Saturday, October 31, 2020

My Dividend Portfolio

I started building this portfolio 18 months ago.  Its been a wild time!  From an "high-and-dry" market in late 2019, where I'd be desperately scrounging around for something to buy like an animal in the desert, to the liquidity crunch in March, where you would buy something a deep value, get punched in the face by Mr Market dropping it 10% in a few days, then repeat the breathtaking recovery which no one believed.  

I was lucky that covid occured when it did.  If you can't build a portfolio when a once-in-50-year pandemic occurs and the end of the economic cycle, you never will.  My timing wasn't great.  I only bought 10% of my portfolio in March.   You can make a lot of mistakes in this business, but as long as you manage risk and don't buy shit that goes to zero, you'll make more than you lose.

My dividend portfolio is now worth SGD 670K, after the last few weeks market hiccups, and can probably pay me SGD 2.7K per month.  Here it is:

What next?

  • Keep grinding away at my day job, plow all my salary back into the market.  Its still fairly valued, I can find things with a 5% yield.  No need to time the market too much.
  • The portfolio is weighted heavily towards Singapore REITS.  Try to diversify away from this.
  • Reits and Utilities weight it heavily towards companies that benefit from low rates.  At some point of the economic cycle, we start expecting higher rates.  Look to buy dirt-cheap banks to balance this out.  "Neither a borrower nor a lender be."
  • Keep learning to trade.  I have a small trading portfolio (~SGD 120K), where I learn to trade by following Hedgeye.  Commodities look like a good bet now.  There may come a time when I can't add to my dividend portfolio (like late 2019), and I may need to learn to trade at different points in the economic cycle.

Monday, October 19, 2020

Bought European Airports

I think we get a covid vaccine in the next 2 weeks (Pfizer & BioNTech) to 2 months (Moderna or Novavax).

It won't mean everything is over.  Pfizer and BioNTech's vaccine requires 2 shots and must be stored at -70 degrees Celsius, making distribution hard.  We can expect most people in the US to be vaccinated by the start of 2023.  Mid-2021 if an easier-to-use vaccine makes it.

News of a vaccine will remove uncertainty and give a timetable for recovery. The stock market hates uncertainty but it can take bad news of a known quantity.  I think this is the turning point for covid plays.  If I'm right on the vaccine timing, its now darkest just before dawn, when uncertainty (when will this virus ever go away?) gives way to bad news (one more year to go).

And even if the all the vaccines somehow fail, I think the virus turns into a normal seasonal flu by the start of 2023.  The 1957 and 1968 pandemics lasted around 2 years.  So 2023 is the last we see of corona chan.

My covid recovery play is airports.  They've made it through the crisis without raising cash so far.

I'm buying European Airports because:

  • Regional Airports (Australia/NZ/Thailand) are too expensive, even based on pre-covid earnings.
  • MAHB has is changing their regulator model to be more capital intensive.  Its probably good for them, but we can't use historical numbers to value it.
  • NYSE-listed Mexican airports have only 28 years concession left
  • is too expensive, as if covid never happened!  The other OTAs are bring squeezed by google.

My picks are Aeroport de Paris and Aena, from Sven Carlin.   Bought 210 shares of AENA @ 118.8 Euros, and 300 shares of ADP @ 82.45 Euros.

Theres still some risk.  From their results, I can't figure out their opex and capex (ie: cash outflows with zero revenue), and Paris has high borrowings.  If the recovery is delayed till 2023, there's a chance they do end up raising money.

Swire Properties

 Hong Kong's largest office landlord.  They hold (mostly freehold) investment properties which they rent out for a steady income stream, using this to develop more property in HK and China.

I want to look at their Investment Property (rental stream) business first, because its recurring and easy to value.  Like a REIT.  So I exclude their lumpy Property Development and Property Trading for now.  Their past ten year Property Investment income is impressive:

Source: 2019 Annual Report, p10

Theres no depreciation in their "Profit Attributable to Company Shareholders" for Investment Properties.  Properties are valued every year and the gain/loss is added to the P&L (HKAS40 does not require depreciation (p17) for Investment Properties held at fair value).  The company helpfully puts revaluations on separate line.  So these profit numbers are purely for rental and 'cost-of-rental', similar to a REIT's Distribution Statement.

These numbers make me excited.  Can you find any SGX listed REIT or property company that doubled their rental profits in 10 years while reducing debt?  And the number of shares issued in this period is only up 3%.  This shows the power of compounding when a company reinvests some of its income, instead of paying it all out as dividends.

Investment Properties

Geographical Segment breakdown

They rent out office and retail properties in HK, China, with a small business in Miami.  They only give breakdown by Total Revenue (not Investment Properties revenue/profits):

Source: 2020 Interim Results Briefing


I estimate that 70-80% of their Investment Property profits are from HK.    I guess 50-60% of HK profits should be from office, with the rest from retail.  Their Office is mostly Grade-A, and saw positive rental reversions in the first half of the year - so not much to worry about here.  Retail is mixed: they have 2 main malls: 

  • The Mall, Pacific Place: an upscale mall surrounded by offices and 4 hotels, next to the High Court and British Council.  It boasts a large number of luxury brands as tenants (p29).  Retail sales fell by 47% here in 1H20 (p11).  Long term, I expect the squeeze to continue even after covid and social unrest are resolved, as luxury brands decrease their HK footprint.

  • The largest regional shopping centre in HK, with MTR access, shopping & dining, cinemas and an ice-skating rink.  Retail sales fell 20% here in 1H20, I expect it to rebound long term, as things recover.
  • Other properties, around 15-20% of HK Investment Property GFA (weighted by ownership).
4.5% of their HK retail rentals expire this year, followed by another 23.5% in 2021.  We can expect continued pressure here.  This may already be reflected in 1H20's numbers (gross 1H20 HK retail rent was down 10% - p11), as Swire has already given concessions to luxury tenants, showing they will cut rent to keep occupants.  Both malls remain 100% leased as of June.

