The Motu Institute recently published new research into the urban productivity premium in New Zealand, or the degree to which firms and workers in big cities tend to produce more and earn higher wages. This is an essential issue for urban and transport policy as it gets to the heart of why we have cities. As we’ve discussed in the past, cities offer opportunities for better connections between firms, workers, and customers, leading to better economic outcomes. (Economists usually describe this as agglomeration economies.)

In the paper – with the enthralling title of “Urban productivity estimation with heterogeneous prices and labour” – researcher Dave Maré sets out to update and extend some of his previous work on the topic. His new research investigates two issues that might bias estimates of the urban productivity premium:

  1. Imperfect competition in small markets: Firms in large cities face more competition and hence will tend to have less market power (ie ability to jack up prices) than firms in small cities. This tends to result in low estimates of the urban productivity premium.
  2. “Sorting” of skilled workers into cities: People with higher skill levels – which could mean more education, more experience, or better ideas – tend to gravitate towards cities. (Similarly, cities tend to have different mixes of firms and industries.) Not controlling for this can result in a high estimate of the urban productivity premium.

Even after controlling for these factors, Maré finds evidence of a non-negligible productivity premium in Auckland. That is,

We document an urban labour productivity premium, with Auckland firms having labour productivity that is 17.9% higher than that of firms in other urban areas, and 17.0% higher than firms in rural areas. Some of this premium is due to the mix of industries in different cities. Auckland has a disproportionately high share of employment in industries that have above average labour productivity. Adjusting for this composition reveals a smaller, but still sizeable, premium of 13.5% relative to other urban areas, and 11.3% relative to firms in rural areas.

Here’s a chart showing how other parts of New Zealand compare to Auckland in terms of productivity, after controlling for industry mix, workers’ skill levels, imperfect competition, and a range of other factors like firm size. (This chart is based on the first column in Table 3 of the paper.) As you can see, firms in Auckland are more productive than firms in other parts of NZ, with Wellington (4.2% less productive) and Tauranga (9.4% less productive) being closest to Auckland.


It’s worth noting that Maré’s new estimates of Auckland’s productivity premium are considerably smaller than his previous ones. In a 2008 paper, he estimated that firms in Auckland were around 51% more productive than firms elsewhere in New Zealand. These are obviously very different numbers! But, as explained in an appendix, the majority of the differences are due to different procedures regarding data selection and processing.

Notwithstanding the exact number, Maré’s new analysis raises a few important conceptual questions about the urban productivity premium. His analysis shows that a large share of the difference in productivity between big cities, small cities, and rural areas is down to the fact that skilled workers tend to sort themselves into cities. When we control for workers’ skill levels, we tend to get lower estimates of the urban productivity premium. Or, if you prefer that in economese:

The meta-analysis by Melo et al. (2009, Table 4) reports that studies that control for labour quality generally yield agglomeration elasticities that are 5 to 6 percentage points lower than studies that do not. In the current study, labour quality adjustment lowers the estimated agglomeration elasticity by 0.057 (from 0.079 to 0.022).

However, I’m not sure it is totally appropriate to adjust for skill levels, as it’s possible that cities’ ability to attract and retain talented, innovative, and motivated people (and productive firms) is in fact a type of agglomeration economy. In other words, we might be controlling away the effect of interest!

Open migration between Australia and New Zealand means that people who can’t find an appropriate place (urban and economic) in New Zealand can easily go to Australian cities. So the alternative for skilled people who are dissuaded from living in Auckland isn’t necessarily that they’ll go and start up a business in Hamilton. Instead, they might head across the Tasman, where their skills are totally lost to New Zealand.

What does this mean for urban policy in New Zealand? I’d tentatively identify two key ideas we might want to focus on in order to allow our cities to get better at attracting and retaining productive people and firms.

First, we need to think hard about whether our policies make it attractive for mobile people to come to (or stay in) our bigger cities. This is a key consideration for, say, urban planning reform, as high housing prices driven by constraints on housing development are an important barrier to people coming or staying. Evidence from the US suggests that, if left unaddressed, high house prices can systematically dissuade people from moving to productive places where they can put their skills to best use.

Second, we also need to think about how to preserve and enhance the amenities that are on offer in New Zealand cities. Our relatively clean air, reasonably well-preserved coastal environment (clean beaches, marine reserves, etc), and accessible forests and natural parks are important attractions, but other areas are letting us down. For instance, rural and small-town water quality is rapidly declining due to expanding dairy farming.

