Yesterday the Productivity Commission hosted a seminar at which the Maxim Institute’s Julian Wood presented his ideas on regional development policy.  The Maxim Institute is a policy think-tank, often seen as towards the conservative end of the spectrum, based in Auckland, and over recent months they have published a couple of papers on related issues.  The second of these Taking the Right Risks: Working Together to Revitalise our Regions, was the focus of yesterday’s presentation.

(The seminar ran under Chatham House rules, which means I can’t name the person who championed the success of planning in Auckland, and lamented that we don’t yet have such a plan in Wellington.)

Wood –  a former Department of Labour researcher and policy analyst –  began his presentation with this chart from the first of the two papers.

The first panel highlights the TLAs where population has been static or is estimated to have fallen between 2013 and 2018, and the second panel is the projections in 25 years time (2038 to 2043) from the SNZ subnational population projections.  On those numbers, the national population will still be growing quite a bit, but most TLAs would be seeing flat or falling populations.   These numbers apparently excite a lot of interest in provincial New Zealand –  or at least in the local authorities and local “economic development” agencies.  There is, we are told, much gnashing of teeth.

It isn’t entirely clear why.  In most cases, the places with (projected) flat or falling population 25 years hence, “flat” is more accurate a description than falling, and most of the projected falls are pretty small (eg a couple of per cent over five years).   Taking the full 30 year period, from 2013 to 2043, TLAs that have currently less than 5 per cent of New Zealand’s population are expected to shrink in population over the 30 year horizon.  And given that New Zealand fertility rates are now well below replacement (about 1.81 children per woman), a future of fairly flat or falling populations seems like one that New Zealanders individually are happy to contemplate.  It is, after all, the situation now in much of the advanced world.   There is a handful of TLAs where the population falls projected do look quite stark – eg Kawerau, Opotiki, South Waikato –  and there may be some specific issues for local authorities in those area (especially dealing with central government infrastructure mandates), but it hardly looks like a case for widespread concern.  And it isn’t as if isolated substantial falls in population are a new phenomenon: Taihape’s population now is about half what it was in the 1960s;  Hokitika’s population is not much more than half what it was in the 1860s.

Not only is population decline not a new phenomenon –  even in New Zealand –  but we can, when we look abroad, see that it also isn’t inconsistent with productivity growth and improved material living standards.  Most eastern and central Europe countries have flat or falling populations, and those countries are typically doing rather well economically (Japan’s population is also flat or slightly falling, and South Korea’s is rapidly getting to that point, both countries that continue to rack up productivity growth.)

It also wasn’t clear whether Wood was framing his proposed policy responses around the prospect of falling populations in some of these areas or around some perception of poor economic outcomes in some regions at present.   And the two don’t seem well-aligned.  Thus, if we look at the regional GDP numbers, the regional councils with the lowest average per capita GDPs are Northland and Gisborne.  And yet on the SNZ projections, the population of Northland is expected to be 20 per cent higher in 2043 than it was in 2013, and the population of Gisborne is expected to be 6 per cent higher (although falling a bit by the end of the period).  Whatever the issues in those two regions, population doesn’t seem set to be one of them (unless, arguably, too little outward migration to regions offering better opportunities).

But whatever the precise motivation, –  and some of it simply seems to be the advent of Provincial Growth Fund and a dedicated Minister of Regional Development –  Wood (and Maxim) seem keen on the potential of regional development policy (or “customised regional development pathways” harnessing “the great potential benefits of spatial policy tools”).   I came away from the seminar –  and from reading their paper –  no more convinced than I was by the evangelical spiel offered up by a former MBIE staffer at a Treasury lecture on this stuff last year.

There was, as far as I could see, no analysis at all of what the market failures were, and why then there might be a role for active targeted measures, whether taken by central or local government.  And even though one of his key themes was that locations matter, it was striking that the overwhelming bulk of the hundreds of studies he drew from were of experiences in Europe.  Thus, featuring prominently in the paper was a table described as a checklist of indicators of regional growth and decline, explicitly stated as being drawn from European experience.  Among the items on the “indicators of decline” were “an economic base founded on resource exploitation and/or the primary processing of this exploited resource”.   Not only does that substantially describe New Zealand (and Australia) as a whole, but it also specifically describes Taranaki –  the region with the highest average GDP per capita in New Zealand –  and Western Australia (highest GDP per capita in Australia) and Alberta (highest GDP per capita in Canada).   Marlborough –  without oil or coal – had much the same average GDP per capita as Auckland last year (the sort of relative performance one doesn’t see in any EU country).

