Weekly Market Summary

Summary:  Trade war rhetoric is driving US equities. This week, for the third time in the past month, the start of a sustained rally was clobbered by administration threats. Conversely, every interim recovery has come on the heels of conciliatory language. Long story short, what happens next in the equity market is very much a function of which trade posture the administration adopts next. Longer term, it's unlikely much of the current rhetoric will make into actual policy as it suits no one's economic interests.

Volatility has shot up in the past two months. Remarkably, investors now view volatility as the "new safe haven" and a "dependable bet." To that end, speculators are now positioned net long Vix futures to a near record extent; in the past decade, that has reliably coincided with at least a near term top in volatility.

This past week, SPX closed below it's 200-dma for the first time in over 400 days. The end of prior long streaks have not coincided with the start of bear market since 1962. Returns after the end of these long streaks have been exceptionally strong.

* * *

US equities gained three days in a row last week for the first time in a month but a massive gap down on Friday and further follow through selling turned the markets negative for the week and (mostly) for the year (from Alphatrends). Enlarge any chart by clicking on it.

The Worry About Indexing is Overblown

Summary:  Investors are clearly shifting away from actively managed funds to those based on index strategies. Only time will tell, but this has the look of a durable, secular change in investment management. But much of the perceived threat to market stability of indexing is overblown. Overall, the stock market is still dominated by active management. And while the number of index products has clearly exploded, 96% of these are of insignificant size.

* * *

Bloomberg recently reported that the number of indexes has exploded and now exceeds the number of stocks in the US.  Enlarge any chart by clicking on it.

The Similarities (and Key Differences) Between 2017 and 2013 So Far

Summary: 2017 is off to a remarkably similar start to 2013. No two years are ever exactly the same, so there's no reason to suggest that 2017 will repeat the 30% gains achieved in 2013. But many of the technical and fundamental similarities between these years suggest that 2017 may continue to be a good year.

There are two watch outs, however, that make 2017 much higher risk than 2013. It's also worth recalling that equities fell 3-8% at six different points in 2013. Expecting 2017 to continue to ride smoothly higher will probably prove to be a mistake.

* * *

2017 is off to a remarkably similar start to 2013. Is it set to be a repeat of that year? It's an important question as stocks gained about 30% in 2013.

Consider some of the following:

2017, like 2013, is the first year in a new Presidential term.

Both years got off to a fast start. By March, SPX had gained 8% in both 2013 and 2017.

Both years started by making a string of new all-time highs (ATHs). By the end of February, the Dow Industrials had closed at a new ATH 12 days in a row. A similarly rare streak of 10 days took place by March 2013.

SPX is often weak in February. But the index gained in both January and February in 2013 and 2017. In the other instances that this has happened since 1945, SPX closed up for the full year every time by an average of 24% (more on this here).

Both 2013 and 2017 came on the heels of long, volatile periods. SPX dropped 20% in 2011 and started 2013 only  2% higher than 18 months earlier. Similarly, SPX dropped 16% in 2016 and started 2017 only 5% higher than 18 months earlier. In both years, a dip at the election in November caused the market to be oversold; in both years, the market was significantly overbought by March (top panel).

Weekly Market Summary

Summary: All of the US equity indices made new all-time highs again this week. Treasuries were the biggest winner. A drawdown of at least 5-8% in SPX is odds-on before year year end, but there are a number of compelling studies suggesting that 2017 will probably continue to be a good year for US equities.

* * *

On Friday, SPX and DJIA made new all-time highs (ATH). During the week, COMPQ, NDX, RUT and NYSE also made new ATHs. All the indices moving to new highs together suggests that this is a broadly based rally. The trend remains up.

For the week, SPX and DJIA gained 1%. NDX notched a 0.4% gain and RUT closed lower. The biggest gain came from treasuries, with TLT gaining 1.4%. We continue to like the set up in treasuries, as explained in detail last week (here).

Little has changed from last week's summary. Instead of repeating those the same messages, we'll highlight four new studies that show a favorable longer term outlook for US equities.

First, SPX has now gone 76 days since the last 3% drawdown ended on November 4, right before the US election. That is the longest streak since July 2006 to February 2007, when the SPX went 150 days without a 3% drawdown. The chart below shows the duration and magnitude of the current rally relative to other long streaks in the past 14 years (yellow highlighting equals the current rally). Enlarge any chart by clicking on it.

Be On Alert For A Pop Higher in Volatility

Summary: The trend in equities continues to be higher, even a very short term basis. As equity prices move higher, volatility is compressing. That, on its own, is not bearish, as volatility can stay low for months as equities grind higher. But it's noteworthy that volatility has popped higher in each of the past seven Augusts. Combined with an unusually tight trading range in SPX and an extreme in the volatility term structure, short term traders should be on alert for a pop higher in volatility. That may well correspond with SPX approaching its next "round number" milestone at 2200.

