**OLD ARTICLE – Originally posted on April 5, 2016**

Investment holding (IH) companies insert an extra layer of cost between the underlying assets and the ultimate investor. In theory, this layer of cost includes management which allocate capital like a fund manager, thus generating alpha with the IH company’s capital.

The reality, though, is neither the structure is free nor the people running do so for charity. Everyone is in it to make money.

Hence, *after* valuing the underlying investments, we tend to *also* value the “holding company”. This holding company tends to be loss-making and, thus, the fair value of it tends to be negative.

Therefore, we set off its negative fair value against the positive fair value of the underlying investments and you arrive at what investors commonly refer to as “investment holding company discounts”.

Consider Astoria (ARA), a locally-listed permanent capital IH company aiming to buy big blue chips mostly listed in the USA markets, with some other unlisted and related investments. Read more here.

A key phrase here is: *permanent* capital.

Therefore, the expenses that Astoria incurs existing are perpetual. Hence, I use the perpetuity formula (read here about it).

Firstly, a starting point is: how much does it cost Astoria to exist?

Well, there is Anchor Capital’s management fees and Astoria’s own costs, but collectively Anchor estimates that those fees will be c.1.6% of Astoria’s NAV (i.e. the Total Expense Ratio or TER). As the TER excludes certain fees, this is likely under-estimating Astoria’s real cost, but I’ll use it nonetheless as a good indication.

Then the key variables to value a perpetuity are:

- Perpetual cash flow: -1.6 cents y/y (i.e. based on the estimated TER with a theoretical NAV of 100cps)
- Cost of Equity: My estimates of a equity-based CoE for USA equity markets is around 7% (read about it here). I believe that yields are artificially low at this point, so let’s adjust this CoE upwards closer to 8.5%.
- Growth in cash flow: As Anchor is Astoria’s major cost
*and*Anchor’s asset management fee is a % of NAV, it means that Astoria’s cost base grows with it (i.e. is mostly variable). And therefore the growth in its cost base will be largely based on the growth in its NAV, thus USA average equity returns are c.10% y/y (read here). In real-terms to match it to our CoE, this is closer to 7% y/y. I believe that forward growth rates may be much lower going forward, so this might be closer to 3% y/y (let’s be nice and conservative).

Hence, per 100c of Astoria NAV, in the end, its cost base will cost 29c (=1.6c / (8.5% – 3.0%)).

That is an effective c.29% discount that Astoria should be trading at, despite what *appears* to be a relatively inexpensive cost-base.

But, let me phrase it this way, if Astoria grows at 3.2% y/y, a 1.6% TER is effectively costing investors *half* of this growth. If you are giving away half of your upside, then I hope that a 29% discount suddenly doesn’t sound too crazy to you.

Now the above numbers are just thumb-sucks and the CoE may be higher, the growth may be quite different and the TER, overtime, may lower as returns to scale kick in with some of Astoria’s fixed costs (listing fees, Board fees, etc), but this gives you an idea how to go about calculating IH discounts.

Getting back to Astoria, the argument is that the structure of the IH has value. Especially in Astoria’s case, it has tax and frictional efficiencies over many alternatives of private investors going direct into foreign markets.

Thus, should this benefit not be worth something?

Absolutely, so let’s say it is worth a “bonus” 10%, which leaves Astoria’s implied IH discount at 19% (= 29% – 10%). Let’s round that up to 20%…

At 31 December 2015 Astoria’s NAV was $0.97 per share. Let us assume it has remained flat since then, which means its current NAV is now around R14,41 per share.

Astoria is currently trading at R14,80, thus not only is ARA *not* trading at a discount to its NAV, but it is actually trading at a premium.

Could the structure be that valuable?

No, and of this I am pretty certain. A lot of Astoria’s IH structure’s advantage is tax-based and tax is a *percentage* of things. So a tax advantage will *lower* the IH discount, but it could never make it positive (as far as I can see).

So it is much more likely that Astoria is just overvalued and should rather be trading at a price closer to R11,53 (= R14,41 x (100% – 20%)), assuming you give the “structure” a 10% benefit. I hope these thoughts make sense? I suggest you play around with your own cost-base, Cost of Equity and average return assumptions to arrive at your own sense of the IH discount that should exist in Astoria.