Sunday, January 25, 2015

All washed up!

Trend Model signal summary
Trend Model signal: Neutral
Trading model: Bullish

The Trend Model is an asset allocation model which applies trend following principles based on the inputs of global stock and commodity price. In essence, it seeks to answer the question, "Is the trend in the global economy expansion (bullish) or contraction (bearish)?"

My inner trader uses the trading model component of the Trend Model seeks to answer the question, "Is the trend getting better (bullish) or worse (bearish)?" The history of actual (not backtested) signals of the trading model are shown by the arrows in the chart below. In addition, I have a trading account which uses the signals of the Trend Model. The last report card of that account can be found here.

Update schedule: I generally update Trend Model readings on my blog on weekends and tweet any changes during the week at @humblestudent.

When models fail
Good investors know their own limitations. As an example, Stan Druckenmiller described himself as being "all washed up", because he doesn`t understand today`s markets and he has therefore lost his edge (via Zero Hedge):
It has become harder for me, because the importance of my skills is receding. Part of my advantage, is that my strength is economic forecasting, but that only works in free markets, when markets are smarter than people. That’s how I started. I watched the stock market, how equities reacted to change in levels of economic activity and I could understand how price signals worked and how to forecast them. Today, all these price signals are compromised and I’m seriously questioning whether I have any competitive advantage left.
Charlie Munger put it slightly differently, but the underlying message is the same. You have to know your own limitations and manage your investment risk accordingly:
Confucius said that real knowledge is knowing the extent of one’s ignorance. Aristotle and Socrates said the same thing...

You have to strike the right balance between competency or knowledge on the one hand and gumption on the other. Too much competency and no gumption is no good. And if you don’t know your circle of competence, then too much gumption will get you killed. But the more you know the limits to your knowledge, the more valuable gumption is.
I am neither a billionaire hedge fund manager nor am I Warren Buffett`s partner, but I recognize that the Trend Model may also be "all washed up". That`s because current market conditions are not conducive to the Trend Model adding very much value. The underlying premise of the model is to spot developing macro trends and jump on them in order to capitalize on the "bandwagon price effect", Unfortunately, there are no trends at the moment.

The chart below of the SPX in the last six months shows how the market environment has changed. Early in this period, the price trend of the market were long-lived. Starting about mid-December, the price swings got shorter and the magnitude of the moves were lower, which change the character of the market from a trending market (shown by the blue lines) to a choppy, whipsaw market (shown by the red lines).

This is an especially challenging environment for trend following models and my inner trader has had to rely more on short-term sentiment and overbought-oversold models for his trading. The markets are experiencing powerful cross-currents, which can be highly treacherous if someone is positioned in the wrong way.

Current conditions are suggestive of a range-bound stock market - at least we start to get more clarity on how fundamentals are developing. Until macro trends start to stabilize, I urge my readers to pay minimal attention to Trend Model readings. This model is "all washed up", at least for the moment.

The bull case: Global reflation
To explain my thesis for a range-bound market, I believe that the market is being pushed and pulled by both bullish and bearish forces, neither of which is able to consistently gain the upper hand. The result is a choppy up and down market, depending on which camp is dominant at the moment.

First, let`s start with the bull case. In my post last week (see The global deflationary storm is abating), I indicated that macro trends were improving from "bad" to "less bad". Indeed, if you tour the global markets this week, that seems to be the case.

Despite the ups and downs of the US equity market in January, the SPX managed to rally above its 50 day moving average this week and it is in a minor uptrend.

On Thursday, the world held its breath and waited for the ECB decision. Mario Draghi did not disappoint with the size and the scope of the QE program and the European markets melted up in response. As the chart below shows, Euro STOXX 50 staged a mult-year upside breakout.

China was also showing signs that it is reflating. Notwithstanding the casino-like atmosphere surrounding the Shanghai market, three of the four of the regional bourses that I monitor staged upside breakouts, indicating better optimism about the regional and global growth outlook. The fourth, which is the South Korean KOSPI, has rallied to test a downtrend line and, given the powerful regional bullish forces at work, may break out next week.

In short, global assets were enjoying a surge in risk appetite and the bulls were clearly in control of the tape.

The bear case: Earnings, macro and technical headwinds
Despite the rejoicing over the latest party thrown by Mario "whatever it takes" Draghi, ominous dark clouds were gathering. In the past few months, the US economy had been seen as the engine of growth while non-US economies sputtered, it may be the turn of the US to pause and catch its breath. Cracks are starting to appear in US earnings reports, employment and technically.

One of my biggest concerns has been the corporate progress shown the current Q4 earnings season. As this chart from John Butters of Factset shows (annotations in red are mine), forward 12 month EPS continues to fall and past episodes have been highly correlated with stock market weakness:

The reports from Earnings Season, so far, have been mixed. With 90 of the SP 500 having reported, the EPS beat rate was 79%, which is well above historical average, and the sales beat rate came in at 54%, which is slightly below. Analyzing the earnings results further, Butters' analysis showed that the 79% beat rate may not be as good as it sounds for two reasons:
  • The misses were much worse than the beats. Even though there were many more earnings beats than misses, the magnitude of the misses were much worse than the beats.
  • The aggregate earnings growth rate is disappointing, largely because megacap stocks tended to miss EPS estimates so that, on a capitalization weighted basis, these misses pulled down the aggregate growth rate. The worse culprits were the large cap banks, which virtually all missed Street expectations.
The combination of these effects led to significant downgrades in aggregate SPX forward EPS estimates - which creates a significant headwind for US large cap stocks. I had suggested in the past that the strong USD effect was creating a drag on large cap multi-national companies and mid and small cap stocks, which had better US domestic exposure, was a better place to be (see Focus on small caps in 2015). Unfortunately, this may not be the case. Analysis from Ed Yardeni show that forward earnings for small and mid-cap indices are declining as well.

