Thursday, October 5, 2017


Updated: 2017.10.06
Data source: SPDR ETF annual report 2017, AIA fund report semi-annual report 2017

I got the pdf files from both web sites, both reports recorded respective funds performance ended Jun 2017. Because AIA's file is password protected, cannot copy and paste the figures into EXCEL file for comparison, so we have to live with the image captures.

AIA Fund performances

Spider STI ETF performance

STI ETF's annualized return is boosted by dividends over the years, approx. 3% improvement.

I suppose that AIA's fund return is before cost, meaning after mgmt cost deduction---assuming a 20:80% split, the unit trust holders of these funds are generally poorer than directly invest into SPDR STI ETF.

Of course, AIA funds are usually provided together with insurance products with certain level of insurance protection.

Sunday, July 16, 2017

NetLink NBN Trust (IPO)

Data source: IPO prospectus
Date: 2017/07/17

I have been reading on this prospectus (total 536 pages) over the last week and entire week end. Netlink is closely related to the industry I'm in, so naturally its IPO attracts my attention.

Given it is one of the largest IPO in the recent years, a lot of financial bloggers have written about their views on its list price of 0.81SGD.

I decided to post my blog only after today's IPO application is closed (at 12pm) so as not to affect anyone's own decision.

The corporate structure is very complicated in this IPO, there are many 'Trust's involved, it is important to understand which and when Trust word is referring to.

To be listed is Netlink NBN Trust (Trust), the underlying business is from Netlink Trust(NLT, another trust, or the original trust). There are two sets of financial statements in prospectus, the audited NLT group's 2015, 2016, 2017 FR; and unaudited pro forma Netlink NBN Trust (Trust)'s 2016, 2017FR. If this has not caused confusion, please refer further to the structure on pg 65.
As IPO subscribers, we are applying to the middle triangle(in dark color)'s units.

The idea of this complication model, is that NLT (the triangle at the bottom which is actual the original trust that owns the operational biz) is to use this method to be in QPDS(qualified project debt securities scheme), so that IRAS can agree to tax exemption on its interest return(further on this point later, NLT Notes) to Trust(the dark triangle, to be listed)

First thing first, who makes a bulk of money?
Answer: Singtel
In order to be listed, Netlink NBN Trust (Trust) is to acquire from Singtel's portion of NLT(bottom triangle) units with a consideration of SGD$1.8878b(pg 63 notes 1), of which $1.095b paid in cash, with remaining in Trusts(IPO, top triangle) Units: 24.99% (or 966m Units), which is to be held under Singtel Interactive (HoldCo, full subsidiary of Singtel).

I wasn't in Singapore when OpenNet(predecessor of NLT) was setup, so not clear what Singtel's original investment was. But quiet clearly to me, when compare two sets of FR in the prospectus, this action created an intangible asset(goodwill I assume) for Trust, at slight above $1b.

It means: Singtel sold its portion of NLT which is NAV at 0.8878b(1.8878-1) to Trust(IPO we are subscribing to) at $1.8878b, got a cash coffin of over $1b, and 966m Trust Units (at 0.81 per piece) in return.
That is what I call Investment. The goodwill created by this acquisition, will be carried on Trust BS, and amortized over 50years at ~$20m each year.

Trust mgr basically get a piece of business at a higher rate(from Singtel), and then float basically the same biz onto the security market at a lower rate (to NAV, whether due to market condition, other whatsoever reason), of course Trust mgr doesn't have to fork out that short fall(then who?)

Do you go IKEA to get a piece of furniture at 188 and sell it later below 188? Maybe yes, because that is your friend's money, and you get a commission cut from such transaction.

(where is customer's yacht)?

How does TRUST fund its DPU(read as attract its IPO investor)?
The projection is DPU at 5.43% and 5.73% in FY2018, FY2019.
As large CAPEX period is over, depreciation of PPE and amortization of licence/intangible assets etc are largely non-cash items, as long as NLT sets aside enough money for future growth ($211m CAPEX for 2018, 2019 projection) and working capital for daily operation, the remaining(majority) cash generated operationally can be funneled back to Trust, thus allow Trust to distribute all its income to Unit holders.

So much of the fanfare is on an non-SFRS number: EBITDA, a much likened figure of Telecom sector, projected at 240m, 150m for 2019, 2018 respectively, some of which will be the source for NLT to pay Trust.