More relevant to transport, street design in New Zealand is pretty retrograde, leading to a lack of high-quality public spaces where people actually want to be. All too often, we insist upon shoving cars down corridors, heedless of the fact that some streets have higher value as places to be. But we know we can do better: places like O’Connell St and Wynyard Quarter give many people joy on a daily basis.

What do you think about the urban productivity premium? And how can we get more of it?

Share this


  1. The Mt Albert example shows that AT at least thinks urbanity comes from having five traffic lanes in a town centre, rather than wider footpaths and cycle lanes.

    1. Yeah, trends matter as well. Can’t rest on laurels.

      For what it’s worth, I doubt that the Auckland Plan targets around GDP and productivity growth are realistic. For instance, the combination of 5% GDP growth and 2% productivity growth implies average population growth of 3%… which is also a lot higher than we’ve experienced over the last decade. Really confuses me every time I think about it.

  2. Excellent post Peter. I too immediately wondered why Maré wants to adjust for sorting? That is exactly the point isn’t it? Productivity of places isn’t about where employees were born, but where they are drawn to to thrive.

    1. well, you could look at it two ways:
      – Cassandra: Auckland has a relatively small productivity premium (when measured in levels and growth), therefore we shouldn’t expect gains in the future to be that high; or
      – Alternative: Auckland has a relatively small productivity premium (when measured in levels and growth), therefore potential gains from more urban friendly policies will be larger.

      I note that Auckland has notably less dense central suburbs than most cities overseas, largely because of planning restrictions. So I think there’s a good case to be made for the fact that Auckland’s relatively low productivity premium compared to other countries is due to policies, rather than a function of our economy per se. I appreciate that we’re located in the south Pacific, but the core-periphery structure is still evident, which would suggest agglomeration economies were still relevant.

      The other thing to mention, which we’ve mentioned before, is that agglomeration economies also bring gains on the consumer side in terms of increased diversity of goods of services. You could argue that perhaps Auckland has seen more gains of this type than agglomeration gains in productivity.

    2. P.s. One other comment, which I allude to above: Not all agglomeration economies will necessarily show up in regional GDP figures. Some will show up in other socio-economic indicators.

    3. Cliff notes version:

      1. Maré’s paper documents differences in productivity *levels*, while Reddell is using regional GDP statistics to make an argument about *trends*. It *may* be the case that Auckland is more productive but also not growing any faster than the rest of NZ.

      2. Maré uses significantly more detailed data to make his estimates, and controls for a lot more stuff. So I would see his estimates as far more reliable than estimates made using statistical aggregates. There are also a number of issues around measuring GDP at the regional level that may (from what I understand) affect those figures.

      And finally, I agree with Stu that there are potentially large benefits to more urban-friendly policies and that’s where we should probably look if we want to enhance agglomeration economies. (Reddell’s preferred solution of reducing population growth to increase real wages seems to be based on economic theories that have been falsified. For instance, Japan, which has virtually no immigration and a rapidly aging population, is finding that it’s unable to increase wages in low-productivity sectors suffering from labour shortages.)

  3. My reading of it is he controls for sorting in order to make the maths work. He is using a regression approach so he needs to avoid mutual causality otherwise the regression simply fails. He says he is not interested in causation, so ok he isn’t trying to figure that out, but endogeneity is still likely and he wants to deal with that. Do cities make people more productive or do more productive people flock together and make cities grow? Probably both. Unless you can find a two-stage least squares proxy or use simultaneous regressions then you are stuck with the leading diagonal or orthogonality problem in your matrix. By removing sorting he can reach a conclusion about the rest of it. He is not saying the sorting is a bad thing. That is my reading of his article.
    My views on these types of studies are a bit different. They are almost all use wages and prices as a measure of productivity. Yes even the multi-factor studies often estimate the mpf’s from wage and price data using an algorithm. In this case the labour quality is measured from wages. Using a price to measure quality? There is an assumption that firms in smaller places operate in less competitive environments. I don’t buy that. A piece of cake costs more in Queen Street than in Whakatane. That is because the inputs cost more, rent is higher and so are wages. But the person selling the cake is probably just as productive in the smaller town, the person in Whakatane probably makes the cake and is more productive. Wage levels don’t just mirror output. I would turn the monopoly argument on its head. Queen Counsel don’t compete on price, nor do judges or CEO’s or most of the big city specialisations that result in the city premium. If there were truly monopoly profits to be had in smaller places then people and firms would be flocking there. The reverse is true.

