- Book as small absolute loss/meaningful relative performance gain in long Domestic Large-Cap Healthcare Stocks (XLV), as the quantitative signal no longer supports being long of that exposure despite its historical outperformance in #Quad4.
- Add Short-term U.S. Treasury Notes (SHY) as we think the market may start to price in what may be the Fed’s final rate hike (in December) prior to what could morph into an elongated pause throughout the first half of 2019.
- Cover our short in Corn (CORN) for a decent relative performance gain as the exposure has lost a considerable degree of volatility since we added it back in early-September, which has historically been a bullish leading indicator in our quantitative signaling process.
- Short U.S. Retailers (XRT) whose profit margins appear increasingly at risk amid a developing downturn in domestic consumption growth and cyclical acceleration in wage growth – a catalyst that has historically not translated into accelerating consumer spending, much to the chagrin of the Macro Tourist crowd.
#Quad4 (introduced 9/27/18):
- September CPI Confirms the U.S. Was In #Quad1 In 3Q18E (10/11): The advent of the SEP Headline CPI data (↓ -40bps to 2.3% YoY; now TRENDING lower) confirmed our non-consensus view that reported inflation is past peak with probable downside to ~1.8% by mid-1Q19E. If that forecast is even in the area code of accurate, we can’t help but call out what will be a then 100-plus basis point decline in YoY Headline CPI in just seven months. Market history suggests that outcome is supportive of a 10yr Treasury yield in the mid-to-low 2% range, not one in the low-to-mid 3% range. Importantly, the sharp downtick in inflation in SEP confirmed a -6bps decline from the 2Q18 quarterly average run-rate, which pushed the U.S. economy narrowly into #Quad1 for 3Q18E assuming our nowcast for Real GDP growth – which ticked up to 3.06% YoY/3.53% QoQ SAAR on the dovish inflation print – is ultimately proven prescient.
- September CPI Also Confirms the U.S. Is Headed Into #Quad4 In 4Q18E (10/11): Our comparative base effect model perfectly explained the variation in inflation during 3Q18 so the +42bps step-up in the 2yr comp here in Q4 represents a real challenge for any investor that is calling for YoY Headline CPI to accelerate on a QoQ basis in 4Q18E. Our updated Headline CPI scenario analyses suggests inflation should tick up slightly in OCT (and in DEC too if we use a 2yr comp stack model vs. a 3yr), but it’s all downhill from there in terms of the progression of the YoY prints. The quarterly average is lower in each subsequent quarter from the 2Q18 peak, which is the biggest takeaway regardless. Looking underneath the hood, Energy CPI (↓ -548bps to 4.85% YoY; now TRENDING lower) was among the biggest drivers of the sharp downtick in the headline index in SEP. We’ve been back and forth with so many investors about the impact of $80+/bbl. Brent Crude Oil on our CPI estimates and our response is the exact same each time: “Last price is already factored into our [dovish] outlook.” As we detail on slide 40 in our Q4 Macro Themes presentation, Brent has to do a lot more than average $80/bbl. here in Q4 (and beyond) for Energy CPI to “comp the comps”. Shifting gears a bit, the flatlining of Core CPI from both a sequential and TRENDING perspective in SEP is confirmatory of our view that underlying inflationary pressures have peaked as well – for at least the NTM – as we detailed in the scenario analysis on slide 43 of the aforementioned presentation. Tariffs are another scarecrow that is not perpetuating the inflationary impulse investor consensus expects they should. It is, however, getting those investors who trust our process paid in long/short #Quad4 portfolio construction terms. The dollar’s ascent is a mitigating factor here, which we detail on slide 44 in that same presentation.
