Category Archives: 2018

2018 Year-end Investment Commentary: Easy come, easy go

As the saying goes, markets go up on an escalator, down on an elevator. January 2018s smooth escalator ride higher was more than outweighed by December’s swift elevator ride lower. In hindsight, the culmination of the January ascension in February’s storm of volatility was a precursor of what was to come. The Old Wall Street rules dictate that a 20% rise from lows in markets is a “Bull Market” while a 20% fall from highs marks a “Bear Market.” By that token, we had been in a relentless “Bull Market” since March of 2009, what many call the “longest on record.” As the saying goes, “all good things must come to an end.” The final nail to the post-Financial Crisis Bull came on Christmas Eve, 2018 when the S&P hit an intraday low that was 20% off highs.

During the fourth quarter we saw historically bad stock market performance stack up on several timeframes. This was amongst the dozen worst quarters ever for markets, with December ranking as the worst December. Included in this mess was a four day stretch of the S&P registering consecutive 1.5% down days, which according to Bespoke Investment Group only happened twice outside of the Great Depression.[1] Looking at lists that stack up the “badness” of this period is a history in true economic chaos featuring the Great Depression, the 1987 crash, the Iraqi invasion of Kuwait, the popping of the Dot Com bubble and the Great Financial Crisis. 1987 is very similar in terms of the prevailing economic conditions with a benign US economic outlook amidst turbulent geopolitical tensions and valuations on the high side of their longer-term ranges. The look of things and the extent of the washout though looks most similar to 2002:

Essentially 2002 was a three-month waterfall. The big difference is how in 2002 the fears at the time were far more severe than we are seeing today. Here’s one such warning during 2002’s bear market

In 1998 it was somebody else’s problem, with their blood splashing on our trousers,” he said. “This time we’re ankle deep in our own blood.” Unlike 1998, the current crisis is rooted in the United States. It’s our stock market bubble that popped. The United States was attacked, and our CEOs cooked the books. In 1998, U.S. investors were staring out at the world and wondering where it would end. Now the world’s investors are staring at the United States and wondering the same thing.

Griffin can imagine, he said, a Japanese-style pullback from risk in the United States, in which investors give up, banks refuse to lend and the economy languishes. He can imagine other countries, accustomed to the United States’ role as the world’s economic locomotive, not taking up the standard. He can imagine a crisis of capitalism of a 1930s order. Griffin thinks the people talking about deflation are probably wrong, but he doesn’t think they’re crazy.[2]

As we all now know, 2002 was merely the warning shot and the deflationary shock came before the decade ended.

A contrast with our expectations:

Our preview for 2018 featured a few appropriate points, but more wrong ones. The appropriate ones remain applicable and we will start there (we will spare you any Bitcoin here this year). First, it was clear that the returns from 2017 were the “pulling forward” of future returns. The year was simply too good to be repeatable. In 2013, when we issued a similar warning, we emphasized that the pulling forward of returns can resolve either with sideways action or downside. Despite this call as 2018 began, we were hardly saddened to see the year on track for a back-to-back repeat of strength through September. This was before Mr. Market suddenly hammered home the idea that a pulling forward of returns can also be corrected with swift downside, in addition to extended sideways stretches.

We were especially wrong on three big things: the slope of the yield curve, the outlook for financials and the reemergence of dispersion. It turns out all three of these expectations were correlated with one another. If you were to look at the yield curve from the day the S&P made its all-time high, the expectation of more slope was mixed: inside of three years there was a steepening but the long end had flattened.

The 10-year is the level to which a considerable amount of corporate and household lending activity is tied. In the week after markets peaked, the 10-year made new highs. When equities bounced again in November, there was yet another new high in the 10-year yield, over 3.25%. This was the highest 10-year yield since 2011! At the time, banks were exhibiting considerable relative strength to the rest of the market, despite a weak year-to-date showing. As the selloff in equities escalated, the main cause was attributed to rising bond yields. We got headlines like “Here’s why stock-market investors suddenly freaked out over rising bond yields this week” which raised the prospect that Fed rate hikes Fed might push us into recession sooner than feared.[3] Interestingly, through it all, credit spreads remained firm. Later in the quarter, credit yields gave way to market pressures; however, despite the worst peak to trough market move in a decade, credit spreads remained within normal bounds and well below the scarier tumult of the winter of 2016.

The context here is helpful. While credit has hardly performed well lately, markets went from incredibly narrow spreads to modestly wide ones. We are well within the “comfortable” zone. It would however be supportive to see the direction of change stop worsening and start improving. Markets grasp onto directional change and sentiment tends to extrapolate momentum.

Housing—“The Scene of the Crime”

The rate narrative and its tie to financials is important for housing–one of the key sectors driving our economic cycle. Housing was the economic “scene of the crime” last decade and while the fundamental backdrop is drastically different today, investors tends to anchor to the past risk in hunting for future risk. While the market down move started this quarter with the spike in rates, the pressure accelerated with the late October release of poor September home sales data.[4]

Several prevailing narratives emerged with respect to housing affordability:

  • Rising rates made mortgage interest a bigger burden for prospective purchasers.
  • Years of upward momentum in Case-Shiller, especially in areas with job growth kept younger would-be purchasers sidelined
  • Tightness in labor markets pushed up the key input cost for homebuilders (40-45% of input cost) leaving the builders in a position of either needing to sacrifice margin or pass on costs

The affordability problem amidst a backdrop of accelerating wage growth is a challenging cross-current for analyzing this key economic sector. For example, the low end of the housing market had been neglected by builders since the crisis; however, just this year, low-end jobs wage gains reached their highest level since the crisis. One of the most fundamental problems is a mismatch in new construction between what worked over the last decade and where the strongest demand resides today. A selection of quotes from recent homebuilder earnings calls highlight this phenomenon:

  • “We are experiencing healthy market conditions across most of our markets with solid demand, especially at affordable price points. And the supply of new homes remains limited. In this environment, we’re reducing sales incentives or raising prices in communities where we’re achieving our targeted sales pace, while striving to ensure that our product offerings remain affordable.” Q3 2018 D.R. Horton 7/26/2018
  • “So there’s not a lot of inventory out there at the affordable price bands, and much of the headlines, I think, are tied to higher price points that are seeing some slowdown. And we’re trying to stay ahead of that at the price points that are slower, are well above us today and where we’re operating. But if it comes down, we have to be prepared for that, and it’s in part why we’re rotating to lower priced communities, positioning smaller models in higher priced areas even to keep affordable and keep some insulation there. But we think market conditions are very good and continue to see a growth opportunity as we head into ’19.” Q3 2018 KB Homes 9/25/2018.
  • “Well, sort of on a general basis and there are some exceptions at the higher price points, it’s gotten a little bit softer than the lower price points have been. And that’s not unusual in a market that there’s a lot of publicity as to what’s going on in mortgage rates and price depreciation. Because the more affluent buyers time their purchase based on where they think the world is going. And so we think, as we said consistently, that this is just a market adjustment to a more normalized market from what was, really, a red-hot market in some of these markets.”Q3 2018 Lennar Corp 10/3/2018.

While some of the signs are troubling for the housing market, we think the shift from high-end to affordable is a mix-shift problem that will resolve without too much residual damage. Further, recent market activity has gone a long way towards correcting two of the big prongs impairing affordability: first, mortgage rates have come down considerably since peaking in early October, along with the 10-year yield; and, second, although housing markets correct more in terms of volume than price  the rise in price will be held in check by rising inventories, sluggish volumes and a down stock market. [5]

The call that remains outstanding is one that has become a mantra of sorts for us since the commencement of RGA Investment Advisors in the wake of the financial crisis and is perhaps the most important. Last year we said:

It is hard to sit in our seat today and say exactly which areas of the market would best withstand the next bear, for there are yet to be obvious areas of overinvestment like the technology sector in 2000 and the financials in 2007 that will lead the way down; however, we have confidence that the nature of the next recession (and let’s not forget, a recession is an inevitability eventually) will be different than the last. To reiterate: “the 2007-09 bear market was an actual liquidity crisis where deflation was the risk that imperiled the economy, whereas the next recession will be a normal Fed induced recession when inflation gets too high and the Fed thus raises rates to cool things off.

China and tight labor are the kindling for broader fears

One of the fears today is that market action can be self-perpetuating and self-fulfilling. In 2018, we saw no real signs that point to an imminent recession in the U.S. but markets have now gone a long way to price one in. The global situation is messy and the idea that “Trade wars are good, and easy to win” is truly being put to the test. [6] We are insistent that they are neither good, nor easy to win. That said, there is merit to pushing a harder line with China on global trade; however, we think a resolution in a reasonable timeframe is imperative.

A conversation we had with the CFO of a manufacturing company was truly insightful on this point. The CFO told us that the biggest challenge is not knowing how to pursue strategy. Companies spent over a decade building their supply chains to run through China–this was not done overnight. For some of the largest multi-nationals, it makes sense to accelerate planned supply chain diversification to areas like Vietnam; however, this CFO warns that Vietnam’s labor pool is not skilled enough for hard durables and the port volume is not deep enough to handle the swell in demand. Further the strategic bind is intensified because if there is a timely resolution to the trade war, a diverse supply chain will end up as a cost that dampens future returns. C-suite strategists are left flat-footed between action and inaction with severe costs on both sides of hindsight proves either side wrong. This leaves many in “wait and see” mode because it’s easier to undo inaction than action. In the near-term, the combination of a strengthening dollar and some tactical maneuvers like batching orders and changing the “importer of record” have helped cushion the impact, affording the luxury of time. This state of affairs cannot last for too much longer.

This CFO conversation featured a second insightful angle on the state of the US economy. While he expressed caution about the China situation, he emphasized that an even bigger challenge was the US labor pool. The CFO’s company manufactures at low wage rates, though with a degree of skill required for some harder tasks. The CFO emphasized how the ability to retain employees was killing efficiency necessitating a change in their compensation structure to enhance retention. The company went to higher base rate and incentive compensation tied to efficiency and productivity. This was echoed a late December survey claiming “82% of U.S. CFOs expect a recession to have started by 2020.”[7] Per the survey, “The biggest concern U.S. CFOs had was the tightening labor market, which makes hiring and, therefore, continuing operations and expansion more difficult.”

Tight Labor and the Fed

Federal Reserve Bank Chairman Jerome Powell has come under fire from various corners of the political and market sphere for raising rates in December despite weakening equity markets. We think his decision was largely motivated by concerns about the labor environment.  2018 saw the swiftest rise in average hourly earnings since the financial crisis and with full employment, the Fed is expressing caution that inflation can stay within its targeted range. This gets back to our point that the next recession will be caused by the Fed, and some seem to already believe this is the Fed’s intent right now (something we believe is an overly hawkish take on the Fed’s intents). We do think it was a mistake to raise rates in December and that the risk from holding on just one more month was far more modest than acting too swiftly, though at the same time we do not think the economy is so fragile as to tip over from one 25 bps move.

It is very hard to call a good employment environment a “headwind” to the economy. After all, the economy relies on consumers, with gainful employment, investing and spending the proceeds of their labor. Rising wages should thus lead to both better consumption and better investment. This should be one of the factors most supportive of preventing a deeper slide in the house market. Importantly, this could help alleviate one of the big mysteries of the wake of the Financial Crisis–why has productivity growth been so poor?

The answer might be as simple as the amount of slack in the labor market that persisted for far too long. Why would an employer invest in productivity when they could simply hire a helping hand for decently less than the inflation adjusted wage a decade prior? Investments in productivity take an economic calculus of opportunity cost. Can I buy technology that will enhance my productivity over the next five years for less than it would cost to train and employee a worker over this time? Today with unemployment in the high 3%s is very different than five years ago at this time sitting just beneath 7%. Below we will tie this force with one of our new investments during the quarter.

