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.
What Type of SELL-off is this?
Having been in a number of meetings with investors this week, the current key debate(s) about markets can be encapsulated within one key question: “What type of SELL-off is this?”
In particular:
“Is the SELL-off all over?” (with the intraday lows on December 24th the final lows for this pullback).
Or:
“Is it a repeat of the 2015/early 2016 playbook?” (in which case there’s one more leg down after this current relief rally).
Or finally:
“Is it something more sinister?” (such that a sustained cyclical bear market is underway, and likely to be enduring).
As is usual when forecasting financial markets, discerning the correct outcome is challenging. The trick is to correctly frame the debate, lay out the evidence and then make a judgement.
As such, we lay out below the evidence (for and against) for each of those three scenarios. Of note, and importantly, we have done that within the context that there are three types of SELL-offs during cyclical bull markets:
- Risk events/waves of risk aversion – which are primarily about the ‘washing out’ of excessively exuberant & bullish positioning. These are short lived events, after which the bull market resumes its normal uptrend (e.g. Jan-Feb 2014; & Sept-Oct 2014; & Jan 2018 in the US);
- bear markets driven by global macro shocks (e.g. 2015/16, the Euro crisis, the Asian crisis etc.) – these are typically, although not always, the smaller bear markets; &
- bear markets driven by US (& global) recessions (e.g. 2007 – 09 GFC & 2000 – 02 bear market) – these are often the largest bear markets.
These three types are explained in more detail HERE (and see HERE for an analysis of all 10% plus S&P500 pullbacks from 1928 through to 2006 – and their underlying drivers).
- Is the SELL-off all over? The evidence for and against
Estimated probability (15%)
There are 5 key points in support of this notion:
1. The message of the models at the lows: Many of our medium-term trading models by late December were generating medium term strong BUY signals, equivalent to those from other major local lows (e.g. in early 2o16 or during the Euro crisis, see FIG 1 below).
FIG 1: Percentage of western stocks above their 200 day moving averages
2. The shift in positioning in portfolios: Portfolio positioning has become notably defensive in recent months (and especially weeks), as this SELL-off has played out. As a result, some of our positioning models (e.g. see FIG 2) are now suggesting that portfolios are defensively positioned. This supports the idea that the selling is done. That is consistent with anecdotal and survey evidence about portfolio positioning (e.g. see latest BaML fund manager’s survey – LINK).
FIG 2: World sector risk appetite model vs. US 10 year bond yields
3. The current strength of this rally: “The stock market just got off to its best start in 13 years” is how one investor website described this January’s stock market behaviour (so far). While that’s as much to do with the unusual nature of December’s price action, it does highlight a fundamental point: i.e. that this relief rally has been impressive. The S&P500, for example, is 10% higher than its 24th December low. Our daily RAG work (looking at 1 – 2 week trends) suggests a pause/consolidation for the next 2 – 5 trading sessions, should then be followed by a resumption of the uptrend (see today’s Daily RAG).
4. Shifting central bank rhetoric: The messaging from central banks, in particular from the Fed, has shifted dramatically since the December press conference (where Powell mentioned that the balance sheet reduction remained on ‘autopilot’). His Atlanta interview this time last week was widely viewed as one of the key reasons for further strength in this relief rally.
5. The non-consensus nature of this view: Up until this week, it would have been particularly non-consensus to suggest that the December lows marked the end of the pullback. Given that the consensus view is usually wrong (along with the crowded trades), that’s another interesting signpost.
The problem with those points above, though, is that each of them can be refuted/contextualised: i) The message of the models, for example, is strong but can get stronger. Indeed in major pullbacks, our models often generate a strong BUY on initial lows, and then reiterate that signal 1 – 3 months later on subsequent new lows during the next wave of selling (e.g. see the behaviour of FIG 1); ii) while positioning is defensive it can become more so or indeed remain defensive for an extended period of time (e.g. see FIG 2 during the Euro crisis); iii) the strength of the relief rally is usually linked to the size of the prior pullback. Given December’s vicious SELL-off, a strong relief rally should/is to be expected; iv) yes central bank rhetoric has shifted but: a) is it enough? & b) Powell’s comments on the balance sheet yesterday* seemed to somewhat roll back his Atlanta comments; v) Whilst that view was non-consensus in late December, it has become increasingly more consensus in the past week.
Over and above those factors, there are the more troubling macro and liquidity indicators laid out below (see points 2 and 3).
