Trader’s Narrative

The NYSE cumulative breadth (daily advance decline) is showing signs of life. with its first victory over a previous high:

Referring to this, recently StockCharts blog wrote:

This show of relative strength in the AD Line reflects broad participation in the current advance and bodes well for the current uptrend.

Sounds good but what they are missing is that the cumulative advance decline line is notoriously deceptive. Although the NYSE started out as an exchange where the biggest and best US companies traded, over the years there has been a significant shift away from operating company common stocks to new and strange securities like municipal bonds ETFs, ADRs, bonds (yes, actual bonds trade on the NYSE!), etc.

At first these non-operating company securities were a small portion of the whole but over time, they have come to take such a large portion of the trading that the breadth numbers from the NYSE should be discarded. The only other option to ignoring the data is to painstakingly filter out these ‘pollutants’. There are very few services that provide such data, Lowry Research being one of them.

To see the wide discrepancy, you need only compare the NYSE cumulative breadth chart to the corresponding Nasdaq chart:

The difference between the two becomes especially large when interest rate changes. This is because most of the non-operating company securities are interest rate sensitive (like the municipal bond funds) and they move as a herd either up or down in reaction to different interest rate environments, skewing the NYSE breadth.

If we step back and look at a very long term chart of cumulative breadth for both the NYSE and Nasdaq, it becomes clear that this indicator isn’t really helpful at all. This is why, instead of the cumulative measure, I prefer to look at a simple moving average of advance decline numbers. This indicator, in contrast, is useful for both the Nasdaq and NYSE exchanges by highlighting both kinds of extremes in the market.

Limited Time Free Access to Elliott Wave
Only a few days left! Download the FREE 120 page report from Elliott Wave covering the US, European and Asian markets as well as interest rates, commodities, currencies and much more. This is the most recent edition of their comprehensive Global Market Perspective and is exactly what their regular paying clients receive. But it is only available free for just a few days. There’s no obligation to purchase anything and you only need your email. I’ll go over it shortly on the blog but get your copy while you still can.

This is a tough market to navigate and the cross currents make things extra slippery. Perhaps it would be better to take a step back from the superficial index numbers and levels and look at the internal market structure. With that in mind, here is a chart for the New High New Low index for the Nasdaq:

It is a lesser known indicator but basically it compares the number of stocks going up and down. You can get more background information here, in case you’re unfamiliar with the new high new low index.

It is surprising to see that this indicator has been stuck consistently below 60 for the longest duration of time since the 2002 bear market. As the chart shows, this indicator stays low when we are going through a bear market and high when we are going through a bull market (as in the green box in 2003).

By the way, I’m showing a chart for the Nasdaq as opposed to the NYSE because the ever increasing number of non-common stock securities traded on that exchange has made most internal analysis of it fairly worthless.

In any case the chart above confirms what most of us know on an intuitive level: it feels like a bear market out there. Doesn’t it? Well, even though technically we aren’t in one. That doesn’t seem to matter. The tape is sluggish and the advance decline numbers also bear that out:

Last Friday’s weak market took us down to negative 1788 - extreme low advance decline breadth levels which we last saw in the swing lows of March (green arrow). But even so, it takes more than just one spike like that to wring out the market and set it up for a rally.

And in case you are wondering why I use the raw advance decline data, instead of the cumulative breadth data like most, it is because cumulative breadth is misleading.

For a strong and healthy bull market, either the advance decline must remain above such lows or when it does fall so low, it is preferable for a bouquet of extreme low readings which serves to flush out the weak hands (green boxes).

If you look closely, you can see what I mean about non-common stock securities ruining NYSE advance decline data. If you were only going by NYSE data, in June 2007, you would have mistakenly thought that the market was oversold.

So we seem to be in a bit of a limbo - not quite a bear market, not quite a bull market. Or maybe I’m projecting my own ambivalence onto the charts. What do you think?

The financial stocks are not doing too well right now. I’m just telling you in case you’ve been hiding under a rock or have been floating in sensory deprivation tank for the last whee bit.

It seems everyone is looking at the poor financials and noting how weak they are relative to the market. The shibboleth is then trotted out that we need the financial stocks for a healthy bull market. I’m sure you’ve heard or read this multiple times.

I caught myself repeating it in my previous post (see above link). But rather than accept it at face value, lets put it to the test.

Does a bull market truly need the leadership or participation of the financial sector?

To begin, here is the chart of the Philadelphia Banking Index (BKX) compared to the S&P 500 Index (SPX):

Remember, this is a relative chart. When the financial sector is doing better than the general stock market, the line trends up, and when they are weaker than the general market, the line goes down. And while the ratio showed incredible volatility between 1998 - 2003, the BKX was relatively unchanged, simply treading water the whole time.

Now lets see how the market actually behaved during the times that the financial stocks were leading and during the times that they were weak:

The bull market didn’t start in 1995 but it certainly did intensify with a sharp upturn in its slope. While the S&P 500 continued to rally - with intermittent corrections - until 2000, the Philadelphia Banking Index (BKX) only kept up its leadership till early 1998.

From then to the S&P 500 market top (otherwise referred to as the “bubble top”) financial stocks actually performed weaker than the general market. It certainly didn’t faze the bull market though.

