Model Portfolio: Is It Safe?
SEE LAST PAGE OF THIS REPORT Paul Sagawa / Tejas Raut Dessai
FOR IMPORTANT DISCLOSURES 203.901.1633/.901.1634
psagawa@ / email@example.com
December 6, 2018
Portfolio Update: Is It Safe?
From Oct. 3rd to Nov. 23, the S&P fell 10.0%, the tech components of the S&P dropped 15.9% and our model portfolio was off 18.5%. Outside of recessions in 2009, 2001, and 1991, we found 10 other periods where tech issues fell at least 10%. Relative to the S&P500, this tech downturn is the most severe since 2006, and at 64 days, presuming we’ve seen the worst of it, it is the longest slide since 2010. However, several factors lead us to believe tech, and particularly those stocks positioned for the emerging Cloud/AI Era that make up our portfolio, will outperform coming out of this downturn. First, despite delivering 374bp faster earnings growth than the S&P500 average, tech’s P/E premium has fallen from 10x to 3x since the 1Q peak, and PEG ratios are now at rough parity. Second, tech issues typically recover quickly from corrections, returning to pre-decline peaks in 138% of the time that they took to fall, 635bp faster than the broader market. Third, we believe our portfolio is well insulated from the concerns that have weighed on the market – interest rates, trade negotiations, political changes, etc. Finally, growth tech drivers are mostly deflationary – the cloud, e-commerce, digital ads, and robocabs, etc. all reduce costs for customers. This, and their robust balance sheets, will be protection in the case of recession. All this said, our model portfolio held up well since our last update in August, up 34bp vs. the S&P500, and 467bp vs. the tech benchmark. We are removing QCOM in favor of TMUS because we see short term weakness in smartphones and more protracted negotiations with AAPL holding up recovery until 2H19.
- Expect a strong tech recovery. The market correction has been painful for tech. Presuming a Nov. 23 trough, the tech components of the S&P500 fell 15.9%, 587bp more than the broader index. This relative slide is the most severe since 2006 and the longest since 2010. However, we see five reasons to believe in a sharp rebound. 1. Valuations are reasonable relative to its growth. 2. Tech has been quick to recover in previous corrections. 3. Our portfolio choices are well insulated from the current market concerns. 4. Tech growth drivers are deflationary for customers. 5. Tech earnings remain healthy.
- Reasonable Valuations. Tech P/Es jumped to a 10x premium vs. the S&P500 early in the year and held the advantage until October. The slide from a 33x average multiple to 23x narrowed this gap to just 3x. While this is in line with the trading relationship over the previous 7 years, it fails to reflect the 374bp tech advantage in EPS growth, both actual and projected. Thus, even without multiple expansion, one should expect tech to outperform the broader index at a very modest valuation premium.
- Tech Bounces Back. In 10 times since 1990 when the tech index fell at least 10% (outside of recessions), prices have returned to the pre-correction peak in just 138% more trading days than it took to reach their trough. In these periods, tech has outperformed the broader market by 635bp and has usually moved to set a new relative peak within 16 weeks. Recessions have more difficult recoveries, but excepting the 2001 internet bubble, tech fared much better than the market in both 1991 and 2009. Software has been especially resilient.
- Growth Tech is Insulated. Investor fears over possible interest rate hikes and contentious US/China trade relations sparked the current sell off. While these concerns cast doubt on the market, the cloud/AI leaders that make up the core of our model portfolio have limited exposure. Most of the top tech names have substantial net cash balances. Most internet names are blocked in China (AMZN, GOOGL, NFLX, FB, TWTR, etc.) and few SaaS companies have established strong presence there (CRM, NOW, WDAY, ZEN, AYX, ADBE, etc.). Enterprise IT has yet to be included in proposed tariff packages. Only consumer electronics and the semiconductor supply chain appear to face specific threat.
- Selling Lower Costs. The shift to the cloud is highly deflationary – hyperscale datacenters have costs as much as a magnitude below private facilities, enabling dramatic savings for cloud-based solutions. The Cloud/AI era also offers potential revolution in many disparate sectors – self-driving transportation, e-commerce, digital advertising, AI enhanced tele-medicine, “blockchain” automation for financial services, and others – each portending substantial advances in cost and efficacy. Investment toward these ends are unlikely to be stymied by recession. New tech has typically performed unusually well through the course of recessions. The obvious exception is the 2000 internet bubble crash, which featured debt-heavy balance sheets, cash burn, and unprecedented sector valuation premia. The current situation is in no way similar.