* * *

This week, the major US equity indices - SPX, NDX and COMPQ - all traded at new bull market highs. Moreover, RUT has traded at a new 12-month high. None of these, nor the DJIA, has closed below its 50-dma since late June. All are trading above their rising 5, 10, 20 and 50-dmas. The trend for US equities remains higher, even on a very short term basis.

As always, the first sign of a weakening trend will be consecutive closes below the 5-dma, which will then flatten or inflect downwards. As of today, that is not the case for any of the US equity indices.

In our last update, we shared several studies related to trend, breadth, sentiment, macro and corporate reports that supported higher equity prices in the month(s) ahead. That continues to be the case. Read that post here.

But there were also reasons to be on alert for a retracement of recent gains in August.  This post elaborates further on some of these reasons with a focus on volatility.

The CBOE volatility index, Vix, which measures implied volatility in the stock market over the next month, has been under 12 the last 4 days and also intermittently under 12 over the past month. This is unusually low volatility.

On its own, a very low Vix is not necessarily bearish: forward returns in the SPX are no different than when the Vix is above its median of 18.6 (data from Mark Hulbert).

Volatility Is Set To Increase

Summary: On Tuesday, VIX closed below its lower Bollinger Band for the first time in a year. In the past, this has very often led to at least a 5-10% increase in VIX in the weeks ahead. But the affect on SPY has been mixed; just over half of instances were followed by a decline of at least 1% in the week ahead.

* * *

VIX measures the market's expectations for volatility over the next month. A low VIX implies that expectations are for little volatility looking ahead. Today's VIX is near 12, one of the lowest levels in the past year. Given the small daily and weekly movements in the SPY over the past several months, it is not surprising that VIX is low.

Bollinger Bands measure the movement of price around its mean. Using the most common set up, a movement outside of the upper or lower Bollinger Band is equal to 2 standard deviations from a 20-day moving average. Price should only fall outside of the upper or lower Bollinger Band only about 5% of the time so when this occurs, it is noteworthy.

On Tuesday, VIX closed below its lower Bollinger Band for the first time in more than a year. In the past 5 years, this happened only 15 times.

What happens next?

VIX itself has a strong tendency to rise in the days and weeks ahead. In 14 of the 15 instances, VIX increased by at least 5% and it increased by more than 10% in more than half of all instances.

Normally, SPY moves opposite to VIX; so an increase in VIX would typically lead to a decline in SPY. But that's not always the case and in the 15 cases where VIX closed below its lower Bollinger Band, SPY fell more than 1% only about 60% of the time. Stocks have a natural tendency to rise, so the likelihood of a decline is slightly elevated, but the edge is not significant.

The charts below show every instance where VIX closed below its lower Bollinger Band since 2010 (vertical lines). SPY is in the top panel and VIX is in the lower panel. A rise of more than 1% in SPY is highlighted in green, a decline in yellow. The same applies to VIX, except the threshold is a rise or fall of more than 5%.

In 2014, there were 3 instances when VIX closed below its lower Bollinger Band. VIX rose every time, once by more than 10% (lower panel). SPY fell by more than 1% twice but rose unabated once. The overall uptrend in SPY was not affected.

A Set Up For Volatility To Soon Rise

Summary: The S&P index is making new highs. The trend is higher. But in the process, the trading range has become very tight. In the past, this has been followed by higher volatility and limited near term upside for equities.

* * *

Over the past five weeks, the trading range for the SPX has contracted dramatically. Just 18 points separates the last 5 weeks’ closes (2118, 2108, 2116, 2123, 2126). 4 of these 5 closes are separated by less than 0.5%.

Moreover, the range between the weekly open and close for SPX the last 4 weeks has been 11 points,  6 points, 7 points and 5 points. That averages to a mere 0.3% open/close range each week.

Periods of contraction in the market are typically followed by expansion during which volatility increases. Presented below are several studies that suggest this is likely over the next week or two.

First, Salil Mehta of Georgetown notes that SPX has gone 8 days without a 1% intraday decline from a high. Streaks this long are rare and typically end after 9 days, meaning that a 1% decline is due next week. This implies an intraday visit to roughly the 2100 level is ahead (his post is here).

Crossing SPX 1000 in 1998 vs SPX 2000 Today

The S&P crossed above 2000 for the first time Monday.

This brings to mind 1998, the year the S&P crossed above the 1000 milestone.

The 1990s are wrongly remembered as a period in the market where volatility was uniformly low and double digit annual gains came in easy succession. The crossing of the 1000 milestone is a case in point.

