Friday, May 22, 2015

What apples-to-apples market breadth is telling us

In light of the recent weakness of the DJ Transports, there has been a greater focus on the idea of market breadth and what it tells us. I have heard a lot about how the Transports don`t represent the US economy or market in the same way when Charles Dow first devised the Dow Theory. When you boil it all down, the divergence between the DJIA and DJTA represents a form of breadth divergence.

The best representation of market breadth can be explained this way. When an army is advancing, you want to know if only the generals are leading the charge or if the whole army is moving forward. In market terms, the generals represent the heavyweights of the market, while the army represent the broader market. A market rally on positive breadth is said to be supported by the broad market. By contrast, a rally on narrow breadth is led only by the leadership of the heavyweights and such advances are interpreted with greater skepticism.

In the past, common breadth measures have included the NYSE Advance-Decline Line or NYSE Composite. In more modern times, one problem with this approach is that the underlying components of NYSE-listed stocks do not include the more growth oriented NASDAQ stocks. In addition, the NYSE Composite has a number of closed-end funds and REITs which may not be representative of the broad market (aka the army).

An apples-to-apples breadth metric
What is needed is an apples-to-apples comparison of market breadth.

Enter the SP 500 A-D line and Equal weighted SP 500 as a breadth benchmark for the SPX. These measures do not suffer from the shortfalls of the NYSE A-D Line or NYSE Composite because they measure the breadth of the same stock universe.

Each is slightly different. The A-D Line is a diffusion index. A single stock advancing will have the same impact on the A-D Line regardless of whether the magnitude of the advance is 0.1% or 10%. By contrast, a 0.1% advance in a stock will affect the Equal-weighted SP 500 differently than a 10% advance. The difference between the equal-weighted and float-weighted SP 500 is that the movements in heavyweights such as AAPL will affect the float-weighted index far more than the smallest stocks in the index.

The chart below shows the 10-year record of the SP 500 and the A-D Line and Equal-weighted SP 500. In order to graphically exaggerate divergences, I have graphed the ratios of the A-D Line to the SP 500 (in green) and Equal-weighted SP 500 to the float weighted SP 500 (in red).

The bottom panel of the chart shows the rolling 52-week correlation of each of the indices to the SPX. As the correlation analysis shows, the A-D Line ratio is highly correlated to the SPX, which makes spotting divergences difficult. The correlation of the returns of the equal to float weighted ratio is relatively high and positive, but fluctuates. This makes the task of spotting divergences a little easier.

These indicators did a relatively good job of confirming stock market trends over the last 10 years. They were able to give early warnings of a negative divergence ahead of the market top in 2007; they confirmed the powerful rebound at the bottom in 2009; and gave early warnings of market weakness and subsequent rally in 2012. No indicators are perfect, however. The equal-float weighted ratio gave a false signal and turned bullish prematurely in 2008 and it was unable to spot the market rally in 2011.

A market distribution warning
What are they telling us now about market internals?

As the chart below shows, but of these indicators are showing negative breadth divergences. The equal-float weighted ratio has been falling while the stock market has rallied in the last two months. During the same period, the A-D Line ratio has been flat, which is another sign of the lack of broad market participation in the current rally. In addition, the bottom panel shows a negative divergence in On Balance Volume, which is an additional warning of distribution.

In light of these negative breadth divergences, it makes the recent divergence between the DJIA and DJTA create concerns for stock prices (see How worried should you be about the weak DJ Transports). As the chart below shows, the DJIA rallied to new all-time highs last week, while the DJTA weakened. Its breach of a technically important support zone is further confirmation of the divergence. These conditions set up a Dow Theory non-confirmation of the new highs in the DJIA.

Taken into a broad-based context of widespread weakness in market breadth, all of these signs are worrisome for the near-term health of the current bull run.

Tuesday, May 19, 2015

The Fed's Magical Mystery Tour

What's going on at the Fed? Notorious dove, Charles Evans of the Chicago Fed, gave a speech in Sweden. In his prepared remarks, he said that he was in no hurry to raise rates:
To give you the punch line, I think the outlook for growth in economic activity and the labor market is good. However, inflation is too low, and it has been too low for the last 6 years. Moreover, my forecast is for inflation to rise at a very gradual pace, reaching our 2 percent objective only in 2018. Based on this forecast, and the risks to the outlook, I think the FOMC should refrain from raising the federal funds rate (our traditional short-term interest rate policy tool) until there is much greater confidence that inflation one or two years ahead will be at our 2 percent target. I see no compelling reason for us to be in a hurry to tighten financial conditions until then.
If it were up to him, he would like raise rates in early 2016 [emphasis added]:
In my view, it likely will not be appropriate to begin raising the fed funds rate until sometime in early 2016. Economic activity appears to be on a solid, sustainable growth path, which, on its own, would support a rate hike soon. However, the weak first-quarter data do give me pause, and I would like to see confirmation that they are indeed a transitory aberration. Furthermore, and most important, inflation is low and is expected to remain low for some time.
Glad we know how you feel, Charles.

