Will Cambridge run out of water?
Probably not.
Britain needs to build more housing, or at least more housing in the places with highest demand; especially London, Oxford, and Cambridge.
The Government recently announced plans to build 1 million new homes, including a ‘a new urban quarter in Cambridge’. Cambridge’s local councils were already targeting north of 50,000 new homes, but the central Government is mooting figures closer to 250,000 new homes for the area.
The reaction of the local (Conservative) MP was not positive:
Call this the water objection to Cambridge housebuilding.
The knee-jerk cynical YIMBY response to the water objection is that it’s a figleaf for NIMBYish opposition to any and all development. In the absence of the water objection, we’d have the endangered newt objection, the ancient monument objection, the noise pollution objection, or anything else capable of being wielded to prevent development.
A cursory check of some claims made doesn’t allay this YIMBY scepticism. For example, Browne mentions the Environment Agency ‘systematically blocking all major new development’. While it’s true that they have objected to planning applications for new developments, citing water concerns, objecting to planning applications is not the same as ‘blocking’ developments.
Further, water objectors equivocate between two distinct points:
There is not enough available water to meet municipal demand.
There is not enough available water to meet municipal demand without avoiding environmental damage.
The first interpretation implies that new housing will result in taps running dry (‘We have run out of water’, ‘There is not enough water for existing housing’). The second interpretation implies that, while there is sufficient water to supply new housing, this will come at the expense of degrading (chalk) streams (‘streams and ponds already run dry’).
Water objectors might claim that meeting municipal water demand only at the cost of environmental damage is unacceptable. Fair enough. However, if that is the claim, it would be better not to frame it in a way that implies insufficient water for municipal demand full stop. (Not least because these different problems might have different solutions.)
On the other hand, the water objection is not some ad hoc post hoc rationalisation to object to the most recent development plans. Browne has been vocally concerned about local over-abstraction of water for years. Further, the water objection isn’t party-political. It’s cited as a limit to Cambridge growth by local politicians and activists of all stripes. At that link, the Liberal Democrat leader of the Cambridgeshire County Council refers to ‘the crisis of water sufficiency in our area’, and the (Labour) Cambridge Mayor claims that ‘Water scarcity is a very real issue that is limiting the growth of our marvellous city’, also citing a ‘water crisis’.
Despite its constant repetition, however, I haven’t seen a fleshed-out analysis of the water objection. So I’ll try to figure it out for myself.1
1. How much water is available in the UK?
Cambridge is the driest city in the UK, so it is prima facie plausible that if anywhere in the country might run out of water, it would be here. On the other hand, the UK is much wetter than many other countries. If countries in literal deserts can supply water to their populations, surely the UK can too?
Let’s first get a sense of the available water resources.
I’ve selected a few local peer countries (plus the US) to compare per capita renewable water resources. I was surprised that, despite our rain, the UK isn’t especially blessed with water resources:
But this doesn’t yet give us a good grasp on potential water scarcity. That depends on how much water we use. And the UK uses far less water than these peer countries:
(Why? Because we use much less water for industry and agriculture.)2
If we divide renewable water resources available by withdrawals made, we can arrive at the factor by which we could multiply existing withdrawals without exceeding the resources available:
The UK could increase water withdrawals 17x before depleting renewable resources, while Italy, Germany, and Spain only have leeway for a 3x expansion.
The water objection is specifically that there isn’t enough water for houses, for municipal use. Building more houses does not necessarily entail more industrial or agricultural water use (indeed, it may even decrease them if heavy industrial or agricultural land is turned into housing). How many times could we multiply municipal water consumption without exhausting water resources available?
First we need to find the municipal share of water consumption. In the UK this is proportionately high…
…but absolutely low:
I took the total amount of ‘untapped’ renewable water resources (ie total renewable resources minus total withdrawals) and divided this by the current municipal withdrawal amount per capita, to find a factor by which we could increase municipal water resources without exhausting available renewable water resources:
By this metric, the UK could increase municipal water resources 23x before depleting renewable water resources. If per capita consumption didn’t change, then we’d need a population of 1.5bn people - more than India or China - to get there. No feasible increase in municipal water use will come close to depleting the UK’s renewable water resources.
