Equities and the Economy:

• Dow loses 1% for first time in 64 sessions.
• Confluence of factors bring in selling.

Yesterday was a rough day for your portfolio or 401K with all the major indexes suffering material losses. The Dow fell 274 points, 1.2%, finishing at 21,750. This is the first time the Dow has fallen at least 1% since May 17th. To put this in perspective, the previous longest streak for the Dow to not lose at least 1% in a day was 69 sessions way back in October 1995. The S&P 500 lost 38 points, 1.5%. ending at 2,430 and the Nasdaq ended off 1.9%, 123 points, at 6,222. It was really a confluence of factors that pushed stocks lower. As I’ve said many times, P/E ratios have been at historical highs, concerns about the Fed tightening monetary policy (even if it is only through shrinking their balance sheet), and heightened concerns about President Trump’s ability to pass a raft of pro-growth policies all added to the bull’s yoke. This was exacerbated by the rumor yesterday that Trump’s economic advisor, Gary Cohn, was resigning. Cohn is considered the glue holding Trump’s pro-growth agenda together and the feeling is if he goes it’s another setback for tax cuts and deregulation. Additionally, and this is from my trading background, we had lost momentum to the upside. That means traders are going sell at the first sign of a decline.

Now you regular readers know I’m a “glass is half full” kind of person. The sky is not falling. Prior to yesterday the Dow has gained 180 points over a 4 day period and finished higher in 14 of the last 18 sessions. Additionally, we’re still close to record highs. The Dow is only 1.7% of its record, the S&P 2.1% and Nasdaq 3.1%. As I’ve said too many times, we’ve had one of the longest bull runs in history. People are spoiled. We need a pullback. It’d be healthy.

On a side note, with riskier assets, which includes stocks, losing value folks fled to safe-have assets, including gold, which finished above $1,300/oz. That’s a 9 month high.

Regarding fundamental economic news, the Fed reported that industrial production rose 0.2% in July. Mining production and utility production rose 0.5% and 1.6%, respectively, which was offset by a 3.5% decline in the auto sector. Once again, further evidence it’s a good time to buy a car. The Conference Board reported its leading economic indicators rose 0.3% in July to 128.3. The Board’s economist said “The LEI improved in July suggesting the U.S. economy may experience further economic activity in the second half of the year.” I’ll take that!

Overnight Japan’s Nikkei 225 closed at its lowest level in 3 months. Stateside stocks remain on the defensive with the Dow down 42 points.

Oil

• Oil prices bounce after 3 days of losses.
• Prices remain range bound.

After 3 days of losses oil prices bounced marginally yesterday with WTI closing up 31¢ at $47.09 and Brent finishing up 76¢ at $51.03. Yesterday’s move was really just a technical move when prices fell below $47 bringing some “chart” traders. At the macro level we have U.S. crude inventories falling but U.S. production climbing and most likely hitting a new record level in September or October. Also, traders are keeping an eye on OPEC and friends compliance to the production cut agreement, which, while still strong at 90%, has been slipping. All that said, we’re still range bound between $45 and $50.

This morning WTI is down 9¢. Yawn.

Courtesy of MDA Information Systems LLC

Natural Gas

• EIA storage report marginally bullish.
• Bakken and Eagle Ford production on the rise.

Yesterday the EIA released its weekly storage report stating 53 Bcf was injected last week. At first blush this was a bearish number for traders were looking for something closer to a 47 Bcf injection. However, drilling into the data the EIA reported that 9 Bcf was reclassified from base gas to working gas. So if you take that 9 out 44 Bcf was injected from actual production. (Base gas is gas needed to maintain pressure in a storage field and is considered not available for withdrawal, i.e. working gas). So once traders did the math they realized the actual injection was 44 Bcf which was below expectations, and they bought it up. A little. At day’s end the September contract closed up 3.9¢ at $2.929. Overall, the trading range is $2.75 to $3.00, which is down from a few months ago when it was $2.90 to $3.15.

Production is rebounding in the Bakken and Eagle Ford. The North Dakota Industrial Commission (which is like a Public Utilities Commission in other states) reported this week that natural gas production in the play topped 1.8 Bcf/d from April to June, a new record. 89% of the production in the Bakken is oil so the gas is associated gas which tells me oil production must be increasing in the Bakken.

Eagle Ford production is also rising. Per S&P Global Platts, production from the area has been on a steep rise over the last month increasing nearly 200 MMcf/d to 1.31 Bcf/d. Eagle Ford production has been the main driver in pushing total Texas gas production up to 17.8 Bcf/d for August, a 317 MMcf/d increase m-o-m.

This morning the weather forecast has shifted a little cooler for the eastern half of the U.S. for the 6-10 day time frame showing below normal temperatures for Ohio and surrounding states which is bringing some bears back in. Natty is down 2.4¢.

Elsewhere

Did you read about the artificial intelligence experiment Facebook did recently? The company connected two artificial intelligence programs, called “chatbots,” and challenged each of them to try and negotiate with the other over a trade attempting to swap hats, balls and books, each of which was given a certain value. The chatbots were instructed to use English and work out how to negotiate between themselves. It also allowed them to improve their negotiating as they went along. The chatbots learned to negotiate in ways that seem very human. For instance, they would pretend to be very interested in a specific item so they could later pretend they were making a big sacrifice in giving it up.

But a strange thing happened during the experiment. The chatbots made changes to the English language developing a language only they could understand, apparently to make it easier for them to communicate. There did appear to be some rules of speech. For example, the chatbots kept stressing their own name which was believed to be part of the negotiations.

Facebook researchers shut down the experiment stating their objective and interest was in having the chatbots talk to people, which did not occur. They said they did not shut down the program because they were afraid of the results or had panicked but because they were looking for the chatbots to behave differently.

Two artificial intelligence computer programs negotiating, learning, evolving, creating their own language. Think through that.