There are few events in modern British history that have had as profound an impact on the construction industry as the Grenfell Tower fire.
Seventy-two people lost their lives in a disaster that should never have happened. The subsequent Grenfell Tower Inquiry concluded that those deaths were avoidable and identified failures across government, the client, designers, contractor, subcontractors, product manufacturers, certification bodies, Building Control and the London Fire Brigade.
Reform was therefore necessary.
But nearly a decade after Grenfell, I believe the construction industry should be able to ask another question without being accused of questioning the importance of building safety itself:
Are all of the regulations introduced in response actually making buildings safer in proportion to their enormous cost?
That distinction matters.
One of the difficulties of making public policy following a tragedy is that proportionality becomes politically uncomfortable. Measures introduced in the name of safety become extremely difficult to challenge because questioning whether a particular measure is effective can easily be characterised as questioning the importance of safety itself.
They are not the same thing.
Grenfell demanded change. The question now is whether the regulatory system that has emerged has concentrated its resources in the areas most likely to prevent another disaster, or whether parts of the industry have become increasingly occupied with proving that buildings are safe on paper while some of the greatest risks still occur during construction and occupation.
What Grenfell Actually Told Us
It is important to start with what the Grenfell Tower Inquiry actually found, because the disaster cannot credibly be attributed to one simple regulatory failure.
The Inquiry found failings across an extraordinary number of organisations. The aluminium composite material used as rainscreen contained a highly combustible polyethylene core and was primarily responsible for the rapid spread of the fire. Combustible insulation contributed to that spread. The Inquiry found serious failures by the architect, principal contractor, specialist subcontractors, fire consultant and Building Control.
It also uncovered deeply troubling conduct within parts of the construction-products industry.
Celotex was found to have manipulated the way in which a system containing its RS5000 insulation was tested and subsequently marketed the product using a test report that omitted important information about the system actually tested. The Inquiry was similarly damning about Kingspan’s marketing of K15 insulation, describing years of misleading claims about its suitability for buildings over 18 metres.
There were failures of design, specification, product testing, certification, procurement, construction, regulation and management.
The lesson from Grenfell should therefore never be that the old system worked perfectly. It plainly did not.
Dame Judith Hackitt’s subsequent review identified something many people within construction would recognise: fragmented responsibility, poor information management, inadequate accountability and an industry in which responsibility for critical decisions could become dispersed between numerous organisations.
Much of the regulatory response to that diagnosis has genuine merit.
The Golden Thread Is a Sensible Reform
The Golden Thread is probably the clearest example.
Construction projects generate enormous quantities of information, yet historically that information has often been fragmented between consultants, contractors, subcontractors and building owners. Drawings become superseded, specifications change, products are substituted and records disappear when project teams disband.
Years later, somebody responsible for managing the building may have very little reliable information about what was actually constructed.
The Golden Thread attempts to address that by maintaining reliable building-safety information through design, construction and occupation and recording important changes as they occur.
That seems difficult to argue against.
If a fire door is changed, a compartment wall penetrated or a fire-safety product substituted, the people subsequently responsible for that building should be able to establish what happened and why.
The principle is sound because it addresses one of the problems identified following Grenfell: people responsible for buildings frequently did not possess adequate information about them.
The more difficult question concerns what happened next.
Gateway 2: Safety Before Construction
For higher-risk buildings, the Building Safety Act regime introduced a fundamentally different approach to Building Control.
Before relevant building work can commence, detailed information must now be submitted to the Building Safety Regulator at Gateway 2 and building-control approval obtained.
Again, the principle is understandable. It is considerably easier and cheaper to identify a fundamental fire-safety problem before construction than after hundreds of apartments have been built.
But the system has experienced serious difficulties.
Government data for February to April 2026 showed a median approval period across all Gateway 2 categories of around 35 weeks. In London the figure was also 35 weeks, while remediation approvals were taking substantially longer. More recent figures show approval rates improving significantly, which is welcome, but they also demonstrate the scale of the problem the new regulator has been attempting to resolve.
For a major residential developer, 30 or 35 weeks is not an administrative inconvenience.
It can represent millions of pounds of land and development finance continuing to accrue interest while construction cannot begin.
There is also a more fundamental question about where regulatory resources should be concentrated.
Gateway 2 necessarily focuses heavily on design information.
But buildings do not fail because a compliant drawing catches fire.
Buildings fail when the physical building does not perform as intended.
A cavity barrier can be perfectly detailed on a drawing and incorrectly installed on site. Fire stopping can be specified correctly and poorly executed. A compliant product can be substituted. Compartment walls can subsequently be penetrated. Fire doors can be incorrectly installed or inadequately maintained.
Grenfell itself demonstrates the danger of assuming that approval of design information is equivalent to delivery of a safe building.
The Grenfell Inquiry explicitly observed that obtaining Building Control approval should have identified errors in the design and selection of materials before work began, but it did not. It also documented subsequent failures in product selection, substitution and construction.
The question therefore should not be whether Gateway 2 has value. It clearly does.
The question is whether the enormous regulatory resources now concentrated at the pre-construction stage are sufficiently matched by scrutiny of what is actually built.