Large, mid and small cap Consensus EPS
Click to enlarge
Forward EPS weakness is concentrated in three sectors: Energy, Financials and Consumer Discretionary, as shown by the chart below (via Factset). These three sector make up roughly 36% of the weight of the index and it would be difficult for aggregate earnings to make significant headway with such heavyweights dragging it down.

The presence of Consumer Discretionary stocks in the list was a big surprise as I would have expected that this sector would have been a prime beneficiary of lower oil prices. While it's still early in Earning Season, Factset reported that six Consumer Discretionary companies had given earnings guidance, with five negative and one positive.

As well, this interview with Ali Dibadj of Bernstein Research indicated that better consumer confidence does not necessarily translate into better growth for Consumer Discretionary companies and that the benefits are much more focused on specific industries, such as restaurants.

In addition to the stubborn headwinds posed by the lack of positive momentum from forward EPS estimates, another nagging worry I have is that employment may be starting to stall. The chart below shows the 4-week moving average of initial claims, which seems to be ticking up. While the rise may just be data noise, it is starting to create a concern for me.

This chart shows stock prices have been highly inversely correlated with initial claims during this recovery. Should initial claims start to rise again, it could spell trouble for the economic growth outlook and therefore stock prices.

The same inverse correlation held up well during past periods when stock prices were largely range-bound until its breakout in 1982:

However, initial claims was not a good explanatory variable for stock prices during the secular bull market (1985-1999):

My third source of bearish angst arise from the nature of the market leadership. The black line in the chart below depicts the relative returns of defensive sectors, Consumer Staples (XLP), Utilities (XLU) and Telecom (IYZ) relative to the SPX. As the chart shows, defensive sectors have begun to outperform and past episodes (which are shaded on the chart) have been associated with either bear markets or corrections.

What does this all mean?
When I put this all together, it is suggestive of a choppy range-bound market. While US equities are likely to enjoy the lift from global reflation effects from non-US sources, my base case scenario calls for a rally to test the all-time highs and a failure at about those levels as the bullish non-US tide diminishes as the major US indices face technical resistance. At that time, the bearish forces are likely to assert themselves and push prices downwards.

Tactically, we are likely to see two major events with binary outcomes and possible sources of volatility in the week ahead. This weekend, we have the Greek election. As I write these words, roughly 95% of the votes have been counted and Syriza has won 149 seats, which is just short of the magic 151 seat count required for a majority government. They therefore need a coalition partner, which may imply some softening of their anti-austerity stance.

The 10-year chart below of Greek stocks relative to the Euro STOXX 50 shows that these stocks are already approaching past crisis lows and therefore expectations are already very low. Even a minor positive surprise could cause European stocks to rocket upwards.

Even though SP 500 futures are deeply red at the time of this writing, here are some scenarios to ponder when interpreting the election results:
  1. The Syriza victory is a terrible outcome and creates more uncertainty in the eurozone as the specters of Cyprus, Grexit, etc. rear their ugly heads. (Bearish)
  2. Based on past polling, this is not an entirely unexpected result. Greece`s European partners were ready to negotiate some sort of accommodation in any case, so why worry? (Neutral)
  3. Given the Greek primary budget surplus, either we get (2), which is the status quo, or some sort of compromise where some of that budget surplus is spent, which follows Mario Draghi`s suggestion that Europe needs a little bit of fiscal stimulus. (Bullish)
  4. How often will the market consensus oscillate between (1), (2) and (3)?
In addition, the FOMC is meeting on Tuesday and Wednesday and market participants will be scrutinizing the statement closely to watch for any change in language. While the market expectations derived from the futures market indicates that the first rate hike will occur in the second half of 2015, both Tim Duy and Jon Hilsenrath have suggested that the Fed is much slower to react and the base case seems to be a June liftoff for interest rates. If that is the case, the Fed could use January FOMC statement to "correct" market perceptions, which could create tremendous market volatility.

Bloomberg recently featured a story that the Yellen Fed, despite its dovish tilt, is philosophically opposed the idea of a "Yellen Put" for the stock market:
Janet Yellen is leaving the Greenspan “put” behind as she charts the first interest-rate increase since 2006 amid growing financial-market volatility.

The Federal Reserve chair has signaled she wants to place the economic outlook at the center of policy making, while looking past short-term market fluctuations. To succeed, she must wean investors from the notion, which gained currency under predecessor Alan Greenspan, that the Fed will bail them out if their bets go bad -- just as a put option protects against a drop in stock prices.
Bill Dudley of the New York Fed outlined that approach explicitly and expects greater market volatility ahead. Investors should as well.
“Let me be clear, there is no Fed equity market put,” William C. Dudley, president of the New York Fed, the central bank’s watchdog on financial markets, said in a Dec. 1 speech in New York. “Because financial-market conditions affect economic activity only slowly over time, this suggests that we should look through short-term volatility.”
My inner investor remains neutrally positioned at policy weight in stocks and bonds in his balanced portfolio. My inner trader is nervously long equities with tight stops and a higher than normal cash position.

Disclosure: Long SPXL

Wednesday, January 21, 2015

Trend Model FAQ

Subsequent to my last Trend Model trading system report card (see An excellent Trend Model report card: (Dec +7.8%, 1 year +40.9%)), I have been getting an increasing number of questions about the Trend Model. Instead of responding to them one at a time, here is the Frequently Answered Questions (FAQ) about the Trend Model.

What is the Trend Model?
The Trend Model is an asset allocation model which applies trend following principles based on the inputs of global stock and commodity price. In essence, it seeks to answer the question, "Is the trend in the global economy expansion (bullish) or contraction (bearish)?"

The trading model component of the Trend Model seeks to answer the question, "Is the trend getting better (bullish) or worse (bearish)?" The history of actual (not backtested) signals of the trading model are shown by the arrows in the chart below and this chart. The chart is usually updated in a weekly on Sunday night:

Trading account management practices
The actual results of the account using the Trend Model trading system is a trading account. That account generally trades 3x leverage long or short ETFs on either the SPX, NDX or RUT. Total portfolio leverage normally doesn't exceed 2x. I have tried 3x total account leverage on a couple of occasions, but found that the volatility was beyond my comfort zone.

