The Provincial Growth Fund is too good an idea to be botched and discredited by a bull-at-a-gate approach.
Or as Maxim Institute researcher Julian Wood puts it in a report on the scheme one year in, it's like the road safety campaign says: the faster you go, the bigger the mess.
There is a respectable case for what in the jargon is called spatial policy, as distinct from the "horizontal" approach of policy which treats all places and all people the same and leaves it to market forces to determine what growth happens where.
The risk for New Zealand in the hands-off, one-size-fits-all approach is that we end up with regions littered with zombie towns and local authorities struggling to fund services while Auckland becomes a kind of sump into which the country's population and capital drain.
The alternative risk with localised intervention in the name of regional development is that some public money will be wasted and the scheme will be seen as a shambolic boondoggle or slush fund.
Some of the criticism of the Provincial Growth Fund so far is absurdly premature. "Where are the jobs?" ask the critics. "There are supposed to be jobs!"
But trawling through the current (as of last month) status of the 163 projects announced so far on the provincial development unit's website reveals that only $25 million of the $3 billion fund has been paid out so far.
Those numbers indicate a pattern of announcing a project, to some fanfare, then getting down to negotiating the terms of an actual agreement with the relevant counterparty, and the fact that those agreements typically have conditionality — milestones which have to be passed before the first or subsequent funding is released.
The relatively modest outlay so far, compared with the sums announced, partly reflects the fact that some forestry-related projects got ahead of the ability of nurseries to deliver, or hillsides to receive, the seedlings required.
But it also indicates that most of the 163 projects so far announced are pretty small potatoes.
The controversial (but still un-finalised, as of February) plan for a soft loan of $9.9m to Westland Milk Products is not the norm.
More typical is the $400,000 contribution to a feasibility study into establishing a hatchery and nursery that might kick-start an aquaculture industry in Southland. That project has the backing of the local authorities and Ngāi Tahu. There is even a little bit of Sanford's money on the table.
It looks like the kind of approach the Wood paper advocates: regional consensus about where some potential comparative advantage lies that requires a bit of monetary assistance from central government to get to base camp, so to speak.
But having looked at the regional action plans the previous Government required, Wood concludes that many require a refresh and a less siloed, more collaborative approach if they are to be the basis of successful applications for funding.
Inevitably perhaps, the fund's approach to regional development involves institutionalised favouritism and rough justice.
Not only are the main centres excluded, but some parts of rural New Zealand have been prioritised as "surge" regions: Northland, Bay of Plenty, Tairāwhiti/East Coast, Hawke's Bay, Manawatū-Whanganui and the West Coast.
In addition, there is a preference for certain sectors: forestry, tourism, and food and beverage.
This approach is too blunt and risky, Wood argues. "Already it would appear some of the sector-based initiatives seem to be at very high risk of failure due to a lack of due diligence," he says.
"Instead all investment activity should prioritise solutions aimed at the discovery of something new, no matter what the sector or industry." If a focus on whole industries is too broad, and on individual companies too narrow, what middle ground does that leave? What would "smart specialisation" look like?
Supporting innovative activities that could create knowledge spillovers, Wood suggests, much as expertise in composite materials has spilled over from boat-building into rocketry and, less successfully, wind turbines.
The report is critical of a lack of transparency and evidence of ongoing evaluation strategy around more significant (say, $10m-plus) initiatives.
Ongoing monitoring and evaluation is important both to identify and learn from mistakes, and to reassure the public that poorly-performing initiatives can be improved or, if need be, dropped and the money used more effectively elsewhere.
So the Maxim Institute report recommends requiring every PGF initiative over $10m to include and fund an evaluation and monitoring programme.
It also challenges the fund's exclusive focus on growth.
"Because growth everywhere is not a realistic or achievable goal, the PGF's focus on fostering economic growth in the regions is a missed opportunity to enable some communities facing decline to consider how they might become smaller and better," Wood writes.
"Just as our growing centres need transitional funding to overcome congestion, density, and pollution issues, places in decline need transitional funding to enable them to become smaller, better, healthier, and more sustainable."
This might mean putting some things which are currently out of the fund's scope, like pensioner housing and health facilities, on the funding table.
Wood notes approvingly that Southland's regional action plan includes creating an "age-friendly community in Gore based on the concept developed for Levin in Horowhenua."
Overall, the Maxim Institute report is clearly right to point to the opportunity cost of spending $3b on regional development — all the other useful things that money could fund — and to the risk that avoidable failure of this investment would cast serious doubts on the use of spatial policies in the future, so that we default to the policies of the past which have left large tracts of the country languishing and left behind.
Better, surely, to slow down and make some cautious course correction now.