Drive through Brightwater at two in the morning and you will see a district lit up for a crowd that went to bed hours ago. Every pole burning, every cul-de-sac, every stretch of rural road where the only traffic between midnight and dawn is a possum and the milk tanker. We are paying to do that. We are paying for the power, we are paying for the shorter life it puts on the lanterns, and that bill comes out of the same pot as the roads and the pipes. On top of it we are paying a quieter price — the night sky over one of the few districts in the country that still has one worth looking at. Tasman District Council is responsible for around 3,800 street lights. Before the next Long Term Plan tells us there is no money for anything, it is worth asking whether all of them need to be on at full output all night.
Start with the money, because that is where a council argument lives or dies. Tasman changed its street light strategy in 2015 and converted the network to LED, and the council’s own Transportation Activity Management Plan puts the saving at $5.95 million in maintenance and power costs over thirty years against carrying on as before — and it calls those energy assumptions conservative. Good. That was the easy win, and it is banked. But the same plan said something else that has quietly gone nowhere. It noted that the new LED fittings make a centralised management system possible “as an optional extra”, and that such a system “can enable greater energy savings through controlling levels of light output to where and when it is required, eg, light dimming between midnight and dawn instead of all lights operating at full output throughout the hours of darkness”. Those are the council’s own words. “Future consideration will be given,” it said. That was in the 2018 plan. The 10-Year Plan 2024–2034 does not mention dimming, part-night lighting or a management system anywhere. Six years, and “future consideration” has produced nothing you can point to. A good idea identified by staff, minuted, and left to gather dust. Business as usual.
How much would it save here? That is exactly the figure the council should have to put on the table, and hasn’t. Waka Kotahi’s guidance is that an LED street light can cut up to about 60 per cent of the old power bill, and that dimming it through a central management system can lift the total energy saving to as much as 70 per cent. Nelson City ran the same LED conversion and expects to save around $200,000 a year in power and maintenance. The dimming layer on top is real money, every year, on top of money already saved, and it comes with lanterns that last longer because they spend half the night at half output.
Cornwall Council in England has already done it. It turns off street lights between midnight and 5am across minor and residential roads, dims the rest, and keeps full lighting in town centres with a night-time economy, on main distributor roads, and anywhere covered by CCTV. Stated saving: about £680,000 a year. Street lighting, on their figures, makes up close to 30 per cent of the whole council’s carbon footprint. Somerset, Dorset, Shropshire and Coventry run part-night schemes on the same midnight-to-dawn pattern. France went further in 2018 and did it by law — new outdoor lights capped at 3000 Kelvin, tight limits on light thrown upward, and curfews on shop-window, office and advertising lighting. The French government’s own figures show a 19 per cent drop in exposure to serious light pollution across mainland France between 2014 and 2023. This is not a fringe idea. It is what councils do when someone finally looks at the power bill.
Now the objection, because it is a fair one and it deserves a straight answer rather than a brush-off. Some submitters and some councillors will say a darker district is a more dangerous one — for the nurse coming off a late shift, for the hospitality worker walking home at eleven, for any woman who has ever had to think about the walk from the car to the front door. That fear is real and it is not irrational, and any policy that waves it away deserves to fail.
So what does the evidence actually say? The largest controlled study of this in the UK is Steinbach and colleagues, published in 2015, which looked at 62 local authorities across England and Wales and compared areas that had switched off, gone part-night, dimmed, or changed to white light against areas that hadn’t. On night-time road crashes it found no measurable change from any of those strategies — switch-off came in at a rate ratio of 0.97, part-night 0.95, dimming almost exactly 1.0, with confidence intervals straddling “no effect” in every case. On crime, same story: no detectable rise in total crime under part-night lighting or dimming, and a weak hint that dimming and white light went with slightly less crime. One strand of the analysis did throw up a faint, statistically shaky suggestion of more robbery under part-night lighting — worth saying, but it did not reach significance. The same project’s cost-benefit work concluded part-night lighting “may represent a large net benefit”. A separate randomised trial in New York public housing, by Chalfin and colleagues, is the honest counterweight: putting bright lighting into places that were genuinely dark and high-risk cut night-time outdoor crime by at least 36 per cent. And closer to home, Waka Kotahi’s Economic Evaluation Manual assumes that improving poor road lighting cuts crashes by about 35 per cent, and before-and-after studies here and overseas show reductions of around 30 per cent or more; a New Zealand study by Jackett and Frith found a clear relationship between how well a road is lit and its night-time crash rate.