Then came the surprise. Reuters reported that he said that a June liftoff was on the table if all the stars were to line up:
Evans, who in his speech argued for rates to start rising in early 2016, told reporters if the FOMC had confidence that inflation was going to move up and that first quarter economic softness was temporary, "you could imagine a case being made for a rate increase in June".

"I think we are going to go meeting-by-meeting to make that decision," Evans, a voter this year on Fed policy and among the most dovish of U.S. central bankers, said after taking part in a panel debate.
WTF! What happened to raising rates in early 2016?

For a confirmed dove like Charles Evans to make a remark like that, he must have been prepped. It is highly likely to have been part of a coordinated Fedspeak campaign to prepare the markets for an interest rate hike - call it the Fed's Magical Mystery Tour as Fed officials fan out around the globe and trumpet their message. Indeed, the WSJ had reported that John Williams of the San Francisco Fed had warned that a rate hike was on the table at every meeting and they weren't going to telegraph any further guidance. Fed Chair Janet Yellen warned about the risks from excessive equity valuation and to high yield bonds if rates were to rise (see Bulls shouldn`t expect help from the Fed).

Fed Presidents like Charles Evans don't speak off the cuff. In an interview after his Fed Chair appointment had expired, Paul Volcker quipped when he went out to dine at a restaurant, he felt compelled to say, "I'll have the steak but that doesn't mean I don't like the chicken or the lobster."

The Fed seems to be trying to nudge the markets to prepare for greater policy tightness and rate normalization. The latest BoAML Fund Manager Survey indicates that the consensus for first liftoff is either 3Q or 4Q, with 3Q being the most likely. My base case scenario therefore calls for a September rate hike, largely because the Fed seems to prefer to guide and nudge markets, rather than to surprise them.

With the SPX at record highs, the market may need further Fed "nudging". Will there be further surprise guidance contained in the FOMC minutes to be released on Wednesday?

Stay tuned!

Monday, May 18, 2015

Why I am bearish (and what would change my mind)

Now that SPX has reached further all-time highs, I received a number of comments to my latest weekly market outlook post (see Where's the new high celebration?) which amounted to "you've been bearish and wrong for the past few weeks and now it seems that you are stubbornly making up reasons to stay that way". Under the circumstances, I feel compelled to respond and explain.

How I got bearish
To explain how I became bearish, follow me on my market research journey in the last few months. In January, I observed that the market environment was becoming difficult for trend following models like mine (see All washed up!). The market had become choppy, which was problematical for identifying a short-term trend:
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.
Since I wrote those words, the US equity market averages continued to do the stutter step above and below the 50 dma and behave in a choppy fashion.

In April, a comprehensive study by James Paulsen of Wells Capital Management identified the current environment as market adherence to the long-term trend as overbought (my words, not his). Paulsen did the rolling 36-month regressions of stock prices and charted the R-squared of the regressions (the higher the R-squared, the tighter the fit). He found that current levels of R-squared is in the top decile of fit, which is consistent with the straight up stock market without a 10% correction that has been observed.

Paulsen then hypothesized that current market conditions had created excessive investor complacency and such periods have tended to not ended well. Consider the return statistics when R-squared is in the top decile (leftmost bar) in the charts below:

At about the same time, Brett Steenbarger had another take on the current market environment. He observed that while the long-term trend remained intact, short-term trends, which he called momentum, was misbehaving:
Overall, chasing new highs and stopping out of long positions on expansions of new lows has brought subnormal returns. We have had a trending environment since 2012, but not a momentum environment. Understanding that distinction has been crucial to stock market returns.
Using Paulsen study methodological framework, I did some more research and compared the 36-month R-squared, or trend, with a shorter 6-month R-squared, or trend. I found that the two had diverged considerably, which confirms the Steenbarger comment (see How to make your first loss your best loss).

Such episodes have tended to resolve themselves in a bear phase, largely because a weakening short-term trend combined with a strong long-term trend is indicative of weakening momentum. In fact, current readings are similar to conditions observed just before the Crashes of 1929 and 1987, though I am not forecasting a market crash as this model is better at forecasting direction than magnitude.
Nevertheless, based on the current 36-month to 6-month R-squared spread of 0.844, I looked at what the return pattern of the DJIA was during past episodes with similar characteristics. The sample size was a more reasonable 16, compared to the minuscule N=4 in the SP 500 study that went back to 1950. The market outperformed initially, but rolled over at between 3-6 months after the first time the spread went above 0.8 (which was March 2015).
And if the trend got even more extended and the 36 to 6 month spread went to 0.9? The results were more dramatic, as the market declined almost immediately.

My research showed that, based on monthly data going back to 1900, such episodes of trend divergence have tended to resolve themselves in market downturns. That is how I came to the opinion that the next major move in stock prices is likely to be down. As the study was based on monthly data, recent market action amounts to mere squiggles.

Another way of depicting the long and short term trend divergence is through the use of MACD. As the monthly chart of the SPX below shows, the MACD histogram has gone negative, indicating a loss of price momentum. Every past instance in the last 20 years has either seen stock prices either be in a bear phase, as measured by the 12 month moving average, when MACD turned negative, or resolved itself into a bear move soon afterwards.

