Are Computers Exploiting MLP Investors?

Energy infrastructure roared higher during the first six trading days of January. After four days, the sector had recovered December’s losses. Two days later, it had almost recouped all of 2018. It was a complete reversal of last month, when slumping equity markets dragged pipeline stocks lower.

It’s looking increasingly likely to us that automated trading strategies relying on complex algorithms (“algos”) are at least part of the reason.

Last year MLPs had already been laboring under the weight of serial distribution cuts. For example, AMLP’s payout is down by 34% since 2014. Incidentally, this must be the worst performing passive ETF in history. Since inception in 2010, it has returned 4% compared to its index of 25%. Its tax-impaired structure is part of the reason (see Uncle Sam Helps You Short AMLP).

For the first nine months of last year, crude oil and energy infrastructure moved independently of one another. Investors who painfully recall the 2014-16 energy sector collapse complained that rising crude prices didn’t help, but as the chart shows, they rallied together in the Spring but parted company in late Summer as the oil market started to anticipate the re-imposition of sanctions on Iran. Crude and pipeline stocks are intermittently correlated, because their economic relationship is weak. Crude sometimes drives sentiment, which can quickly change.

Energy Infrastructure Sometimes In Step with Crude Oil

However, when crude dropped sharply following the Administration’s waivers allowing most Iranian exports to continue, energy infrastructure followed. Pipeline company management teams routinely show very limited cash flow sensitivity to commodity prices, and 3Q18 earnings reported in October/November were largely at or ahead of expectations. Nonetheless, an algorithm incorporating the 2014-16 history would expect MLPs to follow crude when it drops sharply, and would act accordingly. Trading systems bet on falling MLPs following crude, and sold.

In late December crude reversed, and trading systems at a minimum are closing out short pipeline positions if not going the other way. Hence the blistering early January recovery. The fundamentals were good in December, and remained so in January. Any change was imperceptible.

This is conjecture. It’s impossible to obtain hard data to support or refute this theory of market activity. So consider our perspective as a money manager in the energy infrastructure sector, in daily contact with clients and potential investors discussing the outlook. In December, callers were frustrated. The apparent disconnect between fundamentals and stock prices was confusing, troubling. What were they overlooking? What were we missing? Many held, but some didn’t. Frustrated at losses they couldn’t explain, having lost faith with repeated bullish analysis, the sector saw more outflows than inflows. Potential buyers noted compelling values, but usually were dissuaded by continued sector weakness. Unable to comprehend the inability of good financial performance to boost prices, many opted to wait. Tax loss selling towards the end of the year exacerbated.

The turn of the calendar coincided with a modest bounce in crude oil, as reports surfaced that the Saudis were sharply reducing crude exports to the U.S. Current prices are creating for them a substantial fiscal gap.

Conversations with clients and prospects have completely turned. Now callers want to know if there will be a pullback. Is the rally for real? Flows have also reversed. One day last week inflows to one of our funds outweighed outflows by 100:1. In December there were no buyers. In early January there are no sellers.

I was prompted to consider events in this light by a recent article in the Financial Times (Volatility: how ‘algos’ changed the rhythm of the market). Philosophically, I’m inclined to believe that automated trading simply does what humans do, just better and more cheaply. However, there is a less benign feature in that algos are also exploiting the inefficiences of humans. Michael Lewis showed this in Flash Boys by examining how high frequency trading systems will see market orders placed and race to snatch the best price before the limit order is executed. This happens in fractions of a second. But humans can also be outwitted over longer periods. December saw the biggest ever monthly outflows from mutual funds, capping an unusual year in which almost every market was down.

There was some bad news. The trade dispute with China is slowing growth there, and S&P500 earnings forecasts are being revised lower. The government is shut down. Fed chairman Powell sounded more hawkish before walking back his comments (see Bond Market Looks Past Fed). But these developments scarcely seem to justify record mutual fund outflows. There have been far worse environments for stocks.

Hedge fund veteran Stanley Druckenmiller said, “These ‘algos’ have taken all the rhythm out of the market, and have become extremely confusing to me.” Philip Jabre closed his eponymous hedge fund, complaining, “…the market itself is becoming more difficult to anticipate as traditional participants are imperceptibly replaced by computerized models.”