2. How much water is available in Cambridge?
While the UK has sufficient water resources to increase municipal water consumption, the water objection is not about the UK as a whole. It is about Cambridge specifically. And Cambridge is the driest city in the UK. If the UK’s water resources are in the north west, and the increased demand is in the south east (only a slight oversimplification), then national water abundance wouldn’t rebut the water objection to Cambridge housebuilding specifically.
Cambridge water is supplied exclusively from local groundwater, mostly from a chalk aquifer to the south of the city. The water is abstracted via 26 groundwater boreholes. This puts a natural limit on the water available, and is the root source of the water objection.
How limited are the water resources available? English water companies have a regulatory requirement to submit a water resources management plan every five years. Cambridge Water’s draft plan for 2024 is now available, so we can see how they are modelling the local situation.3 Their headline forecast, and presumably the reason for all the concern, is this one:
This shows that supply will roughly equal total demand + target headroom in 2023 (ie now), and that this headroom will be lost by 2025, such that afterwards there will be a water deficit. This, I assume, is the basis for the claim that ‘we have run out of water’.
Three things are worth clarifying about this chart.4
First, most importantly, this is ‘the predicted outcome if existing policies are continued without any further changes’. Obviously it would be a bad thing if there was not enough water. Equally obviously, Cambridge Water will take measures to prevent this outcome. It would be highly misleading to treat this as a prediction about what will happen all-things-considered. (Cf sleepwalk bias.) More on this later.
Second, this forecast includes new housebuilding, specifically 41,250 new homes (with 75,500 more people). While recent plans suggest more housebuilding than this, it’s not the case that Cambridge Water is modelling purely current demand. (And they do include a sensitivity analysis, ie measure how sensitive the model is to assumptions like this.)
Third, why does supply decrease in 2024/25, 2029/30, and 2039/40? The draft report explains that the model includes ‘impacts on supply from climate change, reduced DO [deployable output] from improved modelling and groundwater source availability and reductions in DO to protect the environment’. The 2030 drop is apparently due to ‘licence capping’, and the drastic 2040 drop due to ‘environmental destination licence caps’. This is effectively a requirement to reduce groundwater abstraction, thereby solving the environmental concerns cited by Browne. Note that demand would only barely exceed supply without these enforced supply reductions.
That’s what Cambridge Water’s model predicts absent any mitigation measures. So what mitigation measures do they suggest? The draft plan notes various possibilities for increasing supply: improving extraction at current sites, exploiting new sources (groundwater or surface water), transfers from elsewhere, as well as desalination and other exotic options (including ‘iceberg import’):
After considering various options, they conclude that the feasible options (in order of impact) are building a new reservoir, transferring in water, reusing more water, and ‘licence optimisation’ (a euphemism for taking more water from existing sources). They think that essentially all these feasible options need to be implemented to meet demand:
The vast majority of the proposed extra capacity is via the new Fens Reservoir, (mentioned by Browne):
The problem, as Browne pointed out, is that this reservoir will arrive too late to avoid a short term deficit.5 Construction is planned to start in 2029 and water supply to begin by 2035. (Ie 12 years from now - considerably quicker than Browne’s ‘nearly 20 years’.6)
The main stopgap plan is to transfer in water from Grafham Water, a reservoir located about 20 miles northwest of Cambridge (next to Huntingdon and St Neots), which is run by Anglian Water:
Their model based on these plans, which also include some rather ambitious demand reduction, looks like this:
What’s the takeaway here?
This illustrates and vindicates some of the concerns behind the water objection. There does appear to be a risk of a short term water deficit in Cambridge. While there are plans to fix this with the new reservoir, that is only scheduled to come online in 2035. Moreover, Cambridge Water’s plans do involve extra water extraction via ‘licence optimisation’, which will result in less water available for already depleted streams.
However, against talk of a ‘water crisis’, there is a plan in place to resolve these issues, including short-term transfers from Grafham Water to meet demand. I see no realistic prospect of taps running dry (the first and stronger interpretation of the water objection), but a more realistic prospect of environmental damage (the second interpretation).
If you disagree, you can bet against me here:
(The starting odds of 10% were generated by me betting 100M on ‘No’. This is not because I believe taps running dry is 10% likely - I think the odds are less than 1% - but rather because I didn’t want to put all of my virtual play money into a single market.)