When Compliance Becomes More Expensive Than Construction
I have experienced another consequence of this increasingly evidence-driven environment on projects myself.
On one relatively small 15-unit residential development, the mezzanine floor incorporated timber joists supported within structural steel I-beams. The timber had been designed to achieve the required fire performance through its sacrificial charring allowance and the structural steelwork was protected with an intumescent coating system.
The floor required 30 minutes’ fire resistance.
From an engineering perspective there appeared to be a reasonable basis for believing the individual elements could comfortably achieve the required performance.
The difficulty was the junction between them.
There was no directly applicable test evidence covering the precise interaction between the timber members and the intumescent-protected steel beam.
That evidential gap required a specialist desktop assessment.
The cost was approximately £15,000.
On a 15-unit development, that is the equivalent of £1,000 per home to assess one construction detail.
Importantly, I am not suggesting Building Control should simply have looked at the detail and said, “that seems safe enough”. Grenfell demonstrated precisely why assumptions about the interaction between materials need to be treated cautiously.
But £15,000 to establish the fire performance of one relatively conventional construction junction illustrates how expensive the evidential burden can become.
I encountered something similar with PIR insulation.
The proposed cavity-wall construction incorporated two layers of PIR insulation. The product itself had appropriate BBA certification, but the certificate did not cover the precise configuration involving two layers installed against one another.
Again, the proposed construction was not necessarily defective.
There was simply an evidential gap.
Further specialist analysis was therefore required to demonstrate that the arrangement would not create an unacceptable interstitial-condensation risk between the layers.
Another consultant. Another report. Another cost.
One £15,000 assessment will not determine Britain’s housing output. Neither will one additional condensation analysis.
But multiply the experience across a development and then across thousands of developments nationally and the economics become significant.
A fire assessment here. A condensation analysis there. Another fire engineer’s report. Another specialist consultant. Another product substitution requiring revised evidence. Another construction junction falling outside the precise scope of the tested assembly.
Each can be individually justified.
Collectively, they increase the cost of building homes.
There is therefore an important distinction between requiring a building to be safe and making the process of demonstrating that it is safe unnecessarily expensive.
The first should be non-negotiable.
The second should always remain open to scrutiny.
Then Came the Second Staircase
The requirement for second staircases provides perhaps the clearest example of the need to discuss proportionality.
From 30 September 2026, Approved Document B introduces guidance for a second staircase in new blocks of flats with a storey 18 metres or more above ground level, subject to transitional arrangements.
It is worth making an important distinction here.
The second-staircase requirement was not itself one of Dame Judith Hackitt’s recommendations. It emerged later as part of the continuing post-Grenfell reform of fire-safety guidance.
There is clearly a safety argument for it. A second staircase provides redundancy if one route becomes compromised, offers an alternative route for residents evacuating and provides additional access for firefighters.
But even the Government’s own analysis recognises that this safety measure has a substantial economic consequence.
Its impact assessment estimated that the preferred second-staircase option would impose approximately £2.4 billion in additional present-value capital costs over the appraisal period.
The same assessment acknowledged another problem familiar to anybody who designs residential buildings: stairs consume valuable floor area.
Staircases, firefighting shafts and associated circulation space reduce the net internal area available for homes unless the footprint of the development can be increased.
On unconstrained sites, perhaps the building simply becomes larger.
On a tight urban site, that often isn’t possible.
Government’s assessment estimated around £170 million in lost value from net internal area during the transition period under its preferred option, with representative affected buildings potentially losing around £600,000 to £2.1 million in value each, depending upon their form.
Most strikingly, the Government’s own consultation material acknowledged that the additional cost would affect development viability and was likely to reduce the amount of affordable housing developers could provide.
That deserves considerably more public discussion.
We have introduced a policy intended to make residential buildings safer while acknowledging that one consequence may be fewer affordable homes.
That does not automatically make the policy wrong.
But it certainly means we should be entitled to ask whether the additional safety benefit is proportionate to the cost.
Why 18 Metres?
There is also an interesting question surrounding the threshold itself.
The Government originally consulted on a 30-metre threshold before ultimately settling on 18 metres. The 18-metre figure aligns with the definition used for higher-risk buildings under the Building Safety Act and also brings the English approach closer to that adopted in Scotland.
That provides regulatory consistency.
It does not necessarily mean that risk suddenly changes at 18 metres.
A building at 17.9 metres does not become fundamentally different from one at 18.1 metres.
Thresholds are unavoidable in regulation, but whenever an arbitrary line produces substantial additional construction cost, lost floor area and potentially reduced housing provision, the evidence supporting that line deserves scrutiny.
The Government’s own impact assessment considered alternatives at 11, 18, 30 and 50 metres. The estimated annual equivalent cost to business varied dramatically: approximately £93 million per year at 50 metres, £292 million at 18 metres and £432 million at 11 metres.
These are not trivial regulatory decisions.
They materially alter the economics of residential development.
One Staircase Does Not Automatically Mean an Unsafe Building
It is also worth remembering that a single-stair residential building is not automatically an unsafe building.
