Beautiful Armidale in northern NSW within Barnaby Joyce's electorate
Decentralisation; beautiful Armidale in the federal Minister for Agriculture’s electorate will be the lucky recipient of an agricultural agency relocated from Canberra

There’s lots of talk at the moment about decentralisation and regional population growth. It’s driven in part by the perception that Australia’s capital cities can’t cope with further growth and in part by enthusiasm for exciting projects like High Speed Rail (e.g. see Is decentralisation regional sprawl by another name? and Are regional dormitories the way to grow our cities?).

The discussion is almost entirely about creating regional dormitory communities around capital cities, but earlier this year Deputy Prime Minister and Minister for Agriculture, Barnaby Joyce, announced his intention to push for real decentralisation by setting up Commonwealth agricultural agencies in regional centres. He recently confirmed three Rural Research and Development Corporations were to establish regional offices outside Canberra. Last week he confirmed the Australian Pesticides and Veterinary Medicines Authority (APVMA), along with all 175 jobs, would be relocated to Armidale in Mr Joyce’s New England electorate as early as March.

Now Victorian Premier Daniel Andrews looks like he’s set to announce the relocation of an as-yet unnamed government department from Melbourne to a regional centre, probably Ballarat (or could it be Traralgon?). Mr Andrews would be well advised to hold his fire while he takes a good hard look at what Barnaby Joyce is up to with the APVMA.

While there might be political advantage in country areas from shifting government functions out of capital cities, a cost-benefit and risk analysis of the planned relocation of the APVMA to Armidale prepared by Ernst and Young indicates the non-political benefits are at best minor and might possibly be negative. It concludes that:

  • The net benefit to the nation is “modest”. The consultants say the “strategic and operational benefits of having the APVMA operate out of Armidale appear to be limited”.
  • While Armidale’s 25,000 residents will be better off, the town’s gains will be cancelled by Canberra’s (somewhat higher) loss. Canberra will lose 365 direct and indirect jobs and $157 million p.a. in economic activity. On the other hand, Armidale will gain 350 jobs and $119 million p.a. in economic activity.
  • Only 15% of APVMA’s staff said they’d be willing to move to Armidale; hence there’d be high levels of personal disruption, as well as significant redundancy costs for existing staff and recruitments costs for new staff.
  • Industry groups CropLife Australia and Animal Medicines Australia oppose the move.
  • The consultants don’t say the APVMA couldn’t operate out of Armidale, but they say relocation would create significant risks in terms of the organisation’s ability to:
    • Attract and retain suitably skilled technical and management staff.
    • Deliver efficiently on its core functions – particularly registration of new agricultural and veterinary chemical products – during the transition period.
    • Consult adequately with stakeholders across the country given Armidale’s lower accessibility to the rest of the country compared to Canberra’s.

One thing we don’t know is whether or not Armidale is the best possible non-Canberra location for the efficient discharge of APVMA’s functions, or even if it’s the best regional location.

However there’s no doubt the planned move will provide a significant economic boost for Armidale. Ernst and Young note that although Canberra’s economic loss is higher in absolute terms than Armidale’s gain, the change will have a greater impact in proportional terms on the much smaller economy of Armidale. That’s terrific for Mr Joyce’s electors and for his standing in his electorate, but it doesn’t come free.

Ernst and Young didn’t examine savings in infrastructure costs in Canberra arising from the reduction in population, or additional infrastructure costs in Armidale resulting from the increase in population. If Mr Andrews wants to frame similar action around the virtues of decentralisation, he should start by showing whether savings in infrastructure costs justify relocating Victorian government functions from Melbourne to provincial centres. I’m not confident there would be significant savings, or that they’d exceed the economic costs.

Ultimately though the big problem with these sorts of exercises as a way of driving decentralisation on any sort of scale is that centralised government activities are knowledge-intensive. There simply aren’t that many government functions that can be separated from their Minister, from the rest of government, from a large slice of stakeholders, and from access to a large and specialised workforce. Much of the sort of routine work done by government that’s relatively location-agnostic is already outsourced in various ways to private providers.