- Leaders Are Healthy. In the most recent quarter, the S&P500 tech constituents delivered a 10.6% average upside surprise in earnings, an acceleration from the 6.6% surprise delivered in 2Q18. Sales growth saw a slight slowing in the quarter, but still surprised against expectations by more than 1%. Many higher growth names showed even better results. With analysts reacting to the downturn with more cautious estimates, we believe the potential for future surprises is increased.
- Our model portfolio outperformed its benchmark since August. Despite a 4.6% drop in absolute value, our model portfolio outperformed the tech benchmark by 467bp and even topped the S&P500 performance by 34bp. XLNX and CIEN were stalwarts, up 28.8% and 19.9% respectively, with AYX rising 6.3% as well. These helped offset NVDA (-39.8%), which missed 3Q18 expectations, and NFLX (-18.4%), which didn’t. GOOGL, AMZN, IBM, and QCOM were each down double digits after quarterly disappointments. Weathering this decline, the portfolio is still up 29.2% on the year. We continue to see clear evidence for a generational shift to a Cloud/AI dominated era that will greatly favor companies like our portfolio constituents.
- Removing QCOM, adding TMUS. We continue to be bullish on QCOM’s longer term prospects. Still, the expected catalysts for its rebound – an AAPL resolution and a ramp in 5G device sales – are likely more than 6 months away. Meanwhile, we are concerned for smartphone sales given signs of a global market slowdown and the risk of escalation with China. We will replace it in our model portfolio with TMUS, which posted strong 3Q18 numbers and is poised to gain approval for its synergy-rich merger with S. We will look for an opportunity to add QCOM back as we get closer to its catalysts.
Tech on Sale
For nearly two months, tech stocks were in a freefall, losing 15.9% from their peak on October 3 to an apparent trough just before CRM’s triumphant earnings report and good news on interest rates and trade sparked a post-Thanksgiving rally. Certainly, other sectors of the market were also hit hard, but the FANGs took it hardest, along with AAPL, whose 4QFY18 miss and step back on disclosure knocked nearly $300B from its market cap. This disproportionate whipping of tech is a tradition – of 12 non-recession market downturns since 1990, tech fell more than the market in all of them.
We are expecting a significant tech bounce in tech shares for five main reasons. 1. Reasonable Valuation – The tech components of the S&P500 traded as high as 38x trailing earnings in April, a gap of ~13x vs. the broader index. With tech’s drop to a 23x average P/E, the gap has closed to just 3x. Given expectations for EPS growth 374bp faster than the rest of the market, we believe tech is relatively cheap. 2. Tech Bounces Back – With the obvious exception of the internet bubble recession of 2001, tech has been quicker than the broader market to return to a previous peak after a significant downturn. In 11 previous non-recession corrections since 1990, tech has averaged 58 days from peak to trough, and 102 days to return to that level. This is 17.4% faster than the full index, driving 635bp of outperformance during a recovery period.
3. Growth Tech is Insulated– Projected interest rate hikes and an escalating trade conflict with China (with global economic implications) have been the biggest factors behind the recent market sell-off. As the big tech names had been the engine driving the market ever higher earlier in the year, the change in sentiment hit these stocks particularly hard on the way down. Lackluster results from AAPL, AMZN and GOOGL, along with the ongoing political fumbling at FB contributed to the fall. Still, few top tech names carry enough debt to be concerned about interest rates, nor do their customers typically finance their purchases. Moreover, unlike their device making brethren, web-based businesses, like GOOGL, AMZN, FB, NFLX, and others, have had little success breaking into the China market. Even SaaS companies, like CRM, NOW and others, have small exposure.
4. Selling Lower Costs (along with amazing new capabilities) – The hyperscale datacenters run by cloud-based leaders offer computing and storage costs as much as 80% cheaper than the all-in costs of private facilities, with an even bigger gap vs. the PCs and smartphones running consumer apps. The cloud is not just cheaper computing, it is also the catalyst for AI applications – virtual assistants, autonomous cars, predictive analytic models, etc. – and blockchain-related technologies that can drive better outcomes and lower costs for many jobs. This should insulate the players that are driving this transition from the worst of a recession. 5. Leaders Are Healthy – The top tech companies are churning out cash and carry very little debt. Despite hiccups from AAPL, AMZN and GOOGL, 3Q18 tech earnings were excellent, with the average surprise 1063bp for EPS and 118bp for sales.