What if we go into a recession in 2019?

If we do head into a recession now (and it’s hardly a foregone conclusion), the economy will do so on much stronger footing than in 2008. The employment backdrop is one of the more important differences today. The composition of the workforce both in terms of age (millennial cohort is hitting peak employment rates and ages today) and sector are very constructive. Bank capital ratios are in their healthiest state in decades, household balance sheets are as strong as they have been in decades and the key industries that led to the 2008 recession massive job losses are on much firmer footing. Over any long-term timeframe, recessions are a guarantee. The key is understanding where the real economic and market damage will come from within in. In other words, it’s important to have confidence in a portfolio that there will not be true long-term impairments in value vs interim markdowns in stock prices that will eventually revert to trend once the dust settles.

Market structure matters a lot for how stocks handle shocks–whether they be intrinsic or extrinsic shocks. We spoke about the industry reclassification undertaken by S&P shaking up the “technology” label as a sector. The major shuffle, which was described as “the biggest impact on the sector landscape in GICS history” went into effect on September 21, 2018. Markets peaked on September 20, 2018. We do think that begs the question whether there are structural or fundamental causes to the subsequent market action. While one can never know for certain, the crowding of certain pockets of the active investment community (ourselves included) in the technology sector was vulnerable to the crosscurrents that the rebalancing flows created. Add to this pocket of vulnerability a sluggish housing number and ongoing China questions and you have the makings of a bear market with or without a recession.

This also helps resolve the question: “if the biggest risks are labor tension and the trade war with China, why are some technology stocks with minimal exposure to trade wars and labor an immaterial portion of margin the biggest losers over the last three months?”

Forward P/E’s today are the lowest they have been since 2013, the year in which we had some “normalization” post crisis:

In the end, 2018 marks one of the biggest one-year declines in the forward P/E in the past few decades fueled by a combination of strong realized earnings growth and a decline in the stock market. Some of the earnings growth was the sugar rush from tax cuts, though a healthy amount was realized via higher revenue growth. While one might call the early 2018 levels “the upper end of fair value,” today is on the lower end of the fair value range. Yes, we have been below these levels, but they were in far different environments. This makes now a compelling time for long-term investors to put money to work.

See through the lumps at Cognex’ bright future:

Cognex is one of the most compelling companies we have come across in some time. The company boasts gross margins in the mid-70%s and ROICs upwards of 100%. Yes, that is not a typo on the ROICs. Consequently, Cognex has built a healthy balance sheet featuring a cash stash upwards of 10% of the market cap today. The problem for the company is that demand cycles are incredibly lumpy, and growth is inconsistent from year-to-year. While Cognex targets a 20% compound annual growth rate in revenue, which they have come close to over the past five years, the longer-term realized growth is closer to 15%. This longer-term period includes the crisis.

Cognex is in the business of “machine vision.” They build the sensors and the supporting software for manufacturing, supply chain and logistics companies to process and analyze a high velocity of product and data moving through a field of vision. The industry is somewhat fragmented, with the two largest players (Cognex and Keyence) combining for about 40% of share in the core machine vision segment. The core TAM for Cognex will represent about $4.3 billion in 2019, expected to grow at an 8% compound annual growth rate into the foreseeable future. Vision itself is a mission-critical add-on to other core automation initiatives and is a modest percentage of the cost for a company to automate in their respective field. While there are some functions of vision that are commoditized, the high function areas are both research and relationship intensive. Cognex and Keyence spend meaningfully more than the rest of the industry in R&D (a target around 15% of sales at Cognex, one of the most R&D intensive companies) and they have dedicated sales forces of highly trained engineers who work in close conjunction with their customers to develop uniquely tailored solutions.

Cognex CEO Robert Willett explains this nicely in the company’s Q3’18 earnings conference call:

“I mean, machine vision is very difficult to do, and the competitive advantage of advanced products from us and a few of our competitors is very substantial. So in general, this is not a market driven by price. It’s a market still driven by technology. Of course, we could drop price to try to win share, but that’s never been something that we’ve wanted to do. It’s not what we’re about as a company nor I believe it’s really what about — our major competitors are about.”

Note that while the units sold are measured much like a hardware company would, what Cognex (and Keyence) really sell is the software designed to process what the hardware reads. Customers get an ROI within two to three years of purchase, creating an incredible pricing advantage for these two unique companies.

Willett, as CEO, has the background and incentive structure we admire. He built his own business–Willett International LTD–that he sold to Danaher for $125 million in 2003. Danaher merged Willett’s business into their Videojet subsidiary and made him the head of the business unit. You may have heard of Danaher recently, when its former CEO Larry Culp was picked to help GE out of its mess.[8]  Danaher is known for both outstanding operations built around the “Danaher Business System,” (DBS) a form of the Japanese “kaizen” and a deep capital allocation playbook deployed with prudence. DBS and kaizen feature continuous improvement and efficiency. Despite proceeds that would make anyone comfortable financially, Willett’s drive led to him joining Cognex in 2008 as an executive VP, rising to COO in 2010 and ultimately CEO in 2011 when the company’s founder Robert Shillman stepped back.

Cognex growth has been fueled by a robust secular tailwind driving automation’s increased adoption within and across various industries. The company has been further aided by taking a growing share of the revenue pie. One notably large customer is Apple at over 20% of revenue. This has left Cognex vulnerable to Apple’s iPhone upgrade cycle, further fueling the lumps. Apple is unquestionably a great customer and the willingness of Apple to pay Cognex an exceptionally high margin despite a reputation for pressuring key suppliers is demonstrative of both Cognex quality and importance. Importantly however, we see the future of revenue streams diversifying to Cognex’ benefit.  Two emerging areas of growth are logistics and 3d vision. Logistics is interesting for how low the overall penetration has been. Arguably this has been a direct result of the high unemployment rate over the past decade. With labor tightening alongside the rapid escalation in ecommerce and commensurate rebuild of supply chains to accommodate new market realities, the outlook is as bright as ever for logistics. [9]

Today, about 10% of Cognex’ revenues come from logistics, a market which vision has 10% TAM penetration. We see Cognex’ share of revenues from logistics nearly doubling over the next five years. Similarly, 3d vision will continue its own rapid growth.

We envision Cognex’ revenue base evolving along the following lines:

Notably, the future growth drivers have far less lumpy demand channels than the existing ones. Meanwhile, the existing channels will continue to grow albeit at their consistently inconsistent pace. Diversifying the revenue stream in this way will give analysts more confidence in forecasting out-years and investors more confidence in ascribing an appropriate multiple to the company’s earnings stream.

With over 100% ROICs, capital allocation is an important driver of future value and the hardest to model. The company boasts over $830m in cash, though does not screen that way since around $300m of that resides in “long-term investments.” Given Willett’s background at Danaher, it’s no surprise that capital allocation has been a strong suit. Despite the huge cash balance, Cognex’ share repurchases have been incredibly lumpy with the company demonstrably “buying low.” To that end, repurchases have been reaccelerating in the past year after having paused on the share price’s rapid ascent.

In our base case, we see a clear path to double digit annualized returns with a minimum three-year target holding period, assuming no return is generated on excess cash flow either via strategically timed repurchases or acquisition. We expect the primary driver of growth to be a presumed 12% CAGR in top line, with modest EBIT and EBITDA margin leverage. Note this 12% CAGR falls well short of the company’s 20% target and below the 15% realized growth rate. In our bull case, growth in the high teens for the next four years, settling into the mid-teens thereafter, alongside 200 bps margin leverage supports prices today into the $70s, with the same prevailing assumption on reinvestment of cash flow. In our bear case, a mid single digit CAGR and flat margins give us confidence that a worst-case outcome would yield a mid $30 stock over our targeted 3-5 year timeframe. This bear case also assumes a WACC of 10% vs 9% for the base and bull cases.

Full Stream Ahead with Roku:

In our January 2016 commentary, we detailed our attraction to networks and platforms in the digital age. We were insistent on the appealing valuations and growth prospects for a certain crop of elite company. Today we see a similar opportunity taking shape and we have found our favorite risk/reward setup in a “robust network for the long-term” since then: Roku, Inc. (NASDAQ: ROKU). Many of you heard of Roku and have a streaming device of some kind in your household–likely also a Roku. Our younger readers are likely not just to have a Roku, but to consume a high percentage of video content on one of their devices. Per Anthony Wood, CEO of Roku, “10% of the 18 to 34 year olds watch their television on a Roku,” meaning Roku itself is the exclusive platform through which this key demographic watches TV.

You can get a sense of the future by seeing how this breaks out timewise:

(Nielsen Q2 2018 audience report)

Roku IPO’d in 2017 and the stock has been highly volatile ever since. At the time of IPO, approximately 56% of revenues were driven by low margin hardware sales (“Player” sales) and the remainder from “Platform” sales. In the first quarter of 2018, Platform exceeded Player sales for the first time in the company’s history. This was the inflection point upon which it became clear the company’s future value will be driven by Platform sales. The company strategically views Player sales to drive placement of the Platform. Once a streaming device (whether a stick or increasingly a TV itself) is placed in a household, the company then seeks to monetize an account with a combination of successful subscription and Video-on-Demand (VOD) purchase royalties (1/3rd of Platform revenues) and, more importantly advertisements (2/3rds of platform revenues and growing quicker than the combined segment). Ads take several forms, including home screen placements for audience building, but most important is advertising VOD (AVOD).

Once a household has an “Active Account” on Roku, the Platform side benefits from several levers enhancing monetization over time. First, with a surging abundance of streamable content, hours streamed continues to grow. Second, Roku can “fill” more of the available inventory on the platform. Third, Roku can drive better CPMs (cost per mil/thousand impressions) augmenting advertising inventory with its growing trove of user data. Roku boasts 95% completion rates on commercials, which contrasts favorably with linear TV.


The enhancement to the user experience offered by lower ad loads makes the ads themselves more valuable. Scott Rosenberg, the one who runs the advertising platform at Roku, described it as follows in their Q3 2018 call:

“We’ve been very effective at showing them the ROI math, that the last dollar that they put in the linear TV is reaching a smaller and smaller base, and that if they move that dollar to Roku, it’s going to deliver much more reach and ROI. I’m just — I mean, backing up more broadly and thinking about our advantages as an ad platform, the reason advertisers come to us in the first place is we’re an ad-scale platform with the largest, most engaged user base. It’s a unique audience that can only be reached on Roku. We’ve got that direct consumer relationship and all of the data that flows from it. And ultimately, it’s ads that work.” Roku is building its own “walled garden” with this virtuous cycle.”

In 2018, just shy of $70 billion will be spent on advertising over linear TV. When measured as “linear TV,” advertising is no longer growing.[11] Advertising on digital platforms will grow nearly 19% in 2018. Advertising on the Roku platform will nearly double. Roku was early to realize there was demand for free content, supported by ads in the streaming world, while the rest of the world was focused on subscriptions. As a result, some estimate Roku controls as much as “87% of ad requests” on connected TV today.[12] This kind of scale has advantages for viewers and advertisers alike.