As the Trend Model uses trend following principles and trend following models suffer from a drawback that they do not perform well in choppy sideways markets, I supplement the Trend Model trading signals with some short-term sentiment and overbought/oversold indicators in order to judge the market's upside/downside risk and return. For example, if the Trend Model has a bullish signal but the market is overbought, I would either pull back from a 2x levered long position to either 1x long or go to cash entirely. I generally do not override the Trend Model signal and take the opposite view.

What would returns have been if...
As the actual returns of the trading account have been impressive, I have been asked variants of the question:
  • Can you tell me what the returns would have been if you had only used 1x leverage?
  • Can you tell me what the returns would have been if you did not short the market?
  • Can you tell me what the returns would have been in the year ____, when the market soared/tanked?
I always answer these questions the same way:
I present the results of my trading account for informational purposes only. I am not offering any advisory or management service using the signals of this model. If such an event were to occur, it would be accompanied by the proper legal and regulatory disclosures.

The purpose of the trading account is a proof of concept that the trading component of the Trend Model works. I have optimized the account in such a way that I believe maximizes the risk-reward ratio of this trading technique.

I have no plans to backtest the model based on different conditions or different forms of implementation. If I were to offer either an advisory or management service and the funds are managed with different risk preferences as my trading account, its actual mileage will vary.

Did the Trend Model turn bullish (or bearish) when...
When the markets get volatile, I often get asked question like the following: "You were bullish, now that the market has dropped ___, have you turned bearish?"
I endeavor to update Trend Model signals on a weekly basis on Sundays and during the week on my Twitter account. You can get those updates by following me at @humblestudent.

While I try to be prompt, I make no promises about the timeliness of updates. Remember that this is a free service. You get what you pay for.

Do you offer a subscription service?
There is no service offered. However, I am in discussions with a number of different parties on how to best monetize the signals of this model.

If an announcement is to be made, it will be done in due course.

Tuesday, January 20, 2015

(Sort of) debunking the bear case for stocks

I received a fair amount of feedback to my post last week, A consideration of the bear case. While an intermediate term top is not my base case scenario, it is a possibility that we have to consider. Subsequent to that post, I am revising down the probability of a bearish outcome for stocks in 2015.

First, let me recap that post. I showed this chart indicating the loss of positive momentum in risky assets, which suggested that most past episodes of momentum rollover have seen bear phases in the stock market.

I further showed a similar chart of the Wilshire 5000, which has deteriorated since that post was written:

Recency bias in analysis?
One criticism that I received was that I wasn't looking back far enough in my history. While the last 20 years is interesting, past instances of momentum rollover has not necessarily resulted in bear markets. The chart below depicts the SPX and its rolling 5-year rate of change. While the current ROC peak is higher than the 2007 peak and similar in magnitude to the NASDAQ peak, momentum peaks during the 1955-1980 period have not necessarily resulted in bear markets.

No valuation support
I further argued that valuations are highly stretched. Paulsen at Wells Fargo Capital Management showed that median PEs are very high by historical standards.

...and so are median PB ratios:

I would agree. The WSJ shows that the SPX is trading at 18.8x trailing and 16.7x forward earnings. The market is not cheap.

Earnings growth is the key to the market
If past momentum peaks do not necessarily indicate market tops, but valuations are rich, what should investors think?

The current environment suggests that EPS growth is the key to the market. As the Fed prepares to raise interest rates this year, don't expect gains from multiple expansion in 2015. Instead, we may see some multiple contraction.

In the past, the Fed has only started to raise rates when it believed that the economy was growing fast enough to withstand the withdrawal of monetary stimulus. Such episodes have been associated with economic growth, which has pushed EPS to greater levels. In short, we need to see earnings growth for this stock market to go up.

In my last post (see The global deflationary storm is abating), I voiced my concerns about this chart from John Butters of Factset. Forward EPS are still falling and past episodes of falling forward EPS have also seen either corrections or bear markets.

Nevertheless, I remain optimistic. The American consumer is poised to come back strongly. The Gallup survey of economic confidence saw a surge in December, just as oil prices fell.

In addition, a Gallup survey of life confidence showed that the personal life outlook of Americans is improving.

More importantly, the uptick is occurring across the board in all income categories.

So far, the view on the ground from Earnings Season paints a upbeat view of the American consumer. Here is an excerpt from Scott Krisiloff's summary of the corporate earnings calls:
The big question on everyone’s mind is: how does the oil price drop affect business?

The consensus is that it’s a net positive for the consumer

“America as a whole is a net consumer of energy and American households will benefit from the decline in energy prices, which is positive for the U.S. economy.” ($WFC)

“From any indication that we see it’s a positive experience for the American tax payer, for the American economy.” ($CSX)

“Overall, oil is a reasonable positive for the economy and consumer” ($JPM)

Bank of America is already seeing a benefit from lower oil prices in its credit card data

“in the consumer spending in January on debit and credit cards basically…even the first week or so January we’re seeing the benefit of the consumer very historically when the year-over-year compares” ($BAC)

So far growth is only 3% though, a step down from where it was

“So far January of 2013 versus January of 2014, spending on credit debit cards of about 3% year-over-year and that’s overcoming a drag effect of about percentage and a half from lower fuel prices.” ($BAC)

Wells Fargo is still optimistic on the economy

“when I look at the businesses that we’re in and as I am talking with customers…and frankly looking at the numbers..I am optimistic…I don’t think this is a breakout, but I think we’re on the front foot and consumer’s confidence is at a all time high since the downturn. So the way I read the tea leaves, I’m optimistic.” ($WFC)

CSX maintains a positive view of the economy too

“Looking forward, we expect a positive demand environment in the first quarter with stable to favorable conditions for 96% of our markets and unfavorable conditions for the remaining 4%.’ ($CSX)
Unless oil prices turn around and head back to over $100, these circumstances should lead to better consumer spending and a pickup in employment. I also showed this chart of the relative performance of consumer spending sensitive sectors in yesterday`s post, which indicated that the market is expecting a consumer-led turnaround. Recessions do not start when the American consumer is resurgent.

