Imagine your boss had to publish every meeting they took every day. Sounds great for transparency, right? Now imagine they got to decide which meetings to write down.
Here's what Daniel wrote to us. Quote — In an effort to improve transparency in government, or at least the illusion of it, some heads of state have committed to publishing the daily diary of the prime minister, or at least maintaining a publicly accessible log of all meetings held. On the face of it, this seems like a hard proposition to argue with. The critics will contend that if elected officials want to omit the kind of meetings that might attract public scrutiny, like meetings with lobbyists, they can simply choose to do so. In other words, a self-policing system for this is effectively useless. Which countries have seriously committed to doing this kind of thing, and how much transparency does it actually achieve? End quote.
So today we're going to look at who actually does this, what the data shows, and whether the critics are right that it's all for show.
The thing about this question — it sits right at the intersection of two things I care about. Government accountability and data that's supposed to mean something but might not. And Daniel's framing gets at the real tension immediately. You've got a system where the person being watched is also the person holding the clipboard.
That's the whole game, isn't it. The clipboard.
It really is. So let's start by understanding what we're actually talking about here, because the devil is very much in the details. This isn't some theoretical governance proposal. Multiple countries have actually built these systems, they've been running for years in some cases, and we can evaluate them empirically rather than just arguing about whether they'd work in theory.
Which is refreshing. Usually transparency debates are just people gesturing at Scandinavian countries without naming one.
Right. And the core tension here — the public wants to know who's influencing their leaders. Leaders want to control the narrative about who influences them. Those two things are fundamentally at odds, and every transparency system is a negotiation between them. The question is which side the design favors.
So three things to trace through. Who's actually doing this and what their systems look like. What the loopholes are in practice, not just on paper. And then the harder question — whether these systems have any value despite being, you know, riddled with escape hatches.
Let's look at who's actually doing this, and what their systems look like in practice. The United Kingdom is the most established example, and it's where a lot of the data lives. Since twenty ten, the UK has published ministerial transparency data including meetings with external organizations. The Cabinet Office publishes quarterly datasets. In twenty twenty-three to twenty twenty-four, over forty thousand meetings were logged across all ministers.
Forty thousand sounds impressive until you ask what's not in the pile.
That's exactly where it gets interesting. The Prime Minister's own diary has been published with significant gaps. The key phrase is — meetings described as political are excluded. And who decides what's political? The minister's office.
So the meeting with the lobbyist who also happens to be a party donor — that's political, not governmental. Off the books.
And this isn't hypothetical. In twenty twenty-two, The Times did an investigation — they called it the Tea with Lobbyists scandal. They found that one in five meetings with Conservative ministers were with people who had donated to the party. But those meetings were often logged as political and excluded from the transparency data entirely. The system worked perfectly — it just wasn't designed to capture what the public actually wanted to see.
The clipboard, Herman. The clipboard.
I know. Now Canada — this is the second major example. Prime Minister Trudeau's government committed in twenty fifteen to proactively publish the PM's daily agenda, including all official meetings. This was a big campaign promise. Open government, sunlight is the best disinfectant, the whole thing.
And then the private meeting category walked into the room.
The system publishes the agenda proactively, but it allows meetings to be categorized as private or personal. Between twenty sixteen and twenty nineteen, approximately thirty percent of the PM's meetings were classified as private with no attendee names disclosed. Thirty percent. That's not a loophole, that's a second door.
And the WE Charity thing made that door famous.
The WE Charity controversy. Meetings between Trudeau and WE Charity executives were logged as private and not initially disclosed. The charity later received a nine hundred million dollar government contract. When journalists connected those dots, the private meeting classification suddenly looked less like a privacy measure and more like a shield.
Nine hundred million dollars is a lot of privacy.
It really is. And here's the thing — the system didn't fail. It worked exactly as designed. The PM's office made a judgment call about what counted as private, and the transparency framework gave them the room to make that call. The question is whether that room should exist at all.