Of IPO proceeds, $1.1b will be used by NLT to repay its external ST facility agreement, which carries 2.9(or 4+)% interest rate, since Trust is going to help NLT to repay all that ST facility, NLT in term owes 1.1b NLT Notes (pg 119) to Trust(at a rate of 10.5%), a whopping rate jump.

Why would NLT instead of using a ST facility which is at much lower rate, replacing it with NLT Notes to Trust at 10.5%, until 2037?

On the other hand, because IRAS' tax exemption is on the condition that the distribution from NLT to Trust, must be distributed to Trust Unit holders within 3(or 6)mths, thus means interest earned by Trust on NLT Notes, around $110m has to be distributed semi-annually. It is my understanding that this contributes to the 5.43% DPU projection in prospectus; in other words, the actual operationally generated cash for distribution (CAFD) from underlying business of NLT is much smaller.

Should a unit holder rely on DPU based on organic business growth, or based on return of one's own fund (Trust IPO to raise fund from unit holder, pass it to NLT, NLT to repay ST facility, NLT to issue NLT Notes to Trust so as to be QPDS, Trust get interest from NLT and other operating cash income, Trust to distribute cash to unit holder as DPU).

Nonetheless, given fiber is currently already 86% of residential market, and presumably the capex will dwindle down over the years when it reaches saturation, NLT will turn to a cash-cow type of business, with only large depreciation carried on BS, w/o actual cash outflow.

Few things still puzzle me:
- NLT recved $732m in grants for building the nationwide passive fiber infra(amortized at 29m by NLT over 25years). But upon completion of Trust acquisition, this deferred financial support grant will be derecognized at Trust group level. Why? what is the reason this obligation is no longer required?
- Given it is a high regulated market, will IMDA(SG gov) allow NLT(or thus Trust) to earn an abnormally high level of ROI? If not, how can Trust, who relies on NLT business, earn higher return? Without a higher underlying ROI, how does Trust improve DPU? It is further loaded with all the extra structure(thus mgmt fee, commissions, agent fees). Obviously, from another angle, IMDA won't let Trust be in the red, either.
(main revenue contributor, residential connection fee is $13.8/mth, it is revised from previous $15, IMDA review this every 3,5 years, only downtrend foreseeable)

Cut the crap, straight to the point
Whether $0.81 a piece is attractive? Will price go up or down? I have no clue, market price is influenced by many factors, I cannot predict the price movement, especially the short term ones. There are also over-allocation units, stabilization period, lockup commitment to 'stabilize' this big creature.

As I'm really new to such biz trust, my analysis could be error-prone, read with your own care. I myself learnt a lot during this study.

I'll apply some shares(with my own money), I want to keep 100 units to revisit it in 2019. It must be a good learning experience for myself.

Monday, May 15, 2017

STI at 50, GDP, HDB, Private Property Index and CPI

Draft: 2017/5/15

Source: Straits Times Article by Lorna Tan, her original article can be found online.

I read this article over the weekend, it is from Lorna Tan, editor of Straits Times Business Section.

I like this article, because it gives a lot of info from STI's history. 50-year is a long duration, with Singapore's own relatively short nation history, it is more precious to have such info.

Quick calc of annualized return on STI over 50-year is 7.2%, and for the most recent 15-year, it is 7.3%.

The benchmark index ended last week at 3,255.29 - a long way from its starting value of 100 points in 1966

During this period, the Singapore Exchange (SGX) has gone through many peaks and troughs, with the latter including the Pan-Electric crisis in 1985, the global stock market crash in 1987, the 1997 Asian financial crisis and, a decade later, the global financial crisis.

The first public mention of the index was on Jan 4, 1967

There are two ETFs tracking the STI that retail investors can invest in: the SPDR STI ETF and the Nikko AM Singapore STI ETF. The SPDR STI ETF has the longer track record, dating to April 17, 2002.

"Since inception, an investor would have received an annualised 7.3 per cent return over the past 15 years, not including dividends which are distributed on a semi-annual basis"

@copyright Straits Times

From this source and source. Singapore GDP were:

It grows 300 times(check CPI section in the later part of this blog) from 1965 to 2015(50-year), and it grows 3 times from 2000 to 2015(last 15 years).