A downturn every time on when MACD turned negative

What would make me bullish?
Like every investor, I've been wrong before. In order to change my assessment of the intermediate term trend for stock prices, I need to see definitive signs of improvement in the short-term trend. I am watching for improvement several of the following signs in order to turn more bullish:

  • MACD divergence improvement: It doesn't necessarily have to go positive, but some signs that it is flattening out and starting to rise would help.
  • Stop the chop with better momentum: Now that the SPX has broken out to new highs, I would like to see some follow-through indicating that positive momentum has re-asserted itself. One useful sign would be a series of "good" overbought readings where the market gets overbought on indicators like RSI and stays overbought.

Macro and fundamental
  • An improvement in macro outlook: The Citigroup US Economic Surprise Index has been mired in highly negative territory, indicating a preponderance of economic misses compared to beats. Doug Short's Big Four Recession Indicators are looking a little wobbly and New Deal democrat has been calling for a mild industrial recession, though the consumer sector remains healthy. So is it too much to ask for some signs of improvement in Citigroup ESI?

  • A consistent record of positive EPS estimate revisions. If the economy starts to improve, then the Fed is likely to raise rates, which would hold back stock market gains from PE expansion. The negative effects of a flat to falling PE can be offset by robust EPS growth. From a valuation viewpoint, it will be up to EPS growth that does most of the heavy lifting in pushing stock prices upward at this point of the economic cycle.
Like all investors, I have been wrong before. Admittedly, such periods of negative performance creates valuable "scar tissue" that makes us all better investors - as long as we are willing to learn from our mistakes. For now,  I remain cautious on stocks, but "data dependent".

Sunday, May 17, 2015

Where's the new high celebration?

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

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.

An odd all-time high
Imagine going to a wedding where virtually no one is celebrating. The guests are whispering in hushed tones. What could be wrong? Is the groom flirting with the bridesmaids? Did the bride run off with an old boyfriend (much like the final scene in The Graduate)? That what the stock market felt like last week.

The SPX and OEX reached all-time highs, but market internals were unenthusiastic with exhibitions of positive momentum. As the chart below shows, RSI(14) remained in a tight range and behaved much like it had in the past few weeks when the market oscillated in a tight band. The % bullish metric flashed a negative divergence and % above the 50 day moving average was also range bound.

Other US stock market indices were not confirming the new highs either. The NASDAQ Composite, Russell 2000, the DJIA and DJTA were all below their highs, though the Dow was very close (see my post How worried should you be about the weak DJ Transports?). Moreover, all of the averages shown in the chart below are exhibiting a disturbing sign of having breached an uptrend indicating a loss of momentum. I was informed by another technical analyst that I had turned bearish too early, as such formations often resolve themselves with a sideways consolidation and sometimes a rally to a marginal new high before rolling over.

Looking abroad, SPX strength hasn't been confirmed by the European averages either. Both European averages also display the same characteristic breach of uptrend lines seen in the US averages.

To be sure, not all inter-market analysis is painting a bearish picture. Despite the weakness in Europe, the technical conditions of the Greater China stock markets are mostly healthy. But then if US equities have to depend on Chinese stimulus as a source of global growth, we would be indeed be scraping the bottom of the barrel.

Score breadth, momentum and inter-market analysis as bearish.

Sentiment model readings mixed
I saw a number of bulls get very excited when NAAIM exposure showed a crowded short reading, which is contrarian bullish, and AAII sentiment survey showed that the percentage of AAII bulls had fallen to a two-year low (see analysis from Bespoke).

There is no doubt that the NAAIM exposure index is a bullish sign, but the AAII results have to seen in context as the high level of neutral opinions pushed the AAII bull to bear spread to an overall neutral reading. While I am not a big fan of opinion surveys as opinions can move on a dime, surveys where people put real money on the line are better indications of investor opinion. My own interpretation of the AAII sample and Rydex data are pointing to neutral sentiment readings.

Other indirect indicators of sentiment from option data indicate that the market is complacent and showing no fear. The chart below of the term structure of the VIX, the equity-only put-call ratio and the VIX Index itself, where high readings indicate fear and low readings indicate greed, are all showing complacent readings.

Moreover, the all three versions of ISE call-put ratio (total, equity-only and index and ETF) moved above the 200% level on Thursday, which is a rare occurrence.  Though the sample size is small and there was one important exception at the 2009 market low, such occasions have not been bullish signs in the past.

Dana Lyons looked at the ISE index and ETF data and came to a similar conclusion:

Score sentiment models as mixed to slightly bearish.

The growth scare continue
The US macro outlook is shaping up to be the "mild industrial recession" postulated by New Deal democrat based on his review of high frequency economic data:
The bottom line is, the US economy is in a shallow industrial recession. It is not due to the weather, nor to the West Coast ports strike. Rather, it is driven by a 16% appreciation of the US$ globally, and secondarily by the effects of the collapse of commodity prices on raw materials producers. At the moment, weakness in consumer spending in the Oil patch is outweighing consumer strength elsewhere.
With the exception of initial claims, all of last week's major macro releases missed expectations, such as retail sales, industrial production, consumer sentiment. The latest update of Doug Short's Big Four Recession Indicators is looking a little shaky. Of the four, Employment has consistently been positive< Industrial production was negative. Retail sales came in flat, but we have to wait for the inflation adjustment to see how negative it is. The last, real income, was negative last month.

