Sunday, 23 April 2017

Possibly The Worst Time to Invest – 3 Years On

This is an annual blog series that I started 3 years ago to document the worries about investing at the wrong time, which would bring losses and headaches. The blog series track the performance of 2 passive portfolios invested in index funds using the portfolio rebalancing strategy. Both portfolios comprise of 70% allocation in stocks and 30% in bonds. The plain vanilla portfolio invests in global equities and global bonds while the spicy portfolio invests in US equities and Asian bonds. The first portfolio was started in Dec 2013, while the second one was funded progressively over 2015. 

In the first post in 2014, I mentioned worries about the Dow Jones Industrial Average (DJIA) nearing its all-time high (then) and US Federal Reserve planning to raise interest rates from an all-time low. In the second post in 2015, I mentioned that the same worries persisted, with DJIA touching yet new highs and interest rates moving up in anticipation of Fed's interest rate increase. Not only that, new risks emerged with oil price crashing by more than 50%, China's growth slowing down and the threat of Grexit. Yet, despite all these worries, the plain vanilla portfolio went up by 12% since its inception.

In the third post last year, I mentioned that worries about market declines actually materialised, with major declines in Aug 2015 and Jan 2016. The decline in Jan 2016 was especially severe, with stock markets around the world crashing. At mid Feb 2016, the plain vanilla portfolio was down by 0.7% since inception while the spicy portfolio lost 7.0%. Yet, by the time I wrote the annual post in Apr 2016, both portfolios had bounced back strongly. The plain vanilla portfolio was up by 8.5% while the spicy portfolio gained 0.6% since inception.

With each passing year, more and more risks materialised. Jun 2016 saw Britons voting for Brexit while Nov 2016 saw US citizens voting for Donald Trump as president. Both outcomes were unexpected and led to sharp falls in the stock markets around the world. Yet, barely days later (or hours in the case of the US presidential election), stock markets had recovered fully from their initial falls. Not only that, stock markets went on to scale new heights on optimism that President Trump's fiscal policies would spur faster growth in the US and world economies. Currently, the plain vanilla portfolio is up by 21.6% while the spicy portfolio is up by 13.7% over their respective holding periods of about 3.5 years and 1.5 years.

Personally, I still worry a lot about risks, which I wrote about in a couple of posts last year, such as What Have We Got After 8 Years of Easy Money?, Making America Great Again and Its Impact to Asia, Another Year That Ends with 7, etc. This pessimism is reflected in my active investments. Over the past 1 year, I have been taking some money off the table. Some of the risk management related divestments include Venture at $8.38, Valuetronics (partial) at $0.50, Global Logistic Properties (partial) at $1.81 and a couple of speculative shares (see Meet The Minions). Nonetheless, there are new investments, but these are in more defensive stocks such as dividend stocks, beaten-down stocks and even Gold.

In fact, I was quite tempted to tinker with the 2 passive portfolios given the strong views about the market. But I decided not to do anything about them. Had I rebalanced or withdrawn money from the 2 passive portfolios, they would not have achieved the returns mentioned above. They have built-in defence mechanisms to manage market crashes through portfolio rebalancing if the stock/ bond allocation were to deviate from the original allocation by a pre-defined amount. For these 2 portfolios, I will continue to stick to the pre-defined strategy even if the markets were to crash.

In conclusion, it is difficult to predict where the markets are heading. If you have a well-defined defence mechanism in place, just let the portfolios continue their work.

See related blog posts:

Sunday, 16 April 2017

Early Retirement Maybe A Luxury That I Cannot Afford

I have blogged about early retirement in the past 2 years, but I really do not intend for this to be an annual series. Moreover, I do not intend to retire early and sit back and do nothing. Nevertheless, there are fresh insights on this topic and it is good to write them down for future reference. 

In the past 1 year, I have read a few books such as "Capital in the 21st Century" and "Rise of the Robots: Technology and the Threat of a Jobless Future". I am concerned about automation and robots taking away jobs. By right, this should not be a concern for someone who has considered early retirement. However, there are additional complexity if I think about future generations. I do not have any children currently, but if I have, then any actions on my part now would have an impact on them in the future.