Now to ES3 ETF:
From its inception, the annualized return is ~7+% too, just coincide with the underlying STI, whose performance in term is aligned with underlying SG GDP growth. The performance of the Fund, inclusive of dividends, is net of all charges payable upon reinvestment.. While 7% is nothing fantastic, it is much better than a savings account, a fixed deposit, a usual corporate bond return, SGS, SSB, and even OA/SA in our everyone's CPF account,  although at any one time of a shorter period, it was beaten miserably by one and many of other 'so-called-safer' investment alternatives.

Thus, for any Singaporean who still has a long time horizon before retirement, ES3 is a good investment vehicle, and I'm happily invested into it.

The statistics tells only the history, whether it is to repeat in the future, anyone's guess.

Updated: 2017/5/17
I thought to add another major asset category for comparison, HDB is almost the de facto option for most Singaporeans.

  • 2001Q4: 69.6
  • 2016Q4: 134.6
  • Nearly doubled in 15 year, ~5% annualized.

As usually, HDB flat is brought through a Loan, meaning leveraged, the return is much higher on the original capital outlay.

SG Private Property

  • 1976Q4: 9.8
  • 2001Q4: 87.6
  • 2016Q4: 137.9
  • 14 times in 40 years, ~6.8% annualized
  • Most recent 15 years, 157% increase. ~3% annualized (worse than that of HDB?)
Updated: 2017/5/18
CPI (inflation rate)
With the help from Statistics Gov, I got annual CPI percentage change, a proxy for inflation rate
  • 1962 over 1961: 0.5%
  • Over 55 years, CPI increased by ~4 times
  • Over recent 15s, CPI increased by 1.3 times
Over last nn of Years  55    15  
Average1.026 1.018 
Median1.019 1.010 
Annualized1.026 1.018 
Product (times)4.033 1.310 

Std deviation4.15%2.16%

I love my own study done in this post.

Monday, April 24, 2017

Keppel (BN4) and 2016 AGM

Draft: 4/21/2017, Price: S$6.56, Market cap: S$12b, PB: 1.08; Yield: 3%
Update: 4/25/2017, Source: 2016 AR

With my "First Reit" AGM experience and location of Suntec, I didn't have any issue to locate the meeting venue for Keppel's 2016 AGM. I got registered at counter and was passed a voting handheld device.
Many grey haired retirees were at the scene, most of them ethnic Chinese. It is quite surprised or pondering to me, since I didn't see other races, comparing with what I had seen at First Reit's AGM.

There ought to be 300+ shareholders at the site, a large ballroom was set up in Suntec for this purpose. It was said that this year's turn-up rate is already much lower than previous years' when shareholders turning in with long queue for getting their free buffet and their voucher; both 'goodies' were stopped this year. This change of practice was stated clearly in the 2016 letter to shareholders, nonetheless, the first two shareholders who grabbed the mic started whole sessions with still this topic, one of "Mr. Tan" who spoke with broken English, almost shouted his entire questions at Keppel Chairman Mr. Lee Boon Yang. Questions were asked by many shareholders, some of them well thought, some are simply being deployed to release shareholder's angers or grieves for the loss they might been suffered on the dropping share price. Quite interestingly, a few younger shareholders, in their 30~40s, came up with some relevant questions. Board members decided to take a 10% directorship fee cut, 8 non-executive directors took home S$2m fees this year. Other than "Mr. Tan" who created some 'spice' for an otherwise very dull/routine AGM, there is not much to write about. 

Now, let me take a look at mgmt and board directors' scorecard in 2016.

Keppel is a conglomerate, consists of 4 major business lines:
  • Offshore & Marine
  • Property (Keppel Land)
  • Infrastructure
  • Investment
or its strategy of "Multiple Earnings Streams" on "Urbanization".

It is ironic the cover of AR shows:
when its EVA decreased in a year by S$140m(Pg2) in 2016. A 3-year view shows revenue is steadily coming down..

This, coupled with mgmt's remarks on prolonged headwind faced by O&M sector, shows it still has a long way for Keppel to regain its glory.