The latest update from the Atlanta Fed's GDPNow nowcast of GDP growth is an anemic 0.7%, which is getting close to stall speed:
The GDPNow model forecast for real GDP growth (seasonally adjusted annual rate) in the second quarter of 2015 was 0.7 percent on May 13, down slightly from 0.8 percent on May 5. The nowcast for second-quarter real consumer spending growth ticked down 0.1 percentage point to 2.6 percent following this morning's retail sales report from the U.S. Census Bureau.
Incidentally, Deutsche Bank took a look at the forecast record of this simple mechanical model and it's been pretty good (via Barry Ritholz):

We may be seeing the worst of all worlds. We have a sputtering US economy and a Fed preparing to raise rates (see Bulls shouldn't expect help from the Fed), which would not only dampen growth but push the USD up and create earnings headwinds for companies.

The US shouldn't expect much help from overseas. Even though the latest data shows that European growth is recovering, US growth is faltering and Chinese growth is decelerating. Gavekal has suggested that global sales growth may have peaked for this cycle.

Commodity prices, which have been good indicators of global growth, may be confirming that conclusion. While commodity prices appear to have bottomed in the last few weeks, much of the strength is attributable to USD weakness. A glance at industrial metals, a key input to the global growth engine, in EUR and CAD leads to a different conclusion.

Score the macro conditions as being a bit on the weak side.

The weight of the evidence
To be sure, not all indicators are pointing down. Here are a number of signs that can be interpreted bullishly:
  • Great China stock markets are supportive of growth. While I tend to discount the gyrations of the Chinese equity market because of its retail orientation and casino mentality, the message from HK, Korea and Taiwan is more stimulus from Beijing (see previous comment above).
  • NAAIM exposure came in at a historically low level, which is contrarian bullish. The NAAIM exposure index shows what RIA managers are doing with their assets and it has been a terrific contrarian indicator.
  • Retail sales may not be that bad. David Rosenberg pointed out that some components of personal expenditures no accounted for in retail sales are growing strongly, such as air travel,theme park spending and car leasing.
  • NFIB small business optimism rose and beat expectations. This is a contrary sign of a tanking economy.
  • Forward EPS revisions have become more constructive. The latest update from Ed Yardeni shows that forward 12-month estimates are rising for large caps and small caps, but falling for mid caps, compared with falling estimates for small and mid caps in the previous week. EPS estimates for the large cap SP 500 have risen for two consecutive weeks. Could the growth scare be over?
  • The economic weakness may prompt the Fed to delay the interest rate normalization process. A recent Jon Hilsenrath WSJ article indicates that the Fed would like to raise rates and begin the interest rate normalization process, but the soft patch in the economy is likely to push out the date of the first liftoff. According to the WSJ, the current consensus is for the Fed to first raise rates in September, but we are likely to get greater clarity on this issue when FOMC minutes are released on Wednesday.

Despite the smattering of bullish indicators, the weight of the evidence is intermediate term bearish. Technically, breadth, momentum and sentiment models are tilted bearishly. The economy remains weak and the Fed is about to begin a tightening cycle. When it does, rising rates will be USD bullish, which will create headwinds for corporate earnings.

The week ahead
This review is much too long and it`s time to wrap up. In summary, the new high shown by the SPX last week is unconvincing. Market action continues to be suggestive of a wimpy bull-wimpy bear range bound market. Short-term indicators like this one from show that readings are nearing levels where the market has sputtered in the past few weeks. Indeed, this chart of net 20-day highs - lows rolled over on Friday, even as the index made a marginal high.

My inner investor remains cautious, but not overly panicked. My inner trader remains short, Should a decline materialize next week, he will be watching for signs of negative momentum of prolonged oversold conditions. Otherwise, his base case is a brief market swoon, to be followed by the usual short rally - until the real bear episode begins.

Disclosure: Long SPXU,SQQQ

Saturday, May 16, 2015

How worried should you be about the weak DJ Transports?

Last week, I pointed out the negative divergence between the Dow Jones Industrials Averages and the Transportation Average. The latter was not only weak, but it was testing a key support level. Such an event was a typical non-confirmation of the DJIA bull trend.

The observation prompted push back in the blogosphere and on Twitter. A typical example came from Josh Brown, who said that the Transports don't matter much anymore as the character of the stock market had changed since Charles Dow first wrote about the Dow Theory.

However, Nautilus Research found that negative divergences of the SPX with the Transports were bearish, though the sample size was very small:

Mark Hulbert more or less said the same thing:
How bearish is this divergence? To come up with an answer, Jack Schannep recently focused on periods over the past 25 years that included big divergences. Schannep is the editor of a market-timing advisory service called

Schannep found 14 such instances. In nine of them, he says, the broad market subsequently dropped by less than 10%. But in the remaining five cases, the stock market’s eventual decline averaged 25.7%.

