What worries me now

Below is a weekly chart of the NYSE Composite Index.

It shows price versus several different moving averages. It seems clear that the Index is rolling over. The faster, shorter moving averages are below the slower, longer moving averages -- not good and confirmation of a rolling over. And the Index price is "caught" underneath these moving averages (resistance).

Note how this compares versus 2001 and 2008. Eerily similar.

Also note how this compares versus late 2011. Then the Index price got below the moving averages, BUT the faster moving averages generally remained above the slower moving averages, AND the direction of the overall trend as per the MAs remained flat-to-positive, not negative.

Then things remained relatively bullish, esp. compared to now, where again clearly the trend via the MAs is currently negative -- like in 2001 and 2008.

I remain concerned.

Not Out Of The Woods Yet

Since the swift and dramatic correction in August, equities have put in an impressive rally off of what appears to be a double-bottom.

The S&P 500 has climbed about 11% from its August low and on Friday successfully broke through key resistance at 2050, the breached support level that led to the plunge in August. It will be important for the index to experience carry though this week, to rise clearly beyond the resistance area and towards prior highs of 2100.

It is a bit concerning to see the flatter Russell 2000 Index performing less well.

Whereas the S&P 500 has achieved reaching the low levels of August, the Russell 2000 has yet to get there (2000 level). In fact, the 1170 area looks to be serving as resistance thus far. Also note that the Russell 2000 peaked in June compared to the S&P 500 peaking a month later in July -- a bearish divergence as the R2000 Index made a lower high in July and continues to underperform the S&P 500.

On a longer-term monthly basis, I keep an eye on the following S&P 500 chart:

The chart shows four things that I have tracked over time: 1) the 6-month moving average (MA) vs. 12-month MA, 2) Unemployment rate, 3) stochastic, and 4) MACD histogram. This 4-item checklist has helped to serve as a reliable longer-term barometer for the stock market. Currently, the picture is somewhat worrisome. To run through the list:

1. The 6-month MA has pierced down through the 12-month MA, indicating a possible negative trend change. This occurrence does not always result in a trend change (see late 2010 and 2011) and it's more valid when there's greater daylight between the two MAs. However, it gave plenty of heads-up notice in 2000-2001 and 2007-2008 (blue circles). BEARISH
2. The unemployment rate (black line) remains in a downward trend, bullish. Note in 2000 and 2007 it broke downward trend pre-market corrections. But for now: BULLISH
3. The stochastic indicator (STO) has declined from above 80, but remains above 50, another good sign. Note in 1998, 2010 and 2011 the stochastic similarly eroded to near 50 but did not break down through that key level. For now: BULLISH
4. The MACD histogram is concerning. Currently the monthly reading is -25.4, well below -10 (blue line) indicating momentum thrust, if you will, has decidedly turned negative. Whereas the market corrections in 1998 and 2011 did not result in the MACD histogram submerging below -10, it clearly did so for 2000-2001 and 2008. My fear is when momentum turns negative this abruptly and strongly, it can tend to have legs or carry through going forward. BEARISH

To sum up, the weight of the evidence in the above chart is mixed with a 50/50 bull/bear split. However, given the MACD histogram in particular, I continue to lean on the cautious side.

Another indicator I follow that suggests we're not out of the woods yet is the percentage of NYSE stocks above their 200-day MA.

As shown in the weekly chart above, during the August correction, most stocks could not avoid the carnage. Only 20% of NYSE equities were able to remain above their 200-day MA. With the recent rally this indicator has improved to nearly 35%, however it remains far below the key level of 75% that I prefer to see for an all-systems-go signal.

For stocks to get below the 30% level (blue horizontal line), it means that most stocks have experienced significant technical damage and typically it takes some time for healing to occur in the form of widespread price recovery (breadth). Note in the chart that after declining below 30% in 2002, 2008 and 2011, this indicator eventually returned to the 75% or higher level in less than 12 months time, i.e. relatively soon, inferring a higher probability for an ensuing secular bull market. Yet a glaring exception is 2008, when early in that year this indicator dipped below 20%, then rose to as high as 60% before rolling over.