Our model portfolio survived the downturn only slightly worse for wear. Since our last update in August, the 15 stocks are down 4.6% on average, 34bp better than the S&P500 and 467bp ahead of the tech components of that index. We are making one change to the portfolio. While we are still confident in QCOM’s long term prospects, the catalysts for a rerating – an AAPL settlement and the ramp of 5G devices – look at least 6 months away. Given pressures on the global smartphone market, we are swapping it for TMUS, expecting approval of its merger with S to deliver real synergies and spark new interest in the stock.
For the first three quarters of the year, tech investors enjoyed a fantastic ride. The tech components of the S&P500, the benchmark we use for our model portfolio performance, were up 17.9% (Exhibit 1). Our picks, which focus on the beneficiaries of the shift to a Cloud/AI era (http://www.ssrllc.com/publication/the-cloudai-era-a-perspective-on-the-next-decade-of-tmt-investing-2/), were up 46.3% – a substantial beat against the benchmark and the overall market, which rose 8.1% (Exhibit 2). However, the lead in to 3Q18 earnings revealed a measure of doubt – Would rising interest rates spoil the party? What about the looming trade conflict with China? Could a Democratic congress reverse the tax cuts and regulatory roll backs that had seemed to stimulate the economy? Were the top tech stocks just too expensive? How much trouble was Facebook in?
As the October tech results came in, the story was much more mixed than it had been in the almost universally triumphant 2Q. There was the good: Netflix and Microsoft beat expectations handily, then raised guidance (Exhibit 3). There was the Bad: Alphabet and Amazon missed on revenues. And there was the ugly: Apple missed and then announced it would no longer provide unit sales numbers for its products, while Nvidia revealed that the cryptocurrency meltdown would take its toll after all. Against a skeptical sentiment, investors sold first and asked questions later (Exhibit 4).
From October 3rd to November 23, the S&P 500 fell 10.1%, driven by a 15.9% drop in the IT-related components, which had risen to 18.5% of the value of the index at the peak. Our model portfolio, which
Exh 1: Broader market performance for first three quarters of 2018
Exh 2: SSR TMT Model Portfolio relative returns for first 3 quarters of 2018
Exh 3: Earnings and Revenue Surprises by Top Tech Companies for 3Q18
Exh 4: Broader market performance QTD amidst wider market selloff
Exh 5: SSR TMT Model Portfolio relative returns QTD amidst wider market selloff
counts a few of those mega cap growth names (e.g. Amazon, Microsoft, Alphabet, Netflix, Nvidia and Salesforce) amongst its constituents, suffered even more, down 18.5% in those 8 weeks (Exhibit 5). Our last update had been in mid-August, so the correction was tempered by strong performance in the prior 6 weeks, but the declines were painful nonetheless.
Strong Salesforce results post-Thanksgiving, along with beats from smaller SaaS names, helped stem the flow, as did a more dovish commentary from the Fed and an apparent softening in the President’s China stance. Since November 23, the tech benchmark has added back 3.5% and our model portfolio 5.7%, so perhaps the worst is behind us. If so, this was still the worst tech downturn since the 2008 recession.
We are inclined to call bottom on the correction, although we are aware that the actions of the current U.S. administration have been difficult to predict. The downturn wiped away 914bp of outperformance vs. the S&P500 and 337bp vs. the tech benchmark, but very strong performance outside of that period has our portfolio up 2921bp for the year in absolute terms, 2905bp vs. the S&P500, and 2507bp vs. the tech elements of the S&P. We believe that there is ample room for further outperformance looking into 2019. As such, we see substantial opportunities in technology stocks based on five main factors:
1. Reasonable Valuations – Since 2010, in the aftermath of the last recession, tech P/Es had run tight to the rest of the market, despite delivering EPS growth well more than average (Exhibit 6). As a result, tech PEG ratios ran well below the S&P500 (Exhibit 7). The major tech run, which began roughly at the beginning of the year, took the average tech P/E to more than 38 times by March, opening a 13 times gap vs. the broader market. Arguably, this was excessive, even though tech companies were posting extraordinary sales and earnings growth in the first half of 2018. This gap began to narrow after 2Q18 earnings season, then nearly closed after 3Q18 results. Now the average tech P/E is 23 times trailing EPS, while the rest of the market trades at 20x. Given that tech earnings are outgrowing non-tech 12.3% to 8.6%, we see the premium as more than justified.