- Is #Quad4 Already Priced In? (10/17): We’ve been getting a lot of questions lately on whether or not #Quad4 has already been priced in – specifically due to the fact that we’re not calling for a material downturn in economic growth. What we’ve learned by parsing the individual factors embedded in our GIP Model into their own regimes is that absolute levels of growth and inflation actually do matter – but only when you apply a rate of change overlay. Isolating the U.S. Equity Momentum Style Factor, we see that: 1) The expected value in regime “3” (i.e. growth decelerating from an elevated base rate) in the attached table(s) – although slightly positive – is only half the expected value of regime “2” (i.e. growth accelerating from a depressed base rate, a la 3Q16-3Q17) and less than a fifth of the expected value of regime “4” (i.e. growth accelerating from an elevated base rate, a la 4Q17-3Q18); and 2) The percent positive ratio is decidedly lower at 52% vs. 79% and 89%, respectively. On balance, the latent positioning built up by investors being forced to chase this factor as a function of the persistence of the aforementioned regimesis the real market risk. We’d be sympathetic to any view that argued for a more positive market outcome with respect to this and related factor exposures during this multi-quarter instance of #Quad4 if we did not just accelerate for a record nine quarters in a row. Said the other way around, who is the marginal buyer of Momentum/High Beta/Growth in October 2018 after 2+ years of being forced to chase or capitulate on the short side of these factors? We’d argue there aren’t many – especially after EM/Europe blew up, which likely forced even the most slow-moving int’l capital allocators back into the U.S. growth story during the mid-to-late summer.
#CyclicalPeaks (introduced 9/27/18):
- U.S. Growth = Past Peak (10/16): The advent of yesterday’s disappointing Headline Retail Sales data (↓ -180bps to 4.7% YoY in SEP; now TRENDING lower), flat Retail Sales Control Group data (unchanged at 4.9% YoY in SEP; TRENDING higher), and today’s rock solid Industrial Production data (↑ +28bps to 5.14% YoY in SEP; fastest rate since DEC ’10) rendered our nowcast for 3Q18E Real GDP growth down -1bps to 3.05% YoY and -4bps to 3.49% QoQ SAAR, respectively. The latter figure compares to 3.2% YoY for Bloomberg Consensus and 4.0% for the Atlanta Fed. It’s worth noting that September represented the last easy comp for Industrial Production – currently the third highest weighted factor in our predictive tracking algorithm for U.S. GDP – and that base effects get materially more difficult at nearly every interval for the next year. With that in mind, the “fiscal impulse” better get commensurately larger, at every interval, for domestic manufacturing activity to comp those comps. Looking to our scenario analysis for domestic consumer spending, OCT should see a modest deceleration off of the SEP growth rate, while the NOV print is likely to be a bomb (to the downside) due to the data having to compare against last year’s cycle-peak growth rate of 5.91% YoY. All told, it would be wise to fasten your seatbelts folks, as our math suggests #Quad4 is just getting started.
- CapEx Plunges Right On Time (10/25): Core Capital Goods New Orders slowed -595bps in SEP to 1.86% YoY – the slowest pace since JAN ’17 – on its second consecutive MoM decline. Recall that we’d been highlighting the likelihood that CapEx growth would slow sharply in September against the tax-reform-anticipatory pop seen in SEP ’17, when growth accelerated to cycle-highs of 13.3% YoY and 5.7% MoM, respectively. While CapEx growth has the potential to accelerate rather modestly in OCT, it’s clear from our NTM scenario analysis that a new [lower] progression has been established on both a TRENDING and quarterly average basis going forward. That same model suggests Industrial Production and Retail Sales – the number #3 and #1 weighted factors in our predictive tracking algorithm for U.S. GDP – are the next shoes to drop [off] in OCT and NOV, respectively. Elsewhere in the Durable Goods report, New Orders for Durable Goods decelerated -420bps to 7.86% YoY and, like their CapEx counterparts, are now slowing on a TRENDING basis as of September. Throw in the -20bps deceleration in Wholesale Inventories to 5.1% YoY in the month and it’s clear that what was a strong Q3 ended on fairly weak footing and the ebbing of momentum does not bode well for growth in Q4 amid a step function higher in the comparative base for Real GDP. All told, this is the last of the factor inputs we’ll receive ahead of the release of the advance estimate of 3Q18E GDP tomorrow morning. On that front, we’re settled at 3.03% YoY/3.43% QoQ SAAR, down -2bps and -6bps on the day, respectively.
