Saturday, May 23, 2020

Key takeaways from MOOCs (Part 1)

During this period, plenty of MOOC providers are offering their paid certificates at no cost to learners. I have been racing against time to complete as many courses as I possibly can. I will be sharing the key takeaways from the courses I have taken in this, and subsequent, posts.

For those interested, please refer to this post by Class Central for a compilation of MOOC resources.

Blockchain for Business (NTUC LearningHub & GO1)

  • Blockchain and its impact on Commercial and Investment Banking functions
  • Commercial Banking:
    • Blockchain as an alternative for the unbanked to store their money
    • International money transfers could be sped up using the blockchain
    • Using smart contracts for escrow services, buyers and sellers do not need to use banks as intermediaries
    • Using blockchain for post-trade settlement to reduce settlement time
  • Investment Banking:
    • Capital markets: Equity and debt securities could be represented as tokens on the blockchain for easy clearing and settlement
    • Advisory services: too early to tell how the industry would develop
    • Asset Management: Gaining recognition among institutional and retail investors (e.g. bitcoin futures)
  • Blockchain and its impact on Tech companies (Google, Facebook, Amazon, Apple)
  • Blockchain and Google:
    • Google Search: Very hard to disrupt Google search dominance. Brave Software is attempting it with the Brave Browser and its native token, Basic Attention Token (BAT)
    • Youtube: Similar to Google search, it is hard to compete with Youtube for dominance. Disrupting Youtube requires providing strong incentives for creators and viewers through the offering of tokens. DTube from the Steem blockchain is an alternative to Youtube
    • Adsense: Steemit empowers creators to retain more of their earnings instead of Google taking a cut from it
  • Blockchain and Facebook:
    • Of the 4 tech companies, Facebook is the firm that is most threatened by blockchain technology the most
    • Facebook: Steemit empowers creators to retain more of their earnings instead of Google taking a cut from it
    • Instagram: Same as Facebook above
    • Facebook Messenger & WhatsApp: Blockchain-based messaging apps will challenge their dominance
  • Blockchain and Amazon:
    • Biggest disruption to Amazon is decentralized peer-to-peer marketplaces built on the blockchain. One example of decentralized marketplace is OpenBazaar, which has zero platform fees and does not control user data. It is less feature-rich compared to Amazon
    • Amazon Web Services provide very competitive pricing and is reliable. Will be difficult for blockchain storage services (Filecoin, Storj, etc) to compete with Amazon
  •  Blockchain and Apple
    • Among the 4 companies, Apple is the least threatened by blockchain technology as it deals with hardware technology as well
    • Apple’s centralized ideology is in conflict with the decentralized ideology of blockchain
  • Course syllabus is already outdated as the blockchain space moves very fast

Introduction to Personal Branding (Coursera)

  • Creating Alerts in Search Engines allow you to keep track of keywords and phrases of your choosing; users will be notified of any mentions of the keywords/phrases on the given search engine via email
  • Importance of creating your own “board of directors” whom you would consult for advice/as a sounding board for your ideas
  • Preferably, the number of board of directors should be an odd number in order to resolve ties in decision-making
  • If you do it solo, limit the number of social media platforms to two
  • Research the various social media platforms before launch. Ensure that the selected social media platforms are home to your target demographics before proceeding
  • Do not just copy/paste your social media updates across the various social media platforms you are on. Followers who follow you across multiple platforms will be notified of the same update multiple times without any value-add. Tweak them to differentiate them from one another
  • In your updates, remember the 80/20 rule. 80% of your updates should be content that is of interest to your followers. Cap self-promotion to the remaining 20% so as to prevent yourself from oversharing about your achievements which your followers may not be interested in

Cloud Computing Basics (Coursera)