*(NB yesterday he discussed a ‘substantial’ reduction in the balance sheet over time – see HERE).
2. Is this a repeat of the late 2015/early 2016 playbook?
Estimated probability (50%)
Currently this is our central case expectation. More specifically we expect a continuation of this relief rally into late January/February, followed by a topping out/rolling over and then weakness into March/the start of the second quarter. That weakness is likely to constitute a retest of the S&P500 lows or potentially new lows. In this outcome that would then mark the final low for this 2015-16 playbook scenario.
As highlighted above, medium term models generated strong BUY signals in late December. That, however, doesn’t preclude further market weakness. Medium term models often reiterate strong BUY signals on a retest of the initial local lows. In 2016, for example, the model shown above generated a strong BUY signal in January and then reiterated it in February. During the 2007 – 09 GFC bear market, that same model made repeated new lows over the course of 18 months and then remained at notably low levels for the final six months of the global equity bear market (FIG 3 below). Many other medium-term market timing models behaved in a similar manner at that time. As such, while market timing models are useful for trading relief rallies (and helpful when thinking about the end of pullbacks), ultimately the medium-term trend of the market is about fundamentals: i.e. the macro and liquidity environment as well as potential shock risks.
FIG 3: Percentage of western stocks above their 200 day moving averages during the GFC (2007 – 09)
In that context, therefore (i.e. that fundamentals drive the medium-term market trend), it’s troubling to note how weak money supply measures have become. Our global (GDP weighted) M1 money supply model, for example, has reached its lowest level since the GFC. It’s now at levels consistent with global recessions.
FIG 4: Global (GDP weighted) M1 money supply growth (Y-o-Y %)
Most of that weakness is evident in both the US & China, with M1 money supply growth slowing sharply in both economies over the course of 2018, especially in China (where current M1 money supply growth is at its lowest since 1989). Somewhat encouragingly, US weekly M1 money supply growth has picked up in the past few weeks and grew at 5.6% in week 1 of 2019 (albeit weekly M1 money supply is highly volatile). Tight global money supply growth (as illustrated by FIG 4 as well as tight global financial conditions) needs a response from the world’s central banks. Currently while the Fed has started to signal less tightening (or potentially a pause), other major central banks have barely shifted their stance over recent months, as global liquidity has tightened (indeed the Swedish Riksbank raised its policy rate last month).
Added to the above concerns about overly tight central bank policy, a global economic slowdown is clearly underway. As is traditionally the pattern, these start in the manufacturing sector and, if left unchecked by monetary policy, they then spread into the services sector. The sharp pullback in US ISM manufacturing and new orders is consistent with the commencement of that pattern, and is likely to be indicative of a mid cycle slowdown (NB it also occurs at the start of US recessions – although this is not our central expectation – see HERE for US recession risk analysis). The weakness in German capex expectations is consistent with that (FIG 5 below – and see HERE for full analysis of Euro zone macro outlook). Over and above those factors, the Chinese economy, especially the private sector, is clearly under considerable pressure. Contracting cement production growth, sharply slowing retail sales, contracting shadow bank lending and a dramatic slowdown in car sales are some of the indicators that illustrate that (see HERE for further analysis).
FIG 5: EZ capex expectations (6m lagged) vs. EZ gross fixed capital formation (Y-o-Y %)
FIG 5a: Chinese car sales (Y-o-Y %)
More importantly, aggregate Chinese stimulus to date has been limited. One chart that illustrates that is the aggregate central bank liquidity provision by the PBoC (FIG 6). Of note, whilst liquidity has picked up in the past few weeks, overall since the start of 2017 it has been flat. This is in stark contrast to the dramatic increase in liquidity in 2016 (which started in 2015). Overall the PBOC’s balance sheet also continues to shrink.
FIG 6: Central Bank liquidity proxy (China) – aggregate OMOs plus MLFs
As such, this type of sell-off is our central expectation (at this current juncture).
3. “Is it something more sinister?”
Estimated probability (35%)
Whilst scenario two is our central expectation, this scenario has been assigned a 35% (‘best guess’) probability. If central banks are too slow to stimulate (or are not aggressive enough) bubbles could burst.
The ‘enduring bear market’ playbook is more typical of a late cycle economy, in which excess has built up over the course of the cycle. That excess usually equates to bubbles somewhere. As the financial historian Charles Kindleberger outlines in his analysis (e.g. see ‘Manias, Panics and Crashes’), cheap money creates bubbles (along with 3 other conditions), while the removal of that cheap money bursts bubbles and once bubbles have burst they are hard to control/offset (e.g. TMT bubble, housing bubble etc). As such, if the central banks are too slow to change their monetary course/ease policy, then burst bubbles have the potential to push markets into this more sinister scenario.
