The Government
You Didn’t Elect

When we talk about democracy, we tend to concentrate almost entirely on elections. Every few years we are presented with a collection of political parties, each publishes a manifesto, we put a cross in a box and the winner gets to form a government. An enormous amount of attention is given to that process, yet remarkably little is given to what happens before it. Where did all the ideas in those manifestos actually come from? Who decided which problems required solving, which solutions were politically respectable and which were beyond consideration?

You probably didn't vote for the Tony Blair Institute for Global Change, Policy Exchange, the Institute of Economic Affairs, the Resolution Foundation or any of the other organisations orbiting Westminster. You certainly didn't vote for the corporations that fund political research, sponsor conferences and participate in discussions with politicians. Yet these organisations inhabit an enormous professional ecosystem that operates every day, regardless of whether an election is taking place. They produce research, formulate policy, employ former government officials, provide experts to parliamentary committees, organise conferences and meet ministers. Some of the people involved move between government and the organisations advising it.

None of this is particularly secret, which is precisely why it interests me.

The conventional description of democracy suggests that the public identifies problems, political parties propose different solutions and voters choose between them. There is some truth to that, but it ignores an important stage in the process. Before most of us ever hear about a proposed policy, it may already have spent years being discussed by academics, charities, pressure groups, civil servants, corporations and think tanks. By the time it appears in a manifesto, much of the intellectual work has already been done.

Policy Exchange provides a useful example because it is remarkably open about its role. Generally associated with the British centre-right, it describes its purpose as developing and promoting new policy ideas. More interestingly, it says that ideas from at least 34 of its papers appeared in the 2024 manifestos of the Conservatives, Labour, Liberal Democrats and Reform UK.

That claim needs treating carefully. Think tanks have an obvious incentive to advertise their influence and an idea appearing in two places does not prove that one caused the other. Political ideas rarely have a single identifiable owner. Nevertheless, four supposedly competing political parties drawing upon ideas that one policy organisation claims to have helped develop tells us something useful. There is a layer of politics that exists before party politics, and the organisations occupying it can potentially influence the menu long before voters are invited to order from it.

The Tony Blair Institute for Global Change is an especially interesting example because it makes few bones about wanting to influence government. Its own website declares that, “TBI doesn’t just join the conversation, we drive it.” It says that its ideas strengthen leaders' decisions, shape debate and inform policy innovation, while its teams work directly with governments around the world.

This was essentially Tony Blair's intention from the beginning. In describing the organisation that eventually became today's Institute, Blair talked about combining policy development with networking and providing practising politicians with a “bigger and better policy agenda”. The Tony Blair Institute isn't an academic department observing politics from a respectful distance. Attempting to influence how governments think is part of its stated purpose.

The access it enjoys is also demonstrable. Official government transparency records show that Pat McFadden met TBI in April 2025 to discuss inquiry reform. He met the organisation again that October to discuss current affairs and, in January 2026, while serving as Work and Pensions Secretary, met Tony Blair personally to discuss public-service reform. These aren't allegations extracted from leaked emails or claims made by anonymous sources. They appear in the government's own records.

The relationship between the Institute and government is not limited to occasional meetings either. Dr Laura Gilbert is currently Senior Director of AI at TBI. Before joining the Institute she was Director of Data Science in 10 Downing Street and led the government's Incubator for Artificial Intelligence. She now sits on the Government Digital Service's Responsible AI Advisory Panel. Again, there is nothing covert about this arrangement. Her government biography says so.

What we're looking at is therefore considerably less theatrical than a secret organisation issuing instructions to ministers. It is a network of expertise, relationships and personnel connecting organisations outside government with the machinery inside it. That distinction matters because the mundane reality may tell us more about how modern political power works than any conspiracy theory ever could.

Things become more interesting when corporations enter the picture. TBI operates something called the Global Network for Change, which it describes as giving members opportunities to connect with influential people and “engage in policy conversations that shape the agenda”. In September 2026, the Financial Times reported that corporate membership begins at $25,000 per year and identified members including BP, Uber, Amazon, Microsoft and AstraZeneca.