- U.S. Economic Growth Was Great… Emphasis on “Was” (10/29): I’ll make this quick because the last thing anyone needs on a day like today is another GDP recap note from the sell side. If, however, you’re into that kind of stuff, please refer to the table below for more details. In short, it was a good print, just not as good as the one prior to it in terms of the Headline (i.e. QoQ SAAR) growth rate, which, at 3.46%, came in 3bps higher than our final nowcast of 3.43%. Our YoYestimate of 3.03% was a mere basis point shy of the actual 3.04% print. The key takeaway here is that because our forecasts were proven accurate, the progression implied by our comparative base effect model is largely unchanged from what we were previously anticipating – which is nothing shy of modest, but unrelenting deceleration off the now-cemented cycle peak. Next stop: ~6 months of reported #Quad4 data with the worst of it [in rate-of-change terms] being in the current quarter. Perhaps the market has priced that in already and over-owned “structural growers” are buys now, but I doubt it given that I heard a chorus of similar statements on October 11th too – BEFORE AMZN, GOOG, FB, NFLX all reminded investors that even they are not immune to #TheCycle.
#LongHousing (introduced 9/27/18):
- What Works In #Quad4 If Rates Don’t Come In? (10/18): Historically the Low-Beta/Min-Vol.and High Dividend style factors have held up reasonably well during historical instances of #Quad4, but that’s usually because of the decline in interest rates. Looking to scenarios where rates actually increased during historical instances of #Quad4, there have been 5 of 15 instances where bond yields actually rose during the observation period and bond proxies broadly did better in three of them (e.g. 2Q01, 4Q01, and 2Q09). 4Q05 saw mixed results and 1Q09 saw a final wash-out of equities broadly despite the positive signals emanating from the Treasury yield curve at the time. At the sector level, Utilities seem to have the most downside sensitivity to any bond yield breakout during #Quad4, but that’s seemingly the opposite of what’s been occurring in the MTD – which may be a signal that bond yields don’t have much upside from here.
- New Home Sales │ Awaiting the Washout (10/24): We expect a Quad 4 environment (growth and inflation both decelerating) in both 4Q18 and 1Q19. #Quad4 represents a macro environment typically characterized by falling rates, a more dovish policy lean and outperformance in defensive yield and select interest rate sensitive equities, including Housing. Unprecedented late-cycle fiscal stimulus has cultivated an interesting, somewhat anomalous macro condition set as it elevates the prospects for further acceleration in wage inflation and the potential for households to see improved consumption capacity at the same that both headline growth and inflation are slowing and yields are making lower highs – a prospective setup that would be broadly favorable for housing, particularly against a backdrop of significant underperformance, trough valuation, and a housing cycle which, itself, is still only mid-cycle. You do not want to be long Housing related equities during rate shock periods, period. When the magnitude of short-term changes in rates pushes above 2 Standard Deviations the inverse correlation between rates and housing equities pushes towards 1 and housing investing devolves into a single factor model (i.e. the trade is singularly about rates and broader fundamental considerations get sidelined).
#StrongDollar (introduced 4/3/18 under the title, “#DollarBottoming?”):
- PCE Data Confirms More of the Same (10/26): The state of U.S. economic data has become increasingly characterized by the phrase “past peak” – which is fine for absolutist investors who, inherently prefer to buy high (read: “things are good”) and sell low (read: “things are bad”). It is, however, not fine for those among us who were schooled in the rate-of-change camp – a school of thought that views catalysts like rolling off the cycle-peaks in growth and inflation amid accelerating monetary tightening as a negative catalyst for risk assets. Specifically, Real PCEgrowth decelerated -20bps to 3.0% YoY, Real Disposable Personal Income growth decelerated -10bps to 2.9% YoY, and the Headline PCE Deflator decelerated -20bps to 2.0% YoY. Complicating matters was the +1bps acceleration to 1.97% YoY for Core PCE, a development that should keep the absolutists at the Federal Reserve patting themselves on the back for effectively achieving their inflation mandate, which implies a low likelihood of a near-term pause in their policy normalization drive. Last week was a perfect example of how Mr. Market likes to price in central banks tightening into slowdowns: U.S. Dollar Index ↑ +0.6%, 10yr UST Yield ↓ -11bps, Junk Bond Yields ↑ +12bps, and High Beta Stocks ↓ -7.3%.
