I communicated with one of our shareholder this week about some foreign listed businesses; I thought I might share some of this correspondence with blog readers.
By way of background, the fellow has been an investor with EGP for a number of years, and was our member auditor at the 30 June 2015 balance date. I met him for the first time in Kuala Lumpur this year when I travelled to the UOS AGM (he was deployed in the city for work) & while we were chatting, I suggested as a bright young fellow he might usefully employ some of his spare time while in Malaysia reviewing Bursa Malaysia to see if there was anything that really stood out. Like many of our investors, I suspect there is a bit of successful self-directed investing in his future.
Last week, he sent me through four names and although it’s a hectic time of year, I committed to look at one of his selections.
My letter to him is set out below:
I have spent a little time today on the first of your suggested Malaysian businesses, Apollo Food Holdings.
I have attached the three pages from the 2005, 2010 & 2015 Annual Reports I have used to get the trend figures over the last 15 years.
I have also attached a spreadsheet where I have built a simple model to show two potential futures (I will explain further on).
Price today is (I will use $ as a substitute for MYR in the following – easier) $5.25. I will assume that as the buy price in all return calculations.
If you look at the first of the third page of the ‘Apollo Headline Numbers’ .pdf, you will see I’ve averaged 2001-03 & 2013-15. I prefer to do this when I’m calculating growth rates as it evens out peak/trough years. This allowed me to calculate a revenue growth; I reckon the underlying rate is about 6.7% per annum.
A consumer goods company that exhibits relatively strong revenue growth rates such as this would usually be valued at well above the market median. Unfortunately, Apollo has only turned this 6.7% revenue growth into about 6.1% PBT growth over the period.
It is apparent to me that they listed this at a very opportune time, as sellers very often seem to do (asymmetric information).
If you look at cells B41, B43 & B45 on the first tab of the spreadsheet, you will see in FY2000, they were generating 19 inventory turns and $1.80 of revenue for every $1 of PPE & $0.97 of revenue for every $1 of tangible equity. The inventory turns seems to have stabilised at around 11x and the revenue per unit of tangible assets or PPE have both begun to improve over the last few years.
On the second tab of the spreadsheet, I have modelled things more or less continuing as they have done expenses and revenues each growing at about 6.7% annually. The result of course is that EPS will grow at around 6.7% annually. Recent announcement indicates flat revenue growth for the first quarter of FY16, but we are really looking out a bit further with what we’re trying to do.
A company that earns 6.7% annually in perpetuity and pays out 60% of their earnings in dividends will return the buyer 10.72% (0.067*0.6+0.067) annually. This assumes the starting and finishing multiples are the same. I have worked on a 14x multiple (cells on row 18) for Apollo in 2020, which is about what I estimate the market would pay for a business operating like this. If you look at the 4th tab, the IRR for a 5 year investment is about 9.1%, lower than 10.72% because the multiple being paid today (about 16.6x at $5.25) is higher than the multiple being forecast 5-years out (14x).
The third tab on the spreadsheet models what might happen if Apollo managed to introduce some operating leverage into the business. I have used 6% revenue growth (below historic averages) and 3% expense growth. A business with a large revenue stream that can grow that stream at twice the rate of their cost base is a VERY good business. Given the 60% payout ratio, the EPS growth under the scenario is a high teens number. I have allowed for some multiple expansion to about 18x because the market eventually will pay up for a business that demonstrates it is capable of such value creation.
On the IRR tab of the spreadsheet, you will note that if the above unfolded as modelled, a return more like 23.8% per annum would be realised over 5 years.
Apollo Foods is a good business, no doubt. Over the last 15 years, they have generated about $2.4 billion of sales, about $320 million of Net Profit After Tax, and paid nearly $207 million in dividends. There was a pessimistic time after a few poor results in 2002 when the business could be purchased on market at a market capitalisation of less than $130m. Best of all, the business has done all the above without ever coming to shareholders for more capital. The 80 million shares on issue have remained constant through Apollo’s history; it is a genuine business generating genuine free cash flow for shareholders.
However, in order to be a VERY good business one of two things needs to happen. Either revenue growth needs to start to outstrip expense growth, or a major reduction in the underlying cost base needs to be found. The easier of these two seems to be keeping a lid on future cost growth as major cost reduction programs are hard to implement. Having strong reliable revenue growth on your side is a big advantage.
At current prices, the business does not strike me as remarkably cheap, but if I were to see changes taking place that would lead to operating leverage at anywhere near that provided for in the second model, I might be interested.
That said, there are plenty of people out there for whom a 9.1% annual return would be perfectly satisfactory – I just aim a little higher.
The other way a better investing return could be generated out of the modest operating performance described on tab 2 of the spreadsheet. If you monitor businesses such as this closely and remain alert when there are meaningful price falls, you could get a quite satisfactory result. If for some unexplained reason the price in the tab 2 scenario dropped from the $5.25 of today’s trade to say $4.10 tomorrow, the 9.1% return modelled becomes more like 15.1%. 22% falls in the trading price of a business are exceedingly common, as long as you can satisfy yourself that the underlying fundamentals remain the same.
This analysis, I should point out is fairly cursory, I am modestly familiar with the brand range, having seen them is shops when I visit Kuala Lumpur, but I have no special insights into the markets Apollo operate in, this is more of an overview of how I would initially look at the financials of a newly discovered company.
If you have any other questions, let me know. I am unlikely to get to the other 3 companies you nominated, but I’d be interested if you have the inclination to hear whether you think any of them stand out upon closer review in light of how I’ve looked at Apollo – Tony
I hope that thinking set out prior is useful to anyone considering how they should approach analysing a new business.
The week was a poor one for the fund, our second largest holding Dicker Data was down 7%, I’m not sure why. They apparently didn’t suffer much disruption/damage from the cyclone that struck Kurnell (where the distribution warehouse and call centre operates), which is very fortunate given the very substantial damage their neighbours at the desalination plant suffered. Just serves as a useful reminder that no matter how well you think you understand risks, there always lurks the possibility of the unexpected – Tony Hansen 18/12/2015
|
|
Apr 1st 2011 |
Jul 1st 2015 |
Current Price |
Since July 1st 2015 |
Since Inception |
|
EGP Fund No. 1 |
1.00000 |
1.57872 |
1.62650*1 |
3.03% |
91.30%*2 |
|
37333.23 |
50922.68 |
49504.99 |
(2.78%) |
32.60% |
EGP Fund No. 1 Pty Ltd. Up by 3.03%, leading the benchmark by 5.81% since July 1st 2015. Since inception, EGP Fund No. 1 Pty Ltd is Up by 91.30%, leading the benchmark by 58.70% all-time (April 1st 2011).
*1 after a 31 May 2013 dividend of 2.333 cents per share (cps) plus 1.000 cps Franking Credit, a 31 May 2014 Dividend of 7.000 cps plus 3.000 cps Franking Credit and a 31 May 2015 Dividend of 8.6667 cps plus 3.7143 cps Franking Credit
*2 calculated based on dividends reinvested
