Astrea V Class A-1 PE Bonds

Astrea V is offering 3 classes of bonds of which S$180m of the Class A-1 Bonds ("the Bonds") are available to retail investors. The public offer will end on 18 June 2019 at 12 noon.

The remaining $135m Class A-1, US$230m Class A-2 and US$140m Class B Bonds were already fully placed out through a placement tranche.

Details of the 3 different classes of bonds are summarised below:

Key points to note about the Class A-1 Bonds:

• Interest rate of 3.85% per annum
• Interest is payable every 6 months on 20 June and 20 Dec each year
• The Bonds will be redeemed on 20 June 2024 (5 years) if conditions are met. 
• The Bonds are expected to be rated A by Fitch and A+ by SnP. 
• There will be a bonus of 0.5% of the principal amount for Class A-1 if the Performance Threshold is met.

I am not going to repeat the good work done by other bloggers:

You can read Financial Horse' piece of Astrea V here. 

You can also read about Bondsupermart research on Astrea V here.

In addition, the Issuer has spent a of efforts to educate investors on how what is private equity and how Astrea V PE Bonds work. Go watch the Explainer Video below

and listen to the management presentation below.

You will find all the resources you need on its website here. Don't be lazy! 😆

The structure is also highly similar to Astrea IV, you can read my write up here.

As such, in this post, i will not repeat what the management has already said but will touch on 5 key points that investors will likely want to know.

1.  How was the 3.85% Interest Rate Set and is it relatively attractive?

According to management, the interest rates for all the 3 tranches were priced by institutions and the same rate for Class A-1 was then offered to retail investors, meaning that the interest rate was set by a proper book building process.

The placement tranche was 7.3x subscribed (around US$3.4 billion received) and 70% were placed out to high quality institutions such as Insurance, Endowment and Foundations. The remaining 30% were placed out to high net worth individuals. If you recall, Astrea IV was "only" 4.5x subscribed.

There was a significant step up in orders from Institutional Investors, meaning that the PE Bonds has now gained wider acceptance since its first launch in 2016. This led to more participants for the offering and thereby lowering the rate (sadly for us).

According to my friend working in the fixed income department of one of the leading insurance company here, the Class A-1 book was very very hot investors, with over 10x demand for the S$135m bonds. Many investors were cut back drastically. In his words

Issuer could have tighten the Class A-1 pricing all the way down but have decided to leave "some money on the table", probably to benefit retail investors.

Investment Moat has done some analysis on the current trading levels of Astrea IV. However, he has used the yield to maturity to 10 years which is actually not right as Astrea IV PE Bond for Class A-1 is most likely (99%) going to be redeemed at end of Year 5. According to the same bondsupermart website, the yield to call for Astrea IV Class A-1 is around 2.124%. Including the 0.5% interest rate, the yield to call is around 2.27% based on the last traded price of 1.065.

In this regard, the Astrea V Class A-1 Bonds of 3.85% is actually quite attractive compared to the current trading levels of Astrea IV Class A-1 Bonds even when you add in a one year extension premium.

If I look at where the trading levels of Astrea IV Class A-1 Bond is currently, there is probably some upside but do note that this "upside" is meaningless if one buys at IPO and hold till maturity as the returns are fixed at 3.85%.

If I compare against the risk free Singapore Savings Bonds, the 5 year average interest is now 1.98%. As such, at 3.85%, the spread is around 187 bps. If you intend to hold till maturity, you will have to judge for yourself whether you are better off earning a lower interest of 1.98% in the SSB or earn the 3.85%

2. Is Astrea V portfolio better than Astrea III and Astrea IV?

The Astrea V portfolio is 5.4 years compared to around 7 years for Astrea III and Astrea IV. According to the SnP report (page 3), the younger portfolio is actually a "strength" as it has higher upside potential over the remaining life

We also see similar comments from Fitch whereby its says that 84% of the funds in the Astrea V portfolio are in the top 3 quartiles and 54% of the funds exposure are in the top 2 quartiles. If you superimposed the funds quartile on the historical performance of the funds in the different region, you will know that this portfolio is a "solid" one with great potential to outperform the median cases. 

My conclusion is that this portfolio is even better than that of Astrea III and Astrea IV as a younger portfolio has more upside potential. The cash flow will become generative over the next 5 years and this will be able to meet the cash flow needs of Astrea V.

3. How has Astrea III and Astrea IV performed? Does it has any bearing on Astrea V?

The answer is no and yes. 

Astrea III has performed well and the bonds have been upgraded. The LTV has also fallen to around 31% in Jan 2019. In addition, the Class A-1 bonds of 3 years (3.9%) will be fully redeemed in July this year.

Astrea IV has done well as well with Fitch upgrading the Class A-1 to A+ in May this year. In addition, it has just paid out the second distribution date interest rates. You can see the report here. In fact, the reserves is on schedule at US$79m and the Performance Threshold is at 46% (145m out of 313m) after one year. Meaning there is a high chance that the bonus of 0.5% will be triggered once it hits 100%.

