Energy shock reaches household expectations: German saving at 2008 peak, US inflation expectations at 4.6%; central banks can no longer look through it

Same week, two shores, the same household reflex

In Germany the NIM/GfK consumer climate index fell to −30.6 for October on 25 September; September's revised value was −26.8 and the expectation in the Reuters poll was −27.4. The break is on the income side: the income expectations sub-index fell in one month from 1.7 to −15.0, a drop of 16.7 points, while the propensity to save rose from 15.5 to 21.5, its highest level since the 2008 crisis. The propensity to buy slipped from −9.8 to −10.8, while general economic expectations improved slightly from −3.9 to −3.4; in this survey of 2,000 people conducted between 3 and 14 September, households said they feared the erosion of their own purchasing power more than a collapse of the economy.

On the other side of the Atlantic, the University of Michigan's final September data were released the same day: consumer sentiment fell from 51.7 in August to 48.1, and the expectations index dropped 10.1% to 46.3. The 1-year inflation expectation rose from 4.0% to 4.6%, returning to its June peak; in February this rate was 3.4%. The 5–10-year expectation, more critical for central banks, rose from 3.3% to 3.4%.

The common denominator of the two surveys on 25 September is that the energy shock has moved out of the fuel and utility bill and into households' 12-month plans. According to the Bank of England's 17 September summary, since the July report Brent has risen 36% and UK wholesale gas prices 78%; according to Eurostat's flash estimate, energy prices in the euro area rose 14.3% year on year in August while core inflation stayed at 2.4%.

Why the option of looking through it is closing

In a supply-driven price shock, the standard central bank response is to look through it rather than suppress demand with rates, as long as the shock is temporary; this response depends on inflation expectations remaining anchored to the 2% target. The week of 21–25 September showed this condition eroding simultaneously in 3 major economies. On 17 September the Bank of England held rates at 3.75% by a 6–3 vote, but its summary stated that the risk of second-round effects had increased since July and that the Citi/YouGov household inflation expectation had risen between July and August; its inflation projection for the first quarter of 2027 is slightly above 4%.

Eight days later, on 25 September, Governor Bailey said in Oxford that high energy prices make it harder to hold rates; a day earlier Deputy Governor Lombardelli had defined wage growth consistent with the 2% target as about 3.25%. Depending on the source, the market puts the probability of a November hike at between 75% and 80%. In the US, Philadelphia Fed President Paulson said on 24 September that core inflation was stuck in the 2.5–3% range and that modest further tightening might be needed; the Fed had raised rates to a 3.75–4.00% range on 16 September. The probability of an October hike also varies by source: Briefs, citing CME FedWatch on 24 September, gives 64%, and Trading Economics on 25 September gives 66–70%.

In the euro area, the ECB raised the deposit rate to 2.50% on 10 September; money markets price 2.75% at the end-October meeting and roughly 100 basis points of hikes in total by the end of 2027. Northern Europe has turned the same way: Norges Bank raised its rate from 4.25% to 4.50% on 24 September, the Riksbank held at 1.75% but announced that hikes would begin this year, and the SNB kept its rate at 0% while raising its 2027 inflation forecast from 0.6% to 0.8%.

Stagflationary tightening: saving cools demand, yet rates still rise

Here is the hidden link: households are reacting through 2 channels at once. The 6-point rise in Germany's propensity to save in a single month, to 21.5, is a brake that cools demand by itself; in normal conditions this would give the central bank grounds to wait. But the 0.6-point rise in the US 1-year expectation in a month and the BoE's view that second-round risk has increased since July show that the same households are starting to treat price rises as lasting. The result is tightening in which rates rise while demand weakens: in autumn 2026 the consumer brake and the rate brake are being pressed at the same time.

The yield curve priced this combination at the long end on 25 September. According to Trading Economics data, the 10-year Bund closed at 3.61% on 25 September, heading for a 7th consecutive weekly rise; the 30-year Bund climbed to 3.91%. The same day in the UK the 10-year gilt was at 5.35% and the 30-year gilt at 5.87%; in the US the 10-year yield eased to 5.17% while the 30-year rose to 5.50%. The long end holding up better than the short end indicates that the market is pricing not just 1–2 hikes but the possibility of inflation carrying over into 2027.