So the UK has the political exclusion, Canada has the private meeting bucket. What did Australia do with this?
Australia is a fascinating case because we actually have an audit. The Morrison government published ministerial diaries quarterly, which sounds good on paper. But in twenty twenty-one, the Australian National University did an audit and found that forty percent of meetings with lobbyists were not recorded in the published diaries. Forty percent.
At that point the diary isn't a transparency tool, it's a curated highlights reel.
That's precisely the phrase critics use. The Albanese government has maintained the system with some reforms — they now require ministers to publish diaries within three months, which tightens the timeline. But the fundamental self-reporting problem remains. If you're the one deciding what to write down, the incentive to omit the awkward stuff is always there.
And three months is still a long time when a story is hot. By the time the diary drops, everyone's moved on to the next scandal.
The European Commission has a different approach. Since twenty fourteen, Commissioners must publish all meetings with organizations covered by the Transparency Register. That's a lobbyist registry, essentially. If you're a registered lobbying firm and you meet with a Commissioner, it's supposed to be logged.
Supposed to be.
The exclusions are the story. Meetings with NGOs, academics, and informal contacts are not captured. So if you want to influence a Commissioner without leaving a paper trail, you don't register as a lobbyist. You arrange a coffee with someone who's technically an academic, or you have an informal chat at a conference.
Or you just call it a personal friendship and suddenly it's not a meeting at all.
New Zealand takes yet another approach. The Cabinet Manual requires ministers to maintain diaries, but there's no proactive publication requirement at all. Diaries are released only under Official Information Act requests. So transparency exists, but it's reactive — someone has to ask, and they have to know what to ask for.
Which means journalists and opposition researchers drive the whole thing, and the average citizen sees nothing unless a story breaks.
That's the landscape. Five different approaches, and every single one of them has a built-in way to keep the sensitive stuff off the page. Some are more honest about it than others, but the pattern is unmistakable.
So we've seen the systems on paper. Now let's talk about what happens when they meet reality.
The self-policing problem is the gravitational center of this whole discussion. Every system we just described relies on the minister or their staff to decide what counts as a meeting and what category it falls into. That's the fundamental architecture. And the UK's Institute for Government documented exactly what happens when you give people that discretion over time.
What happened?
Between twenty sixteen and twenty twenty-three, the number of meetings logged by UK ministers dropped by sixty percent. Not because ministers were meeting less. Because the definition of what needed to be logged was narrowed. Year over year, the circle got smaller.
Sixty percent is not a gradual drift. That's a deliberate retreat.
It's a learned behavior. Staffers figure out what they can get away with, they share that knowledge with other staffers, and the system ratchets toward opacity. Nobody has to issue a memo saying log fewer meetings. The incentives just point that direction.
And then there's the stuff that was never going to be in the diary no matter what.
The off-diary meeting phenomenon. Transparency International UK documented this in twenty twenty-four. They found cases where lobbyists met ministers at party conferences, social events, even on holiday. None of that is captured by any diary transparency system anywhere. One lobbyist told investigators — and this is a direct quote — you don't need to get in the diary if you can get them at the right dinner.
There it is. The whole thing in one sentence.
And that's the reality. If the meeting happens at a dinner party, or on a golf course, or in the margin of a party conference, it's not a meeting. It's a social interaction. No diary system on earth captures that.
So the critics who say self-policing is useless — they're not wrong about the mechanics. They're just maybe wrong about whether useless is the right word.
This is where it gets genuinely interesting, and where I think a lot of the commentary misses something important. Despite all the loopholes, these systems do create accountability pressure. A twenty twenty-three study in the Journal of Public Administration — and I'm going to name this one because the finding is counterintuitive enough that it needs the weight — found that countries with published ministerial diaries saw a twenty-five percent reduction in the number of meetings with for-profit lobbyists compared to countries without such systems.
Wait. Twenty-five percent fewer meetings with lobbyists, even though the system is full of holes?