  • Cloud computing allows firms to save cost on hardware by renting only the amount of computing power they need from cloud providers
  • Function-as-a-Service (FaaS) provides app development environments in the form of microservices. Customers only pay for the time that the environment is used. There is no storage for FaaS. Hence, if data needs to be stored, it would be at an extra cost. FaaS also requires a few seconds of start-up time
  • Docker provides OS environments called containers and applications could be installed on these containers. Containers run independently and it could be difficult to keep all containers synchronized
  • Kubernetes automates app deployment, scaling and maintenance. It supports Docker containers as well
  • Edge computing is a variation of cloud computing where computing services are brought physically closer to the customer. Popular in gaming, IoT, and healthcare where response time is important

Marketing Analytics (Coursera)

  • Brand value could be destroyed if the acquirer does not know how to manage the acquired brand within their existing portfolio of brands (e.g. Quaker acquiring Snapple and treated it the same way as Gatorade; quirky products, Wendy the Snapple lady, and mom-and-pop distributors for Snapple in contrast to Gatorade for sports)
  • Brand Personality: what if the brand is a person? What age would the brand be? Would the brand be male or female? What traits would the brand have?
  • Interbrand Brand Valuation Model: Interbrand is a consulting company that came up with a ranking system to rank brands. Their valuation model is used for mergers and acquisitions
    • It comprises of three aspects: (a) financial analysis of balance sheet for residual earnings, (b) market analysis of the sector the brand is in, and (c) brand analysis through conducting surveys on consumers’ perception of the brand. Brand value is then determined through future earnings and the volatility (risk) of future earnings materializing
  • Young & Rubicam Brand Asset Valuator: More for marketers, based on consumer surveys
    • Four elements of a brand: (a) relevance to consumers, (b) product differentiation from competitors, (c) esteem of the brand by the public, and (d) consumers’ knowledge of the brand. Each brand will have varying levels on those 4 elements
  • Importance of knowing brand value: have to know the trade-offs in marketing investment between long-term brand value vs short-term price promotions
  • Revenue premium as a measure of brand equity: attempts to combine Interbrand Brand Valuation Model and Young & Rubicam Brand Asset Valuator. More suitable for valuing grocery store products. Central idea is to identify the premium that a brand earns over a private label
  • Customer Lifetime Value (CLV): the dollar value of an individual customer relationship with the firm. Use past data in order to project forwards. Allows the firm to make marketing investment decisions (e.g. if the cost of acquiring or retaining customers is greater than the value customer brings to the firm, it would not make sense to proceed)
  • For marketing experiments, have to try your best to make sure the test and control groups are comparable. If possible, capture the pre-existing differences (if any) between the test and control groups
  • The internet age has helped marketing experiments to flourish. Experiments could be done quickly, with minimal cost, and allows for the manipulation of many variables. Moving from the brick-and-mortar world to the digital world has resulted in marketing cost changing from a fixed cost to a variable cost, allowing firms to have more flexibility in their marketing investments
  • Online marketing campaigns can have a carry-over effect in sales even after the campaign has ended
  • A marketing mix regression model could include variables such as Product, Price, Placement, and Promotion (4 Ps of Marketing)
  • Important to differentiate between statistical significance and economic significance in regression models used in the marketing context. Economic significance refers to whether the benefits from the marketing campaign outweighs the cost of the campaign. Statistically, economic significance is known as “effect size.” It could be determined by plugging in numbers into the regression equation and identifying whether the benefit is greater than the cost of the marketing campaign
  • Tips in modelling retail data (e.g. grocery stores): running regression analyses by customer types (e.g. a regression model for price-sensitive customers who tend to use government subsidies/coupons, reduce prices for them, and reach out to them)

Monday, April 6, 2020

Q1 2020 Portfolio Update

It has been a while.

Assignments have been submitted and exams are up next (or would they be postponed in light of the current virus situation?).

The important question that I have asked myself is what sort of pattern do I see the markets and the real economy settling into. I have come up with three possible scenarios:

  1. A V or a U shape recovery
  2. A GFC-like scenario
  3. A Great Depression-like scenario

My focus is on its impact, not on the probabilities of each scenario. That is, if a given scenario materializes, how should I conduct my life such that I am not blind-sided and caught off-guard. Personally, I lean more towards scenarios 2 and 3 due to my reading of both bullish and bearish investment articles. 

