---
title: "Purra, ministerial resignations, taxes and opinion polls | Sami Miettinen | Negotiator 203"
summary: "A July 2023 summer episode in which Sami Miettinen goes through the season's surveys alone: MTV's online panel on Riikka Purra's resignation, Jethro Rostedt's and Jessikka Aro's Twitter polls, the Helsingin Sanomat poll and the channel's own votes. This write-up treats the sampling question openly — none of the surveys named is a probability sample — and unpacks the episode's weightiest single claim, the eight billion euro error on the tax rate, so the reader can check it themselves. The episode is the host's own commentary, and that is disclosed."
datePublished: 2023-07-24
dateModified: 2023-07-24
originalLang: en
section: society
sections: ["society","economy"]
authors: ["Sami Miettinen"]
tags: ["Neuvottelija","EP203","Gallupit","Mielipidemittaus","Otanta","Veroaste","Riikka Purra","Politiikka","Sami Miettinen"]
canonical: https://www.neuvottelija.com/ai/ep203-purran-ja-ministerien-erot-gallupit-sami-miettinen/
---
# Purra, ministerial resignations, taxes and opinion polls | Sami Miettinen | Negotiator 203

# Purra, ministerial resignations, taxes and opinion polls | Sami Miettinen | Negotiator 203

> **Summary:**
> A July 2023 summer episode in which Sami Miettinen goes through the season's surveys alone: MTV's online panel on Riikka Purra's resignation, Jethro Rostedt's and Jessikka Aro's Twitter polls, the Helsingin Sanomat poll and the channel's own votes. This write-up treats the sampling question openly — none of the surveys named is a probability sample — and unpacks the episode's weightiest single claim, the eight billion euro error on the tax rate, so the reader can check it themselves. The episode is the host's own commentary, and that is disclosed.

## A note on reading this

This is the **host's own summer episode**, about seventeen minutes, recorded on
**24 July 2023**. There is no guest and no counterparty, so nothing stated in it is
challenged within the episode. Three caveats:

- **The episode is opinionated and concerns day-to-day politics.** This write-up reports
  Miettinen's positions as his and takes no view on them.
- **The surveys are not comparable with each other** — neither in the episode nor here.
  Section 1 sets out why, because that is the episode's whole subject.
- **The source.** A MacWhisper transcript (Finnish, 101 cues). Proper nouns were corrected.
  In a few places the transcript renders a number indistinctly; those are flagged in the
  text rather than guessed.

---

## 1. Four different things, all of them called a poll

The spine of the episode is four surveys, and the differences between them matter more than
their results.

| Survey | Type | N | Result |
|---|---|---|---|
| MTV online panel, 14 July | self-recruited online panel | 1,927 | Purra should resign 47 %, should not 40 %, don't know 14 % |
| Jethro Rostedt's Twitter poll | open, self-selected | 19,100 | should not resign 77 %, should 23 % |
| Jessikka Aro's two Twitter polls | open, self-selected | 10,233 and 7,439 | government is functional 67 % and 69 % |
| Helsingin Sanomat poll | research institute sample | — | support for the governing parties holds |

Miettinen's first observation is **sample size**: MTV's panel had about a tenth of the
respondents that Rostedt's poll did. The second is **direction**: the smaller sample gave a
narrow majority for Purra resigning, the larger ones a clear majority for the opposite. The
third is **weight**: the official-looking survey seems to count for more in public debate
than a Twitter poll two orders of magnitude larger.

**This is an observation about journalism, not about statistics — and it is honest to say
so.** Sample size does not make a self-selected poll representative. The respondent pool of
a Twitter poll is determined by whose followers see the question; Rostedt's and Aro's
followings are politically very different, and it is precisely because **respondents from
both leaned the same way** that this is the episode's most interesting single finding —
stronger than either poll's number on its own. An online panel, meanwhile, is not a random
sample but a recruited group, and its representativeness depends on weightings the episode
does not go into.

The only one of these that is a measurement based on a probability sample is the
**Helsingin Sanomat poll**, and that is the one the episode spends least time on.

**A comparison the episode mentions in passing** and that is worth noting: the same MTV
survey had earlier been run on Pekka Haavisto's resignation during the Al-Hol affair, and
the call to resign was stronger then. Results for two different questions cannot be lined
up without knowing the wording, the timing and the weightings.

