Street lighting in suburban London: a parable for digital government

The London history website LCC Municipal has a story about the history of street lighting in the south London borough of Croydon.¹ For reasons I’ll explain in a minute, it serves as a parable for digital government and civic use of technology today.

Croydon is London’s most populous borough and, due to its location at the southernmost edge of Greater London, was absorbed into the wider city relatively late in 1965. Before that, it was a ‘county borough’, a more autonomous form of government, within the county of Surrey.²

As the LCC Municipal article explains, in part due to this autonomous status, Croydon built and operated an extensive power grid at the turn of the last century, as well as several large power stations, and that grid survived until very recently:

County Boroughs were imperious beasts which enjoyed far more clout and responsibility than other forms of local government such as municipal boroughs and urban/rural district councils. The original “Croydon A” power station on Purley Way was built by Croydon Corporation and opened in the late 1890s. One of its functions was to generate electricity for lighting the streets, and it is this early 1900s network, much amended and added to, that remained in operation across large parts of the Borough until very recently.

The operation of utility networks by municipalities was a political issue in the 1900’s. The Metropolitan Water Board had been setup in 1903 to take over the operation of the water system in London from an adhoc, often unsafe and unaccountable set of private companies. The government commission that led to the water board’s creation recommended that:

[The water supply should be] brought periodically and automatically under the observation of Parliament, and that it should include delegates of the Local Government Board, in order that the influence of the executive government should be continuously felt.³

And, in 1905, the Fabian Society was arguing for further public control of the new electrical power networks, and the risks of private ownership of critical civic infrastructure:

If gigantic and tyrannical trust monopolizing the production and use of electric power are not to dominate our children as the railway companies dominate us, we must see that the community secures at the outset effective and systematic control over the new force.⁴

As such, Croydon wasn’t being eccentric in running its own power network. Other local governments did the same — Fulham and Hackney councils both opened power stations in 1901 and operated for many decades before being absorbed into the wider, national system.⁵ ⁶ The scale and independence of the network in Croydon ensured greater longevity. As a nationwide system slowly emerged, Croydon remained a network within a network, such that, according to the LCC Municipal article, in 2011 it still ran 70% of the borough’s street lights (albeit minus the original power station, which closed in 1973).⁷

The result was that when a contractor was secured to decommission the network they faced a complex task:

Like the dendritic pattern of ice fracturing across a frozen pond, the CCS had spread throughout the Borough by way of a complex system of branches and spurs. Its piecemeal development was poorly documented and therefore not well understood …. What was clear is that not all street lights were equal. Across the network were “feeder” lamp posts upon which other lights depended for their power supply. … Mapping the CCS, and identifying these feeder lights was at the heart of the complexity.

That upgrade was part of a 25 year-long ‘Private Finance Initiative’, under which the Swedish company Skanska will operate the lighting system for both Croydon and neighbouring Lewisham until 2036.⁸ (Private Finance Initiatives are long-term contracts where private firms deliver and operate public projects while theoretically keeping the debt off the public books.)⁹

The final aspect of the LCC Municipal article that is relevant to our parable is about something less tangible than physical infrastructure, it’s an almost whimsical, sense of place:

Street lighting is a functional thing, there to guide us and keep us safe — there is no question that it is right to keep it current and effective. We shouldn’t subordinate our very real needs to nostalgia. But I also think street lighting plays a (probably small) part in creating the identity of a place … Cross a borough boundary and the street lighting was one of the things that changed. Although uncommon now, lamp posts were often painted in Corporation colours — in Croydon’s case, silver (the colour of childhood spaceships) was entirely in keeping with the town’s 1950s futuristic vision.

While it’s hard to disagree that this is a nostalgic viewpoint, there is a sense of place and of understanding that comes from the small changes in the built environment. I’d challenge anyone to walk from Clerkenwell in the City of London, through the Inns of Court (technically separate local authorities), across Kingsway to Covent Garden without spotting the boundaries through the subtle changes in streetscape. Nostalgic or not, the visual differences in the built environment are things that people care about, and for reasons that are sometimes hard to articulate. How else do you explain a blog that documents the ‘Bollards of London’, including the silver Pegasus of the Inner Temple and the Griffins of The City?¹⁰ Or the #GovBins project which documents the design of wheelie bins around the UK?¹¹

People attach meaning to design and place. Even if, as London blogger and journalist Peter Watts puts it, psychogeography is a way for people to make “grandiose statements about people, place and history that you are fairly sure won’t stand up to any great inquisition but look fucking brilliant on paper”, it “doesn’t mean there’s not a lot of it about”.¹² So, why not space-age lamp posts as municipal markers?