Scenario 2 is what I have (kinda) prepared for for the past couple of years. A balanced mix of cash (excluding emergency funds since they are accounted for separately), precious metals, and equities. I have indirect exposure to bonds as well through the giving of SSBs to my mum. My equities allocation will obviously be destroyed in scenario 2, but that is a given and I have made peace with it a few years back. There are ways to reduce the impact of equity destruction. The first is to take partial profits (which I have done with varying success in the past 2 years) and the second way is to market time, which I decided not to do as an income investor.

In terms of the impact of scenario 3, it would be all the more devastating. If I were to give my honest opinion, I think, by then, most people would be having trouble with daily subsistence; investing would be the last thing on their minds. To mitigate the impact of scenario 3, I have been building up an emergency fund for my mum and myself, a concentrated position in precious metals, and (soon) a concentrated position in cryptocurrencies. This is not new; regular readers would know that I have been building up my emergency fund for a couple of years already and that I have a sizeable position in precious metals as well. If scenario 3 unfolds, we would be hurt, but we wouldn't be the first to hit rock bottom.

But before we get into the knitty-gritty, we'll start with the usual.

Dividend Income

Q1 2020 SGD Dividend Income

Dividends received in Q1 2020 fell slightly compared to the same quarter in the previous year. This was due to a reduction in my Singtel position some time back.

Q1 2020 USD Dividend Income

In contrast, USD-denominated dividends increased slightly due to a dividend increase from Blackrock.

Moving forward, I expect dividends to fall further. Dividends received would be used to pare down credit card debt from buying cryptocurrencies.

In this quarter, all of my investment actions were cryptocurrency-related.

I have added to my positions in bitcoin and ether when the price crashed.

I have initiated a new but smallish position in Binance Coin (BNB), the in-house coin of Binance (the company). The rationale is that Binance (the company) has been growing by leaps and bounds in the cryptocurrency sphere. Their business model involves being the one-stop shop for pretty-much everything cryptocurrency-related. Would you like to purchase/sell cryptocurrencies? You could use BNB to defray your trading cost. Would you like to participate in initial coin offerings (ICOs) or initial exchange offerings (IEOs) on Binance? You need BNB to participate as well. Would you like to swap coins on Binance dex? You need BNB as well.

Hence, Binance (the company) stimulates demand for BNB by offering a compelling variety of choices to users in the ecosystem. These choices compete for a limited number of BNB that are available. Theoretically, demand would encourage the growth in BNB price over the long haul.

For risk management, I have kept the size of my BNB position small. It is a centralized coin and there are mixed comments on Binance (the company) and BNB online.

Next, I initiated a new position in Chainlink (LINK). Oh boy, this will be something pretty complicated to explain. Lemme try using a simplified example to illustrate how chainlink works instead.

Imagine that there is no such thing as SGX and you have to trade shares with your neighbours. How do you know what price is a fair price to use?

For example, Singtel shares last changed hands at Bedok at $3.02 between John and James. Concurrently, Singtel shares last changed hands at Ang Mo Kio at $2.98 between Mary and Julie at Ang Mo Kio. How do you know what is the "actual price" of Singtel shares? How do you reconcile the differences in the prices of trades at different geographical locations? Enter Chainlink, a decentralized data provider/aggregator/relayer of information.

In decentralized blockchain systems, it is very important to ensure that your data provider/aggregator/relayer of information provides accurate and timely information. If information comes from a single source, it is more susceptible to manipulation if there is malicious intent behind it. Chainlink attempts to solve this issue by requiring multiple independent providers/aggregators/relayers of information and introducing financial penalties if any of these providers/aggregators/relayers provide inaccurate information.

Lastly, I initiated a small position in Kyber Network Crystal (KNC), the underlying token of Kyber Network, a liquidity protocol, in order to test out something.

I have also added to my position in Synthetix Network Token (SNX) following its fall from grace from its all-time high due to BBs taking profits, leading to a cascade after cascade of sells (there goes my 2nd double-bagger in my portfolio, lol). There have been plenty of developments at Synthetix Exchange to the point that I can't keep up with their updates anymore. Just recently, they launched the synthetic FTSE 100 and Nikkei 225 indices. Coming up shortly is synthetic brent crude oil.