In the meantime, investors will have deal with the volatility that occurs during Earnings Season. Energy companies will see some EPS downgrades and so may large cap multi-nationals as they disclose the degree of headwind they face with a strong USD. Patient investors should be able to benefit from the benefits of higher consumer spending from a growing economy and lower oil prices. A better way to get exposure to this theme of a resurgent American consumer would be to tilt towards mid and small cap stocks, where the companies tend to do more business in the US rather than abroad (also see Focus on small caps in 2015).

Sunday, January 18, 2015

The global deflationary storm is abating

Trend Model signal summary
Trend Model signal: Neutral
Trading model: Bullish

The Trend Model is an asset allocation model which applies trend following principles based on the inputs of global stock and commodity price. In essence, it seeks to answer the question, "Is the trend in the global economy expansion (bullish) or contraction (bearish)?"

My inner trader uses the trading model component of the Trend Model seeks to answer the question, "Is the trend getting better (bullish) or worse (bearish)?" The history of actual (not backtested) signals of the trading model are shown by the arrows in the chart below. In addition, I have a trading account which uses the signals of the Trend Model. The last report card of that account can be found here.

Update schedule: I generally update Trend Model readings on my blog on weekends and tweet any changes during the week at @humblestudent.

Trend Model whipsaw
In my last monthly report card of the trading account that uses the trading model component of Trend Model, I stated that turnover averaged about 200% a month, which indicates that the typical holding period for a position is roughly two weeks. 

In the last two weeks, however, we have seen tremendous volatility in the US equity market. By the book, I would have had two or three signal changes in the trading model which flipped the signal from bullish to bearish and vice versa. The current environment exposes one of the weaknesses of trend following models - you need to have a trend in order to profit from them. Sideways choppy markets create whipsaw in trend following systems, with the resultant hit to the bottom line.

To mitigate this effect, I have overlaid a number of short-term sentiment and overbought/oversold models to the trading model signal. These short-term models identify the risk and reward of being a bull or bear by seeking the answer the question, "Is the market overbought (oversold)? Are bulls (bears) in a crowded long (short)?"

Despite the choppiness of the last couple of weeks, it seems that both the longer term trend and short-term trading models are aligning in a bullish fashion. The world is reflating and US equities are oversold enough that they are ripe for a rally.

A tour around the world
My framework of analysis has always been to focus on the three major trading blocs in the world, namely the US, Europe and Asia (or China).

In my last post (see Decision time in the eurozone), I identified a wedge formation in the Euro STOXX 50 and I was watching for the direction of the breakout, largely because the markets were worried about global deflationary fears. The source of deflation was coming from Europe and, to a lessor extent, China, not the US. Therefore monitoring non-US markets made more sense. After I wrote that post, eurozone stocks resolved the wedge in a bullish fashion.

Indeed, the ECB seems to be set to announce a QE program as Der Spiegel reported that Draghi briefed Merkel and Schaeuble on the details:
Spiegel magazine (without citing sources) reports that European Central Bank President Mario Draghi has briefed German Chancellor Angela Merkel and German Finance Minister Wolfgang Schaeuble on quantitative-easing plans 
  • Plan is national central banks would buy bonds issued by their own country
  • Envisages limits of 20 percent to 25 percent on purchases of each country’s debt
  • Says Greece will be excluded from the program because its bonds don’t fulfill the necessary quality criteria
An ECB spokesman declined to comment on the design of any QE program

A German government spokesman said earlier that Merkel and Draghi met on Jan. 14 for “regular informal talks,” while declining to comment on the topic.
European bulls still have to contend with the hurdle of a Greek election on January 25. However, the chart of the relative performance of Greek stocks relative to the Euro STOXX 50 shows that this ratio is nearing the levels seen in the last crisis low. In effect, the market is already betting on a collapse. Barring the announcement of a Greek withdrawal from the euro, which is not in Syriza's electoral platform, how much more downside is there?

Less awful in China = reflation
Conventional wisdom says that Chinese economic growth is decelerating. Shorter term, however, Beijing seems to have blinked and begun a stimulus program this is starting to see results (via Business Insider):
While attention has been fixed on the slow down it seems the world has taken its eye off one of the most important figures in China — total social financing (TSF). After slowing all through 2014 it made a comeback in December.

TSF is a measure that the Chinese government invented in 2011 to figure out how much debt non-state entities (like people and private companies) have taken on. Jim Chanos, founder of Kynikos Associates and on of the biggest China bears out there told Business Insider TSF is "still the most under-appreciated number in global finance."

Why is it so important? Because it gives you a picture of the condition of both borrowers and lenders, and includes things like China's infamous shadow banking sector, where interest rates can hit 20%.
As the chart below shows, TSF ticked up in December, which is suggestive of a near-term "less bad" scenario for Chinese growth.

Indeed, the December 2014 survey of property prices in Chinese cities confirm this theme of "less awful" as we see a deceleration in the pace of decline when compared to November.

A glance at market based indicators tell a similar story of improvement and turnaround. The chart below shows the stock markets of China and its major Asian trading partners. While the parabolic rise of Shanghai is likely a localized anomaly, much like the latest news about the authorities clamping down on margin account. The action in Shanghai can therefore be ignored. On the other hand, the stock markets all of the other Asian trading partner countries, with the exception of South Korea, are turning up.