    1. While there are a lot of rent-seeking industries present in cities, I wouldn’t overestimate the degree to which these will affect the numbers. This is because these studies will (usually) control for within industry effects. So you’d have to argue that within the legal industry, the lawyers in AKL are more likely to be participating in rent-seeking behaviour.

      I think that’s possible, but be aware that there’s a lot of rent seeking in small towns too, because often there is only one supermarket, electronics store etc. In my experience, the average household (thankfully) spends more on consumer goods than things like legal fees, so I don’t think the divergent effects would favour small towns.

      I think your cake example is actually more prescient than you think, but your conclusion is a bit off. That is, most studies find only weak agglomeration economies in things like food retailing. So this industry will tend to suffer from increased competition for factors of input (rent etc), without benefiting from higher productivity — hence leading to higher rather than lower prices.

      That’s why it’s almost better to talk first about industry-specific AE, and then aggregate to a certain geography based on its industrial composition.

      1. Extend the cake selling analogy to any job at the same level in both the city and small town, electrician, panel beater, dentist. The city ones cost more because their inputs cost more, not because they are more productive. Using prices to measure real productivity isn’t going to work.

        My view is that the sorting is the greater reason for cities. We came here for the chance to do easier, more interesting jobs. In effect the city grows because people show up rather than the city making people more productive. As for your views on small town rent seeking, I totally disagree. Some of the best prices you will get are in the regions. Buying a second hand car is cheaper, buying an appliance is similar but you get better service. Very distant places you will have to pay more for freight but that is a real cost not due to economic profits. The flaw I see with these studies is even using industry level data the managers tend to be in the big places. Look at Fonterra, they moved to Auckland from Hamilton because the people on salary preferred Auckland. It had nothing to do with access to markets or financial institutions or any supplier to their business.

        1. I don’t disagree on the importance of sorting.

          I do disagree on the cost of many items in small towns, mainly because many goods and services simply don’t exist. Yes, you may be able to find a widget cheaper in Te Kuiti.

          But you won’t find the widget dongle USB adaptor gizmo that allows you to charge your robotic dog via bluetooth. That requires a trip to somewhere else. And that costs more time money etc.

          Ultimately I think we’re arguing about two things: 1) income and 2) preferences. The latter is a normative question rather than a positive one. It comes down to what people want: If people just want the basics, then small towns are great. On the other hand if you have a preference for a wider variety of goods and services, and the income to afford them, then larger cities are probably better. And that’s great, because then people can choose.

          And as the evidence currently stands, people are on average choosing to live in cities — more so than small towns.

        2. Widget dongle USB adaptor gizmos are available on Amazon and Trademe. I agree with your point it is about both income and preferences. Income is easy to aggregate but preference can’t be. The same things that attract some people to a city can be the things others put up with as a cost of living in one.

        3. Yes such dongles are available on TradeMe, but the transport costs will often be higher. And arguably the search costs will be too — there’s still an advantage (small, I agree) from being able to observe products in person, especially for highly differentiated products.

          Also don’t forget about services: The ability to access good educational services, for example, is very important to some people. And such services are more readily available in cities, if not impossible to get in rural towns. Yes people in rural areas can send their children to boarding schools etc, but then again the costs will be much higher.

          Moreover, agglomeration economies in things like public services, such as schooling and recreational activities, are not priced and hence won’t show up in productivity differentials.

          Basically, I think the real monetary and non-monetary costs of accessing many (not all) goods and services is higher in areas with less agglomeration. One thing that is not is recreational activities, where many small towns in NZ do really well. Mind you, for a large city Auckland does pretty well there too … Waiheke anyone?!?

        4. Yes if you were buying a cello you would want to play it before you buy it. I doubt there are many shops that sell cellos outside of the main cities. But those are not every day purchases and are the reason country people make a trip to the city once in a while. I accept your point about education . Even Ed Glaeser has said he lives in a suburb because the schools are better.

        5. I’d argue that once you get past basics like housing, food, and energy then most products people buy are highly differentiated.

        6. “The city ones cost more because their inputs cost more, not because they are more productive.”

          To an extent. Maré’s done some things to try and control for spatial variation in input prices and their impact on measured productivity. He finds that: “Implied prices in Auckland are only 0.3% higher than in other urban areas, and 7.4% higher than in rural areas. Wellington prices for intermediate inputs are estimated to be 6.1% higher than in Auckland.”

          The productivity estimates seem to account for this variation.

    2. Yes, that’s all correct. FYI, I had a chat with Dave about the paper a few weeks ago; he’s very aware of the economic importance of sorting and also (obviously) the endogeneity issues that it creates for econometric analysis.

