Chapter 06 of 13

Suspect 6: Government overspending

The charge is unusually broadly supported. Viðskiptaráð Íslands ran the headline "Ríkið kyndir undir verðbólgu" — "The state is fanning the flames of inflation" — as the title of a formal consultation to the Althingi finance committee on 13 December 2021, before the rate cycle had even properly begun. They repeated the line almost verbatim in every subsequent umsögn on the budget bill, most sharply in May 2024: "Útgjaldavöxtur síðustu ára hefur kynt undir háa verðbólgu og valdið bæði heimilum og fyrirtækjum búsifjum" — "Spending growth in recent years has fuelled high inflation and caused hardship for both households and firms." The Samtök atvinnulífsins (SA) employer confederation ran a parallel line continuously from 2022, most recently through deputy CEO Anna Hrefna Ingimundardóttir in December 2025 calling the 4.5% December print "afleit þróun" driven by the government's tax and spending mix. Fjármálaráð, the constitutionally non-partisan fiscal council, reached the same conclusion from its independent perch on 30 April 2024: "Vaxta- og verðbólgustigið bendir til að þensla sé enn til staðar. Útgjaldavöxtur síðustu ára er ósjálfbær." Miðflokkurinn leader and former prime minister Sigmundur Davíð Gunnlaugsson supplied the populist version on the Althingi floor on 5 February 2026, framing inflation as the product of a government that blames everyone but itself while raising "prices, fees and taxes" in lockstep with the firms it chides.

The highest-status voice belongs to the central bank itself. Peningamál 2024/2, published 8 May 2024, formally lists fiscal slack as an upside risk to inflation: "efnahagsumsvif gætu vaxið hraðar og verðbólguþrýstingur því verið meiri ef slakinn í aðhaldi hins opinbera reynist meiri en grunnspáin gerir ráð fyrir." The MPC statement in the same issue put recent wage agreements and fiscal-policy measures on the same line as the two forces whose demand effects had not yet fully materialised. The Seðlabanki is, on the record, endorsing the hawk charge.

Five voices — Chamber of Commerce, Confederation of Employers, Fiscal Council, central bank, opposition — saying Iceland's inflation is fiscal in origin. As broad as Icelandic political economy gets. It deserves to be tested.

The prima facie case

The case is real. Iceland's pandemic response was large by every available measure. The general-government deficit swung from −1.4% of GDP in 2019 to −8.8% in 2020 and −8.0% in 2021, cumulatively 4.7 percentage points deeper over those two years than the euro-area aggregate (−7.0 and −5.1). The fiscal rules were suspended in 2020 and remained suspended through 2025 under a series of extensions. The Ministry of Finance appropriated emergency wage-support schemes, tourism-rescue loans, and business grants on a timeline measured in weeks. Iceland's two-year cumulative borrowing requirement ran at 16.8% of GDP, one of the deepest Nordic outlays of the episode.

The Grindavík volcanic sequence began on 10 November 2023 with the first Svartsengi intrusion and has continued in cycles since. The state bought essentially the entire housing stock of the town through Þórkatla ehf, a dedicated SPV, at a gross cap of 61 billion ISK. It built defensive walls around the Svartsengi geothermal plant, kept Almannavarnir civil-protection operations running at emergency intensity, and paid wages to displaced workers through a state scheme. The IMF's 2024 Article IV puts direct Grindavík property purchases and volcanic spending at roughly 1.2% of GDP; Hagstofan's own reconciliation puts the full 2024 Grindavík-related outlay above 1.9%.

And on 7 March 2024 — the same day the Breiðfylkingin coalition signed the Stöðugleikasamningurinn — the government unveiled the Vaxandi velsæld welfare package, headline value 80 billion ISK over four years, with increased child benefits, expanded housing benefits, free school meals from autumn 2024, and a one-off vaxtastuðningur interest-rate relief payment for indebted households.

On the headline balance, this looks like a four-year fiscal loosening that has never fully reversed. Iceland's 2024 general-government deficit came in at −3.7% of GDP, wider than the euro-area aggregate of −3.1%. On that single number, the hawks' case stands up.

The decisive test

What the headline balance cannot tell you is how much of any given deficit is an active policy choice and how much is cyclical or one-off. For that you need the cyclically-adjusted primary balance — the deficit stripped of the business cycle and of interest payments — which is the measure central banks, the IMF and the OECD use when they want to judge whether fiscal policy is leaning into the cycle or against it. OECD Economic Outlook publishes it for every member.

Cyclically-adjusted primary balance, % of potential GDP. Source: OECD Economic Outlook.

