To their credit, Kinder Morgan (KMI) responded to our recent blog (see Kinder Morgan’s Slick Numeracy). We exchanged several emails, although they declined our invitation to write a rebuttal which we promised to publish unedited. The company stands by their presentation, but did concede that some slides might have been more clearly labeled.

We had noted in our blog that the 5.9X EBITDA multiple on $12.3BN invested didn’t foot to the $1.8BN EBITDA from growth projects in the Bridge Chart. KMI explained this was because some investments they made had been sold in the meantime.

They believe that attractive returns on more recent investments have been masked by headwinds in their existing business. The EBITDA Bridge Chart blames the 2014-17 energy downturn for $0.6BN in reduced EBITDA from their CO2 business, which is net of new investment (i.e. CO2 growth projects are included in this figure). The $0.3BN drop in the Midstream segment was from lower volumes and tariffs in their pipelines.

KMI’s return on invested capital has drawn questions from others. A January 6 research report from Morgan Stanley placed KMI’s 2017-18 Return on Invested Capital (ROIC) at 4.5%, worst out of 12 peers. In October, Wells Fargo calculated a 2013-18 cash return on investment of 4%, 2^nd worst in the group and declining.

In response, KMI referred us to a slide showing ROIC by segment. They say they have discussed the Wells piece with the firm, and make a distinction between recently invested capital and returns on legacy assets.

The ROIC slide incorporates some complexity. The footnote reminds that pre-2014 returns are from Kinder Morgan Partners (KMP) and El Paso (EPB), which is where KMI’s operating assets were held before being rolled up into the parent. Those were the days of Incentive Distribution Rights (IDRs), when KMP and EPB both paid a share of their returns back to their controlling general partner, KMI.

Once the IDRs went away, returns might have been expected to jump. That they didn’t suggests that the chart treats IDRs as a cost of capital and not as an expense to KMP and EPB, which they most assuredly were. So we think the returns are on the assets, not on what KMP and EPB unitholders earned on those assets.

KMI believes they are making a genuine effort to present their case, and in providing so much detail they create opportunities for investors like us to look for inconsistencies. But we think that EBITDA multiples aren’t a good way to do it. The declining ROIC chart is hard to reconcile with higher recent returns. It also highlights the volatility of the CO2 business, which they evidently believe can get back to the returns it generated a decade ago. The fact that they haven’t yet received a sufficiently attractive offer for this segment means few share their optimism.

The company uses IRR in allocating capital. They say new projects require unlevered pre-tax returns of 15-20%, but their ROIC chart shows returns sliding towards 10%. At some point the high return investments of recent years must lift their overall ROIC. EBITDA multiples can flatter – a project with declining EBITDA (like a CO2 investment) might look superficially attractive based on Year 2 returns but ultimately not cover the cost of capital. The company is adamant they’re not allocating this way. So why not show expected IRRs on new investments?

We appreciate KMI’s effort to reach out – “slick numeracy” probably didn’t gain us any additional friends there. Along with countless other long-time investors, we’re frustrated that KMI remains well below the highs of 2014. Their stable, fee-generating assets ought to draw a higher valuation.

A little noticed sentence in President Trump’s State of the Union speech was that the U.S. would join the Trillion Trees Initiative. He didn’t mention climate change – most people like trees anyway, and the link was obvious. But planting trees to combat climate change is so seductively simple that it must be impractical. A trillion trees is a lot, and mentioning it in a prime time speech will have caused at least half the country to dismiss a political gimmick.

So we looked at the plausibility of harnessing nature to consume the excess CO2 humans are generating.

To grow, trees require CO2 and water, which through photosynthesis they convert into carbon and  glucose to form the tree, and oxygen which they emit. Growing 1 ton of wood requires around 1.55 tons of CO2. By atomic weight, carbon is 27% of a CO2 molecule. A dry tree is typically around 50% carbon.

Of course trees vary enormously in size. The University of Arkansas Division of Agriculture publishes a handy guide for estimating the weight of a tree given the measured circumference.

From this, we estimate that the average hardwood tree weighs 3.4 tons dry, or about 3.1 metric tonnes (the unit that’s used for CO2 emissions). Assuming it grows over 100 years, it’ll require half its weight in carbon (1.55 tonnes), which will be extracted from 5.7 tonnes of CO2 (i.e. 1.55 divided by 0.27).

The world generated 33 gigatonnes of CO2 last year. Divided by the 5.7 tonnes consumed by the average tree, planting 5.8 billion trees would, over their life, consume all the CO2 emissions of last year.

Is this possible? Earth currently has 3.04 trillion trees, so we’d need to increase the stock of trees by just 0.19% per annum to take a giant step towards combating global warming.

Last year, one million Indians planted 220 million trees in a single day, as part of a government campaign against global warming. Ethiopia planted 353 million trees in 12 hours.