As China has been the major marginal consumer of commodities, commodity prices is another way of gauging the trajectory of Chinese economic growth. Despite the concerns expressed by many market commentators that cratering commodity prices stems from a lack of demand, which denotes slowing global growth, there are signs of hope. First of all, the above chart shows that the Australian stock market seems to be making a bottom, despite tanking commodity prices as measured by the CRB Index.

The chart below of commodity prices and commodity-related indicators tell a constructive story of global growth. The top panel shows the prices of industrial metals, which are highly sensitive to global growth, compared to emerging market equities (EEM). EM equities have been very correlated to industrial metals, but the latest divergence where EEM has rallied while industrial metals have continued to tank may be an indication that the de-financialization of commodities is beginning. Observant investors will recall that a number of institutions committed funds to commodities several years ago on the basis that their returns were uncorrelated to other assets. This led to the financialization of commodities as an asset class and pushed up prices. As commodity prices have begun to falter, we may be seeing the start of a de-financialization trend where investors exit their commodity investments.

The bottom panel of the AUDCAD exchange rate is another sign that Chinese growth may not be as bad as many people expect. Both Australia and Canada are major commodity producing countries, but Australian exports are more sensitive to Chinese growth while Canada is more sensitive to US growth. The rally in the AUDCAD is another indirect indication of a turnaround in the Chinese growth outlook.

A resurgent US consumer?
In the US, the Fed is getting ready to raise interest rates in 2015 as economic growth has begun to liftoff. Despite the dismal retail sales number announced last week, there are signs that the American consumer is ready to drive the economy.

First the blowout UMich Sentiment figures can be thought of an antidote to the retail sales miss. Gallup produces a rolling survey series of economic confidence and employment. As the chart below shows, economic confidence shot up in December just as energy prices tanked.

The chart below shows the relative performance of Consumer Discretionary stocks to the market, as well as the relative performance of industries that are sensitive to consumer spending. Taken together, these form a picture of a resurgent/ consumer and continued economic growth.

US equities oversold
In the meantime, many of my short-term indicators show that the risk/reward is skewed to the upside. As an example, this chart from show the 50-day highs-lows. This indicator is highly oversold and it is at or below levels where the market has bottomed out in the past (circles are mine), though there is no guarantee that the market can`t get more oversold from current levels.

My option based sentiment indicators are also flashing bottoming signals. In the past, the following combination has been reliable indicators of short-term market bottoms:
  1. An inverted VIX term structure (VIX/VXV ratio more than 1)
  2. VIX more than 20
  3. NYSE TRIN more than 2
The chart below shows past signals in the last three years marked with vertical lines. Last week, we saw two of the three signals and may have seen the third, namely TRIN more than 2.

While TRIN did not close above 2 at any time during the last week, this hourly chart shows that TRIN did move above 2 on an intra-day basis on Wednesday.

Brett Steenbarger also made the following observation before the market open on Wednesday, when Tuesday`s closing levels on the SPX was about the same as Friday`s close (emphasis added):
I have mentioned in the past the Stock Spotter site of John Ehlers and Ric Way. They make use of cycle analysis to generate buy and sell signals for individual stocks and ETFs. Notably, they publish their track record of signals and have done well overall. I note that, as of yesterday's close, they had 138 buy signals on stocks. Since late 2013, when I began tracking the service, there have only been five occasions in which we've had 100 or more buy signals. It's a small sample, to be sure, but all five occasions were higher in SPY five trading sessions later, by an average of 1.59%.

Indeed, when we divide the buy signals by quartiles and look at the highest quartile (most buy signals), we see an average next three-day gain in SPY of +.33%. The lowest quartile (least buy signals) yields an average next three-day gain in SPY of +.06%. This is clearly not how John and Ric designed the service to be used, but I do find it interesting that the broad market has tended to perform best when individual stocks are giving the greatest number of buy signals.
In a more recent post, Steenbarger analyzed his trading indicators and concluded that stocks are ripe for a short-term bounce (emphasis added):
The top chart depicts a ten-day moving average of all stocks across exchanges making three-month new highs vs. lows. That breadth has been deteriorating since late October, but notice also that this week's price lows saw fewer shares making fresh net new lows than at the mid-December bottom. For that reason, I'm viewing the market as a range one defined by the December highs and lows.

The middle chart takes a look at the balance between buying pressure (upticks) versus selling pressure (downticks) across all NYSE shares. Note the recent intensity of selling pressure ahead of price lows, followed by Friday's buying surge. This is a pattern that has been common at intermediate market lows and is consistent with the range perspective noted above.

Finally, in the bottom chart we see the 10-day average of changes in shares outstanding for the SPY ETF. This has been an excellent sentiment gauge, as we have tended to see expansions in shares outstanding when traders have been bullish and contractions in shares outstanding when traders and investors have been bearish. We finished 2014 with considerable bullishness and recently have swung to the opposite extreme, again consistent with the range notion.

Stars are lining up for a rally
In short, the stars are lining up for a short-term stock market rally. The deflationary storm spooking the market is abating, which should be supporting of the risk-on trade.

I would caution, however, that the skies aren't all clear. Forward consensus EPS continue to fall, even as in this young Earnings Season, the EPS beat rate last week was 84% and sales beat rate was 60%. Here is the chart from John Butters of Factset (annotations in red are mine):

As the chart shows, past episodes of weakness in forward EPS have been associated with market weakness. While I fully expect forward EPS to start rising again as Earning Season progresses, largely because of greater optimism from the American consumer, it hasn`t happened yet.

My inner investor is therefore neutrally positioned with an asset mix equal to his policy position mix of stocks and bonds. My inner trader remains bullish on stocks.

Disclosure: Long SPXL, TNA

Tuesday, January 13, 2015

Decision tine in the eurozone

It's decision time in the eurozone, at a whole host of levels. From a technical viewpoint, the Euro STOXX 50 is forming a wedge and I am watching for the direction of the breakout.