How bad can it get?  Back-of-envelope calculation: Lets say that The Mall accounts for 40% of HK Investment Property Retail profits, or 20% of HK Investment Property profits.  Then lets say The Mall's Profits halve (following a further 30% decrease in gross rental to 1H20 results).  So we could see a 10% drop in HK Property Investment income, which is maybe a 5-8% drop in total Investment Property income.


28% of their Gross rental is from Mainland China, I guess it accounts for 15-20% of profits.  Most of the China rental is from retail.  Skewed towards luxury (p15): 40% if their rental is from "Fashion and Accessories", and 20% is from "Jewellery and Watches".  They are optimistic: "Footfall and retail sales in the Chinese mainland have recovered strongly since March 2020. Retail sales are expected to continue to improve for the rest of the year, led by sales of watches, jewellery and other luxury brands." 


Since I'm valuing based on property rental, I need to know their property tenure.  62% of their completed HK investment properties have leases of 800+ years (by GFA), 23% have leases 30 years of less.  All China properties have 40 year leases.

Future Growth

Their development pipeline calls for a ~10% increase in GFA by 2023:

Spending on this is forecast to range from 2bn to 5bn a year:

Assuming 1H20 is as worse as it gets - Investment Property profits were 4bn in six months - they should have no problem covering this expenditure.

Other Segments

They have a small hotel segment, which usually operates at a loss (pre-covid).  Hotel P&L numbers do include depreciation.


Company-wide, net debt (including leases) is around 17bn.  Or 2.1 times annualised 1H20 Investment Property Earnings.  This debt is so low compared to their earnings/assets, its not even worth looking at the debt repayment schedule (p28).

Property Development Segment 

"Property Trading" and "Sales of Investment Properties" have been lumpy over the past ten years:

Instead of analysing the history of property transactions, lets look at where they are now.  It means have some residential properties to dispose of after covid:

My guess is that they write down some Miami condos as the market crashes, but most are already sold anyway.  For Singapore and Indonesia, worst case is they sell at a discount if theres a recession.

Second, they are developing new properties:generally, the residential units will for sale in their "trading portfolio" (p18), and office/retail units will be added to Investment Properties.


I'd be happy to buy it at a 5% yield, or HKD 18.  What really gets me excited is that this dividend is only 63% of their Investment Properties profits (1H20 numbers, annualised).

Theoretically - as a comparison to SG Reits - a 100% payout ratio with a 6% yield would give them a target price of HKD 23.


This company has a great long term track record, and is conservatively geared.  The risks come from the external environment, principally HK Real Estate, and China.

Hong Kong property prices have boomed for the past 10 years:

Source: Trading Economics

This is a bubble.  There are no signs of it deflating, and even signs the government may continue to blow it up.  It must pop one day - one way or another.  High property prices are one of the causes of unrest, and there's always a political risk the government will enact policies against property developers.  This line of thought is that HK's unrest is a structural problem, and one of the fixes required is lower property prices.

Another risk is the China bubble bursting.  The giant China bubble pops, and it becomes like Japan in 1989.  Or the Soviet Union.  The problem is best summarised here.  

Its a book.

I don't think "Work From Home" is a risk to this company.  I think that most people in Hong Kong will not work from home as their units are too small.

Medium term, I'm not sure where we are in the HK property cycle.  HK office rents have fallen for 6 consecutive quarters.


While Swire Properties and HKL gave concessions to luxury tenants, Wharf REIC did not, causing LVMH to plan to close its 10,000 sq ft Times Square store.

Monday, September 7, 2020

Ascendas India Trust

Quick look at this company.  

The Numbers

Long term, the numbers are excellent.  They have been growing.  In INR:

And DPU has grown too, meaning they did it without too much dilution:

The payout ratio has been 90% since 2013, and (I think) 100% before that.  I think it's the only SGX REIT/trust that keeps a little bit of its income for future growth.

Just to double check, lets compare the growth rates of debt, units issued and distributable income:

Most of their past growth has been funded by debt.  Their last big issue of new units was in November 2019, as a private placement.  They are still only at a 29% gearing, giving plenty of room to grow.

There is one catch: the Indian currency (INR) is constantly falling against the SGD:

Over ten years it has depreciated at a CAGR of -4.3% a year against the SGD.  Its like the INR has high inflation build into it - if you hold this currency, you are running just to stay in the same place.  AIT's growth has to outperform this depreciation.  When we look at the DPU in SGD, its still growing:

The bad performance from 2010 to 2013 was mostly due to currency.

The Properties

When I look at a REIT's property, I just look at two simple common sense things: is it freehold?  And where is it located, in the middle of a city or the middle of nowhere?

Here I look at AIT's top 4 buildings by value, and also describe their property location from google maps:

All of the property is freehold, or for long enough that it doesn't matter.  The important buildings, making up 75% of the Trust's property value are all in outlying areas that are quite built up - medium to high density, some with a little vacant land nearby.  They are what we would expect from business parks, the point here is to make sure they are not in the middle of nowhere.


The majority is due in 2022 and 2023:

Source: July 2020 Results Presentation

35% is denominated in SGD, as SGD loans have lower rates.    There's a risk here, if INR drops, the debt rises compared to the asset value.  If the SGD rises 30% against the INR in a few years - like it did in 2010 to 2013, then that 29% gearing ratio becomes 32%.

The Trust's rules specify at least 50% of debt should be in INR.

Tenants and Leases

WALE is 3.6 years:

The fine print at the bottom says the retention rate from July 2019 to June 2020 was 57%.  Pretty low.  If we assume most of those who left did so during covid (Jan onwards), the retention rate may be low till Covid is over.

