Three Paths, Not Two – Corrections and No is the right path

Smári McCarthy wrote a detailed rebuttal to my post of 10 August, and this week he made a wider version of the same argument in Vísir, responding to Carl Baudenbacher. Both are worth reading before this.

He is right about several things, including some errors of mine. I’ll deal with those first, because I don’t think you should have to take the rest on trust.

The corrections

The 0.08% figure was wrong. I wrote that membership buys Iceland “roughly 0.08% of the vote.” That is Iceland’s share of the EU’s population, not its weight in any institution. Six seats in a 726-seat Parliament is 0.83%. One Commissioner of 28 is 3.57%. In the Council, a qualified majority needs 55% of member states as well as 65% of population, and Iceland would be one of 28 states on the first leg. The number made my case look stronger than it is.

I contradicted myself on sovereignty. I wrote that critics of “fax democracy” are right, and four paragraphs later that Iceland holds “100% of the national control.” Both cannot be true. The correct, narrower claim is that Iceland retains full authority in the domains outside the EEA — fisheries, agriculture, trade policy, the currency.

I left out the capital controls, which ran from November 2008 to March 2017, in an essay about sovereignty. That omission is indefensible and I make it good here. I don’t accept the inference, though: Cyprus imposed controls in 2013 and Greece in 2015, both inside the euro. Controls are what happens when a banking system fails, not a feature of small currencies.

And I should have named who paid for the 2008 adjustment. It ran through a fall in real wages of roughly an eighth and a rise of about a third in indexed mortgage principal. Calling that a shock absorber without saying whose shoulders absorbed it was too clean.

One thing I don’t concede, and it is the central one. Smári’s headline figure — 598 basis points nominal, 323 real — is the wrong measure of what the euro delivers. A short-rate differential contains four things: the inflation gap, the equilibrium real rate gap, the cyclical stance of policy, and the currency risk premium. The euro removes the last of these. It does not remove the others. Iceland’s higher equilibrium real rate reflects higher trend growth, higher investment demand, a younger population and less capital per head, and Frankfurt changes none of that. A euro-area real short rate of minus 0.44% is largely an artefact of a decade at the lower bound — a symptom of chronic demand deficiency, not a prize on offer.

The number I used is not mine. Chapter 3 of Currency Matters — the 219-page review chaired by Catherine L. Mann for the Ministry of Finance and Economic Affairs — finds that euro adopters saw cost-of-capital reductions in a range of 25 to 100 basis points, with an Iceland-specific estimate of around half a percentage point from the Central Bank’s own 2012 modelling. The report then pairs that gain with redenomination risk, and the Greek, Italian, Spanish and Portuguese spreads of 2012 are what redenomination risk looks like when it arrives.

So the government’s own experts put the prize at roughly 50 basis points for Iceland, bought at the price of a tail risk you can no longer devalue your way out of. My own independent estimate is somewhat higher — 50 to 150 basis points on sovereign and large-corporate funding, more for startups financed abroad. Either way it is a modest, one-time convergence. It is not 323, and it is not 598.

Where he is right, and it matters

His strongest argument is not about numbers. It is this: the status quo is not a neutral default. It is a path. Iceland has been walking it for thirty years through the EEA — continuous adoption of EU law, no seat, no vote — and treating that as “standing still” is a mistake.

I accept this completely. Anyone arguing No who pretends the EEA is a resting state is not arguing seriously.

But the conclusion doesn’t follow. “The status quo is a path” does not give you “therefore the only other path is accession.” That is a trichotomy presented as a dichotomy, and the missing third option is the one Iceland has never actually tried: the EEA plus the domestic reform we have spent five decades postponing.

The asymmetry nobody is pricing

If we are going to argue about path dependence, the operative variable in path dependence is exit cost. On that measure the two paths are not comparable, and the Yes case gets its symmetry by never mentioning exit at all.

Article 127 of the EEA Agreement: a contracting party withdraws on twelve months’ written notice. Article 50 of the Treaty on European Union: ask Britain what that cost, and how long it took.

Both are paths. Only one has a door.