I said this was a parable about digital government, so what lessons can we draw?

Commoditization and fast followers

The first is about the commoditization of technology.

Researcher Simon Wardley talks about technologies evolving along a spectrum from an original idea (genesis), through custom-built or bespoke products until the technology ultimately becomes a well-known commodity.¹³ Today a local government organization building it’s own power station and separate power grid seems like madness. That’s because, today, power is clearly something best treated as a commodity. But at the turn of the 20th century, it was far from that. There was no National Grid for Croydon’s government to buy electricity from.

Any form of electrical lighting had, until recently been seen as a novelty — used to attract shoppers to the likes of Electric Avenue.¹⁴ Mass, public electrical lighting was a new phenomenon, although by this time the technology was enough of a known quantity, and the public need clear enough, that it made sense for a local government like Croydon to be engaged in building electrical lighting networks and power stations. In this sense, Croydon was displaying ‘fast follower’ behaviour in the adoption of technology. As David Eaves and Ben McGuire from the Harvard Kennedy School have pointed out, this is quite a good strategy for governments to adopt:

government doesn’t need to be at the bleeding edge of technology to provide valuable services to constituents. Leaders should instead invest in the skills and organizational flexibility that will let them be fast followers of proven private-sector practices and get smart about how to integrate technologies into efficient, effective, equitable governance.¹⁵

The development of power networks is also an example of how new technology can change the work of government. The move from ad hoc experimentation in the private sector to local government demanded new capabilities. The move from local provision to a standardized national grid shows how, as technology evolves and standardizes, some things are better done at the national level than the local.

Public ownership and accountability

The debate from the early 20th Century about issues of safety, exploitation, ownership, and the new power dynamics of municipal utilities map pretty well to today’s debates around the ethics and accountability of digital platforms. Take for example the connected sensor networks of the LinkNYC and InLink programs in New York and London. As the campaign group Rethink LinkNYC point out, although often described as a public good made available for free, there is a hidden cost:

The city boasts that there is zero cost to taxpayers. That is, however, quite misleading. The business model of these kiosks is advertising revenue generated by drawing on the data of people who use the wi-fi, and through bluetooth beacons, sensors, cameras, and location services.¹⁶

With the exception of groups like Rethink LinkNYC, the idea that the ownership and accountability of digital infrastructure is a political question is rarely acknowledged. It represents a contested zone where the interests of private companies and the different public institutions play out. As a society, who do we want running digital services and infrastructure? Where are the risks of new defacto monopolies emerging? Or of the disintermediation of democratic institutions in the name of innovation?

Some cities are beginning to answer these questions. For example, Cambridge City Council in Massachusetts has voted to require council approval of all new ‘surveillance technology’ by city agencies and police. ¹⁷ While it doesn’t extend to use of data from all public services, this represents a democratic body putting public oversight in place to ensure that digital infrastructure will be operated in the public interest.

Design and the separation of powers

The move towards single digital platforms like GOV.UK bring with them a consistency in design that aims to ensure people don’t need to understand the structure of government to complete a given task.

From a utilitarian point of view, this makes sense. People shouldn’t have to understand the internal structure of government to claim welfare, pay taxes or apply for a fishing license. But what, as with the design of lampposts being real-world markers of the boundaries of the civic world, might we be losing in the process?

Jeni Tennison, now CEO of the Open Data Institute, recognized this question earliest, writing this shortly after the launch of GOV.UK:

the relentless focus on user needs leading to a future of identikit pages, with no individuality, no character, no clue that behind these pages — which, remember, under the Single Government Domain policy becomes the single authoritative view, the site that represents the department on the web — is a living and breathing institution that manages hugely important parts of our lives¹⁸

The complexity of government is abstracted away in the name of functional expediency. The institutional character subsumed.