3. Is stopping housebuilding the best response to water scarcity?
There are legitimate concerns behind the water objection: increasing municipal water use will result in somewhat more water abstraction in the short term, and this will probably impact the amount of water available for chalk streams. Stopping housebuilding will reduce demand for water in the area and thereby protect these streams.
Stopping housebuilding is one solution to water scarcity. But is the best solution?
Almost certainly not.
A. The value of housing
Imagine that you win the lottery and want to build your dream house in the country. You find the perfect spot with great views. But there’s a catch. Given how remote your plot is, the water company won’t hook you up, as doing so would cost them £100,000.7
What do you do? Easy. Just pay them £100,000 to hook you up. It’s easily worth that to you, given how perfect your plot is in every other way. (Or figure out a cheaper solution, like building a well.) While £100,000 would normally be a crazy amount to pay for a water connection (it’s usually a few £1000s) it might be worth it in your unusual case.
Likewise, if the value of new housebuilding is sufficiently high, then previously uneconomical solutions to the water problem will become economical.
Didn’t Cambridge Water’s draft plan say that it had no such options? No, it did not. Cambridge Water looked at ‘feasible’ options to balance supply with demand. Feasible, in this sense, is relative to something like current prices. They did not try to figure out the disvalue of stopping housebuilding. They therefore did not consider responses which would be proportionate to that disvalue.
The price of agricultural land is roughly the same everywhere in the UK: about £25,000 per hectare. Wales is cheapest at just over £23,000/acre, while Cambridge’s superior farmland pushes us to the lofty heights of almost £27,000/hectare. The reason for the similarity is because, given modern fertilisers and technology, an acre of land can produce roughly the same value of crops anywhere in the country.
By contrast, the price of residential land varies enormously throughout the UK. According to the Government’s land value estimates, residential land in the cheapest regions is worth £370,000 per hectare, while residential land in the most expensive area, Kensington & Chelsea, is worth £161 million per hectare.
(This estimates the unimproved value of the underlying land for residential use, not the buildings themselves, assuming a building density of 35 units/3150 square meters per hectare outside of London (denser within London).)
South Cambridgeshire clocks in at a more sane £5.4 million per hectare, a bit shy of Cambridge proper’s £6.25mn.
This means that South Cambridgeshire residential land is 200 times more valuable than agricultural land. Converting a hectare of South Cambridgeshire agricultural land to residential land generates £5,375,000 of land value. Divided by 35 (for the number of assumed units per hectare), and this is just under £155,000 of value unlocked per home.8
(For a broader accounting of the costs of stopping housebuilding, see the housing theory of everything.)
(Plenty of proposed Cambridge development is brownfield rather than greenfield, but I don’t think this makes a large difference. The two possible candidate locations for a ‘new urban quarter’ in Cambridge are the barely-used airport, which may as well be agricultural land, and the Cambridge waste water treatment plant, which is planning to relocate onto (presently) agricultural land outside of city bounds, making the ultimate incidence of the land-use change agricultural-to-residential. Most other major developments are slated for similarly non-valuable brownfield land (eg former barracks at Waterbeach and Northstowe), while others are greenfield projects.)
If the water objection was the sole reason not to build more houses in South Cambridgeshire, then stopping building would be the best response to that objection only if other solutions would cost more than £155,000 per property. And that is extremely unlikely.
What solutions to the water problem might be cost-effective, given the enormous costs of stopping housebuilding?
Happily, the Government is already establishing a ‘Water Scarcity Working Group … with the Environment Agency, water regulator Ofwat, local government and industry to try to address the issue.’ But until they report back, here are some options.
B. How to increase water supply
i. Transfer
The obvious solution is just to pipe in water from elsewhere.
Water transfer can be achieved with literally ancient technology. The Romans completed their first aqueduct, the Aqua Appia, in 312 BC.
Closer to home, London was supplied with water via the (lead) Great Conduit in the 13th century and the New River from the early 17th century. Importing water is standard practice for every city.
The main question is how far the water can be transferred, how economically.9
Cambridge Water’s draft plan already proposes to transfer in 15 megalitres (million litres) per day from Grafham Water, starting in 2027.
If we built more houses, could we just… transfer in more water?
Let’s see whether this would be economical by considering a completely over-the-top water transfer scheme.