Put those together and the answer is not “darkness is safe.” It is narrower than that: adding light where it was genuinely missing and dangerous does real good, and dimming or killing a light nobody is using at 3am on an empty residential street does no harm anyone has been able to measure. Those two findings only look like a contradiction if you think this is all-or-nothing. It isn’t. Nobody serious is proposing Tasman go fully dark. The proposal is to keep the lighting where people actually are after midnight — Richmond’s centre, the state highway junctions, the known crash blackspots, the routes between late workplaces and the bus stops, the CCTV areas — and dim or switch off the rest.
The evidence does carry one real cost, and it should be named. In that same English research, residents living under part-night lighting said they felt less safe walking alone at night, even where the crime and crash numbers had not moved. That matters. It changes whether people go out on foot at all, and it does not land evenly — it lands hardest on women. A council that waves that away is not being brave, it is being careless. The answer is not to drop the idea; it is to design it properly. Keep the pedestrian routes lit. Consult the people who do the late walk before you draw the map, not after. And where lights do go off, say so publicly and say why, rather than letting people discover it by walking into it. The places that got this wrong flicked a switch across the whole network with no exceptions and no warning. England’s national highways agency reversed its overnight motorway switch-offs in 2018 after casualties on unlit stretches jumped. The AA, the loudest critic of switch-offs, says street-light reductions have been recorded at inquests as a contributory factor in eleven road deaths since 2009, and that in blackout areas 35 per cent of people feel less safe after dark, rising to 43 per cent among women. But look at what the AA actually asks for: full lighting kept on where the speed limit is 40mph and over — about 65km/h here — and on slower residential streets, lights that were switched off turned back on with energy-saving technology “unless residents specifically request otherwise”. Even the loudest critic of switch-offs is arguing about which roads and what technology, not about whether to touch the network at all. Doing it badly is worse than not doing it. That is an argument for doing it carefully, not for leaving every light burning at full power until dawn.
Then there is the part that doesn’t show up on the power bill. Tasman District Council owns the Wai-iti Dark Sky Park near Wakefield — 135 hectares of council land recognised in July 2020 as the first International Dark Sky Park accredited in New Zealand. The council already tells the world that the night sky here is an asset worth protecting and worth visiting. Aoraki Mackenzie has built a genuine tourism economy on exactly that. Meanwhile the glow off Richmond and Motueka climbs into the same sky every night. Worldwide, the rate at which stars are disappearing from view puts the increase in skyglow at close to 10 per cent a year — fast enough to double in under a decade — and the Milky Way is already invisible to more than a third of humanity. A council that has staked a claim on dark skies and then floodlights its own empty streets until dawn is working against itself.
The wildlife side of this is not sentimental either. A 2021 study found street lighting cut the number of moth caterpillars in roadside hedgerows by around half, worse under white LED than under the old sodium lamps — among the first field evidence that street lights are pushing down wild insect numbers, and insects are what everything else eats. The Department of Conservation has published advice on keeping artificial light away from pekapeka, our native bats, which call less and avoid flying where it is lit. New Zealand research has documented the same avoidance in wētā and links between coastal lighting and seabird groundings. DOC runs a campaign asking people to dim lights for migrating birds. The council’s own draft Environment Plan already proposes to “introduce controls to maintain and enhance the darkness of the night sky”, naming upward light spill, colour temperature and “curfews / time limits on some lighting”. If rules like that can bind a farmer’s security light, they can bind the council’s own network.
The health question is the least settled, so take it as caution rather than alarm. Blue-rich white light at night suppresses melatonin, and the American Medical Association in 2016 told councils to use warmer light — 3000 Kelvin or lower — shield it, and dim it where they can. The lighting industry disputed parts of that report. But warmer, shielded and dimmed off-peak is now the mainstream position; the warmer and shielded part is written into French law, alongside curfews on advertising and building lighting, and it costs nothing to adopt when lanterns are being replaced anyway.
So where does the cost-benefit land? On the side of turning the lights down. Dim the network after midnight, switch off the quiet residential and rural roads, keep the lighting where people genuinely are after dark — and you save money every year, with no rise in crashes or crime in the largest controlled study of it. It cuts carbon. It protects a night sky the council already markets to visitors. It eases the pressure on the native wildlife the council is already spending money to bring back. The costs are real but manageable: design it around the late-shift walk, tell people plainly where and why, and keep it well clear of the fast roads where darkness does cost lives. Do it as a blunt all-lights-off switch and it will fail, and deserve to. Do it street by street, with the map drawn by the people who use these routes at night, and it is one of the cleaner wins on the table — permanent savings, no repeated fight, and a sky worth stepping outside to look at.
The lights are on tonight over streets where nobody is home, burning at full power, and we are paying for every one of them. The council’s own plan raised dimming them years ago and then let it sit. So the question for this Council, before the next round of “there is no money”, is a simple one: a district that owns a Dark Sky Park, that is already writing night-sky rules for everyone else’s lights, and that was told by its own engineers six years ago that the fittings could already do this — what exactly is it waiting for?