ECJ decision
The ECJ will render a decision tomorrow on the ECB`s OMT program, which could have an impact on the proposed QE program likely to be announced next week. This Bloomberg report puts the decision in context:
European Central Bank President Mario Draghi will get a legal readout tomorrow on a predecessor to the quantitative easing plan that he’s set to reveal later this month.

An adviser to the EU Court of Justice will say whether the European Central Bank’s Outright Monetary Transactions program overstepped the law in a non-binding opinion that may signal whether QE must also be reined in.

A negative opinion “would make the ECB’s life much tougher,” Carsten Brzeski, chief economist at ING-DiBa in Frankfurt, wrote in a note to clients yesterday. “It would be welcome grist to the mills of Germans’ opposition against QE. In our view, given the political and economic sensitivity of the Court’s verdict, an outright condemnation of the legality of OMT is highly unlikely.”

To QE or not QE?
As well, we have the ECB Governing Council meeting next week on January 22, where they are expected to announce some form of QE-lite (via CNBC):
The European Central Bank (ECB) could be ready to announce a quantitative easing (QE) program based on the contributions made from national central banks, a source close to the central bank has told CNBC.

The source said that the central bank is planning to design a sovereign debt purchase program based on the paid-in capital contributions made by euro zone central banks.

Every national central bank pays a certain amount of capital into the ECB. For example Germany pays in 17.9 percent of the total contributions, while France contributes 14.2 percent. Cyrpus, meanwhile, pays the least with 0.15 percent of the total.

Greece is the word
Shortly after the ECB meeting, Greece will hold its election on January 25. While Syriza is still leading in the polls, it is unclear whether they will be able to form a government, according to Marc Chandler. Needless to say, such a development would be hugely market bullish:
In Greece, the latest polls continue to show Syriza winning a plurality of votes at the January 25 election. Prime Minister Samaras’ New Democracy has not been able to close the 2-3 percentage point lead. Even with the 50 bonus seats given to the party that gets the most votes, it appears Syriza may not secure a majority in parliament. If it fails to do so after three days, the party with the second highest vote count will get a chance. New Democracy may be able to do so, avoiding another election.
I would note that the SPX is already forming a wedge formation similar to the Euro STOXX 50. Given the kinds of deflationary winds emanating from Europe, it may be eurozone developments that drive the next direction in the stock market, rather than Earnings Season in the US.

So instead of just watching US stocks, savvy investors and traders might be better served by watching Europe.

Monday, January 12, 2015

Some outside-the-box thinking on US employment

Here at Humble Student of the Markets, we like to look for outside of the box thinking about the markets and the economy. In view of some of the continuing angst about last Friday's NFP report, I present some unconventional thinking about the US employment situation.

Lack of wage growth
As a single data point, I would tend to agree with Tim Duy that the -0.2% hourly wage growth reported for December was a blip. Longer term, however, Justin Lahart suggested slowing wage growth is related to the lack of aging in the average worker.

Falling labor participation rate
In addition, there has been much concern expressed over the falling labor participation rate.

David Kelly at JP Morgan Asset Management cited the usual reasons, namely aging Baby Boomers, the propensity of the younger cohort to stay in school longer, more people on disability and more discouraged workers. Then he added a unique take of his own:
The criminal record problem. A large, unexplained part of the decline in participation may stem from the growing number of Americans with a criminal record., The percentage of the U.S. population with a criminal record rose from 13% in 1991 to 22% in 2012 (2) — meaning that nearly one in four Americans has a criminal record. As many employers conduct background checks on prospective employees, the possession of a criminal record could remove many workers from the labor force, effectively permanently.
I am not necessarily endorsing any of these views, but these creative approaches are to applauded. Each of these explanations are intriguing but more work is needed to validate each of these views.

Sunday, January 11, 2015

A consideration of the bear case

Trend Model signal summary
Trend Model signal: Neutral
Trading model: Bullish (upgrade)

The Trend Model is an asset allocation model used by my inner investor. The trading component of the Trend Model keys on changes in direction in the Trend Model - and it is used by my inner trader. The actual historical (not back-tested) buy and sell signals of the trading component of the Trend Model are shown in the chart below:

Update schedule: I generally update Trend Model readings on my blog on weekends and tweet any changes during the week at @humblestudent. In addition, I have been trading an account based on the signals of the Trend Model. The last report card of that account can be found here.

A tactical buy signal
I wrote last week that market "moves have tended to be sharp and brief". I didn't realize how sharp or brief until the trading model reversed itself and moved from bearish to bullish last week. In the wake of the strong stock market rally, I moved to cash on Thursday:

Tactically, the SPX is behaving much in the same way as it did in the last two V-shaped bottoms. The 5-day RSI moved from an oversold reading to neutral (top panel) and the VIX Index moved from above 20, which is an oversold reading, to below (bottom panel). As the market closed on Friday, the index is sitting right at its 50 day moving average, which should act as support for a rally to test new highs in the next week or two.

In addition, I am seeing a reliable tactical buy signal from the VIX term structure, which are shown by the vertical lines in the chart below. In the past two years, whenever the term structure has inverted and the VIX-VXV ratio falls below 1.0 and then further falls below 0.92, that has been a very consistent buy signal with little downside.

Steve Burns summed up the short-term technical picture perfectly this way:
$SPY broke out above the 50 day sma on Thursday: In the past 4 years after $SPY crossed over the 50 day, 5 days later there were 21 wins/8 losses Average win 1.6% average loss -1.33%.

$SPY bounced and regained the 50 day by the end of day on Friday...

It is normal the majority of the time for price to pull back to a major moving average at least once before a new trend takes off.
In addition, the CNN Fear and Greed Index is tilted towards a fear reading, indicating that downside risk is relatively low.

As Earnings Seasons kicks off next week, I would expect the market to respond positively to better corporate guidance from lower energy prices, both in the form of lower input costs and higher consumer spending. Currently, forward estimates are falling largely because the negatives of lower oil prices are obvious, but the benefits are more diffuse. 