Even the Mayor of London, while supporting the introduction of second staircases, acknowledged that buildings with a single staircase are not inherently unsafe.
Fire safety in residential buildings relies on a system of measures: compartmentation, fire-resistant construction, cavity barriers, smoke control, sprinklers where required, protected escape routes, fire doors, detection systems, appropriate materials and effective building management.
A second staircase adds another layer of resilience.
But the wider lesson of Grenfell is surely that all of those systems have to work in reality, not merely exist on a drawing.
Two staircases do not compensate for failed compartmentation.
A second staircase does not make incorrectly installed cavity barriers acceptable.
An evacuation strategy cannot compensate indefinitely for poor maintenance.
And an immaculate Gateway 2 submission cannot guarantee that the contractor ultimately builds what was approved.
Lakanal House Points Towards the Same Problem
The 2009 Lakanal House fire is instructive for precisely this reason.
Following the fire, considerable attention was placed on compartmentation, fire-risk assessment, refurbishment and maintenance. Subsequent recommendations included ensuring that those carrying out fire-risk assessments had access to information about the building’s design and previous refurbishment, and that personnel procuring or supervising maintenance and refurbishment understood the importance of fire protection and compartmentation.
There was particular concern that compartmentation in existing buildings could be compromised through refurbishment, maintenance or later alteration.
That is fundamentally a building-in-use and construction-quality problem.
A wall designed to provide 60 minutes’ fire resistance only provides that protection if it is built correctly and remains intact.
The drawing can remain perfectly compliant while somebody drills through the wall ten years later and fails to reinstate the fire stopping.
This is why I believe the debate following Grenfell needs to mature.
The question is no longer whether Britain needed stronger building-safety regulation.
It did.
The question is whether the regulatory resources now available are being directed towards the places where they produce the greatest improvement in actual safety.
The Building Control System Is Still Evolving
There is some recognition within government that the system is not yet where it needs to be.
The Building Safety Regulator has been making changes to improve Gateway 2 performance and approval rates have increased. By late June 2026, the BSR reported a 77% approval rate across Gateway 2 categories, while the approval rate for remediation applications had risen to 85%.
That progress should be recognised.
But the latest data also illustrates the extraordinary workload involved. Internal refurbishment work alone represented 63% of the BSR’s live caseload, with a median determination period of 28 weeks.
Meanwhile, an independent panel established following the Grenfell Inquiry reported in May 2026 that Building Control continues to face longstanding problems including fragmentation, conflicts of interest, uneven capacity and inconsistent oversight.
So this is not an argument for simply turning the clock back to 2016.
The old system had serious weaknesses.
The challenge is to create something better rather than simply something more bureaucratic.
What Would Proportionate Building Safety Look Like?
I would retain much of what has changed since Grenfell.
The Golden Thread makes sense. Clear dutyholder responsibility makes sense. Better control of product substitutions makes sense. Improved competence requirements make sense. Proper scrutiny of fire strategy and critical design decisions before construction makes sense.
But those measures should be accompanied by an equally serious focus on construction quality.
More regulatory resources should be directed towards risk-based site inspection of safety-critical work. Fire stopping, cavity barriers, compartmentation, structural fire protection and critical interfaces should be inspected and recorded while they remain visible.
Product verification should ensure that what arrives on site is what was specified and approved.
The Golden Thread should then record not merely what somebody intended to build, but reliable evidence of what was actually installed.
And competent professionals should have proportionate routes for exercising engineering judgement where construction details fall outside an existing certificate or test configuration, rather than automatically requiring developers to commission expensive bespoke assessments for every minor evidential gap.
That would not weaken building safety.
Done properly, it could strengthen it.
Safety and Proportionality Are Not Opposites
Grenfell should never be forgotten, and the construction industry should never return to the culture that allowed the failures identified by the Inquiry to occur.
But neither should Grenfell become a reason why every subsequent building-safety proposal becomes immune from scrutiny.
The Government itself estimates that the second-staircase requirement will impose billions of pounds of additional cost and acknowledges that it may reduce affordable housing provision.
Gateway 2 has produced determination periods measured in months rather than weeks.
Developers increasingly spend significant sums obtaining specialist evidence for details that fall outside the precise boundaries of existing test certificates.
Those consequences may sometimes be justified.
But they should always have to be justified.
There is nothing anti-safety about asking whether £15,000 spent proving one construction junction represents proportionate regulation. There is nothing anti-safety about asking whether an additional staircase provides sufficient additional protection to justify millions of pounds in construction cost and lost residential floor area. And there is nothing anti-safety about asking whether a regulator’s time might sometimes be better spent inspecting critical fire-safety installations on site than reviewing another document before construction begins.
In fact, asking those questions is exactly what good regulation should require.
The lesson from Grenfell should not simply be that Britain needs more building-safety regulation.
It should be that Britain needs better building-safety regulation: regulation that understands how buildings are designed, how they are procured, how they are constructed and how they are subsequently maintained.
Because ultimately residents are not protected by the number of documents uploaded to a regulatory portal.
They are protected by the building around them actually performing as it was designed to perform.