Various ambitious long-distance water transfer schemes have been proposed at various times, including from Kielder reservoir (in the far north of England) to London:
These schemes generally didn’t proceed because they weren’t cost-effective. But whether they are cost effective depends on the costs and benefits in question.
How much would it cost to transport water from Kielder reservoir to Cambridge?
For reference, Kielder is the biggest reservoir in England, and about as far from Cambridge as it’s possible to be in England. It’s currently used to supply the Teeside region (Newcastle et al), but the enormous capacity means that plenty could be siphoned to the south. (According to Wikipedia, it has a volume of 200 billion litres, as compared with, eg, Grafham Water’s 60 billion.)
Zhou and Tol (2005)10 note that estimates for the cost of water transport are not very precise and ‘should be treated with great caution’. Nonetheless, they cite an earlier study (Kally (1993)) for the figures of 6.1 cents/m3/100 km for horizontal transport and 5.2 cents/m3/100m for vertical lift.
Adjusted for inflation, 6 cents in 1993 would be 30 cents in 2023, ie 5x higher. However, Zhou and Tol claim that Kally’s estimates ‘seem to be on the high side’, and that a more recent project (Hahnemann (2002)) reported 15x lower estimates. So I’ll split the difference on the conservative side by using the Kally figures, but without adjusting for inflation.
Very crudely using today’s exchange rates ($1 to £0.78), and estimating the distance and elevations required via a Google maps cycling route, I make the distance from Kielder to Cambridge 518km horizontal and 1,576m vertical lift.11 Plugging in our transfer cost figures:
Horizontal: 518km/100km * ($0.061 x 0.78 = £0.04758) = £0.246
Vertical lift: 1576m/100m * ($0.052 x 0.78 = £0.04056) = £0.639
Total: £0.66/m3.
That is: using the most commonly cited water transportation cost figures, with some hokey but relatively conservative assumptions, it would cost £0.66 to transport a cubic metre (1000 litres) of water from the (abundant) Kielder reservoir to Cambridge.
Taking the British average household consumption of 95 cubic metres/person/year, that means that shipping in all of a person’s annual water from Kielder to Cambridge would cost them an extra £63 per year. The average UK household is 2.4 people, so this would be an added cost of £151 per household per year.
Now recall that we estimated that the value of unlocking land to build one house in Cambridge is about £155,000.
If our only objection was water supply, then transferring in water absolutely ridiculous distances is still worth it.
Indeed, it would be worth it even if my estimate was too low by an order of magnitude— if transferring in water (extraordinary distances) cost £1500 per household per year. At those prices it would still take over 100 years to break even with the unlocked land value. And, given that I selected the furthest possible major reservoir in England, and assumed that all new water would come from there, the real cost of transferring in sufficient water would be dramatically lower than this.
I’m not suggesting that a Kielder transfer is a good idea. There are environmental objections to long-distance water schemes (eg due to mingling invasive species), and at any rate far closer sources would be more appropriate. Plus, it might take too long to build the relevant pipes and pumps to avoid a short-term deficit.
My point, rather, is that even absurd schemes like this would be more cost-effective solutions to water scarcity than stopping Cambridge housebuilding.
ii. Faster reservoirs
In addition to the plan to transfer water from Grafham, Cambridge Water’s main plan is to build a big reservoir in the Fens. This is their main and biggest solution to Cambridge’s water issue, planned to generate 87 megalitres of capacity per day, which should effectively solve the water scarcity issue:
The problem was that it will arrive too late to avoid a short-term deficit.
Can we just… build it faster?
We’ve already talked about Kielder Water, the UK’s largest-capacity reservoir. This was built in 6 years. Rutland Water has the second largest capacity, and has the largest surface area. It was built in 4 years. The new plans to build the (smaller) Fens reservoir are to do it in 6 years (2029-2035), which is well within the historical rate. But the construction itself is only scheduled to begin in 6 years’ time (2029), doubling the total lead time.
This is not especially surprising. It seems that the UK hasn’t built any reservoirs since 1992, and there are only three new reservoirs planned in the entire country: the two mentioned by Browne (Fens and Lincolnshire) plus one in Oxfordshire. But ‘not surprising’ doesn’t mean ‘justified’ or ‘can’t go faster’.
In the original water objection I quoted at the beginning, Anthony Browne claimed that he had ‘written to the water companies to speed up construction, but they say they can’t’. So what’s causing the delay?
