There are two primary issues facing the market right now. The first issue is what level of tariffs Trump will impose on China (rollback, 10% on $250bn, 25% on $250bn, or 25% on $500bn) and consequently, what level of earnings the S&P will generate next year. If it’s the full $500bn, I estimate earnings will be $5-$8 lower than otherwise, before taking into account dollar strengthening and a slowing global economy. The second issue is how hawkish or dovish the fed will be in the wake of Powell’s “nowhere near neutral” comment, his shift away from reliance on models that estimate the neutral rate to guide policy, and hawkish commentary from most other fed members (eg. Brainard writing that short term neutral rate is above the long term neutral rate, which is code for we should invert the yield curve). Given that these two issues are defining the market, a 3 by 2 matrix with different trade and fed scenarios represented in each box is a good way to visualize different market outcomes (see below). Short Nasdaq (QQQ) and long SPY is a lower risk way to bet on either a resolution or escalation of trade tensions, and I don’t believe you will lose much in a status quo scenario.
If there is a trade deal, I believe QQQ will underperform in a manner similar to the underperformance in the first half of 2016 when China hard landing fears were averted and investors flowed to low valuation cyclicals/emerging markets that were already pricing in a significant slowdown. Today is an even worse setup for QQQ if a slowdown is averted as the QQQ is at a 22% P/E multiple premium to the S&P 500, versus a 8% premium at the beginning of 2016, while industrials are at 5% discount to the S&P multiple versus a 9% discount in 2016. Another way to say this is that QQQs multiple is 28% higher than XLI today versus 20% higher in 2016. If a trade deal is announced this spread will compress.
Alternatively, if Trump passes 25% tariffs on $500 billion of Chinese goods, its very likely that cyclical will underperform despite low valuations as earnings estimates will come down and bond proxies will outperform as investors seek safety. While not as clear cut, I also believe QQQ will underperform as AAPL will start to price in the potential for a China ban against their products (plus their guidance/commentary on Thursday shows that slower EM growth is already hurting AAPL’s volume growth and a strong dollar means extra pricing is required) and MSFT and AMZN will price in an accelerated slowdown in cloud spending. Even without additional tariffs, Western Digital’s commentary suggests the large hyperscale cloud guys are already seeing a slowdown in spending: “We are experiencing a temporary slowdown in Data Center capital spending, particularly by large cloud service providers after several quarters of growth above the expected long-term exabyte growth rate of 40% for capacity enterprise.” In a callback with WDC they explained that this dynamic really started accelerating in late September so I don’t think this was fully captured in the numbers AMZN and MSFT reported. OLED and TXN also described an accelerated late September and October slowdown. The last point is simply that the market tends to move from one extreme to the other, and with QQQ still near cycle high relative multiples and having outperformed the S&P 500 by 22 percentage points since both 2016 and 2017, its reasonable to expect a period of mean reversion as the macro narrative shifts to something new.
Obviously this chart is far from exact but it’s helped me think about this idea and the market more broadly. Green boxes represent scenarios where I think QQQ underperforms SPY, orange where I think they do about the same, and red where QQQ outperforms SPY.
No Tariff Escalation in January
25% Tariffs on $500 bn
-SPY = ~3100 = 17.5*178 (consensus 2019 eps).
-No recession fears. Consensus doesn’t have to be revised down in the normal pattern.
-SPY outperforms QQQ as cyclicals rally. EEM does best.
-SPY = ~2790 = 16.5*169 (consensus 2019 eps -5%).
-Small multiple expansion from Powell being more dovish but still a lot of earnings uncertainty due to potential for more tariffs.
-QQQ outperforms SPY as dovish fed and pause in trade tensions tends is just an extension of the status quo, which supports the momentum stocks represented in QQQ.
-SPY outperforms QQQ as cyclicals valuation already assumes some slowdown, bond proxies in SPY will outperform, and QQQ will underperform as relative valuation still near highs, full tariffs represent a big shift in the macro narrative, cloud spend slowdown, dollar strength, and potential targeted retaliation should hurt AMZN, MSFT, and AAPL. IP theft issues make people worried about owning tech.
-SPY = ~2900 = 17*171 (consensus 2019 -4%).
-SPY outperforms QQQ as cyclicals have relief rally and tech hurt by higher rates until market worries about recession again.
-SPY = ~2675 = 16.0*167 (consensus 2019 -6%).
-This is what was implied by SPYs price a week ago before Trump had positive trade tweets.
-SPY performs in line with QQQ as QQQ relative multiple still near record, momentum now negative and breaking long term trend lines, money continues to slowly shift to true defensives and don’t have low rates to support high multiples.
-SPY = ~2220 = 14*158
(no growth over 2018 EPS)
-This could lead to a recessions and be worse.
-Similar to above scenario.
-Just want to be outright short in this rather than hedged.
I hold a position with the issuer such as employment, directorship, or consultancy. I and/or others I advise hold a material investment in the issuer's securities.