Bull or Bear, Banks Don’t Care
As the tone changed and a bear market took hold, the financial stocks re-awakened and took leadership once again. They actually performed better than the S&P 500 as the bear market raged on. In retrospect, they were a decent hiding place. You didn’t make much money on an absolute level but you didn’t lose money either. If you had borrowed a tactic from a hedge fund playbook, I suppose you could have made money going long financial and short the S&P 500.

Finally, as the bear market subsided and a new bull market was born, the financial sector lost its luster and started to underperform again. Little at first but more recently, at a torrid pace.

I know this is a short slice of market history but even from such cursory analysis it seems that there isn’t much stock in the common belief that financial stocks need to lead a rally. Nor that they need to perform better than the general market for us to enjoy a bull market. In fact, there is no real relationship that I can discern. If you see one, please edumacate me.

I believe there are conditions that precede bull markets - this just isn’t one of them.

Sacred cows make the best burgers and this one is of kobe proportions. If your blood-lust is not satiated, let the intellectual slaughter continue by checking out how cumulative breadth can be misleading.

Last week I suggested that market breadth doesn’t matter, until it does. By which I meant that inspecting every twitch of the cumulative breadth measure for the market isn’t all that useful.

Most of the time, this indicator is brought up because there is a “negative divergence” which then is used to argue that the market is floating on air and will come crashing (or correcting) down because not enough constituents are supporting its rise.

As I mentioned, the problem with this logic is that for the most part, the Nasdaq cumulative breadth has been in perpetual free fall:

The only time this indicator was able to mount a feeble come back was in 2003. And even then, it didn’t last long. While the market continued to rise, the cumulative breadth soon fell and broke through the low set in early 2003.

To see the recent graph of Nasdaq cumulative breadth, check out the link above.

The long term chart of the NYSE cumulative breadth is even more enigmatic. From 1995 to 1998 it rose along with the S&P 500. Then it decoupled and became it’s mirror opposite until the bear market bottom in 2003. And since then it has again, walked in agreement with the market index.

To anyone who proposes the theory of “negative/positive divergence”, I would ask, when should I have bought or sold? and why?

For example, should I have sold in the spring of 1998? and missed the massive run up to 2000? should I have bought in early 2000 because cumulative breadth was turning up and breaking the downtrend? wouldn’t that have resulted in massive losses?

Cumulative breadth simply doesn’t provide any sort of actionable insight. Unless I’m missing something huge. In which case, someone please forgive my elephantine ignorance and rescue me from myself.

Market breadth is what goes on inside the stock market. Most people pay attention to price, like the Dow or S&P 500 index. Market breadth looks at the number of stocks that are advancing or declining within an index or an exchange. It is a great way to measure the “health” of the market. After all, if the majority of securities on an exchange are falling, we can’t expect it to keep rising, right?

Or can we?

Every once in a while the bears point to the “negative divergence” in the Nasdaq index and the Nasdaq cumulative breadth. They get worked up over the fact that market breadth does not correspond to the market price. Here is the recent Nasdaq breadth, showing a waterfall decline, in contrast to the Nasdaq Composite index:

It sure looks ominous. Once you zoom out though, you realize that there’s something seriously wrong with this way of looking at the market.

NASDAQ Cumulative Breadth
Just for kicks, let’s go back to 1998. From there, the Nasdaq cumulative breadth fell consistently until October 2002. That’s right. Even though the Nasdaq was screaming higher, then topping out in early 2000, its breadth just barreled down paying it no attention.

Breadth continued falling until it made a sort of double bottom in early 2003, just as the Nasdaq was ending its bubble bear market. The recovery in breadth was short lived because it again started to fall in early 2004 and has been falling consistently since!

So it is obvious from this slice of history that Nasdaq breadth and the Nasdaq composite are completely decoupled. In fact, if we go back further in time we see that breadth has been falling continuously since, well, since I have data for it.

NYSE Cumulative Breadth
The other broad measure of breadth, for the securities on the NYSE, is not all that different. From a top in 1998 it fell continuously until early 2000. For the rest of the year it stabilized and in late 2000 started to rise. NYSE breadth found a top in May 2002. Yes, you read that right! As the market was going to hell in a hand basket, breadth was rising! For some reason, it decided to not rise in 2002 - I guess in sympathy to the stock market. But then in early 2003 as the market was rising, so did NYSE breadth. And it has continued to rise to this day.

My point is that we have so many positive and negative divergences between breadth and indices they purport to represent that cumulative breadth is basically useless.

I agree that a trend simply can not continue if less and less securities are participating in it. Eventually it exhausts itself and crumbles under its own weight as it becomes unsustainable.

The problem is that no one knows when, exactly, this will take place. And cumulative breadth certainly provides no insight whatsoever into this.

This is why I prefer taking a much simpler measure of breadth: the moving average of net advancers and decliners:

The above chart is the 30 day moving average but you can use any number you like, as long as it doesn’t introduce too much lag into the equation. It gives you not only timely signals, but it also pinpoints most, if not all, intermediate bottoms with ease.

I’ll show the long term charts of cumulative breadth for both Nasdaq and NYSE in an upcoming post. It really is eye opening.