Worrisome signs of a top
Even though a bear market is not part of my base case scenario, good investors should always be aware of the risks to their forecasts and understand what could go wrong. It is with that in mind that I would like to explore a bear case scenario for US stocks in 2015.

From a technical viewpoint, the most worrisome sign for US equities are signs of diminishing momentum. The chart below shows the 20 year monthly history of the Wilshire 5000 with the MACD histogram in the bottom panel. In the last 20 years, there have been four times when the monthly MACD histogram has dipped into negative territory. Three of those instances have marked the start of a bear phase for the stock market. 

While the month isn't over yet, MACD is marginally negative for the Wilshire 5000. Is this a warning sign that a bear phase may be about to start?

Here is my interpretation: MACD is a momentum indicator. When it dips into negative territory, it indicates a loss of price momentum. In plain English, risk appetite is waning - and that`s a red flag.

As confirmation of this investment theme, the chart below shows the SPX to US long Treasury bond price ratio as a measure of risk appetite. When the ratio (middle panel) is rising, stocks are outperforming bonds, indicating rising risk appetite; and when it is falling, the reverse is true. As the bottom panel of this chart shows, MACD already turned negative in late 2014.

Another way of thinking about the loss of momentum for economically sensitive assets is to look at the oil to gold ratio. Both oil and gold are commodities and therefore have inflation protection qualities, but oil is more sensitive to global growth. The chart below shows the oil to gold ratio (in grey) compared to the SPX to long Treasury bond price ratio (in green). These two ratios have been correlated to each other in the last 20 years. What does the plunge in the oil to gold ratio telling us? Is this just a special case of excess supply of oil, or is the drop also indicative of slowing global demand, which would negatively affect the growth outlook?

The above chart suggests that the fall in oil is attributable to more than just supply issues. James Hamilton ran a regression of WTI to copper, USD and 10-year bond yield. He concluded:
That is, of the $55 drop in the price of oil since the start of July, about $24, or 44%, seems attributable to broader demand factors rather than anything specific happening to the oil market. That’s almost the same percentage as when I performed the calculation using data that we had available a month ago.

So what’s been happening on the supply side of oil markets is important. But so is what’s been happening on the demand side.
Mark Hulbert recently wrote a column warning that risk appetite is at excessive levels that puts the stock market at risk:
The risk of a major bear market in stocks is now higher than it’s been in years.

That, at least, is the message of an indicator that keys off the amount of risk incurred by the top-performing advisers. On average, the top 10 finishers in the 2014 Hulbert Financial Digest performance scoreboard recommended portfolios that were more than three times riskier than the stock market itself.

That is one of the highest levels to have emerged from the Hulbert Financial Digest’s more than three decades of performance tracking. In calendar 2011, in contrast, the comparable risk level among the top 10 finishers was less than a quarter as much. It’s been growing more or less steadily ever since.
What makes this trend so alarming is that the stock market has been near a major top whenever the top performers’ risk levels were at or close to current levels. In 2006, for example, the last calendar year prior to the 2007-2009 bear market, it rose to slightly higher than current levels: 3.85 times riskier than the market versus last year's 3.32 times.

In 1999, the last calendar year prior to the bursting of the bubble, the comparable level was 2.58.
In other words, top advisors have to take on more and more risk to outperform and be in the top 10. That works, as long as you are in a bull market, but current levels are historically excessive, indicating market euphoria.

Bottom line: These indicators are telling a story of waning risk appetite and possibly falling global growth, which could be negative for stock prices.

What causes a bear market?
While I use a lot of technical tools, I am not an investor who only reads the charts without regard to fundamentals. Despite the warnings sounded by the charts that I have shown, the case for a significant bear market needs fundamental support in the form of a credible bearish scenario.

I outline the causes for bear markets, from the most serious to the least serious from a market impact viewpoint, which is an enhancement of David Rosenberg's list:
  1. Recession: Recessions cause earnings growth to fall and stock prices to tank..
  2. Tight money: Past examples occurred in 1982, when the Fed tightness engineered a recession to cool off inflationary expectations, 1987, when the Fed tightened into a market crash, and 2000, when the Fed started to withdraw liquidity after flooding the system with money in anticipation of Y2K.
  3. Financial crisis: Think Russia Crisis (1998), which was a sharp but brief bear, and Subprime/Lehman (2007-2008), which led to a global recession (see 1).
We also have to consider two factors that exacerbate the magnitude of bear markets. Those are the presence of excessive leverage and high valuation. The NASDAQ bubble that ended in 2000 was financed mainly by equity with little leverage and therefore the ensuing economic damage was relatively contained. By contrast, the 2008 downturn was financed by excessive leverage caused by real estate speculation. That result was a global financial crisis that the world is still trying to recover from, seven years later.

Similarly, unreal market valuations can cause a minor market crack to cascade into a major bear market. In 1987, the Fed tightened into a market crash. In 2000, the withdrawal of Y2K liquidity may have been the spark that ended the NASDAQ bubble.

No recession in sight
Fortunately, we can rule out the possibility of a recession. New Deal Democrat has done excellent work in this area. In addition, Jeff Miller watches a number of pundits and indicators, none of which are showing a recession risk. The US economy remains in the mid-cycle phase of an expansion.

Tight money: It`s all about earnings growth
Tight money, which is one of Rosenberg's causes of bear markets, is a concern as the Fed prepares to hike interest rates in mid-2015. In the past, however, the first rate hike has not sparked bear markets as the negative effects of rising interest rates have been offset by the positive effects of an acceleration in economic growth.

