Below is a round-up of Longview related views/research & trade ideas – the intention is to publish this most Fridays, updating key themes and highlighting key pieces of (often contrarian) research. Feedback, as always, is appreciated. Equally just ‘unsubscribe’ at the bottom if you don’t want to receive the email.
Beware the ‘palpable’ relief…..
With the S&P500 having rallied 215 points (i.e. 8.3%) from its intraday lows on 29th October into Wednesday’s high, and with equities across the globe following the US lead, the relief amongst money managers this week has been palpable. Key global equity benchmarks have now recovered notable proportions of their October losses and are honing in on key technical levels. The S&P500, when it touched its intraday highs on Wednesday/Thursday this week (at 2,817/18), had recovered 61.8%1 of its losses (FIG 1). It had also punched back above its 200 day moving average and was closing in on its 50 day moving average.
FIG 1: S&P500 (shown with key Fibonacci retracement levels, i.e. of the October losses)
*That ratio (i.e. 61.8%) is, of course, the key Fibonacci golden ratio, named after the famous 12th century Italian mathematician, regarded by many as "the most talented Western mathematician of the Middle Ages". The Fibonacci sequence, from which the ratio is derived, is seen by many as ‘Nature’s code’. See Benoit Mandelbrot ‘The Misbehaviour of Markets’ published 2004 for more analysis.
If history is any guide, though, this is not a rally that should be chased (at this specific point in time). Yes, it’s correct that over the next 1 – 3 months, we’d anticipate continued gains (e.g. see this week’s ‘Tactical Equity Asset Allocation’ – available HERE). Ahead of those gains, though, a near term retest of the October 29th lows is likely. That was the pattern of recent major SELL-offs in equity markets. We laid out the evidence for that in the Longview Alert we published Friday last week (see Longview 'Tactical' Alert: "Analysing Price Action Post a Crash a.k.a. how likely is a 'wave three'?" – available HERE). In that piece we highlighted the 19 SELL-offs since 1928 (outside of recession related bear markets) whereby the initial wave of selling in the S&P500 was 10% or greater. Of those 19, only one example did not have a third wave of selling (with most of those examples resulting in a retest of the initial low or indeed a new low).
Recent examples include the 2015 (July – August) SELL-off: After the initial wave of selling into late August 2015, the S&P500 staged a relief rally up to 17th September and then sold off through to 29th September (to within 31 points of those initial lows – FIG 2). This latest pullback is mapped onto the 2015 pullback below.
FIG 2: This pullback mapped against the one starting on 20th July 2015
Other examples include the Euro crisis SELL-off that started in July 2011. The price action of the S&P500, during this phase of the Euro crisis, is mapped below against this current pullback. Once again, the shape of the price action is similar2 with a wave three retest occurring in 2011 on the 62nd trading day of the pullback – see FIG 3.
FIG 3: This pullback mapped against the one starting on 7th July 2011
*albeit they are on alternative scales, given the greater depth of the 2011 sell-off
The April 2010 pullback is another example. This also suggests that further near term weakness is likely (FIG 4).
FIG 4: This pullback mapped against the one starting on 26th April 2010
Naturally there are, as always, exceptions. The SELL-off in early 2012, for example, was different in that: i) it was more drawn out and less precipitous in its initial phase; & ii) once the index made its lows (after 47/48 trading days – FIG 5), there wasn’t a retest3.
FIG 5: This pullback mapped against the one starting on 27th March 2012
*Of note though the Euro crisis was meaningfully advanced at this stage. As such the shock factor was (somewhat) reduced.
There remains, though, plenty to be worried about that could be the spark for the wave three retest of the late October lows. Those points of concern include, amongst others: i) the Italian budget situation (with Italian bond yields still trading at 294bps spread over bunds); ii) ongoing tight financial conditions in Asia ex Japan and increasingly so in Europe (with some modest deterioration in the US) – FIG 6; & iii) continued signs of stress in the Chinese economy (especially in the private sector), e.g. see Chinese high yield bond yields (FIG 7).
