Quick Comment: Riksbank to Hold Steady for Now
Photo by Karin Gerlach
Punchline: The Riksbank is expected to pause in August, holding the policy rate at 2.0%. Inflation is still high after the summer spike, but underlying pressures are softening. With growth stagnant and the labour market weak, a September cut is the base case.
The Riksbank will announce its next policy decision on 20 August. Like several other central banks, it faces two opposing forces. On the one hand, inflation rose in early summer and remains above target. On the other hand, the real economy is weak. That combination makes it difficult to justify either a rate increase or an immediate cut.
This is an interim meeting without new forecasts. That gives the bank a reason to keep rates on hold and wait until September, when new projections will be available. By then, it will be clearer whether the unexpected rise in inflation in June was temporary or a sign of renewed price pressures.
It seems likely that inflation will abate. The tariffs introduced by the Trump administration are likely to slow global growth and, by doing so, lower inflation in the US’ trading partners, including Sweden. The likely outcome is no change in interest rates this month, but a signal that a cut is coming later in the year.
Inflation
In July, headline CPI rose to 0.8% y/y from 0.7% in June. CPIF inflation, the Riksbank’s target measure, increased to 3.0% from 2.8%. Core inflation, measured by CPIF excluding energy, eased to 3.2% from 3.3%.
Source: SCB
The recent inflation overshoot does not appear to be driven by domestic overheating or currency weakness. Demand remains subdued, and the krona is markedly stronger than at the start of the year. This supports the view that the summer’s higher readings reflect temporary factors rather than a shift in underlying pressures.
The key question is how persistent the increase in inflation will be. Since the Riksbank and most other central banks underestimated the persistence of the rise in inflation in 2021, it is reluctant to make the same mistake twice.
Furthermore, with both the June and July readings above the Riksbank’s June forecast, it is difficult to cut interest rates now. The bank will want confirmation in the August data that the trend is downward before acting. If inflation falls as expected, a September cut remains likely.
Economic Activity
Economic growth stalled in the second quarter. The flash GDP estimate showed an increase of only 0.1% QoQ and 0.4% YoY. The weakness of activity is underscored by an average QoQ growth rate of -0.2% since 2023Q1.
Source: SCB
Flash GDP figures are prone to revision, often upward once more complete data become available. Even so, the underlying message is clear: growth was weaker than the Riksbank had anticipated. The National Institute of Economic Research expects the sluggishness to persist into the third quarter. This means the bank is likely to mark down its full-year GDP forecast at the September meeting.
The lack of growth has been accompanied by labour market weakness. Unemployment fell in the second quarter and employment rose, but in both cases the change was limited. The labour market is plainly weaker than in recent years.
Source: Ekonommifakta.se
Sentiment and External Developments
Sentiment has improved modestly but remains weak. Consumer confidence has risen in recent months, with a strong gain in July, but it is still well below its long-term average. Business sentiment has been broadly stable for the past five quarters.
Source: Ekonomifakta.se
The recent EU–US trade agreement reduces the risk of escalation and removes some uncertainty for exporters. However, tariffs remain high and risk slowing economic activity.
Reaction function
In several past posts (starting from here), I have complemented the analysis of upcoming policy decisions with predictions from so-called reaction functions. Updating that analysis for this meeting, the estimated probability of no change is 60%, while the probabilities of a cut or an increase are each about 20%, as shown in the graph below. Since some readers may feel that this adds little to the discussion, the details are provided in the appendix.
Source: my estimates
Policy Outlook
In June, the Riksbank made a notably dovish move, lowering the policy rate to 2.0% and signalling in its projections roughly a one-in-two chance of another cut before year-end. Since then, incoming data have diverged from those forecasts, showing a softer real economy but also somewhat higher inflation than expected.
The unexpectedly high inflation this summer is the main reason for the Riksbank to wait before cutting again. But the weakness of the real economy, and the risk that the shift in US trade policy will aggravate it further, means that additional easing will probably be needed. With no new forecasts this month, the bank can afford to wait for more data.
Market pricing adjusted sharply after the June inflation surprise. Ahead of that release, a September rate cut was close to fully priced, but the market now expects a later move, possibly towards the end of the year. The August decision could therefore also serve to guide expectations back towards an earlier cut, should the bank wish to keep September in play.
The most likely outcome is no change in August, with a dovish message pointing to a cut in September. If the August inflation data show that price pressures are abating, the Riksbank may cut the policy rate to 1.75% next month. Depending on incoming data, further cuts could follow later in the autumn.
Appendix: Reaction Functions
A reaction function captures the average interest rate setting behaviour of a central bank over a given sample period. In the case of the Riksbank, the period starts in 2021. The idea is that policy changes can be predicted using a small set of observed economic variables.
This type of statistical analysis is not intended to replace a careful economic assessment. Rather, it provides a point of comparison. If the policy prediction derived from economic analysis differs from that produced by the statistical model, the difference should prompt further investigation. Often, but not always, this leads to the conclusion that the reaction function is poorly specified.
The model used for the Riksbank is particularly simple. It includes the policy rate set at the last meeting and the inflation rate in the month before the current meeting. In earlier versions, I found—somewhat to my surprise—that using CPI inflation gave better in-sample predictions than CPIF inflation, even though the Riksbank targets the latter. Using more recent data, that finding has been reversed. The reason is that while CPI inflation has fallen in 2025, CPIF inflation has risen. This change has made it easier to identify which variables best predict policy changes.
The analysis was initially based on 19 policy meetings but has now been extended to 27. This is still a small sample, and the results should be viewed as preliminary. The model will need to be refined as more data become available.
That balancing act between stubborn inflation and a sluggish economy feels like walking a tightrope in a windstorm. Holding steady for now makes sense - it gives the Riksbank room to gather better data before making a move that’s hard to reverse.