On Near-Misses via Barry Ritholtz

I found the following article (via Barry Ritholtz's Big Picture blog) poignant and thought-provoking. Enjoy.
--------------------------------------------------------------------------------------------------------

“It is the paradox of the close call. Probability wise, near misses are not successes. They are indicators of near failure. And if the flaw is systemic, it requires only a small twist of fate for the next incident to result in disaster. Rather than celebrating then ignoring close calls, we should be learning from them and doing our very best to prevent their recurrence. But we often don’t. People don’t learn from a near miss, they just say, “It worked, so let’s do it again.” Other studies have shown that the more often someone gets away with risky behavior, the more likely they are to repeat it. Modern disaster prevention can and should be about stopping trouble before it strikes, not cleaning up afterward.”
-Ben Paynter, Wired Magazine

The near miss is a misnomer. As George Carlin once said, what we really mean is a “near hit.” Carlin, whose wonderful use of language made his comedy that much sharper, used to say “A collision is a near miss . . . Boom, they nearly missed — but not quite.”
The markets have had a series of “near misses” over the past few months — assorted single stock flash crashes, trading glitches, even the Facebook IPO were all market structure collisions. The snafu with Knight Trading as they were testing new algo trading execution software was reputed to cost them $10 million per minute — about $167,667 per second. (more here)

--------------------------------------------------------------------------------------------------------
Leaves you wondering if all the near misses we have had lately, be it economic or some computer-induced "glitch" causing the markets to roil, are not a systemic issue.

Traders Edge 8/8/12- The SPY/IWN rejected the high

The market continues its trek upward that began in or around Draghi's insistence that the ECB would do whatever it takes to save the Euro. The SPY closed over the 140 level, corresponding to the 1,400 level on the S&P 500 index, in yesterday's trading. Although the market has been trading up (and may continue to do so over the near-term), I cannot get over that the rally appears to be on fumes.

For one, you will notice that the advance over the last three trading days has occurred on what can only be considered anemic volume. The price on the SPY is also running into swing points with higher volume, suggesting there is no conviction in this move. For instance, yesterday's advance of the SPY occurred on 109.5 million shares. This is running into a swing point on May 1, where 138.8 million shares traded. Today's share count is not enough. In addition, the market looks to have rejected the $141 price level, as the SPY closed near the lows of the day. This may be a sign of a failure, noting that this morning's futures suggest a lower open. One last note, the higher move in the SPY is not being confirmed in the more economically-sensitive small-cap stocks. The IShares Russell 2000 Trust (ticker IWM) remains 7.5% off its $86.39 high logged in April 2011. I also note that the IWM also showed a price failure running into a higher volume swing point (May 3, $80.62, 61.2 million shares vs. 50.1 million shares yesterday)

The portfolios I manage have been up in recent days, due largely to the exposure to gold stocks and coal. I remain largely in cash with a small short position (largely as a hedging tool). I may consider upping my short exposure in the coming days or weeks.

The S&P 500 versus Cyclical/Non-Cyclical Companies Revisted

I really cannot get over how well the price of the Morgan Stanley Cyclical Index relative to the Dow Jones Consumer Non-Cyclical Index has tracked the price and performance of the S&P 500 over the last decade (which I spoke about here and here). That said, I also still cannot get over the feeling there is some statistical aberration or other quality inherent in both indexes, that is not otherwise readily apparent, that results in this relationship.

To that end, I further expanded my research and expanded both the time frame and the available universe. First, I used monthly price and performance figures for a period including the end of 1991 through August 2012. In addition, I expanded the available universe to include all of the stocks (excluding shares of Berkshire Hathaway, as the high share price skews the results) in the Russell 3000, a broad universe of stocks covering everything from the mega-cap to small-cap stocks.

In this study, I defined cyclical and non-cyclical stocks using two different classifications. The first being the overall volatility in sales and the second being the volatility in the stock price, the later being used on the premise that the market will anticipate fundamental inflection points. Using the standard deviation of sales and stock prices, I broke the Russell 3000 into quartiles. Companies with either low volatility of sales or stocks prices were designated as non-cyclical stocks while I designated cyclical companies has having high volatility in either sales or stock prices. For any monthly period, I added the stock prices for cyclical and non-cyclical companies, calculated the relative value relationship, and compared this relative price to the S&P 500. The results are show below.

The first chart derives the relative cyclical/non-cyclical price performance based on the volatility of sales, delineating the Russell 3000 into two separate halves.

The second chart derives the relative cyclical/non-cyclical price performance based on the volatility of sales, with cyclical (non-cyclical) companies defined as the highest (lowest) volatility quartiles. (The field in the below chart is truncated due to the tech bubble pushing the relative relationship to extraordinary highs in late 1999.)

The third chart derives the relative cyclical/non-cyclical price performance based on the volatility of stock prices, delineating the Russell 3000 into two separate halves.

The last chart derives the relative cyclical/non-cyclical price performance based on the volatility of stock prices, with cyclical (non-cyclical) companies defined as the highest (lowest) volatility quartiles.

A few items worth noting.
- The structure of all four relative price charts are very similar.
- That said, the relative price of cyclical to non-cyclical stocks based on sales has more of a flat structure over time (some of this may be due to the scaling) as compared to the relative price chart derived from stock price data.
- I think the stock price derived models are probably more robust, as stock prices capture other important factors- over and above just sales- including earnings, income and balance sheet structures, cash flows, and most important forward-looking expectations. 
- With that said, my comments will be more focused on the stock price derived results.

