Digital 9 Infrastructure Share Discussion Threads

[invisage] 29/3/2024 10:50

Stifel downgrades Digital 9 Infrastructure after casting doubt on valuations

Stifel has downgraded Digital 9 Infrastructure from ‘positive̵7; to ‘negative̵7; due to what the broker said appears to be “materially incorrect” valuations, as well as a loss of confidence in the board’s decision-making process.

In a stock exchange notice today (28 March), DGI9 said the valuation of the portfolio as at 31 December was £1.08bn, which represents a 5.6% decline since 30 June 2023.

The decline principally relates to a 16% fall in the value of Verne Global, which has been valued at £398.3m, including gross proceeds already received and deferred consideration due to be received from its sale to Ardian, as well as a possible earnout.

Excluding Verne, the portfolio was valued at £684.1m, up 1.6% since 30 June and down 1.2% since the end of 2022.

Based on the £1.082bn portfolio fair value, the board also released an unaudited NAV of 84p per share (£728m) as at 31 December 2023, which represents a 16% decrease, or £138m, in the second half of 2023.

In a research note published this morning, Stifel analyst Sachin Saggar said an additional asset appears to have been created in EMIC -1, with a valuation of £36m. In the interim results to 30 June 2023, Aqua Comms including EMIC-1 was valued at £227m, he noted.


"This figure based on today's RNS is now £253m as at 30/06/23. Hence, either the interims or RNS are inaccurate. We think it is damning for a board that presided over so much chaos and value destruction," he said.

"We have had confirmation that figures in today's RNS are accurate, but the interim figures were incorrect. In our view, this is a bad look for a portfolio where there are less than 10 assets."

Saggar also said it is "inconceivable" that the portfolio excluding Verne Global has not been materially written down.

"On the surface, it appears as if Aqua Comms is being valued at an EBITDA multiple of c.48x. This is a sector which traditionally does not trade at a multiple of greater than 10x," he said.

"The multiples should be taken with a pinch of salt given costs incurred during ramp up of new cables – but the discrepancy is far too significant and just looks incorrect.

The analyst also drew comparisons to fellow investment trust Hipgnosis Songs. "Overall, the NAV appears to have been prepared based solely on a DCF without any practical sense check against transaction multiples. This is along the lines of what we have seen at Hipgnosis," he added.


hxxps://www.investmentweek.co.uk/news/4191067/stifel-downgrades-digital-infrastructure-materially-inaccurate-valuation

[skinnypope] 28/3/2024 16:26

@keith95 - the VLN very much is ON the balance sheet of D9, it was issued by one of D9s subsidiaries.

The NAV for Arqiva has always been the fair value of the investment netted with the value of the VLN.

While the VLN is often spoken of as "non-recourse", I find this to be mostly semantics as the holder of the note (CPP) have a charge against D9 for the amount of the VLN. In other words, the loan is "non-recourse" in the traditional sense, as the lender does not have recourse to the general assets of D9 plc, but they do have recourse to one specific asset held in one of the subs.

When D9 bought Arqiva back in 2022 for £300m cash + £159m VLN = £459m total enterprise value, they immediately revalued their investment up by £55m in the year end accounts. I think they have largely held that valuation steady and then the NAV has just come steadily lower as the VLN has accreted.

So current NAV = £459m [original EV] + £55m [2022 FV uplift] - £175m [VLN grossed up]

This is £339m versus the NAV today at £340m so I'm happy with the numbers.

[ghhghh] 28/3/2024 16:19

So, I imagine the two non execs resigned because they didn't want their names attached to this valuation which is clearly detached from (current) reality.

That's the obvious concern but why were they trying to continue the business rather than wind up? And why did the BoD turn them down rather than quickly dump a great pile of the brown stuff?

And as WC says, GS and other advisors are discussing realisable valuations with them.

Perhaps the difference is realistic realisable versus potentially realisable?

[rambutan2] 28/3/2024 16:10

"In arriving at their fair value conclusions, for which the Board is solely responsible..."

So, I imagine the two non execs resigned because they didn't want their names attached to this valuation which is clearly detached from (current) reality.

[donald pond] 28/3/2024 15:17

£2.4m on legal fees in just 6 months though. That's insane.

[keith95] 28/3/2024 14:50

"it wouldn't be prudent to take £325 million for it "

From the original sale:

"for an equity purchase price of up to US$575 million ("Purchase Price") (approximately £456 million). The Purchase Price is comprised of the following"

I'm wondering why someone has forgotten both the earn out and the deferred consideration payment.

Problem reading past first sentences because of the red mist maybe?

Easy solution - get an independent reader to explain the details.

[bdbd11] 28/3/2024 14:16

I can see this slowly rolling forward here. Fortunately as a newcomer i'm sitting on paper profit. The discount to nav is just too large to ignore. I added a few more this morning and probably will add again. I'm not expecting 84p back, probably no where near it, but 22p has little risk imo. Realistic 50% to double bagger from here with patience.

[loglorry1] 28/3/2024 14:06

I think the BoD are a bit behind the curve. They wanted to get the wind up vote done before they start marketing the other assets. Up to now I think they've been focussing on Verne sale and things like VAT insurance.

They basically used the same methodology and the same numbers from TP and came out with the same NAV - give or take. That's not surprising since TP get paid based on NAV and also why trash the value of your own assets before you sell them.

I think we are worth more than current share price but I think achieving NAV is not going to happen. Would love to but its not gonna be that.

[williamcooper104] 28/3/2024 13:56

You'd have thought that the BoD would sense check them with GS who given they'll likely do the selling and don't have to compete to win a mandate would give a conservative view

[ghhghh] 28/3/2024 13:47

I doubt very much they have a clue if it's achievable or not.