There is some room for new rents to rise: each buildings' average rental is below its recently renewed rental rates:

(The numbers in red at the bottom are the percentages of each each building's property value for the trust).

They have high quality tenants:

  • 86% are MNCs
  • "Collections for office rents remain healthy with 99% of April, 95% of May and 92% of June billings collected." (slide 6)

The top 10 tenants account for 38% of rent, the top tenant accounts for 9%.


The biggest risk is "Work From Home".  In their latest update, the company says "Park population remains below 10% across all parks."  If the people in India figure out they can all work from home, less office space is required.  I believe most middle class Indians live in condominiums with a domestic helper for the children, so they should be able to WFM.

Next is the currency risk.  If we get a big INR depreciation like from 2009 to 2013.

Lastly, de-globalisation.  India is less affected than China, but the issue of outsourcing and H1B's may come to the attention of US politics.


Great long term growth, management, properties and tenants.  Manageable debt.

The currency risk is the biggest one.  No way of predicting these.  But at a yield of  over 6%, I think they are priced in.  I am wondering if I am comfortable with a 5% position or 10% position?

DPU from 1st April 2019 to Dec 2019 is 6.45c.  Annualised its 8.6c.  Thats a 6% trailing yield at $1.43 and a 7% yield at $1.23.


A good blog post that brings up some points about India's demographics, AITs Management Fee structure and currency:

A series of good posts by Snowball about competitors in the Indian commercial property marketIndian Interest rates & Lease terms, and comparison with Embassy REIT.

Saturday, September 5, 2020

Hong Kong Property Companies

 Quick notes on 3 HK property cos:


  - Yield 4.55% (at HKD 63, year ending March 2020 results), 100% payout ratio, mostly 50 year property leases

  - Neighbourhood shopping centers, a bit like FCT in Sg.  Largely unaffected by economy/covid.

  - Good LT track record growing assets without raising cash from shareholders, but yield too low for me, esp with 100% payout ratio.

Wharf REIC:

  - Yield 6.3% (at HKD 32.10, end Dec 2019 results), only 50% of CFO paid out, mostly freehold property. 

  - 75% operating profit from one mall: Harbour City.  Overall high exposure to luxury/tourism: 81% rental from leather, fashion or jewellery. 

  - Luxury brands are reducing their footprint in HK: eg: LV closing their Times Square store after Wharf refused a rent reduction.  Aug 19: (pre-covid) Prada's Causeway Bay landlord "willing to cut rent by 44%" after Prada leaves.

  - If Wharf's 2019 rents fall by 40%, CFO is down roughly 50%.  I think the market is pricing this in.

HK Land:

  - 65% of 2019 operating profit from investment properties portfolio, 35% from development.  Development is lumpy, so ignore it now.  Conservative estimate 515m operating profit (excl D&A (not clear?) and change in properties' values) from investment properties in 2019.  Or USD 22c/share.

  - Dividend is also 22c/share (in 2019 and 2018), thats a 5.6% yield (@ USD 3.89/share).  Mgt said they intend to keep it constant.

  - For investment properties: 83% operating profit from HK central, 11% from Sg.  Mostly office (banking/legal).

  - Stock is cheap based on recurring income from Investment Properties alone.  Their development are is highly leveraged, but lets them expand their portfolio without raising more capital.  So cheap value with some growth.  Worth looking further.

What I am doing

On Friday's market opening I sold all my trading portfolio at a small profit.  Think the correction will be short and sharp, maybe 2-4 weeks.  Sharp enough that I don't want to sit through it.  It is not a liquidity event like March, but the market was too frothy - some RobinHoodies and a big options-buying whale need to be cleared out first.

For the trading portfolio, I look to buy back copper, oil ad gold miners.  Plus maybe a US housing stock.

I kept my investment portfolio holdings, look to add Ascendas India Trust.  Hong Kong Land also looks interesting.

Saturday, August 1, 2020

What I'm doing

I'm now 95% invested.  85% if we count gold as cash.  Probably for the first time since this blog started.

There's always the risk of another market crash, but if we hold cash, we will lose to inflation.  The Fed is always printing.

Dividend Portfolio

Bought large positions in WMB and KMI several weeks ago, after news of Buffet buying into the sector.  These are boring, 'old man' utility stocks...but with some growth potential as their cashflows can fund expansion.

Also bought a half position in Delfi - the USD is no longer rising and threatening EMs.  More of a growth stock, but still pays a (trailing) 4% dividend.  Could grow.  Could fall too.

I am probably not yet earning enough in dividends to live off, but its getting close.

Growth Portfolio

This is all in commodity copper miners (SCCO, Lundin, COPX) and and oil producers (EQNR, CNQ), as inflation plays.  

Cash and Equivalents

I'd like to convert my gold to gold miners (GDX, GDXJ, FNV), but the miners are way overbought now.

Other Ideas

For the first time in a while, I have more ideas than cash!
  • For my dividend portfolio, I may add small positions in Ascendas India Trust or Link Reit.
  • Airports look interesting as cyclical plays.  We will get a recovery sometime, think it will be sharp when it happens.
  • Crypto looks interesting, as a small speculation.

Friday, July 3, 2020

US Pipeline Companies: Part #2

Williams Companies

Their business: Mostly Natural gas, they break down their segments by geographic region:
  • West: Gas gathering, processing and treating in several Western US shale oil fields.  27% of 2019 EBIDTA.
  • Northeast G&P: Gas gathering, processing and fractionalising in the Appalachians.  30% of 2019 EBIDTA.
  • Transmission and GOM: Transport along their Transco pipeline, with a little gas/oil gathering in GOM.  42% of 2019 EBIDTA.
Transco is irreplaceable - more than one fifth of US natural gas consumption flows through it.  for Northeast G&P, Morningstar estimates that Williams collects about a third of overall gas volumes across the Appalachian region.  Their other assets are more exposed to market forces, especially declining crude.

