People sometimes worry that investors (or foreigners) are buying up properties and leaving them empty, speculating on capital gains instead – but if this is happening at all, it seems to be on a very minor scale. In a post last year, I looked at unoccupied homes in Auckland and other NZ cities, using census data. Most of them, including Auckland, had similar percentages of unoccupied homes – roughly 5% to 8%.

In this post, I’ve dug deeper, looking at the unoccupied homes figures for each Auckland suburb. I’ve also put together an interactive map showing the percentage of unoccupied homes for each area:

Across the Auckland region, there were 33,201 unoccupied dwellings on census night 2013. This sounds like a lot, but as per my earlier post, the fraction of homes that are vacant is pretty similar to other NZ cities, and to the other ‘snapshots’ taken in the 2001 and 2006 censuses.

Unsurprisingly, there are plenty of vacant homes on some of our offshore islands – Kawau, Great Barrier, Waiheke, with around 2,500 between them. Most of these, of course, are used as baches.

Most of the other ‘ghost areas’ are also bach hotspots – the highest percentage of vacant homes are in places like Leigh, Omaha, Snells Beach, Waiwera and Point Wells. Around 2,700 vacant homes are in these northern holiday/ lifestyle areas.

Still, most vacant homes in the Auckland region are in the urban area. This is what we’d expect given that’s where most of Auckland’s 507,000 dwellings are.

Within the city, Hobsonville stands out as an area with a lot of empty homes, presumably because many of the homes have just been built and not yet occupied. In fact, the Hobsonville East area unit had 363 occupied dwellings, 120 unoccupied, and another 18 under construction. That construction figure seems implausibly low, so my guess is that many of the “unoccupied” homes hadn’t actually been completed and certified yet, even if they looked habitable.

Auckland Central has 2,200 unoccupied dwellings, according to the 2013 census. It’s quite common for CBDs to have lower occupancy rates, due to several different factors:

  1. With a ready market of people wanting to stay in the CBD for a night, a week or a month, apartments are more likely to be rented out as short-term accommodation than a house in the suburbs
  2. Some people can afford to keep a CBD apartment for weekday or occasional use, and spend the rest of the time in a house elsewhere in Auckland, or somewhere else in New Zealand, or even overseas.
  3. Although legally everyone is supposed to fill out a census form, not everyone does. The ‘response rate’ is likely to be lower in the CBD (language barriers, privacy desires, age cohort factors etc), plus it’s harder for the census collectors to tell whether an apartment is occupied or not, compared to a house. Some apartments will have been incorrectly tagged as ‘unoccupied’.

In my previous post, I quoted North & South magazine, who give “Mt Albert and Mt Eden, Grey Lynn and Herne Bay” as examples of suburbs where ‘ghost houses’ are apparently common. But the census data doesn’t show that at all. Most of these suburbs are spectacularly normal. They have vacancy rates at very typical levels, in line with the averages for Auckland and other NZ cities. Herne Bay is slightly high at 9.1%, but it was 9.3% in 2006 and 7.9% in 2001. And those ‘higher’ rates seem to be typical for the wealthiest suburbs – Remuera is similar.

I initially thought that wealthier suburbs might be more likely to have unoccupied homes – since people living there might have more mobility, travelling overseas for example. However, there’s no evidence of this in the data. I’ve done a few really basic regressions, all of which fail to show a relationship between incomes and the percentage of unoccupied homes. Here’s one example:

Vacant homes vs income

Finally, I’m open to the idea that the number of empty homes has crept up since the 2013 census – that this is becoming an emerging issue – but it doesn’t seem likely. We’re not building enough homes, and there’s plenty of demand from people looking for places for to live (giving rise to issues like overcrowding). If anything, there’s more incentive to rent out an empty house now than there was in 2013. My two cents is that we need to put our energies elsewhere if we want to get the housing market into better shape.

New Zealand isn’t the only place to be concerned about empty homes. This article from Sydney uses 2011 census data for the city, similar to what I’ve done here. And this article from Melbourne uses a methodology I quite like – they’ve estimated the number of empty homes, based on homes with abnormally low water usage.

Share this


  1. I wonder if you examined the relationship between household wealth, or net worth, (as opposed to household income) and unoccupied houses whether there would be a correlation? as this would capture retirees and other financially independent types who are not tied down to one location, as breadwinners tend to be.