Those are sobering odds. If we average all 14 instances of prior divergences similar to the current one, the market eventually fell more than the 10% threshold for a correction. If that turns out to be the case this time around, it would be the first correction since 2011.

Even more ominous is that in five of the 14 cases, or more than a third of the total, the divergences presaged a full-scale bear market. In fact, Schannep points out that when the broad market hit its bull-market highs in 1990, 1998, 2000 and 2007, the Dow Transports in each case had already turned down several months before.
Notwithstanding the bearish analysis from Nautilis and Schannep, I would philosophically inclined to agree with the naysayers about the importance of the Transports. No indicator is perfect, particularly an indicator as old as the DJIA-DJTA link. Technical indicators have to be viewed as part of a mosaic and the analyst has to see what the big picture is telling him.

The comparison of the DJTA with the DJIA is a well-known technique of looking for breadth confirmation of a move. So why all the hate? Good technical analysts need to consider the big picture on indicators like breadth. Even as the SPX achieved new highs on Thursday at Friday, indicators like RSI and % above the 50 dma remain range bound, indicating a lack of momentum. In addition, % on point and figure buys is in a downtrend showing a negative divergence.

Weakness in the Transports is only one data point. On the other hand, when you look at the big picture on breadth and momentum, does how much confidence would you have about the all-time highs achieved by the SPX?

Thursday, May 14, 2015

What the COT data really tells us about the stock market

Business Insider featured a chart from UBS indicating that speculators were in a crowded short position in the SP 500 e-mini contract:
It's been almost a year since traders were betting against the SP 500 like this.

According to data from UBS, as of last week, traders continued to add to short positions on SP 500 E-mini futures, meaning these folks were betting that stock prices would fall.
By implication, traders are in a crowded short and the stock market is likely to rise. The chart from UBS is shown below. I have annotated past "crowded short" readings in red and "crowded long" readings in green. Can anyone seriously tell me that they want to use this Commitment of Traders (COT) data to trade the stock market? First of all, the CRTC has three broad classifications for traders, namely hedgers, or commercials, large speculators (institutions and hedge funds) and small speculators (mostly individuals). It is unclear whether the "traders" referred to in the UBS report are large speculators, small speculators, or both.

I used to analyze COT data in the past and I found that SP 500 COT data was useless for calling market direction. However, large speculator and leveraged position (read: hedge funds) positions on the NASDAQ 100 (NQ) contract was a reasonable contarian indicator. The most likely reason is that the fast money uses NQ positions to make market direction bets as the NASDAQ 100 is perceived to be a high beta index.

Here is the NASDAQ 100 e-mini COT chart from

Currently, large speculators and hedge funds have started to unwind what was a crowded long position. As this data is reported weekly and with a lag, it is unclear how the COT data has shifted as the SPX has achieved an all-time high.

The moral of this story? Know how well your model works before jumping to conclusions.

Tuesday, May 12, 2015

Bulls shouldn't expect help from the Fed

I received more than the usual amount of comments in response to my weekend post about unemployment claims (see A scary thought about US employment). With unemployment claims falling to multi-decade lows, I rhetorically asked if this is as good as it gets.

Not mentioned in my post is the close correlation between weekly claims (blue, inverted scale) and stock prices (red).

The real subtext is how the Fed interprets the employment numbers. At what point does the employment picture compel the Fed to act and start tightening?

A steady tightening path
I am seeing more and more Fedspeak and other signals indicating that the stock market should not expect any help from the Fed. On Monday, the WSJ reported that San Francisco Fed President Williams stated that the markets can't count on any further Fed signals about an interest rate hike, as it could happen at any time:
Federal Reserve Bank of San Francisco President John Williams told a television channel Monday that the U.S. central bank is unlikely to provide much warning ahead of an increase in short-term interest rates.

Instead, the official told CNBC that officials will need to go into every policy meetings with an open mind. “We should be coming together every six weeks, discussing what the outlook looks like, and what the right appropriate policy decisions at that meeting are, and adjusting policy” accordingly, Mr. Williams said. The Fed needs “to get out of this business of telegraphing our decisions in advance” and make sure that monetary policy is truly driven by and responsive to incoming economic data.

Mr. Williams said rate rises are “on the table at every meeting” but he would not suggest action is likely to happen at any particular point, although he did say a year from now he believes short-term rates will have moved up off of their current near zero levels.
Telegraphing intentions, especially when the decisions are "data dependent", would create excess volatility:
“You don’t want to make a decision three months in advance and announce that decision when you really have more time to collect data and make the most informed decision you can,” Mr. Williams said.

The official also noted that he accepts that a lack of firm guidance from central bankers could generate more volatility in financial markets.