## 2. The channel's own polls — and what they measured

Miettinen says he had asked about the shape of the coalition before it was even formed, in
connection with the episode of *Politiikan puskaradio* hosted by **Jari Hanska**. What the
channel's audience wanted:

- **51 %** National Coalition + Finns Party + Centre — did not happen
- **25 %** the current right-wing government — did happen
- **19 %** the same base with Liike Nyt in place of the Swedish People's Party
- the rest (red-blue combinations, National Coalition–SDP–Finns) marginal

This is **the channel's audience**, not the electorate, and the episode says so plainly by
addressing viewers directly ("your favourite government"). What matters for this write-up
is that it measures one audience — and that this audience predicted the outcome reasonably
well as its second preference.

The Helsingin Sanomat numbers as given in the episode: National Coalition **+1.2 percentage
points** from the election result to about 22 per cent, Finns Party and SDP roughly flat (a
few tenths down), the **Centre below ten per cent** after a drop of about a point and a
half, Left Alliance and Greens up about a point, Swedish People's Party and Christian
Democrats stable.

## 3. Ministerial changes in context

Vilhelm Junnila's resignation raised the question of whether it would be the only one.
Miettinen does not believe the government falls over it, and sets the changes against a
benchmark:

> **The left-wing government replaced 23 ministers over its term.**

He breaks down where the number comes from: the switch from Rinne's government to Marin's
produced a large rotation at once, the finance minister changed several times, and on top
of that came ordinary reasons such as Krista Kiuru's maternity leave.

The conclusion cuts two ways. On one hand a single ministerial change is not a dramatic
event against that historical background. On the other — and this is the reader's point,
not the episode's — **the figure of 23 includes a structural change of prime minister**,
which is a different thing from one minister resigning over a controversy. The comparison
is therefore indicative rather than symmetrical.

## 4. The weightiest part: the tax rate and the eight billion

This is the analytically most valuable section of the episode, and it is worth unpacking,
because it can be checked.

**Miettinen's claim has two parts.**

1. The finance ministry's tax rate forecasts, he argues, carry a **systematic downward
   bias**: the forecast lowers the tax rate and the outturn does not. He says he went through
   the method with **Elias Erämaja**, and points to how the deflator and the GDP ratio are
   calculated, and how index adjustments to taxes are handled.
2. The figure repeated in public, that a **44.0 per cent tax rate would produce eight
   billion euros more in tax revenue**, is in his view wrong.

**The numbers in the episode.** The realised total tax rate for 2022 was **43.2 per cent**.
A guest column by **Jaakko Kiander**, cited in Helsingin Sanomat, estimated that 2023 would
land at **41.2 per cent**. Miettinen does not believe in a two-point drop.

**The arithmetic the reader can do.** One percentage point of the tax rate is one per cent
of GDP. Finland's GDP in 2022 was on the order of **270 billion euros**, so a percentage
point is about **2.7 billion**.

- **43.2 → 44.0 is 0.8 percentage points**, roughly **2.2 billion** — not eight.
- **41.2 → 44.0 is 2.8 percentage points**, roughly **7.6 billion** — which is about that
  eight billion.

*A clarification to what is said in the episode.* Miettinen calls the figure "complete
rubbish", and his own calculation matches the first line. But both numbers are
arithmetically correct — they simply start from different baselines. Eight billion appears
if the reference point is the **forecast** of 41.2; two billion appears if it is the
**outturn** of 43.2. This does not refute the episode's claim so much as sharpen it: **the
dispute is not about the multiplication but about which figure is the right starting
level** — and it is exactly that forecast whose validity Miettinen denies. Looking back at
which estimate proved right settles the question without any politics.

The episode's own conclusion is rhetorical and concerns public debate: when a
forecast-based figure is repeated as a quotation, it creates the impression that raising
the tax rate would dissolve the fiscal problem. That impression is built on two different
number series.

## 5. Side threads worth reading

**Finland is a poor Western country.** Miettinen refers to a Helsingin Sanomat piece
reporting that about **20 people** in Finland reached a **cumulative gross income of over
a hundred million euros across 22 years** — around four million a year. His point is the
comparison: Sweden would produce hundreds on the same method. That is his estimate; no
Swedish calculation is presented in the episode.