The Financial Times quoted a senior JPMorgan strategist as saying that, “…we just have to accept that equity markets are almost fully automated.” JPMorgan estimates that only 10% of trading is generated by humans.

The Wall Street Journal reported that trend-following trading systems shifted “…from bullish to bearish to a degree not seen in a decade, according to an analysis of algorithms that buy or sell based on asset-price momentum.”

The WSJ also blamed volatility in commodity markets on computerized trading, citing a report that “…Goldman Sachs attributed the recent volatility to algorithmic traders exerting more influence in the oil market.”

The growth in algorithmic trading seems to be coinciding with a drop in discretionary trading by humans, which probably reflects that computers are beating humans in certain areas.

On one level, algos are designed to exploit human frailties by anticipating them. Perhaps they even cause them. It seems many people had little reason to sell in December beyond fear of losing money. Successful algos by definition attract capital, increasing their ability to move markets and hunt for more inefficiencies. Humans, at least those who adjust their positions too frequently, are becoming prey. One result could have been systematic shorting of pipeline stocks in response to crude’s sharp drop, because it worked in 2014-15. Sector weakness certainly couldn’t be traced back to recent earnings reports. Investors who sold during this time because they tired of losing money added to the downward pressure, reinforcing the trend. This would also have provided confirmation to the algorithms that the signal worked.

Since computerized trading isn’t going away, survival for the rest of us lies in being less sensitive to market moves. Examining fundamentals and staying with a carefully considered strategy that doesn’t rely on the market for constant confirmation of its correctness is the human response to algos.

Energy infrastructure remains attractively valued. Free cash flow yields before growth capex (i.e. Distributable Cash Flow) are over 11%. We estimate dividends will grow on the broadly based American Energy Independence Index this year by mid to high single digits. The long term outlook for the sector remains very good.

We are short AMLP.

SL Advisors is the sub-advisor to the Catalyst MLP & Infrastructure Fund.  To learn more about the Fund,  please click here.

SL Advisors is also the advisor to an ETF (

Print Friendly, PDF & Email

Important Disclosures

The information provided is for informational purposes only and investors should determine for themselves whether a particular service, security or product is suitable for their investment needs. The information contained herein is not complete, may not be current, is subject to change, and is subject to, and qualified in its entirety by, the more complete disclosures, risk factors and other terms that are contained in the disclosure, prospectus, and offering. Certain information herein has been obtained from third party sources and, although believed to be reliable, has not been independently verified and its accuracy or completeness cannot be guaranteed. No representation is made with respect to the accuracy,  completeness or timeliness of this information. Nothing provided on this site constitutes tax advice. Individuals should seek the advice of their own tax advisor for specific information regarding tax consequences of investments.  Investments in securities entail risk and are not suitable for all investors. This site is not a recommendation nor an offer to sell (or solicitation of an offer to buy) securities in the United States or in any other jurisdiction.

References to indexes and benchmarks are hypothetical illustrations of aggregate returns and do not reflect the performance of any actual investment. Investors cannot invest in an index and do not reflect the deduction of the advisor’s fees or other trading expenses. There can be no assurance that current investments will be profitable. Actual realized returns will depend on, among other factors, the value of assets and market conditions at the time of disposition, any related transaction costs, and the timing of the purchase. Indexes and benchmarks may not directly correlate or only partially relate to portfolios managed by SL Advisors as they have different underlying investments and may use different strategies or have different objectives than portfolios managed by SL Advisors (e.g. The Alerian index is a group MLP securities in the oil and gas industries. Portfolios may not include the same investments that are included in the Alerian Index. The S & P Index does not directly relate to investment strategies managed by SL Advisers.)

This site may contain forward-looking statements relating to the objectives, opportunities, and the future performance of the U.S. market generally. Forward-looking statements may be identified by the use of such words as; “believe,” “expect,” “anticipate,” “should,” “planned,” “estimated,” “potential” and other similar terms. Examples of forward-looking statements include, but are not limited to, estimates with respect to financial condition, results of operations, and success or lack of success of any particular investment strategy. All are subject to various factors, including, but not limited to general and local economic conditions, changing levels of competition within certain industries and markets, changes in interest rates, changes in legislation or regulation, and other economic, competitive, governmental, regulatory and technological factors affecting a portfolio’s operations that could cause actual results to differ materially from projected results. Such statements are forward-looking in nature and involves a number of known and unknown risks, uncertainties and other factors, and accordingly, actual results may differ materially from those reflected or contemplated in such forward-looking statements. Prospective investors are cautioned not to place undue reliance on any forward-looking statements or examples. None of SL Advisors LLC or any of its affiliates or principals nor any other individual or entity assumes any obligation to update any forward-looking statements as a result of new information, subsequent events or any other circumstances. All statements made herein speak only as of the date that they were made. r