I do not think I will be replaced by automation and robots any time soon. However, the same cannot be said for the next generation. If the doomsday scenario of robots replacing workers on a wide scale were to materialise, it means that we are back to the very old days when how well we live does not depend on how hard we work, but who our parents are and/or whom we marry. In the case of my children (if any), that parent would be me. Thus, when seen from an inter-generational perspective, the window of human employment is closing soon and early retirement at a time when jobs are still available seems a luxury. Hence, instead of saving enough for my own retirement and retiring early, I should work for as long as possible to maximise the income from human capital and build up sufficient financial capital upon which my descendants could lead a decent life. Early retirement maybe a luxury that I could not afford in the face of automation and robots.

Of course, this is not a fool-proof plan. Whatever I save could be squandered away by future descendents. So, I do hope that the doomsday scenario of robots replacing workers will not happen. Or perhaps the prevailing governments of the day would understand the social implications and implement some basic income for citizens as suggested in the books mentioned above. If I have to pay more taxes for this to happen, I would grudgingly pay them. It is a small price to pay for social insurance for my future generations.

Having said the above, if I can have the option of not relying on some external parties to bail us out, that would be the best. Thus, I will have to use my own efforts and earn as much as possible. Sorry, folks, I have to go back to work tomorrow.

See related blog posts:

Sunday, 9 April 2017

Breaking My Valuation & Position Limits

It is official! I have broken my valuation limits on buying & selling stocks and position limits on individual stocks! Previously, I mentioned in What is My Target Price? that I have valuation limits of 1.8 to 2.0 times book value for buying stocks and 3.5 to 4.0 times book value for selling them. In Jan this year, I had broken these rules with the purchase of M1 at 4.7 times book value and Singtel at 2.5 times book value!

Also broken were my position limits on individual stocks. I have an initial position limit of $15K to $20K on each stock, depending on what type of stocks they were. These limits could be doubled to $30K to $40K if I need to average down on the stocks. The position limits were first broken in Nov 2015 with the purchase of Global Logistic Properties (GLP). Initially, I thought this would be the exception rather than the rule, considering the long-term growth prospects of GLP. However, after I invested in M1 and Singtel beyond the initial position limits, it is confirmed that the position limits have been broken. 

What caused the change in my valuation and position limits? To understand the reasons for the change, you need to understand why the valuation and position limits were put there in the first place. For a very long time, I have been using quantitative methods to analyse and value stocks, looking at only earnings, dividends, cashflows, debts, book value, etc. This approach has served me well in the past, but there are times when this approach turned up value traps whose share price keeps on declining. Thus, it makes sense to have valuation limits to ensure that I do not overpay for stocks identified using this approach and position limits to ensure that whatever mistakes I make do not become so large that I cannot recover from them. Quantitative limits on valuation and position size go hand in hand with quantitative methods.

It is also important to realise that quantitative methods have some underlying assumptions -- either (1) the stock will close the gap between price and intrinsic value, (2) the stock will recover to its past earnings and price (mean reversion), or (3) the stock will continue to generate good earnings and dividends (extrapolation). Sometimes these assumptions do not hold. Some stocks just do not recover in earnings and price after a decline, such as the few Oil & Gas stocks that have gone into judicial management. Other stocks are unable to sustain the good earnings and dividends, such as Starhub and M1. The problem with quantitative methods is that you cannot tell whether the assumptions will hold or not until the results are announced. By that time, it is probably too late to sell the stocks. Valuation and position limits make a lot of sense when you cannot see what is ahead.

Over the past 2 years, I have been gradually moving away from quantitative analysis into qualitative analysis, looking at issues such as business strategies, competitive environment, corporate governance, etc. This approach has the advantage of providing a glimpse into where the business is heading instead of extrapolating from past performance. Thus, if the business looks good, I could take up positions ahead of the market. Conversely, if the business looks bad, I could sell in advance. Valuation and position limits are less useful if you can see accurately what is ahead.