TBI says explicitly that it does not lobby governments on behalf of these companies, and there is no evidence that paying membership fees purchases government policy. That qualification is important. What we can demonstrate is access.

At the 2026 Labour Party conference, for example, TBI ran an invitation-only reception attended by Work and Pensions Secretary Pat McFadden and Trade Minister Anas Sarwar. The event was supported by OpenAI. The following morning it co-hosted another invitation-only event discussing Britain's AI infrastructure strategy with Growth Minister Dan Tomlinson, supported by technology company Nebius. Other TBI conference events placed ministers into discussions supported by organisations including Google DeepMind and venture-capital firm General Catalyst.

All of this is openly listed by TBI itself. It doesn't demonstrate corruption, improper lobbying or corporate control of government, and pretending otherwise would weaken the argument. What it demonstrates is that access to the process by which political ideas are discussed and refined is distributed very unevenly.

You and I can write to our MP. Google DeepMind can support a policy discussion attended by a government minister.

Those are both forms of political participation, but it would be rather silly to pretend they are equivalent.

This is where Elite Theory becomes useful, because none of this would have particularly surprised Gaetano Mosca.

Mosca was an Italian political theorist who developed a deceptively simple observation about political organisation. In every sufficiently large society, he argued, a relatively small organised minority governs a much larger majority. Governments and political systems change, but the organisational problem remains. Millions of people cannot personally administer a state, formulate an energy strategy, negotiate international agreements, design a tax system or regulate artificial intelligence. Modern societies therefore require specialists, administrators and organisations.

Organisation itself consequently becomes a source of power.

This doesn't require those people to meet in a smoke-filled room and agree upon a common agenda. In fact, elites frequently compete with one another. Mosca's observation was more fundamental: a relatively small number of organised people will inevitably possess advantages over a vastly larger but comparatively unorganised population.

James Burnham later approached a similar problem from another direction. In The Managerial Revolution, Burnham was interested in the growing separation between ownership and control. Millions of shareholders can technically own corporations while possessing practically no influence over their operation. Day-to-day control belongs instead to managers, administrators and technical specialists who understand and operate the machinery.

There is an obvious parallel with government. We say that elected politicians control the state and, constitutionally, they do. Yet a minister arriving in a complicated government department cannot personally possess the accumulated knowledge of thousands of permanent officials, lawyers, regulators, economists, scientists and technical specialists. Government requires expertise, and around that requirement has grown another ecosystem of consultants, academics, NGOs, corporations, advisers and think tanks supplying information and proposed solutions.

This doesn't render the politician powerless. It simply means that actual political power is considerably more complicated than the diagram we were shown at school.

Interestingly, the British government itself now uses remarkably similar language.

In 2026, former senior civil servant Philip Rycroft completed an independent government review examining foreign financial influence in British politics. The resulting Rycroft Review describes how “the world of politics sits within a rich ecosystem of influencing, lobbying and policy thinking.”

Rycroft doesn't present this ecosystem as inherently sinister. Quite the opposite. Most of this activity is legitimate and can improve government. Politicians cannot possibly be experts in everything, and organisations outside government frequently possess knowledge that the state genuinely needs. Businesses understand their industries, charities understand the communities in which they operate and specialist researchers can spend years studying problems that ministers have only recently encountered.

The review also contains a warning particularly relevant to this subject: think tanks sometimes exaggerate their own influence. If an organisation proposes something and government subsequently introduces a similar policy, it does not necessarily follow that the organisation caused the decision. Several people can independently reach the same conclusion, and politicians themselves are not empty vessels waiting for think tanks to fill their heads.

That is an important limitation because without it Elite Theory can deteriorate into an unfalsifiable conspiracy theory in which every meeting demonstrates control, every similar policy demonstrates influence and every disagreement somehow proves that the conspiracy has successfully hidden itself.