Clearly aware of the slow pace of new water projects, the three main water regulators (Ofwat, the Environment Agency and the Drinking Water Inspectorate) created the Regulators’ Alliance for Progressing Infrastructure Development, aka RAPID. They explain their timeline as follows:
There are four gates during this period. At each gate, companies submit information about their work on a solution, which is assessed to ensure companies are making progress on investigation and development of solutions.
Ofwat provides the following timetable for each ‘gate’ on their website:
It seems, then, that the slow pace of infrastructure development is partly baked in by the regulators’ timetable. While RAPID frames this as ‘ensuring the water companies are making progress’, it seems that there is no mechanism for the water companies to make faster progress than anticipated by the regulators, because they can only begin construction once they pass through each of the pre-scheduled ‘gates’.
The obvious solution for the Government here, then, is to accelerate this process on the regulatory side. Go faster!
iii. Desalination
So far I’ve focused on the exploitation of natural water resources (from rivers, streams, springs, aquifers, etc.) But could we instead desalinate water?
The Cambridge Water draft plan mentioned desalination in their longlist of options, but not the shortlist.12 Presumably this was because it was not cost-competitive with other options. Once again, though, this doesn’t mean that it isn’t worth it relative to preventing development.
Zhou and Tol (2005), whose estimates on water transport costs I mentioned above, provide a desalination cost estimate of $1/m3 or £0.78/m3 to desalinate water. But that was in 2005. Israel only started desalinating around then, and is now the world’s leader in the technology. According to Energy Monitor, Israel’s ‘Sorek B desalination plant currently under construction is contracted to produce water for $0.41/m3, while, away from the cutting edge, the US Department of Energy guidelines ‘suggest that operators should target… $0.50/m3 for… sea water’. The UK may not be able to match this, as desalination is energy-intensive, and the UK has much higher energy costs than other countries. But it seems like the price of desalinated water would be somewhere between £0.40 to £0.80/m3.
That desalinated water would still have to be transported from the nearest coast (at King’s Lynn), which is around 80km if we account for some bends. Plugging in our transportation costs from earlier, this would add around £0.08/m313 . This gives us a total cost of between £0.48 and £0.88.
Converting that to an annual household cost (95 m3 per person x 2.4 people per household) we get £110 to £200.
These estimates straddle my estimate for my bonkers Kielder transfer scheme (£155/household/year), and would again be a bargain relative to the costs of preventing Cambridgeshire development (£155,000 per home).
Desalination also has the benefit of not being quite so bonkers, and potentially quicker to implement. Israel’s original Sorek desalination plant took under 3 years from start of construction to operation. An Australian project took 5 years from announcement to completion. One Texas plant took 6 years from planning to completion. And the UK’s first (and so far only) desalination plant - the Thames Water Desalination Plant - took 8 years from order to completion, including delays due to objections from the London mayor at the time, Ken Livingston.
(A second proposed desalination plant, to be built near Southampton, was scrapped in the face of local environmental objections. Because of course it was.)
iv. The Evian solution?
So far I’ve offered several (relatively) out-there solutions and claimed that they are all cost effective relative to preventing housebuilding. You might be suspicious that I’d say this no matter the proposed solution. Are there any limits?
Here’s a test case: would it be cost effective for all new Cambridge residents to rely exclusively on premium bottled water? Let’s work it out.
Tesco sells 9 litres of Evian for £4, or £0.44/litre. (By contrast, tap water in Cambridge costs £0.001/litre, or 440 times cheaper.) At the average per capita annual water consumption of 95m3, the Evian solution would cost £41,800 per person or £100,000 per household per year. So there we have it: the Evian solution is not cost-effective, even relative to stopping housebuilding.
C. How to decrease demand
So far I’ve focused on supply-side measures. But these are relatively unpopular. Cambridge Water’s draft plan involved surveying customers on how they should balance supply and demand. The results are not surprising:
Across all qualitative and quantitative engagement customers from all demographics have and continue to consistently prefer demand side options, rather than increasing supply side options.
Likewise, a spokesperson for Cambridge offered the following response to the ‘water crisis’:
We urgently need the government to commit to all necessary measures to resolve these problems with the council and our partners. This includes innovative measures to reduce domestic, industrial and agricultural water consumption and investment to clean our rivers and chalk streams.
