[LINK] NBN to cost 24 times South Korea's faster network, says research body

[Richard Chirgwin]

On 10/02/11 6:15 PM, Tom Worthington wrote:
> Bernard Robertson-Dunn wrote:
>> NBN to cost 24 times South Korea's faster network, says research body
>> ...http://www.theaustralian.com.au/news/nbn-to-cost-24-times-s-koreas/story-e6frg6n6-1226002952747
> The report is ‘Full speed ahead: The government broadband index Q1 2011'
> from the Economist Intelligence Unit (the research unit for The
> Economist magazine).  The full report costs US$2,950 so I just read the
> free summary:
> <http://www.eiu.com/public/topical_report.aspx?campaignid=broadband2011>.
>
> The report criticises the Australian government for "... spending a
> colossal 7.58% of annual government budget revenues on its National
> Broadband Network.

Which is, as I have written for The Reg, just wrong. The $27 bn is 
compared to ONE YEAR of government revenue. It's deliberately 
misleading, designed to get media to over-state the size of the spend - 
and in that sense it worked.

Also, the South Korean government's *current* spend is on a network 
upgrade, not a new network.

To call this report a pile of steaming dung is flattering: you can at 
least use dung to fertilise flowers.

> In South Korea, by comparison, the government is
> spending less than 1% of annual budget revenues to realise its broadband
> goals, achieving targets by encouraging the private sector to invest in
> the country's broadband future. ...".
>
>   From the summary of the report it appears that the EIU's researchers
> developed an index to rank national broadband schemes on speed, coverage
> and rate of rollout.
But: the index is "opaque" - "secret sauce" analysis where verifiable 
facts such as the "7.58% figure" are plain wrong. If the stuff we *can* 
see has errors, what can we say about the stuff we can't see?
> More controversially they also include in the index
> the "... most appropriate regulations for realising targets and
> fostering a competitive broadband market". This makes the assumption
> that a market is possible and appropriate.
>
> On these measures Australia ranks 9th out of 16 countries, just under
> Denmark and above New Zealand and the USA. South Korea ranks top and
> Greece bottom.
>
> Any deployment of broadband across Australia which attempts to achieve
> equity will be difficult and expensive. Australia has large cities where
> deployment is easy and then sparely populated areas where there is no
> technology which can provide cost effective deployment. If everyone in
> Australia was prepared to move to Sydney, then broadband could be
> provided at comparable speeds and costs to countries like Singapore. ;-)
I agree. But the Economist doesn't: the analyst described geographical 
considerations to me as "red herrings" (I unpick this little load of 
nonsense as well: urban Oz is 1/4 the density of Korea in total).
> The cost of the NBN (about $43B over 8 years) should be seen in
> perspective with other public expenditure. As an example Australian
> public expenditure on education each year is 4.5% of GDP (from
> "Education in Australia", OECD, 2008), or about $56B (based on the OECD
> States Extract estimate of Australian GDP of $1,253,121.0 for 2009). If
> the NBN achieved a 10% saving in the cost of education, this would pay
> the entire capital cost of the network.
Note my little rant above. Education is 4.5% of GDP, or about 8% of 
government revenue (rough). The NBN is, annualised, less than 1% of 
government revenue. The Economist deliberately picked a measure that 
overstated this.

RC