A more useful investigation asks narrower questions. Who repeatedly gets access to ministers? Who supplies government with expertise? Where do senior personnel move when they leave public service? Which organisations are represented on advisory panels? Who funds the organisations producing policy proposals? Most importantly, can particular ideas actually be traced as they travel through these institutions and eventually appear in government policy?

Those questions can be answered with evidence.

Perhaps the easiest way to think about modern democracy is to imagine a restaurant. Every few years we're handed a menu and invited to choose what we want. We spend enormous amounts of time arguing over which meal to order and congratulating ourselves on having a choice.

But somebody decided what went on the menu in the first place. Someone selected the ingredients, somebody trained the chefs and several suppliers have spent years persuading the restaurant that their particular ingredients ought to be used.

That doesn't make your choice meaningless. You genuinely can choose between the meals being offered. But understanding how the restaurant actually works requires occasionally looking inside the kitchen.

Nor is this phenomenon exclusively associated with the political left. That would be a convenient conclusion for an article centred upon Tony Blair, but the evidence doesn't support it. Policy Exchange operates broadly within the centre-right. The Institute of Economic Affairs has spent decades developing and promoting free-market ideas. The Centre for Policy Studies was founded by Margaret Thatcher and Keith Joseph and became part of the intellectual environment from which Thatcherism emerged. The Adam Smith Institute promotes market-oriented reforms. On the other side are organisations such as the Fabian Society, IPPR and numerous institutions historically associated with Labour and the wider left.

These organisations disagree with one another, compete for influence and frequently watch governments ignore their recommendations entirely. The existence of influential elites does not imply the existence of one unified elite, which is something the cruder versions of this argument often miss. Mosca's elites competed too.

The more interesting possibility is that electoral politics represents only the visible end of a much longer process. By the time an idea appears in a party manifesto, it may already have spent years travelling through universities, think tanks, charities, pressure groups, corporations, conferences, parliamentary committees and government advisory bodies. Politicians still choose between those ideas and voters still choose between politicians, but neither group necessarily controls the earlier process through which certain ideas become credible possibilities while others remain outside respectable political discussion.

That distinction between electing a government and producing political ideas deserves far more attention than it receives.

Rather than asking who secretly controls Britain, therefore, I think there is a much better question: who consistently gets close enough to government to help define what government considers possible?

Fortunately, we don't have to guess. Ministerial diaries are published. Employment histories can be traced. Think tanks publish reports. Advisory-panel memberships are recorded. Corporate sponsorship can sometimes be identified. Manifestos can be compared with earlier policy papers, and people moving between government, business and policy organisations leave a paper trail.

We should also record when governments reject the recommendations of supposedly influential organisations, because otherwise we're merely collecting evidence that confirms what we already believe.

Perhaps, after following that paperwork for long enough, we'll discover a healthy and competitive marketplace of ideas. Perhaps we'll discover that a surprisingly small professional class repeatedly circulates through the institutions surrounding political power. The answer may well lie somewhere between the two.

Either way, if we want to understand who governs Britain, watching elections isn't enough.

We need to start watching the people who prepare the menu.

Tony Blair Institute for Global Change, Shaping Policy and Debate.

Tony Blair Institute for Global Change, Global Network for Change.

Tony Blair Institute for Global Change, TBI at Party Conferences 2026.

UK Government transparency records, Cabinet Office ministerial meetings, April–June 2025.

UK Government transparency records, Department for Work and Pensions ministerial meetings, October–December 2025 and January–March 2026.

UK Government, GDS Responsible AI Advisory Panel.

Philip Rycroft, The Rycroft Review, 2026.

Policy Exchange, About Policy Exchange.

Financial Times, reporting on TBI's Global Network for Change, September 2026.

Gaetano Mosca, writings on the political class and ruling minorities.

James Burnham, The Managerial Revolution and The Machiavellians.