Note the absence of concern about increasing supply.
I think that focusing on demand reduction is a mistake. Recall that the UK already uses much less water than comparable countries, and somewhat less for municipal uses:
Misgivings aside, how can we reduce demand?
Ask any economist how to balance supply and demand and the answer will be obvious: prices. The price of water could be set to achieve sufficient demand reduction to avoid environmental degradation.
(Municipal water pricing is partly hindered by the absence of metering in many properties, which is a curious feature of the English water supply system. The Cambridge Water draft plan duly spends a long time discussing metering to achieve this. Ultimately, however, this is a relatively simple regulatory fix: mandate meters.)
The degree to which price increases reduce demand depends on the price elasticity of demand. Waddams and Clayton (2010)14 found no studies which measured the price elasticity of demand for municipal water consumption in the UK specifically, and that international estimates vary widely. But they do offer a mean estimate of -0.38. This means that a doubling in price (100% increase) leads to a reduction in municipal demand of 38%:
Reducing household consumption by 38% would be more than enough to bring demand in line with current supply to a degree which would protect the chalk streams. And the median estimate here suggests that this could be achieved by doubling the price of water. (Albeit with wide error bars in both directions.)
Doubling the price of water sounds scary. But Cambridge Water’s tariff as of 2023 is £1.05 per cubic metre (1000 litres). This is already anomalously low. Anglian Water, our next-door neighbour, charges £1.68 per cubic metre. And Norway and Germany apparently charge over £4.00 per cubic metre. The price of our water could more than double while remaining within the limits of other comparable locations.
Would doubling the price of water make it unaffordable for the poor? No. Even though a 100% price increase is extreme in relative terms, it is not particularly extreme in absolute terms. We saw above that average municipal water consumption per capita is 95 cubic metres. At Cambridge prices this comes to almost exactly £100 per person per year. (Actual water bills are higher due to standing charges, but we’re only interested in the marginal/variable portion of the bill.) £100 per person per year is small relative to other household costs. For comparison, average yearly spending on alcohol averages between £200 to £300 per person per year. Further, water companies, including Cambridge Water, already offer discounts for those in receipt of certain benefits. Further still, poorer customers will on average use less water to begin with (no irrigating large lawns or filling swimming pools, for example).
And would we need to double the price of water to achieve that level of demand reduction? Also no.
The first demand reductions will come from the marginally least productive uses of water. You’ll stop hosing your lawn before you stop drinking water. And whose use of water is less valuable than household users? Agricultural users.
Eyeballing the charts above, it seems that for Cambridge Water the non-household share of water consumption is around 33%. Presumably this is primarily agricultural use. And the value of water in agriculture is presumably much lower than the value of water for households. Farmers can plant less water-intensive crops while consumers can only economise so much.
This is borne out by greater price elasticity of demand for water in the agricultural sector. Waddams and Clayton cite Wheeler et al (2008)15, which found a price elasticity of demand of around -2.00 for Australian farmers. That is, every doubling (100% increase) leads to a 400% decrease in the quantity of water demanded.
With these elasticity figures in hand, we can work out by how much water prices would have to rise to reduce demand to sustainable levels without supply increases:
Recall Cambridge Water’s supply/demand forecast chart:
I can’t find the underlying data tables in the report nor in the supporting resources, but just eyeballing the graph I estimate average 2019-2030 demand as:
32 Ml/d metered household
15 Ml/d unmetered household
28 Ml/d non-household consumption
15 Ml/d leakage
= 90 Ml/d totalOn the supply side, the 2030 licence capping results in about 75 Ml/d of total water available for use. Assuming no improvement on leakage, this gives us 60 Ml/d of usable water. So we need to reduce demand by 15 Ml/d. Conservatively assuming that no metered properties install meters nor see price rises, and using our elasticity estimates from earlier, how much would prices have to rise to achieve that 15 Ml/d demand reduction? I’m no economist, so I can’t figure out the right equation for this. Instead I tried a few figures using a price elasticity calculator. And the result is that a ~ 30% price increase would be required:
32 metered household at -0.38 price elasticity leads to 29 = 3 reduction
28 non-household at -2.00 price elasticity = 16.4 = 11.6 reduction
= 14.6 Ml/d reductionAt these elasticities, only 20% of the demand reduction would come from households, with 80% coming from non-household sources. (And this is very conservatively assuming no reduction from unmetered households).
