power & institutions · 3 min
Can we still trust the institutions?
Trust in the people who run things sits at record lows. Yet the planes land, the courts sit, the money clears. What actually broke?
The collapse is real — but it’s local, political and mostly at the top. What died globally isn’t the machinery. It’s the mystique.
Our observer finds a species that built extraordinary cooperation machines — courts, parliaments, central banks, universities, a free press — now reporting historic levels of distrust in them. And then it spots the oddity: the same humans keep boarding the planes, accepting the verdicts, taking the medicine and using the money. They distrust the machines while betting their lives on them, daily.
Here’s the honest layout: the three stories told about the great trust collapse, the analyst who says the institutions never changed — our view of them did — and what the numbers actually show.
What you're usually told
They failed us first
Populists of both flavours — and much of the public
Distrust isn’t a mood, this camp says: it’s a memory. Iraq, the 2008 crash nobody went to prison for, the expenses scandal, Partygate — and in Britain the Post Office and infected-blood scandals, where institutions didn’t just fail but closed ranks for decades. The record lows are rational updating: 45% of Britons now say they’d ‘almost never’ trust a government of any party to put country before party — the highest ever recorded — and 58% say politicians in a corner won’t tell the truth.
The trust recession is overstated
The OECD and most political scientists
Look past the shouting and the state you actually touch is doing fine: across the OECD, two-thirds are satisfied with public services, trust in the police, courts and civil service runs well above trust in national government — and the headline number has stabilised at about 40%, improving in half the countries where it had fallen. The deficit is specific, not general: it’s politicians, parties and parliaments people have downgraded. A politics problem, this camp says — not an institutions problem.
They were captured from within
Conservatives and the heterodox centre
Institutions spent the century’s first decades trading neutrality for mission — universities, newsrooms, public-health bodies and HR departments started taking sides, and the public noticed. The proof, this camp argues, is that trust now swings with the jersey: in Gallup’s 2025 US read, 80% of Republicans expressed confidence in the presidency against 3% of Democrats, with the military swinging twenty points by party. Referees who play don’t get trusted as referees — and shouldn’t be surprised when the crowd stops clapping.
The contrarian read
The ‘nothing broke — you just got glasses’ read
Martin Gurri (ex-CIA media analyst), The Revolt of the Public
The analyst who called this earliest argues the institutions didn’t decay — the distance did. Twentieth-century authority rested on an information monopoly: the institution spoke, the public couldn’t check. The digital tsunami ended that, and every ordinary failure — always there — became visible, recorded and shareable. A networked public now punishes the gap between the institution’s story and its performance, without offering a programme of its own. On this read you can’t restore what’s been lost, because what’s been lost is the mystique, not the machinery — the only way forward is institutions that promise less and show their working.
Our read
Most likely: not a general collapse — a targeted strike on the political core of rich democracies, while the everyday machinery stands (and quietly keeps being used).
Our confidence: that the fall is real but concentrated — high. That it amounts to civilisational trust collapse — low (≈20%). Read from surveys, not prophecy.
Start with the shape the headlines miss. Globally, confidence in armies, courts, banks and elections is flat or higher than twenty years ago — the military polls at a median 72%, banks at 63% — and OECD-wide trust in national government has stabilised rather than slid. The genuine, grinding declines concentrate in a handful of rich democracies — the US, Britain, France — and, within them, in the political core: parties, parliaments, the press. Britain’s record 45% ‘almost never’ sits beside courts and police that most people still trust, and services most people still rate. That split — the nurse trusted, the minister not — is the single most consistent finding in the field.
But concentrated is not harmless. Edelman’s 2026 read finds seven in ten people now unwilling or hesitant to trust anyone with different values or politics — the distrust has gone horizontal, person-to-person, which is harder to fix than a bad government and feeds the vote-them-all-out volatility now routine across the West. And the two camps are both half-right about cause: the failures were real (the first camp’s receipts are genuine), and the exposure is new (Gurri’s point) — visibility outran performance. The honest scoreboard from here isn’t the trust question in isolation; it’s whether the distrust of the political core starts contaminating the parts still standing — the courts, the elections, the money. Mostly, so far, it hasn’t.
- ≈55%Two-speed trust — the everyday state stays trusted and used, the political core stays distrusted and noisy, and anti-incumbent landslides become the pressure valve.
- ≈25%The rot spreads — in one or more major democracies political distrust reaches the courts, elections or currency, and the two-speed settlement breaks (the US is the live test).
- ≈20%Delivery rebuild — governments that visibly deliver, open their working and give people a real say claw trust back, as it already has in about half the OECD countries where it fell.
How sure are we?
‘Trust in institutions’ bundles wildly different things: the same person who’d almost never trust a party leader still trusts the fire brigade, the courts and the nurse. Before believing any collapse headline, ask which institution, where — the global military/courts/banks numbers are flat or up over twenty years, and the collapse is concentrated in the political core of a few rich democracies. One more honesty note: autocracies reporting high trust in government partly reflects the risk of saying otherwise.
Where this comes from
- OECD (2026) — Survey on Drivers of Trust in Public Institutions — 2025 results, 38 countries The stabilisation case: 40% trust national government, services satisfaction at two-thirds, police/courts/civil service trusted more than politicians — and 31% feel people like them have a say.
- NatCen (2024) — British Social Attitudes 41 — Damaged Politics? Britain’s records: 45% ‘almost never’ trust governments to put country first; 58% say cornered politicians won’t tell the truth; 79% want the system of governing improved.
- Gallup (2025) — Global Confidence in Institutions: A 20-Year Review The against-the-panic numbers: global confidence in militaries (72%), banks (63%), courts and elections flat or higher than two decades ago — with the US, UK and France the exceptions, not the rule.
- Edelman (2026) — Edelman Trust Barometer 2026 — Trust Amid Insularity Where the distrust went next: seven in ten now unwilling or hesitant to trust someone with different values or politics; only 32% think the next generation will be better off.
Go deeper — read the book
The ex-CIA media analyst who saw it first: a 2014 diagnosis of why networked publics revolt against every institution — that has kept coming true ever since.
The book that named this crisis. If you read one on why authority stopped working, it’s this.
Written with AI and editorially curated. Talescout always labels AI text (EU AI Act, Art. 50). Numbers are considered judgements, not measurements.