High Fees Unlikely To Interfere With Brookfield Infrastructure Partners' Solid Unitholder Returns

Summary

• We calculate the present value of fees paid by BIP to BAM.
• About 40% of the cash flows produced by BIP goes to BAM, much more than is apparent from the 25% highest tier of IDRs.
• While the fees are high, they need not to interfere with very satisfactory returns for ordinary unitholders; it is reasonable to expect at least 10-11% at present.
• Based on sheer numbers, the incentives are much less aligned for BIP compared to Brookfield Property Partners. With BIP, one relies more on the integrity of Brookfield.

A topic rather neglected in articles on Brookfield Asset Management's (BAM) listed partnerships (BIP, BEP, BPY and BBU) is that of fees. As those partnerships will grow over time and increase their distributions, fees will gradually become more and more of a burden on cash flows. I have analyzed the fees paid by Brookfield Property Partners (BPY) in my recent articles (and I really encourage the reader to read "Incentives To Increase Return On Equity Should Lead To Nice Gains At Brookfield Property Partners" to better understand the method, its assumptions, shortcomings and implications). Now I will apply it for Brookfield Infrastructure Partners (NYSE:BIP). I will start with a list of the most surprising and most important discoveries for BPY.

• BPY, over its 5-year public existence, has not paid any incentive fees on distributions yet thanks to some mysterious "accumulated credits".
• Even without those credits, the current rate of incentive fees would be about 3% of distributions to unitholders (compared to the 33% maximum allowed).
• Majority of BAM's fee income will be earned not through the seemingly high 33% incentive fee on distributions, but via the innocuously looking 1.25% equity enhancement fee (calculated from market cap). The reason is the growth of both distributions and market cap thanks to the earnings retained within BPY.
• The current unit price of ~$18 corresponds to a projected 8% return on equity. That would correspond to NAV of $25 out of which $7 will go to BAM as fees.
• Despite these seemingly obscenely high fees, unitholders would do great if return on equity was 12%+ as the management targets and has achieved in the past decades (but not over the last 5 years).
• In particular, if one aimed for a 12% return and assumed 12% ROE, then the current unit price would still offer a 25% margin of safety.
• The fee structure, together with BAM's large stake in BPY, incentivizes BAM to increase ROE (e.g. by recycling capital), not to issue new units. And that is what actually they have been doing most of the last 4 years.

Brookfield Infrastructure Partners

The fee structure at BIP resembles that at BPY, but there are minor differences. The fees are as follows:

1. Base management fee of 1.25% p.a. of total capitalization (paid quarterly), where total capitalization is the sum of the values of the outstanding units (i.e., market cap, but a portion of the units is held directly by BAM and thus not part of the market) plus market value of the preferred units plus all recourse debt minus cash. At present, total capitalization consists of $15B of ordinary units, $1B of preferred units and $2B of recourse borrowings. We will henceforth calculate the management fee as 1.25% of market cap plus $3B, or 1.25% of unit price plus $8 if interested in per-unit fee.
2. Incentive distribution (we might call it fee as well). From the first $0.203 of a quarterly distribution, nothing goes to BIP. Between $0.203 and $0.22, 15% goes to BAM, 85% to unitholders. Above $0.22, 25% goes to BAM and the rest to unitholders. (These thresholds are updated by amendments to reflect unit splits.) We are now at $0.47, way above the highest threshold, and can make the following simplification: the yearly fees paid on top of what unitholders receive will be $0.01 plus 33% of the difference between annual distributions and $0.88.

One can find this (and much more) in the 2017 annual report. The base management fee was $126M, $158M and $230M in the 2015, 2016 and 2017, respectively; the incentive fee was $64M, $80M and $113M in those years. (You can compare it with what ordinary unitholders received in distributions; in 2017, it was a bit less than $700M.) BAM owns about 30% of BIP (116M of 394M units outstanding); its stake is currently worth about $5B.