America plants 1.6 billion trees a year – half by forest product companies. Costs are estimated to be as low as 30 cents a tree.

Even if this analysis is out by a factor of 10X, that would still leave the world needing to add 1.9% to global forests every year – certainly viable if embraced as a solution.

Planting trees is labor-intensive. But two companies, Droneseed and Dendra Systems, are developing plans to use drones that can plant seeds on hundreds of acres a day, versus the two acres that a professional tree planter typically covers. You can watch two interesting videos explaining how here, and here.

So it would seem that a global effort to add around 6 billion new trees every year is achievable.

Articles like World losing area of forest the size of the UK each year, report finds in the UK’s Guardian stoke fears of enormous global tree loss. Brazil is widely criticized for deforestation in the Amazon, but overall the portion of the world covered with trees has been growing. This is partly because a warming planet is raising the tree-line in mountainous areas, and allowing forests to creep into tundra.

Environmental extremists have instinctively rejected the Trillion Trees Initiative, for mostly predictable reasons: it doesn’t require overhauling our energy supply, or erecting millions of windmills, solar panels and tens of thousands of miles of ugly high voltage electricity lines. It seems so much more attractive than the Green New Deal.

Renewables, nuclear and natural gas are all part of the solution to climate change, along with adding billions of trees. Burning natural gas produces water and carbon dioxide, the two inputs trees need to grow. Burning coal releases nitrogen oxides, sulfur dioxide, particulate matter, mercury and other hazardous substances that the local population inhales. Everyone should be able to agree that coal use must drop.

Most of the criticism of the Trillion Trees Initiative stems from concern that it’ll weaken the case for dramatic interventions to the economy promoted in the Green New Deal. That’s precisely why it’s appealing.

Last week Kelcy Warren, CEO of Energy Transfer (ET), defended the MLP structure. He’s definitely correct that MLPs possess a powerful tax advantage over corporations, in that their profits are only taxed at the investor level. Tax-deferred income free of the double-taxation to which corporate profits are subject is very appealing, and for years it drew countless buyers. Unfortunately, Warren is part of the reason that the MLP structure is losing favor. Midstream energy infrastructure and MLPs used to be synonymous, but widespread distribution cuts and investor abuse have left the old, rich Americans who used to be the investor base betrayed.  The names Kelcy Warren and Rich Kinder still elicit strong reactions from longtime MLP investors.

The Alerian MLP ETF, a good proxy for how MLPs have performed, has cut its distribution by 34% since the market peak in 2014. Companies chose to finance growth projects in excess of free cash flow, and ultimately resorted to either outright distribution cuts or “backdoor” distribution cuts by merging with a their lower yielding corporate general partner. Many MLPs abandoned the structure, and income seeking investors in turn have abandoned the remaining ones.

The result today is that MLPs represent 36.5% of the sector by market capitalization, as defined by the Alerian Midstream Energy Index — AMNA (see MLPs No Longer Represent Pipelines). Kinder Morgan, ONEOK, Enbridge, Targa Resoures and Williams Companies are among those that have fully adopted the corporate structure.

MLP-dedicated mutual funds and ETFs were originally designed to offer sector exposure to retail investors who didn’t want to deal with K-1s. They saddled their investors with a ruinous tax burden, because funds with over 25% of their portfolios in partnerships (which is what MLPs are) have to pay corporate tax. It seems odd to take a tax-efficient vehicle and add taxes to it, but showing how few investors read the fine print, these products took hold. And they’re now focused on just 36.5% of the sector (see Are MLPs Going Away?).

To illustrate how much things have changed, just two names, Energy Transfer and Enterprise Products, represent 43% of the market cap of all MLPs.  Dedicated MLP funds are forced to drastically underweight these two, which leaves them with outsized exposure to the smaller MLPs. They’ve moved a long way from diversified portfolios of large, fully integrated “toll road” pipeline systems that originally attracted investors.

The biggest of them, the Alerian MLP ETF, has since inception delivered less than one third of its index. This is probably the worst performing index ETF in history. Corporate taxes have taken a bite, and when the sector delivers a couple of big years the tax hit will be even more noticeable (see MLP Funds Made for Uncle Sam). The 1.04% since inception annual return is not far from the 0.85% advisory fee, putting AMLP in the company of the hedge fund industry in making profits while the clients don’t.

If you ever meet one of the hapless souls who’s chosen AMLP, you’ll find they’re probably unaware of the tax drag.

The shrinking number of MLPs has rendered MLP-dedicated funds less representative of the sector. Of the ten biggest North American pipeline companies, six are corporations and so excluded from AMLP and its cousins. Every time another pipeline company leaves the publicly-traded MLP universe, these funds are left with fewer names and a preponderance of small ones. The market has shifted since many of these were launched a decade or more ago (see AMLP’s Shrinking Investor Base).