    1. Hi Deborah, it’s possible, but we’ll never know as NZ doesn’t really have good data on wealth. There are other socioeconomic variables I could use – percentage of households who own their own home, or median house prices – but given that I’m not getting any correlation with income, I imagine that the correlations to other variables wouldn’t be very strong anyway.

  2. A local agent told me it is happening in our area now as people buy to resell. They dont want the hassle of tenants to get rid of. Some of the houses are being renovated but some are just sitting waiting. It is inevitable as the returns are now all about capital gain and not the rental yield.

    1. In that case, they’re likely to get pinged by the two-year “bright line” rule, so they’ll need to hang onto the house for quite a while if they are just playing the capital gain. Definitely unfortunate if that is happening, but surely most investors (the word ‘speculators’ is probably appropriate here) would prefer to get a tenant in on a fixed-term tenancy.
      It gets plenty of coverage in its own right, but never in the same articles as the ones which look at ‘ghost homes’ – Auckland has a shortage of homes for rent, so rents are rising ( That means investors have stronger incentives to rent their properties out.

      1. I think they expect to pay tax on the gain and yes they are straight out speculators. The worry is they can sell quickly so when the market turns they will be first out and the price plunge will be dramatic. Doesn’t worry me, we unloaded two Auckland rentals and now only have the Auckland home we live in and a rental in Wellington where the yield is still very good. Much easier to sleep not having to guess the peak.

        1. Dob them in to the IRD. If they are buying for capital gains then they have to pay tax whether the keep in 20 years or 20 minutes.

    1. That Herald article left me with steam coming out of my ears.

      John was the only independent analyst consulted on the article. Instead of listening to him the Herald decided to totally ignore his advice and instead decided to be a mouthpiece for the Labour Minister, screaming an alarmist headline across it’s front page. It was a disgusting piece of journalism that has no place in a democratic society. Though it’s nothing the Herald hasn’t done before and won’t do again as the Herald continues it’s series of politically bias attacks against the current government.

      Thanks for the article John and thanks for speaking some sanity into the debate.

      1. Yes, the fact that they went looking to Sydney and Melbourne for a comparison is telling. If they were to compare with other NZ cities and previous censuses (or censii?), which would be logical and as John did in his post last year they wouldn’t get the answer they were looking for.

      2. Bollocks! The Herald is Nationals biggest cheerleader. It has only been lately that they have seen the publics annoyance at the out of control housing market that they have been headline grabbing on the issue. If the media had done their job properly and held the government to account over failed policies then National would have likely lost the last election.

        1. National could have lost the last election had there been an actual strong alternative something that is still lacking.

        2. I’d disagree with both of your statements. In my opinion the Herald publishes roughly 90% left wing viewpoints. I have trouble thinking of failed National policy, the flag referendum wasn’t really a failure as such, and it is also labour party policy. They haven’t really done that much. I think the lack of a cohesive, comprehensive alterative is the real reason National is still in government.

          1. I reckon I hear about 50/50 comments that the Herald is biased to the left or right, so it must be doing a reasonably decent job then. It has definitely got on the bandwagon with housing prices though, but I think that is more to do with easy headlines than anything else, I have noticed the number of articles talking about the risk of a crash lately, not front page yet though.

  3. Easy answer is to ask around honest real estate agents. Central city ones will tell you many hones are empty as the owners ( investors) are effectively making 100k plus per year in capital gains without the hassle of tenants. Ask around. It is happening more

      1. No not even that; he’s claiming to know what they would say, and with figures! Nonsense on nonsense.

        Anyway as an owner of an inner city investment apartment I can tell you without our tenants we would be going backwards with body corporate fees and rates real fast. We make sure any time between tenants as short as possible. Luckily the place rents it self, never have had to advertise, so desirable is quality inner city living….

      2. ‘Actual data’. My reading of the data is that is 3 years out of date. A considerable amount has happened in the market in that time. To me one of the major problems is that there is no up to date statistics on occupancy (or many other things for that matter). This is typical with the current govt who suppress any statistics gathering on things that do not suit so they can say ‘We don’t know’ whenever things look bad.