Instead of NAV and return on equity, the management refers to invested capital and return on invested capital (ROIC) - the ratio of adjusted AFFO to invested capital. Currently, invested capital is ~$8B, or about $20 per unit, and ROIC has very consistently kept between 10 and 13%, inching up over the years. However, ROIC of at least 11% is necessary just to cover the current distribution, and at 14%, the distribution growth based on retained cash flows would amount to only 3%, a far cry from the 6-9% anticipated by the management. Note that FFO is based on net income, thus already all gains on asset disposals, IFRS fair value gains, inflationary increases, etc., are included.

It took me considerable time to figure out this apparent contradiction. The key are the following two slides from the 2017 ID presentation. (I strongly encourage readers to study the financial considerations extensively discussed during that presentation. Read the transcript, compare it to the data in ARs and supplementals, read it again. It is a lifelong education on how to think about returns from long-life assets.)

The point is that with small incremental investments, a large growth of AFFO is achieved because the cash flows generated by the existing assets are growing with no need for additional capital. Even with no new investments, significant AFFO increases would come every year (75% is indexed to inflation). This also explains why ROIC has been creeping upwards over the years. For the purpose of my calculations, I will consider ROIC in the range of 12-24%; that is what is likely to be achieved when retained AFFO is reinvested. (Since inception, they managed to achieve 17% of this "incremental" ROIC.)

There is one other important difference. BIP already subtracts the base management fee when calculating FFO (it is part of the G&A expense). The management fee is roughly 3% of invested capital per unit, thus the correct rate to use should be those 3% higher than what BIP reports.

Finally, a word of warning. The calculations are performed to illustrate the effect of fees, and must not be misinterpreted as an attempt to pinpoint a precise value on BIP units. The assumptions I have used are too simplifying and the values derived highly depend on ROIC which is unlikely to be high and constant - the last decade was quite favorable for infrastructure, with growth coming from increased volumes, inflation escalators, etc.; it might not go as smoothly in the future, especially not during periods of economic unrest or much higher interest rates. Despite these drawbacks, I believe that the results at least illustrate the range of possible outcomes. The table below uses a 10% discount rate for the calculations of present value.

As we can see, BAM takes about 40% of the present value of cash flows, and this percentage does not depend much on the ROIC. The high fee is not necessarily bad if they provide ordinary unitholders with adequate returns. For instance, the present unit price of ~$40 roughly corresponds to the historical ROIC that has been achieved. That would mean that if future conditions do not change much, then distributions will grow by ~6% p.a. and total returns of about 10-11% p.a. can be expected practically ad infinitum, plus some upside from accretive issuance of new units. The fact that BAM extracts $36 (at present value) in fees over time does not interfere with it much.

The only problem could be with incentives: BAM might want BIP to extensively issue units to increase fees. In the last decade, they have not done it with value-destroying consequences, and I find it unlikely to happen in the future. The key reason is that they would not want to destroy or even just dent their reputation; even a couple billions of fees extracted from BIP is negligible compared to what they can earn on hundreds of billions of AUM in other listed partnership and institutional funds over the next decade (and to the loss of that opportunity caused by screwing unitholders over).

The "numerical" alignment is, however, much weaker compared to BPY: with BIP, BAM's unitholder stake is $5B, but the overall fee stream is worth $15B. Thus, if they decided to double the size of BIP by mediocre acquisitions financed by new units, decreasing overall ROIC by 2% in the process, they only lose $1.5B in invested capital, but gain $5B in additional fees.

Analyst’s Disclosure: I am/we are long BAM, BPY. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.

Comments

[edgardo-edmondo]

The ROIC definition in the 2017 ID presentation looks strange.

On first take, it looks like a scheme to inflate ROIIC, return on incremental invested capital, as it muddles contributions to cashflow from pre-existing assets with those of assets acquired in the five year period.

