is not assured, but in our opinion EU jurisdiction (versus German) increases the odds to well over 50 percent. And as for the valuation? Nearly incomprehensible, even if there is little growth in the markets that remain. Consider that pro-forma for the deal closing, the VOD proceeds would equal ~80 percent of LGI’s current market capitalization or that LGI’s EV/EBITDA multiple (5.4x 2019E EBITDA) would be roughly half the valuation levels at which it is selling exiting assets.

Although we promised a “quick summary,” we can’t resist one detailed point: We have seen multiple sell- side reports assign near random number generator multiples to various “sum-of-the-parts” LGI valuations, based on bombed-out public market valuations. Could LGI’s Virgin assets (which will represent roughly 60 percent of pro-forma EBITDA) really be worth only 4x EBITDA? Virgin may not grow much outside its Lightning expansion, but it will still generate large amounts of free cash flow. Virgin’s Lightning expansion (where it is passing 4 million additional homes) gives a real-world test case of replacement value. If it costs £700 to pass a home that is often within 25 meters of its existing plant and another ~£250 in CPE costs, this would imply that Virgin’s replacement multiple is somewhere between 5-6x 2019E EBITDA, assuming 40 percent penetration. This value gives no credit for cash flow that the cable assets would generate while alternative fiber systems are constructed. Even using higher weighted average cost of capital estimates (despite existing inexpensive longer-term debt financing) and no growth, it is difficult to derive estimates much below 7-8x once any consideration for opportunity costs is considered. We think that the deal will be approved and that LGI could conceivably retire nearly 30-40 percent its share base this year.

“Risk Off” Financials

Jefferies: We covered both JEF and TSB in our Q3 letter and presented both names for The Manual of Idea’s Best Ideas conference earlier this month. JEF recently reported Q4 earnings (Nov 30 fiscal year) which showed some expected investment bank weakness, mostly concentrated in fixed income. However, the company also disclosed that it bought back nearly as much stock in October/November of 2018 as it had in the first nine months of 2019. In total, JEF repurchased 13 percent of its shares at just under $23 and authorized another $500 million repurchase plan. We detailed our thoughts on JEF’s net-asset-value in previous letters. We believe these buybacks will prove highly accretive over the next several years.

Permanent TSB: As discussed in our Q3 letter, TSB has a ~€700 million market capitalization but only a roughly €175 million float after taking into consideration the Irish government’s 75 percent interest. To say it is illiquid is an understatement. We have moved the stock with 15,000 share purchases, so volatility needs to be expected. As previously discussed, TSB has announced two securitizations that take non-performing loan (NPL) below 10 percent. We think further NPL reductions occur over 2019 and we still believe TSB can return roughly half its market capitalization through capital return sometime in the next year. In December, the Irish Ministry of Finance was authorized to sell its entire stake in Irish banks as it deemed fit. A widely discussed merger with similarly sized Ulster would offer the greatest upside. While it “feels” like it will never happen in our lifetime, it is metaphysically possible that European interest rates could rise and provide a substantial tailwind to this deeply unpopular sector. That said, we believe that substantial returns can be realized without either of these scenarios and occur simply as a result of the previously discussed capital return and modest loan growth in a red-hot housing market. Consider that the widely detested Euro Stoxx Banks Index trades at ~60 percent of book value. Capital return plus valuation improving to this level could drive substantial returns. While the price will fluctuate, the <0.4x tangible book value valuation provides substantial downside support.

Blackstone is the epitome of “risk off” name, and we used the volatility to increase our position. Market declines will impact the timing of asset sales and ultimately the collection of performance fees. Of course, every time we turn around, BX seems to raise another $10 billion which is locked up for 10+ years or sometimes permanently. While other private equity names may screen cheaper, BX has by far the strongest fundraising franchise. Additionally, investors perpetually underestimate the value of private equity’s performance fees and we suspect history is repeating itself here. Ultimately, the stock price should follow AUM levels. As the stock sold off in December, its forward dividend yield (blend of 2019 and 2020) was probably north of 8 percent. As discussed in previous letters, we think buybacks would create more long-term value when these names plummet, but we accept the capital allocation policy given the powerful secular tailwinds. We would vote against willingly paying more in taxes (aka C-Corp conversion), but we have little doubt the stock would pop should such a decision be made. Please see our Q2 2018 letter for more details on BX.