From sector info in AR, here is the table of  each segment's business performance

There are also updates on Q1/2017 from the mgmt:
  • O&M order book S$3.5b (exclude Sete), it was 3.7b for year end of 2016
  • Property revenue stay stable, but profit boosted by divestment of two properties in Wuxi and Chengdu/China
  • Infrastructure, profit boosted by divestment of GE Keppel Energy Services.
  • Investment, profit boosted by sales of three pieces of land in Tianjin Eco-city project, and

Keppel's value can be summarized as aggregation of each segment's business value (one of the valuation methods).
  • O&M has an order book that lasts ~3 years, with 3b+ each year, with a barely break-even outcome at current Oil price level, what is the value of this segment, everyone's guess on Oil sector's future, difficult to value.
  • Property, constant residence flat sales, 2b revenue, with 11% growth(if last 3-y history record holds), if we use 13PE or 16EBITDA(from that of Capitaland), I value it at S$xxx.
  • Infrastructure business is stable, since it is more likely to get support from goverment, for power, gas, desalination plant works, I value it at S$xxx.
  • Investment, Keppel internal asset's unlocking vehicle, up to mgmt's plan & financial team's skill to show/release/hold profit, difficult to value.

Keppel is, in Chinese, a "昔日的贵族", strategy is "卖家产" to tide over.
In addition, I feel 10% directorship fee cut, after reading the AR, is aligned with O&M segment, the HC has been reduced from 30k to 22k, and per HC cost reduction is just ~10%, so the directors are with the same percentage of cut.

However, one is directorship fee, the other is employee's lifeboat, can they be the same?

Current yield, not attractive to me, with little light on O&M segment, the biggest contributor of Keppel in hisotry, thus My current plan: wait and part down my position where opportunities arise.

I originally intent to write more, but...

Wednesday, April 19, 2017

QAF (Q01)

Date: 4/19/2017, Price S$1.36, Market Cap: S$775m, PB:~1.45, Yield: 3+%
Source: QAF 2016 AR

QAF's most known brand to me is Gardenia bread. It focuses on:
Bakery  Manufacture and distribution of bread, confectionery and bakery products
Primary production  Production, processing and marketing of meat; feedmilling and sale of animal feeds and related ingredients
Trading and logistics  Trading and distribution of food and beverage products and provision for warehousing logistics for food items
Investments and others  Investment holding and other activities

A few things puzzled me after reading QAF's 2016 Annual Report, which I think QAF can improve in its report in the future:
  • Pg44, many different brands, why not consolidate into a few well known ones to leverage marketing power?

  • Pg 47, Independence of Mr. Didi Dawis, Chairman who is holding 8+% shares, is considered by Nominating committee as independent because of the % is below 10%, Really independent? (in comparison, it is considered significant shareholder for >5% in CN)
  • Pg 84, Advance payment to subsidiaries with no plan of repayment, why would group provision Adv. payment without a plan/expectation for return term, is that causing (at least the concern of) money to be siphoned out of the Group (if such subsidiary is not 100% owned)? The details of this Loan arrangement is surprisingly scarce.

  • Segment reporting, only until Note 41. then I found such info I'm long looking for, and it is not really format friendly. Seems QAF is trying to leave it less noticed in AR.

By sectors:

Too little cost details per sector.

By regions:

It is an Australia company?!?.

There is an one-time profit(divestment of  20% of  Gardenia Bakeries KL in 2016, and remaining 50% of joint-venture's fair value gain), if to normalize it by removing ~60m exceptional item, the PAT will be ~56m(or more, considering the tax impact); ROE 11~12%.

My Plan: to be updated.

Tuesday, April 18, 2017

Side effect of 2G shutdown in SG

This post is not about investment.

I have a dual-sim mobile phone: SIM1 for my SG number, SIM2 for my CN number; SIM2 is in roaming mode, anchored with 2G signal without data service, so that I could continue recv call and SMS from CN.

Due to recent 2G radio shutdown by Singapore's local operators, I got the following SMS yesterday from my CN mobile operator when I was on level 6 of a 5-star hotel at Orchard Road.

Quite amused, as I was no where close to Indonesia.

When Singtel, StarHub, M1 shut down their 2G network, has this use case been thought of?

Side effect: turning of 2G radio signal, renders SIM2 100% useless, and effectively turns it into a single SIM phone.

Only dual active SIM supported mobile is spared from this situation. For example: Huawei Ascend Mate 7, but too outdated a model.