If any of these funds decides to reduce their MLP exposure below the 25% threshold, so as to be more representative and avoid corporate taxes, it’ll depress MLP prices because many will have to sell three quarters of their holdings. A quarter of the $135BN in public float for all MLPs is held by $34BN in MLP dedicated-funds. It’s a crowded space.

Moreover, AMLP reflects the worst of MLPs – AMNA is 21% Gathering and Processing (G&P), the more risky end of the midstream business because it’s more dependent on production from specific areas. Due to limited choices, AMLP has 26% exposure to G&P. Even worse, natural gas pipelines are a big underweight in AMLP, even though the long term prospects for natural gas are more visibly positive than for crude oil and liquids. Natural gas pipelines represent 46% of AMNA, but only 27% of AMLP. So AMLP investors have an overweight towards crude oil and liquids.

Investors are starting to act on the many flaws of MLP-dedicated funds. Over the past year, $4.1BN has left the sector. The American Energy Independence Index is investible (you cannot invest directly in an index) and has weights that are more reflective of the industry. Its holdings are mostly corporations, which reflects today’s pipeline business. Several names are ESG holdings for Blackrock and other big fund managers, but MLPs don’t pass ESG screens because of poor governance (watch ESG Investors Like Pipelines). The broader investor base and ESG qualities helped pipeline corporations outperform MLPs last year.

Disclosure: our affiliated investment products are structured to reflect the insights listed above.

A Chief Financial Officer needs to know her way around a financial statement. Presenting operating performance in the best possible light is a highly valuable skill. We watched Kinder Morgan’s (KMI) Analyst Day last week via webcast, and my admiration is split between (1) KMI President and former CFO Kim Dang’s deft maneuvering among numbers, and (2) my detail-oriented partner Henry Hoffman for spotting the sleight of hand.

The story begins where KMI uses a “bridge” chart to show how EBITDA changed from 2014 to 2020. Dang introduced the slide as answering a common investor question: if you’re doing so well, why isn’t EBITDA growing more?

The chart includes the much criticized CO2 segment, which sells CO2 for Enhanced Oil Recovery (EOR) as well as being used by KMI itself for that purpose. CO2 contributed a $0.6BN decline to EBITDA from 2014-20. Since its decline is shown on the chart, it must also be part of the $1.8BN EBITDA increase from Expansion Projects, because the $7.6BN 2020 Adjusted EBITDA is for all of KMI.

Another slide shows an attractive EBITDA multiple on capital invested, of 5.9X. In other words, $100 invested will return its capital invested over 5.9 years. The reciprocal of 5.9X is the return, a juicy 16.95% in this case and well over KMI’s assumed cost of capital. The 5.9X multiple implies these projects are generating $2.08BN in EBITDA.

But the small print notes that on this slide, CO2 segment Expansion Projects are excluded.

To find out what KMI invested in CO2 projects 2015-19, we turn to another slide, which shows annual figures.  Adding up the Capex for 2015-19 gives $2.18BN.

So KMI’s total capex over the period covered was not $12.3BN but $14.48BN,  once you add back CO2.

This makes the EBITDA multiple on new investments a little less stellar. The $1.8BN in EBITDA from Expansion Projects that required $12.3BN in investment equates to a 6.78X multiple. But since we know the $1.8BN EBITDA is from all Expansion Projects, including the CO2 business, it’s appropriate to add back the $2.18BN invested in CO2 Expansion Projects to get the true total.

That means the $1.8BN in “growth” EBITDA was earned on $14.48BN in capex, or a multiple of 8.04X. That’s 12.44% rather than the 16.95% calculated earlier. It’s pre-tax, and excludes interest expense as well as depreciation.

KMI uses more than one EBITDA definition through these slides, so you have to follow carefully. When calculating the EBITDA multiple, they like the “Year 2 Project EBITDA”, with its implied $2.08BN in EBITDA. The EBITDA bridge chart simply shows 2020 EBITDA of $1.8BN from all new projects initiated since 2015. Why are they different, and what does it mean?

Investment outlays have been falling since 2015, which means that the projects funded since then are, on average, past the two year mark. 2020 capex is almost a third lower than it was in 2015.

Since the $2.08BN “Year 2 Project EBITDA” is above the $1.8BN 2020 EBITDA from Expansion Projects in  the bridge chart, and we know the average project is more than two years old, it means that EBITDA from new projects is declining. Using the “Year 2 Project EBITDA” flatters the results. Management teams persist in using EBITDA multiples, which are easily manipulated, rather than NPV and IRR. It’s as if they don’t think their investors will understand them – or maybe they’ve concluded that EBITDA multiples are easier to obfuscate.

Following this is admittedly a complicated exercise, but it illustrates the agile numeracy required of today’s energy sector CFO, and of the investors interested in properly understanding their business. KMI is certainly not alone in using financial complexity to their benefit. In this case, we think it’s driven by their continued retention of the CO2 business, which they should sell. Apart from that, KMI has some great assets.

We are invested in KMI