Compass Diversified Holdings is a middle-market focused owner of 10 businesses, from baby carriers (ergobaby) to quick-turn production circuity boards (Advance Circuits). We will not detail all the businesses (we can hear the cheers), but we’d simply note that each business is among the leaders in some smaller niche industry in which it operates. Think of CODI as a permanent capital public private equity firm with fantastic disclosure on each portfolio company. We had owned shares at varying points over the past 8 years but had generally exited over the past 1-2 years. We selectively re-entered as the stock was crushed during the fourth quarter.

CODI came public with a larger dividend partially because of a usual corporate structure (Delaware Trust) which required a payout to attract investors. Additionally, the original backer of CODI (Teekay shipping family) used the dividend to support its charitable endeavors. Historically, CODI funded acquisitions with cash on hand and with proceeds from business sales, as a large portion of its current cash flow was used to fund its payout. A chink in CODI’s business model was periodic lags between when CODI sold a business (generating nearly $800 million of gains since its 2006 IPO) and when it could deploy proceeds into new acquisitions. While the company was thoughtful on the timing of equity issuances, there was still dilution and, in our opinion, this limited the ultimate upside. We continued to follow the company and noted two favorable preferred issuances at attractive prices (for CODI) which should reduce equity issuances going forward. CODI probably would be better served by paying a reduced dividend and simply repurchasing shares during market selloffs, but its shareholder base (including its original backer) requires the dividend. While the stock has always traded at a discount to fair value, its discount can balloon out during market selloffs. During these periods, CODI’s management team and board members have consistently bought stock and this pattern repeated in December. Seeing a fully covered 11-13% dividend yield (which backers again need to make charitable contributions), multiple insider purchases and a breakup value 45-80% above current prices, we started buying as shares retreated back to 2011 levels.

Cheap, Less Leveraged Consumer Discretionary Names

Discovery Communications (DISCK) and Qurate (QTREA) have both been discussed in past letters. These names are probably among the most exposed to a recession that we own. As noted in our last letter, we reduced our DISCK position in the third quarter but started buying again as the stock plummeted. DISCK has significantly reduced its leverage since the Scripps acquisition and has gained traction on some of the virtual platforms. While not as statistically cheap as it was during the fall of 2017, DISCK still has a nearly 15%+ free cash flow yield and arguably has less downside risk now, given the traction on merger synergies and debt paydown. While we would prefer that CEO David Zaslov spend more time growing Eurosport subscribers versus proclaiming to be the “Netflix of golf, field hockey, skiing, ping pong, etc.,” the virtual deals he secured with Hulu and Sling provide substantial downside protection should cord cutting accelerate. We believe Comcast meaningfully overpaid for Sky, but the bid certainly suggests strategic value for international content assets. In public appearances, Zaslav has actively discussed the benefits of partnering with large technology companies and he has essentially done everything except place a post-it note on his forehead saying, “We are open to partnerships or takeovers.” We still think DISCK days as a standalone public company might be numbered.

Qurate: As lamented in past letters, QRTEA is something akin to the red headed stepchild of the Liberty empire and has often been shown little love, whether called LINTA, QVCA or now QRTEA. While there have been blips in quarterly performance that reignite the secular concerns, flagship asset QVC still boasts a sticky customer base and fantastic free cash flow conversion. It is more difficult for us to understand the 2018 rally in multiple brick and mortar stores when QRTEA appears to have a superior model, including: no operating store leases, limited capex requirements, an international franchise and 57 percent ecommerce penetration, with roughly 66% of this mobile. QRTEA can also generate substantial value from the HSN deal even in limited/no growth scenarios if HSN simply hits synergy numbers. QRTEA held up far better than other retailers during the 2008/2009 recession, but we concede its shares would likely suffer during a downturn. But at 14%+ free cash and debt levels low relative to historic levels, we think QRTEA is better positioned to take advantage of a downturn than it was entering the 2008/2009 recession.