The shape is unambiguous. Iceland ran a deeper pandemic impulse in 2020–2021 — the hawks are right about that window, and the gap to peers was genuine. Then from 2022 onward Iceland consolidated harder and faster than any peer in the panel. The underlying stance moved from −2.7% of potential GDP in 2021 to +0.6% in 2024, a 3.3-percentage-point tightening in three years. The euro area managed 1.6 points over the same interval. By 2023, the year Iceland's core-inflation gap to the euro area peaked at +2.7 percentage points, Iceland's cyclically-adjusted primary balance was 1.6 points tighter than the euro area's. By 2024, with the Grindavík response in full flow, it was 1.8 points tighter. Iceland was the fiscal disciplinarian of this panel from 2022 onward, not the laggard.

And the Seðlabanki knows it. The same institution that flagged fiscal slack as an inflation risk in its May 2024 Rammagrein had published the retrospective accounting in the previous issue. Peningamál 2023/4, Chapter III: "Slökun á aðhaldi ríkisfjármála, mæld sem breyting í hagsveifluleiðréttum frumjöfnuði, er áætluð að hafa numið samanlagt 4,9% af landsframleiðslu á árunum 2020-2021 … Það gekk að nokkru leyti til baka í fyrra þegar aðhald ríkisfjármála jókst um rúmlega 3% af landsframleiðslu og útlit er fyrir að það aukist um 1,3% til viðbótar í ár." The 4.9% of GDP pandemic loosening was partly reversed by a 3% of GDP tightening in 2022 and a further 1.3% in 2023. The central bank's own numbers — in the same publication where it warns about fiscal slack — describe an active consolidation. The hawks and the CBI both believed fiscal was too loose. The CBI's own cyclically-adjusted measure showed Iceland was consolidating harder than the euro area through exactly the years the inflation gap opened. Both things cannot be true; the quantitative side wins.

The 2024 decomposition

The 2024 headline balance of −3.7% is the main plank of the late-cycle hawk argument: on the face of it, Iceland is loosening again. It isn't, once the components are named.

• Grindavík direct cost: roughly 1.9 percentage points of GDP (Hagstofan and Ministry of Finance reconciliation). Þórkatla property purchases 51.5 bn ISK, defensive walls at Svartsengi 7.2 bn, wage support for displaced workers 9.2 bn, Almannavarnir civil-protection operations 4.5 bn, remainder scattered. The IMF uses a narrower 1.2pp figure covering the property piece only.

• Vaxandi velsæld incremental 2024: roughly 0.4 percentage points. The 80 bn ISK headline is a four-year pledge, not a single-year number; the incremental 2024 cost is closer to 20 bn, with the single largest line a one-off vaxtastuðningur interest-rate relief payment that does not repeat.

• Underlying: roughly −1.4% of GDP — materially tighter than the euro-area aggregate of −3.1%.

Strip Grindavík out — not because a natural disaster is discretionary, but because the question on the table is whether Iceland's fiscal stance was chosen to be expansionary — and the 2024 number is a continuation of the 2022–2023 consolidation. The IMF's 2024 Article IV reaches the same conclusion in so many words, calling the stance "broadly neutral."

One further detail kills what remains of the demand-impulse argument. Of the homeowners compensated through Þórkatla, approximately 90% rolled the proceeds directly into a new primary residence. The Grindavík buyback rotated a housing stock across owners; it did not inject a flow of consumption spending. The state paid households for their homes and those households handed the money to developers selling other homes. At the household level it was a balance-sheet transaction, not a spending boost. The disaster appears in the fiscal accounts as 1.9% of GDP of outlay; its demand effect on the broader Icelandic economy is close to zero.

The energy-mix contrast

There is one further comparison that makes Iceland's position look tighter still. Peer economies ran loose in 2022–2023 specifically to shield households from an energy-price shock — Germany's Strompreisbremse and Gaspreisbremse, France's bouclier tarifaire, Italy's bonus energia, the UK's Energy Price Guarantee. These were large, named, budgetary programmes explicitly framed as inflation-compensation, and they are the reason a significant share of the euro-area expenditure expansion in 2022–2023 appears in the accounts as discretionary fiscal loosening.

Iceland ran none of them, because Iceland did not need to. The domestic electricity system is 100% renewable; residential heating is geothermal; the petroleum import bill is small and did not spike in the domestic retail market the way gas bills did across mainland Europe. The 2022 Iceland HICP housing-and-utilities component (CP04) ran at +6.0% year-on-year against a euro-area aggregate of +17.5% — a gap of eleven and a half percentage points driven almost entirely by the absence of the gas-price shock. Iceland got the cheapest possible pandemic-era pass on the biggest single inflation shock of the decade.