The contradiction in Germany sharpens the picture. On 24 September the Ifo business climate rose to 89.9, its highest since May 2023, and a day earlier Germany's composite PMI had come in at 53.8; yet on 25 September households' income expectations collapsed to −15.0. Optimism on the corporate side is fed by a fuel tax cut of 0.17 euros per litre from 1 October and by the five leading institutes raising their 2026 growth forecast from 0.6% to 1.3%. For the ECB, a strong PMI is an argument for a hike, while a consumer climate of −30.6 means consumption lost in the coming quarters.

The winter buffer either anchors expectations or unmoors them

Where expectations go in the 2026–27 winter will largely be determined by gas. On gas day 23 September EU storage held 796 TWh of 1,132 TWh capacity, 70.3% full; that is 11.7 points below the 82.0% on the same day of 2025, 15.7 points below the 2021–2025 average of 86.0%, and also behind the five-year low of 72.6%. Germany, at 57.12% full on 25 September, is 19.3 points behind 2025.

Sources diverge on the filling path: Global Energy Flow writes that a daily pace of about 0.21 points falls slightly short of the relaxed 80% target for 1 November, while Voltstack says winter preparation looks on track at the current base. Europe's benchmark gas price, TTF, had risen to 81 euros/MWh on 10 September, its highest since December 2022. If a cold winter drains storage quickly, the 1-year expectation now at 4.6% and the propensity to save at 21.5 could become entrenched with a new wave of bills; a mild winter, on the other hand, would give central banks room to pull back in early 2027.

Two channels to Türkiye: the order book and the yield floor

For Türkiye, the significance of this week is that its 3 largest export markets showed a break on the household or rate side at the same time on 24–25 September. According to an Anadolu Agency report dated 18 January 2026, Türkiye's exports to Germany in 2025 rose 9.57% to a record 19.834 billion dollars, and Germany kept first place; the second market was the UK at 14.2 billion dollars and the third the US at 13.2 billion dollars. The three markets total 47.2 billion dollars, and in all three either household confidence fell or expectations of a rate hike strengthened on these dates. Automotive exports to Germany rose 36.1% in 2025 to more than 6.5 billion dollars; durable goods are the first spending item households postpone in an environment where the propensity to save has risen to 21.5.

The second channel is the yield floor, and it became clear on 25 September: according to Bloomberg HT's market summary for 25 September, Türkiye's 5-year CDS exceeded 250 basis points for the first time since 27 May; the 24 September value on Investing.com was 245.17 basis points. The same day the euro/lira rate stood at 55.84 and the dollar/lira rate at 48.97. In an environment where the 10-year Bund is in its highest zone since 2009 and US long-term yields in their highest since 2004, Türkiye's external borrowing cost is being pushed up independently of its own risk premium.

These two channels do not offset each other; they stack: as demand cools in Europe, export revenues come under pressure towards 2027, while over the same period the global rate floor makes external financing more expensive. The euro/lira level of 55.84 gives exporters a partial currency buffer; but the fall in Germany's propensity to buy to −10.8 shows the limit of offsetting volume losses through the exchange rate. Türkiye's own 2-year yield having exceeded the policy rate at 37.09% on 24 September means domestic demand is being squeezed at the same time.

What to watch and uncertainties

Three thresholds will test this reading. The 30-year Bund exceeding 4% (3.91% on 25 September) would show that the term premium in Europe continues to be fed by the expectations channel. EU storage remaining below 80% on 1 November would turn winter into an expectations risk. Türkiye's CDS exceeding 275 basis points would mark the zone where the global rate floor turns into Türkiye-specific credit pricing. On the calendar, 28 October is the day of both the FOMC decision and the UK budget.

The probability of an October Fed hike varies by source between 64% and 70%, and that of a November BoE hike between 75% and 80%. For the 25 September closes, our event records give different figures: the 30-year Bund at 3.91% versus 3.92%, and the US 30-year at 5.50% versus 5.51%; the report uses Trading Economics values. Because the GfK survey was conducted on 3–14 September, it does not reflect energy price moves over the following 11 days. Sector-level order data on how far the fall in German demand will feed into Turkish exports in the last quarter of 2026 could not be verified tonight.