        1. There are costs involved with gathering statistics, are you advocating for undertaking a census every 6 to 12 months? This is a classic example of a ‘flavour of the month’ issue, if we gathered stats on every issue like that the census would be a small book to fill in. Even if it is an issue, it is likely to dissapear as it is completely reliant on ongoing capital gains, and the market is showing a lot of signs of being at or near the top

          1. No doubt statistics gathering costs.
            I recall a brief news item a year or so ago describing that the Statistics Department requested funding for gathering figures relating to overseas purchase and ownership of NZ residential property. The govt turned down the funding enabling those in power the ability to down play the ownership profiles with repeated ‘We don’t know’ every time it was raised. They had made certain of that!
            Very convenient.
            Plenty of new flag diversion funding though.

  4. Rather than the number of unoccupied houses and apartments, a more relevant measure of the shortage of affordable homes would have to be the increasing number of houses used for business purposes ONLY , especially in the suburbs – i.e. not situations where the business owner’s family also live upstairs or at the back of the premises. Presumably there are financial reasons such as tax-free capital gains and the ability to deduct expenses relating to the part of the building used for business.

  5. 2013 was a while ago now. Since that time the buy up of Auckland property by Chinese (in particular) property speculators reached a peak meaning that those numbers are hopelessly out of date. Estimates bandied about suggest numbers of vacant houses are now 2-3x that 2013 number… the same as what has happened in Vancouver where they now have ghost suburbs with hardly anyone living in them (which in turn means no more of the lovely cafes and shops loved by this blog… but on a positive note it means less cars since there is nobody around to drive them!).

    *Edit… less cars in the particular areas but more overall as people are forced to live further from the city centre where they can find a semi-affordable house. These areas tend to have less access to PT so result in more cars overall across the whole city.

    1. Bruce this is just exaggerated stories; where are these ghost suburbs in Auckland? Do you mean the city centre? It’s fuller than ever. Sure brand new dwellings are changing hands on completeltion as people who bought off the plans take a quick gain in a rising market, but they are soon taken up by occupiers or holding landlords. This is observable in higher volume and poorer design new builds like SugarTree. Better apartments like The Dylan are full of owner occupiers.

      1. In Vancouver Patrick… as per the post. But there is a good chance of this happening here too if foreign buyers keep snapping up properties and not renting them out.

    2. I think a lot has changed in my area since 2013. The local school now has two ESOL teachers working with mostly Chinese kids where just a few years ago they had 1 part time to help a few Korean kids. They are currently building classrooms to handle a boom.
      Also we have noticed in the last 2 years that there are now a lot of cars parked in the street where there used to be none. Most of the houses opposite us have changed from being families to younger NZers where one or two have bought and have filled the house with flatmates presumably to help with the mortage. Down the end of the road in the rich bit there are at least three houses sitting empty. None of this proves a trend, it just indicates some changes- maybe we just became a suburb of Auckland rather than a village.

  6. For brand new property speculation, resell value is affected if there was a tenant.

    Adding that oversee investor would rather avoid the taxable risk of finding tenant (due to do taxable income, which involves IRD).
    It makes them easier to hide their tax, especially if they purchase the property from 100% cash, and before the rule requiring IRD number and bank account.

  7. The areas of high unoccupied houses are in the case of Hobsonville not yet occupied as they have just been or are still being built. Wiri is a question mark as it is not a big residential area, the rest will have a reasonable percentage of holiday homes so is to be expected.

  8. between people on holiday, and buying & selling, the vacant house rates aren‘t newsworthy.
    given the amount of money real estate agent put into the likes of the herald, its not surprising we‘re forever being bombarded with beat up stories about housing

  9. Quote from Gareth Morgan :’Look at me, I own six houses. I don’t have tenants; they just make carpets dirty’.
    To say this is not happening is just ridiculous.

    1. No one is saying that it isn’t happening, they are saying that the evidence clearly shows that it isn’t happening at a significantly higher rate than anywhere else.

      1. Come on its unfair to confuse people with facts and figures when they have already made up their mind based on what their mate told them at the pub.

  10. Do the figures include the built but uninhabitable apartments such as the ones in Richmond road, Symonds and Beresford Streets? These are now being repaired /replaced but all contribute to the housing shortage.

  11. It’s likely mainly only the very high end houses that won’t get rented out, capital gain is huge but cash flow is relatively less. Any damage/wear in a high end house is more costly/important to hold value. It’s a general property investor rule that an area that has no or little or even negative capital gain (eg Wairoa) will have the highest cashflow % (cheap house, relatively high rent). High capital gain area like, say Parakai Drive in Auckland, you may not bother renting out perhaps.

Leave a Reply