What if cars had no number plates?
If the government can’t do identification, it can’t do identity.
Recently I suffered a broken wing mirror while away visiting friends. This isn’t the first time this has happened to me and from experience I thought finding a replacement might be a pain. Fortunately this time it was mostly intact, but the replacement still needed to be the right size and shape for my car. As well as supporting the various lights and gizmos to be found in a modern wing mirror.
But surprisingly it was an easy process. I put my number plate into various websites. Instantly they knew the make, model and year of my car and could tell me what replacement parts were in stock. Replacement ordered, dispatched, delivered and then fitted after a bit of jiggery-pokery. A simple process from end to end.
Systems are all linked together using your number plate, whether buying parts for your car, comparing insurance, checking when your MOT is due or paying your road-tax. It works well and we take it for granted. But not everything works this way.
How would this work if number plates don’t exist? Where there isn’t one unique ID number for your car. Where there is instead a multitude of different unique identifiers: one for your MOT, another for your insurance, another for your road-tax and so on.
When you tax your car, how would DVLA check whether it had a valid MOT?
The road-tax system would only know your road-tax vehicle ID, the MOT system would only know your MOT vehicle ID. The DVLA would need to build a way to match them up, based on storing copies of some of your car’s details in both systems. If enough details match, it can assume a specific car in one system is the same vehicle as a specific car in the other.
But what details to use, to narrow all the possible cars down to one? Make, model, year and colour would clearly not be unique enough. Dates of sale or previous MOT tests would certainly narrow it down. Adding the owner’s personal details, their name, address and date of birth, would probably narrow it down to one in most cases.
Complex IT systems cost money to build, they cost money to run, they cost money to maintain and update over time. Any automated matching service only works if both systems store the same data, and both copies are accurate. One record might have the wrong date for a previous MOT, a spelling mistake in the owner’s name or an out-of-date address. Then the automated system falls down.
This may be resolved by building yet another online service. Asking you to confirm even more of your car’s details, or your personal details. Yet more details that will have to be stored in the system so they can be checked against.
Almost always, some of these cases will have to be handled by a manual, clerical process. Teams of civil servants manning call centres, sending and receiving letters or processing paper forms. And like any manual process, it will create more errors along the way.
However, if no number plates existed it is not just one matching service that would be needed. Road-tax alone would need several more, to check insurance, SORN status and other things. Each with its own copies of common data to match on. And each with more civil servants manning more phones, sending and receiving more letters, creating more errors and ever more cost.
But the cost of all this complexity isn’t just in all the additional things that must be done. It would be in the additional things that are not done, the increased cost of change.
New ideas would need to overcome the inertia of changing complex existing systems, or prove that their benefits outweighed the considerable cost of building new ones. Any bright spark with such an idea, a way to make an existing service more effective or more efficient, would face a much tougher fight to get stuff done. Many would wonder whether the effort of the fight itself was worth it.
Thankfully, in a world of number plates, we avoid the bureaucratic nightmare of countless matching systems. For cars at least.
For people, the nightmare is real. Different parts of government have different ID numbers for an individual. Individuals have National Insurance numbers, NHS numbers, Driving Licence numbers, passport numbers, Unique Learner Numbers, Unique Taxpayer References and armed forces Service Numbers, among others. And these are just the ones we see on the outside; internal systems have their own separate person identifiers. Whenever data needs to be joined up, a matching service must be built. Multiple copies of personal data must be kept in multiple databases across multiple departments. Endless teams of civil servants must be hired to catch when people cannot be matched. Creating endless opportunities for error and endless cost.
We all feel the cost of this in our day-to-day lives. When you move house you need to update the government. You need to update each government service separately. Each of these services needs to build a way for you to update it. Inevitably you’ll miss one, or the update fails to go through. Almost anyone who has lived in a rented flat knows the pile of post that accumulates for long departed former tenants. These may be tax bills from HMRC, court summons or hospital test results. Personal data, leaking out via Royal Mail.
The government has no single source of truth. It doesn’t know who is who. It doesn’t know whether a person is a citizen or not, and if not whether they are a tourist, temporary worker, student or asylum seeker. The problem doesn’t stop at central government; it is repeated in the devolved administrations and in local councils. More matching systems, more errors, more inertia and more cost.
Criticism of the government’s digital identity card proposals focuses on privacy and security. But the risks from digital identity cards are small relative to those that already exist. And these are risks that identity cards alone do nothing to solve.
The government will still need multiple copies of your data in multiple expensive matching systems. As well as thousands of civil servants plugging the gaps when these systems don’t work.
Each system is another point that could be attacked. And the more data it holds, the more value there is in attacking it.
The more data someone has on you, the easier it is to assume your identity. Gaining access to one system can mean access to others. Checking your answers to questions, called ‘knowledge-based verification’, only works if the answers are sufficiently secret. All it takes is for someone else to know the answers, your mother’s maiden name, the name of your first school or how much tax you paid last year. Then that person can assume your identity and gain access to more of your personal information. And every additional bit of knowledge potentially helps them gain even more.
The spectre of ‘Big Brother’ arises whenever identity cards are suggested. Because they’ll make it easier for the government to access your data and link it all together. This is true. But this linking is already happening using expensive and complicated matching systems. There is no legislation controlling precisely what data each public body can hold on you and what they can use it for. It is controlled by general principles, not specific limits. The complexity of the current system makes real oversight and accountability effectively impossible.
There is a solution. A single, unique, trusted identifier for individuals, used across all government departments and public bodies. Eliminating the need for both expensive matching systems and the thousands of civil servants that accompany them.
Knowing your car’s number plate does not allow everyone to find out your name and address. But it does allow the police to do so.
Regulated by statute, a universal identifier doesn’t need to mean universal access. Access can be democratically accountable and controlled. Ensuring that public bodies can only store or access data that is needed for their specific purposes.
The government’s digital ID proposals create their own problems of privacy and security. But they do not solve the significant issues that already exist. We need to solve the problem of identification. Not as a matter of convenience, but of privacy, security, cost, and democratic accountability.