FIG 6: Asian and European financial conditions indices
FIG 7: Chinese vs. Asian high yield corporate bond yields
Over and above those factors, and as illustrated by the financial conditions in Asia and increasingly in Europe, central banks need to adjust their course: Money is tight; & the global economy is weak in certain parts (i.e. China), slowing in others (i.e. Europe) and likely to slow next year in the US (see Longview on Friday 2nd November 2018: “Time to BUY Europe?” – available HERE).
In the absence of that change of stance from the world’s major central banks, then this pullback is likely to be a more drawn out affair (i.e. more than just a three-wave pullback and closer to the late 2015 – early 2016 six month playbook).
Rotation into defensives – half way done….
That expectation of a ‘drawn out affair’ is consistent with the ongoing rotation that’s been occurring across global markets in 2018. By the start of 2018, cyclicals, EM, other reflation trades and equivalents had become crowded LONGs, while defensive assets were being heavily shunned. Reflecting that, many of our positioning models had reached extremes. The model below, for example, which looks at net speculative positioning in classic cyclical assets (e.g. AUD, CAD and copper) versus positioning in major defensive assets (including JPY and CHF currencies), reached an extreme pro-cyclical positioning in late 2017/early 2018 (e.g. see FIG 8). Since then, that positioning has been unwinding and is now just above NEUTRAL. If this pullback becomes a drawn-out affair (as discussed above), then we can expect that positioning to shift to an overall net defensive stance (as in early 2016). Currently therefore, while this rotation is advanced, it’s probably not finished (particularly if the Fed remains reticent to change its policy stance).
FIG 8: Positioning - Reflation vs. safe haven shown against consumer staples-materials relative performance
The conclusion is similar when examining the breadth of US dollar trading models. Sentiment towards the USD, for example, has become bullish again, having been bearish at the start of 2018 (e.g. see Consensus Inc. dollar sentiment). Positioning has also become bullish, having been neutral at the end of 2017/start of 2018. Just as described above, though, these models can become more extreme, i.e. implying further USD strength - see FIG 9.
FIG 9: USD index net speculative positioning vs USD
The picture is also similar if we look at global sectors: Rotation has been underway for several months amongst global equity sectors. As such, many of the models are now mid-range: Healthcare, for example, has been outperforming since May 2018 (FIG 12). Back then on a global PE heatmap, the sector was notably cheap and out of favour. Since then it’s outperformed by 13.4%. As a result, on our global sector PE heatmap, healthcare is valued in a mid-range (53rd) percentile versus the index, and at similar levels versus other sectors (see FIG 10 below)……
Consumer staples is similar (51st percentile – FIG 13)…..
FIG 10: Global PE ratio Heatmap
FIG 11: Healthcare valuation premium/discount relative to the market - shown with long term average
FIG 12: Global S&P1200 healthcare sector relative to the global S&P1200 index
FIG 13: Consumer staples valuation premium/discount relative to the market - shown with long term average
…looking more broadly across all sectors, our sector risk appetite model has been moving lower all year. At the start of this year, this model was at high ‘risk-on’ levels. Currently it’s just below mid-range, but not at levels typically seen at the end of the major sector rotations (i.e. 2015-16, the Euro crisis years and the GFC years – in chart below).
FIG 14: Longview global sector risk appetite vs US 10 year bond yields (%)
Curiously the key exception to this rotation across global assets is sovereign bonds. US 10 year bonds, for example, remain close to a multi year yield highs. And indeed have experienced a breakdown in correlation with our sector risk appetite model (shown above)…
…as well as with our narrower relative cyclicals vs. defensives performance model (below – FIG 15).
FIG 15: Global cyclicals vs defensives rel. performance vs. US 10-year bond yields
As laid out in our recent macro trade recommendation (see HERE), we expect bond yields to rediscover their correlation with sector risk appetite and relative performance, and move lower over coming months.