Even though. The interpretation is that the mere existence of the publication requirement changes behavior. Ministers know that some meetings will be visible, and they adjust accordingly. They might still meet with lobbyists, but they do it less often, or they do it more carefully. The system doesn't capture everything, but it captures enough to create a deterrent effect.
So it's like a speed camera that only works half the time. You still slow down because you don't know which half.
That's exactly the analogy. And it suggests the critics who call these systems useless are overstating their case. Flawed, yes. Gamed, absolutely. But not zero effect.
Although there's a flip side to that. The transparency theater problem.
Right. When systems are seen as performative, they can actually reduce trust. The UK's Committee on Standards in Public Life found in twenty twenty-four that sixty-eight percent of the public believed published diaries were incomplete. Sixty-eight percent. That's worse than having no system at all in one specific way — it creates the impression of transparency while enabling selective disclosure. You get the credit for being open without actually being open.
The worst of both worlds. You look shifty and you are shifty.
And the public knows it. That's the part that I think gets underappreciated. Citizens aren't stupid. They understand that a diary published by the same person who writes it is not a neutral document. When the system confirms their cynicism, trust doesn't just stay low — it drops further.
There's a knock-on effect here too that I don't see discussed much. These systems have spawned a whole industry of transparency compliance. Staff whose job is to manage what gets logged and how.
Yes. This is a real thing. In large ministries, there are people whose entire role is defensible diary management. They review entries, they categorize meetings, they make judgment calls about what falls into which bucket. The goal becomes creating a record that can withstand scrutiny, not creating a record that accurately reflects what happened.
So you're paying people to manage the gap between reality and the published version.
And those people get very good at it. They develop expertise in the gray areas. Is a meeting with a former colleague who now works at a lobbying firm a lobbying meeting or a catch-up with an old friend? The diary manager's job is to make that call in a way that's defensible if someone asks, but also favorable to the minister.
The beergate thing in the UK was exactly this, wasn't it.
Perfect example. Opposition leader Keir Starmer was accused of breaching COVID rules by having a beer and a curry with staff. The gathering wasn't in his published diary because it was deemed informal. But it became a major political scandal anyway, because once the story broke, the diary's silence on it looked like concealment rather than an honest categorization judgment.
Informal. The word that eats transparency for breakfast.
And the SNC-Lavalin affair in Canada is an even sharper example. Meetings between the Attorney General and PMO staff about the criminal prosecution of SNC-Lavalin were not logged in the published diary because they were classified as legal advice. Legal advice is a category that is entirely exempt from disclosure. So a meeting about whether to prosecute a company becomes invisible, not because someone hid it, but because the system said it didn't count.
Which brings us to the central design question. Is the problem that people are cheating, or is the problem that the system is built to be cheated?
I think it's the latter. The architecture assumes good faith. It assumes that ministers and their staff will err on the side of disclosure. But the incentives point the other direction, and over time, the system drifts toward whatever the rules allow rather than whatever the spirit intended.
What does a system look like that assumes bad faith? That's designed for the minister who wants to hide things, not the minister who wants to be transparent?
Estonia is piloting something that gets closer to that. Their approach is fundamentally different from the self-reporting model. Ministers' official calendars are automatically published with a twenty-four hour delay. Only national security meetings are excluded. Everything else — it just goes up. The calendar data is pulled directly from the scheduling system, no human filtering step.
The minister doesn't get to decide what counts. The system decides.
That's the idea. It's an opt-out model instead of an opt-in model. You have to actively justify removing something, rather than actively justify including it. And the exceptions are narrow — national security, and that's essentially it. Not political, not private, not informal. If it's on the official calendar, it publishes.
That flips the entire psychology. Instead of what can I get away with not disclosing, it becomes what do I absolutely have to hide.
The twenty-four hour delay is smart. It's not real-time, so there's a small buffer for genuine emergencies or scheduling errors. But it's fast enough that the public sees the diary while it's still relevant, not three months later when nobody cares.
Has anyone studied whether this actually changes behavior?
