Why the Poor Return on Equity?, page-4

I understand your question and concern, but i again politely suggest your focus on ROE isn't hugely relevant to an assessment of the merits of investing in GXL today. Why do i say that? Because ROE is a highly reductive balance sheet metric that's a crude summary of years of accumulated corporate history, yet what i am interested in as a prospective investor today is the future cash flows of the business relative to today's market capitalization. The focus on future marginal ROIC - i.e. what will GXL generate from each marginal dollar of capital spent - is far more important to understanding what future cash flows could be, and i personally don't forecast out detailed balance sheets so i can't tell you exactly what i think GXL's ROE will be in 1, 2, or 5 years' time, but suffice to say i think it'll be higher than it is today. And in terms of looking at balance sheets of overseas comparable companies, you have to be particularly careful with US companies as they use US GAAP vs. the ROW GAAP accounting system - so, for example, while ROW companies (incl. Australian companies) can mark-to-market their physical assets annually, the US system prefers historical cost.

To give you an example: this year, my guess is that GXL will grow its invested capital, as displayed by the balance sheet, by about $35-$40m. I get to this figure by saying they'll invest ~$35m in greenfield stores and co-located vet stores (+35), there'll be about $20m D&A (-20), they'll spend that same $20m to keep their existing store network maintained (+20), and maybe there'll be a small investment in working capital (+5). From that $35-$40m invested into the business, i estimate there'll be a ~$10m uplift in EBIT over FY16, because that's roughly what management are guiding to. So, on what i consider the far more relevant metric of marginal return on each new dollar of capital, GXL will generate something in the order of 25% this year (10/40). And if marginal ROIC is ~25%, then marginal ROEC will be even higher because GXL's debt definitely is cheaper than 25% p.a. But that's not all, because the FY17 EBIT number will be depressed by the 15-20 new stores and 15-20 co-located vets opening as most of these greenfield sites run an EBIT loss in the first year of trading, so the $35-$40m doesn't capture the full investment because these trading losses run through the P&L rather than the capital account (but, obviously enough, GXL expect those newly opened stores to deliver profits from Y2 onward).

To look at it from another angle: Mammoth, i.e. the GXL retail business pre-acquisition, did $8.5m NPAT on $38m net assets in FY13 (note 3.3.2, Explanatory Memorandum 2013), i.e. a ROE of 22%, and that's being harsh because you should use average equity balance and not just the closing number (as the profit is earned on the equity balance throughout the year, not just on the closing equity balance). And, on a standalone basis, Mammoth was forecast to do FY14 NPAT of $10.9m, and i calculate their rough closing FY14 forecast equity to be ($38m starting position + $29m investment in PP&E - $16m debt draw) = $51m, meaning their average FY14 equity balance was (38+51)/2, or ~$45m. So their FY14 standalone ROE in the hands of the former owners, i.e. the PE groups that owned Mammoth, was in the order of 25%. Using what i think is the more salient metric of ROIC (stripping out effects of capital structure), Mammoth was forecast to do $24m FY14 EBIT on FY14 average total assets of perhaps $175m (they started with $164m, then were forecast to add ~$25m net of depreciation), which gives a ROIC of ~14%.

Now, obviously, GXL came along and bought that business at a substantial premium to book value - the book value of Mammoth's assets were $164m when the transaction occurred, yet GXL effectively ascribed $403m to the value of this business in the merger, so the pro forma ROIC of Mammoth went from 14% pre-transaction, to ~6%, quite literally overnight and with the stroke of an accountant's pen. Did Mammoth become a ~60% worse business because it's ROIC dropped 60% following this deal? Absolutely not. Did GXL overpay? Maybe a little - i don't know exactly what the right ROIC or EV multiple is for an immature but category leading retail-platform with years of growth ahead of it, but a 17x EBIT multiple (24/403) isn't as absurd as it seems at first blush. Moreover, the sellers (Mammoth) took the entire consideration as equity in the newly merged group, meaning that the potential impact of overpaying wasn't imposed solely on existing GXL shareholders but was instead diffused between old GXL shareholders and Mammoth's owners.

Did GXL overpay for past acquisitions and has that pushed down what is reflected in today's balance sheet and ROE? Yes, i think they probably did, but i think more so on the vet side of the business than the retail side of the business, and it's partly why i watched GXL from the sidelines for years. Does it really matter to me (or any prospective investor in GXL) how much GXL paid for its many acquisitions anywhere from 2-10 years ago? Maybe at the margins, but i personally think the relevance of historical ROE or ROIC pales into insignificance when compared to an assessment of the business' future earnings potential relative to its current market capitalization.