Ownership is not the tax rate
The first axis, and the test that separates a large state from a socialist one.
The commonest error on this axis
Almost every mistaken claim about economic position comes from measuring the wrong thing. The axis does not measure the size of the state, the level of taxation, or the generosity of public provision. It measures who owns productive assets and who allocates them.
Consider two arrangements. In the first, the state owns the mines, railways, banks and steelworks and directs their output by plan. In the second, everything is privately owned and traded, and the state taxes heavily to fund universal healthcare, education and pensions.
The second may have the larger state by every fiscal measure. The first is unambiguously further toward common ownership. A scale built on the size of the state reports these backwards.
The test, in three questions
Who holds title to productive assets? Who decides what is produced and for whom? And can either of those be changed by a market transaction?
Redistribution does not move an organisation along this axis. Taxing income and transferring it leaves ownership and allocation untouched — the assets stay in private hands and the market keeps allocating. That is a substantial commitment, and it is a different one from socialising production.
The anchors
A number means nothing unless the scale is fixed. Here is the economic axis with its five anchor definitions and the real cases that sit at each.
| Score | What it means | Anchor cases |
|---|---|---|
| −10 | All productive assets held in common or by the state; money and market allocation abolished or vestigial. | Khmer Rouge, anarcho-communism |
| −05 | Large-scale redistribution and universal public services inside a mostly private, price-allocated economy. | Nordic social democracy, Die Linke |
| 00 | Genuine mixed economy, or a programme that subordinates ownership questions to something else entirely. | Italian Fascism, Deng-era China |
| +05 | Market allocation is the default; the state corrects failures and funds a safety net but does not direct investment. | Christian democracy, Kuomintang |
| +10 | Private property is near-absolute; taxation, regulation and public provision are treated as illegitimate in principle. | Anarcho-capitalism, La Libertad Avanza |
Two cases that go the wrong way
Sweden is not socialist
No wealth tax, abolished 2007. No inheritance tax, abolished 2004. An extensive school voucher system with for-profit providers. A partially privatised pension system. One of the most open trading economies on earth. Productive assets overwhelmingly in private hands.
What Sweden has is a very large redistributive state on top of a market economy — which is a real and distinctive arrangement, and is not what the ownership test measures.
Vietnam is not communist, economically
Governed by a Communist Party under a constitution naming socialism as the objective. And: collectivisation dismantled from 1986, private enterprise legalised, the private sector now the majority of employment, comprehensive free trade agreements signed.
On ownership it sits near the centre and moving. On the organisation of power it has barely moved since 1975. The label reports one of those and discards the other.
Zero is a position
The commonest misreading of any scored scale is that the midpoint means nothing was found. On this axis zero is a genuine mixed economy — a deliberate settlement in which substantial private ownership coexists with substantial public provision, held by people who believe that arrangement is correct. It was the explicit programme of postwar Christian democracy and it built most of the institutions readers of this live inside.
Now code one yourself
The instrument below draws a cell at random, shows you the axis, its anchors and the evidence attached to that cell — with the score withheld. You assign a score, and then the atlas score appears beside yours.
The rationale is held back until you commit, because most rationales state their own score. The evidence artefacts are shown, because almost none of them do.
Code three cells blind
Open the blind coder and work through three cells. For each one, write your score down before you reveal the atlas score — committing on paper is the whole point, because a score you have not committed to always turns out to have been the right one.
Then record two numbers: how many you were within two points of, and the largest single gap. The first tells you whether the codebook is precise enough to work in your hands. The second tells you where it is not.
If you disagree by more than two points on any cell, the challenge form is the right next step. Thirty scores in this atlas have been changed that way, each with the reason recorded publicly.
This is the only honest way to find out whether the codebook works. If you disagree by more than two points, the challenge form is the right next step — thirty scores in this atlas have been changed that way.
Open the blind coderBrowse the corpusOne or two emails a month, or none at all
Nine free modules, nothing gated, nothing to fill in. The update carries codings that changed, new claim checks, and at least one finding that went against the atlas. Nothing else, and one click to stop.
Site owner: set FORMSPREE_ID in offers.py to activate this form.