We think inevitably most TV will be consumed through one of the digital platforms and as of today, two clear leaders are emerging: Roku and Amazon. Each boasts unique advantage and each benefits from considerably more share in the relevant metrics than the peers on the outside looking in (like well capitalized Google and Apple). One of Roku’s most important advantages in this competition stem from the platform’s openness. On Roku, you can watch Amazon’s Prime TV, YouTube, access the Google store and channels with add-ons ranging from ESPN to NBC, whereas on Amazon there have been extended periods of time where YouTube is inaccessible. Roku’s combination of interface, quality and low price are garnering favor with customers and reviewers alike. For TV manufacturers competing amidst this monumental shift in the content distribution landscape the answer is rather simple. Wood summed up the advantage of Roku software for a manufacturer as follows:

“So for example, so what’s better about Roku software? Well, a big one is that phones are expensive, like they cost a lot of money, they’re super computers. TVs are cheap, the main board on a TV is $25 and TVs are brutally cost competitive, no one makes money in the TV business. So our software runs on low-cost TVs, it costs less to build a TV with Roku software. When you’re trying to get 50 cents off your bill of materials so you can win a Black Friday special at Walmart, the amount of money you save by cutting your RAM in half and your CPU in half by running Roku software — which actually has great performance and more content — is huge. It’s the difference between getting distribution and not getting distribution in Walmart.”[13]

Roku is led by Anthony Wood and its origin story sounds more Hollywood than Silicon Valley. Wood himself owns 25.9 million shares, or 24.2% of the company. Wood’s ownership is via the Class “B” voting shares, of which he controls 42.3% of the vote. As such, investors in Roku are effectively business partners with Wood, granting him an extremely high degree of control. While the separation of voting control is not ideal, we like how much Wood’s own fortunes are tied to Roku’s stock and view this as an important alignment of interests.

Wood’s background is interesting and worthy of such co-investment. He founded a DVR company that was an early competitor to TIVO, called Replay TV and learned much about the box placement and content challenges along the way. The successor company to Replay TV is now owned by DirectTV. The proceeds from this early foray helped Wood bootstrap Roku early on and has led to a determination to conservatively finance the company while investing in growth. Roku’s history is tied to Netflix and streaming itself. In fact, it’s first name was “The Netflix Player by Roku.” The idea within Netflix was to create a platform for easier distribution of streamed content, but Netflix instead decided to pursue licensing arrangements with various hardware-based distributors ranging from gaming consoles to TV makers. Wood speaks with grand ambition when he talks about making Roku the operating system for the future of TV:

So the big picture is if you think about when new competing platforms have emerged, the software platform has always changed. So if you go back, we were talking about PCs back in the early days, well before PCs, there were mini computers like that PDP 11. Those had their own operating systems. Those operating systems didn’t make the transition to PCs. Instead of operating systems designed for PCs, Windows became the dominating operating system on PCs and then when phones became a computing platform, Windows didn’t make that transition. No one’s running Windows these days on their phone, they’re running Android or iOS. [14]

Since Roku places an actual piece of hardware in a house, increasingly in the form of a TV, the churn associated with the platform is inherently lower. Per Roku’s analysis, one out of four smart TVs sold in the US are powered by Roku’s OS. Evidence of this is provided by TCL’s rise from a non-factor in the TV industry (24th in market share) to number three overall share, driven almost entirely by the Roku relationship.[15] The Washington Post recently called a Roku-powered TCL TV “the best deal on a premium TV I’ve ever seen” when comparing it to a high-end Samsung [16]

It is important to consider both the actual and the strategic value of Roku. Disney, which has every reason to downplay the value of Hulu as it seeks to acquire the 40% it does not own recently valued Hulu and it’s near 30 million users at $9.26 billion[17]. In contrast to Roku, which is modestly cash generative in today’s high growth mode, Hulu continues to burn cash. Further contrasting the two is how Roku places hardware whereas Hulu sells subscriptions. Digital subscribers tend to churn with regularity.

As of today, Roku boasts nearly 24 million households. In its Q3 2018 report, Roku reported trailing twelve-month annual revenue per user (APRU) of $17.34, up just shy of 48% yoy. CPMs on the platform are said to be upwards of $30, rivaling some of the best networks on linear TV. Roku tastefully inserts ads at approximately 1/4th the industry average ad load (4 minutes of advertising per hour on Roku vs 16 minutes on linear) and have banned pre-roll ads before movies or shows in an effort to “premiumize” the experience watching content monetized with ads. The effort to make advertising more personal and to consume less time per hour makes the platform and its supporting ads appear less intrusive to consumers. This makes viewing content on Roku increasingly appealing driving further growth in hours viewed per account, thus making the entire platform increasingly attractive for advertisers to be on.

The Roku Channel (TRC) itself is becoming a key driver of value. TRC was first launched in October 2017 and is already one of the top 5 most watched apps on Roku. Originally, TRC was solely available on Roku itself, but now the company has a distribution agreement with Samsung covering their smart TVs (perhaps a precursor of a bigger agreement), has enabled web-based login and a forthcoming mobile app. These changes have untethered TRC from the Roku hardware and unleashed TRC as a streaming asset with universal availability. The increased engagement with TRC has driven ad revenue growth which has empowered Roku’s ability to acquire more desirable content. There are also signs Roku will use TRC as a home screen and a modern aggregator/bundler of content. Roku started the New Year by announcing the availability of subscription content within TRC that will be payable all via one simple, monthly bill.[18] Our guess is that this subscribable content is monetized like the standard royalty Roku takes on subs over its platform. There have also been suggestions The Roku Channel might eventually become fertile grounds for Netflix to release past seasons of some of their own shows as advertising supported content, to both further monetize that content and tempt new subscribers to join.

At the end of 2019 we expect unit economics to look like the following:


These assumptions are also unfair to the Player business, which does generate margin. The company views this “margin” as a form of “negative customer acquisition cost” and for the purposes of our analysis, we use this as an offset to selling & marketing expense per new active user. Wood explained the margin rationale as follows: “And if we felt like…it’d be better for us to go to 0 gross margin or even negative on players, we would certainly consider that. But right now, at some point, there’s diminishing returns. I mean, the cheaper you make the players, the more likely that someone buys it and doesn’t use it.”[19] Laying out the money up front for a Roku device is effectively a commitment that it will be used. This is the same rationale behind a Costco or even Amazon Prime membership.

If Roku can get to 40 million Active Accounts (32% penetration of the 126.22 million US households, leaving aside the international opportunity for now) and an ARPU of $25 by 2020, the company can hit $1 billion in platform sales (approximately 66% growth for Active Accounts and 50% growth for APRU over expected 2018 final results, both implying a fairly large slowdown from present trends). This $1 billion expectation is slightly ahead of the consensus expectation, which has been rising consistently as the year progressed (Roku is one of those rare cases where a stock will finish the year down 40% all the while forward revenue expectations rose). If we assume no value or margin for the Player business, we can use today’s roughly $3 billion market cap and think about what the 3x P/S means. P/S = net margin * payout ratio * (1+g) / (r-g). In solving for net margin, assuming a 10% WACC and either 4% or 5% growth, you get an implied net margin at maturity of 17.3% or 14.3% respectively.

We think 3x what is an achievable 2020 platform sales targets a unique wager. If beyond that timeframe Roku achieves one of the following, returns on an equity investment today will be comfortable in the double digits annualized:

  • Greater than 40m households long-term
  • Greater than a $25 ARPU
  • Greater than a 20% net margin
  • Growth above terminal rate in 2020 and beyond

It’s possible for Roku to exceed each of these hurdles, with ARPU being the most likely and an answer on net margin farthest in the future as the company invests to drive growth beyond the forecast period. To put the 3x P/2020 Platform sales in context, many great high growth digital platforms trade with P/S upwards of 8 (see NFLX, SHOP, CRM, SQ, ETSY, and TTD to name a few). We expect the primary driver of returns to be Roku’s delivery of exceptional growth; however, we also see a path to multiple expansion as the market comes to appreciate the company’s uniquely advantaged business model, scale and free cash flow generation ability on a recurring style of revenue base.

Thank you for your trust and confidence, and for selecting us to be your advisor of choice.  Please call us directly to discuss this commentary in more detail – we are always happy to address any specific questions you may have.  You can reach Jason or Elliot directly at 516-665-1945.  Alternatively, we’ve included our direct dial numbers with our names, below.

Warm personal regards,

Jason Gilbert, CPA/PFS, CFF, CGMA
Managing Director
O: (516) 665-1940
M: (917) 536-3066

Elliot Turner, CFA
Managing Director
O: (516) 665-1942
M: (516) 729-5174






[5] Leamer, Edward. The Housing Cycle IS The Business Cycle










[15] and




[19] Roku Inc at Citi Global Technology Conference 9/6/2018

Past performance is not necessarily indicative of future results.  The views expressed above are those of RGA Investment Advisors LLC (RGA).  These views are subject to change at any time based on market and other conditions, and RGA disclaims any responsibility to update such views.  Past performance is no guarantee of future results. No forecasts can be guaranteed. These views may not be relied upon as investment advice.   The investment process may change over time. The characteristics set forth above are intended as a general illustration of some of the criteria the team considers in selecting securities for the portfolio. Not all investments meet such criteria.  In the event that a recommendation for the purchase or sale of any security is presented herein, RGA shall furnish to any person upon request a tabular presentation of:
(i) The total number of shares or other units of the security held by RGA or its investment adviser representatives for its own account or for the account of officers, directors, trustees, partners or affiliates of RGA or for discretionary accounts of RGA or its investment adviser representatives, as maintained for clients. (ii) The price or price range at which the securities listed.

Q3’2018 Investment Commentary

The second quarter was outstanding for our portfolios; however, the market absolved us of the opportunity to take comfort in those results rather swiftly. While this is our third quarter writeup and we typically confine this space to a discussion of the events during the course of the titled quarter, we feel it is important to address the market’s gyrations thus far in the fourth quarter.

Seemingly on a dime, market sentiment changed from fairly constructive to rapidly destructive. When markets enter a period such as this, we attempt to identify the excuse. For most of the October selloff, there were a multitude of explanations ranging from China to interest rates to housing to Italy to tariffs. We try to know the prevailing narrative as to why the market is dropping, because once we get a sense for the why, we can then hone in on assessing the impact:

  • How big a risk is it? Can we quantify the effect on the economy and most importantly, our core holdings?
  • Has the market already discounted the extent of the potential risk?
  • Is there a way for policymakers (whether that be politicians or the Fed) to fix the problem?
  • How does the specific threat come to an end or come to pass?

The Federal Reserve’s interest rate hiking cycle has certainly caused concerns, specifically in housing. The selloff started alongside the long end of the interest rate curve steepening and rates moving higher. While this certainly matters, we would caution reading too deeply into the impact, especially considering the fact that before the selloff hit its most extreme levels, much of the long end interest rate move had already reversed.

We think the foremost reason for this October selloff has been the impact of the trade war with China. This has been evident throughout earnings season, where companies talked about the drain the tariffs have placed on margins and the uncertainties the entire trade war with China engenders over years of supply chain strategy and investment.[1] Although investors had time to prepare for these potential problems, we think far too many expected them to pass before visibly seeing any impact in earnings. Further, while the effects themselves are rather modest in what we would call the largely insular US economy, they are rather pronounced in the earnings of global multinationals—exactly the kind of companies which populate the S&P 500.

For years, manufacturing companies built their processes around at least some degree of Chinese production capability and with that imperiled, it places a severe stress on planning and willingness to invest going forward. While housing seemingly is an interest rate story, there also is a degree of Chinese tariff risk: in some talks with homebuilders, we heard that relatively small costs like lighting fixtures have seen price increases upward of 20%. If several of these small costs jump apace, homebuilders must face several tough choices concurring with a rise in interest rates. Do homebuilders raise prices? That might be tough for the market to swallow. Do homebuilders accept lower margins? Shareholders certainly don’t like that. Do homebuilders push down the feature-set of new houses to lower their cost base? That’s certainly a realistic possibility which results in less overall demand for product generally.