I have also taken partial profits off Small Love Potions.

Opportunistic Trades
As an income investor, I rarely engage in opportunistic trades.

In this quarter, bZx exchange suffered an exploitation and there was a "bank run." The yield for Ether shot up to around 70% per annum in the early hours of the bank run as depositors began to withdraw their assets. While people began to rush out, I rushed in to eat the high yield. The rationale is to eat the yield for a day or two and exit before the remaining mass of people realize what was going on and exit.

Management has been prompt in responding to the crisis. A day after, a second exploit happened, revealing that management did well to allay fears but were less capable in handling technical matters. Uh oh. Trouble, indeed, comes in pairs (insolvency and management incapability). For a while, I thought that I would have to write-off my principal. Thankfully, there were more aggressive individuals who added their assets into the pool to arbitrage the situation, which allowed me to escape with my principal intact. For all the trouble, I came out with a net negative SGD $2 due to the transaction fees.

A second opportunistic trade came when all the markets crashed (equities, crypto, metals, etc). Leverage always appear to be fine when the sun's out and everything is fine and dandy. When dark clouds suddenly gather, one might not have enough time to react and delever.

This was what happened when the cryptocurrency market crashed along with the equity markets. Those who leveraged up by minting the DAI stable coin found their ether collateral values going south. Unfortunately, the ethereum blockchain was clogged up at that point. Hence, they could not delever even if they wanted to. As programmed, the inbuilt system auctioned off their ether assets at fire sale prices. If I am not wrong, bidders couldn't even bid on these assets due to the network lag. The end result was that assets were priced at 0. Those who managed to high-ball the transaction fees and get the Ethereum blockchain to prioritize their transactions would get these ether collateral for "free."

makerdao liquidation

This is crazy man. Assuming the lowest ether price of SGD $127, the affected user just "gave" SGD 217k for "free."

For those folks whose assets did not get caught up in the fire sale and auctioned off at $0, they had to quickly get their hands on the DAI stable coin to delever. 1 DAI is pegged to 1 USD. Due to the insane demand for DAI, 1 DAI went up to even USD 1.20.

DAI coin price

As I have some DAI with me, I took profits. In other words, I managed to sell USD $1.00 for USD $1.06. 

Networth breakdown
Due to the sharp fall in equity prices, asset allocation to equities fell from 41% at Q4 2019 to 33% currently. Cash level is similar to Q4 2019 (29%) even though I have been purchasing cryptocurrencies throughout this quarter. Percentage of cryptocurrency as part of networth doubled from 6% to 12%.

Networth breakdown

As per before, "the pie chart depicts the breakdown in my net worth across the various asset classes in percentage (pie chart neither includes my CPF nor my emergency fund). To be conservative, I computed my precious metals allocation at spot price even though I am holding everything in physicals."

The curious case of physical and paper metals decoupling
Recently, while on twitter (yes, I recently got a twitter account), I came across a curious tweet stating that the physical gold/silver market has decoupled from their paper equivalents. Shortly after, I received an email from GoldSilverCentral, a local bullion dealer, bringing attention to the same thing.

A cursory review of the websites of all major bullion dealers in Singapore showed that physical metals (especially silver) are trading at a much higher price than their paper counterparts.

From GoldSilverCentral's website:

From Bullionstar's website:

The price of paper silver when the screenshot was taken:

The price of physical silver on that same instance:

From SilverAG's website:

With the closure of the 3 largest refiners in the Swiss Canton of Ticino due to the virus, supply is not able to keep up with the increased demand. As a result, the prices for physical metals have increased to reflect this.

This is a pleasant surprise. As my exposure to precious metals is entirely physical and I have a concentrated position in the asset class, it has helped to completely negate the sudden crash in equity prices.

