Jamie Rumbelow

James Scott argues that the state seeks to control what it can observe. But today’s government is flying blind, unable to make legible the things it wants – or has been asked by the electorate – to control.

This means that the British state does not have accurate answers to many important questions:

• How many people are in the country?
• How many houses were built last year?
• How many houses are left vacant?
• How effective are the charities that receive state funding?
• How much energy do individuals use?
• How many buildings still fitted with the type of cladding that exacerbated the Grenfell fire?
• What is the scale of child sexual exploitation, especially group-based (“grooming gangs”)?

There are lots of well-documented problems with the UK's official statistical body, the ONS. As a result, we don't really know how much the economy is growing or what are the inflation figures really are.

Here are some international examples:

• The DOGE debates reveal, if nothing else, that the federal budget is obscure even to the people responsible for overseeing it
• Lots of other countries struggle with population counts too

Why is this important?

• Democratic legitimacy and trust in government. Decision-makers need to be able to effect the changes for which they are elected.
• Value for money. We are spending a lot of money to operate the state. It would be good to know that money is getting the outcomes we're paying for.
• Quality of governance. Policy should respond to the way the world actually is, not the way the world appears to be at its most legible points
• The private sector depends on reliable state data to plan investment
• There's historical precedent: successful statebuilding efforts in the past are those where state capacity enables observability (Domesday book, Napoleon’s cadastre, 19th century Prussian civil service); deep failures have been where observability was limited (Scott's examples in SLAS, the Gosplan, mid-century US urban renewal à la Jacobs)

What’s going on?

Things we definitely know are problems:

• Measuring these things is difficult
• Legacy IT systems, data silos, devolution, latency, all create issues
• Opacity rents. arbitrage opportunities from informational asymmetries. If charity effectiveness isn't measured properly, organisations can continue receiving state funding regardless of actual outcomes.
• Lack of institutional capacity. Even if the private sector is delivering something, the public sector needs to be a competent client. A lot of the British state lacks good technical capability, the civil service is not structured to reward it, and doing these sorts of measurements well requires technical capability.

More speculative explanations:

• British aversion to state control? Maybe we find it quite satisfying that the state is distant – c.f. the ongoing debate around ID cards – and so this creates political economy problems for fixing it