We have referenced Michael Mauboussin’s distinction between uncertainty and risk in the past:

“Risk: We don’t know what is going to happen next, but we do know what the distribution looks like. Uncertainty: We don’t know what is going to happen next, and we do not know what the possible distribution looks like.”[2]

Here we know there are distinct risks and this past earnings period saw them quantified. We also must cope with the uncertainty of not knowing directionally how investment strategy will be engineered going forward for companies with global supply chains. Given this landscape, early in this October, we sold our one holding with the most risk to these specific problems (Williams Sonomia, NYSE: WSM). The position had reached our sell range and the Trade War risks and uncertainty provided the right impetus to exit, reassess and move on. Beyond this one position, the rest of our portfolio has pockets of China risk and exposure though it is far smaller, more esoteric, and in no instance requires a major revamp of corporate strategy and planning.

PayPal: A core position for the long-run

PayPal has been and remains our largest position for years now. We commenced our position via eBay before the companies split in two in the summer of 2015. Since that time, much has changed at PayPal and the company has performed well; however, we think there continues to be an extreme misperception from Wall Street in analyzing the business itself. Specifically, analysis tends to be detailed and thoughtful, but lacking in creativity. Most analysts pick apart take rate every which way in an attempt to deduce underlying net revenue and gross margin trends, while the bigger picture opportunity is given little thought. We like simplifying an investment thesis to its essence and while take rate is certainly important here, we view engagement as the end-all, be-all metric to assess our investment in PayPal.

Let’s first take a step back and look at the “take rate” equivalent of some of our most important payment companies over time. First, American Express:[3]

Then Visa:[4]

Note the persistent downward trend in the “take rate” of each American Express and Visa over the course of over 40 years in American Express’ case and 22 years in Visa’s case. These trends have persisted and meanwhile each company grew mightily during the covered timeframe and continues to enjoy a strong competitive position today. What force overcome these “take rate” headwinds? It was engagement—engagement in the form of an increase in usage by the average user. Let us now look at engagement trends over time at PayPal. Here are the most recent quarterly trends from PayPal’s Q3 earnings release:[5]

And here are the customer engagement trends going back to 2012:[6]

A few points are worth noting:

  • In 2012, the typical active user at PayPal used the service once every 2.4 weeks, or less than twice a month. In 2018, the typical user will use the service about once every week and a half, or a little more than three times a month.
  • Notice the acceleration in engagement starting in late 2015 and early 2016. We attribute much of this acceleration to the benefits of independence from eBay with PayPal’s mission pivoting from driving sales on eBay to creating the ultimate consumer experience for online checkout. A singular focus on PayPal’s essence was crucial for driving meaningful improvements in the experience.
  • Engagement growth pulled back in 2017 though this came from a step-change in the growth rate in new active users. In 2016, net new actives checked in at 10.06% vs 15.23% in 2017. The acceleration in growth was driven by “choice” following the Visa and Mastercard partnerships, enabled by easier onboarding and less friction throughout the user interface. New users take time to reach the engagement of more mature users. What’s notable is that the customer acquisition cost (CAC, measured by sales & marketing spend per net new user) dropped from nearly $54 to $37.60. In PayPal’s third quarter 2018 earnings report, CAC checked in at less than $36. Note: here we are looking at CAC on a net user basis, which overstates true CAC. Gross active user additions would account for churn. For the sake of this analysis we are focusing on simplicity.
  • In 2018 the company will maintain its new user growth acceleration, while engagement growth re-accelerates.

Why is engagement so important? The significance is both qualitative and quantitative. Qualitatively, engagement is the purest measure of the value that active users are getting from PayPal. When engagement is rising, PayPal’s users are expressing the fact that PayPal’s service is more valuable and more useful to them. The extent of the rise is significant, as the average user thinks about the company more than three times as frequently on a monthly basis as they did five years ago. This increase in mindshare is important for how it influences the network effects operating within the business and forges a moat that will be impenetrable for the long run. The benefits with respect to the network are twofold:

  1. The more frequently the average customer uses PayPal, the more valuable accepting PayPal becomes to the merchant side of the network. More engagement thus leads to more merchant uptake.
  2. The more value each user gets out of PayPal, the more people who don’t yet use PayPal end up seeing reasons to engage with the company. In part, this is driven by peer-to-peer (whether it be core PayPal or Venmo) and in part this is driven by the positive sentiments the frequent users express because of their experience with PayPal. Users thus become PayPal’s best promoters.

The quantitative benefits are meaningful as well. Growth in customer engagement is more valuable and higher margin than growth in new actives. New actives require customer acquisition cost, whereas increasing engagement from existing customers has no such cost against it. New actives are obviously an important driver of top line growth & economic profit long-term; however, the Street consistently underestimates the importance of engagement as both a profit and margin driver. One way to think about this is to contrast the lifetime value (LTV) of a customer to PayPal with average engagement versus one with double the engagement.

For illustration purposes, we used a 10 year lifespan, but note that the ratio of LTV enhancement would be the same were the lifespan 5 years. The “other costs” line above includes the full brunt of product development (R&D), customer support and operations and general and administrative expenses. We would make the case that much of the R&D line in particular goes towards growth investments and is not thus part of the individual customer economic value analysis; however, for the purposes of this illustration we are inclined to include it. The key takeaway here is that when engagement doubles, there is a nearly 3.5x improvement in LTV.

This is an important exercise for two reasons. First, while we can analyze averages across the enterprise, there are cohorts in PayPal who are far more engaged than others and these engaged customers are far more important for the overall value of the business. While PayPal does not disclose any number approximating churn, we must therefore make assumptions. Our assumptions can target the average, but it’s also important to think about cohorts. For example, were the 80% of PayPal customers extremely sticky with customer lifespans a decade or longer and the churn existent in the 20% around the edges, average/aggregate churn might be fairly high, while the LTV of the sticky 80% would be far greater than the LTV of the average. Second, if PayPal can continue to deepen engagement from its average customer, this would be a meaningful driver of value.

Between the big drop in customer acquisition costs starting in 2017 and continually rising engagement, the lifetime values of existing customers continue to trend higher at PayPal and the value of each incremental user continues to rise. Unlike the typical subscription service that has a price and sticks to it (or raises it modestly), PayPal enjoys two levers on the customer level that the typical SaaS within the LTV framework does not. We’ve spoken to engagement, but also the average transaction price itself rises at the rate of inflation. As inflation rises, PayPal’s unit economics improve at a faster rate than its expense base grows. Considering our ongoing theme of the economy normalizing and inflation returning to normal levels, this degree of pricing power is a powerful lever for the stock’s value longer-term. Once the analyst community realizes they can move past their fear of take rate erosion and embrace the powerful levers of engagement, the stock will be treated quite differently and far less skeptically by the investment community.

A Quick Comment on Walgreen’s:

Walgreen Boots Alliance entered the third quarter at its lowest share price in four years on the heels of Amazon announcing a late second quarter acquisition (on Walgreen’s earnings day no less). Amazon acquired PillPak, an online pharmacy with a differentiated offering.[7] We addressed the Amazon risk for Walgreen’s head-on in Elliot’s presentation at the Manual of Ideas Best Ideas Conference in January of his year.[8] While the market instantly focused on the risk, over the course of the quarter Walgreen’s shares recovered their lost ground and more. This offers an important lesson in sentiment.

While Walgreen’s sold off each time there were whispers of Amazon entering the pharmacy business, the materialization of an entry ended what has been a nearly two yearlong running speculation of the ecommerce giant’s intents. In the speculative phase, prognosticators allowed their imaginations to run wild with upside scenarios for Amazon and harsh consequences for the incumbents in Walgreen’s and CVS, with timeframes for these predictions seemingly irrelevant. Upon actualization however, the focus now must shift to execution. Before the PillPak acquisition, Walgreen’s the stock was fighting a perception battle with hypotheticals where price action was driven by narrative. Now Walgreen’s is fighting a competitive battle where market shares and margins will be the arbiters of success. On this front, Walgreen’s continues to gain share, though the company is sacrificing margin at the pharmacy to do so, while the front end continues to stabilize around a SKU rationalization and margin enhancement strategy.

Thank you for your trust and confidence, and for selecting us to be your advisor of choice.  Please call us directly to discuss this commentary in more detail – we are always happy to address any specific questions you may have.  You can reach Jason or Elliot directly at 516-665-1945.  Alternatively, we’ve included our direct dial numbers with our names, below.

Warm personal regards,

Jason Gilbert, CPA/PFS, CFF, CGMA
Managing Director
O: (516) 665-1940
M: (917) 536-3066

Elliot Turner, CFA
Managing Director
O: (516) 665-1942
M: (516) 729-5174



[3] Evans, David and Richard Schmalensee. Paying With Plastic: The Digital Revolution in Buying and Borrowing.

[4] Ibid.


[6] Sentieo data, RGA Investment Advisors analysis and estimates



Past performance is not necessarily indicative of future results.  The views expressed above are those of RGA Investment Advisors LLC (RGA).  These views are subject to change at any time based on market and other conditions, and RGA disclaims any responsibility to update such views.  Past performance is no guarantee of future results. No forecasts can be guaranteed. These views may not be relied upon as investment advice.   The investment process may change over time. The characteristics set forth above are intended as a general illustration of some of the criteria the team considers in selecting securities for the portfolio. Not all investments meet such criteria.  In the event that a recommendation for the purchase or sale of any security is presented herein, RGA shall furnish to any person upon request a tabular presentation of:
(i) The total number of shares or other units of the security held by RGA or its investment adviser representatives for its own account or for the account of officers, directors, trustees, partners or affiliates of RGA or for discretionary accounts of RGA or its investment adviser representatives, as maintained for clients. (ii) The price or price range at which the securities listed.

Q2’2018 Investment Commentary

We are now at the halfway mark of the year, which always sets the stage nicely for a review of expectations coming into the year.[1] We will refrain from pounding our chests too mightily on our Bitcoin outlook considering we have no “skin in the game.” Our learning has continued amidst the price collapse, though we continue think patience and knowledge accumulation is worth far more than action in the crypto world. One of our primary expectations was for markets to spend the period digesting 2017 gains. Were we to draw a straight line from the S&P 500s closing price for 2017 to the 2018 first half closing price, it would certainly look like digestion:

However, as we know, markets tend to move in fits and starts instead of straight lines. Digestion takes many forms and the chop we have experienced thus far in 2018 certainly is one of the more prominent forms. Alongside this expectation for digestion was our call for renewed dispersion between sectors and stocks. This looks prescient thus far and has been both rewarding and a source of opportunity that we are now hopefully taking advantage of.

Our most misguided expectation thus far has been the slope of the yield curve. We expected some steepness to begin emerging in 2018 and to date, the yield curve has actually compressed. The spread between the 2-year Treasury and 10-year Treasury has neared but not gone through the “inversion” level. This has been a headwind for net interest margin at banks, though it has not placed too big a hurt on the financial sector yet. This is largely due to abundant positives for the financial sector that have so far outweighed the negatives of a flattening yield curve.[2]

An Active First Quarter:

First quarter volatility gradually eroded during the course of the second quarter of the year. Markets were far calmer and technology stocks in particular exhibited considerable strength. We used the early quarter volatility to make five notable moves, closing out of two positions, doubling down on one and commencing two new holdings. The second quarter thus marked our most active quarter of portfolio activity since the second quarter of 2016.  We are especially excited about one of the new additions, for how we think this purchase elucidates our philosophy, process and discipline—Equinix, Inc (NASDAQ: EQIX).

We first learned of Equinix years ago and as many were wont to do, we dismissed the stock as merely another datacenter marketing itself for having recurring revenue while the primary asset depreciated rapidly and required constant replenishment. Slowly our view became more informed as we spoke with various investors and industry participants. The key factor that influenced our desire to truly dig in on in this company was the suggestion to read Tubes: A Journey to the Center of the Internet by Andrew Blum, in late 2015. We featured this as one of the “Best books we read” in our year-end 2016 commentary.[3] The central thesis of Tubes is that despite the Internet appearing rather abstract and terms like “the cloud” seemingly implying the Internet itself exists in the air, the entire edifice is built on a highly tangible, physical infrastructure that is essential to global connectivity.