Why would you take cashflows from existing assets that are growing with no additional capital and include them in a calculation to determine the ROIC of new investment? You can't attribute the growing cashflows from a port acquired a decade ago to the farrago of infrastructure assets acquired in the last five years in any sane what's-yours-is-yours-and-what's-mine-is-mine measurement methodology.

On second take, just how different is this BIP ROIC from a plain old yield-on-cost? The denominator is supposed to be "reflective of the capital we have been allocated", i.e. units and preferreds cash proceeds. While the numerator just keeps on growing without reference to the increasing value of the underlying assets.

Mgmt revaluation of assets is a continual process (that exacerbates issues with AFFO vs net income, cf. 2016 ID presentation) that they conversely choose to avoid when it comes to ROIC.

BIP defines Invested capital as: "We define Invested Capital as partnership capital removing the impact of the following items: non-controlling interest in operating subsidiaries, retained earnings or deficit, accumulated other comprehensive income and ownership changes."

I would assume that under this definition the denominator then excludes the ~20% of retained and reinvested cashflows and any gains on disposals. Net income is small compared to AFFO (D&A >> maintenance capex) so retained earnings (and gains) probably don't exist after distributions well in excess of earnings (net income = earnings). Revaluation of assets would be put into accumulated other comprehensive income and removed from invested capital.This leaves the unit and preferred proceeds, which is why I wonder just how far from a yield on cost this bespoke ROIC definition is.

In the 2018 ID presentation mgmt highlights lower 7% FFO yield assets being sold to reinvest in higher 11% FFO yield assets. What happened to the 2017 ID presentation's 20% AFFO based ROIC on the last five years of incremental investments? They're now excited about investing at an 11% FFO yield? An AFFO yield has to be lower than an FFO yield if there's any maintenance, and the numerator of the ROIC ratio is AFFO. Of course if the ROIC is an all-flash-no-dash yield on cost in disguise, it all makes sense.

What will they think of next? (w/ apologies to Joe Campanella...)

[Ján Mazák]

Yes. For evaluating fees, it was good enough -- I work with just rather large range for ROIC anyway -- but for evaluating unit price, one has to be much more careful. Thanks for the valuable comment.

My present take is that the 17% incremental ROIC mentioned in the 2017 presentation (since inception, not just over the last 5 years) is the upper limit on where their reported adjusted AFFO yield (basically on cost) can grow. It got from 10% to 14%, so not much room to grow, and if economic conditions worsen, will likely go down instead.

But YOC is a dangerous measure to use over time -- can grow to absurd numbers if retained earnings are not added to the denominator. (I do add them in my calculation.)

I have not done the necessary work for valuing the units. BIP is not really awash in capital without BAM and/or equity issuance, so they can enjoy cheap acquisitions during a recession, but not much of it. BAM is in a way better position in this regard. Or BRK.

[gastro4]

Thanks for the article. Very interesting analysis

[Dry Powder]

Jan, I remember being surprised how BIP thinks about ROIC. Honestly, I think that the DeltaAFFO/DeltaIC is a measure of earnings growth but it is misleading to call it ROIC. Long term, I think that true ROIC will converge to some limit near FFO/Equity. In other words, I think that looking at the deltas is a measure of recent success but that it is a derivative and might not be meaningful long term. Can you expand on how you think about DeltaAFFO/DeltaIC?

[Ján Mazák]

It is hard to put a precise value on BIP (or, basically, on any rather fast growing cash flow stream). Consider, for instance, that if the market price changes by 20%, the yield changes by 1% (i.e. from 5% to ~4% or ~6%). Since growth from all those inflationary escalators and retained earnings is the same, the total return for a prospective unitholder also changes only 1%. And even this is mitigated by the change in the ability to issue capital -- if price went down (thus yield went up), they are much less likely to issue new shares, and vice versa.