This also underscores the differences between the different "classes" of precious metals. Many people do not seem to get this because of their lack of familiarity with the precious metals market. At the top of my head, there are: (a) paper metals, (b) bullion bars/coins, (c) "semi" bullion bars/coins, (d) antique bullion bars/coins, (e) numismatics/proof coins, (f) bullion jewellery, (g) jewellery, and (h) jewellery with low workmanship cost.

For example, one "semi" bullion silver coin that I have trades at 900% premium above paper silver price at one of the local bullion dealers. 

Depending on the class of precious metals purchased, your returns may vary.

Debt Levels
Interest Coverage Ratio: 24
Debt-to-Equity Ratio: 0.012

Similar to the previous quarter, leverage in relation to overall networth has been increasing due to my use of credit card to purchase cryptocurrencies.

Current Holdings
After converting all my USD and HKD holdings to SGD at the end of the quarter, the following table shows the percentage of each stock from only the equities allocation of my net worth (arranged in descending order).

Stock Name
Hongkong Land
Parkway Life REIT
Hong Kong Tracker Fund
Frasers Centrepoint Trust
Mapletree Industrial Trust
First REIT
Thai Beverage
Mapletree Commercial Trust
JM Smucker
ST Engineering
Raffles Medical
Capitaland Mall Trust
Dairy Farm
Sheng Siong
Japan Foods
Frasers Commercial Trust
Frasers Property
General Mills
Frasers Logistics Trust
Kraft Heinz
Hormel Foods
Welltower REIT
Abbott Laboratories

Quite a number of changes in positions, but I am not too worried about it.

Areas for Improvement
Some blogger friends indicated that I could improve on my generally buy-and-hold strategy. Upon reflection, what I could have done better is to include more instances of partial profit taking. I have executed on a number of them in previous years for non-REIT counters in my portfolio. What I could have done better is to apply the same to REITs in my majority REIT portfolio. My hesitancy stems from my preference for income and taking partial profits would mean killing the golden goose for dinner. To facilitate partial profit taking, I could increase the position sizes of each transaction. That way, capital gains would be more meaningful (in absolute terms).

Another recurring point of contention is the presence of cryptocurrencies in my networth. From a portfolio perspective, it is but one among many other asset classes. Sure, volatility is a concern, but in the grand scheme of things, each asset class balances the other out since I have exposure to all major asset classes (equities, precious metals, cryptocurrencies, cash, and, indirectly, bonds). 

In the initial stages, it was a wild ride of insane ups and downs. There was angst when it went up and I haven't accumulated enough. What helped me at this point was a concept that I have learned from Finance Smith on Value Cost Averaging. When prices came down again, it was another opportunity to accumulate. When it went down further, I wasn't spooked. I have kept myself up-to-date with cryptocurrency developments and it was time to accumulate more, with a larger time interval between transactions to manage risk from buying too fast, too furiously. The feeling I got back then was not fear, but sian that prices were not reflective of developments.

After awhile, you will get used to it. You wouldn't bat an eyelid at 50% ups or downs anymore.

The Same Game plan
Dividends received, albeit reduced, goes back into investments. REITs and employment income continue to form and strengthen the financial base from which I expand into other non-REIT income stocks, growth stocks, or cryptocurrencies. 

To strengthen the financial base further, continue adding to emergency funds and investments at the same time. 

Next up is to observe how my US dividend growth stocks fare. Will there be dividend cuts or will they remain as stalwarts? We shall see.

It has been a long post.

Thanks for reading!

Wednesday, January 1, 2020

Year 2019 Portfolio Performance and Dividend Income

Alrighty. The markets have closed for 2019 and it is time to review my investment performance for the year.

Dividend Income

Q4 2019 SGD Dividend Income

Dividends received from my SGD-denominated portfolio in Q4 2019 was an improvement over Q4 2018. Regular readers would know that I have been pruning the weaker counters from my portfolio for over the past 1+ year and substituting them with stronger counters (mainly large caps). This appears to be bearing fruit but I'm not sure how sustainable this is going forward.

Q4 2019 USD Dividend Income

As SG REITs and income stocks have been trending higher and higher, I have been increasingly hesitant to add to them. Instead, I have been placing some spare cash into boring old US dividend growth stocks from 2018 and early 2019. This appears to be bearing fruit now.