You should immediately notice that, over the last 20+ years, the relative price of cyclical stocks relative to non-cyclical stocks tends to capture the directional momentum of the market very well, as the relative price tends to move up and down in tandem with the market. Interestingly, the relative price performance peaked ahead of the market tops in both the late 2000's and in late 2007. In addition, the relative price turned up ahead of the market bottom in early 2002. Most interestingly, you see that the cyclical stocks never recovered their value, relative to non-cyclical stocks, since the market bottomed in early 2009. The same dynamics, although less pronounced, is seen in the sales-derived models. One particular aspect I find noteworthy is that following the market bottom in 2009, the sales derived model that is based on the top and bottom quartiles of sales volatility showed cyclical stocks increasing in value relative to non-cyclical stocks from the market bottom in 2009 through February of 2011. Since February 2001, this model has not been confirming the market rally, but has fallen off. This corresponds with the results seen in the study using the Morgan Stanley Cyclical and the Dow Jones Consumer Non-Cyclical Indexes. It also fits the pattern seen in the weekly figures from the Economic Cycle Research Institute (ECRI), the firm who has repeatedly stated their call for a recession.

Getting back to the data, despite the market approaching highs again, cyclical stocks appear to have underperformed their non-cyclical counterparts. Is this a result of a Fed manipulated market where stock prices rise on the hope of monetary stimulus and not real economic growth. I think the answer is largely yes, but more evidence is probably needed.

Clarke and Dawe - Quantitative Easing Too Funny

Cyclical vs. Defensive Indexes- Unprecedented Divergence

I stated in a previous post (http://tradingjournalinvestmentjunky.blogspot.com/2012/08/growth-and-cyclical-companies-not.html) that I wanted to look at a long-term analysis of the S&P 500 versus the relative price of the Morgan Stanley Cyclical Index and the Dow Jones Consumer Non-Cyclical index. To that end, I compiled a longer-term chart that compares these indexes S&P 500 versus the relative price of Morgan Stanley Cyclical Index (ticker ^CYC on Yahoo Finance) and the IShares Dow Jones Consumer Good Index (ticker IYK), which tracks the performance of the Dow Jones Consumer Non-Cyclical Index. I present two charts below: the price trend since 2000 and the year-over-year change for the indexes.

The one thing that really stood out to me is that, except for a period early last decade, the price and the performance of the S&P 500 and the relative price of the Cyclical and Non-Cyclical indexes have never significantly diverged. That is up until now. It may just be an issue of the composition of the indexes or tracking error, but the divergence we are seeing is unprecedented versus the historical relationship for the last 12 years. Very Interesting.

Growth and cyclical companies not participating in the rally

I thought I would highlight a dynamic in market that many may be overlooking. Below, you see the three-year chart of the S&P 500 versus the price the Morgan Stanley Cyclical Index relative to the Dow Jones Non-Cyclical Good and Services Index.

Now, I preface the following comments on the fact that I have not done a significant, long-term analysis of this apparent relationship (which I hope to do soon), but in my mind there is an intuitive appeal to co-movement in these indexes. The Morgan Stanley Cyclical Index tracks the performance of stocks of economically sensitive companies. In contrast, the Dow Jones Non-Cyclical Good and Services Index follows the performance of companies that are less economically leveraged, hence is defensive in nature. A period of rising relative performance in the cyclical index means either continued economic growth or an outlook for gains in the economy. You would only expect that the S&P 500 would rise in this environment. I would also think the opposite would hold true. A decline in the relative performance means falling economic growth (or an outlook there of) and probably suggests a broader market decline.

Since March of this year, the Morgan Stanley Cyclical index has underperformed the the Dow Jones Non-Cyclical index. In my opinion, this highlights investors' concerns with economic growth. It also suggests that the recent market rise is more related to investors bidding up large-defensive stocks, a flight to safety trade if you will, and that the market is rallying on fumes. I also think that without some resumption in growth, or at least an improved growth outlook, that the market is just building cause for a further downside move.

Sunday Link Dump

Seasonal adjustments and Birth/Death Model were THE cause of rising employment figures according to Zero hedge

.....while the quality of jibs is very low

Shiller is saying his index is up due mainly to seasonal factors, bubbles forming in some regions 

However, the Case-Shiller Index shows housing prices have stabilized, and may indicate a housing recovery...

...While Ed Harrison from Creditwritedowns.com does not think this will be the case.

My take is that housing is in a bottoming process (much like a bottoming process following other bubbles) and there will be fits and starts to the recovery. On the positive side, investment home buyers have the ability to be cash-positive. On the negative side, I would suspect a great number of potential home sellers are sitting off the market waiting for prices to rebound. These potential sellers will keep prices low.

According to a post on theTrader, a major sell signal has been tripped

James Bianco via Barry Ritholtz's The Big Picture showing that market may be overvalued when capitalization is compared to GDP- at least as compared to history.

The world industrial sector is in decline

New car sales a disappointing -  this comes as reports of car companies extending credit to sub prime lenders (here, here, and here) to boost car sales. Have we not gone that path before?

Bill Gross on  the death of the cult equities

Coal inventories as a way to look at what is actually going on in the Chinese economy

Time to bone up shadow banking 
      ..... and their influence on money market funds

The pros-cons of Fed Policy

Mapping Municipal Bankruptcies- this could be an interesting app going forward.

Too much to think about while wrapping my head around in the India blackouts- It highlights the infrastructure problems we all face and the massive investment that is needed. Interesting in that most problems with infrastructure are man-made. Also interesting as it shows the structural energy supply deficit that India's economy faces, especially in regards to coal.

U use poore grammer on the job and the jobe hunt, dont

Great quake distributions likely be random through time

Boost your memory

Improve your professional network

Trillion dollar lies