Why do projects go over budget so often – and what can we do about it?

You don’t have to look far from the ICE’s headquarters in Westminster to spot infrastructure projects struggling to stay on time and budget.

A few yards away, under the Thames, the Tideway project started life with a £2.2 billion price tag. The latest figures suggest the final cost is closer to £5 billion.

Across the river, at HS2’s Euston station, delays and cost overruns are being paid for directly by the taxpayer.

Why does this keep happening?

This was the subject of a recent ICE presidential roundtable where I had the honour of giving the opening remarks.

The questions at hand: why are cost overruns and delays so common in infrastructure projects? And what can we do about it?

Projects can be predictable

Let’s start with the good news: some places and projects are getting it right.

Research increasingly shows that project predictability is not only achievable – it’s already happening.

Road projects in Portugal, Ireland, the UK, and Sweden tend to perform much more reliably than the megaprojects dominating headlines.

And in Hong Kong, the difference is even more striking: infrastructure projects there tend to finish, on average, 15% below their approved final cost estimates.

So, what’s going on?

Smaller, smarter (and more local)

The first factor is scale.

Projects tend to be more predictable when they’re smaller. Smaller projects often mean more of them, which creates learning curves and efficiencies.

They’re also more likely to be delivered by local firms with a better understanding of the environment, a closer relationship with stakeholders, and, crucially, a greater connection to the project’s purpose.

But Hong Kong presents a different story. There, strict controls on cost rises mean that budget increases require formal approval from the Legislative Council.

This is a powerful incentive to start with high estimated costs. The downside? Hong Kong’s construction costs are some of the highest in the world.

This is the trade-off: predictability versus affordability.

What can we do about it?

So, how can infrastructure projects be more predictable and affordable?

At the roundtable, we explored how we might learn from successes and failures. I took away three key themes from our discussion:

Client-ing with focus

Successful projects often rely on small, focused client teams.

Teams that understand production, not just governance. We need less bureaucracy and more boots on the ground that understand how to build during planning.

Owning the production system

Too often, ‘joint'’ project analyses don’t go beyond the top two tiers of the supply chain. Risk isn’t appropriately allocated across the full array of actors involved in project delivery.

We need shared accountability and better discipline. For example, by starting with clear risk-sharing agreements that don’t get rewritten halfway through.

Focus on value, not process

‘Value’ is a word we throw around too easily, but we must take it seriously.

Too many projects get bogged down in managing processes rather than delivering outcomes.

We need to focus on what matters to users and stakeholders, not just what ticks a governance box – and we should also probably reduce the number of boxes that need ticking.

Where the ICE comes in

If we want to fix the problem of predictability and cost in infrastructure, we need to change how the industry learns. That means learning not only within projects, but between them.

The ICE is uniquely positioned to lead this charge by building networks, sharing insights, and providing professional development that enable us to learn faster and build better.

The full roundtable report is available below. I hope it gives you food for thought. Let’s keep this conversation going.

How can we close the infrastructure cost gap?

*The ICE welcomes guests to share their views about infrastructure policy issues on the Infrastructure Blog. These views are the views of the individual.

If you're interested in writing for the Infrastructure Blog, please email [email protected]. The ICE reserves the right not to publish articles that have been submitted.