Certain hyperlinks or referenced websites on the Site, if any, are for your convenience and forward you to third parties’ websites, which generally are recognized by their top level domain name. Any descriptions of, references to, or links to other products, publications or services does not constitute an endorsement, authorization, sponsorship by or affiliation with SL Advisors LLC with respect to any linked site or its sponsor, unless expressly stated by SL Advisors LLC. Any such information, products or sites have not necessarily been reviewed by SL Advisors LLC and are provided or maintained by third parties over whom SL Advisors LLC exercise no control. SL Advisors LLC expressly disclaim any responsibility for the content, the accuracy of the information, and/or quality of products or services provided by or advertised on these third-party sites.

All investment strategies have the potential for profit or loss. Different types of investments involve varying degrees of risk, and there can be no assurance that any specific investment will be suitable or profitable for a client’s investment portfolio.

Past performance of the American Energy Independence Index is not indicative of future returns.

Print Friendly, PDF & Email
2 replies
  1. Elliot Miller
    Elliot Miller says:

    There are some serious deficiencies in algorithmic trading.
    One is that the software is based on the past, historical facts and patterns, and cannot deal with future prospects. For example, SMLP has spent substantial sums on projects that will just now begin to produce substantial EBITDA, but that does not exist in the brain of software–only past trading based on prior EBITDA reports are taken into account by the software.
    Another deficiency in computer based trading is that the software does not distinguish between MLPs whose operations are crude oil based and those whose operations are based on natural gas and its derivatives. If the word “energy” is involved in the name or in the sector in which it trades, they are treated equally. If PAA is hit WGP will decline as well.
    Finally, rather than considering myself “prey” to algorithmic trading, I like to think of myself as a predator, taking advantage of the anomalies built into that system and buying the irrationally high yields provided by declining unit prices. I invest in MLPs for the tax deferred income they provide, so I benefit from high yields from MLPs that I consider core.

  2. Mike
    Mike says:

    I would suggest that we should take a more nuanced look at computerized or algo trading impact on markets.

    The term algo generally refers to a set of rules or instructions, coded into software. The code is largely a set of IF/THEN logic trees that take event and market based inputs and turn them into trading actions. Some of these algos are passive, with behavior analogous to a human market maker working to scalp pennies with lots of trades and agnostic to market direction. Other algos are directional and taking are looking to profit based on moves up or down in price.

    It is unwise or confusing to assume that this algo code must be logical, based upon equity fundamentals or market history. Some of this code is simply probing or testing price levels where other market participants will be sellers or buyers and pressing any advantages or opportunities found. And indeed, some of these can become self fulfilling and reinforcing. For instance, a possible shorting algo tactic could be to add volume on the short side at certain times of day or near technical price support levels. If buyers do not show up, then the positions size is further pressed or sized up and the price moves lower.

    Let us remember too that many of these MLP units and midstream shares are held within leveraged structures, such as CEFs run by Kayne Anderson and Tortoise. Even though these funds have committed capital, they also borrow money and then have leverage limits and rules and so they need to buy or sell based upon price action in combination with their debt coverage ratios and cannot just sit while a sharp down move occurs. So too does the money in dynamic structures such as traditional mutual funds and ETF change on daily or even real time basis. As these fund flows occur for various reasons, then the supply demand for units and shares changes. If people want cash, the funds have to sell. This was going on in lots of corners of the market in December and into January, not just midstream. As an example, look at BV landscape company, LCII RV supplier and ALGT discount airline that have had 15-40% moves in the last month.

    To me, algos are a tool. A software way to codify trading rules and actions. Trading accuracy (based on the coded logic) and speed are improved. This is not necessarily bad or good. And algos are not necessarily, smart or dumb.

    Owners of equities need to know what they own and why with the ability to hold through volatility while remembering that Mr. Market can offer up prices which reflect ‘moodiness’ that does not make sense or even need a reason.


Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.