Wednesday 29 May 2013

NBN: (Dis)Economies of simplistic charging

The central problem for Telcos since the convergence of communications into All Digital is charging.
How do they differentiate products and charge different rates for identical bits on the same pipe?
In 1988, I first wrote/talked about the problem which in 1991 I phrased to a journalist as "an embarrassment of riches". With huge, cheap pipes available, how could a Telco construct a rate-card for both 32-64Kbps voice and 4Mbps video which didn't either make low-rate services "nearly free" or high-rate services unaffordable?


An artificially constructed rate-table, like expensive long-distance phone calls, will force consumers to find substitutes: users will spend money to access services with costs closer to the underlying cost-of-supply. In the 1990's, Australia became CISCO's global testing ground for Voice-over-IP because of Telstra's monopoly pricing of long-distance calls.

Telcos haven't appreciated this problem, nor found good solutions. The current NBN Co model offers 3 variations:

  • Traffic Class, (TC) or prioritisation and set Quality of Service (QoS).
    • Telephony traffic is high-priority "TC-1" while data is "best-efforts" TC-4, the default.
    • The other classes, TC-2 & TC-3
  • Differential pricing for (Fibre) Access speeds.
    • Users don't want 'speed' in itself, but they can signal their desire for, and the utility of, the service to them by paying more to have the spigot opened wider. NBN Co could just give everyone the highest available access rate, but then it loses product differentiation.
  • Volume charging via wholesale CVC (Connectivity Virtual Circuit) access rate.
    • This is the real revenue raiser and allows wholesalers to use a fraction of the raw access speed of a link at a Point of Interconnect.
  • There are a fourth & fifth variant that could be used for differentiation: multicast and IPv6.
    • It's uncertain what NBN Co will do with these.
The thing about these 3 variables is they are artificial, but of perceived value to customers. Customers get to choose what they pay: they signal with their wallets what's important to them

On top of this, NBN Co have modelled a conservative download growth rate (30%pa, not 50%-63%) and will only increase total charges by 5%/year (by dropping $$/GB by 19%). This per-unit price reduction applies to all traffic and will stimulate usage/demand through Price Elasticity of Demand. NBN Co don't have to be very smart to maximise profits and revenue (Total Revenue Test), something that neither Telstra nor the Coalition do or talk about.

Any producer that can charge many prices for the same product (Price Discrimination) will maximise its revenue because it minimises consumer surplus, when what a product costs a consumer is less than they are willing to pay. Additional price-points reduces the Deadweight Loss of the product, benefiting everyone.

This works at both ends of the market: at the high-end, there are a small number of consumers who will pay a lot for the service, at the low end, you can attract a large number of users at "budget" prices, whom you'll never get at the mid-point prices.

Why this works so well in Telecomms is because almost all the Average Total Cost (Fixed Costs + Variable Costs) are Fixed Costs: it costs milli-cents per hour for a telephone call, it's all but zero variable cost.

Think of current Telstra charges and the proposed Coalition VDSL/FTTN charging: a single price only. They are deliberately foregoing both the upper end (consumer surplus) and foregoing all the low-end consumers.

Not only does a highly differentiated charging model, like NBN Co's FTTP,  increase total revenue, and hence profitability, significantly, a side-effect is raising the ARPU (Avg. Revenue Per User), it allows very cheap entry-level services.

As every supermarket will tell you, "loss leaders" and discounts bring customers into your store and you increase total revenue. That's the same for Telcos: there is a strong positive correlation between per-customer revenue and time-on-service. People get "hooked" on the new services and become habituated to them and increase their willingness to pay.

Telstra and the Coalition's VDSL/FTTN proposal ignore economic fundamentals, behaving like old-school inefficient Monopolies leaving huge amounts "on the table" and not fostering and growing demand.

No comments:

Post a Comment

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