PM Daily Market Commentary - 5/15/2014

davefairtex
By davefairtex on Thu, May 15, 2014 - 9:03pm

Gold closed down -9.20 to 1296.30 on heavy volume; silver was off a larger -0.28 on moderately heavy volume.  Gold was pounded down $11 at 0830 EDT at the time of the jobless claims report - which was positive, along with the Empire State manufacturing survey released at that same time which was even more positive.   Silver followed gold lower, and it was hurt even more, dropping as far as 19.42 before rebounding slightly into the close.

I'd leave it at that, except that the US equity market tipped over and sank soon after a negative Industrial Production report hit at 0915 EDT, which left PM prices largely unaffected.  SPX closed down 18 on some heavy volume.  How can manufacturing and jobless claims be positive, but INDPRO turn negative?  You've got me.  And why does the stock market drop on bad economic news 30 minutes after gold & silver drop on good economic news?

Sometimes markets just want to drop.  Other times - I think they get a little push.  Or a big push, as the case may be.  My article of faith is, it won't change the trend.  But intraday activity is another matter entirely.

The USD climbed higher during London trading, touching 80.40 at one point, until all that good economic news hit at 0830 EDT, at which point the buck started sinking, not stopping until it had retraced all its gains. The buck closed down -0.04 to 80.08.

GDX dropped today, closing off -1.62% on moderate volume, GDXJ was off -1.98% on moderately heavy volume.  Miners were sold today, underperforming the metal.  Buyers seemed scarce until the last 90 minutes of the trading day, when some dip-buyers appeared and the miners rallied modestly into the close.   It was the only bullish note on the day in PM, and it was modest.  GDX is periously close to support, and all those ratios are still pointing lower.  Momentum in the miners remains to the downside.

20 year treasury bonds made yet another new high today, closing up +0.80%, in total up about 13% on the year.  Its enough to turn me into a paper bug.  Seriously, the chart does look strong.  The 10 year yield dropped to 2.5%.  A full 50 basis points is a good-sized move over the past 5 months.  Eventually that should help the housing market, since mortgage rates tend to track the 10 year rate pretty closely.

The Fed is quite happy with these results, I am sure.

Going back to the Industrial Production (INDPRO) number for a moment - this is one of my big four economic indicators that - if they all point in one direction - seem to be good predictors of the US stock market.  While its possible to have one-month drops in INDPRO without things melting down, it is also a warning of stock market weakness ahead.  See how it turned down at the start of 2008?  One-month moves in INDPRO do tend to lead to weakness, but its the multi-month moves lower in INDPRO that are more concerning.

Other indicators (Real Personal Income - up, Nonfarm Payrolls - up, Real Final Sales - neutral) haven't tipped over yet, however, so the macro evidence of a downturn is not conclusive.  And INDPRO is only down this month, so the verdict is - "too soon to tell."

But likely, other traders know these correlations too, so the fact that INDPRO dropped and the market sold off is not all that surprising.

SPX remains above its 50 day MA.  Downside protection is still quite cheap, with the VIX at 13.  Complacency still reigns.

3 Comments

davefairtex's picture
davefairtex
Status: Diamond Member (Offline)
Joined: Sep 3 2008
Posts: 5064
housing starts number

So today there was a big jump in "housing starts" which sounded really exciting to me.  Except the big jump was basically in housing starts for multi-family units, i.e. apartments & condos.  There aren't enough rentals, apparently.  What's more, the numbers they used were non-seasonally-adjusted.

Every spring, there is a big bounce in housing starts, as winter ends and the weather improves.  This is news?  Here's a chart that is seasonally averaged.  What does the trend look like to you for the red line - SFH housing starts: a housing market recovery?

cmartenson's picture
cmartenson
Status: Diamond Member (Offline)
Joined: Jun 7 2007
Posts: 5569
How good are the numbers?