Furthermore, SGX is a small market. There are very few stocks in some industries such as banks, telcos, shipping, etc. But the amount of work necessary to analyse the industry is independent of the number of listed companies in that industry. For example, I wrote 8 posts on the telco industry but there are only 3 telco stocks, out of which I selected 2 for purchase. If I could only invest $15K on each stock, it really does not do justice to the amount of efforts put in. Position limits become constraints when there are limited number of stocks in a particular industry. Thus, my position limits were officially broken with the purchase of M1 and Singtel in Jan.

Having said the above, I have not fully discarded the valuation and position limits. There are dividend stocks that I purchase using the quantitative methods. For these stocks which I have no insights or time to analyse deeply, valuation and position limits will continue to be in place.

Will breaking the valuation and position limits lead me to make mistakes that I cannot recover from? I certainly hope they would not. I will still need to improve my skills at seeing the future prospects of the companies. 

See related blog posts:

Monday, 3 April 2017

I Didn't Let My Alma Mater Down

How time flies. This is post no. 208, which makes it the 4th birthday for this weekly blog. This is a time for celebrations and reflections. Today's story is about my studies in the Singapore Management University (SMU)'s Masters in Applied Finance (MAF) programme.

I enrolled in the programme in Jul 2004. The stock market had just recovered from a 3-year slump due to the crash in 2000, Sep 11 terrorist attack in 2001, accounting scandals in 2002 and the Severe Acute Respiratory Syndrome (SARS) in 2003. After a prolonged slump, the market staged a strong recovery in 2004 and I made my first (small) pot of gold. I decided to reinvest the profits, not in stocks, but in a formal education in investment. Through good fortune, I heard of the MAF programme in SMU and decided to enrol in it in Jul 2004.

The MAF programme catered to professionals who wanted to advance in their careers in the finance industry as well as people who were looking to make a career switch into the industry. Thus, it took in graduates who were trained in other disciplines. However, being the introvert I am, I had totally no idea what the intent of the programme was all about, nor the hot prospects of jobs in the finance industry then. I was not even aware of the Chartered Financial Analyst (CFA) programme! I was simply happy to be accepted into the programme even though my basic degree is in engineering. It was only after I enrolled into the programme and met my new classmates, some of whom were planning to switch into the finance industry, did I realise what the programme was all about. At one stage, I wondered whether I had deprived anyone with the intention to switch of a place in the programme. Anyway, by then, it was too late. So, I continued my studies.

Frankly speaking, studying for the purpose of gaining knowledge instead of getting good grades is a joy. I enjoyed the lessons and the projects despite having to rush to class from work every Tue and Thu, besides spending the whole of Sat in classes. How could projects be boring if they were on analysis of companies such as Informatics, CK Tang, SIA, etc., since whatever insights we gathered for the projects could be useful for our own investments? Besides, the American education system that SMU adopted encouraged lively exchanges of ideas instead of just copying notes from the lecturer. I even had the curiosity to read up the "Greeks" (i.e. Black-Scholes model for options) outside of the classes! When we were about to graduate 1.5 years later, I felt a sense of loss. Never had I enjoyed classes so much in my life!

After I graduated, I applied for a few jobs in the finance industry, but I was not accepted, partly also because I was over 30 years old by then. Nevertheless, I did not feel too dejected, since my intention of enrolling in the programme was to gain knowledge rather than studying for grades.

A couple of years later, in 2012, I started this blog to share my knowledge and experience about investing. I believe I have a unique combination to offer to readers -- someone who is old enough to have 30+ years of experience in the stock market and has witnessed many stock market crashes but still young enough to blog and share knowledge. Furthermore, that practical experience is complemented by academic knowledge from the MAF and CFA programmes.

It is 4.5 years since I started blogging and 4 x 52 posts later, I believe I have established a useful resource for readers who wish to learn more about investing. What I have shared in this blog has sometimes gone beyond what is found in books. While I did not contribute back to society through working in the finance industry, through this blog, I am returning to society what I have gained from SMU. I like to say that I did not let my alma mater down. Thank you, SMU, for giving me a wonderful education in investment. Last, but not least, thank you readers who have encouraged me to continue writing through your visits to this blog.

