Stats included long-term graphs which put the latest numbers in some perspective, but I've gone back a bit further again (as far as the data series go on Stats' Infoshare database). Here's what the story looks like.
The 'actual' numbers in the graph are an excitable series, mainly because of chunky apartment block consents turning up in some months but not others. They're volatile enough to turn the seasonally adjusted and trend series into best stabs at what's going on behind the noise, rather than definitive sightings of the underlying reality, so you can't be entirely sure they're on the right track. But after some wavering around the start of this year the trend series (also shown in the graph) now looks to be definitely heading in the right direction.
I've been trying to guess - no stronger - whether the numbers tell us anything about whether we can up the pace any further or whether we've hit capacity constraints.
On the downside, over the past 25 years we've only briefly been able to keep up 1,000+ periods of monthly consenting. Something or other has always knocked it back again. Either we've hit some sort of capacity constraint, or the business cycle has put the kibosh on it, with the GFC in particular decimating activity. The latest ANZ business survey wasn't much fun in that regard: only one month, the election and all that, but I particularly didn't like the finding that "A net 26% of businesses expect it to be tougher to get credit". That's bad news for housing developers.
On the plus side, the Auckland labour force has grown quite a bit, and there are more people around these days with building trades skills. Here's what the Household Labour Force Survey shows for people employed in construction in Auckland since 2009 (as far back as Infoshare went, and I haven't time to fossick the Census if it's got longer/better figures).
There wasn't a lot of movement till 2014, but since then employment has lifted from around 50,000 to around 80,000. Recent immigration probably plays some part in this (and maybe people shifting back from Canterbury building sites): where we live on Auckland's North Shore, a good deal of the in-fill development is by Asian developers with Asian crews, including the one literally across the road.
The unknown unknown is probably the impact of the planning process. I've no idea whether the number of consent-approvers is keeping pace with consent applications; I can surmise that (certainly over a time-scale of 25 years) the regulatory requirements to get a consent have risen; I would bet that land-use constraints have got a lot tighter. Net net - who knows, but I'd lean towards a view that planning is at a minimum no less a constraint than previously.
Overall, I'm glad to see the recent pick-up in consents to over 1,000 a month, and it might be a bit curmudgeonly to add "at long bleeding last": it's here, and it's something. But I'm not exactly jumping for joy yet. At this pace (12,000 dwelling units a year) we're still a little adrift of the numbers (14,000-ish?) that people think we need to meet new demand, let along eat into the existing shortfall (20,000-ish?). And while I'd be pleased to be proved wrong this time round, our recent track record suggests we can't keep delivering, or are blown off course from delivering, before we get the job fully done.