**What device do you watch on.** YouTube's own analytics said about **18 per cent** watched
on television; the channel's own survey gave a different distribution in which **the phone
was by far the most common**, with the rest splitting as computer 12 %, television 9 %,
tablet 4 %, VR headset 3 %. *The phone's percentage is not given here*, because the
transcript renders it indistinctly. The gap between analytics and survey is itself a
finding: the same phenomenon measured two ways gives two answers.

From it follows a practical production conclusion: watched on a large flat screen, the
weaknesses of the editing show, so resolution and framing need work. Subscriber numbers in
the episode: about **16,500 video subscribers** and about **7,000 audio subscribers**, with
the balance shifting towards video.

**Prigozhin.** The channel's poll on Yevgeny Prigozhin's fate: **51 per cent** thought he
would not be killed. Miettinen bet the other way and put the timing in late 2023. *This is
a forecast made in the summer of 2023 and should be read as one.*

**The grain deal.** After the trilateral Turkey–Ukraine–Russia agreement lapsed, the
audience's overwhelmingly most popular wish was that **Turkey, as a NATO member, would
secure Ukraine's exports with its navy**. Miettinen's own assessment is that Erdoğan would
demand large concessions for it, because the risk would be both political and military —
but his position as master of the Black Sea would be strengthened. The second preference
was concessions to Russia, such as arrangements for its own grain exports and possible
SWIFT relief for the proceeds.

**Guest appearances elsewhere.** On *Puheenaihe* with **Leevi Leivo** and Iltalehti
political journalist **Elli Harju** on the coalition negotiations; an autumn episode on the
central bank crisis and commercial banks was planned for **Arto Koskelo's** and **Jussi
Heikelä's** *23 Minuuttia*. Miettinen mentions having worked at **Credit Suisse** in London.

## 6. The negotiating problem left to the end

The final section is short and it is the only one where the episode touches its own core
subject, negotiation. Miettinen was involved in opening the **Helsinki Stock Exchange
Club** to women members, and he puts the question to himself: why push for that while at
the same time taking apart wokeness on Ivan Puopolo's *Viikon viisasteluklubi*.

The answer is left open in the episode — it is promised to the subscriber side — but two
things are said: the outcome **succeeded**, and what was interesting about it was the
**negotiating technique**, that is, how a majority decision was reached in a short time. As
comparisons he notes that nothing equivalent has succeeded at the women's rights
association Unioni, the Suomalainen Klubi, Svenska Klubben or Handelsgillet.

This is the only point in the episode where the question "what does the public want" moves
from measurement into actual decision-making — and also the only one where the answer was
negotiable rather than measurable.

## Claims presented as claims

- The assessment that the public wants the government and Riikka Purra to continue.
- The claim of a systematic bias in the finance ministry's tax rate forecasts.
- The assessment that the campaign to bring down the government is partly orchestrated.
- The comparison of the number of high earners in Finland and Sweden.
- The forecast about Yevgeny Prigozhin's fate.

## What stays with you

1. **Sample size is not representativeness.** The episode's strongest finding is neither
   Twitter poll's number but the fact that **two politically very different followings**
   pointed the same way.
2. **The tax rate dispute is a dispute about the baseline, not about the multiplication.**
   Eight billion and two billion are both right — from different base years.
3. **The number of ministerial changes does not on its own say much about stability.** The
   benchmark figure of 23 includes a structural change of prime minister.
4. **The same phenomenon measured two ways gives two answers** — as true of viewing devices
   as of the tax rate.

## How the episode runs

- **00:00** — Summer greetings and the theme: opinion polls
- **00:33** — MTV's survey on Purra's resignation and the small sample
- **01:18** — Rostedt's and Aro's Twitter polls say otherwise
- **02:40** — The channel's own pre-election polls on the coalition
- **03:59** — The HS poll: the right-wing government still has support
- **05:12** — Junnila's resignation and the left-wing government's 23 changes
- **06:39** — A wish from the Platinum community: company analyses from smallest to largest
- **07:24** — Prigozhin's mutiny and a bet on his fate
- **08:15** — Guest appearances on Puheenaihe and Credit Suisse
- **09:28** — Finland is a poor Western country
- **10:07** — What device do you watch on? Thinking about 4K
- **11:40** — Helsingin Sanomat's tax rate error and the eight billion illusion
- **13:57** — The Ukrainian grain deal: what you would like to see
- **15:49** — Women joining the Stock Exchange Club as a negotiating problem