“We don’t want to be adding noise,” but at the same time, “we don’t want to be absolutely eliminating all uncertainty about our future actions. It’s healthy for the future actions to be uncertain because future conditions can change,” Mr. Williams said.
In a subsequent CNBC interview, Williams argued for an earlier preemptive rate hike to fight inflation, so that further hikes are either not necessary or delayed:
The Federal Reserve hiking interest rates "a bit earlier" allows the U.S. central bank to increase rates more gradually, a top Federal Reserve official said Tuesday.

Inflation shooting above the Fed's 2 percent target would force a more "dramatic" rate hike, San Francisco Federal Reserve President John Williams said Tuesday in prepared remarks before the Harvard Club of New York. Market watchers have eyed indications of when the Fed may abandon its near-zero interest rate policy.

Preparing the market
In addition, Janet Yellen's remarks last week were generally interpreted as preparing the markets for a rate hike (via Reuters):
I would highlight that equity market valuations at this point generally are quite high. There are potential dangers there...

We’ve also seen the compression of spreads on high-yield debt, which certainly looks like a reach for yield type of behavior...

When the Fed decides it’s time to begin raising rates, these term premiums could move up and we could see a sharp jump in long-term rates. So we’re trying to ... communicate as clearly about our monetary policy so we don’t take markets by surprise.
In other words, better let the markets get a little unsettled now and adjust to the idea of higher rates now, than to have it surprised and crash later.

New York Fed President William Dudley more or less said the same thing in a speech delivered on Tuesday, which appears to be part of a coordinated Fedspeak campaign to prepare the markets for interest rate normalizaton.
To be as direct as possible: I don’t know when this will occur. The timing of lift-off will depend on how the economic outlook evolves. Since the economic outlook is uncertain, this means the timing of liftoff must also be uncertain.
In other words, the Fed isn`t going to give us any more signals because they don`t know. But then, they`ve has told us repeatedly what needs to happen for rates to rise, what happens next shouldn`t be a big surprise:
At the same time, though, I can be clear about what conditions are needed for normalization to begin. If the improvement in the U.S. labor market continues and the FOMC is “reasonably confident” that inflation will move back to our 2 percent objective over the medium-term, then it would be appropriate to begin to normalize interest rates.

Because the conditions necessary for liftoff are well-specified, market participants should be able to think right along with policymakers, adjusting their views about the prospects for normalization in response to the incoming data. This implies that liftoff should not be a big surprise when it finally occurs, which should help mitigate the degree of market turbulence engendered by lift-off.
We should be able to make an educated guess as to liftoff timing. Indeed, Matt Busigin modeled wage growth based JOLTS data and the latest release is pointing to rising wages, which is a key metric of inflationary pressure watched by the Yellen Fed.

What normalization looks like
A glance at money supply growth, which has had a relatively high level of correlation with stock prices, show that both MZM and M2 growth has been decelerating, which is indicative of the normalization process.

Here is my key takeaway from all the Fedspeak. Rates are going to rise this year. When that happens, don't expect a Yellen Put. Don't expect the Plunge Protection Team cavalry to appear at the crest of the hill should the stock market start to fall.

Is is any wonder that the bond markets are acting up?

Sunday, May 10, 2015

What I learned from Bundegeddon

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

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.

Signs of distribution
I have been warning about the signs of distribution in the US equity market for several weeks and those signs are continuing. I wrote about how the intersection of a strong long-term trend and a faltering short-term trend was a sign that complacency may be getting its comeuppance (see 3 secrets from the Book of (Trend Following) Revelations and How to make your first loss your best loss). One graphical way of illustrating this effect is to see how the MACD histogram, which is a measure of long and short term trend divergence, has gone negative on the monthly SPX chart. Every past instance of this condition in the last 20 years (marked by vertical lines) has seen the market go down, either with the MACD as a leading, coincidental or lagging indicator.

A shorter term chart of the SPX is also showing disturbing signs of distribution. The SPX has broken a key uptrend that began last October and appears to be tracing out a rounding top formation. In addition, we are seeing a negative divergence in On Balance Volume, which is a volume-based measure of accumulation and distribution.

In addition, Urban Carmel tweeted the following observation during the day on Friday:

Also see Andrew Nyquist's posts about negative breadth divergences, which is another sign of distribution.

A wobbly growth outlook
The rally on Friday was sparked by the combination of a surprise Conservative majority win in the UK election and a not-too-hot-not-too-cold Employment Report, which the market interpreted as the Fed would stay on hold for a little longer. Despite the positives seen in the Employment Report, New Deal democrat cautioned that he was still seeing signs of a mild industrial recession based on his read of the leading indicator portions of the report:
But the leading portions of the employment report are sounding an alarm. The manufacturing workweek was down. Overtime was down. Unemployment from zero to 5 weeks increased significantly. Revisions to prior months were negative. This is the third month of this negative trend in the leading indicators in the employment report. In my "Weekly Indicators" column, for the last month I have said that the US is in a shallow manufacturing recession. Today's report confirms that.
As well, rising bond yields from the mini Bundegeddon panic last week (more on that later) is a negative for housing. Bond yields directly affect mortgage rates, another component of the long leading indicators that New Deal democrat monitors on a regular basis. In his latest weekly summary of high frequency economic releases, he confirmed his assessment of the risks to the US economy:
In the last several months the theme has been poor coincident indicators with generally positive long and short leading indicators, reflecting a shallow industrial recession due to the strong dollar and oil patch weakness, but a resilient domestic, consumer economy. The big addition this week was the move in interest rates. Should interest rates spike much more, they will begin to drag down housing again, and that will leave money supply as the only positive long leading indicator, as corporate profits have already turned down.
There were other calls highlighting the dark cloud overhanging the silver lining of the Employment Report. Greg Ip of the WSJ pointed to signs of a slowing economy:
Look closer and the jobs data still sends a worrying signal about the economy’s underlying strength. Because little of the data in the payroll survey goes directly into the expenditure-based measure of GDP, it can serve as a useful “check” on that broader economy-wide number. According to Chris Varvares of Macroeconomic Advisers, total hours worked is a good predictor of output both in the current quarter and the next.