2. Tech Bounces Back – Presuming November 23 as the trough, this downturn was 64 days, the longest correction since the 123-day slide in 2010 (Exhibit 8). Of course, things were more drawn out during the previous decade and the relative declines were more painful as well. Still, after the pain has come gain. Of the 11 non-recession market drops that we analyzed, going back to 1994, tech was faster to rebound to its pre-decline peak in 9 of them, on average recovering in 82.6% of the time needed for the broader market. During the trough to recovery period, tech outperformed the broader market by 635bp (Exhibit 9, 10).
3. Growth Tech is Insulated – Market declines happen for different reasons. In this case, the catalysts seem to be expectations of interest rate hikes, the aggressive US trade stance toward China, weakening emerging economies and the mid-term US elections. While these factors certainly loom over the world economy, their impact on tech issues (particularly the Cloud/AI stocks in our portfolio) is likely to be less severe than the rest of the market. Interest rates? Tech companies tend to have cash far in excess of their debt, and, increasingly, are not selling capital assets that require financing by their customers. China? Most web-based
Exh 6: S&P500 & Tech Components Avg. Trailing PE Trend for last 5 Years
Exh 7: S&P500 & Tech Components Avg. PEG Ratio Trend for last 5 Years
Exh 8: S&P 500 and Tech Components Rebound Stats after major sell-offs
Exh 9: S&P 500 and Tech Components Peak-to-Trough drop during major sell-offs
Exh 10: S&P Tech Components recovered much faster than S&P 500 broader index post major sell-offs
US companies have little or no exposure. This includes portfolio constituents like Amazon, Alphabet, Netflix, Salesforce, Service Now, Zendesk, and Alteryx. Consumer devices are another story, but investors can steer clear of those names. Similarly, emerging economies are a minor opportunity for most enterprise-oriented tech companies and are still small contributors to most consumer web franchises as well. (Netflix and Amazon have more exposure here). Finally, despite tough talk about “big tech” we don’t see enough consensus to make meaningful legislative or antitrust moves against the likes of Amazon or Alphabet. Facebook continues to be on the hotseat for its possible role as an agent for cross-border psych-ops – the pain is more likely to come from employee dissent and customer fatigue than from government actions.
4. Selling Lower Costs – The hottest tech narratives – Enterprise’s move to SaaS and cloud hosting, Digital media and advertising, E-commerce, Autonomous Transportation, Digital banking, Health care automation, etc. – all involve cloud/AI/5G solutions that will reduce costs across the economy (Exhibit 11). This is another way in which we believe growth tech will be insulated from the potential downside of a recession. In the 2008 recession, the broader market took more than 4 years to bounce back to its pre-decline level. Tech was back in less than 3. Obviously, the 2001 Internet Bubble recession was particularly harsh on tech investors, but then again, early 2000 P/Es for tech companies were more than triple the level of non-tech companies. During the Internet Bubble, even seemingly healthy leaders like Cisco were highly dependent on sales to fast growing but heavily indebted customers building networks on spec, whose eventual bankruptcies and cancelled orders cued the 2001 crash. This time, the products are services, transformational capabilities
Exh 11: SSR’s Leading Investment Themes for TMT Companies
offered to consumers and enterprises across all facets of the economy, built on a cloud computing hyperscale architecture with 80% all in cost savings.
5. Leaders are Healthy – Significant net cash balances? Check (Exhibit 12). Strong record of sustainable revenue growth? Check. An obvious path to fat operating profit margins? Check. Structural competitive barriers (e.g. scale, first mover advantage, consumer reach, branding, etc.)? Check. Diverse and healthy customers? Check. We believe the tech business is far healthier than it was during the Dot Com era. When online companies without evidence of revenues were funded for “eyeballs”? When telecom balance sheets overflowed with debt even as they double ordered with their equipment suppliers to stay ahead of the competition? When equipment suppliers were sure that the wave of demand would never end and stepped up production to try to get ahead of it? Things are PROFOUNDLY different than that today.
How’d We Do?
Despite the carnage from October 3 to November 23, our stocks did relatively well since our August 14 update (Exhibit 13). All together we were down 4.58%, which is 34bp better than the S&P 500 and 467bp better than the tech benchmark. Looking squarely at year end, we are up a comfortable 29.2% for 2018 vs. the 0.2% rise in the full market and 4.1% increase in the IT components of the index (Exhibit 14, 15).
Last quarter, we stood pat on our portfolio selections. This was mostly a good decision. Xilinx and Ciena crushed their 3Q18 results and were huge winners, up 28.8% and 19.9% respectively. Alteryx added another 6.3% to be up almost 75% since we added it earlier in the year. Keysight and Accenture were others that
Exh 12: Highlight of Solvency Metrics for SSR TMT Model Portfolio Constituents