This makes the head-to-head comparison harsher on Iceland in principle and somehow still exonerating in practice. The Icelandic fiscal stance was already tighter than the euro area on a cyclically-adjusted basis and was doing none of the energy-subsidy work that peer stances were doing. Same headline number; different composition; the Iceland number is pure underlying stance. The European number has a large energy-shield wedge inside it. Properly adjusted, Iceland was not loose relative to peers; Iceland was, from 2022 onward, the tightest underlying stance in the Nordic cluster.

A small second Grindavík channel

Grindavík shows up in the CPI in one place the rest of this piece does not examine: home insurance. The global catastrophe-reinsurance market hardened sharply through 2023 and 2024, and Icelandic insurers — Sjóvá, VÍS, TM — passed the cost through to domestic premiums. Sjóvá's earned premiums grew +11.6% in 2023 and a further +7.4% in 2024; ROE climbed from 11.6% to 20.7% and then 17.5%; combined ratios sat at 94–97% against a Nordic peer benchmark of 82–86%. The insurers ran near break-even on underwriting while pushing through double-digit premium hikes to cover the reinsurance bill.

In the CPI this lands in sub-component CP121 insurance — weight 0.97% of the basket, running at 7.8% / 8.2% / 7.7% across 2023–2025. It is the only line inside the CP12+CP13 "miscellaneous services" bucket still running above 5% in late 2025. The archival 4-digit split confirms the composition: housing insurance (IS1242) 8.1% → 4.8% → 9.1% and vehicle insurance (IS1244) 7.8% → 10.6% → 6.9%. The 2025 re-acceleration in housing-insurance inflation lines up with the timing of Grindavík reinsurance renewals hitting the books.

The direct contribution to headline CPI is roughly 0.08 percentage points per year — small at the headline level, but the transmission is mechanical and worth naming. A natural-disaster shock to a global reinsurance market passed through domestic insurance pricing into the CPI, and from the CPI into every CPI-indexed mortgage in Iceland via verðtrygging. Grindavík produced a second, smaller inflation channel that bypasses the fiscal balance entirely and runs through the accomplice identified in Suspect 3. The volcano cost the taxpayer 1.9% of GDP through the budget and added a small persistent drip to headline CPI through the insurance market and then propagated that drip economy-wide through indexed contracts. The same machinery that converts a housing-rent shock into household debt also converts a reinsurance shock into household debt.

What survives

Suspect 6 is cleared on the level charge. Iceland's fiscal stance was not loose during the period the inflation gap opened, and on a cyclically-adjusted basis it was materially tighter than the euro area from 2022 onward. The CBI said fiscal was an upside risk to inflation; the CBI's own accounting showed Iceland in active consolidation. The IMF said 2024 was neutral. The 2020–2021 pandemic stimulus was real and larger than peers — and it was fully unwound by 2023 on the underlying measure. The hawks correctly identified a loosening. They then mislabelled it as an ongoing condition when the data showed it had reversed.

What survives is narrower and more interesting. The hawks are half-right, and the half they're right about is not the half they themselves emphasise. Iceland did spend at considerable scale through the pandemic, and some of that spending did leak into the very market that would later produce the persistent inflation — the capital-area rental market and the new-build queue. The mistake was not the magnitude of the outlay; the mistake was where it landed. Pandemic fiscal capacity that could have gone to scaling stofnframlög, buying land in the capital area, or directly capitalising non-profit housing operators went instead to transfers, business support, and wage shields. The money was spent. It did not build houses.

That is a composition critique, not a level critique, and it connects Suspect 6 to the retrenchment analysis later in this piece. Iceland's fiscal story is not "the state spent too much." It is "the state spent, at considerable scale, on everything except the constraint that would later bind." Real per-capita general-government spending on housing and community amenities fell 26% over the decade through 2024 while the population grew 19%, and the non-market rental waiting list for working-age low-income tenants doubled. The same fiscal apparatus that could mobilise 4.9% of GDP of emergency pandemic stimulus was the apparatus shrinking its housing line in real terms throughout. The 80 bn ISK Vaxandi velsæld package in 2024 is the purest distilled form of the error: the state taking on the distributive work of a housing shortage the state had declined to fix on the supply side, by writing cheques to renters instead of building units for them.

Suspect 6 walks on the direct charge and is booked as a supporting witness on the configuration charge. The problem was never the size of Iceland's deficits. The problem is what the deficits bought.