Does a reading below 30% suggest a very oversold condition for equities? Yes. In the long-term scheme of things (20+ years), do such extreme oversold readings make for good entry points? Arguably yes, they often do. HOWEVER, as I illustrated with 2008, there could be more ugliness to come and some may wish to exercise patience, waiting for the 75% threshold to be obtained.

I'll finish by showing one more chart:

S&P 500 trailing 4-quarter EPS (Y/Y %) is fast approaching the 0% level (orange line). Net profit margins remain at record levels, but typically when 4-quarter earnings get this low, net profit margins have reversed trend, heading lower. Worse yet, note the grey-shaded vertical bars indicating recessions and how the red line is often at 0% or lower before recessions take hold. Yes, this time may be different (as it was in 1985 and 1998), but the above chart remains another item of concern.

UPDATE: After a Down August, What Happens Next? Part 2.

After posting the blog entry on what occurs after down or negative-return Augusts, I received a few emails suggesting that I contrast these results with up or positive-return Augusts. By showing results for the 17 up-August years to compare versus the 13 down-August years, we would be able to see if the results differ. If the two sets of post-August returns were similar, then returns following down-Augusts would be less meaningful. Good suggestion.

Here are the results for all 30 years:

On the left are results from my prior blog post, showing September-December returns in years when August has been negative. The average return in those 13 Augusts was -4.7% and the average September-December period return was +9.2%, with notably all 13 Sep-Dec periods being positive. I also give the median percentage figures, -3.9% and +6.2% respectively.

On the right are results for the 17 years when August had a positive return. The average return in those 17 Augusts was +2.7% and the average September-December period return was -0.7%. The median percentage figures were +2.0% and +0.6%, respectively.

As I always warn, the usual caveats apply (sample size is limited, past results do not guarantee future results, etc.), but clearly years with down or negative-return Augusts tend to have better subsequent September-December periods than do those years with up or positive-return Augusts. 

After a Down August, What Happens Next?

With equity markets going from bad to ugly in just a matter of days, one has to wonder what the near-term future has in store. So far this month the S&P 500 has plummeted -11.2%, putting this August on track to be the second worst performing month since the start of 2005. Topping the list is the month of October 2008 with a -16.9% return and currently in second place is February 2009 at -11%. The fact that performance this month is as awful as some of the worst months during the catastrophic market meltdown of 2008 into early 2009 helps to put this correction into proper perspective.

We're accustomed to thinking that September and October are typically more tumultuous months than August, but it's not true. On average over the last twenty years, August has fared worse than September or October, as have June and July. Summer months have been less than kind to investors and tends to support the "sell in May, go away" refrain.

Given this month is more than likely (!) to be negative, what has happened after down Augusts?

To answer this question, I looked at S&P 500 monthly returns for the last 30 years (1985-2014). In that time, the S&P 500 experienced 13 negative Augusts, shown below.

The average decline was -4.7% for those August months.

For the years that had a negative August, how did the rest of the year fare? Answer: quite well. Over the next four months (September-December), the S&P 500 rose by an average of 9.2%, not bad! Also note that the September-December returns were positive for all years, an impressive 13-0 hit rate.

The usual caveats apply (the sample size is limited, this time may be different, past performance does not guarantee future results, etc.), but given the painful price action and scary headlines of late, it's at least somewhat reassuring to know that the near-term future can be bright(er).

Risk Appetite Indicator Not Confirming Recent Market Move

Along with several other indicators, I use the ratio of high-yield "junk" bonds (HYG) to TLT to help gauge the risk appetite of investors, i.e. is the current market environment risk-on or risk-off? When the HYG:TLT ratio is trending down, it infers a risk-off bias is taking hold and is generally bearish for equities.

Source: Stockcharts.com

The chart above shows the S&P 500 in the upper inset and the HYG:TLT ratio in the lower inset. Note that with the market's recent attempts to make new highs, the HYG:TLT line has not confirmed this move, instead continuing to head south. 

Such a non-confirmation or divergence has proven to be something worth keeping in mind. Bearish divergences in 2007, 2010 and 2011 were quite timely as the HYG:TLT ratio did not confirm the trend in the S&P 500. Also, the bullish divergence in late 2008 / early 2009 suggested that the stock market's vicious decline was coming to an end, with the HYG:TLT bottoming in December 2008 and making a higher low in March 2009.

