The Big Rotation
The recent rebound in the S&P 500 index is approaching a first area of resistance now. If it continues to roughly follow the 1962 analog which we have previously discussed (here is the chart), it should briefly back off, and then move slightly higher before rolling over again (we picked the analog because it featured a weak January open, and was preceded by a distribution phase in 1961 that looked very similar to the 2015 trading range; there is no other reason than pattern similarity, which has to be taken with a grain of salt).
We have actually been remiss in that we have failed to mention that Frank Roellinger informed us that the Modified Ned Davis Model has covered its Russell 2000 short position on Friday Feb. 26, and has switched back to a 100% long stance – however, readers hopefully caught our remarks published on the very same day:
“[W]e believe that the current market rebound could easily go further. Not only are there a number of historical patterns which suggest that market weakness in January is usually followed by a multi-week recovery, but the current positioning and sentiment backdrop also indicates that stocks should manage to trade firmer for a while.”
Now that the rebound has actually gone a bit further, we want to discuss its quality. There has been an improvement in market internals – mainly because previously beaten down sectors (such as transportation, industrial and commodity stocks) have begun to outperform the broader market. We have already briefly discussed some of the suspected reasons for this at the end of last week (stronger money supply and credit growth in China, and the cheapness of commodities in relative and absolute terms, regardless of still weak fundamental conditions).
Improving internals have gone hand in hand with rotation: this time, previously strong sectors are actually weak and lagging, especially technology stocks (including biotech stocks). When the market began to turn down in late December/ early January, the main beneficiaries were initially defensive sectors like utilities and consumer staples. A little later, a further shift in preferences has taken hold, with buyers rotating into the previously biggest losers, many of which have been beaten down to prices that are historically quite low. Here is a chart that is illustrating this theme nicely – it shows the ratio of XME to PNQI, i.e., the ratio of base metal mining stocks to internet stocks.
We could have shown similar looking charts comparing e.g. steel stocks, energy stocks or transportation stocks to big cap tech stocks. The main point we want to make is that the recent improvement in internals is this time not supported by a revival in the formerly leading momentum stocks in the technology sector (the bull market upside leader that has held up the longest). Even compared to the S&P 500 index, big cap techs have begun to lag:
Considering the valuations of the so-called “unicorns” trading in private markets in conjunction with the valuations accorded many of the leading Nasdaq stocks, we think it is fair to say that there has been another major technology bubble. This bubble may actually be bursting now. Its previously strongest sub-sector biotechnology has already become a downside leader, which is rarely a good sign.
The improvement in internals is in fact a bit uneven as well. For instance, the recovery in the new highs/new lows percentage has been weak, while the NYSE A/D line has been very strong of late. The number of S&P 500 stocks exceeding their 50-dma has jumped quite a bit (in fact, it looks now “overbought”), but far fewer stocks have managed to regain their 200 dma, creating a short term divergence in the process:
Taking a look at XME itself (which we use here as a proxy for all the beaten down sectors that have come up for air in the recent rebound), we can see that it is by now strenuously overbought and has experienced a blow-off in trading volume just as it crossed above its 200 dma line – in conjunction with a potential reversal candle (confirmation/ follow-through is still required to declare it one):
For all we know, this rotation may be the start of a new medium to longer term trend, regardless of what near term gyrations occur – it is difficult to tell at this juncture (commodities e.g. tend to bottom while their fundamentals still look lousy, so one cannot come to firm conclusions from their current fundamentals).
However, per experience it cannot be good for the big cap-weighted indexes in the medium to longer term when the largest tech momentum stocks are underperforming. Note that financial stocks, in spite of joining the bounce recently, still aren’t much to write home about either.
We remain therefore on “rally failure watch”. If the recently stronger sectors begin to correct and the technology sector concurrently remains weak, broader-based indexes will turn down again – though it might not happen right away.
A Mixed Sentiment Backdrop
It is difficult to come to conclusions about short term direction from assorted sentiment and positioning data. Some of them would suggest a likely continuation of the rally (such as e.g. futures positioning and small trader buying of put options), whereas others have reached neutral positions and yet others are already indicating that complacency has returned. One of the latter is e.g. the CBOE equity volume put-call ratio:
How to best interpret a batch of mixed sentiment data depends mainly on one thing: whether the primary trend is up or down. Can the direction of the primary trend be discerned with certainty from what has happened recently? Probably not yet – similar to recent deliberations about the likelihood or timing of a recession, there is quite a bit of evidence that suggests the larger trend has changed, but there can be no apodictic certainty yet (from the perspective of the price trend, this would require all the indexes that haven’t broken below their previous reaction lows to do so).
There are many longer-term sentiment indicators that show the market remains at a dangerous juncture though. We have previously discussed margin debt, mutual fund cash and retail money fund assets in this context (the links lead to the respective charts), but want to look at something different today, namely the total Rydex bull/bear asset ratio. The important thing about this ratio is that it has indicated a never before seen surge in trader optimism in 2014/2015. Such extreme values often have long term implications.
Here is one more sentiment indicator published by sentimentrader that illustrates why sentiment data often work slightly differently depending on whether a trend is changing or depending on what type of overarching trend is underway. This one is the so-called smart/dumb money confidence spread, which compares an amalgamation of indicators that are traditionally associated with good, resp. bad market timers (mainly this involves hard data rather than opinions). We have picked a longer term chart and highlighted a few areas on it:
As can be seen here, when “bad” market timers were excessively pessimistic in 2007/early 2008 and excessively optimistic following the 2009 low, they were actually correct. In other words, speculators were correctly betting on the overarching new trends, while hedgers were providing the liquidity to accommodate their bets. Hedgers are usually not betting on direction: they are hedged with the underlying instruments and are merely making money on spreads. It’s just that speculators are as group often wrong when they are predominantly betting on the same outcome – evidently though, this is not the case after major trend changes – click to enlarge.
We would conclude that if the market turns down again from resistance amid “mixed” sentiment data, it would be another negative signal increasing the chance that a major trend change has occurred.
It will be interesting to see if the market continues to follow the analogs we have discussed earlier this year (other occasions when stock markets were exceptionally weak during the normally seasonally strong period in January). We have so far no reason to believe that it won’t – very likely the current rebound will simply go as far as it needs to go to look sufficiently convincing. Obviously, this idea would have to be abandoned if previous highs are exceeded.
Here is one last chart that continues to suggest to us that an overarching longer term trend change remains more likely than a resumption of the previous valuation expansion:
Junk bond yields are currently correcting their recent advance as well, but remain extremely elevated – and are confirming the increasing deterioration in outstanding bank loans – click to enlarge.
Charts by: StockCharts, SentimenTrader, St. Louis Federal Reserve Research