There is a related point about control. Iceland holds reservation rights under Article 102 of the EEA Agreement over what gets incorporated into the Annexes, and has effectively never used them. “We have no real control” is not quite true. “We have declined to exercise the control we have” is true, and it points somewhere different — at reform, not at accession.

On security, the EU adds nothing

Smári bundles four risks together — the EEA fraying, the American security umbrella, market access, and the arrival of the next windfall — and asks the No side to defend all of them at once.

Take security on its own. Iceland’s defence rests on NATO membership since 1949 and the 1951 bilateral agreement with the United States, which remains in force even though American forces have not been permanently stationed here since 2006. Add the Joint Expeditionary Force, which Iceland joined in 2021, and NORDEFCO, which is explicitly a capacity and cooperation framework rather than a collective defence commitment.

None of this is on the ballot. None of it changes on 29 August either way. What the EU offers on top is Article 42(7) of the Treaty on European Union — weaker than Article 5, with no independent force projection into the North Atlantic behind it.

Iceland is not choosing between security and insecurity. That leg of the bundle is already carried, and it is not carried by Brussels.

What “distraction” actually means

I have called the Yes campaign a distraction, and I owe you the object of that sentence, because a vague accusation is worthless.

The binding constraints on Icelandic prosperity are domestic and they are nameable. Verðtrygging, which is a legal construct, not a monetary one, and which Frankfurt cannot touch. The statutory real return assumption embedded in the pension system, which sets a floor under the cost of capital by law. A banking sector concentrated in three institutions. A pension system holding assets near 200% of GDP with almost nowhere domestic to put them, and a listed market too shallow to absorb them. None of these is caused by the króna. None is fixed by the euro.

Read Currency Matters itself rather than the press release about it. The Ministry’s statement said the costs of the króna likely exceed its benefits. The 219 pages underneath are scrupulously a trade-offs document, and they do not recommend adopting the euro.

One finding in particular has gone almost entirely unquoted. The report’s variance decomposition concludes that shocks originating in the króna itself explain very little of the variation in Icelandic output, inflation or the policy rate — that the currency’s volatility is largely self-contained. Sit with that. The instability Icelanders actually feel does not come from the currency. It comes from external shocks and from our own domestic structure: wage growth running ahead of productivity, indexation, procyclical fiscal policy. Abolishing the króna removes some noise. It does not touch the sources.

And the report says the quiet part directly: the decisive reforms — wage-setting, indexation, fiscal discipline — are required under any currency regime. That is not my argument imported into their document. It is their conclusion.

The obvious reply is that we can do both — vote, negotiate, and reform in parallel. In principle, yes. In practice, reform requires a constituency, and every person in Iceland who might have formed one has spent this year arguing about Brussels instead of about indexation. That is what a distraction is. Not a thing that is false. A thing that is true and consumes the attention required for the thing that matters more.

The windfall argument, and why I reject it

Smári describes Icelandic prosperity as a sequence of windfalls: Marshall plan, herring, aluminium, banking, tourism. It is a good line. It is also question-begging, because the word “windfall” smuggles in the conclusion.

The hydro build-out that made aluminium possible took decades of deliberate capital allocation. The Keflavík hub was a strategy, executed. The quota system, whatever you think of its distributional consequences, made this fishery one of the most profitable on earth by design. Calling these lottery wins presupposes that Iceland has no agency — which is precisely the thing in dispute.

And it sits badly with his own earlier argument to me, which was that pension assets near 200% of GDP show a mature, capital-rich economy. That is not the profile of a country waiting for its next ticket.

What I am voting for

Not against Europe. I want the tariff-free access, the research programmes, the mobility, the standards. We have them.

I am voting for the third path, on 29 August, because I think it is the one nobody has tried and the one that is actually ours to walk. Iceland can fix indexation. Iceland can reform the pension reference rate. Iceland can break the banking concentration and build a market with enough depth to keep its own capital at home. None of that requires anyone’s permission, and none of it is on the ballot.

These posts were drafted with AI assistance. The way the first one failed — symmetrical structure, confident figures, precision on one side and adjectives on the other — is exactly what Smári described. Checking whether the shape was true was my job. The errors are mine.

I am voting No. Not because standing still is an option — it isn’t — but because the path I want us on is the one neither campaign is offering.