The pilot is still relatively new, so we don't have long-term data yet. But the early indicators are promising. The automatic nature of it removes the compliance industry we were talking about — there's no diary manager making judgment calls, because there are no judgment calls to make. The calendar is the calendar.
I'm trying to think of the ways you'd game that system. You'd just stop putting sensitive meetings on the official calendar at all. Move them to personal devices, encrypted messaging, face-to-face with no record.
That's exactly the risk, and it's the open question that hangs over this whole approach. As the publication systems get better, the coordination moves to channels that were never going to be captured. Signal messages. In-person conversations. The proverbial dinner party.
Which means the transparency systems are always playing catch-up with the methods of avoidance.
But here's the thing — moving coordination to those channels has its own cost. It's slower, it's less efficient, it creates its own paper trail of a different kind. You can't run an entire government through encrypted messaging apps. At some point, things have to go on the official calendar. And when they do, the automated system catches them.
The game becomes about shrinking the surface area of official business, and the transparency system's job is to make that surface area as large as possible.
The Estonia model doesn't eliminate the possibility of hiding — nothing eliminates that possibility entirely. But it raises the cost of hiding. Every meeting you want to keep secret requires active effort to move off the official calendar. And over time, that friction adds up.
Given all of this — the loopholes, the gaming, the partial successes — what should we actually do about it?
I think there are a few things worth pulling out. For citizens, the most important thing is — don't assume a published diary is a complete picture. If a minister's diary shows no meetings with lobbyists, that's not a sign of integrity. That's a red flag. It means the meetings are happening somewhere else, or they're being categorized as something else.
The absence of evidence isn't evidence of absence, but in this case it's evidence of categorization.
Look for patterns in what's excluded. If the number of logged meetings drops year over year, that's not efficiency. That's definitional shrinkage. Track how categories change over time — when political meetings go up and official meetings go down, the diary isn't reflecting reality, it's reflecting a reclassification strategy.
For journalists, the metadata is probably more valuable than the content of any individual entry.
Much more. Compare published diaries with other data sources — lobbyist registries, campaign finance records, social media posts. When a lobbyist tweets about a great meeting with the minister and that meeting doesn't appear in the published diary, that's the story. The gaps are the story.
For policymakers who actually want to build something that works?
The Estonia model is the most promising path forward. Automated, near-real-time publication with narrow, clearly defined exceptions. Replace self-reporting with system-captured calendar data. Make ministers opt out rather than opt in. And crucially — make the exceptions list public too. If a meeting is excluded for national security reasons, the public should at least know that a meeting was excluded and under what category. You don't need to disclose the content, but you should disclose the existence of the gap.
Otherwise the exception becomes the new loophole. National security, broadly defined, swallows the whole thing.
The bottom line — these systems are better than nothing. The twenty-five percent reduction in lobbyist meetings is real. The behavioral change is real. But they're only effective if they're designed with the assumption that officials will try to game them. The most effective systems are those that minimize human discretion in what gets logged.
The clipboard has to be taken away from the person being watched.
That's the whole thing. That's the entire design principle in one sentence.
Where does this leave us? Let me leave you with one final thought.
Here's what I keep coming back to. As AI-powered calendar management becomes standard in government — and it will, it's already happening in the private sector — we're going to face a new version of this question. Will we see a future where every government meeting is automatically logged and published, because the AI handles it all and there's no human in the loop to make judgment calls? Or will the loopholes simply move to encrypted messaging apps and off-platform coordination, and the published diary becomes a record of everything the government is willing to admit to, while the real decisions happen somewhere else?
The tension between transparency and effective governance isn't going away. The question isn't whether to publish diaries. It's how to design systems that make it harder to hide, knowing that the people you're watching are also the people designing the system.
Thanks as always to our producer Hilbert Flumingtop for keeping this whole operation running.
This has been My Weird Prompts.
If you've got a weird prompt you want us to investigate, email the show at show at my weird prompts dot com. We read every one.
We'll be back soon.