This week – start BUILDing LONG OIL positions
Another crowded trade at the start of this year was oil, with net speculative LONG oil contracts moving over 700k contracts in January. Oil, however, has not been immune to the recent phase of ‘risk off’ and has experienced its own bear market (with Brent off 17.5% from its highs; WTI 20.6% lower).
With that sharp change of price direction, the narrative amongst market commentators has also changed: “A shift in market sentiment” is how Bloomberg labels it. As is so often the case, though, the ‘price drives the narrative’ (post price rationalisation) and as such headlines, explaining asset price movements, are largely to be ignored.
Our work on the fundamentals points to a higher oil price looking forward 6 – 12 months. Supply in the market is tight, the Iranian sanctions have not helped (despite waivers offered by the Americans last week), while ongoing US growth should contribute to a broadly normal level of global demand growth in 2019 & 2020. As such, we continue to forecast a balanced supply-demand outlook in 2019 (and a deficit in 2020), with the main extra supply coming from more US shale. As argued in this piece (see HERE), that will require a higher oil price to compensate/incentivise US shale producers. This is therefore an attractive entry point (on a 6 – 12 month timeframe) into oil and oil related assets. That view is reinforced by our oil market timing models, most of which are now back on BUY (from SELL earlier this year), e.g. see FIG 16 below. As such we have started recommended building LONG oil futures positions (see HERE)…
FIG 16: Brent market timing model
…that view is also reinforced by the valuation of the global energy sector which has reached below average valuations relative to the global equity market (FIG 17 below).
FIG 17: Energy valuation premium/discount relative to the market - shown with long term average
…while the energy sector’s forward PER is close to its lowest quartile…….
FIG 18: Global energy PE ratio (based on consensus 12m rolling forward EPS)
…and its earnings growth remains robust, both on a standalone basis and relative to the global earnings growth outlook…
FIG 19: Consensus rolling 12m forward energy and global index EPS outlook
Elsewhere….
Elsewhere, one other snippet that caught our attention: Bloomberg today published an article highlighting some analysis of empty housing in China (i.e. vacancy rates). This, of course, was a key narrative that gained traction initially back in 2010 – 12 (with discussion, at that time, of 70 million empty houses).
This study suggests that nearly a quarter of Chinese housing (i.e. 22.4%) is empty compared with a range of 4% to 14% in the other countries cited. Perhaps more interesting, though, and a gauge of the level of speculation in Chinese housing, first time buyers in Q1 2018 are only 31% of all buyers (with the remainder purchases of 2nd or 3rd homes, by existing home owners). That’s the lowest share on record (i.e. since 2008, when over 70% were first time buyers).
Bubbles, of course, typically burst once everyone has stopped talking about them bursting (i.e. given up on the trade). Evergrande, the largest and most heavily indebted new home builder in China, is a case in point. It’s been one of the most heavily shorted stocks as a play on Chinese housing in the past 5 – 8 years. It’s only now, though, that it’s really starting to struggle. Just last week, its corporate bond yields reached their highest since on record.
Article here: LINK
FIG 20: Evergrande corporate bond yield (%)
Have a great weekend.
Kind regards,
Longview
Longview research recently published
This week:
Macro Trade Recommendation No. 94 , 6th November 2018:
“Start BUILDing LONG OIL Futures Positions” – available HERE
Tactical Equity Asset Allocation No. 178, 6th November 2018:
“Relief Rally to Continue” – available HERE
Weekly Market Positioning Update, 5th Nov 2018:
"(Positioning) Unwind Continues"
Last week:
LV on Friday, 2nd November 2018:
"Time to BUY Europe?" – available HERE
Longview ‘Tactical’ Alert No. 48, 2nd November 2018:
"Analysing Price Action Post a Crash a.k.a. how likely is a 'wave three'?" – available HERE
Longview Letter No. 117, 31st October 2018:
“High House Prices and 30 Years of Land Speculation a.k.a. It’s not about supply!” – available HERE
Weekly Market Positioning Update, 15th October 2018:
"Positioning in risk assets – not yet fearful"