SPX first came within 2% of 1000 in early October of 1997. In the month before, it had risen over 9%. After attempting to close higher for 3 weeks, SPX instead dropped over 10% in late October. So much for low volatility.

It was back near 1000 by early December, 1997. It again failed to close higher and dropped 6% by the end of the month.

Volatility Is Now At Multi-Year Lows

One indicator that has remained consistently positive in our weekly summary table is volatility (Vix). Volatility has remained below 20 since the start of 2013, with the exception of about 5 days. This correlates with strong equity returns.

A Volatility Set Up For Trading SPX Into Year End

Volatility as measured by Vix spiked higher today for the first time in two months. There is a useful trade set up that is therefore worth noting.

One of the least interesting but most useful indicators in 2013 has been volatility.

Throughout 2013, Vix has been telling investors to stay long. When Vix is under 20, average monthly returns in SPX are 1.5% and returns are positive 77% of the time. In comparison, when Vix is 25-30, average monthly returns in SPX fall to just 0.2% and returns are positive only 52% of the time. The higher the Vix, the worse the returns. Details are here (post).

There were two prior periods like today when Vix was consistently below 20. Both lasted 4 years and are notable for having infrequent and shallow (5%) corrections.  Coincidentally, the prior periods were 1993-1997 and 2003-2007. Is this a 10 year cycle starting with years ending in '3'? Details are here (post).

A more frequently useful way to use Vix is to trade long equities when a volatility spike fades. Specifically, the set up is note when Vix has closed above its upper Bollinger Band and then to go long SPX after Vix has closed back below its Bollinger Band in the coming days.

We are noting this because today Vix closed above its upper Bollinger Band for the first time in almost two months.

Below are examples of this set up from 2012: the top panel is SPX, the middle panel is Vix and the bottom panel tracks closes above and below the upper Bollinger Band. The vertical green lines are signals to go long.

What Monday's Jump in VIX Means For SPX

On Monday, VIX jumped to 19, an increase of 34% in one day and 55% over the prior week. What does this imply for SPX?

We have previously noted that VIX was, like today, in a period where it was sub 20 between 2003-07. Overall during this time, returns for SPX were mostly very good. However, VIX would occasionally jump 50-80% higher and $SPX would decline more than 5% over the next 1-4 months. We are potentially repeating this pattern now.

In the first chart below, we have updated the chart from 2003-07. The bottom panel shows VIX moves of greater that 30%, of which there were 9. In the top panel are corresponding moves in SPX. The percentage drop in SPX is noted in the text.

In the second chart, we have done the same analysis from 1992-97, also a period of low volatility. The conclusions are the same, with corresponding pullbacks of 5-10%.

Low Vix Means 5% Corrections Are Few And Far Between

In the past 20 years, there have been two long periods where $VIX was mainly below 20. During both of these periods, 5% corrections were few and far between. The chart below shows only changes in direction of >5%. Some observations:

1. Uncorrected uptrends during these times can last 4-12 months. 1995 went a full year. There are several lasting more than a full quarter. Low volatility has meant higher prices, long bull runs and shallow corrections.
2. The two periods ran from 1993-97 and 2003-07. A third period may be starting now, in 2013. Is this a 10-year cycle, lasting 4 years?
3. 1993 and 2003 came after economic recessions that corrected the market. We are now already in the 4th year of an uptrend. The starting point for this third era is very different.

VIX during 2004-2007

While a $VIX under is associated with excellent returns in $SPX, it is not accurate to say that intermediate periods of time will not see solid sell-offs. 

For example, $VIX was sub-20 from 2004-07. 2004 was a dog year for $SPX, the other years were very good. Note, however, that $VIX would occasionally 'burp' 50-80% higher during that period, and $SPX would decline more 4-7% for 1-4 months. The notion that this was a period of consistent low volatility is wrong

Buy When the VIX is High (Above 40)

Having just shown that the best risk/reward is when the VIX is below 20, you'll find a surprisingly good bounce in $SPX when VIX spikes above 40. In the chart below, from 1996 to 2012, there were 5 great buying opportunities when VIX over 40. Unfortunately, buyers using this strategy were destroyed  in 2008. So there you go: it works 83% of the time.

A great chart from Short Side of Long below:

Best $SPX Risk/Reward When VIX is Under 20

The best risk/reward in $SPX occurs when VIX is under 20. Average monthly returns (1.5%) and the maximum downside (-4.4%) occur during these periods. On average, monthly returns are positive 77% of the time. 

For comparison, when VIX is 25-30, average monthly returns drop (0.2%), maximum downside doubles (-9.1%) and the monthly returns are positive only 52% of the time. 

It gets worse the higher the VIX. A nice report can be found here with a the summary table extracted below.