I’m not sure we can (usefully) talk of anything as pretentious as psychogeography of digital services, but what will the homogenization and intermingling of services from different parts of government do to how people understand and act towards their government? Can we have utility and understanding?

One recent example from the UK offers further concerns that it might. There has been considerable effort to move court proceedings online, as part of GOV.UK. One of the services being put online allows people to appeal decisions by the Department for Work & Pensions (DWP). DWP’s services are also on GOV.UK, with the same branding. Lawyers have raised concerns about this and led to some small design changes to try and make clear that the court system is separate.¹⁹

This raises some big questions. How should the separation of powers work in a digital government? How should it be communicated to the public, and can it be done without confusing them? What are the units of government that people need to understand, and how can the seams between them be clear?

Just like the lamp posts, what subtle design elements might help people orientate themselves, or give them an ambient awareness of how their government is organized?

Transformation and lock-in

Finally, the story above is an example of why changing existing systems is just harder than the development of new ‘green field’ projects. The loss of institutional knowledge about how the original system worked and its inherent complexity made it hard to replace.

This is precisely the problem that many developed economies face when trying to digitize their services. They are rarely building from scratch; they are generally replacing decades worth of legacy technology.

Croydon’s solution to that problem has left it locked into a 25-year contract. This, presumably, means some of the novel uses of lamp posts that other cities are experimenting with, such as connected environmental sensors, will remain an impossibility for the foreseeable future. San Diego’s street lights, for example, detect gunshots, monitor pollution and spot empty parking spaces.²⁰ Although, given the privacy implications, maybe Croydon will be safer without.

Either way, the point stands that even seemingly known technologies can change and long lock-ins to suppliers come at a risk.

As ever, we’ve been here before

Public electric lighting (and other civic utilities) represented the adoption of new technologies in the civic domain. With them came approaches to adoption, and questions of ownership and accountability. Once entrenched, came those of transformation and lock-in. They changed the work of government and demanded new institutions to govern them.

These are many of the same issues that governments central and local are facing as they digitize their public services. As they look to solve them, they should look to their own history as much as the promises of sales-people or innovation specialists.

This article is part of the Platform Land project led by Richard Pope, a senior fellow at the Harvard Kennedy School digitalHKS project.

  1. LCC Municipal, “Silver Service”, LCC Municipal blog, 2nd August 2017,
  2. Croydon Council, “Timeline of Croydon Borough”, Croydon Council website, retrieved 28th January 2019,
  3. Metropolitan Water Board, “London’s water supply, 1903–1953: A review of the work of the Metropolitan Water Board”, 1953, London, Staples Press. p.12
  4. S G Hobson, “Public control of electric power and transit : Report of the Committee of the Society appointed to consider the Control of Electrical Power and Transit”, 1905 , London, Fabian Society, p.4–5
  5. “Hackney Power Station”, Wikipedia, retrieved 28th January 2019,
  6. “Fulham Power Station”, Wikipedia, retrieved 28th January 2019,
  7. “Croydon power stations”, Wikipedia, retrieved 28th January 2019,
  9. Croydon Council Streets & Environment Scrutiny Sub-Committee , “2014, Street Lighting PFI”, 2014,
  10. “Bollards of London”,
  11. Peter Watts, “New London writing, or What the fuck is psychogeography anyway?”, The Great Wen – A London Blog, 18th February 2014,
  12. #govbins,
  13. Simon Wardley, “Evolution”, Bits or pieces, 4th January 2013,
  14. Mike Urban, “The rise and fall of Electric Avenue, Brixton”, Brixton Buzz, 22nd August 2013,
  15. David Eaves & Ben McGuire, “The Fast-Follower Strategy for Technology in Government”, Governing, 27th August 2018,
  16. “Quick Facts”, Rethink LinkNYC,
  17. Kade Crockford, “Cambridge City Council Passes Surveillance Oversight Ordinance in Unanimous Vote”, ACLU Massachusetts, 11th December 2018,
  18. Jeni Tennison, “Precious Snowflakes”, Jeni’s Musings, 10th March 2012,
  19. Joshua Rozenberg QC (hon), “Justice Online: Are We There Yet?”, 21st February 2019,
  20. Tekla S. Perry, “San Diego Installs Smart Streetlights to Monitor the Metropolis”, IEEE Spectrum, 1st January 2018,