See related blog posts:

Sunday, 26 March 2017

The Investigative Approach to Stock Investments

There are a couple of quantitative methods for analysing stocks, such as the Dividend Discount Model (DDM). A lot of people use them for stock analysis and investment as they are relatively simple to use and do not require qualitative analysis of the business strategies, competitive environment, corporate governance, etc. For a very long time, I was also a keen user of such methods, looking at only earnings, dividends, cashflows, debts, book value, etc. to identify value stocks. Such an approach has served me well in the past. However, there are times when this approach turned up value traps whose stock price keeps on declining. Over the past 2 years, I have gradually moved away from such quantitative analysis.

Let us use the DDM as an example of the quantitative approach. A simple form of the DDM is:

where P    = Intrinsic value of stock
           D1  = Dividend for the next financial year
           r     = required rate of return
           g    = perpetuate rate of growth in dividends

It is simple to use, as there are only 4 parameters to estimate. A lot of times, in the absence of qualitative analysis, these parameters are estimated from past performance. However, past performance do not necessarily represent future performance. An example of this is Starhub. Since 2010, Starhub has been paying a constant dividend of 20 cents every year. The dividend has been so regular that it is commonly assumed that the 20-cent dividend will continue every year. Last month, Starhub dropped a bombshell by announcing that the dividend will be cut from 20 cents to 16 cents in FY2017. This is the perils of looking just at the financial numbers and extrapolating past performance into the future.

An alternative approach to stock investment is to carry out a qualitative analysis of the company and the industry it is in. One of the best known techniques in this approach is the scuttlebutt technique, which is made famous by Philip A. Fisher in his book "Common Stocks and Uncommon Profits". For the past 8 weeks, I have attempted the use of such an investigative approach in the analysis of telco stocks, looking at the business strategies, competitive environment, (my own) customer experience and industry trends. My skills are still rudimentary compared to the scuttlebutt technique, but the investigative approach does provide a glimpse of where the business is heading rather than extrapolating from past performance.

It is tough work reading through and comparing all the telco price plans, financial results, annual reports, industry statistics and trends, technology news, etc. But the end result is a better understanding of the prospects and risks of the company and whether the money can be safely invested in it. 

So far, 2 industry analyses have been completed, namely, Oil & Gas and Telcos. I hope to complete more industry analyses in time to come.

See related blog posts:

Sunday, 19 March 2017

Challenging Times Ahead for Starhub's Dividends

When M1 announced its results in end Jan, I went to buy both M1 and Singtel, but I did not buy Starhub. The conventional wisdom is that between M1 and Starhub, Starhub would be better able to manage the competition from the fourth telco, as it has Pay TV, broadband and enterprise fixed services besides mobile services. That is true provided the other business segments are generating stable, recurrent cashflows. However, is that true?

First of all, let us look at the revenue contribution from its 4 business segments.

Fig. 1: Starhub's Service Revenue Breakdown

Excluding equipment sales, mobile services constitute the bulk of Starhub's service revenue in FY2016, contributing 55% of the revenue. The second largest segment is enterprise fixed services, contributing 18%. Pay TV is third with 17% contribution while broadband services is smallest with 10% contribution. Let us look at the prospects of each business segment.

Mobile Services

Over the past couple of weeks, I have tried to understand the mobile services business by analysing M1, because M1's mobile services constitute 79% of its service revenue and is closest to a pure mobile services company. You can refer to the following blog posts for more information about the mobile services business:

Generally, based on my assessment of M1, mobile services is facing a decline in revenue due to the introduction of SIM-only plans and data upsize plans in the short run. However, the effects are transient and should disappear in the later half of the year. Over the long run, mobile services is still a viable business despite the threat of the fourth telco. 

Pay TV

An entire blog post is dedicated to the analysis of Starhub's Pay TV business in Is Pay TV Still A Reliable Cash Cow? Generally, the Pay TV business is facing intense competition from over-the-top (OTT) providers such as Netflix, CatchPlay, Apple TV, etc. The cost structure for the business is mostly fixed cost for the cable network infrastructure and premium TV content. Thus, the declining Pay TV subscription means that there are less subscribers to spread out the fixed cost. Furthermore, there is also the threat of TV content owners moving away from Pay TV providers to OTT providers. Starhub's response to OTT providers is to roll out similar video on-the-go services, which are priced lower than traditional Pay TV offerings. It also invested in mm2 Asia to acquire locally made video content.