Why Not Wait for 20% Declines Before Buying? …The Power of Hindsight

We want to quickly address two final topics. As noted at the beginning of the letter, we initially bought some of the above names at prices 20-25 percent higher than current prices and then continued adding as the names declined. While we understand this can be frustrating, our experience has been that this is rarely a unique experience. Certainly, we would prefer to perfectly time our initial purchase and have the stock rip higher the nanosecond after we enter. Unfortunately, this rarely occurs. As we won’t invest more than a set cost percentage in any single name, we try to anticipate declines and leave ourselves enough room to take advantage of corrections. The rallies can be just as swift as the declines, and feelings of loathing about buying too soon are replaced by twinges of regret for not buying enough – the investment equivalent of the view that the food is terrible and yet portion sizes are too small.

To take one example, the market valued CODI’s 10 different businesses roughly 4 percent higher on November 30 versus today. But, between November 30 and Christmas Eve, the stock fell nearly 18 percent before jumping 25 percent over the subsequent 3 weeks. There is certainly a debatable range around the value of the various businesses, but these estimates have not materially changed over the past 45 days. It was impossible to know that the stock would decline so rapidly before Christmas and equally impossible to know it would rally so quickly afterwards. In fact, a couple of ill-received Fed comments, Presidential Tweets or institutional redemptions could have sent shares down another 20 percent. While we didn’t bottom tick the Christmas Eve low, we added as shares declined and prepared for additional purchases should the selling continue. While individual purchase decisions will look better or worse with the benefit of hindsight, we do strive to follow a consistent methodology.

As previously noted, many of our names repurchase sizeable amounts of stock. At given points in time, these repurchase decisions can look very foolish short-term, but this may not be the same conclusion looking back several years after the buybacks. In 2012/2013, we made a rare technology investment, purchasing a tiny Redmond based software company named Microsoft (MSFT). We knew little about Azure (speaking of hindsight bias…that turned out to be a nice grower!), but we calculated that if Windows went to zero, this ~$26-$27 stock would still produce over $2.00 per share in free cash flow. Interestingly, one of the big debates at the time was over the amount of capital that MSFT “wasted” repurchasing shares. With shares virtually unchanged over the previous five years, an exhausted shareholder base was tired of looking at charts like this:

So how was the IRR on those share repurchases? Well, not so hot if you measured it in July of 2013. At current $100+ levels (we sold the last shares before $60…insert hindsight bias comment), the picture drastically changes. We suspect a similar story could unfold at several of the names discussed in this letter. While we aren’t predicting 4 baggers from current levels (…although we weren’t aware of any human predicting this type of return for MSFT back in 2012), we are confident that intrinsic value is meaningfully higher across the portfolio. And whether it is CHTR retiring shares at $290, JEF at $23 or QRTEA at $19, we suspect that criticisms over lower or negative share repurchase IRRs might be replaced in a couple of years with complaints about stingy repurchase decisions when shares are meaningfully above current levels. In this spirit, we would like to formally reprimand MSFT for only buying ~$40 billion of its stock between 2009-2013…since it is obvious (circa 2019) that shares were far too cheap.

In conclusion, we understand that periods of volatility like that experienced in the fourth quarter can be unsettling, especially as sizable gains can appear to be erased in a matter of days. That said, we do not believe these arbitrary declines reflect changes in the value of our underlying businesses. While we cannot predict when the price/value gap will converge, we think the gap has become unsustainably wide in multiple cases and this generally is a good omen for future returns. As always, we continue to spend large amounts of time stress testing our assumptions and will change our opinion as new information becomes available. We continue to invest sizeable amounts of personal capital in the names described above with the goal of compounding our investments with clients.

^[1] Hites Ahir and Prakash Loungani Can economists forecast recessions? Some evidence from the Great Recession; as source data, the authors used Consensus Economics, which provides forecasts of real GDP for a group of professional forecasters for a large group of countries.
[2] In October 2018, LILAK purchased Searchlight Capital Partners, L.P’s 40 percent interest in Liberty Cablevision of Puerto Rico in exchange for 9.5 million LILAK shares.