West could be badly affected by crude oil shut-ins, as 30-40% of US natural gas production is associated with crude oil (the gas is an unwanted by product).  Northeast G&P would not be affected  and may even benefit if gas prices rise due to a falls in associated gas.  Transco would likewise be unaffected or benefit.  I believe more than half of GOM's natural gas production is associated with oil wells, so they may be hit too.

Leverage: 2019 debt was 5.4 times EBIDTA.  Very high.  They aim to reduce it to 4.2 times.

For operating leverage, 2019 CFO was 40% of revenue (excluding product sales).  ie: Service revenue would have to fall by this much before they start losing cash.

Valuation: Trading at a 7% trailing yield, with CFO at 1.8 times their dividend.

Growth: Still growing.  Capex in 2018 was 4.2bn, 2019 was 2.4bn, 2020 is expected to be 1.5bn.  For comparison, 2019's CFO was 3.6bn.  They have not announced further cutbacks to 2020 capex.

Near term they have earmarked 3.2bn:

Longer term they have other opportunities:

Management recently said they expect to have "positive free cashflow" from now.  Specifically: without any asset sales, their operating cashflows should support both their capex and dividends.

Long term or political risks: Don't see any.  Natural gas is environmentally clean and has low carbon emissions.  Oh wait...all fossil fuels are bad - one of their proposed pipelines was just killed by NY.

Worst case scenario:  Again, lets say US crude production halves. I'm going to assume that all gas production in their 'West' segment is associated with oil (1) (2).  So if gas production halves, revenue halves, and the company's CFO drops by 900m or 25%, to 2.8bn.  After the projected capex, they would have to cut their 2020 dividends to ~ $1.05,

Kinder Morgan

Their business: Mostly Natural gas:
  • Natural Gas: interstate and intrastate natural gas pipeline and storage systems, gathering, NGL fractionation, and LNG liquefaction & storage.  57% of EBDA.
  • Refined Pipelines: pipelines that deliver refined product, and some crude.  Including some terminals and mixing facilities.  15% of EBDA.
  • Terminals: Terminals, and Jones-Act qualified tankers.  18% of EBDA.
  • CO2: produces, transports and sells CO2, used for crude oil production.  8% of EBDA
KMI owns the Tennessee gas pipeline, one of three large interstate pipelines supplying Eastern US natural gas.  The company says that 40% of US gas passes through its pipelines.  They would be considered irreplaceable,

CO2 would be badly affected by crude oil demand, while natural gas gathering would be affected by oil production.  Refined pipelines would be affected by covid.

Leverage: 2019 debt was 4.8 times EBIDTA.  Pretty high.

For operating leverage, 2019 CFO was 60% of revenue (excluding product sales).  ie: Service revenue would have to fall by that much before they start losing cash.

Valuation: Trading at a 7% trailing yield, with 2019 CFO at 2.2 times their 2019 dividend.  It would cover the newly raised dividend ($1.05/year) by 2 times.

Growth: Still growing.  They announced they were reducing 2020 capex from around ~3bn to 2.2bn.  For comparison, 2019's CFO was 5bn.

Their new projects before the announced reduction are here (slide 13).  The reduction was probably in CO2:

Long term or political risks: The usual ESG stuff.

Worst case scenario:  Again, lets pessimistically say US crude production halves.  And KMI gets corresponding reductions in revenue as customers go broke.

For Natural Gas, how much would be affected by the halving crude production?  They say that gathering and processing was only 10% of EBDA (slide 21):

I can't relate this to revenues, but assume it cashflows from this fall to zero, as revenue is halved.    Thats an 800m reduction in earnings/cashflows.

For CO2, its easier - halving 2019 revenue subtracts 600m from earnings/cashflows.

So total, we lose 1.4bn from cashflows, which is now 3.6bn.

This gives us enough to cover the 2.2bn capex, but not enough to cover the 2.4bn in dividends - the $1.05 dividend would have to be cut to 60c to be cashflow neutral in 2020.

Lets also say that their refined revenue halves over 2020, due to covid.  This is realistic - not pessimistic - but its only temporary.  KMI's refined revenue also halves, as refined fees are usually volume based.  That removes 900m from their Products revenue, they will still have 500m they could use for dividends, or around 20c per share.


I like Williams and KMI the best, as natural gas is not affected by covid, and is only partially affected by crude.

Although both have high leverage, I can't see any way either company fails.  The dividend is comfortably covered under normal conditions.  And both have growth potential.

KMI has slightly more short-term downside due to more refined exposure.

In general, these types of businesses are predictable and profitable.  Its the black swan risks to watch our for (like Deepwater Horizon).  Or the political ones (Green New Deal).

Wednesday, July 1, 2020

US Pipeline Companies: Part #1

US pipeline companies are trading at 8-10% yields.  When stocks trade at that valuation, usually it means theres something wrong with them.  But these are exceptional times.  Lets take a quick look.

All of these companies own pipelines, with most revenue from long term take-or-pay contracts that don't depend on volume.  Their revenues should hold up - unless their customers go bankrupt, which is what the market is concerned with.

I'm trying to see how badly their businesses may be affected by the plunge in crude prices.  The crude market looks terrible now, I expect US crude production to drop in a seesaw pattern for 1-2 years, then grow.

Magellan Midstream

Their business: 62% of 2019 profits are from Refined Products (pipelines from Texas/GOM, up to Wisconsin/North Dakota/Wyoming), 38% from crude (pipelines covering Permian and GOM).

Their Refined business is large enough to be irreplaceable: they provide more than 40% of refined product in 7 of the 15 states they serve, and can access nearly half of nationwide refining capacity.