#ShortEM (introduced 1/4/18 under the title, “#UnderweightEM”):
- Beijing Remains Handcuffed (10/18): With respect to Chinese monetary policy, we obviously track a deluge daily/weekly/monthly data – e.g. OMO, MTLF, RRR, BPR, 7DRR, SHIBOR, etc. – but for those of you who want to boil it all down to what it actually means for economic growth on the mainland, there’s one number you need to track on a monthly basis – i.e. the annual growth rate of bank loans. Why? Because that’s the best way to track incremental liquidity in the Chinese economy, which remains structurally challenged from an existing-supply-of-likely-unserviceable-but-technically-not-nonperforming-credit perspective. Yesterday’s monthly monetary figures were telling in that bank loan growth actually slowed -4bps in SEP to 13.17% YoY, with the TRENDING momentum essentially flat as well. These numbers are well shy of the 15-16% average annual growth rates seen throughout Beijing’s prior liquidity wave, which persisted from late-2015 through 1H16. Recall that said wave of liquidity helped catalyze the SOE fixed asset investment boom that rescued the “Old China” economy from the brink of collapse and ultimately catalyzed the “Globally Synchronized Recovery”, as well as Xi’s “coronation”. The latest numbers – including the sharp deceleration in M0 Money Supply growth (↓ -110bps to 2.2% YoY in SEP; TRENDING lower) continue to suggest Beijing remains handcuffed by international monetary pressures from providing the degree of monetary stimulus it would need to comp cycle-peak comparative base effects over the next two quarters. The A-Shares (↓ -2.9% overnight and ↓ -31% from its 1/29 YTD high) tells you all that you need to know about the likelihood for China coming to the rescue of global cyclical exposures in the near term. In that light, it should come as no surprise to see Japanese Export growth plunge -780bps to -1.2% YoY in SEP – the slowest growth rate since OCT ‘16.
- The Most Important Growth Number In Macro Slows Again (10/19): Last week we discussed Japan’s Machine Tool Orders as the single most relevant leading indicator for the global manufacturing economy. Today, we received the most important coincident indicator for said economy – Nominal GDP growth in China’s Secondary Industries, which slowed another -61bps to 9.88% YoY in 3Q18. It’s the sixth consecutive sequential slowing off the 1Q17 cycle-peak growth rate and, more importantly, is slowing into cycle-peak comparative base effects in 1Q19E. Looking underneath the hood, there’s nothing in the monthly data for September in terms of Industrial Production (↓ -30bps to 5.8% YoY; TRENDING lower) or Fixed Assets Investment – particularly SOE Fixed Assets Investment ( ↑ +10bps off the all-time low to 1.2% YoY; TRENDING lower) that would suggest Beijing is done anything to counteract the ongoing slowdown in growth throughout the “Old China” economy. As such, we feel confident pounding the table with respect to our longstanding bearish bias.
- Beijing Doesn’t Want To Risk Pushing On A String (10/23): We don’t see any reason for China-sensitive exposures to bottom until Beijing pulls out the stimulus bazooka, which isn’t likely to happen unless the prior condition of Fed easing being met. A strong(er) dollar severely constrains what Beijing can do on the monetary easing front given that Chinese nonfinancial corporates are the EM world’s largest single issuer of dollar-denominated debt. If they start to fear Beijing will let the CNY slide well past the vaunted 7-handle due to excessive easing amid ongoing tightening by US monetary authorities, they will start to prepay debt and increasingly retain earnings overseas. That would lead to a strain on M1 growth that hampers Chinese banks’ ability to grow their loan books on the mainland (refer to slides 5 and 6 HERE for more details). What happened in the Turkish and Argentine economies prior to their central banks ultimately responding with higher rates is a classic example of the economic and financial market destabilization that can be born out of a credit crunch that itself is perpetuated by having too lax monetary conditions relative to [changing] global financial conditions. Absent a change in monetary conditions, a widening of budget deficits risks crowding out China’s private sector, as Chinese banks own ~3/4ths of Chinese sovereign bonds. Regardless of what they do, they’re going to have to do a lot more than they did in 2016 to comp the mountain of comps. So even if they do announce more meaningful easing than they have to date, to the extent it’s shy of what’s required, it may not equate to a positive impact on growth in the near-term. The last thing Beijing wants to do is appear like it’s pushing on a string (a la U.S. in 2008).