My 'First' AGM experience

Date: 4/18/2017
Updated: 4/19/2017 (on REIT DPU composition)

There are two counters I'm holding have planned AGM today: CCT@10am and First Reit@9.30am, well, since it is my 'First' AGM, First REIT, by its very name, sounds better to me.

I hold First REIT only for less than a year, one of the reasons I like First REIT is its quarterly dividends.

Not being in SG for a decade, I was a bit disoriented when I walked out of Somerset MRT station and was lost in 313@Orchard shopping mall, the security lady pointed me the wrong way, I had to walk out of the mall to Orchard road under the hot sun to figure out my direction to the right venue.
Never attended any AGM  in China before, I'm totally new to the SG AGM and its procedure.

Registration is pretty easy, there are few clerks helping at the counter with IC presented to verify my number of units, the outgoing chairman Albert Saychuan Cheok shook hands with the earlier birds, including me:)

There are few interesting(and few very basic or strange) questions fired by unitholders during the Q&A session, Dr. Ronnie Tan Keh Poo(CEO & Deputy chairman) hold the mic answering majority of those, which I think it will be better if he let his fellow board directors the chances to answer some of the financial questions which he is not so acquainted with.

A show-up of 100+ shareholders, flocking to the adjacent ballroom for refreshment when Dr. Tan adjourned the meeting for counting of the votes on resolutions.

A few takeaways:
  • A "Greenfield" private hospital needs 3-4 years to reach Op CF positive
  • A hospital makes money when occupancy rate stays above 60%
  • First REIT issued a perceptual notes(MAS classified as security) of 5.625% interest, renewable every 5-year
  • Total debt S$400+m, gearing at 31+%(still some room to 45% MAS regulation)
  • First REIT's current average interest on bond is slightly below 4%
  • Growth strategy: acquire 1-2 hospital per annum from Sponsor (LK group)

One thing I don't really get from Dr. Tan's presentation (dare not to ask) is:
Is "-Capital" in DPU a return of unitholder's own investment? if so, it casts a concern to me, while I'd much prefer a 6.6% yield, I definitely wouldn't like it to be achieved in this manner.

Read more DPU articles after STE highlight some of his comment, First REIT's dividend history, but no idea of % of 'capital' return from it.

REIT's distribution comprises the following types:
i) Taxable income
ii) Tax-exempt income
iii) Capital distribution
iv) Other gains distribution

The Trustee and Manager will deduct income tax at the prevailing corporate tax rate (currently at 17%) from the distributions made to Unitholders that are made out of the taxable income of REIT. This applies unless, the beneficial owners are individuals or qualifying Unitholders. In such instances, the Trustee and Manager will make the distributions to such Unitholders without deducting any income tax.A qualifying Unitholder refers to:
  1. A company incorporated and tax resident in Singapore;
  2. A non-corporate Singapore constituted or registered entity (eg. registered charities, town councils, statutory boards, registered co-operative societies and registered trade unions);
  3. A Singapore branch of a company incorporated outside Singapore;
  4. An international organisation that is exempt from tax on such distributions by reason of an order made under the International Organisations (Immunities and Privileges) Act (Cap. 145).
  5. An agent bank or a Supplementary Retirement Scheme ("SRS") operator who act as nominee for individuals who have purchased Units in the Trust under the Central Provident Fund Investment Scheme or the SRS respectively; or
  6. A nominee who can demonstrate that the Units are held for beneficial owners who are individuals or who fall within the classes of Unitholders listed in (a) to (d) above.
Distributions made to foreign non-individual Unitholders will be subject to a final withholding tax rate of 10% (until 31 March 2020 unless otherwise stated).

The above tax ruling does not apply to gains from the sale of real properties. Such gains, if they are considered as trading gains, are assessable to tax on the Trust. Where the gains are capital gains, the Trust will not be assessed to tax and may distribute the capital gains to Unitholders without having to deduct tax at source.

Any distributions made by the Trust to Unitholders out of tax-exempt income and taxed income would be exempt from Singapore income tax in the hands of all Unitholders, regardless of their corporate or residence status.

Capital distribution is regarded as "return of capital" in the hands of the Unitholders for Singapore tax purposes and is not subject to Singapore income tax.

CCT's hardcopy AR doesn't even print a section on "Notice of AGM", what a strange thing; I have to find its date/timing via web search.