Refined usage has taken a beating due to covid.  I expect it to get worse for the year - the previous outbreak hit New York, the next outbreaks will equally spread in rural/red states. Probably get a recovery in 12-18 months, as the US situation changes from lockdowns into a normal recession.  For now, refined product demand has bounced a little after falling off a cliff:

Source: Macrovoices #122 Art Brennan, slide deck

Leverage: 2019 debt was 3.1 times EBIDTA.  Pretty low.

For operating leverage, operating expenses (excluding D&A, including interest and G&A) were about half 2019 transport/terminals revenue.  ie: revenue would have to fall by half before they start bleeding cash.

Growth: Limited.  They are paying out most of their cashflows as dividends.

Valuation: Trading at a 9+ percent trailing yield, estimated payout ratio 90-100%.

Long term or political risks: Gradual decrease in fossil fuel usage, replaced with electric cars/trucks/planes.

Worst case scenario: Lets say US crude production halves, and Magellan's crude revenues (620m in 2019) halve with it (half their customers go bankrupt).  Rough guess, this removes 300m from their profits, reducing the CFO by the same amount, and dividends by 1/3rd (slide 10).   So even if you think US crude production is permanently and badly impaired - which I don't - the dividend is still 6%.

Long term, I don't see Refined dropping.  They are under long term take-or-pay contracts.  The problem is if their customers go bankrupt during the lockdown period.  I estimate lockdowns will be on/off in different states for another 18 months.

I can't see any risk to this company's long term prospects, and I think their cashflows/dividends may take a temporary hit before going back to 2019 levels.  Short term, if their dividends drop, so does the share price.  Long term, its a cyclical play where you're paid to wait.  The only downside is limited growth.


Their business: 2019 EBIDTA breakdown:
  • Liquids Pipelines.  A series of pipelines transferring crude through Canada to the US.  Also covers most US shale basins, handling 25% of all US crude.  Tolls on the Canadian Mainline are based on volume (pp14-15), tolls on the US interstate pipelines are long term take-or-pay.  56% of 2019 EBITDA.
  • Gas Transmission and Midstream.  Long continent-spanning pipelines transferring gas from Canada to the Vancouver/US, and from US producing fields (especially the Marcellus) to consuming states.  Take-or-pay.  25% of 2019 EBITDA.
  • Gas Distribution and Storage.  A regulated utility consisting of last mile distribution of gas to Canadian households.  Also an unregulated storage business.  13% of 2019 EBITDA.
  • Others.  7% of 2019 EBITDA.
They are heavily exposed to crude oil volume:
  • Its 56% of EBITDA.  
  • Canadian Mainline pricing is volume based, so lower volumes will be felt immediately. Morningstar estimates that Mainline accounts for 30% of EBITDA.
  • It will have new competitors: TCE's new Keystone Phase 4 is expected to be operating in 2023 (unless Trump loses) and will compete with them.  Same with the Trans Mountain pipeline.
  • Meanwhile, Canadian crude output has fallen due to low prices.  Canadian crude usually trades at a large discount to WTI, because it is landlocked, and must go to the US for refining.
  • Canada's crude production is also energy intensive (like using steam to melt bitumen), so have high fixed costs, despite unlimited reserves.
  • So I think Canadian crude producers are marginal producers.  They may do very badly during this crude crisis.
Enbridge's US interstate gas pipelines (TETCO) distributes gas from the Marcellus down to the Southern US, and is irreplaceable. It would also be unaffected by crude.

So we have 30% of EBIDTA serving marginal (ie: Canadian) producers on volume contracts, in a market crunch, with long term competition coming up.  And another 26% on the US side under take-or-pay (where we would only be worried about customer bankruptcies).  With the 38% of EBIDTA being stable.

I don't see this as a dividend stock, but more cyclical, levered to WTI (actually to the WTI and WCS differential).  If I want to play an oil price recovery, its probably better to buy Enbridge's customers (CNQ & Suncor).  Right now I'm looking for steady dividend stocks.

Enterprise Products

Their business: Their business segments by 2019 Gross Operating Margin (similar to EBITDA - p83) is 49% NGLs, 25% crude, 13% Natural Gas, 13% petrochemical & refined.

What are NGLs?  Natural Gas Liquids extracted from natural gas at the wellhead.  Different NGLs have different uses, in industrial, heating and transportation:

Source: EIA

In general, the 'wetter' the gas, the more NGL's it has.  So dry gas (from the Marcellus) has less NGLs than associated gas (from shale oil):

Source: IHRDC training course: Gas Processing and Fractionation.

EPP's NLG pipeline covers most shale (oil and gas) in the west and south US:

Source: EPP System Map

EPP's NGL processing revenue is currently mostly fee based (top of p5)... ie: based on volume, with contracts lasting one to ten years.  Their NGL pipeline revenue is also volume based (p7).  Their NGL fractionation revenue is a mix of volume, and commodity prices (bottom p11).

Given these, their EPP's NGL earnings are dependent on crude prices.  First, as associated gas production falls, NGL production falls with it, affecting their volume based payments.  Second, associated gas has more NGLs than dry gas.  The effect on falling crude on NGL supply/demand is very complex.

Lets look at crude (13% of Gross Operating Margin).  Their crude pipelines cover the Permian, Eagle Ford, Haynessville and GOM, with pipes running to Cushing.

I'm guessing that most of their crude is from shale, and will decrease as prices drop.  Their crude pipeline profits are volume-based (bottom p15).  So dependent on crude oil prices.

Their Natural Gas assets (13% of Gross Operating Margin) seem to be around shale oil fields (except for GOM):

And profits are volume-based (bottom p20), so this is again dependent on crude prices.

For Petrochemical and Refined (13% of Gross Operating Margin), they do not say specifically, but I guess they are dependent on the economy.