In addition, economic environment has been rather favorable for infrastructure over the last decade, and that might not repeat. On the other hand, Brookfield infrastructure vehicles (funds and BIP) can excel even during recessions --- with capital locked for many years, they are in no hurry, and can deploy additional capital at good prices (but BIP would have trouble getting much new capital if units traded low, albeit temporarily). The problem with ROIC is that it would not be static. Perhaps the historical average of 17% mentioned by management is a reasonable value to consider --- that is what they achieved by turning over almost all of the initial assets (deltaAFFO / deltaIC). I would have to think about it more, but BIP is taxed very unfavorably for me, so I have avoided it for the last few years.

Overall, BIP has many very favorable characteristics that are hard to beat by an average company. I am thus a bit surprised that despite great track record and bright future prospects of BIP people are willing to accept lower yields / expected total returns from lots of businesses much more risky, with growth much less secured. (E.g. many MLPs.)

But I still favor BAM: I am not sure where the next dislocation will be and they can buy stuff left and right, e.g. units of BPY now, or can do meaningful repurchases. They retain much more capital and thus have better capital allocation flexibility. And the expected growth rate is much higher.

[Guy at Work reading SA]

Very insightful, thanks!

[me66]

Nice article, Ty, I am watching BIP intently. It is a top candidate for my 2019 Roth contribution. Perhaps $4K to BIP, and $2.5K to BAM, might be a prudent choice. There can be more than one winner, I like the underlying story w/BAM, they seem to have a good business model.

[mound_dweller]

I initially invested in BIP in August of 2013 at around $22.95/share. Using BIP's closing price yesterday of $39.01 and taking into effect the 3/2 share split in September 2016, my total return on invested capital, including dividends is around 208%. That works out to an annualized return of about 39%.

In the 5 years I've held BIP quarterly distributions have increased from $0.48 to a split adjusted $0.705. That's a 47% increase and equates to an average annual increase of 9.37%.

I think the fees BIP pays to BAM are quite reasonable!

[retired+guy]

Great article.
I have long suspected that BAM was the smart money and your analysis seems to bear that out.

However, as a holder of both BAM and BIP I keep waiting for BAM to out perform. It has not? I am still waiting.

Am I missing something or does Mr. Market know something I don’t

[Visual Capital]

@Ján Mazák Great article as always. I have a few questions :

I'm curious, what is accumulated credits and how are they paid out? Shares, cash

Also, you said : " The current unit price of ~$18 corresponds to a projected 8% return on equity. That would correspond to NAV of $25 out of which $7 will go to BAM as fees."

I didn't follow your argument.

[Ján Mazák]

The argument on 8% is in the article
seekingalpha.com/...
which should have been published before this one, but I messed it up a bit. Sorry. This article just states it as in interesting conclusion.

Basically, the 8% ROE at BPY corresponds to 2% distribution growth, and the value of the ordinary distributions to a uniholder is $18 per unit (at a 10% discount rate), while the value of the fee stream for BAM is $7, so altogether the assets of BPY will create a cash flow stream worth $25 which I called NAV (determined by a discounted cash flow calculation).

[Jan Blanckaert]

Pyramid Plays generally last not 20 years.Way to complex in order not to bring conflicts of interest. I own Bep but will reconsider.

[nsm0002]

Strongly feel that BAM is the correct play in the Brookfield universe.

[HunterKiller89]

I'd agree for any long term prospect that doesn't also require dividends. Short/mid-term, some of the LPs can likely appreciate faster than BAM will, but 20 years out? I'm sure BAM will have grown much faster than any of the LPs even with reinvested dividends

[Ján Mazák]

You might consider going even higher in the hierarchy.
seekingalpha.com/...
PVI (PVF.UN) strips out dividends from BAM but gives more exposure to capital appreciation (half of the insider stake in BAM is held via PVF.UN). Or warrants on PVI (PVF.WT) -- perhaps the best derivative security available in public markets (more than 7 years to expire, might be a 10-bagger depending on how BAM performs).

[HunterKiller89]

I wouldn't mind either of those personally, but they're incredibly illiquid unfortunately.

[Stocks for the Long Run]

Very interesting perspective.