- South Korean Growth Surprises to the Downside (10/25): Comping up against its most difficult comparative base of the current cycle, South Korean Real GDP growth slowed -80bps to a 9-year low of 2.0% YoY. Our GIP Model has the South Korean economy effectively hugging the line between #Quad2 and #Quad1 here in 4Q18E, followed by the always-precarious #Quad1/#Quad4 straddle in 1H19E. Ahead of this “less bad” outlook, we closed our EWY short admittedly a bit early back on 10/5, but with market signals having turned increasingly bearish since (e.g. recent new lows in the KOSPI and KRW), we’re keen to wait this one out prior to turning bullish. If the SMH chart is any indication, it’s going to get worse before it gets better for global cyclicals. We’re not in a rush to lose money – and our willingness to obey market signals is the #1 reason we haven’t yet hit the “buy” signal on a factor exposure like U.S. Homebuilders despite having a positive fundamental bias.
- Bolsonaro Optimism Likely Reaching Its Crescendo (10/29): Jair Bolsonaro’s victory in the 2ndround runoff for Brazil’s presidential election has been met with incremental buying pressure, though both the BZA and BRL are now trading well off their initial highs of 90,375 and 3.5864, respectively, as investors begin to actually digest the likely path of policy going forward. Recall that we’ve been in the camp that meaningful fiscal reform was largely a pipe dream given the composition of Brazilian Congress – which, after the recent elections, remains as fragmented as ever. Specifically, pro-market parties like Bolsonaro’s PSL and PSDB now hold 131 (of a total 513) seats across seven parties in Brazil’s Chamber of Deputies, up +23 seats from the previous Congress. That compares to 247 seats (↓ -36) for centrist legislators across 18 parties and Independents and 135 seats (↑ +13) for statist parties like the PT and PSB. Investors should be aware that Bolsonaro’s victory was more of a repudiation of the corrupt PT – which lost the presidential race for the first time in 20 years – than it was for his liberal economic proposals; a third of Brazilians abstained from voting – the highest level since 1989. “We are going to accelerate privatizations,” Bolsonaro’s economic advisor Paulo Guedes said publicly, only to be subsequently undermined by Bolsonaro’s decision to rule out privatizing the core operations of oil giant Petrobras, as well as the generation units of state power utility Eletrobras. With Brazil’s Unemployment Rate at a structurally elevated level of 12.2% as of 3Q18 (it was as low as 4.6% as of 4Q14), we can’t see how the kinds of austerity measures investors are hoping for are going to fly politically – especially in the context of a persistent #Quad3 economic backdrop through 1H19E. Speaking of hope: Citigroup, UBS, and Banco Bradesco are out this morning with 104,000 (+20% from last price), 119,000 (+37%), and 120,000 (+38%) price targets on the Bovespa Index by YE ’18, mid-2019, and late-2019, respectively. We haven’t received full confirmation of upside capitulation in our derivatives market analysis just yet, but to the extent we do receive it tonight’s volatility data and in Friday’s CFTC report (data through tomorrow’s close), we’ll be looking to reestablish EWZ shorts.
#GlobalDivergences (introduced 1/4/18):
- Japan’s Machine Tool Order’s Growth Confirms New Lows in Global Growth In 4Q18E (10/10): If I were forced to rely on one high-frequency indicator to make money as a global cyclicals analyst, it would undoubtedly be Japan’s Machine Tool Orders YoY growth figure, which has lead cyclical peaks and troughs in YoY global Real GDP growth by roughly one quarter over the past 20+ years. It has also lead cyclical peaks and troughs in the JPM Global Manufacturing PMI by 1-2 months as far back as we have access to the data (T3Y). Moreover, the SEP data (↓ -230bps to 2.8% YoY; TRENDING lower rapidly) is confirming of our existing view that global economic growth is likely to take another leg down here in 4Q18E. Respecting the cadence of the data (i.e. the various lags by which hard data is reported broadly), we should expect to see new lows in global equities and new highs in the USD potentially well into 1Q19E.