I like EPP's long term track record, but all their business segments could be exposed to falling crude production.  I can't estimate how much.  Too risky now.  I may look at them later as a cyclical play if they are hit by falling crude production.

Friday, June 26, 2020

Oil Prices

Two MacroVoices interviews with Dr Annas Alhajji and Art Berman give an optimistic long term view on the crude oil market.

 Dr Annas Alhajji (22 June 2020):

  • The surplus is now 180m barrels in inventories in OECD (excludes China), Saudi Arabia is trying to eliminate it.
  • Comparing now and 2017: in 2017, oil inventories decreased 152m bbl in 10 months,.  This was done by the Saudis cutting production, US oil production increased by 1.2m barrels per day in that period.
  • This time, the rest of the world is cutting production.  US production is down by 1.8m/day compared to 2017.  Libya is at 30-40,000 barrels/day, down from 1m in 2017.  Venezuela now 600k down from 1.9m.  Iraq exports are half of 2017.  Brazil, Norway, Ghana have increased, but small amounts.  Overall, we will get serious supply side issues in the future.
  • Estimates the rebalance of the oil market will take a year.  Picture is way brighter in the next few months than people believe.
  • Shale: only 25% of oil has being brought back, (due to price differentials between WTI and oil areas), once the differentials improve, will see major comeback.  Expects all major companies shale wells to come back online, but there is a problem due to lack of new drilling, needed to offset rapid shale declines.  Major decline may last for 2 years.  Thinks will bottom around 9.8-10m US production, then see a recovery.
Art Bernan (18th June 2020):

  • US Production is down from 13.2m bpd to 10m
  • Unlikely to see negative WTI again.  The govt has opened up strategic reserve space (to lease to store oil).  And the market should resolve any issues if it happens again.
  • Can oil production be switched on and off immediately?  For shale (45% of US production), yes - barring occasional repairs, you can switch on/off production with a few clicks on an ipad.  For conventional, no. 
  • The  current rally is just a relief rally, too many people were short oil.  There is still too much oil around (slide 11).
  • Decline rates have been increasing with newer wells. 
  • Time from rig to first oil production is around 10-12 months.  Estimates 16 months lag from when oil prices rise to make shale profitable, to the time the first shale can be drilled.
  • So he expects the oil price to recover, longer term: "And that next down is going to be a buying opportunity. Because it sounds to me like maybe there’s a few more waves up and down along the way, but eventually we get a moon shot when there is a full economic recovery from this crisis and the industry is just not able to respond quickly enough."   [My notes: And I guess the recovery could be around 16+ months....]
  • Slide 13 shows current recovery in oil consumption so far.  It has not recovered yet.  And the recovery has been mostly in gasoline.

Saturday, June 20, 2020

What I am doing

Similar to last time:
  • For my dividend stocks, added Frasers Centerpoint for full position.  Now ~43% invested (all in dividend stocks).  The dividends are not yet enough to live off.
  • For the remaining ~40% of my portfolio allocated to macro, half is in gold (AAAU) and TIPS (ETF).  I consider these to be like cash.  Small 2% position in COW.
My trades are recorded in my Investing Note account.  This blog is for in-depth analysis.  I am now working again after a short time off, so less time to post here.

I think we get a correction or crash soon.  I'll look to add:
  • BIP plus some Singapore REITs (MCT, FLIT).
  • I think we get stagflation.  If the market corrects, then stocks in gold miners, copper miners, oil producers and fertiliser/farming companies look interesting.  Longer therm, my REITs should do OK in stagflationary periods (low interest rates and increasing nominal property prices, offset by lower rental demand).

Quick thoughts about inflation.  There are 3 causes:
  • Too much money printing.  Like adding extra zeros to your currency.  Zimbabwe.
  • Too much demand.  Overheating economy when production at full capacity.  Not now.
  • Supply constraints.  Not enough investment (eg: finding new copper/oil reserves).  Production cost increases (de-globalisation).
I don't know exactly what form the inflation will take, or how it will be transmitted through the economy.  Too much detailed theory doesn't help you trade.  I look to play it broadly through holding real estate (REITs), commodities or commodity producers, and things that don't devalue with paper currency (gold, TIPS).  Residential housing may be another thing to consider.

Wednesday, May 27, 2020

Bought Gold (AAAU)

Have bought 10% gold in my portfolio, or around SGD 80k.  Average price is USD 17.058.

I am holding the Perth Mint Physical Gold ETF (AAAU).  It is backed by the West Australian State Government, and (theoretically) allows retail investors to redeem gold.  GLD only allows you to redeem 100,000 shares, and the gold is stored with HSBC.

Day to day AAAU moves in line with GLD.  It has a lower expense ratio (0.18% compared to 0.4%), so it degrades less:

AAAU is liquid enough to buy or sell occasionally, but if I was short term trading, I would use GLD.

Why buy gold?

  • Its in a long term uptrend, since late 2018
  • It will do well in a deflation environment (recession this year), and with stagflation later (inflation from money printing but no growth).
  • I am uneasy buying long term Treasuries with rates so close to zero.  We do not know if the Fed can allow negative nominal rates.
The time to sell will be when we get expectations of economic growth.  No sign of that now.  I may hold this for months, quarters, or even years.

Gold is money that can't be printed.  Before I become a die-hard gold bug and doomsday prepper, I remind myself of the risks:
  • Paper gold fell in the March panic.  Everything fell against USD and physical gold.
  • There is a constant counter-party risk: you never know if you can redeem until you do it.
  • When we start getting real economic growth again, gold once more becomes a 'barbaric relic'.

For the rest of my portfolio, I am still waiting for the current bear market rally to die before I begin to buy.  May be waiting a long time.

Thursday, May 7, 2020

What I am doing

I am using two strategies: 'dividend stocks' and 'market timing'.