      2SLS or SEM seem to be challenging to implement here, so my preferred analytical approach would probably be something like:
      1. Control for sorting (etc) to get an estimate of the pure urban productivity premium
      2. Run a separate analysis to estimate the propensity of high-skill workers or high-productivity firms to sort themselves into certain types of places
      3. Add the two effects together to get an estimate of the overall agglomeration effect.

    3. “Queen Counsel don’t compete on price” – Sorry but they do. They might not give much of a discount but unless that QC has no competition in terms specialisation (so a monopoly on that area), they will definitely have to be competitive on price.

      I recently engaged a QC for a client and I can tell you that the price was a factor and was discussed.

      Being a QC doesn’t necessarily mean you are the best either, it is very political. Many senior barristers are competing with a QC for work.

      And of course judges don’t compete, they are appointed public servants paid a salary. QCs are not – they are barristers in private practice.

      And you don’t think CEOs compete on price? Of course they do!

      1. I’ll bet you didn’t call for tenders. You will have selected someone who you think can do the job and then had a bit of a discussion about fees to make sure the client could cover them. CEO’s are appointed by a board who are answerable to shareholders. That is an example of the agent-principal problem beloved by game theory students. At the end of it the shareholders lose if the CEO is no good, not the board. They can simply say “but we employed the best available, see he cost more than anyone else”.
        In a small town businesses have to graft for each dollar. In the city it is easier to find a rent seeking specialisation.
        To get back to the point, Judges senior lawyers, CEO’s and anyone else on a good wicket are people located in a city. That means when someone uses regression to look at salaries (or productivity calculated on an assumed function of salary) they will show a premium in cities. My point is that premium is probably more due to the sorting than it is to any transformative effect of the city itself.

        1. I don’t think tendering is relevant. To me an absence of tendering says more about transaction costs than the presence of market power.

          Most private firms I know don’t engage in tendering all that often, and if they do then it’s at the start of a large piece of work and is by select invitation only. Mainly for reasons of transactions costs.

          I actually want to write a research paper on how costs of tendering is likely destroying the effectiveness of NZTA’s research programme.

  4. This seems to be saying that talented and productive people liking living in cities should be discounted as a factor in the success of cities. Maybe this is interesting in studying the behaviour of networks of automata, but for people? For me this factor is one of the best arguments for cities like Auckland.

  5. I’m not entirely sure that I understand this productivity premium thing, and nor do I agree with its findings. In my (very) humble experience, Wellington is a far more productive place than Auckland because we don’t have the large travel distances / travel times that the Queen city has. Everyone in Wellington knows everyone else and meets them for coffee in Cuba St or Midland Park. It honestly is far easier doing business in this town than in a bigger city. And certainly more enjoyable. Less dodgy tradies for starters!

    So saying however, the increasing focus on Auckland means that local merchants no longer hold anything in stock – everything slightly out of the ordinary is ex-Auckland, and so therefore inevitably at least one day’s courier away. That slows down our productivity quite a bit…

    And re Mfwic’s comment above at 9.36am, pieces of cake do not necessarily cost more in Queen St than Whakatane. For instance, meals at a good restaurant in Hawkes Bay often cost more than the equivalent at a good restaurant in Wellington or Auckland. I don’t know why that is so, but it is true…

    1. My inuition goes the other way: Wellington’s economy is primarily government employees, and associated rent-seeking consultants (like me), whereas Auckland has a larger private sector focus which tends to operate in tradeable sectors, and hence be more competitive to price pressures.

      In saying that, the data above does suggest that Wellington does *relatively* well for a city of its size; the productivity gap is only 4.2%. That’s relatively small compared to, say, Christchurch. Although I’m not sure if the earthquakes affect those numbers …

  6. I’m surprised there’s not more mention of the effect on productivity of tertiary learning institutions including universities. Dunedin is an interesting one that deserves closer investigation. At -18.3 it’s a dog compared to Auckland. Why is that? Given that part of Otago University strategy should be around incubating business and innovation as students transition to Post Grad study and employment. I am aware there has been encouragement for high value creative and tech businesses to be based in Dunedin. Why is this not showing up in the productivity comparison to Auckland? Not enough runs on the board yet?

    I had always thought that the location of AUT and University of Auckland should be providing a great starting point to exactly the types of high value, highly-skilled-people-focused industry that one associates with big cities. If creating a big city vibe, and all the positive economics that brings is the ambition, we need to find out more about where the students go following study. And Why.