- European Growth = Still Slowing (10/16): One welcome of the side effects of getting our fundamental research process to its most commercial point ever is that clients are now better equipped to debate the future state of any economy in GIP Model quadrant terms. For example, we have plenty of investors looking ahead to #Quad1 forecasts in Germany, Italy, and Spain for 4Q18E and pondering when it the right time to buy the dip in European equities. As Keith discussed in this morning’s Early Look note, however, one of the bigger mistakes we can make as informed investors is looking ahead to the quadrant outcome our portfolio prefers, while simultaneously overlooking the quadrant outcome that is still being reported. That is precisely the case with the latest batch of European economic data, which confirmed the ongoing slowdown has spilled over into the start of Q4: Eurozone ZEW Expectations of Economic Growth Index(↓ -12.2pts to -19.4 in OCT, the lowest since AUG ’12), Germany ZEW Expectations of Economic Growth Index (↓ -14.1pts to -24.7 in OCT; the lowest since AUG ’12). Italy’s Industrial Orders data for AUG was putrid as well: ↓ -190bps to 0.9% YoY, the lowest since APR ’17; the market doesn’t believe the Italian government’s rosy growth projections and neither does the data. We cannot stress enough that our comparative base effect model – which extrapolates current levels of momentum embedded in our nowcasts for growth and inflation over the NTM – is merely a probabilistic exercise that is designed to inform our views in the absence of reported data for any quarter. We cannot, however, disrespect the cadence of that data once it starts being reported – especially when the base effects imply a reversal of a longstanding trend, as is the case with European economic growth. Investors are better off waiting for either (or both) the data and market signals to confirm said inflections – neither of which is the case with Europe currently.
- Europe Still Slowing… This Is Why You Wait (10/24): The preliminary PMI data for the month of October confirmed ongoing TRENDING deceleration in European economic growth. Specifically, the Eurozone’s Composite PMI (↓ -1.4pts to 52.7) slowed sharply to the lowest reading since SEP ’16 and was led lower by broad-based weakness in the German economy (Manufacturing PMI: ↓ -1.4pts to 52.3; Services PMI: ↓ -2.3pts to 53.6), the region’s key driver of growth. France was an outlier on the Composite PMI front (↑ +0.3pts to 54.3; TRENDING lower still), but its Business Confidence Survey data (Headline Index: ↓ -2pts to 104 in OCT; TRENDING lower) suggests investors would do well to exercise patience in any attempt to determine an investable bottom in French economic growth. The same can be said for both Germany, Spain, and Italy – three economies that our comparative base effect model suggests are likely to move into #Quad1 here in 4Q18E. Investors would also do well to internalize the following two very important points regarding that revelation: 1) Our comparative base effect model is probabilistic in nature and is designed only to provide investors with the highest-probability path forward for growth (and inflation) in the absence of reported high-frequency data in any forecast period; and 2) The initial batch of reported high-frequency data for Europe here in Q4 has been nothing shy of rancid, which suggests the 68% probability German growth bottomed in rate-of-change terms in 3Q18E merely as a function of static base effects is receding at the margins. New YTD lows in both the DAX and the Eurostoxx 600 yesterday are confirming of that view, as is the EUR’s recent failure at immediate-term TRADE resistance.
- Canadian Dollar Breaking Down On Time (10/25): From its late-JUN lows to its highs at the start of this month, the CAD rallied over +4% vs. the USD amid and in anticipation of two quarter-point rate hikes (to 1.75% currently) out of the BoC. The aforementioned monetary tightening that was indeed well-supported by the data, which the predictive tracking algorithms underpinning our Canadian GIP Model have tracking well into #Quad2 for 3Q18E. What’s driving the current downdraft in the loonie (-1.4% MTD) is our projection for the Canadian economy to inflect into #Quad4 here in 4Q18E. Moreover, the monthly Headline and Core CPI figures have each inflected off their respective cycle-highs from a TRENDING perspective and our model anticipates further downside with respect to the next ~6 months. As such, we’re keen to reiterate our bearish bias on the CAD with respect to the intermediate term. That’s a key risk to earnings for U.S. corporations that have meaningful exposure to Canada.
Senior Macro Analyst
After building this base of knowledge, we can now select what we like and don’t like based on our historical back-testing of the different asset classes that perform best in each of the four quadrants. As you can see in the chart above, our GIP model suggests that the U.S. economy is narrowly in Quad 1 for 3Q 2018 before it ultimately slips into Quad 4 for 4Q 2018.