Norges Bank — Policy Preview

Decision due: Thursday, 13 August 2026, 10:00 CEST · Current policy rate: 4.25%

1. Executive summary

The Committee is expected to leave the policy rate at 4.25%, and the July inflation print has all but settled it. Core CPI-ATE held at 2.7% against a 2.9% consensus — a downside surprise, and the second consecutive print at 2.7% after June came in at the same level. Monetary Policy Report 2/2026 said "the four-quarter rise in the CPI-ATE is expected to remain above 3% in 2026"; that is an annual profile, which a single monthly reading below 3% would not falsify, but two consecutive months at 2.7% make the annual average materially harder to reach. Market-implied odds of a move on 13 August fell from 25% to 6% on the release. The June assessment that "it will likely be necessary to raise the policy rate further at one of the forthcoming monetary policy meetings" still stands, and the June path sits a little higher than March's at just above 4.5% at end-2026, so a 25bp move remains embedded and undelivered. But it is no longer this meeting's business. The question on 13 August is whether the tightening language survives an inflation profile running below the forecast that produced it.

2. The call

Hold at 4.25%. Surveyed economist consensus is 4.25%, and market pricing for a hike at this meeting is around 6%, cut from 25% before the 10 August inflation release — an analyst estimate off NOK forward pricing rather than an in-house curve, dated to that release. Conviction: high on the rate, low on the language.

The case for the hike being deferred rather than abandoned is the path. June's is a little higher than March's and sits just above 4.5% at end-2026 — one more 25bp move — and three meetings remain after this one to deliver it: 24 September, 5 November and 17 December, of which September and December carry a new Report and a revised path. The Committee also raised the rate at the 6-7 May interim meeting with no Report and no new path, so an interim move is demonstrably within its habits when the data call for it.

The data no longer call for it. Core inflation has undershot both the consensus and the Report's own above-3% profile, and the krone is firmer than the assumption the June statement recorded. The live question is therefore the statement, not the rate: whether the June assessment is repeated intact, softened toward a conditional, or dropped for a data-dependent framing. A hold with the language untouched would read hawkish against 6% pricing; a hold that dilutes it would confirm the September round as the decision point and is the outcome most likely to move the krone.

3. The committee

Lean labels summarise RoboMacro's read of each member's public communications over the past six months, from Hawkish to Dovish. A lean is only assigned where recent remarks carry a monetary-policy stance: 1 of 5 members currently qualifies (Bache).

Norges Bank decides by consensus and publishes no votes. The May record states the Committee "decided unanimously"; for June no dissent is recorded, which is a weaker claim than unanimity and should not be read as one. Neither the June statement nor Monetary Policy Report 2/2026 attributes a view to any member or group of members — the Report is signed by all five without differentiation — so the public record offers no way to read the Committee's internal balance beyond the chair.

The Governor is therefore the only member whose stance can be read with any confidence, and hers has been consistently inflation-vigilant through 2026. Longva carries no current stance: his one scored intervention in the window, the 25 February response to the Norges Bank Watch report, reviewed the 2025 easing rather than signalling a direction for 2026, and a retrospective assessment of a cycle the Committee has since reversed is not evidence of a lean. Bjørnland took Ingvild Almås's seat on 1 January 2026 for a term running to end-2029 and has not spoken on policy since joining.

4. Data since the last decision

No new release since the last decision for: Mainland GDP (QoQ), Real GDP growth (QoQ), Output gap — the committee sees the same print(s) it saw last time. Market rows are measured from the 18 Jun close — the last observation common to every market series on or before the last decision. WTI crude oil (USD/bbl) traded an inter-meeting high of 93.08 on 23 Jul and a low of 68.30 on 5 Jul.

Data notes. Consumer confidence: our vendor's scaling does not reconcile with the Finans Norge/Verian series other providers carry — read the direction and the change, not the absolute level. Registered unemployment: NAV unadjusted; we do not hold the seasonally adjusted variant. House prices: the prior in that row is May, not June, so the delta spans two months. Market rows: all as of 7 August, the latest common observation date across the FX and oil series.

The July inflation print is the decisive input and it came in soft. Core CPI-ATE was unchanged at 2.7% against a 2.9% consensus — a 0.2pp downside surprise on the measure the target is defined over. It is the second consecutive print at that level: June also came in at 2.7%, itself below the run of readings that preceded it. That pairing is the argument. MPR 2/2026 projected the four-quarter rise in CPI-ATE to remain above 3% through 2026, an annual profile that one sub-3% month would not overturn; two consecutive months at 2.7% make the 2026 average materially harder to reach, and it is that arithmetic, not the level of any single print, that markets repriced. Headline rose 0.3pp to 3.0%, but headline carries the energy prices the target measure excludes by construction.