But they are in the process of selling these assets, by now they must be talking to professional advisors about marketing and pricing?

[loglorry1] 28/3/2024 13:17

I think you'd know more than most if a vulture sniffed - perhaps only when they have a cold ;-)

I think the BoD just fed the TP numbers to an "independent" firm who crunched a few discounted cash flow models and came up with a NAV. I doubt very much they have a clue if it's achievable or not.

[ghhghh] 28/3/2024 13:06

It's pretty clear nobody wants to buy the company!

Bit early to write this off!

The Verne sale only just completed, the revised NAV out today and yet the shares stuck at 21p. There must be a few opportunistic vultures sniffing around, assuming vultures sniff?

[ghhghh] 28/3/2024 12:46

Log

But if Miller and Boleat want to reboot D9 why wouldn't the existing BoD jump at this if today's valuations are so flakey?

[b1nky] 28/3/2024 12:09

No idea why no one is attempting to bid for the whole company. Judging by the fact that there are sellers at this level, 22p, I would imagine they would pick the entire company up for 30-40p. Then, sack the many advisors and banks, etc. then go about breaking up the company and selling off the various parts. I'm sure any bidder could easily double their investment in quick order.

[williamcooper104] 28/3/2024 12:07

Still quite unusual when the biggest bears can come up with a realisation thats >2x current share price Anyways happy Easter all Alas it'll be at least another Easter before we get out of this

[fisternator] 28/3/2024 12:00

Agreed Skinny.

Alan.
That's my point. The valuation set out today are here nor there, they can not be taken seriously. Even if they prove to be correct, can you trust the BOD to extract that value?

The previous example i provided of Verne makes it clear that the bod nor their execution of realising valuations are remotely competent.

have a great Easter all.

[skinnypope] 28/3/2024 11:59

Oh dear me, what a wasted opportunity for the company to "reset" the financials.

1. Aqua/EMIC-1, Elio and Arqiva have all been valued using the same numbers and methodology as previously used. This is classic "Garbage In Garbage Out" modelling. I had expected/hoped that an INDEPENDENT valuer would use INDEPENDENT numbers?

"...valuation was informed by an assessment of portfolio companies' operating models and cash flows provided by the Investment Manager, based on the five-year business plans prepared by the portfolio company management teams and approved by their respective Boards"

2. What IS interesting is that SeaEdge has been valued externally, and that the value shows the NAV was 20% too high:

"The value of SeaEdge has been provided by a real estate adviser as part of the sale preparation processes..."

3. Still no disclosure on the Verne earn-out. I see £52.7m valuation equalling about £73.6m future value [using 11.8% discount rate], so implying ~70% probability of a full payout. How useful are these numbers, we have no idea, so I will continue to value this at zero.

As an ex-colleague of mine was fond of saying when he received useless communications, "how many electrons died writing this pointless report?", which is pretty much how I feel about this RNS.

Looking ahead, I've now re-run some numbers. Let's use Verne as the model, and the June23 valuation of £480m as the baseline.

Sale proceeds = £325.8m {up front] + £19.5m [deferred payment] - £17m [costs] - £23m [VAT indemnity provision] = £305.3m

This is 64% of NAV, and all I have used are stone cold facts, no estimates, and also I would say a worst-case outcome. SeaEdge has been written down 20% in this RNS so maybe all-in post costs could be similar worst case outcome of 64% of June23 NAV.

Applying the same 64% number to the new portfolio NAVs, RCF dropping by the Verne deferred payment, and holding everything else constant gives a new "kitchen sink" number of 46.3p per share.

[ghhghh] 28/3/2024 11:58

But Verne was distressed.

Re Elio, if they are intending to sell this year then they would be nuts to grossly inflate the NAV since will destroy their credibility and encourage other potential buyers of the remaining assets to make derisory offers.

I assume these valuations will be the prices D9 will market at so offers will be less but within 10% to avoid egg on faces

And the first asset to sell will be closest to the above valuation

[riverman77] 28/3/2024 11:55

Surprised not bounced too, but providing they can sell assets anywhere close to NAV (in fact even at a 50% discount) then substantial value will eventually come through - that's just a mathematical certainty.

[alan pt] 28/3/2024 11:51

Regardless how we got here, my question now is: If the Verne completion didn't make it bounce and the valuation didn't make it bounce, what will cause a bounce?

I struggle to see further good news on the near horizon. Any of the additional asset sales must surely take some time to agree?

I can't really see the whole thing as a takeover target, surely if anyone was interested in doing that then they would already have done it?

[fisternator] 28/3/2024 11:36

Keith.
If i had an Asset valued at £517 million in December 2022 and the market remained stable , then for reasons unbeknown to me it was revalued valued at £398 a year later just before a sale. I honestly don't believe it wouldn't be prudent to take £325 million for it ; granted, like i already stated with a wing and prayer for an extra £100 million.


The example i provide is how you should take these sorts of NAV provided today at buyers beware status. Wasn't the Verne NAV provided by the company in Dec 22 as earnest as todays NAV on their assets?

Verne was valued at a NAV of £517 million in December 22 and the actual amount after all sorts of fees is less than £300 million to shareholders. Again. Yes i know the carrot of a further payment in 2 years exists also.

Make sense who can of that discount to NAV from December 22. For all i know the same thing will happen to the values provided today!

[williamcooper104] 28/3/2024 11:35

Shame as rather than selling Verne they should have sold everything else - giving then Verne capex funding and then internalised to create what would then have been one of the best data centre REITs (there aren't many) But that boats sailed now

[williamcooper104] 28/3/2024 11:23

Plus when the asset values fall as assets sold external management would be cheaper than internal - as you can't reduce salaries just as easily