Current bubbles are multiple and include: The Chinese housing (infrastructure and credit) bubble; the ‘everything’ bubble (e.g. see FIG 7); certain western housing market bubbles (including Australia, Canada, UK, NZ and Sweden – all of which depend on further cheap money for their continuation and all of which are currently experiencing a recent tightening of money** and subsequent house price weakness); as well as US ‘corporate credit and leveraged loan’ debt bubble; and the ‘emerging market debt’ bubble (i.e. our ‘pass the debt’ parcel theme). All of these are significant bubble risks in the global economy and they all depend on cheap money to sustain them. A bursting of one/several of these bubbles would add to the current growth weakness in the global economy (and therefore add to the likelihood of a more sustained/enduring bear market).
FIG 7: The ‘Everything’ bubble – US household net worth relative to GDP
**albeit in some economies, like Australia, part of that has been driven by regulatory tightening.
See below for extracts of this week’s publications…
UK: Where Housing Goes, So Goes the Economy – extract from piece
While Brexit related uncertainty and risks are clearly negative (and receive plenty of airtime), the UK economy is under growing pressure from other, more troubling, sources of cyclical weakness.
Most notably, UK financial conditions are tightening sharply (fig 1), as they are in other parts of the world. The UK, though, has one of the world’s structurally weakest economic models. That model is based upon cheap money, rising asset prices (particularly house prices), and credit fuelled consumption growth. It’s therefore especially vulnerable to monetary tightening. As such, and in addition to assessing the ‘cheapness’ of money and the outlook for housing, we find it instructive to concentrate on forecasting spare household cash flow in the UK (which provides strong insight into consumption and ultimately broader economic trends).
Fig 1: UK financial conditions (Bloomberg index)
In that respect, it’s increasingly clear that monetary/financial tightening is resulting in house price weakness (and undermining the cyclical health of the consumer, see points 1 & 2 for detail). Coupled with growing pressure on the corporate sector (point 3), household cash flow (and consumption) growth should therefore remain weak this year and next (point 4), see Key Points below for detail. As such and whichever way Brexit risk manifests itself, the UK continues to ‘limp towards recession’ (particularly in the absence of policy easing by the BoE).
See HERE for full analysis….
Oil: Technicals & Fundamentals (starting to) Align – extract from piece
From a technical perspective, there is a strong case for staying LONG oil (we increased oil weightings in strategic portfolios in late December1). Net LONG speculative positioning, for example, has unwound significantly; the forward curve is in deep contango; and sentiment is at its most bearish level since 2015/16 (figs 9 & 10). In addition, our medium term technical models suggest that oil prices, despite rallying recently, remain overextended to the downside (see figs 11 & 12). While signals from these models have become less extreme in recent weeks, such strong signals usually mark the start of sustained, multi-month oil price rallies.
The key question, therefore, is whether fundamentals support that message from the technical indicators. In other words: Is a convincing and significant supply response to low oil prices now underway and, if it is, how tight will the oil market be in 2019? Or, once a technical relief rally has played out, will prices continue to trend down, perhaps in an environment similar to the 2014 – 2016 bear market?
Fig 1: US shale margin model2 (USD/bbl, 6m lagged) vs. oil well completions (shown with forecast)
1 For detail see 20th December Quarterly Global Asset Allocation No. 36
2 Shale margin model = WTI price LESS shale breakeven price (as a proxy for shale profit margins)
Our analysis points to growing evidence that a significant supply response from major producers (OPEC and the US) is underway. That response is probably significant enough to prevent an ongoing/prolonged phase of oil price weakness.
See HERE for full analysis….
Have a great weekend.
Kind regards,
Longview
Longview Research recently published
This week:
Commodity Fundamentals Report, 10th January 2019:
"OIL: Technicals & Fundamentals (starting to) Align, a.k.a. Increase OIL LONG positions on weakness” – see HERE
Global Macro Report, 9th January 2019:
"UK: Where Housing Goes, So Goes The Economy" – see HERE
Last week:
LV on Friday, 4th January 2019:
“Tactical & Strategic Outlook on the Equity Markets" – see HERE
Tactical Outlook, 3rd January 2019:
"Counter Trend Rally Expected" – see HERE