The great thing about limiting demand via prices rather than other methods is that the extra revenue can be used to implement supply-side solutions of the sort mentioned above, and to bring down costs in due course.
Conclusion
I’ve heard a lot about water scarcity in Cambridge, but I wasn’t sure how much of the water objection to Cambridge housebuilding was a function of NIMBYish grasping for any tool to wield against development, as opposed to a genuine concern that expanded housing simply will not be viable without severe environmental degradation.
After researching the issue, the water concerns do seem real and grounded. But the best solution does not seem to be to prevent housebuilding. It would be better to reduce demand to accommodate the new residents, and much better still to expand supply to ensure plentiful water for current residents, new residents, and nature alike.
While new supply might take some time to bring online, those who are genuinely motivated to solve the water scarcity problem, as opposed to using it as a convenient anti-housebuilding cudgel, should be concerned to find ways to speed up the deployment of new water resources.
Can we transfer in more water (from Grafham or elsewhere) more quickly? Can we expedite the Fens reservoir so that construction begins before 2029? If that fails, can we build a desalination plant to serve the area before the new reservoir is operational?
All of these options are more proportionate than strangling development in one of the most in-demand areas of the country.
I didn’t know anything about water supply before writing this, so caveat lector.
The draft plan runs to 129 pages, packed full of obscure acronyms, with 32 separate appendices plus extra documentation, meaning it is quite possible that I’ve missed or misunderstood some points. Caveat lector.
Eyeballing the chart, the share of household consumption (ignoring leakage) looks to be around 66%, which is a bit lower than the UK average. (Probably because of the relatively large and presumably relatively water-heavy agricultural industry here.)
Where does the water come from to fill the reservoir? If its filled from the same sources that people are worried about, how does this make things better? I assume that its filled from those sources, but filled in times of high water availability (eg winter) to store for times of lower water availability (eg summer), and thereby don’t deplete natural water resources. But I might be missing something here.
It’s possible that the Lincolnshire reservoir will take longer, but, if so, citing just that figure remains disingenuous. It’s also possible (indeed, very plausible) that the intended start date for the Fen reservoir is overly ambitious, and that, in reality, it will take more like 20 years (ie until 2043) to come online. But, once again, if this is the claim it would be better to say so explicitly.
If you’re interested, you can find water connection charges from water companies. See, eg Anglian Water’s charging documents. Glancing over the summary of charges, a rural connection might cost a few thousand pounds.
Obviously it’s an oversimplification to assume that we can always convert agricultural land values to residential land values. If restrictive planning rules were abolished overnight, the UK’s total land value wouldn’t instantly multiply by the ratios implied here: a lot of residential land value is high precisely because of artificially restricted supply. But this simplification is approximately true for a small region like Cambridge, where building even 250,000 homes wouldn’t really affect the national price of land. Moreover, any local decrease in land value due to increased land supply would be (probably more than) offset by local increases from agglomeration effects. This is why land values in Greater London (or Manhattan, or Singapore) are much higher than those in rural Wales (or upstate New York, or Borneo).
Does transferring water amount to robbing Peter to pay Paul? No. As I explained in section 1, the UK as a whole has plenty of renewable water resources. Transferring it from water-rich areas to water-poor areas will not result in a national water deficit.
Zhou and Tol, ‘Evaluating the costs of desalination and water transport’ (2005) 41 Water Resources Research W03003.
This is gross elevation, not net elevation, to err on the side of caution. The route also involves 1,393m descent, for a net elevation gain of just 183m.
Likewise for iceberg importation :(.
Horizontal: 0.8 * £0.05 = £0.04
Vertical lift: 1 * £0.04 = £0.04 (King’s Lynn route would only require 100m lift).
Total: £0.08 /m3.
(The next-closest side is around Colchester, which would involve c. 300m lift, raising the total transportation cost to £0.12/m3.
Catherine Waddams and Kerry Clayton, ‘Consumer Choice in the Water Sector’ (2010).
Sarah Wheeler, Henning Bjornlund, Martin Shanahan and Alec Zuo, ‘Price elasticity of water allocations demand in the Goulburn–Murray Irrigation District’ (2008) 52 Australian Journal of Agricultural and Resource Economics 37-55.