Thus, from the cashflow perspective, there is reduced cashflow as subscribers move away from its Pay TV network and if Starhub continues to make investments in media companies.

Wired Broadband

The wired broadband business can be categorised into cable broadband and fibre broadband. The total no. of broadband subscribers have remained stable in the last 2 years. However, on closer inspection, there is declining subscription for cable broadband, offset by a corresponding rise in fibre broadband subscription. See Fig. 2 below.

Fig. 2: Starhub's Broadband Subscriptions

If the total no. of subscriptions stays the same, does it matter if one type of subscription is declining while another is increasing? Yes, it matters. As explained in the analysis of the Pay TV segment (see Is Pay TV Still A Reliable Cash Cow? for more details), the cable network infrastructure is a fixed capex cost, thus, the less cable broadband subscribers Starhub has, the less customers to spread out the fixed cost. Although Starhub has raised the monthly rates for cable broadband, that is still insufficient to offset the declining subscription. Fortunately, since the cable TV/ broadband businesses have been in operation for many years already, the infrastructure cost should have been mostly depreciated.

On the other hand, fibre broadband has a variable cost structure. Fibre broadband is provided through the Next Generation Nationwide Broadband Network (NGNBN) which is owned by NetLink Trust. Starhub leases the network capacity from NetLink Trust. Thus, the more fibre broadband subscribers Starhub has, the more money Starhub has to pay NetLink Trust. It is still a profitable business nonetheless, but the profit per subscriber is smaller than for cable broadband.

Moreover, the fibre broadband business is also facing intense competition from other telcos as well as broadband service providers like MyRepublic. M1 and Singtel have to bundle mobile broadband and digital home voice to stand out from the competition.

Thus, from the cashflow perspective, there is reduced cashflow from the wired broadband segment as subscribers move away from its cable broadband network.

Enterprise Fixed Services

Perhaps the only rising star is the enterprise fixed services segment. This segment is a cluster of InfoComm Technology (ICT) solutions such as managed security services and analytics, as well as infrastructure solutions such as data centres and islandwide fibre network. Revenue from this segment has increased by 3.9% in FY2016. However, to earn increasing revenue from this segment, Starhub has to continue to invest to build up the network infrastructure, which is expensive. In Jun 2016, Starhub issued a medium term note of $300 mil. In addition, Starhub has announced that dividends for FY2017 will be cut by 20%.

Moreover, this segment is not without competition. ICT solutions are in direct competition with Singtel, while infrastructure is in direct competition with the NGNBN owned by NetLink Trust and Singtel's own network. Thus, even this bet is not guaranteed.

From the cashflow perspective, this segment will require a lot of capex instead of contributing to the cashflow, at least in the short term while the network infrastructure is being built up.


Going back to the opening paragraph, the conventional wisdom is that between M1 and Starhub, Starhub would be better able to manage the competition from the fourth telco, as it has Pay TV, broadband and enterprise fixed services besides mobile services. Yet from the analysis above, it can be seen that the Pay TV and broadband segments are no longer the reliable cash cows that they used to be. Although enterprise fixed services is a rising star, it is soaking up cashflow at a time when the other 2 business segments are not generating as much cashflow as before. Starhub is facing challenging times ahead.

P.S. I am vested in M1 and Singtel.

See related blog posts:

Sunday, 12 March 2017

Is Pay TV Still A Reliable Cash Cow?

For the past 6 weeks, I have been blogging about the mobile services segment of telcos. It is time to move on to the next segment -- Pay TV. I will analyse Pay TV using Starhub's results, as Singtel has more business segments and operates in many countries.

The figure below shows the no. of Pay TV subscriptions in FY2015 and FY2016. Starhub is kind enough to disclose its market penetration rate in its financial results, which allows me to work out the total size of the Pay TV market in Singapore.

