After making bull market highs last week, equities lost 1% this week, led by small caps (RUT) which shed another 2.4%. With RUT now down 8% from its July high and under its 200-dma, investors are starting to wonder whether a larger correction is underway.
One reason to expect a larger sell off is that that has been a pattern during mid term election years. As an example, in the past four mid-term years, SPX has sold off by 8%, 16%, 20% and 34%; from its high to its eventual low has taken 2-6 months. In the past 6 months, the largest correction was 4.5% and took just 11 days.
Another reason is that SPX will have traded above its lower Bollinger band for 22 months by the middle of this coming week. That is four months longer than even the powerful 1995-96 period.
The powerful and sustained move higher in equities has not gone unnoticed. Investors have gone longer without even a brief period of bearishness than any period in the past 25 years. This even includes the hyper bullish 1990s bull market (chart from Yardeni).
Why have US equities been so exceptionally strong? It's not the macro environment. The revised 2Q GDP data released on Friday shows growth of 2.6%; in the 1990s, it was consistently over 5%. "Real final sales" (GDP less changes in inventories) shows consumption growing at just 2.3%, an essentially unchanged level of growth over the past 2 1/2 years and quite a bit lower than during prior bull markets.
One reason for the strong performance of equities is liquidity. This liquidity is probably reaching a saturation point. Let's review several sources of liquidity - from the Fed, corporates and traditional investors - that have contributed to high demand for equities.
Perhaps the most contentious is from the Fed through its permanent open market operations (POMO), in which the Fed buys securities in order to increase banking reserves. Either some portion of this liquidity has influenced the equity market directly or investors' belief that it will has indirectly led to higher prices.
Either way, monthly POMO flows are slowing (table). In 2012, monthly flows were $45b, equal to $540b over the course of a year. These have been reduced since the start of 2014. In September, it was down 67% to $15b. It's possible that the Fed will announce the end of POMO in October. A tailwind for the market is going away.
Second, corporate share purchase programs provide a boost to equity demand, but these programs are also waning in size. Over the past 12 months, buybacks have amounted to $539b; coincidentally an amount equal to POMO in 2013. In 2Q14, share buybacks dropped by $50b (annualized). The flows are still positive, but the rate may be starting a period of decline. In the chart below, higher buybacks (blue bars) correspond with higher equities (red line). The last two drops in buybacks were in mid-2011 and mid-2012, both periods where equities also fell (chart from FactSet).
Finally, it seems that investors' disposition for leverage in buying equities is also waning. Leverage as a proportion of market capitalization reached an all-time peak in February; since then, it has stalled. Unsurprisingly, rising leverage corresponds to rising equities, and vice versa. With the amount of leverage at a peak, further strong increases are less likely (chart from Doug Short).
As an aside, it is worth noting that households' financial portfolios already include a very high proportion of equities. That was not the case even one year ago. Note in particular how strongly those equity allocations have recently risen. This corresponds directly with the sharp rise in share prices (chart from the Fed).
This is also the case for pension funds. Again, the point is not to imply that equity allocations are, de facto, set to fall. But, equities rise the most when investors have been underweight. That's no longer the case: both the chart above and the one below show a dramatic rise in the past two years. Also take note of cash levels (black line), which have not been lower in the past 65 years. New equity purchases have to come from sales of another asset and both cash and bonds holdings are relatively low.
So, the liquidity tailwind to equities is probably easing. Selling RUT first makes sense. We showed last week that their valuations are at a significant premium. There is also a reasonable concern that small cap profit margins have peaked.
That's a legitimate source of concern for the broader market. In the past two years, SPX is up approximately 45%. About 70% of this gain is from multiple expansion. Another 15% is from margin expansion and the remaining 15% from sales growth. Stagnating margins when multiples are already at a premium leaves the market reliant on only sales growth. If the trend in macro growth continues to be a good barometer for corporate sales growth, then the rate of share appreciation will likely approach 2-3% (real).
The Week Ahead
The set up coming into this week was squarely bearish: SPY up against a verified resistance line, a break in RUT below 1160, poor seasonality, a likelihood for higher Vix, rising equity put/call ratios and a spike higher in Skew all suggested a downside edge (post). In the event, markets had their largest sell off in two months.
The week ahead is not as clean. Surprisingly, after just 3 days of selling from an ATH, a number of breadth indicators became oversold. On the one hand, the current bull market is nothing if not persistent, which favors further upside. On the other, strong down momentum, as seen this week, usually takes more time to dissipate, and this favors a continuation lower, even if just to retest the Thursday low.