In the current instance, the market cannot expect much downside support from valuation. Multiples are elevated, though not stupidly high as they were in 1987 or 2000. However, James Paulsen of Wells Capital Management makes a case that valuations are dangerously high. Here is the chart of trailing PE, which shows my interpretation that market valuations are elevated but not excessive:

Instead of analyzing the capitalization weighted PE ratio, Paulsen showed that the median PE of US stocks is at a record level:

The median PB ratio is also highly stretched:

Unlike the Tech Bubble, where the excessive valuation was concentrated in only a few sectors, Paulsen warned that stock are expensive across the board, which is a more insidious form of over-valuation:
Is the current widespread valuation extreme more dangerous than a concentrated extreme simply because concentrated extremes tend to be more obvious and eye-catching? A concentrated valuation extreme tends to loudly announce itself whereas a broad-based valuation extreme seems more stealth and, therefore, perhaps more dangerous.
Paulsen's analysis of historically high PE ratios is confirmed by Goldman Sachs, who analyzed forward PE ratios instead of trailing PEs. The team at Goldman Sachs indicated that current SPX forward PE of 16.7 is high by historical standards, but the median forward PE of 17.6 is even higher as it in the top 2% of history from 1976.

I would temper those alarmist remarks by pointing out that both inflation and interest rates are very low compared to the study period shown by Paulsen and Goldman Sachs. While market multiples aren`t low by historical standards, stocks can continue to rise as long as earnings continue to rise. The analysis from Wells Capital Management showed that the market was expensive in 2005 across the board as well, but the market did not peak several years later. The grey bars depict the PE valuation percentiles (a low percentile is expensive compared to its own history) in 2005 and the black bars in 2014.

Here is Paulsen`s caveat:
The 2005 stock market also exhibited excessive valuations across all P/E percentiles. However, the stock market did not suffer any major decline until almost three years later (and from higher levels) in 2008.
In the end, earnings matter as much as PE multiples. I am therefore watching carefully how forward EPS estimates change in the near future. This chart from John Butters at Factset (annotations in red are mine) show that past pauses in EPS growth have coincided with either corrections or bear markets. I believe that forward EPS should get revised upwards soon because of the boost from lower oil prices, but we need to see how that evolves as I am cognizant of the fact that US large caps are vulnerable to the negative effects of a rising USD (see my recent post Focus on small caps in 2015):

Financial dislocation: Always a wildcard
The third cause of a bear market has been a financial dislocation somewhere in the world sparking a financial crisis. Not all financial crisis spread and cause bear markets. For example, the market shrugged off the Dubai default of 2010, which occurred during a period when the financial system was still frail from the Lehman Crisis of 2008.

The most likely sources of financial crisis and contagion are China and Europe. China has always been a wildcard, largely because its economy is so opaque. I believe that China risks are relatively low in the medium term as the current regime has undertaken to enact reforms. Should the Chinese economy really tank, Beijing have also shown the flexibility to stimulate and kick the can down the road. That strategy works as long as the road is long - and it seems to be at the moment.

The bigger risk of financial crisis and contagion comes from Europe, but the source is not as obvious as the headlines show. While the Greece-Germany-ECB tug of war will continue to be a source of drama for months to come, it is the usual European theatre and noise that can be safely ignored. The parents fight and the kids hear everything, but in the end they still want to stay married.

The biggest source of worry is the British election scheduled in May. As I pointed out before (see 2015: Bullish skies with periods of volatility), that election has the potential to shatter the British social compact and render the UK ungovernable. Here is the warning from Anatole Kaletsky (emphasis added):
Britain could become literally ungovernable after the election, with no single party or coalition of parties able to form a majority government. Current public opinion polls predict that neither the Conservatives nor the Labour Party will win enough seats to form a majority government — even in a coalition with Liberal Democrats.

Conservative-Liberal and Labour-Liberal majorities may both prove arithmetically impossible because of the rise of previously insignificant fringe parties. The Scottish Nationalists look able to boost their six seats in Parliament to anything between 20 and 50, largely at Labour’s expense. The United Kingdom Independence Party is threatening dozens of Conservative incumbents. Meanwhile, the Liberals are almost certain to lose about half their 56-seat representation. As a result, a ruling coalition may have to include not just two parties but three or four, including fringe nationalist groups.

The Scottish National Party is sure to demand another Scottish independence referendum as its price for supporting a coalition, while the UK Independent Party will likely insist on Britain’s withdrawal from the European Union. It is hard to imagine either Labour or Conservatives agreeing to such terms.

This means that a government may have to be formed without a majority in Parliament. While minority governments are quite common in continental Europe, the British Parliament has only once failed to produce a government majority — during a brief interlude in 1974 under Harold Wilson. It created seismic upheavals in Britain’s adversarial politics.
The presence of fringe parties like the Scottish National Party or UKIP in the government will raise the specter of another Scottish referendum, or worse, the risk of Brexit from the EU. While the political commitment of the European elite is very much pro-Europe, there is no such consensus from the British.

The mere whiff of the possibility of Brexit would put into question of not just the euro, but the European Union. Such political instability in a major trading bloc would raise risk premiums around the world. Moreover, this scenario is not on the radar screen of most market analysts and strategists and would be a major shock to the financial system.

Should such a scenario unfold, the good news is that the bear market is likely to be short and sharp, much like the Russia and LTCM Crisis or the Crash of 1987, rather than long and drawn out like a recessionary bear market like 1980-82. For buy-and-hold investors who can keep their nerve and maintain their investment, that would be good news. It means that if you blink and ignore the downturn, you would barely feel its effects.

My inner investor and trader diverge
In light of this analysis, my inner investor is more cautious than he has been in the past. He has moved back from his aggressive position from an overweight position in equities to a more balanced portfolio between stocks and bonds. He is watching forward EPS estimates carefully as Earnings Seasons progresses to see if there is earnings support for US equities. In addition, he is monitoring the polls in Greece and the UK to see how the political situation develops in Europe over the next few months.

By contrast, my inner trader is seeing numerous technical signs that the brief market hiccup is over. He has covered his shorts and gone long the market, but with tight stops.

Disclosure: Long SPXL, TNA