The author has been around long enough to have a certain scepticism.  As he notes

Spatial policy introduces “serious risks” like “misallocating resources, creating a dependency culture and favouring rent-seekers over innovators.” Even the Minister of Regional Economic Development has outlined that the new Provincial Growth Fund is a “bloody big risk…”

But in any rational calculus, big risks require a reasonable prospect of big rewards to make the punt worthwhile.   And nothing in the report suggests any real basis for confidence that such rewards are in prospect, no matter how well targeted, designed, and governed the interventions are.   The author knows the pitfalls –  and so he can write sensibly about the need for clear and explicit goals, for a heavy investment in evaluation, for a governance model that blends top-down and bottom-up perspectives, and also about the need to recognise that any experimentation involves allowing for the possibility of individual failures.

But as I listened to him talk, and as I read the paper later, I was still at a loss to know what he really favoured.  There was enthusiastic talk of R&D tax credits –  including by reference to Israel, a country with as poor a productivity performance as our own –  but nothing to indicate why such a measure was particularly suited to regional development (let alone any analysis of why firms don’t find spending on R&D more attractive).    There seemed to be some enthusiasm for immigration, although he knows some of the caveats there.  Weirdly, the concluding paragraph of his entire “smart growth” section is all about labour supply –  which seems mostly to put the cart before the horse, as people will typically be ready to move to where the opportunities are (indeed if the opportunities are in the provinces, more of their own talented young people will stay or come back).  And any policy approach which includes as one of its key items –  as this one does –  requiring local authorities to include even more pages in their long-term planning documents (vapid enough anyway) will struggle to be taken seriously, at least outside government departments.

My own take on these issues is that people who talk about regional development –  whether under the previous government or the current one –  are usually looking in the wrong place.  There seems to be a knee-jerk political need to “do something” and to be seen to do something, even when the action isn’t based on robust analysis specific to New Zealand (and thus the laudable call for good governance, careful targeting etc is mostly a forelorn hope, whistling in the wind).    I searched both Maxim documents and was struck (if not greatly surprised) to find no reference at all to the way in which the real exchange rate –  persistently high even in the face of our relative productivity decline  and itself a reflection of domestic demand pressures –  has reallocated resources away from the regions (generally with quite export-oriented production bases) to Wellington and (in particular) Auckland.   A real exchange rate that was 30 per cent lower  –  and that is the sort of change implied by real interest differentials –  would make a huge difference to the relative prospects of places like Hawkes Bay, Nelson, Otago, Southland, Gisborne, and so on –  orders of magnitude more so than the best of the smart active initiatives Maxim seems to be calling for.   (I was also struck by the fact that although there were numerous references to tax incentives and R&D tax credits, there was nothing at all about the basic rates of business taxation –  if you want more of something, tax it less heavily.)

But as I look at the New Zealand data, I’m also struck by the way there isn’t an overall New Zealand regional story, and even to the extent there is, the differences between the richest parts of the country and the poorest seem no larger (and generally smaller) than those elsewhere.    I had a look through the EU regional data this morning.  GDP per capita in London, for example, is 150 per cent above that in regions like Durham, South Yorkshire, Lincolnshire, and West Wales.   The margins are almost as large between Paris and some of the outer French regions.   Margins of 100 per cent seem pretty common looking across EU countries.   And what of New Zealand?    Northland and Gisborne last year had average GDP per capita of almost 65 per cent of that of Auckland (and 58 per cent of that of Wellington) –  ie Auckland is about 50 per cent higher than them.   And as I noted above, Marlborough had much the same GDP per capita as Auckland –  and there is nowhere in provincial France, UK or Germany with anything like the average GDP per capita of Paris, London, or Hamburg respectively.

Regional development policy, however cleverly designed or governed, isn’t what this country needs –  arguably it never has been (and Maxim has a nice appendix on past failures).  What it needs is hard-headed policy focused on lifting overall economic performance, notably productivity growth, based on a compelling and carefully scrutinised narrative that explains how we got where we are, not just grabbing bits from some generic OECD handbook, from a need to do something/anything.  In practice, that approach –  adopted in New Zealand for a quarter of a century now, at least –  responds to symptoms not causes, and if it sometimes seems to produce benefits (albeit rarely) it is by chance rather than by the inherent merits of the policy approach.  I suspect that a better-designed set of policies in New Zealand would tend to boost the regions relative to Auckland and Wellington, but that wouldn’t (and shouldn’t) be the goal: the goal should be lifting opportunities for better material living standards for all New Zealanders, and enabling New Zealanders to move to take advantage of those opportunities wherever they are to be found.