Fig. 1: Pay TV Subscriptions

As shown in the figure above, Starhub's Pay TV subscriptions has been on a gradual decline, dropping by 8.6% from 1Q2015 to 4Q2016. In contrast, the overall market for Pay TV has grown significantly by 20.5% over the same period. In particular, the growth picked up speed in 1Q2016. What happened in 1Q2016? The answer is Netflix. Netflix entered the Singapore market in Jan 2016, offering viewers an alternative way of watching videos over the internet instead of subscribing to Starhub's or Singtel's Pay TV.

How does Netflix's pricing compare with Starhub's Pay TV pricing? Starhub has a basic subscription of $26.75 per month for a few groups of channels. If you wish to watch sports, you will need to add another $21.40 for the sports channel. In contrast, Netflix is available at $10.98 per month for the basic plan. Nowadays, you can even watch sports on M1's fibre broadband or even Mediacorp's Toggle channel. You no longer need to subscribe to Starhub's or Singtel's Pay TV to watch videos or sports.

What is the cost structure of the traditional Pay TV segment? Firstly, there is a cable network infrastructure to deliver video to customers' set-top boxes. This is a fixed capex cost which does not vary with the number of subscribers. Thus, the less subscribers Starhub has, the less customers to spread out the fixed infrastructure cost. Fortunately, since the cable TV segment has been in operation for many years already, the infrastructure cost should have been mostly depreciated. In fact, Starhub mentioned in its financial results that some assets have been fully depreciated, although it did not mention which infrastructure it referred to.

The second cost for Pay TV providers is programming cost, or cost for the TV content they are distributing. For regular channels, this is likely to be a variable cost that varies according to the number of subscribers for that channel. However, there could also be premium content such as sporting events that charges a fixed cost for the broadcasting rights (see How the costs of sports broadcast rights have shot up in Singapore). Again, fixed cost is bad news for Pay TV providers facing declining subscription rates.

It is not only subscribers who can move away from a Pay TV provider. Content owners also can. Starhub and Singtel have been competing for the broadcasting rights for sporting events and the rights are awarded to the highest bidder. Prior to IMDA introducing cross-carrier rules in 2011, subscribers had to switch between different Pay TV providers to watch their favourite sporting events, which were exclusive to the winning bidder. Thus, there is no reason why content owners that are currently providing content to Pay TV providers could not switch to over-the-top (OTT) providers such as Netflix, which will further erode the reason to subscribe to Starhub's or Singtel's Pay TV.

Having said the above, Starhub is also not idling. It has introduced Starhub Go, which allows subscribers to watch videos for $9.90 per month on-the-go, like Netflix. Not only that, subscribers who are also Starhub's mobile services customers do not need to pay mobile data charges while streaming video on 4G. This is an advantage over OTT providers, whose customers incur mobile data charges while streaming video.

Starhub is also riding on the popularity of Netflix by offering Netflix as one of its channels on fibre TV (but not available on cable TV, because it requires an internet connection). I understand that there is minimal overlap between the content on Netflix and Starhub so far, but that could change in the future.

Starhub has also invested 9.05% in mm2 Asia for $18 mil in Mar 2016 to acquire locally made video content to better compete with OTT providers. This investment also has the advantage of ensuring mm2 Asia's content do not end up in rivals' Pay TV networks. Perhaps we might see Starhub continuing to invest in other media companies in the region. However, if that is the case, there will be less cashflow for dividends. This creates a difficult balacing act for Starhub. If Starhub invests in media companies, it can better safeguard its Pay TV business, but at the expense of dividends to shareholders. However, if it allocates more cashflow to dividends, it is less able to safeguard its Pay TV business from further competition.

Thus, from the cashflow perspectives, there is reduced cashflow as subscribers move away from traditional Pay TV providers like Starhub and Singtel. Not only that, there is less cashflow available for distribution to shareholders should Starhub continues to make investments in media companies.

In conclusion, the Pay TV segment is facing challenges and is no longer the reliable cash cow that it used to be.

P.S. I am vested in M1 and Singtel.