Consumer confidence improved to -16.6 in Q3 from -18.1, published on 12 August — the freshest release in this note, landing two days after the CPI print. It cuts mildly against the softening-demand read and is the one series pointing the other way.

The labour market is softer, though less so than the table suggests. Registered unemployment rising from 1.9% to 2.1% in July is the ordinary summer pattern in the NAV series and should not be read as loosening on its own. The LFS rate at 4.5%, up 0.1pp, is the cleaner signal: a gradual drift, at a level that is unremarkable rather than tight by Norwegian standards. Norges Bank's own June characterisation was that capacity utilisation "appears to be close to a normal level but is drifting down".

Activity is mixed: retail sales rebounded 1.8% in June after a 2.4% fall in May, and manufacturing production fell 1.0%. House-price growth slowed to 2.2%.

The 8.5pp jump in PPI to 23.4% should be discounted rather than led on. Norwegian producer prices carry oil and gas extraction, which CPI-ATE excludes by construction, and the series fell steeply into June before rebounding — the July move reverses that dip rather than establishing a new impulse. It is not the cost measure the Committee works from, and the Governor drew the distinction herself on 8 May: "Our decision to raise the policy rate this week is not only related to the increase in prices for oil and other commodities."

The krone is firmer: EUR/NOK from 11.15 to 10.97 and USD/NOK from 9.72 to 9.49. The June statement recorded the krone as broadly in line with the March assumption; it is now stronger than that, which dampens imported inflation.

Oil is the one genuinely new macro development since June that is not in the CPI print, and it cuts against the disinflation read. The June statement's stated premise was that oil and gas spot and futures prices had fallen since March. That premise has reversed: WTI closed at 77.08 on 7 August, the common as-of date for the market rows above, and had traded up to roughly 84.8 by 11 August — above the 80.35 decision-day close. For an oil exporter the channel runs both ways, lifting terms of trade and supporting the krone while feeding energy costs that CPI-ATE excludes but headline does not. It is the reason to treat the July undershoot as softening rather than settled.

5. What the speeches say

No member has spoken publicly on monetary policy since the June decision, so the record below is what the Committee already had when it last decided. Its value now is as the benchmark against which the 13 August language will be read — and every line of it predates the inflation print that repriced the meeting.

Governor Wolden Bache's 18 June remarks set that benchmark: "Inflation is too high, and the rapid rise in business costs in recent years will contribute to keeping inflation elevated ahead", adding that "We expect that a somewhat tighter monetary policy stance will be needed to bring inflation down to target within a reasonable time horizon."

The line held all year. In March she said the Committee "judges that it will likely be necessary to raise the policy rate at one of the forthcoming monetary policy meetings", with the outlook indicating "that inflation will be higher ahead than previously projected". Justifying the May hike she said "Inflation, however, is too high and has run above target for several years", warning that a failure to react systematically to prospects for high inflation would erode confidence in the target. A research address on 21 May noted that the bank's estimate of the normal policy rate "has increased somewhat in recent years". June added the word further to the March formulation, reflecting the May hike already delivered, and revised the path beneath it higher.

Deputy Governor Longva's 25 February appearance is the only other scored intervention, and it is procedural rather than directional: responding to the Norges Bank Watch report he reviewed the 2025 easing on the grounds that underlying inflation had fallen faster than expected: "we reduced the policy rate twice in 2025, to 4.25 percent in June and then to 4 percent in September."

6. Scenarios

7. Into the meeting

With no Report and no new path, the decision is the language. The test is whether the June assessment — "it will likely be necessary to raise the policy rate further at one of the forthcoming monetary policy meetings" — is repeated, diluted, or dropped, now that core inflation is running below the above-3% profile that assessment was built on.

Expect the statement to acknowledge softer underlying inflation while noting that core remains above target, and to weigh the firmer krone against inflation expectations among households and business leaders, which MPR 2/2026 describes as still elevated. The next inflation print is 10 September, two weeks ahead of the 24 September meeting — a full projection round with Monetary Policy Report 3/2026 and a new rate path, and the natural venue for a decision this meeting defers.

Full analysis on the RoboMacro site → /central-banks/NORGES