Blum offered some of the following background on Equinix:

  • “Adelson relied on a crucial hunch about how the structure of the Internet would evolve: networks would need to interconnect at multiple scales. They had to not only occupy the same building but the same building in several different places around the world.”[4]
  • “The “ix” in Equinix indicated an “Internet exchange”; the “equi,” their intent of being neutral and not competing with their customers.”[5]
  • “Adelson loved that idea: that an engineer responsible for a global network would feel at home in Equinix facilities everywhere. There are about one hundred Equinix locations around the world and all of them carefully adhere to brand standards, the better to be easily navigable by those nomads in endless global pursuit of their bits. Ostensibly, Equinix rents space to house machines, not people; but Adelson’s strikingly humanist insight was that the people still matter more. An Equinix building is designed for machines, but the customer is a person, and a particular kind of person at that. Accordingly, an Equinix data center is designed to look the way a data center should look, only more so: like something out of The Matrix. “If you brought a sophisticated customer into the data center and they saw how clean and pretty the place looked—and slick and cyberrific and awesome—it closed deals,” said Adelson.”[6]
  • “The rationale for an Internet exchange is straightforward, and not very different from the founding principle of MAE-East: get your packets to their destination as directly and cheaply as possible, by increasing the number of possible paths.”[7]
  • “The two most used criteria are the amount of traffic passing through the exchange (both the peak at a given instant, or on average), and the number of networks that connect across it. In the United States, exchanges tend to be smaller; mainly because Equinix has been so successful in allowing networks to connect directly to each other. The big IXs, in contrast, rely on a centralized machine, or “switching fabric.””[8]

Following our intrigue, we did considerable work on Equinix. Our model suggested the price was fair and offered some upside, though not quite enough. Meanwhile the stock headed higher with hardly a pullback. As this happened, we remained current on the business and kept updating our thesis accordingly. We developed this hope that given Equinix’s prominence in REIT indices (it’s structured as a REIT) it would hit a growth hiccup concurrent with an uptick in rates and a yield-induced panic in yield sensitive sectors. As luck would have it, the confluence of negativity struck Equinix in the first quarter of this year, with investors in yield-sensitive stocks running for the exits. Equinix investors started fearing a slowdown in the long-term growth rate of the company due to “hyperscale” providers building more of their own infrastructure and comments alongside their first quarter earnings report that suggested indigestion of the Verizon assets acquired in Miami.

Needless to say, Equinix as it is positioned is a toll-road of sorts on the internet. While most people colloquially call this a datacenter company, Equinix is actually selling connectivity and specifically in scarce, but high value locations. The advantage Equinix boasts compounds as the company increases in scale and shows up in the company’s industry-leading margins and returns on invested capital. A quote from Interim CEO, President and Executive Peter Van Camp (PVC as they call him at Equinix) is emblematic of just how essential Equinix’ locations are:

“the subsea landing cables are, certainly, a great manifestation of the business model as well. There are actually more subsea cables being laid right now than there have been in the last 20 years. So these are cloud service providers forming consortiums, also with the networks and we’re the traditional ones that would lay these cables. But there have been 18 awarded in the past 18 months, I would guess, at that time, something close to that. Equinix has won 17 of them. And why have we won them? As that cable lands, all the traffic on that cable then needs to diversify and be spread to the networks and end destinations for whatever that data may be about. And so the logical place for a cable landing station is an Equinix data center, where it has the most immediate route, lowest latency to all of the end destinations for the data it carries. So this just continues to support this business model of interconnection and why we think we will always be in a strong, very relevant place about being the aggregation points for the digital world.”[9]

This is extremely important, because it is advantageous in and of itself for how it premiumizes the real estate within an Equinix exchange and Equinix in particular is the only company with true scale, globally. Moreover, it enables Equinix to push its business model even further. Recently, Equinix has been highlighting its ECX Fabric, which enables clients to be “in two places at once.” As the company explains:

ECX Fabric connects Equinix data centers all over the world to create an interconnected fabric of extremely fast connections. The connections are both virtual and physical. They make it possible for a Finnish company to connect to their brand office in Singapore almost as if they were physically in the same building… You can now connect to the cloud provider of your choice directly, bypassing the congested public Internet altogether. For example, AWS’s Frankfurt region is available in Equinix’ interconnection exchange in Helsinki.[10]

Two and a half years after really digging into this company and taking a liking did we finally feel we could underwrite the investment in pursuit of our return expectations. We think buying this stock with an approximately 6.25% yield on adjusted funds from operation with a 45% payout ratio and a long runway of double digit growth will be looked upon years down the line as an outstanding bargain.

We know there are more such future opportunities lurking in our watchlist and we are excited by how with each passing year our watchlist grows in multiple directions. We are increasing the number of companies that we know and we are accordingly increasing our depth of knowledge on each company populating our watchlist. We maintain this list alongside the price at which we would be ready and willing to step in and buy any one particular company.

Thank you for your trust and confidence, and for selecting us to be your advisor of choice.  Please call us directly to discuss this commentary in more detail – we are always happy to address any specific questions you may have.  You can reach Jason or Elliot directly at 516-665-1945.  Alternatively, we’ve included our direct dial numbers with our names, below.

Warm personal regards,

Jason Gilbert, CPA/PFS, CFF, CGMA
Managing Director
O: (516) 665-1940
M: (917) 536-3066

Elliot Turner, CFA
Managing Director
O: (516) 665-1942
M: (516) 729-5174




[4] Tubes, page 87.

[5] Ibid, page 87

[6] Ibid, page 95.

[7] Ibid, page 109.

[8] Ibid, page 111.

[9] REITWeek, NAREIT Investor Forum, June 6th, 2018.


Past performance is not necessarily indicative of future results.  The views expressed above are those of RGA Investment Advisors LLC (RGA).  These views are subject to change at any time based on market and other conditions, and RGA disclaims any responsibility to update such views.  Past performance is no guarantee of future results. No forecasts can be guaranteed. These views may not be relied upon as investment advice.   The investment process may change over time. The characteristics set forth above are intended as a general illustration of some of the criteria the team considers in selecting securities for the portfolio. Not all investments meet such criteria.  In the event that a recommendation for the purchase or sale of any security is presented herein, RGA shall furnish to any person upon request a tabular presentation of:
(i) The total number of shares or other units of the security held by RGA or its investment adviser representatives for its own account or for the account of officers, directors, trustees, partners or affiliates of RGA or for discretionary accounts of RGA or its investment adviser representatives, as maintained for clients. (ii) The price or price range at which the securities listed.

Q1’2018 Investment Commentary

The first quarter of 2018 began much the same way 2017 ended: with markets steadily grinding higher daily, while expected volatility plummeted. It seemed almost natural as analysts updated and revised higher their bottoms up and top down estimates to account for the impact of the Tax Cuts and Jobs Act of 2017. Alas, the trend was too good to be true and swiftly reversed without even a modest pause. The VIX, a popular measure of expected volatility in markets, went from exceedingly placid levels to its most extreme positioning in years over the course of a few days. So harsh was the sudden surge in volatility that two popular volatility ETFs essentially perished in the storm: XIV and SVXY.

In markets, volatility clusters; it does not mean revert. This distinction matters, but is often mistaken. Mean reversion implies that an average level of volatility exists and exerts some degree of gravity when conditions move too far from the mean. In contrast, as Benoit Mandelbrot first observed, with clustering, “large changes tend to be followed by large changes, of either sign, and small changes tend to be followed by small changes.”[1]

Many pundits have called this upswing in volatility a natural reversion following 2017s exceedingly low levels of volatility. This lends to the belief that the volatility we are now seeing is only temporary. Instead, when thinking of this as a cluster, what we have witnessed since February of 2018 is what should be called “regime change” and the expectation from here on out should be that volatility will be persistently higher than what we have experienced over the past few years.

It’s challenging to isolate a specific cause for it, but in Q1 we believe that markets have grappled with a new reality that will remain, hence a regime change in volatility. Flash back to five years ago and the foremost risk in financial markets was the prospect of deflation. Central banks around the globe were either pursuing or commencing aggressive actions to ensure that deflation would not materialize. Just one year ago the risks were far more balanced—some argued that deflation might be the foremost risk and consequently, central banks should continue a dovish stance towards interest rates, while others argued inflation was the foremost risk and a “normalization” of interest rates was thus warranted. Today it is clear in the United States that a normalization is underway and as such, the interest rate curve took a mostly parallel jump higher (there was a degree of flattening between the two and ten year maturities but it was pretty modest during the quarter-as depicted by the 2nd chart below).

The Treasury curve:

The slope between the 2s and 10s:

Why is this happening? Look no further than core CPI. 2017 was “goldilocks” in markets for a reason:

There is good news and bad news out of all this and the two are interrelated. The bad news first: from here on out, the Federal Reserve Bank will continue raising interest rates as long as inflation is rising. Ultimately, the Fed will put the economy into a recession when inflation gets too hot for their tastes. The good news is that the Fed has far more tools and flexibility to deal with inflation today than it does deflation. A Fed-induced recession will be far more like the typical post-WWII recession than what we experienced in Financial Crisis. While the risk of inflation is real, its imminence and extent remain rather modest compared to historical norms. A recession is never fun for stocks, but with valuations comfortably within reasonable bounds the damage should be far more modest and the recovery swifter than the past recession. We do not know when this will take place, but we do know it is an eventuality that must be considered and find it helpful to think about how it should play out in advance. Given this outlook, we used the volatility to purchase two new positions, which we explain in more detail below.

The Power of Pricing Power:

During the quarter, Disney announced it was raising single-day prices at its US parks by up to 9 percent.[2] With inflation reemerging as a risk for markets, companies that can raise prices faster than the rate of inflation become increasingly valuable in two respects: first is the obvious, as these companies can expand margins and make more money; second, the risk to these companies from an inflationary environment is far less, thus warranting a better relative cost of capital. One way to think about relative cost of capital is the P/E ratio of one company verse another. The company with a higher (lower) P/E ratio enjoys a lower (higher) cost of capital. Given what we have laid out above and Disney’s recent price action at theme parks, you might suspect that the relative cost of capital at Disney has been driven lower; however, as the following chart demonstrates, the opposite has been the case.

The blue line on the chart is the Disney forward P/E divided by the S&P 500 forward P/E going back to 1990. Historically, Disney has traded with a decent premium to the S&P. Since the crisis that premium has been less, but still existent. In the last two years, Disney has come to trade at a discount.

Our point above about the Parks is not exactly representative of the entire company, though it remains noteworthy. Parks and Resorts collectively account for one third of the company’s revenue base and EBITDA and one forth of operating profit. This is up nicely from 31% and 21% respective just five years ago. The big loser and focus of much attention here has been ESPN, which was over one forth of operating income five years but now represents about 15% of the total. ESPN is but one part of a broader narrative whereby people fear the end of the broadcast business as we know it and the colossal changes in the media industry.

We think these fears are both overblown and already discounted into Disney’s share prices. Plus, we think Disney with its lush trove of high quality franchises and content makes the company uniquely positioned to thrive in the push to a direct-to-consumer (DTC) relationship. DTC by nature is higher margin and less volatile than the traditional media business for how the relationship works via monthly recurring revenue subscriptions and by cutting out the middle-man in distribution. Some fear that Disney has been too slow and/or acting with too much sensitivity towards distribution. While we do think Disney should have moved faster, we do not think the window has shut by any means and the company’s complete acquisition of MLBAM was smart and necessary for creating a scalable and enduring platform.