No - Each Astrea is different and unique and the assets are ring-fenced. Past performance is also not an indicative of future performance.

Yes - You can see that the Manager has established a strong track record. I guess that is part of the reason why we see more institutional participants in Astrea V as track record do mean something. 

4. Will we likely see a default similar to Hyfliux or Swiber? Is there a possibility that i will lose my capital? 

The answer is that default risk is very remote. While the schedule call is on year 5, the legal maturity is 10 years. In the event that investors get extended, then Class A-1 bonds holders are compensated with a step up in interest rate. 

Let's analyse what the default risk is for Class A-1. The Issuer has done some severe stress testing using historical data. (Read page 141 onwards). You can see from the table that under severe stress testing, Class A-1 will be paid on year 5.5 and Class A-2 in year 7 (due to the timing priority in redeeming Class A-1 ahead of Class A-2). 

In addition, that was validated by an independent 3rd party research consultant from Harvard. You can see from the probability table below that Class A-1 will at most be delayed till year 6 (12th period) but 99.3% chance that they will be redeemed in year 5. 

SnP  also concluded that its modeling results gave Astrea V Class A-1 a "AAA (sf)" rating but was unable to give it such a rating due to its rating constraints. "AAA sf" rating is actually quite close to a "Temasek" rating and pretty rare in today's world. I would venture to say that the risk of low for Class A-1 is very very remote. If you think about it, given that the reserves is on a straight line basis over 5 years, we only need 2.5 years of scheduled reserves to fully de-risk Class A-1 Bonds as they have a timing priority in the redemption at end of year 5! 

I was trying to tie the portfolio value to the loans issued and was looking at the financial statements and cash flow of Astrea V. Given that the portfolio is newly acquired, i noted that there is a $70m cash in the balance sheet. This is probably the "leftover" cash from the portfolio acquisition. In other words, the real "Loan to Value" of Astrea V is actually much lower, at 43% if we include the beginning cash balances. 

In prior Astreas, the structure always keep the cash received from 1 April to the Issuance date. See the first distribution report for Astrea IV page 6, where $14m was from beginning cash balances. In other words, the first period of cash flows are already well covered unless all the cash are drawn for capital calls from the underlying funds (which are unlikely).  

Finally, the last question that you want to know:

5. What is your IPO Chilli ratings and how do I get some shares?

As you are aware, my chilli ratings are designed for equity IPOs and bonds are not meant for "flipping". If you force me to use the same ratings on Bonds, i will give it a 3 Chilli for the very very low likelihood of default and a 2 chilli rating for the interest rate. While it is priced attractively relatively to the current trading levels of Astrea IV, i would nevertheless prefer to see the bonds paying a more attractive coupon of at least a 4% handle. 

However, beggars cannot be choosers and i truly appreciated the work done in bringing such innovative products to the market.  There is hardly any rated bonds out there and I would buy this bond at 3.85% anytime over the unrated 3.03% Singapore Airlines Retail Bond as my risk is diversified over 862 companies.  

So coming back to allocation, this is the paradox we are facing. Nobody "wants" the SQ bond and you could have gotten close to what you asked for as Company can easily upsize the issuance. In Astrea V, you can't. The bonds are fixed at S$180m and this will have to be shared with all the applicants.

So what is the application strategy?

In the gatefold, the Company actually stated its allocation policy. This is the first time i ever see that an allotment policy being mentioned. This means that the Issuer probably know that based on prior subscription of 7.4x, Astrea V is going to be similarly over-subscribed and would like to let investors know about this "allocation policy".

I have no issue with such an allocation policy, in fact, i am truly glad to see that they are being upfront about it and listen to feedback from investors. I am just not sure if all the applicants who applied for $50,000 or more will know that they are subject to ballot as they may have wrongly assumed that the allocation policy will be the same as Astrea IV. 

So my prediction is this (assuming same subscription rate as Astrea IV):

If you apply for less than $50,000, you are going to be allotted $5,000 or less. (maybe slightly more due to the bigger issuance this time)

If you apply for more than $50,000, you should get at least "double" of what you got last time in each band. 

If you control many CDP accounts (wife, kid, grandfather, grandmother) and you intend to hold till maturity, then it is better to spread your applications (say $20k per application) around to as many applications as you can. If you don't have the luxury and you trust only your own CDP and yet still want to get more shares, then I would encourage you to just go for it.  😂 

Using the example of applying $20,000 each using 5 accounts, you will likely get 5 x $5,000 = $25,000 bonds. If you apply for $100,000, then you are subject to ballot and if the lucky stars shine on you, then you may get $14,000 worth of bonds. 

Do you want to bao tiok (100% get allocated) or do you want to tikam (ballot) in exchange for a more meaningful allocation? This is the vote for today!

Polling Time - Are you going to apply for Astrea V Class A-1 Bond?

You can assess the poll here or click on the picture below.