Am now 40% invested, all in dividend stocks/REITs/trusts, mostly SGX listed. The big ones are Netlink Trust and Manulife US REIT, with smaller positions in Frasers Centerpoint Trust and Mapletree Commercial Trust.

I think the last month's rally is a bear market rally.  I am waiting for a correction to continue buying Frasers Centerpoint Trust and Mapletree Commercial, plus Brookfield Infrastructure Trust.  This will take me up to 60% invested.

I'm willing to catch falling knives when buying dividend stocks.  Even though they will be affected by the recession (except Netlink), all the SGX-listed stocks above should survive without raising capital, unless revenue drops by more than half.  Once the economy recovers in a few years, I should have a dividend stream that I can live off.  Then figure out what I want to do with the rest of my life.

For the remaining 40%, I follow Hedgeye for market timing. They have called the cycle well, moving to bearish in mid Feb, and warning of a likely 20% correction on 4th March.  I wait for them to call a turn in the cycle before buying.

I've been covering a lot of stocks here lately, so I have a list of things to buy when the market does turn.  The shopping list:

  • Inflation plays (growing inflation, slowing GDP): Oil (Equinor), TIPs, maybe Natural gas cos.
  • Growth plays (growing GDP, slowing inflation): Copper (SCCO),...maybe the travel companies (Groupo Aeroportuario, Booking, Safran, Rolls), maybe Delfi
  • Growth+Inflation plays (growing GDP, growing inflation): Interest rate plays (Banks, Computershare, Interactive brokers), maybe Oil above.
  • The travel companies may also be buys after some uncertainty from the virus clears up, then we are just dealing with a normal recession.
If you put a gun to my head and forced me to buy something today, it would probably be Delfi, Groupo Aeroportuario and Berkshire Hathaway.  I haven't found any high quality stocks that are cheap.

Sold Straco

Sold two days ago at a small loss of SGD 400.

1) Want to buy BIP instead.  Wait for a correction first.
2) Straco take longer to re-open that I thought.  Even when they reopen, will be at reduced capacity.  This is a company that has high fixed costs and relies on crowds.  China's domestic travel may be recovering, but I can't see international travel recovering.
3) There is an increasing political risk of anything connected with China.

Used the current rally to sell Straco.  Evenafter one month, I still think it is a bear market rally.  I wait for a correction to buy BIP.   Waiting is the hardest part.

Tuesday, May 5, 2020


Crude oil trading at negative prices is not sustainable.  How do I play it when it recovers?

Equinor (formerly Statoil) is Norway's state owned oil company.  It has very low production costs, operates in stable jurisdictions, and has reasonable finances.

I got this idea from Vitaliy Katsenelson's Contrarian Edge.

All numbers are from their 20019 Annual Report ending December (pre-virus, and pr oil crash).


What do they do, and where are they exposed to?
  • Almost all their profit came from Norway E&P, which consists of 70% oil, 30% Natty & NGLs (by revenue).
  • Next are their operations in the US (p38), which produced around 10-20% of their oil, natural gas and NGLs.   What and where do they produce?  By revenue, most should be GOM oil.  By BOE, most is Marcellus natural gas.  There is very little shale oil - there's some from the Bakken, but their Eagle Ford assets were sold off in a well-timed November sale.
  • Their International E&P (including the US) made a small loss.
  • This is offset by their Marketing, Midstream & Processing making a small profit.
  • They have a stake in 7 wind farms.  Makes for nice ESG photos, but meaningless for profits.
  • They own an insurance company!  I think it insures the parent against operational risks (eg: workmen's compensation).  Again, not meaningful for profits.
So basically: Norwegian offshore oil production, followed by US and international (mostly) offshore oil production, followed by natty.  Minimal US shale oil.


Net debt (including IFRS 16 Leases) are 22bn, of which 4bn is due this year.  Another 4.2bn is due in 2021 and 2022.  They recently raised 5bn very long term debt on good terms.  I am ignoring long-term financial investments, which are required for insurance.  2019 interest payments were 1.5 billion.  All debt is fixed rate.  Its in a mix of major currencies (p198), but "normally swapped into USD".

2019 operating profits covered interest payments by 6 and a half times.  CFO covered it 15 times.

Despite the large debt, I don't think they have to issue new shares.  I think their status with the Norwegian government allows them to issue debt cheaply.

Net Debt has dropped from 33bn in 2016 to 22bn now.

Its unclear to me how much of their production is hedged.

Production Costs

They state they have breakeven prices for different projects between $11 and $40.  Not sure if that is extraction or full cycle costs.

In March they stated they can be "organic cash flow neutral before capital distribution in 2020 with an average oil price around USD 25 per barrel for the remaining part of the year".  I interpret this to mean their overall cash breakeven is $31 Brent.


Charting their oil (not BOE) reserves:

Relationship with oil price

EQNR's stock price seems to follow Brent:


Good company that meets my criteria:
  • Operates in stable, lawful jurisdiction.  No geopolitical risk.
  • Low cost of production
  • Minimal exposure to US Shale oil.  Product priced in Brent (seaborne), not WTI (landlocked).
  • Reasonable financials, although a bit too much debt.
This is not an exhaustive look, I won't buy-and-hold forever.  And the oil industry is very difficult to study.  Who really understands break-even costs, or accounting rules for reserves and E&P?  If I buy, it is just as a trade for rising oil prices.

I would prefer to buy a basket of oil companies, but am unlikely to find many (any?) that fit the above criteria.

I use Hedgeye for timing when to buy.  I am not buying now.  The economy is still shit.  Oil producers may keep overproducing for a long time because of fixed costs and hedges.  And there's a chance Brent crashes, maybe below zero, after all the oil tankers fill up.


The Norwegian government owns 2/3rds of Equinor.  Norway has the worlds largest sovereign wealth fund, so I don't think they will push Equinor to pay dividends.