The penultimate read on Disney’s opportunity in DTC is Matthew Ball’s two-part piece on “Disney as a Service.” We encourage you all to check it out.[3] [4] The following image that Ball included in his piece is incredibly important. It is an “iteration of Walter Elias Disney’s corporate vision…codified back in 1957:”

This image demonstrates the flywheel that Walt Disney created within his company. As Ball explains, “Only two years after the company’s first theme park opened, Walt detailed an expansive vision for Disney – one where every segment of the business worked in concert. They would develop shared IP, foster shared creative talent and use shared managerial, promotional and financial infrastructure to tell stories that would define generations.”[5] This opportunity is even greater when a direct-to-consumer relationship is forged. Key to our bullishness on Disney is how deeply embedded these values are in the company’s culture. In fact, these values were essentially born with the company itself. It is a big part of why the recent acquisitions of Marvel Entertainment and Lucasfilm have been such immense successes for the bottom line. While many perceive Disney’s push to DTC as a wholesale strategy shift, it is more realistically a natural extension of an enduring corporate culture. We do not think “any” media company can pull off this transition and to that end, the uniqueness of Disney ultimately results in the company itself being a scarce asset that the market will reward in the long-run.

Data, “the new oil” hits an oil slick:

While news of Facebook’s shoddy data practices broke with respect to Cambridge Analytica and the 2016 election, we were patting ourselves on the back for having bought an under-the-radar data company playing an integral role in the online marketing ecosystem. Almost immediately after purchase, our shares in Acxiom were trading higher, until the last day of the quarter when Facebook decided to cut off Acxiom’s data broker as a third-party data provider on Facebook. The stock plummeted and at its worst was down over 40% from the prior week. Nonetheless, our conviction in Acxiom remained steadfast amidst the panic and we have since purchased more shares and dropped our average price accordingly.

This entire situation is one of the more unique and exciting ones we have seen in recent times. Before the Facebook news hit, Acxiom was lifting off on the heels of its intent to sell its legacy business—Acxiom Marketing Solutions—leaving its Connectivity segment (LiveRamp) as a standalone, pureplay growth company in an intriguing industry. Acxiom will surely lose some business from the Facebook change, but the vast majority of the value in this investment comes from the LiveRamp segment—a largely hidden jewel within the broader company. Acxiom acquired LiveRamp in 2014 and has scaled the business rapidly since.

LiveRamp is what is known as a data onboarder. Data onboarders help companies store, manage and use their data in constructive ways. A 2014 FTC report on data brokers defined “onboarding” as follows:

“Onboarding” refers to a process whereby a data broker adds offline data into a cookie (the process of onboarding offline data) to enable advertisers to target consumers virtually anywhere on the Internet. It allows advertisers to use consumers’ offline activities to determine what advertisements to serve them on the Internet.

Onboarding clients either (1) provide data about their customers to a data broker to facilitate the process of finding those consumers on the Internet to deliver targeted advertisements; or (2) use a data broker to identify an audience of consumers who are likely to share particular characteristics and find those consumers on the Internet to deliver advertisements. Three of the data brokers offer an onboarding product.

Onboarding typically includes three steps: (i) segmentation; (ii) matching; and (iii) targeting.[6]

There are elements of this definition that are not entirely accurate and nuances to how onboarding is deployed, but LiveRamp is essentially the only onboarder with robust offline to online capabilities. This being central to the FTC’s definition is demonstrative of how dominant LiveRamp is in the space. Importantly, it’s the kind of industry where network effects are key determinants of customer stickiness and growing value. The more data an onboarder has, the better its match rate will be (LiveRamp’s are the best) and the more use-cases that can be deployed on the platform. LiveRamp boasts a market share over 2x the next largest competitor and importantly, the key competitors including Oracle’s OnRamp (purchased in the $1.2b Datalogix acquisition in 2015) and Neustar are both heavily reliant on LiveRamp as key customers. Essentially, LiveRamp competitors cannot compete without access to LiveRamp itself and the role of competition has been relegated to either white labeling LIveRamp’s pipes or serving specific niches with unique, but not scalable value propositions.

LiveRamp makes money by charging its users subscription fees and tiered pricing depending on use. The majority of revenue comes from usage fees, and as such, the more use-cases LiveRamp can develop, the more it can grow its relevance and revenue base from customers new and old alike. Once the disposition of AMS is complete, LiveRamp will enjoy enhanced financial flexibility to deploy in developing and acquiring tuck-in applications that can expand the capabilities users will have on the platform. A recent example of such a move is the company’s acquisition of Pacific Data Partners to grow the B2B use-cases for LiveRamp.[7]

LiveRamp has been nurtured under smart, strong leadership. Scott Howe, Acxiom’s CEO, came to the company in 2011 from Microsoft, where he was the company’s top ad executive in charge of advertiser and publishing solutions, including Bing. As Howe explains, “The Axiom I walked into four years ago was really a legacy direct-mail database company but had developed some really great assets that could be extended to other channels and can be repurposed for the entire industry, and that’s the transition we’ve been making over time.”[8] Howe’s CFO, Warren Jenson, was an early CFO at Amazon where he is credited with helping lead the company to profitability in the wake of the dot com bust. The management team has been focused and determined in driving shareholder value and has held on to a material equity position in order to position for the upside they ultimately intend to achieve.

While we have adjusted our expectations for the full brunt of the Facebook hit, we think there are very real mitigants to this loss. Companies like General Motors which advertise on Facebook by nature rely on third party data—dealers sell the cars, not GM, so as such, GM needs to stitch together a profile of its own end customers. In the past, Facebook enabled Acxiom’s data to be sold directly on the platform.That will stop in the second half of this calendar year; however,this revenue can be replaced in the following process, by way of example:

  • GM can now buy this data directly from Acxiom.
  • GM can then create its own custom audiences, in its own files
  • GM can upload those newly created customer audiences as “1st party data” for Facebook’s purposes
  • Advertisements can be targeted exactly as they had been on Facebook before

LiveRamp’s revenue run-rate has grown from $16m annualized in Q1 of 2015 (shortly after Acxiom acquired it) to $224m annualized in Q3 of 2018.  The average customer spends over $1.7m per year on the platform and revenue retention is at 110%. At the shares’ worst price on Friday, March 30th, we think LiveRamp itself was worth more than the entire company even though the legacy company will do around $130m in EBIT after accounting for the Facebook hit.

If the AMS business fetches 4-5x EBITDA, the company will get between $1 and $1.4 billion in proceeds. Despite the Facebook news, the Acxiom remains intent on selling its AMS business and focusing purely on LiveRamp. Should a sale not materialize (though the company sounds confident it will) they can consider a spinoff instead. One way or another, LiveRamp will come to be independent in the near future. It’s large enough, self-sustainable on its own cash flow generation, and poised to benefit from strategic flexibility and customer relationships that were limited by its corporate parenthood. Assuming AMS fetches the low-end of our expected proceed range, that leaves half of the company’s value to be accounted for by a $224m run-rate business growing at rates upwards of 40%, likely to grow in the mid-30s for the upcoming year, with the potential to reaccelerate growth with the strategic flexibility afforded by being a standalone pure-play. Our bear case on a sum of the parts is that Acxiom is worth $27 per share, base case is $40 per share and bull case is $58 per share. Looking out further, standalone LiveRamp has the potential to capture a large and growing total addressable market and will very likely catch the eye of the well-capitalized behemoths who facilitate online advertising with their software solutions. It’s only a matter of time before this Facebook news is far in the rearview mirror.

Thank you for your trust and confidence, and for selecting us to be your advisor of choice.  Please call us directly to discuss this commentary in more detail – we are always happy to address any specific questions you may have.  You can reach Jason or Elliot directly at 516-665-1945.  Alternatively, we’ve included our direct dial numbers with our names, below.

Warm personal regards,

Jason Gilbert, CPA/PFS, CFF, CGMA
Managing Director
O: (516) 665-1940
M: (917) 536-3066

Elliot Turner, CFA
Managing Director
O: (516) 665-1942
M: (516) 729-5174

[1]Mandelbrot, Benoit. “The Variation of Some Other Speculative Prices”









Past performance is not necessarily indicative of future results.  The views expressed above are those of RGA Investment Advisors LLC (RGA).  These views are subject to change at any time based on market and other conditions, and RGA disclaims any responsibility to update such views.  Past performance is no guarantee of future results. No forecasts can be guaranteed. These views may not be relied upon as investment advice.   The investment process may change over time. The characteristics set forth above are intended as a general illustration of some of the criteria the team considers in selecting securities for the portfolio. Not all investments meet such criteria.  In the event that a recommendation for the purchase or sale of any security is presented herein, RGA shall furnish to any person upon request a tabular presentation of:
(i) The total number of shares or other units of the security held by RGA or its investment adviser representatives for its own account or for the account of officers, directors, trustees, partners or affiliates of RGA or for discretionary accounts of RGA or its investment adviser representatives, as maintained for clients. (ii) The price or price range at which the securities listed.

Q4’2017 Year-end Review and 2018 Preview

2017 was a banner year for US equity markets with the S&P returning 21.83% and the technology sector leading the way. Markets rose steadily through the year, which stands in stark contrast to the four distinct periods experienced during 2016. First and foremost, it is essential to emphasize that years such as this are not repeatable and in effect are the result of pulling returns forward from future years. We explained this reality in our 2013 year-end commentary in a way that is worth repeating:

While we did not see the economy’s traditional metrics of success reach “normalized” levels in 2013, it has been clearly positioned for the long awaited, but elusive escape-speed breakout from the Great Recession.  In many respects, the stock market has now surpassed the real economy in its estimation of improvement.  Given that reality, it is quite possible, almost probable that 2014 will be a better year for the economy than it will be for the stock market.

One reality we all must accept is how great years in the stock market tend to borrow from future returns, rather than enhance them.  Heading into 2013, one could have reasonably expected to earn an annualized return of 6.67%  – on par with the market’s average since 1925 – over the course of the next decade.  The S&P 500 started 2013 at 1426.  Factoring in a 6.67% annualized return the index would be at 2720 at the end of ten years.  Assuming that despite last year’s near 30% return, this 2720 ten year S&P index target does not change, the expected return over the next nine years would be 4.39% annually, instead of 6.67%.[1]

In many respects, 2017 is the year that should have happened in 2016 for stock markets. The path to the underlying drivers of economic strength were all paved in advance of this actuality. That said, there were some unique stories that accompanied the market’s ascent along the way which we will cover below. It would be impossible to talk about 2017 in financial markets without broaching the topics of Bitcoin and tax cuts.

Throughout the year, Bitcoin and its fellow cryptocurrencies performed exceptionally well; however, in the fall, the crypto world seemingly captured the imagination of everyone. This peak in interest corresponded with pronounced strength into Thanksgiving and the imminent launch of Bitcoin futures contracts at the CME. The innovation behind Bitcoin, specifically the blockchain, is fascinating and holds tremendous potential; however, as often happens in markets, what started as a story backed by a solid fundamental thesis has warped into a speculative frenzy the likes of which we have not seen in nearly two decades.

The passage of tax reform towards the end of the year added the final bit of gas to propel markets higher. More interestingly, tax reform helped inject a greater degree of dispersion into financial markets than we have seen since before the crisis period. We think tax reform as a topic has been thoroughly picked over by the mainstream media; however, we would like to add a few incremental points to the conversation that often end up omitted:

  1. Think of industries and the playing field within industries on a sliding scale of competition from perfect competition to monopoly. The closer to perfect competition a given industry is, the less impact tax reform will have on the earnings power of the various businesses. It is only for those businesses who face a smaller degree of competition that tax reform will make a meaningful long-term impact on earnings power and thus valuation.
  2. One exception to the above is the situation of legacy retail competing with Amazon (NASDAQ: AMZN). While the prevailing forces are strong and hard to mitigate (i.e. the secular move of brick-and-mortar retail sales to ecommerce), tax reform is uniquely supportive of the legacy retail sector. Think of it as follows: Amazon manages its income statement to invest all profits in growing its business and as such, realizes little to no reported net income. With almost no net income, Amazon is not a Federal corporate income tax payer. In contrast, despite retail’s woes, the retailers generate considerable earnings and pay a higher tax rate than the average S&P constituent by virtue of having predominantly domestic businesses. Simply put: Amazon gets no benefit from tax reform, while legacy retailers get considerable benefit. This does not mean retailers can change the inevitability of the tide flowing from brick-and-mortar to online; however, it does mean that retailers will have proportionately more resources to dedicate towards competing with Amazon.

