Wednesday 17 August 2011

A number I don't believe

Another plausible BERL estimate:
Given the make-up of the Auckland and Bay of Plenty regional economies, we illustrate the overall impact at a regional economy level of fast broadband from early roll-out and uptake. We estimate the productivity benefits of fast broadband could lift GDP by 7-9 percent above its business-as-usual level by 2025 if early roll-out, adoption and uptake is achieved. 
BERL's report on the benefits of broadband for Auckland is attached to the minutes of an Auckland Council Meeting.

I haven't the time to put into fisking this one. But a few bits of interest:

  • They reckon the benefits to primary industry in the Bay of Plenty region is greater: "the estimated total regional benefit of broadband applications to the dairy production industry in the Bay of Plenty region is about $71 million. This is equivalent to about 7 percent of the export value of dairy product in the Bay of Plenty." Dairy can be made more productive by moving from zero internet to half-decent internet. Just imagine the benefits to Fonterra in being able to solve the Salesman problem if they have real-time access to milk volumes in the tanks at all the farms in the area. But I can't see how that's substantially better achieved at 10 mbps than at 500 kpbs.
  • They figure the Auckland region's GDP, by 2025, is 1.8 billion dollars higher if broadband rolls out in 2012 instead of 2015. For NZ as a whole, we get a $4.7b boost. That's about a thousand dollars per capita. I can't believe that there's any difference a decade on that's discernible from noise.
  • I can't help but suspect that there's rather a lot packed into the last two bits of "if early roll-out, adoption and uptake is achieved."
  • They spend a fair bit of time highlighting the economic benefits of future growth in a bunch of industries where broadband hardly seems the binding constraint. Efficiency benefits of Kiwifruit farmers being able to file reports online with Zespri rather than paper forms hardly requires really fast broadband. I'm also not sure that broadband access is a bigger constraint for aquaculture than are regional consenting issues.
If the Auckland and Bay of Plenty Councils believe the results of the study they commissioned, I really hope that they go ahead and make the broadband investment using their own ratepayers' money rather than using the report to pressure Key into providing Auckland with what's mostly a private good.

HT: Henry Ergas, who's done great work debunking similar claims about Australia's national broadband network.


  1. Don't you mean, 'Another *implausible* BERL estimate' in the first line (it reads 'plausible' at the moment. If its sarcastic, it doesn't come though!).


  2. For some more credible assessments of the benefits of fast broadband, see
    "Feeding a need for speed or Funding a Fibre 'Arms Race'?"

    Superfast: Is It Really Worth a Subsidy?

    Note that versions of both of these papers addressing the productivity benefits of fast broadband have been published in the peer-reviewed academic literature on broadband policy. I am unaware of any publications by the authors of the BERL report in this body of work.

  3. Hmm. I'd thought the headline indicated my general view of the figure.

  4. A very quick skim - who would have thought Waiuku deserved singling out for High-Speed demand, apparently more so than Albany... So how did they come up estimates of demand for high speed broadband?

    "To examine the potential demand for broadband, we examine the pattern and geographic density of industry and residences in the Bay of Plenty and Auckland Regions...

    ...Given the pace of progress in the IT sector, it is not useful to define broadband as a particular speed or capacity, except within a short timeframe. For example, the government’s current definition of broadband as 256 kbps, and ultrafast broad as 100/50 mbps download/upload. Our analysis, below, provides a more flexible framework. We expect that the market will deliver businesses and households products that continually improve (and increase the technical requirements), but that the demand for these products will continue to differ by their relative speeds. For example, the need for speed is likely to be lower for households than for knowledge-based businesses, regardless of the definition of “low” and “high” speed.
    We segment demand into that for low (DSL), medium (fast broadband) or high (ultrafast) speed, although the rates are likely to be higher for businesses than for residential users. What these speeds are in terms of bps can change over time, as the technology improves, and applications are invented and are picked up by end-users."

    Very robust analysis then. (Sorry for the long quote - I am sure there are many more goodies though).

  5. @eric , oh yes the headline did indicate your overall sentiment but i was just a little confused when I read the next line. That must've been my literalist, pedantic mode. Just thought I'd bring it to your attention in case it was a typo.


  6. The 5% of businesses in the Auckland CBD who BERL estimate would benefit from fast broadband probably already have it - and have had it since it was first available - in the form of a leased line from a telco (Telecom, TelstraClear, CityLink, Vector - take your pick).

  7. Does BERL stand for Bad Economic Research Ltd?