    1. Yes, that’s a really good question and one that bears investigation!

      A fair number of my friends studied at Otago but subsequently drifted north. Sometimes because that’s where their family and friends live, but usually for work. Dunedin could be a really great, dynamic city but probably suffers from poor economic geography. It’s a bit too cold to attract retirees and lifestyle migrants (unlike Tauranga or Nelson), and a bit too distant from the upper north island to attract economic spillovers from Auckland. And it was heavily specialised in import-protected manufacturing prior to the 80s, which has not been a growth sector.

      A bit hard to imagine a quick solution but I would argue that the university will be an important part of the solution.

      1. I did a heap of reading a few years ago on the economics of education. The University town vs non- university town comparison was one of the ways that had been used to try and figure the impact of education on earnings. Unfortunately most of the literature indicates that universities are not transformative. They are merely a selection and signalling mechanism for high earning people. By comparison primary education is transformative. Learning to read and write really does make a difference.

        1. Sorry I couldn’t edit that. The old idea was you could plot earnings as a function of years of education (Jacob Mincer) and people would do curves at different levels by gender, race etc. But more recent work indicates that without educational institutions those same people would have achieved higher earnings anyway. It doesn’t mean education is worthless, it just means a degree is a way of showing an employer you are good at working through boring stuff successfully over a prolonged length of time.

        2. This is where natural experiments and instrumental variable approaches have a lot of value. Angrist and Pischke have a really good discussion of some research in this vein, which does seem to suggest that there are positive effects from more education, but smaller ones than you’d expect based on the simple charts.

        3. The conclusion most people had come to when I did the reading was that the benefits were personal. Being educated lifted you up and enriched you in the way the UN Human Development index suggests. But keeping people at school for two more years than they wanted to be there wasn’t going to increase wealth.

        4. Interesting commentary and thank you for your feedback.

          It leads to some other questions also related to that productivity premium effect with larger populations. Are tradesmen (and supporting industries) building houses faster and on a bigger scale than in the provinces contributing to the Auckland productivity advantage? If so, that is going to be placing considerable weight on that Auckland number.

          Perhaps looking at London in comparison, it might be found that banking and associated service industries might dominate that economy in a similar way. And help London to have what I suspect is a vastly greater productivity than the regions of Great Britain.

          It is reasonable to suggest that the numbers of workers with a university degree is likely to be higher in banking than building houses. The respective value to each economy is arguably equivalent – that of generating wealth into the community. In both cases, critical mass and a degree of specialization drives productivity. Size matters, the exact industry, and the nature of the job training less so?

  7. Amazing how some break down life into “productivity”. A bit sad really.

    I would suggest those furtherest from Auckland’s figure, are closest to living a better life. A more laid back lifestyle is more desirable to living in a rat race, and if business earning less is a byproduct of that, so be it.

    But workers don’t have less money outside Auckland. Anyone who has lived in provincial New Zealand and Auckland understands full well you have more money in your pocket after expenses if you live outside Auckland. A dollar goes a lot further in Wanganui than it does in Auckland, far outstripping higher wages in Auckland.

    1. Auckland a rat race? You having a laugh? The work environment is so laid back it’s not funny. In summer at 5pm the office here is deserted. In winter they’re all in the boardroom having drinks. People move to Auckland to get away from the rat race, a quiet backwater branch economy kind of place in a lovely natural setting with an alright climate.

    2. The thing with Auckland Geoff is that can have both. As long as you’re fortunate (bust your arse long enough?) to earn a decent amount.

      Just a shame it’s so badly managed. The weather’s a bit shit in winter but there’s always Fiji I spose…

    3. I agree you can have more money left in your pocket if you live outside Auckland. But the problem with that is you have to live outside Auckland. That is fine if you fit into those places, but if your interests are different, or if you are simply different yourself then provincial places can be difficult. The benefit of population is you are more likely to find people you get along with in amongst the crowd. Similarly try having any sort of health issue outside of Auckland. I don’t think these things show up in productivity numbers and certainly not productivity derived from wage data but they are real issues.

    4. “I would suggest those furtherest from Auckland’s figure, are closest to living a better life.”

      Property prices disagree with you. People are willing to pay twice as much on housing to be in Auckland.

      I tried leaving: it’s horrible, everything is closed at 5.30 and even the restaurants close at 8, I have to drive everywhere beyond walking distance because the public transport is rubbish, the bars price gouge because they are all owned by one company, and there are precisely 0 cultural events. I can’t wait to come back because, just like you, I prefer Auckland.

Leave a Reply

Your email address will not be published. Required fields are marked *