The good news, then, is that total hours worked per week in the private sector, a measure that captures both the number of people on the job and how many hours they work, rose 0.2% in April from March. The bad news is that the level of hours is still lower than in February. As the nearby chart shows, hours over the last three months rose at a tepid 1.3% annual rate compared to the prior three months. Hours showed a similar dip a year earlier but began to recover by March.

The latest update of Doug Short's Big 4 Recession Indicators are now in for March. The composite score has now been negative for two consecutive months, with two of the four component in the red for the month of March. Weakness in industrial production and real income dragged down the overall score, compared to an extremely negative Real Sales print in February, which rebounded strongly in March. These figures are arguably backward looking and the story of 1Q weakness is well known, but they nevertheless paint the picture of an economy that is near stall speed.

Gavyn Davies also warned of a global growth slowdown:

Overall, the growth rate of the global economy has therefore slowed further, according to our models. Our estimate of activity growth in the major advanced economies plus China, which we use as a proxy for global activity, has dropped to 3.0 per cent at the end of April, from 3.7 per cent a month ago. This measure of global activity has now broken below the roughly 4 per cent rate that had been established since mid 2014.
According to Davies, the odds of a strong US expansion are falling and recession risk is starting to tick up, though readings are nowhere near panic button levels.

For the stock market, it all boils to the earnings growth outlook, which remains clouded. Ed Yardeni reported that while forward EPS estimates for the large cap S+P 500 have resumed their upward climb, the mid cap S+P 400 and small cap S+P 600 are still seeing downward EPS revisions. In a separate blog post, he openly fretted about the combination of stretched valuation and downward estimate revisions:
Although the dollar might have peaked on March 13 for a while, and the price of crude oil might have bottomed on January 13 for a while, industry analysts who cover the S+P 500 are still lowering their earnings estimates for both 2015 and 2016. They now expect $119.02 and $134.18 per share for this year and next year, down 5.9% and 5.2% from their estimates at the end of last year. For this year, they’ve been lowering their Q2-Q4 estimates as earnings have beaten expectations during Q1 with a 4.7% “hook-up” move so far over the past two weeks, which is typical during earnings seasons.

It’s getting harder to find much if any GAARP (growth at a reasonable price) in the US given the latest lofty valuation ranking for the S+P 500 sectors: Energy (28.8 vs. 14.2 year ago), Consumer Staples (19.7, 17.4), Consumer Discretionary (19.0, 17.4), Health Care (17.8, 16.2), Materials (17.2, 16.7), S&P 500 (17.2, 15.3), Utilities (16.4, 16.1), IT (16.2, 14.4), Industrials (16.2, 16.3), Financials (13.9, 13.5), and Telecom Services (13.4, 13.1).
Incidentally, the latest update from John Butters of Factset on Earnings Season shows that, with 447 companies having reported and Earnings Seasons nearly over, the EPS beat rate came in at a respectable 71%, but the revenue beat rate was an abysmal 45%. Corporate management have learned over the years to play the EPS expectations game, but the revenue beat rate does not pain a picture of robust growth.

Lessons learned from "Bundegeddon"
My main takeaway from the signs of distribution is that technical analysis is telling us that stock prices are poised to fall. We just need a trigger, whose cause is unknown.

The surprise selloff in Bunds last week is instructive as to how a bearish trigger can come out of nowhere (a longer summary here). In hindsight, all the signs were there. European bond yields had gone negative in a substantial portion of the front end of the yield curve. While those conditions left many people scratching their heads, fast money was pouring into eurozone sovereign bonds, and Bunds in particular, as a way to front-run the ECB QE program. We then saw a disorderly unwind sparked by the inflationary fears and exacerbated by the lack of liquidity in the bond market. Joseph Cotterill of FT Alphaville made an apt analogy to NYC rooftop bars at closing time to describe to situation:
New York rooftop bars are rather fine things. They especially seem so after one or two (or three) martinis. The trouble comes at closing time. A small elevator has gradually taken sober patrons up 40 or so floors throughout the evening. Suddenly everyone is three sheets to the wind, and they all want to leave at once.
The Bund selloff induced comparisons with the 2003 spike in JGB yields.