Let's review the facts:
The longer term trend in SPY remains intact despite this week's retreat. Recall that SPY made a new ATH just a week ago. It's now down one week. SPY has not been down two weeks in a row since January. On the weekly chart, SPY remains in a strong uptrend until its 20-wma (blue line) is breached.
The 20-wma is now at 195. A close below, based on the past, would likely lead to the aforementioned touch of the lower Bollinger, at 187.5. From the recent high, that would equal a 7% drop.
The 195 level is also important on shorter time frames. On the daily, this marks a key uptrend from the March 2013 low. This line has not been breached in 18 months. Note the pattern in RSI (top panel): weakness in the coming week(s) to 195 would resemble April 2014 and October 2013.
On the hourly chart, 195 corresponds to a trend line from February as well as trading lows during July. The rally on Friday leaves the short term pattern unclear. Note the two trading ranges highlighted in yellow. SPY ended the week between them without first touching the 195 area. There's a series of lower highs and lows from last week and the falling 5-dma (green line) ended the rally on Friday as it did on Wednesday. On balance, a return to the low seems more likely, but it's far from clean. Note that a gap from mid August would fill at 194.8.
Tipping the scales in favor of continued weakness is that the 13-ema is downtrending for all four US indices, suggesting that the short term trend is weak. We will be watching for an inflection higher to confirm the reversal.
The strong selling on Thursday produced a number of extremes in breadth. Most of these suggest the bottom is either in or close. Let's look at each in turn.
First, down volume on Thursday was 10 times greater than up volume, making a so called major distribution day (MDD). This indicates significant downward momentum. In the past, this has carried the indices down a bit further in the next few days into a low. This suggests some more selling before the ultimate low. There was only one exception in the past two years and this occurred after a month of selling (green arrow).
Dana Lyons also looked at MDDs when they occur, as this one did, near a relative peak in SPY. Returns ahead were weaker than after MDDs occurring at a low. This makes sense as a MDD typically is sign of exhaustion which, near an ATH, would be unusual. Dana's full post is here.
Second, we commented on weakness in breadth last week. At this week's low, 20% of SPX companies were above their 20-ema and 37% above their 50-ema. In the chart below, note that nearly every low occurred after the percent above their 50-ema reached 30% (arrows). It's possible that SPX will bounce or trade sideways for a week before making that low. The point again is that downward momentum usually takes time to dissipate. April 2014 was one of only two exceptions in the past 4 1/2 years.
The 10-day average of the SPX cumulative advance-decline line also reached an extreme on Thursday. Like the studies above, this indicates a low is near but there is usually a bit more selling as momentum dissipates (chart from Bespoke).
Thursday also created a spike over 2 in Trin (the advance-decline ratio for issues divided by that for volume). Of the studies we looked at, this is the most squarely bullish. A high Trin last occurred at the August low and marked the bottom to the day (post). As we noted then, the biggest failures in Trin being close to the low have been when stocks have been rising in the prior week. That was not the case on Thursday. The spike in Trin and the jump in SPX on Friday might well indicate that a low was made.
According to Ryan Detrick, SPX has been higher one and two weeks later ~66% of time after a spike in Trin over 2.
Seasonality favors further weakness. September is typically weak into month end. According to Chad Gassaway, the last day of September (Tuesday) has been higher only 3 of the last 14 years (21%). And (also from Chad) the entire week leading to Yom Kippur (Saturday) has typically exhibited poor seasonality, up just 33% of the time.
Weakness ahead of Friday might be a nice set up. NFP (employment) is released on Friday. These have been very bullish: SPX is up 18 of the last 22 NFP days.
Finally, a few words on the strength in the dollar. The dollar has rallied strongly, up 10 of the last 11 weeks. On Friday it met a possible 2-year trend resistance line. The move looks parabolic, which is typically not sustainable.
Sentimentrader shared a study showing that dollar hedgers now hold a record short. In the past, this has marked at least a short term peak in the dollar index.
Recognizing that the dollar is in a powerful rally, should it weaken shorter term, there are several beneficiaries. One is emerging markets, which have faired poorly over the past month as their currencies have struggled against the dollar.
Commodities would be another beneficiary. The rapid rise in the dollar has had a similar impact as it did after a rapid rise in 2011 (chart from John Murphy).
Within commodities, energy stocks might be the best set up on a reversal. Every stock in XLE is trading below its 50-dma, an unusually oversold level. XLE is also, conveniently, on a long term trend line. Overall, it's a clean set up should the dollar reverse.
Our weekly summary table follows.
If you find this post to be valuable, consider visiting a few of our sponsors who have offers that might be relevant to you.