Friday, 27 September 2013

Copper cheaper than Fibre? Only if you "cook the books" radically!

Comparing costs is only half the picture, profits are what drives business. Turnbull has omitted anything to do with revenue, charges or profits from everything he's released.

Using the correct Discount rate of 4.6%, nearly all sets of FTTP/FTTN costs put Fibre ahead, or at the same cost as Copper, even with the highly unrealistic costs chosen by Turnbull: Fibre costs 100% higher than actual and leaving out all Telstra payments for Copper.

If you include payments to Telstra, Copper is always more expensive than Fibre.

If you use the actual Fibre costs of NBN Co, then at worst the difference is negligible ($17/year) or the Copper FTTN is from 50%-60% more expensive than Fibre over 15 years.

Comparing the Costs of Fibre (FTTP) and Copper (FTTN) networks

Turnbull hasn't just viciously and aggressively attacked anyone and everyone, he was very publicly accused of lying, a claim he's not refuted in nearly six months. When does that become an admission?

It's a given that any politician will distort, select and manipulate figures to support their cause. Turnbull has done this in spades with his NBN Node Plan. This piece will focus on just one point where we can pin him down.

Previous blog posts here have included why Fibre makes a Profit whilst Copper cannot,  a discussion of real costs to taxpayers and critical policy points, like charges, are still unaddressed by the Coalition.

The first outrageous cheat by Turnbull was to not pay Telstra anything for their copper! Turnbull estimated this at $1500/line, considerably more than the $900/port he allows for FTTN.

The next cheat was using fabricated FTTP install costs of $3600/service when very reliable costs of ~$2350/service had been released by NBN Co in good time for him to add to his documents.

The last is more subtle, Turnbull used a usurious rate-of-return, or "Discount Rate", such as Telstra or Optus might use. The "Discount Rate" is what you might otherwise earn from your money. For Government, it's zero, they don't invest or earn profits. NBN Co has a very modest target of ~7% placed on it by its shareholder, as it's prime concern is not lining the pockets of investors, but providing affordable services to taxpayers.

Another wrinkle is that Turnbull omitted the effects of inflation.The "Discount Rate" needs to be further reduced by the inflation rate for these figures to be meaningful.

Lastly, the most blatant and colossal distortion of reality is to never, ever speak of revenues, charges or profits! Any business person knows that there are two sides to the ledger, and focusing on just one side, either revenues or expenses, leads inevitably to ruin. Businesses survive solely by making a profit. A more subtle point that has killed many a small businesses, and one that Turnbull trades on, is a positive cashflow (cash surplus) which is not a profit. Business have to cover non-operating expenses, like Depreciation, and payback their loans, something you can't do with positive cashflow and no profit, as Turnbull has as his target for the NBN.

In this "example", as elsewhere, Turnbull focuses solely on costs, where he can manufacture an advantage, pretending that is all that matters to a business. He then proceeds to choose outrageous, unrealistic and unjustified assumptions to create the outcomes he wants. Very easy to do if you don't submit your full figures to independent scrutiny and verification.

Even the manufactured cost advantage of Copper over Fibre fails if you use real, Australian data, not fictions dreamt up in the backrooms of Turnbull's offices.

Source: Background Papers, The Coalition’s Plan for Fast Broadband and an Affordable NBN.

Coalition Net Present Value example of FTTN/FTTP expenses, page 14.

These figures are also available as a downloadable spreadsheet.

The first line, in blue, is the Coalition example. Values marked in red are where Fibre is cheaper than Copper. You'll notice for the majority of cases, Fibre is cheaper than Copper, even when unrealistic costs are used.

Notes on reading table:
The first seven columns are the parameters input to the two NPV values calculated in the Coalition example to compare which approach is cheaper over the period.

The two NPV calculations are the cost of building FTTP versus building FTTN initially, then upgrading later to FTTP, with half of the FTTN cost wasted, as per the Coalition "50% CapEx Reuse".

Disc- ount Rate
Year of FTTP upgrade
Years of NPV
8% 3 3 $900 $90 $3600 $60 -$3755 -$3633 $122

FTTN to FTTP upgrade cost

The Coalition Background Papers use a FTTP CapEx cost of $3600 and $60/year OpEx, with FTTN costs of CapEx: $900 and OpEx $90.

The real-life difference is actually much worse for FTTN than is detailed here:
  • For FTTP, payments are only made to Telstra when active services are transferred to NBN Co.
  • While for FTTN, we can assume every copper pair cut, active or not, will require payment to Telstra.
    • For FTTN, Turnbull estimates a markedly lower take-up of 75%, increasing the Telstra payments per active copper service by 10%-25%.
    • In the Telstra customer access network, there are many unused copper pairs . This increases Telstra payments by another 10%-25%.
Assumptions in the Coalition "Net Present Cost" model presented are:
  • There will be an upgrade from FTTN to FTTP at some future date.
  • The example given uses an unrealistic replacement time of 3 years. This is well before the $900 investment can be recovered, somewhere between 10 and 20 years.
  • Half the FTTN investment is wasted, $450 is deliberately wasted in the Coalition Plan.
  • An unrealistic 8% discount rate is used. This is a figure that Telstra and Optus may obtain as they expect far higher returns for their shareholders. NBN Co, because it is not driven by profit-hungry shareholders, seeks a very modest shareholder return of 7.1% in comparison.
  1. -($3,600 - $450) - $90 =  -$3,240 [Coalition model figures]
  2. -($2,350 - $450) - $90 = -$1,990  [Actual costs for NBN Co]

1. Telstra "PSAA" Payments estimated at $1500 by Coalition and used in their modelling.
a. FTTN $900 excludes lead-in cost
b. FTTP $3600 includes lead-in cost

Comparing FTTN & FTTP CapEx costs on same-same basis:
c. $900 v $2100 without lead-in
d. $2400 v $3600 with lead-in

2. Real FTTP CapEx Costs
From  NBN Co report, 19-Apr-2013.
a. $1100 lead-in
b. $1100-$1400 Fibre ($1250)

3. FTTP CapEx (Capital Expenditure)
a. $3600 = $2500 + $1100 leadin
b. $2350 = $1250 + $1100 leadin

4. FTTP OpEx to CapEx ratio. Lead-in costs excluded.
a. $60 claimed by Coalition for $3600 FTTP
b. For current FTTP cost of $2350, same ratio is $30. ($60 * ($1250÷$2500)

5. Discount Rate (the earnings/interest rate the money could otherwise make in the business)
a. 8% ignores inflation, long-term rate of 2.5%
b. 4.6% is NBN Co IRR (Internal Rate of Return), adjusted for inflation
 4.6% = (7.1% - 2.5%)


  1. Steve I think your column headers (FTTP/FTTN) in the table are the wrong way around.

    With regards to Delimiter's claim that Turnbull was lying, let's borrow from Abbott: "The claim has been out there six months and nobody has been able to criticise it, it is bulletproof".

    1. Dean,
      Thanks for the feedback. I've reworked the column headers and put a note in there on how to read it.



Note: only a member of this blog may post a comment.