The risks lurking in the bond market
By the end of last week, the mini-panic appeared to be over and Bund yields had retraced part of their move. Will they continue to fall next week or is this just a pause? I have no idea, but this rush for the exits in "risk-free" bonds is a lesson for stock market bulls. Thing can turn south a lot faster than you think.

What I do know is that bond yields are extraordinarily low compared to its 600 year history (via Business Insider):

Fed Chair Janet Yellen warned last week about the risks to stock prices should long bond yields start to tick up:
I would highlight that equity market valuations at this point generally are quite high. There are potential dangers there...

We’ve also seen the compression of spreads on high-yield debt, which certainly looks like a reach for yield type of behavior...

When the Fed decides it’s time to begin raising rates, these term premiums could move up and we could see a sharp jump in long-term rates. So we’re trying to ... communicate as clearly about our monetary policy so we don’t take markets by surprise.
Incidentally, there are plenty of other warnings about valuation and interest rates. Warren Buffett indicated that stocks look cheap compared to bonds today, they won't look so cheap should interest rates normalize. Recently, longtime market bull Abby Joseph Cohen conceded that stocks are no longer as attractive.

In a recent presentation, Jeff Gundlach also warned about the crowded long in the fixed income market. Investors have been lengthening their duration in reaching for yield and that increases the price sensitivity to interest rate changes in their portfolios.
Credit quality is also falling as another sign of stretching for yield.
Despite the rhetoric about data dependency, the Fed has shown a historical tendency to keep hiking once it first starts to raise rates. When it all unwinds, watch out!
HY spreads also tend to widen dramatically when interest rates rise.
Could the recent Bundegeddon episode be a dress rehearsal for the real event later this year? What happens if we get one or more data points, or Fedspeak, pointing to an imminent rate hike? What if we were to see a Greek default without a Grexit, how would the fixed income markets react?

So far, credit market risk appetite remains healthy, but the behavior of HY bonds will be an important canary in the credit market coalmine.

The stage for a risk-off induced downturn is set, both from a technical and macro viewpoint. None of this means that a bear phase will start tomorrow, or ever. It just means that risks are elevated and we need a catalyst, which can come out of nowhere as the Bund selloff did.

No sentiment support for stock prices
If a market downturn were to occur, current investor sentiment models readings are not supportive of stock prices. The chart below shows the AAII Bear-Bull spread (black) and Rydex bear-bull ratio (green). Readings had moved off a crowded long reading and into neutral. However, they are nowhere near the panic levels seen at major market bottoms.

The NAAIM exposure index show managers to be slightly bullish, but remain in neutral territory.

Market based sentiment metrics, such as the VIX term structure, are still showing a fair degree of complacency (a ratio above 1 shows rising fear). We did see a brief bout of fear last week when the VIX Index rode the its upper Bollinger Band, which was a sign of a "bad oversold" reading where the markets gets oversold and stays oversold.

It would be instructive to observe what happened with these indicators during the market selloff in 2011.

Similarly, the CNN Money Fear and Greed Index is meandering in neutral territory. It has yet to break out either upwards or downwards, reflecting uncertainty about direction.

The week ahead: Top of the range?
Last week, we saw a Bundegeddon induced downleg in stock prices. The market had stabilized on Thursday when I began to short the market. Friday morning saw a rally sparked by the surprise David Cameron majority win in the UK election and the favorable Employment Report, though the GBP rally was cut short by the prospect of a UK-referendum on EU membership under a Tory government.

What`s next? My interpretation is that the stock market is back to the choppy back-and-fill pattern. Upside potential is limited and the risk-reward ratio is tilted to the downside.

As the chart below shows, the SPX had moved to a near overbought reading in the space of two days where it has topped out in the past few weeks (see RSI(5), top panel). The more standard RSI(14) indicator remains tightly range bound and refuses to get either overbought or oversold, but readings are at the top of the current range where the market has staged a retreat. In addition, the SPX is displaying a pattern of lower highs, indicating a loss of momentum.

As well, this chart from IndexIndicators is just one of many examples of the swift reversal seen in breadth and momentum indicators. Upside potential is limited given the recent wimpy-bull and wimpy-bear pattern.

If we were to zoom into an hourly chart of the SPX for a more tactical view, we see a familiar pattern of an overbought reading on RSI(5), where it has struggled to see much upside action in the recent past. We also see the short-term downtrend that I identified early of lower highs. If the market were to rally further, it will have to contend with first resistance at the 2120 level and then further resistance at the all-time highs.

My inner investor remains cautious, but not overly panicked. Markets are vulnerable to a downturn of unknown magnitude, sparked by an unknown catalyst. The Bundegeddon episode taught him that possible bearish triggers could come from anywhere, as a backup in German bond yields was not exactly on his top 5 market risks.

My inner trader has a slight bearish bias. Despite going short a little early on Thursday,he recognizes that this has been a forgiving market - as long as you don`t get panicked out of your long or short position. He remains short and he is waiting for a bearish trigger to signal the start of a downleg in the major equity averages.

Disclosure: Long SPXU, SQQQ