Bitcoin and the Superexponential Price-Mediated Feedback Loop

The following is a chart of Bitcoin on a logarithmic scale from the beginning of 2015 through the end of 2017:


The chart cleanly moves in the up and to the right direction, but there are two additional observations worth making:

  • The Y-axis—it does not move up 1 unit at a time, but instead jumps according to a scale factor a little shy of 2. This is a power law in action, just like the Richter Scale for earthquakes whereby each 1 unit up the scale has 10 times the amplitude of the 1 before it (ie an earthquake that registers 7 on the Richter Scale is 10 times as strong as an earthquake that registers 6).
  • The moving averages—look at the magenta, green and yellow lines following the price higher. These lines have a parabolic arch and are curved instead of straight. A straight line from bottom left to top right on a log chart shows something increasing at an exponential rate. A curved, upward-sloping line on a log chart shows something growing at a superexponential rate.

Very rarely in financial markets do we see something maintain a superexponential arch for so long. These phenomena are rare for a reason and deserve attention when they happen. In addition to being rare, they are also unsustainable. Stated simply and bluntly: this is exactly what bubbles look like. In 2017 one of the best books we read was Scale: The Universal Laws of Growth, Innovation, Sustainability, and the Pace of Life in Organisms, Cities, Economies and Companies by Geoffrey West. West is a theoretical physicist and professor at the Santa Fe Institute who has studied scale and growth in great depth. West’s explanation of the consequences of superexponential growth are incredibly relevant to Bitcoin:

[Superexponential] growth behavior is clearly unsustainable because it requires an unlimited, ever-increasing, and eventually infinite supply of energy and resources at some finite time in the future in order to maintain it. Left unchecked, the theory predicts that it triggers a transition to a phase that leads to stagnation and eventual collapse…

…there has to be a transition from one phase of the system to another having very different characteristics, analogous to the way the condensation of steam to water which subsequently freezes to ice epitomizes transitions between different phases of the same system, each having quite different physical properties….Unfortunately, for cities and socioeconomic systems the phase transition stimulated by the finite time singularity is from superexponential growth to stagnation and collapse, and this could lead to potentially devastating consequences.

….A major innovation effectively resets the clock by changing the conditions under which the system has been operating and growth occurring. Thus, to avoid collapse a new innovation must be initiated that resets the clock, allowing growth to continue and the impending singularity to be avoided.

….to sustain open-ended growth in light of resource limitation requires continuous cycles of paradigm-shifting innovations.[2]

There is a lot to unpack in the preceding excerpt. The most obvious point that carries through the entire section is the unsustainability of superexponential growth. The subtle point that we would like to expound on is the energy needed to drive Bitcoin. Bitcoin requires two essential forms of energy to move upwards:

  1. It needs net inflows of money.
  2. It requires increasing amounts of power to maintain the network.

We won’t go in detail on the power needs necessary to fuel Bitcoin, because the actual energy is the area most ripe for an innovative phase change that can engineer sustainability. This is not the case on the money side in any reasonable timeframe. (If you are interested in reading on the power requirements of the network, Wired offered a great primer in December titled “How much energy does bitcoin mining really use? It’s complicated.”)[3] The higher Bitcoin goes on this superexponential trajectory, the more money it requires to keep driving it. Price moves based on simple supply and demand. With an essentially finite supply, each new dollar coming in is used to cash out an existing holder. In other words, the newest investors in Bitcoin in effect are cashing out the earliest holders of Bitcoin who are intent on holding unless and until it hits a price worth selling at. If this sounds a little like a ponzi scheme it’s because there are very real parallels, though that is likewise the case with any bubble. This is the essence of speculative markets where the intent of the buyer is not to  value the fundamentals underlying the asset, but instead to participate in the Keynesian Beauty Contest in search of a “greater fool” who will eventually pay a higher price.

Bitcoin today is in what Soros calls a price-mediated feedback loop. The newest buyers of Bitcoin are not people with a fundamental understanding of the technology or even a basic theory on what it might be worth. Speculators are merely buying because the price has gone up a lot with the hope that it continues to go up a lot. Elliot went into great detail on this in a blog post titled “The Rise and (Inevitable) Fall of Bitcoin.” [4] We covered a similar phenomenon, though far less extreme in our April 2015 Commentary on ETF Fund Flows.[5] Specifically, we looked at the relationship between fund flows and price in a dozen different sector-based ETFs and noticed something unique and odd in IBB-the biotech ETF. Here was the chart with the accompanying explanation:


Of the dozen plus ETFs we looked at, this is the only one where price and flow moved upwards in tandem. This is exactly what a price-mediated feedback looks like. The rise in price brings more flow, which drives price higher, thus attracting yet more flow. This is also the explanation for how momentum works and what it looks like in action. While in the past we asserted biotech was not in a bubble based on valuation metrics and the fundamental outlook, this right here is a very concerning development. The longer this persists, the more troubling it will be. While price and flow can keep driving each other higher, when one breaks down, the other too will follow. One of the foremost points we learned from Soros is that when price-mediated feedback loops break, they do not simply find a new equilibrium at the price’s present plateau. Rather, the feedback loop reverses and works in the opposite direction. Note that the move up and down in price and flow are both positive feedback loops—the disequilibrium-seeking forces in markets. As such, these relationships are inherently unstable [emphasis added].[6]

Sure enough, 2 months after the date of publication on our ETF fund flow report (5/11/2015), the biotech sector ETF peaked nearly 13% higher and promptly shed 40% over the next seven months. We are now nearing two years since the bottom of the reflex move lower in IBB and yet the sector has recouped only half of its losses from peak to trough.

Considering the Bitcoin move has been many times more extreme than biotech, we expect a more extreme flush out that will similarly take an extended period to recover from. In fact, the Bitcoin situation is far more akin go the Dot Com bubble in 1999 and to emphasize the point, consider that Priceline first neared $1,000 per share in 1999 and took a whole 14 years to finally get back to those levels. At its worst, Priceline had declined over 99% from peak to trough.


Priceline may not be a perfect example for what Bitcoin will do from here, but it does offer a good lesson. Cryptotechnology holds tremendous potential, but today the values are premised on speculative fervor rather than fundamental valuation. Speculative fervor is fleeting and when it ends, the price of the asset needs to find a floor and build a foundation where fundamentals can support the valuation. From that floor, to once again begin an ascent, the promise of the speculative fervor needs to be translated into tangible worth.

We are sure that in laying out our perspective in cryptocurrencies, the nuance will be misinterpreted by bulls and bears alike. Let us all remember one key fundamental fact here: there is little actual utility for Bitcoin today aside for international remittances. The initial promise was at least in part premised on Bitcoin as a transaction protocol and the combination of volatility and painfully slow clearing times make that a virtually moot point barring some phase change in the consensus of the Bitcoin community. To that end, the potential fundamental value is even lower today than when the transactional element was core to the thesis. It takes real utility and real value to develop a fundamental valuation to underpin the price of such an asset, not merely the potential for such development. We will continue to learn about the technology underlying Bitcoin and blockchain and explore investment opportunities and their respective merits. This pursuit will be taken with immense patience and caution. In anything with incredible long-term promise, there is far more risk in being too early, during a clearly euphoric phase, than in being a little late. Fear of missing out is not a fear that we at RGA actually feel anywhere in our bones.

Long end of the yield curve debt ceiling move unwind?

The following is a graph of the US Treasury yield curve today vs January 1, 2011:


Notice that the yield curve is far flatter today than it was in 2011, despite a strong economy and inflation gaining in velocity. This is counter to what one would expect, though there is a sensible narrative: a flattening yield curve today theoretically is an expression that bond investors expect the Federal Reserve Bank will raise interest rates and put the economy into a recession before inflation gets ahead of itself. Many are wondering whether this is the case. We think there is a different view worth exploring. Let’s look at a proxy for Treasuries that will show another angle to underpin our narrative on what we think happened and how this might play out from here:


The circled area above shows exactly when rates on the long end of the yield curve came down dramatically. Two external shocks happened within the circled space alongside a change in the Federal Reserve’s quantitative easing strategy:

  1. The US debt ceiling crisis
  2. The European sovereign debt crisis entered an especially acute phase.
  3. The Federal Reserve Bank hinted at and then commenced operation Twist

One of the big changes that occurred is European financial institutions, especially in the worst performing economies, started buying more Treasuries to balance their risk profile. Ireland is especially telling. In the beginning of 2011, Ireland was the 16th largest foreign holder of Treasuries. As of today, Ireland is the 4th largest holder, registering a 643% increase in their Treasury holdings.

Over the last six years, in aggregate, the Fed was by far the biggest purchaser of Treasuries as part of its quantitative easing (QE) policy. QE2 ended in the early summer of 2011. The Fed had been purchasing relatively short-dated Treasuries in the beginning of QE, because that is where the majority of issuance and thus liquidity is situated; however, in the Summer of 2011, the Fed introduced and then shortly after commenced “Operation Twist”—a name borrowed from a 1961 Fed initiative applying the same strategy. The stated intent of “Operation Twist” was to pull lower long-term interest rates. Up until 2011, the market’s implied interest rate expectation was decently higher than the Fed’s own expectations. To pull longer-term interest rates lower, the Fed announced it would use the proceeds from maturing short-term Treasuries (3 years or less) owned via QE1 and QE2 into longer-dated Treasuries (6-30 years). Today we can see clearly that the strategy accomplished its objective though it now looks like the effect has outlasted the Fed’s intent.

Today, with Europe healing and with the market now behind the Fed in expecting interest rate normalization, we think the time is right for the 2011 move in the long-end to entirely reverse. Europe’s sovereign yields have normalized and the Fed today is slowly unwinding their book of Treasuries by letting maturing bond proceeds revert to cash. This would be welcome news as equity valuations are near the upper-end of the fair value range and some yield from bonds would be a helpful addition to our portfolios should we decide during the year to move some of our exposure out of equities in anticipation of a more favorable return setup long-term.

The Reemergence of Dispersion

During 2017, while markets rose, dispersion collapsed. The following chart from KKR’s “Outlook for 2018” is insightful:[7]


Dispersion measures the degree to which stocks correlate with one another. The higher the dispersion, the more uniform the moves across markets. Beginning with the financial crisis and through 2015 correlation between stocks stayed above its long-term average (the horizontal red line in the chart above). At the end of 2016 dispersion started to emerge and then accelerated as 2017 went on. At the end of the year, there was as large a degree of dispersion between stock as there had been over the last 30 years. This is a great development for active investing.

Financial crises tend to inject volatility into markets during which time “all correlations go to 1.” In other words, during financial crises, everything tends to move together. It is notable how long this condition persisted even past what many would regard as the “end” of the Financial Crisis itself. We believe one of the primary forces behind this extended period of little dispersion has been the move towards increased indexation. Many forces have driven indexation, though with this too we can tie to the Financial Crisis itself: as markets recovered from the Crisis, many advisors and people alike concluded that indexation in buy-and-hold fashion is superior to an attempt to actively pick winners and losers. We explored this topic in greater detail in our October 2013 commentary.[8] For our purposes here, the most important point is related to one we made above about Bitcoin: financial markets (and by derivative, financial market participants) tend to take good ideas way beyond their logical extreme.