If I buy, I have to buy the ADR since Interactive Brokers does not access the Oslo stock exchange.  One ADR represents one share.  There is a fee of 0.5c per ADR for each dividend distribution.  Interactive brokers has Equinor as STL (Statoil).  Witholding tax for dividends should be 15% for Singapore residents (p8).

Friday, May 1, 2020

Brookfield Infrastructure Partners (BIP)

BIP is a trust that invests in worldwide infrastructure and pays out the cashflows as dividends.

Business Model

They raise money through debt or equity, then form joint ventures as a minority (but controlling) partner to invest in ports, rail, roads, power transmission, gas lines and others.

Brookfield likes to buy distressed assets, for example:


I want to check their long term financials to see how profitable successful their investment have been.  How do we do this?

Their consolidated financial results are meaningless because they hold small controlling stakes in many joint ventures  We need to look at their results on a proportionate basis instead (ie: their share of profits, debt, etc) from their supplemental information.  They key figures are below.


What returns do their investments generate?

Look at cash generation rather than asset values, since cash is more objective.  'Adjusted Funds from Operations' is 'Funds from Operations' minus maintenance capex.  Its not an accounting (IFRS) number, just an estimate they give us.  But most accounting numbers are estimates anyway.

AFFO has risen nicely:

The maintenance capex estimates for their different segments are interesting.  eg: Road and Rail are high maintenance, Energy Transmission almost none.  Useful for when looking at other infrastructure trusts.


One source of funding for investments.

They have high debt, but 88% of it is non-recourse (at the project level only, the company is not liable for it).  The non-recourse debt seems to be fixed rate (F-75 to F-77).  The corporate level debt is tied to LIBOR.

Debt has risen in line with AFFO:

Non recourse debt means their debt is essentially unlimited.  Each project, if it is run well, supports its own debt (and equity too...?).


Along with debt, how much equity have they raised for their investments?

To simplify: they have two types of units: normal units and preferred shares.  Both pay dividends.  The 'normal units' are the majority: they pay around 95% of the dividends:

  • The 'normal' unit pay distributions based on FFO.  The company targets 60-70% payout of FFO.  818m distributions were paid out in 2019
  • Additional 'Incentive Distributions' are also paid out to Brookfield Asset Management based on the normal distributions above.  Generally, 25% of any additional distribution is paid as IDS - see page F-92 (eg: If next year's distribution/share is $1 higher, an additional 25c is paid to Brookfield).    158m Incentive Distributions were paid out in 2019.
  • Preferred distributions pay out a fixed dividend for a certain period of time.  Currently issued units pay out 4.5-5.4%, expiring June 2020 to Dec 2023.   49m fixed dividends were paid out in 2019.

Putting it together

Most of their funding is from issuing equity.  From 2012 onwards, the cumulative money raised:

Since the preferred dividends are fixed, we deduct them from AFFO first.  They only make a small part of AFFO:

And since they issue a lot of 'normal' units, we want to look at this number (AFFO minus preferred dividends) on a per unit basis:

This is the key metric for measuring their performance.  Very impressive.

Asset Lifespan

Some of their assets have short lifespans, in particular:

  • Their Brazilian regulated gas transmission operations (NTS) form around 30% of proportionate 2019 CFO, and expire in 20 years:

  • DBCT has a remaining concession period of 30 years, and forms 8% of their CFO.
  • Westrail has 29 years and forms 4% of their CFO.
  • Their road tolls (Peru, Chile, India) form 4% of CFO and have an average of 15 years (p88). 
I estimate they lose an average of around 2-3% of their CFO (or AFFO) a year due to asset ownership expiry.  Over the next 30 years.

I am only looking at when legal ownership of the asset expires here.  I'm not looking at how much money they earn or when customer contracts are renegotiated.

Payout Ratio

How much FFO or AFFO is paid out as dividends?  The payout ratio has risen over the years:

At 94% in 2019, there is not much space for it to rise, especially when we account for the 2-3% every year needed to make up for expiring assets.  So long term dividend growth will be slower over the next ten years than the past ten.

This payout ratio covers all distributions, including preferred dividends and incentive distributions.

They say they target "retaining 15-20% of FFO", and a payout ratio of 60-70% FFO.  Maintenance capex is generally 20% of FFO.  Anyway, the payout ratio now looks sustainable, but I don't think they can raise it.

They note that their 2019 FFO "payout ratio exceeded our target range of 60-70% due to the time between raising and deploying the proceeds from the July equity issuance and the depreciation of the Brazilian real. Removing the impact of these two factors reduces the payout ratio to 70%".  And the 2018 ratio was affected by "the 2018 financing at our Brazilian regulated gas transmission business".


Some of their sectors, particularly ports and natural gas pipes/processing are market driven.  Others (eg: UK residential connections, toll roads) are regulated, but affected by the economy or volume.  Others, like electricity transmission have fixed revenues.  95% of their 1H 2019 cashflows are regulated or fixed.

Theres a chance that North American natural gas throughput drops a lot over the coming years, due to a shale oil production decline.  This forms about 7% of 2019's CFO.

Negative rates in Europe and Japan make more everyone willing to invest in infrastructure (or anything, for that matter).  Spoils the market.

They do not have direct exposure to China.  They have indirect exposure in some Australian operations (DBCT and Westrail).

Their financials are fucking impossible to understand due to consolidation.  Its hard enough to just list their assets.


This company has a good track record based on AFFO/unit.

I would probably be happy to buy at a 6% trailing yield, even before paying the 15% tax on dividends from a Canadian company.

They aim to increase dividends by 5-9%.  I would take the bottom range of this due to the currently high payout ratio.  5% growth is reasonable based on their track record.  They do not have gearing limitations like a REIT, so could grow forever.

There's a risk the dividend yield drops this year due to the economy.


Brookfield bails out Donald Trump's son?