Why i think SHJ is fool's gold

I took my first, fresh look at SHJ yesterday and basically came to the conclusion that it looked prima facie cheap in terms of both balance sheet and P&L metrics (e.g. discount to NAV, EBITDA multiple etc.), but i struggled with the accounting so basically came to the conclusion that, although it may be cheap, it wasn't for me. I had a few spare hours to run some basic diagnostics over SHJ this morning just to see whether i was missing anything, and what i found actually debunked my first impressions - even at today's market cap of ~$100m and EV of ~$140m, i don't think SHJ is cheap.

The first diagnostic i ran was to see whether the underlying business is actually growing - an aggressive acquisition strategy can very easily hide deterioration in the underlying business. Broad strokes:

- SHJ reported $28m EBITDA in FY13. Acquisitions conducted in FY13 (Shannon Donaldson, Ron Kramer and Irena Penchant) contributed $2.6m to that period's result, so stripping those out, the underlying business did $25.4m EBITDA in FY13.
- From FY13-FY16, SHJ acquired 8 separate businesses with combined annual EBITs of $17.5m - this is disclosed in the combinations section of the footnotes to the financial accounts. I've extracted the relevant data below:

	Column 1	Column 2	Column 3	Column 4	Column 5	Column 6	Column 7
1
2
3
4	EBIT of acquisitions (annual)		2013	2014	2015	2016	TOTAL
5	Shannon Don.		0.60
6	Ron Kramer		2.40
7	Irena Penchant		0.70
8	Emanate				4.26
9	Browne and Klein				4.61
10	Sciacca				0.94
11	Bradley Bayly				2.80
12	Best Buckley					1.20
13	TOTAL		3.70	0.00	12.60	1.20	17.50

- So, we can deduce from this data that, if there's zero underlying/organic growth in SHJ business', they should be earning approximately what they started with in FY13 ($25.4m) plus what they've since acquired ($17.5m), which is $42.9m.
- According to 1HFY17 guidance, FY17 EBITDA will be at the lower end of a range of $36m-$40m. Let's go with $36m.
- Therefore, not only is SHJ's business not growing, it is 100% clear that it has gone backwards, and quite significantly - something in the order of 10-20% shrinkage in underlying EBITDA over the last 4 years.

So, taking their $36m FY17 EBITDA guidance at face value, that puts the business on a FY17 EV/EBITDA of (140/36) = ~3.9x. A ~4x EBITDA multiple isn't actually that cheap for a business that's actually going backwards - one can look at the multiples of other business which are going backwards (PMP, PRT and SPO are three off the top of my head, although there's many more examples) to verify that.

But, in SHJ's case, their EBITDA-to-cash conversion is woeful, so that 4x EBITDA multiple actually flatters them. Below is what i think the cash EBITDA of SHJ is, using the very basic method of looking at the cash flow statement (taking gross operating cash flow) and adjusting for working capital movements:

	Column 1	Column 2	Column 3	Column 4	Column 5	Column 6
1
2
3
4		2012	2013	2014	2015	2016
5	Receipts from customers		92.9	101.7	134.6	152.9
6	Payments to suppliers		-84.2	-91.1	-121.6	-134.1
7	CF pre working cap adj		8.7	10.6	13.0	18.8
8	Receivables	6.0	8.6	5.4	15.2	17.1
9	Unbilled disbursements debtors	20.0	20.8	19.5	24.2	28.7
10	Payables	-9.3	-15.2	-16.4	-10.1	-13.3
11	Unbilled disbursements creditors				-16.7	-21.0
12	NWC	16.8	14.2	8.6	12.6	11.5
13	Change in NWC		-2.6	-5.6	3.9	-1.0
14	Operating CF adj. for NWC change (adj. cash EBITDA)		6.1	5.0	16.9	17.8
15	Adj. EBITDA		28.0	34.5	44.4	38.4
16	Conversion of adj. EBITDA to CF		21.81%	14.45%	38.14%	46.27%

The key line in the above table is that last one - it tells me how much of their adjusted EBITDA (EBITDA as reported, less non-cash impairments made in FY16) is coming through as cash. Even on their best cash flow conversion year of FY16, they still converted only 46% of adjusted EBITDA to cash flow - a truly woeful metric.

So, going back to FY17 guidance of $36m EBITDA and applying a generous 50% assumed cash conversion rate, i get to FY17 cash EBITDA of around $18m. Applying that against the business' $140m EV turns out an FY17 cash EBITDA/EV multiple of ~7.8x.

My conclusion here, therefore, is that SHJ is fool's gold - it looks cheap, but it isn't. A near 8x cash EBITDA multiple for a business that is very obviously going backwards (net of acquisitions) is nowhere near cheap enough for me.