Transversal
Thanks for taking the time to put together that excellent post - i've never taken a look at GEM and your post is an excellent introduction for me.
A couple things i'd clarify or query:
- Do you think the declining employee cost:revenue ratio is as a result of scale? I'd say it has more to do with the fact that pre-kindergarten childcare costs (i.e. GEM's revenue line) have consistently grown at greater than inflation over the last decade, while wages for childcare center workers have grown less (i'm not 100% sure on the latter but it seems to me to be the case, given that there's intermittent mass media coverage and outrage at how little childcare workers are paid). Therefore, all other things (such as child:carer ratio and occupancy levels) being equal, i'd expect that to be the main driver of things here.
- If GEM has a NIBD/EBITDA ratio of 2.0, i don't think that means they can extinguish their entire debt in two years. Let's say the numbers there are (200/100); that $100 of EBITDA services, say, $10 of interest ($200 * 5%), a few (say, $3) dollars of D&A/maintenance capex, and then the tax man gets hits 30%, so cash flow available to pay down debt is more like (100 - 10 - 3 - 26) = $61, and that assumes no additional investment in working capital and a 100% EBITDA-to-cash flow conversion metric. So, at 2x NIBD/EBITDA, the business would actually take more like 3-and-a-bit years to extinguish its debt.
- On your cash flow conversion metric, you may also want to adjust for net working capital movements to get a more accurate measure of how much EBITDA is, in fact, converting to cash. I suspect (or would hope) that there's not much working capital movement going on for GEM given the nature of its business, but NWC movements can make substantial additions or subtractions to cash flows in any given period so best to keep an eye out for them.
- As a general principle, i'm not sold on the idea of taking a 3-year forward estimated earnings number and comparing it today's price to justify something's good value or cheap. Firstly, a lot can (and always does) happen in 3 years; secondly, using your FY19 EV metric, i can justify about 95% of the stocks on the ASX being cheap at ~10x FY19 forecast EBIT. Best just to use 1-year forward, i think, unless you have supreme certainty about the growth profile of the business (which i think is difficult for any business), or the 1-year growth profile of the business substantially undersells the more medium term growth profile.
With those nits picked, perhaps the biggest potential value-add suggestion i have for you is to investigate whether the law of diminishing returns is likely to apply to GEM. This is a fundamental point of importance for considering all businesses engaged in roll-up strategies; the logic goes that, as the business gets bigger and bigger, it requires more and more acquisitions each year to maintain the same level of growth (which tends to get harder and harder, particular for granular businesses like childcare centres). Not only does the business have to acquire more and more each year to maintain the same earnings growth rate, but roll-up players tend to pick the low-hanging fruit first (i.e. they buy the better businesses first), and the acquisition multiple tends to slowly creep up over time (probably because more and more acquirer targets hold out, or because management get increasingly willing to push the envelope on multiples so they can maintain their earnings growth rate).
So, with that in mind, where i think you could extend your analysis is in a few areas focusing on GEM's acquisition strategy:
1) How many centers have they acquired each year going back, say, 10 years?
2) How has the multiples for their acquisitions changed over those 10 years?
3) What's going on in the underlying business? I don't know exactly what GEM disclose, but the way i attempt to deduce this with roll-ups is as follows:
Say FY15 EBIT was $110m. If the acquirer did a bunch of acquisitions in FY15, it should be disclosed in the footnotes to their financials a) how much those acquisitions contributed to the group's FY15 EBIT, and b) if all those acquisitions occurred on 1 July, how much those same acquisitions would have contributed to the group's EBIT. So, let's assume GEM did $110m EBIT in FY15 (i'm making numbers up), and footnotes to the FY15 financials say: "acquisitions, all done on 31 Dec, contributed $5m to the group's FY15 EBIT. Had all these acquisitions occurred on 1 July, they would have contributed $10m EBIT for FY15". With this knowledge, we can deduce that the new earnings base for FY16 is $115m (being $105m of what the underlying business did in FY15, plus $10m of what FY15's acquisitions should contribute on an annualized basis). Once you start isolate the impact of acquisitions like this, you can start getting a good feel for what's going on in the underlying business (and whether management are using acquisitions to compensate for deteriorating underlying growth), and you can ran multi-year analysis to isolate underlying/organic vs. acquired growth.
Hope that has been of some use and, again, thanks for your post.
Expand
(20min delay)