DaveF:

Going back to the Industrial Production (INDPRO) number for a moment - this is one of my big four economic indicators that - if they all point in one direction - seem to be good predictors of the US stock market.  While its possible to have one-month drops in INDPRO without things melting down, it is also a warning of stock market weakness ahead.  See how it turned down at the start of 2008?  One-month moves in INDPRO do tend to lead to weakness, but its the multi-month moves lower in INDPRO that are more concerning.

Going back through time, how good is this number?

I know that at one point they decided to re-classify hamburger assembly as 'manufacturing'

There are serious shifts to the underlying data all the time.  A biggie is coming in 2017 when the so-called 'factory-less manufacturing' (a.k.a. outsourcing) shift happens  which will serve to BOTH reduce imports and increase factory output,' a handy statistical trick if ever there was one.

Globalization Forces Major Change In Business Classification System

Aug 20, 2013

U.S. federal agencies involved in economic data are on the verge of a major and transformative change in the way they classify companies that have outsourced their U.S. production to foreign manufacturing contractors.

The change could radically increase U.S. production statistics by classifying "factoryless goods producers" as domestic manufacturers. Companies like Apple will no longer be considered "wholesale traders," and their sales would be counted as U.S. production, even though none of their manufacturing is in the United States.

Imports by American companies that outsource their production to foreign manufacturers also would no longer be counted as imports, thereby impacting the balance of U.S. international trade accounts.

The idea is for the federal government to determine how much production has been offshored and to pinpoint the number of American companies that are linked to manufacturing, even though they don't make the products they design and sell.

The changes now being finalized by the U.S. government would be implemented in the 2017 North America Industry Classification System when factoryless goods producers will be classified as U.S. producers.

The new classification system of manufacturers would introduce "significant discontinuity" to a wide range of statistics gathered by the government, say those involved.

Given that 'making electricity' in an electrical plant is manufacturing, does not the crazy weather of 2013/14 need to be factored out to get a clear picture of how much actual, value add production is happening?

 

davefairtex's picture
davefairtex
Status: Diamond Member (Offline)
Joined: Sep 3 2008
Posts: 5064
INDPRO - what's inside

Have to be frank, I don't know what's under the covers in INDPRO.  It correlates pretty well with market movements, but that's about all I know.

I have heard mutterings in different news stories that the accounting of the more iconic US-branded manufactured items is undercounted, mainly because (for example) AAPL tends to make the vast bulk of the profit, while the Chinese plants that assemble the things squeak by on thin margins.  I don't know if that's true, but looking at AAPL's margins I can believe it.  AAPL sure seems to have the whip hand.  Should AAPL be counted as a "US Manufacturer?"  AAPL can make Macbook Pros without Flextronics, but the reverse is definitely not true at all.

How this sorts out in all the economic measurements, I have no idea.  Is this the same sort of scam like the "hedonic adjustments" we're all so fond of in CPI, or is this a real change that makes sense? I don't know.

I suppose it depends on what you're trying to measure with the timeseries.  A Macbook Pro is definitely a US product that is a manufactured item, but its also clear that it provides few if any US "manufacturing jobs."

Of course, my goal in watching INDPRO is an attempt to project forward what the US equity market will do.  If they muck around with the timeseries, that will likely confuse things, so I'm glad you're bringing it up.  Best not to count too much on any one timeseries I suppose.

At some point, the market will correct, but I do wonder if we're going to see a replay of 2008, or if it will be a more dramatic break.  In 2008 it was this slow tipping over, with signs visible for months.  INDPRO dropped, total bank credit dropped, NYSE margin loans gave us a six-month warning, Nonfarm Payrolls trailed a bit as an absolute value but its derivative was definitely a concurrent warning sign, and the junk bond spread popped a few months before the top too.

Few of these signs are showing up now.  INDPRO is staggering a bit, NYSE Margin credit as well, but none of the others are signaling anything.  All Quiet on the Macro Front.

Well, minus Hussman's projections of "very low returns for the next 10 years" based on historical valuation metrics, of course.  If I were focused on the 6-12 month timeframe, I'd be following Mr Hussman very closely indeed.  Cash is a position too, as they say.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.
Login or Register to post comments