I have also received my first dividend payout from the Tracker Fund of Hong Kong in Q4 2019. No graph for my HKD-denominated portfolio this time as it's just one bar on a bar chart.

Q4 2019 Transactions
In the last quarter of 2019, I subscribed to the Mapletree Commercial Trust rights issue (with excess) to enlarge my position in the counter.

I have also chosen to receive scrip dividends from OCBC instead of dividends.

I have accepted the general offer from ISEC. For a holding period of 2 years and 9 months, my annualized returns were 11.20%.

That's all for "traditional investments." In contrast, I have been making plenty of strides into the cryptocurrency markets. For those interested, do refer to this post I specifically made to share what I have been doing in cryptoland.

For my SGD-denominated portfolio:
Returns for Year 2019: 22.82%
Annualized returns since portfolio inception (March 2015): 10.03%

For my USD-denominated portfolio:
Returns for Year 2019: 4.55%
Annualized returns since portfolio inception (March 2016): 4.88%

For my HKD-denominated portfolio (portfolio started in May 2019):
Returns for Year 2019: 9.15%

Unrealized gains from gold: ~15%
Unrealized gains from silver: slight profits

Unrealized losses from Bitcoin: ~ -10% (excluding dividends)
Unrealized losses from Ether: ~ -30% (excluding dividends)
Unrealized gains from SNX (Synthetix Network Token): ~ 70% (including dividends)

Thoughts on Performance
Generally, I am satisfied with my performance across all my portfolios. When one portfolio or asset class is down, another pulls it up. This is what I have been working on; a diversified portfolio that emphasizes stability and survivability over returns. Come rain or shine, it is dependable, a bastion of strength with layers upon layers of defence.

REITs and income stocks provide me with the cashflow. Cashflow provides me with the liberty to swing for alpha (cryptocurrencies). Gold and cash provides me with peace of mind, a hedge against catastrophic crashes and dry powder to use when wildfires grow out of control and liquidity (pun intended) is scarce.

Camaraderie is important. When HongKong Land, our star player in the USD-denominated portfolio, fell from grace this year, fellow USD-denominated soldiers like BlackRock along with Medtronic did a good job in holding the fort. The larger positions in my portfolio are higher-conviction plays. If they flop, the mid-tier and small-tier positions are there to lend support. A bundle of sticks is not easily broken. They will survive and then thrive.

Note: returns for SGD/USD/HKD portfolios are computed using Excel's XIRR function. In contrast, gold/silver/cryptocurrencies returns are computed as the difference between market price and average purchase price.

Net Worth breakdown
Not much changes to my networth breakdown. Still in the accumulation phase for cryptocurrencies.

Net Worth Breakdown

As per before, "the pie chart depicts the breakdown in my net worth across the various asset classes in percentage (pie chart neither includes my CPF nor my emergency fund). To be conservative, I computed my precious metals allocation at spot price even though I am holding everything in physicals."

Debt Levels
Interest Coverage Ratio: 22.28
Debt-to-Equity Ratio: 0.013

Leverage in relation to overall net worth has been increasing due to my use of credit card to purchase cryptocurrencies.

Current Holdings
After converting all my USD and HKD holdings to SGD at the end of the quarter, the following table shows the percentage of each stock from only the equities allocation of my net worth (arranged in descending order).

Stock Name
Hongkong Land
Frasers Centrepoint Trust
Parkway Life REIT
Hong Kong Tracker Fund
Mapletree Industrial Trust
First REIT
Mapletree Commercial Trust
Thai Beverage
Capitaland Mall Trust
ST Engineering
Raffles Medical
Japan Foods
Frasers Property
JM Smucker
Frasers Commercial Trust
Dairy Farm
Sheng Siong
Frasers Logistics & Industrial Trust
Kraft Heinz
General Mills
Hormel Foods
Abbott Laboratories

BlackRock joined the top 10 stock holdings due to its 2019 good performance, kicking First REIT off the list.

That's all for now. Thanks for reading!

Non-financial updates coming up next!