Will The Real US Dollar Please....

One could argue that the DXY Index is perhaps the most common representation of the US Dollar (USD). However it's not the only representation, and depending on which one is chosen can often dramatically alter one's view of the USD.

The DXY measures the performance of the USD versus a basket of six foreign currencies, deriving a geometric mean based on preset weights. Currently the currency with the largest weight (by far) in DXY is the Euro at a whopping 57.6% weight. The currency with the next largest weight is the Yen at a much more meager 13.6% figure.

An alternative measure for the USD is the Fed Reserve's trade-weighted representation (USTWBROA on Bloomberg). It offers a depiction of the USD that takes into account a much broader basket of currencies, currently comprised of 20+ regions and/or countries. And the weights are flatter with the largest weight, China's Yuan, set at 20.8% and the next largest weight, the Euro, at 16.2%.

Over time the DXY and USTWBROA have tracked fairly closely, but as shown below, that's not been the case since the latter half of last year.

Source: Bloomberg

Whereas the DXY broke down through a rising trend in the 3Q of 2013, the USTWBROA has maintained its steady ascent within the rising trend. Obviously the strength of the Euro and the variance in composition and weights of the two USD representations have much to do with this fairly recent disconnect. But depending on which measure one chooses to portray the USD, quite a different view can emerge. In the chart above, the DXY appears bearish and yet the USTWBROA appears quite the opposite. Food for thought.

Market Outlook and Materials vs. Consumer Cyclicals

First, apologies for the recent lack of blog posts, but I've explained in the past my current situation (ongoing job search due to my employer shutting down). With regards to the market, frankly I haven't had much to say since my posting on February 11th. On February 3rd, when the S&P 500 was around 1780, I wrote that I wanted to see at least two things occur to confirm a bottom, 1) the S&P 500 put together three consecutive up days off this low or one significant up day with a bar/candle range spanning 15-20+ points, and 2) the MACD to trigger a buy signal, i.e. for a bullish crossover to occur in the 12- and 26-day EMAs, also shown with a positive histogram. Both of these conditions were achieved on February 11th, see chart below (orange line).

Since February 11, the S&P 500 has risen 4.3% and is fast approaching the upper-bounds of an ascending channel (red lines). Longer-term the market remains in a solid uptrend, however several oscillators currently reside in overbought territory and I would expect this rally to slow down, consolidate sideways or even pullback from these levels.

The weekly S&P 500 chart below also suggests an interim peak may be at hand:

Despite the fact the S&P 500 is up YTD, attaining a new all-time high, note that the weekly MACD has yet to confirm this rise in the Index, instead remaining in a downtrend and establishing a bearish divergence. As shown in the chart above, a few negative divergences have developed over the last few years and all eventually resulted in a market pullback or correction, some more severe than others. 

One of the more interesting developments YTD is the erosion in relative performance of consumer cyclical stocks (I wrote about on January 21 and February 14) and the resurgence in what has been a longtime laggard, the materials sector.

The weekly chart above plots the relative performance of the XLB vs. XLY (or Materials SPDR vs. Consumer Discretionary SPDR). In late December, the XLB:XLY relative price successfully broke through a two year declining trend line and XLB's outperformance over XLY has continued into 2014. 

Will this trend continue? A significant factor will be the strength of the US dollar.

The weekly chart above shows the relative performance of the XLB vs. S&P 500 (lower inset) and the US dollar (USD, upper inset). It's fairly plain to see an inverse relationship exists as the relative return of material stocks tends to fare better when the USD is weak and declining. 

Regular readers know that I turned bearish on the USD last October and I have yet to see a compelling reason that warrants changing that stance. The US dollar is down since early October and the weekly chart above shows the USD has been carving out a descending triangle formation of sorts since the 2Q of last year. If the USD breaks to 79 or worse, it would signify a clear breach of support as indicated by the gradually ascending trend line. I would also point out the developing two-year bearish divergence between this modestly rising trend line (higher lows) and the downward trending MACD indicator, which also resides below zero. Finally, the Death Cross condition remains in place for the USD, i.e. the 50-day MA is below the 200-day MA.

(Source for all charts: Stockcharts.com)