There were three distinct catalysts that helped dispersion reemerge:

  1. The Presidential election in 2016
  2. The passage of tax reform in 2017
  3. The rise in short-term rates

The first two of these events had traders betting on companies and sectors whose prospects were specifically changed by either actual or expected policy changes. Some of the most distinct winners in both were sectors whose primary business are domestic (in contrast to multinationals) and whose businesses are subject to regulatory oversight. The rise in short-term rates were a big boon to the financial sector and a source of pain for the sectors whose allocations over the prior decade had been driven by yield. Specifically, the worst sectors last year (from worst to best) were the utilities, the REITs and the staples. Each of these were treated as bond-like proxies and with the rise in rates and the expectation for more hikes yet to come, they were on the receiving end of considerable selling pressure.

The issue of dispersion raises an important question: is dispersion increasing a side-effect of markets rising? Stated another way, will dispersion remain high in the next market decline? We are inclined to believe that from here on out, dispersion should be more “normal” than had prevailed in the Financial Crisis and its wake. In the 2000-02 bear market there were abundant opportunities to make money being invested on the long side, so long as one stayed clear of the carnage in technology. In contrast, in 2007-09 everything collapsed together. It is hard to sit in our seat today and say exactly which areas of the market would best withstand the next bear, for there are yet to be obvious areas of overinvestment like the technology sector in 2000 and the financials in 2007 that will lead the way down; however, we have confidence that the nature of the next recession (and let’s not forget, a recession is an inevitability eventually) will be different than the last. To reiterate: the 2007-09 bear market was an actual liquidity crisis where deflation was the risk that imperiled the economy, whereas the next recession will be a normal Fed induced recession when inflation gets too high and the Fed thus raises rates to cool things off. By virtue of this difference in nature, a Fed induced recession will hit sectors and underlying economic fundamentals in a very different way than was evidenced in 2007-09.

If dispersion remains at these high levels, expect our portfolio activity to increase considerably from 2017. While 2016 was one of the most active years ever for us, 2017 was our single least active year yet. Amidst high market valuations, dispersion is the primary force behind uncovering new opportunities. We need not necessarily wait for broader market declines to make moves, but instead can find fertile hunting grounds in either out-of-favor sectors or those sectors on the wrong side of flows and rotation trades.

Everything is now Tech, Tech is now everything:

In the past, we have bemoaned the use of labeling and categorization with respect to industry and sector classification. While a necessary evil, the rise of indexation has led to pronounced mispricings where the categorization does not fit the business reality. These problems have been most acute with the existence of the technology sector. We most recently made this point in our Q2 2017 commentary by asserting that “Investors are realizing that companies either use technology to their advantage or risk obsolescence….The natural endpoint is where people stop thinking of companies as ‘technology’ or ‘retail’ and think about how a particular company is developing technology to entrench their business in the new commercial landscape.”[9] We maintain heavy “technology” exposure as a result of this worldview, yet the overlap with the label has at times led us to experience a higher degree of volatility due to certain exposure and factor unwinds.

In November 2017, the S&P Dow Jones Indices and MSCI announced that they will revise the industry categorizations pertaining to technology companies. The “Telecommunication Services Sector” was relabeled the “Communication Services” and will now “include[s] the existing telecommunication companies, as well as companies selected from the Consumer Discretionary Sector currently classified under the Media Industry Group and the Internet & Direct Marketing Retail Sub-Industry, along with select companies currently classified in the Information Technology Sector.” Further, “online marketplaces…regardless of whether they hold inventory” were moved into the “Internet & Direct Marketing Retail Sub-Industry.”

In effect, the Global Industry Classification Standard (GICS) has moved closer to acknowledging that we live in a world where technology is not a standalone sector, but rather a crucial piece of infrastructure enabling and impacting each and every facet of the economy. We welcome a world where investors think more critically about the underlying drivers of a business separate and apart from the adoption of “technology” generally speaking. For example, while Alphabet (NASDAQ: GOOG) is regarded by most as a technology company, as the company achieves immense scale, its earnings will increasingly be driven by the cyclicality of the advertising industry, Expedia (NASDAQ: EXPE) and Priceline (NASDAQ: PCLN) already are in position where their earnings are subject to the fluctuations in global travel and sentiment and PayPal’s earnings will eventually reflect the flow of payments. These companies maintains considerable runway for secular growth with the share of business moving from offline to online in each respective domain; however, the reality cannot be escaped in analyzing these businesses that they are indeed beholden to their end source of demand rather than some abstraction called technology.

Thematic trade in the financial sector—low valuations, strong fundamental backdrop, yield sensitivity, business cycle, easing regulatory environment, capital cycle in full swing!

We buy each position based on our bottoms-up analysis of a given business. We want to know that a company has a strong management team, excellent or improving unit-economics, reasonable growth prospects embedded in the market’s expectations and a fair price. One of the most useful screening criterion for us to find new investments is the development of a thematic thesis on a given sector or industry. While any given theme is of secondary import in actually allocating to a security, our use of themes as screening criterion are a) thoroughly fleshed out with data and analysis, and b) an explanatory force behind why our positioning tends to cluster around certain sectors. You can think of our overexposure to technology in this way—we have a prevailing thesis that technology is not actually a sector, but rather a collection of companies using technology to do business in a better way and we sought out companies within technology where that distinction was most acute between the price of the stock and our estimation of fair value for the business.

One other sector where we have maintained excess exposure to is the financials. In the most acute phase of the Financial Crisis there were bankruptcies and a frenzy of M&A activity. While such actions have been few and far between lately, globally, the financial sector is far more concentrated than it has been in the recent past. Setting aside the debate about the merits of such a situation from macroprudential considerations, we think this is incredibly supportive of the long-term profitability in our leading financial institutions. With underwriting standards remaining largely rigorous since the crisis, there is further support to the fundamentals underlying banks today. Despite these forces, financials as a sector continue to trade at nearly a four turn forward P/E discount to the S&P 500—a clear reflection of the residual stigma from the crisis period.

Financials stands to benefit from all three forces behind dispersion referenced above more so than any other sector. First, Financials by-and-large were on the receiving end of increased regulatory scrutiny following the financial crisis and that has reversed with the new presidential administration. Second, financials in the US tend to be predominantly domestic businesses taxed at the all-in U.S. rate. With tax reform, financials as a sector stand to receive nearly the full benefit of the change in statutory tax rate for businesses. Third, financial sector earnings are sensitive to the direction of interest rates. The Federal Funds Target Rate started 2017 at 0.50-0.75% and ended the year at 1.25-1.50%. This is a substantial increase that boosts potential growth tremendously at many banks and insurance companies alike.

We expect each of our financial holdings to benefit from the prevailing environment. The Charles Schwab Corp’s (NASDAQ: SCHW) earnings are the most sensitive to interest rates and will continue to grow well into the double digits. Envestnet Corp’s (NYSE: ENV) (technically an Information Technology, not Financial company) earnings are influenced by asset valuations, therefore rising markets will be a powerful tailwind to revenues, with earnings leverage in the coming year. JP Morgan (NYSE: JPM) stands to benefit from easing capital standards which should result in a better Return on Equity and a higher payout ratio, supporting strong dividend growth. The situation in Europe is perhaps even more advantageous on some of these fronts, where ING Groep (NYSE: ING) should be able to pay a considerable special dividend, expunging the excess capital regulators required during the crisis period and leading to a step-change in ROE. Meanwhile our insurers—Markel Corp (NYSE: MKL) and Exor (BIT: EXO)—will benefit from investing the bond portion of their float in higher-yielding securities, while the equity side benefits from alpha-generating investment management continuing to beat the S&P.

What do we own?

2017 was a strong year, with the vast majority of our holdings ending in the green. Of the three laggards below, only one remains a current position and we have increased our allocation to it considerably during 2017—Walgreens Boots Alliance. We covered Walgreens in our Q3 2017 commentary[10] and on January 12th, Elliot gave a detailed presentation on Walgreens at Best Ideas 2018, Hosted by MOI Global.[11] You can find his slides at the following link for our full thesis. Total returns are indicated based on the stock’s performance during our holding period within 2017 and are denominated in the US dollar.

The Leaders:

IAC/InterActiveCorp (NYSE: IAC) +88.73%

Trupanion (NASDAQ: TRUP) +88.6%

PayPal Holdings, Inc. (NASDAQ: PYPL) +86.52%

The Laggards:

Under Armour, Inc. (NYSE: UA) -24.95%

Walgreens Boots Alliance, Inc. (NASDAQ: WBA) -12.25%

AmerisourceBergen (NYSE: ABC) -9.10%

Best books we read in 2017:

The Structure of Everyday Life: Civilization and Capitalism, 15th-18th Century Volume 1 by Fernand Braudel – A far-reaching study of what daily life would look like for an average person in each region of the world covering diet, work, household, possessions and relationships and their respective evolutions across the covered time period.

Scale: The Universal Laws of Growth, Innovation, Sustainability, and the Pace of Life in Organisms, Cities, Economies and Companies by Geoffrey West – Scale offers several powerful mental models with broad applicability across the social sciences.

The Attention Merchants: The Epic Scramble to Get Inside Our Heads by Tim Wu – A fascinating look at the news, media and entertainment industry over the past century and its quest to capture increasingly large swaths of our attention. The historical templates offered in the book for the rise of nascent industries and the detailed color on the tactics used by information predators to capture our attention or insightful.

Merchants of Grain: The Power and Profits of the Five Giant Companies at the Center of the World’s Food Supply by Dan Morgan – The intersection of history at the industry and global level is a captivating perspective on how cornerstone industries become enmeshed in some of the most important geopolitical events of any epoch.

Adaptive Markets: Financial Evolution at the Speed of Thought by Andrew Lo – Lo offers an entirely new economic paradigm incorporating the efficient market theory, behavioral economies and the collective lessons of the financial crisis. The work is ambitious, but accessible and a worthwhile read for anyone interested in finance and how it impacts society and touches our lives.

Thank you for your trust and confidence, and for selecting us to be your advisor of choice.  Please call us directly to discuss this commentary in more detail – we are always happy to address any specific questions you may have.  You can reach Jason or Elliot directly at 516-665-1945.  Alternatively, we’ve included our direct dial numbers with our names, below.

Warm personal regards,

Jason Gilbert, CPA/PFS, CFF, CGMA
Managing Director
O: (516) 665-1940
M: (917) 536-3066

Elliot Turner, CFA
Managing Director
O: (516) 665-1942
M: (516) 729-5174



[2] West, Geoffrey. Scale: The Universal Laws of Growth, Innovation, Sustainability, and the Pace of Life in Organisms, Cities, Economies and Companies. Pages 414-416.




[6] Ibid






Past performance is not necessarily indicative of future results.  The views expressed above are those of RGA Investment Advisors LLC (RGA).  These views are subject to change at any time based on market and other conditions, and RGA disclaims any responsibility to update such views.  Past performance is no guarantee of future results. No forecasts can be guaranteed. These views may not be relied upon as investment advice.   The investment process may change over time. The characteristics set forth above are intended as a general illustration of some of the criteria the team considers in selecting securities for the portfolio. Not all investments meet such criteria.  In the event that a recommendation for the purchase or sale of any security is presented herein, RGA shall furnish to any person upon request a tabular presentation of:
(i) The total number of shares or other units of the security held by RGA or its investment adviser representatives for its own account or for the account of officers, directors, trustees, partners or affiliates of RGA or for discretionary accounts of RGA or its investment adviser representatives, as maintained for clients. (ii) The price or price range at which the securities listed.