The third Russia–Africa summit ended in Moscow on Thursday with representatives of 48 African countries, 22 of them heads of state, according to The Moscow Times². Opening the plenary in the Grand Kremlin Palace, Vladimir Putin said Russia’s trade with African states “exceeded $27 billion last year” and had grown by 50% since the previous summit¹. A day earlier he said it could pass $50 billion in the near future, which, as Vedomosti noted, means at least doubling it³.

Now read the same numbers from the other side of the desk. The Moscow Times reports that the $27 billion was 2.4% less than in 2024, and that trade with the entire continent accounts for just 3.9% of Russia’s foreign commerce². Putin has promised a doubling since at least 2019². None of that makes the relationship small for the countries involved; it makes the target a speech, not a forecast.

The more interesting line was about factories. “Our companies are simply interested in moving certain production operations to African countries,” Putin said, adding that this fits Russia’s interests “when scores of sanctions are placed on the Russian economy”². He named no company². If it happens, it will show up in the dull places first: investment approvals, customs codes and certificates of origin in the host countries.

The paperwork that did appear is real enough. Presidential adviser Anton Kobyakov said more than 60 agreements were signed, not counting documents classed as commercial secrets, and that some 1,200 business representatives from about 630 companies took part³. In the transcript, Putin also put the rouble’s share of trade settlements with African countries at 85% and Russian grain deliveries to Africa at more than 18 million tonnes in 2025¹.

For you, the practical point is modest. Summits produce targets; customs offices produce statistics. …

Continue in reading view →

Thursday handed markets three pieces of news, and only one of them moved a price much. The euro edged down to $1.1125 after the European Central Bank held its deposit rate at 2.50%, Reuters reports¹. It has lost almost 2% against the dollar this month and 2.5% against sterling¹. Investors hold their biggest short position on the euro in a decade, almost $19 billion¹.

What moved it. Two Frankfurt items came on the same day: a hold that read as cautious, and Christine Lagarde’s announcement that she will leave on 7 February 2027². Markets had gone into the meeting pricing about a 76% chance of a December rise, ActionForex notes, so a statement stressing stable underlying inflation read as relatively dovish². CNBC, citing LSEG data, had money markets at about 67% for December after the announcement³. Euro-area government bond yields edged lower too¹. ActionForex is careful on cause: because the resignation came before the rate decision, the two effects “cannot be separated precisely”².

What did not move. The United States reported third-quarter growth of 3.4% annualised, above the 3.2% expected, and headline PCE inflation of 3.6% in September². Either could have lifted yields. The two-year Treasury stayed near 4.687% and the ten-year near 5.00%, with a December Fed rise priced at about 83%, against about 85% a day earlier². Desk arithmetic: the market had already paid for the news, so the receipt changed little.

The calm point. Europe’s stocks did not join the euro’s mood: the STOXX 600 remained about 1% higher as the ECB held, Reuters reports¹. A weaker currency on a leadership date, with bonds and shares orderly, is a market doing its accounting in public, not a run. For you, a softer euro makes dollar-priced fuel a little dearer and European exports a little cheaper; both effects arrive slowly. Nothing here is investment advice.

Israel’s Central Elections Committee said on Thursday that the counting of the double envelopes, the ballots of soldiers, diplomats, patients and prisoners, had been completed, with final results to follow¹. By then results from 12,644 polling stations had been entered, containing 5,589,836 votes: 5,564,764 valid and 25,072 invalid¹. Desk arithmetic: fewer than one ballot in 220 was spoiled.

The seats did not move. With 97% counted, Haaretz reported Netanyahu’s bloc at 53, the Eisenkot-led bloc at 48 and the parties representing Palestinian citizens of Israel at 19, a record, according to the Guardian². Neither camp reaches the 61 needed for a majority on its own². The opposition faces a plain choice: an arrangement relying on the Arab parties, which Eisenkot’s bloc has ruled out, or a unity government with Netanyahu, which Eisenkot vowed never to join². The Joint List has offered tactical support to an Eisenkot minority government without joining it².

What you can actually hold on to is the timetable. The committee has asked the Shin Bet to look into alleged foreign interference and wants findings within a week, ahead of the 4 November deadline for certifying results². President Isaac Herzog said that once official results are in he will meet representatives of all political groups, and asked them to “keep an open mind”². The new Knesset is due to be sworn in on 10 November, and Eisenkot warned against using the interference allegations to delay it².

Each of those steps belongs to an institution, not a candidate: a committee that counts, a president who consults, a parliament that convenes on a fixed date. Coalition talk before 4 November is bargaining. The certified result is the only document that binds.

“Look on the bright side” has a technical name, cognitive reappraisal: changing how you think about a situation to change how it feels¹. It sits at the heart of cognitive behavioural therapy, and for years brain-imaging studies have tied it to activity in the prefrontal cortex and calmer responses in the amygdala¹. The quiet assumption was that one could read off a person’s ability to calm down from those regions.

A consortium of 32 research teams has now tested that assumption at scale. They pooled 40 brain-scanning datasets with 2,175 adults from non-clinical samples, about 20 times more than earlier efforts, and preregistered the analysis¹. They compared three common yardsticks: what people say about their own habits in questionnaires, how much their mood ratings drop when told to reappraise, and how much their amygdala quietens¹.

The yardsticks barely agree. Mood ratings and amygdala change were linked, but weakly, with a correlation of 0.08; the authors calculate that a single study would need 1,105 participants to detect it reliably¹. Questionnaires were linked to neither¹. And none of the three showed the expected link to the prefrontal “control” circuits; those circuits were active both when emotions were being generated and when they were being regulated¹. The authors conclude that strong claims about a person’s emotion regulation based on any specific brain region “should be made with considerable caution”¹.

There are limits. All the samples were non-clinical, which the authors say limits how far the findings carry to people with more severe symptoms, and most studies asked about reappraisal habits rather than ability¹. A commentary in the same issue describes the work as investigating “an assumption that researchers have held” about measuring reappraisal².

Continue in reading view →

Days make headlines; months make the bill. So here is the month. Between 1 and 28 October, German solar panels generated 4,742.6 gigawatt hours of electricity, according to the Bundesnetzagentur’s SMARD figures, against 3,244.9 GWh on the same days last year¹. Desk arithmetic: that is about 46% more, in a month when the sun is low and the days are short.

Bar chart of German electricity generation from 1 to 28 October, in gigawatt hours, 2026 against 2025. Lignite generated 6,081.3 GWh in 2026 against 5,111.5 GWh in 2025, up about 19%. Solar generated 4,742.6 GWh against 3,244.9 GWh, up about 46%. Natural gas generated 4,223.2 GWh against 4,852.7 GWh, down about 13%. The latest day included is Wednesday 28 October 2026.

Source: Bundesnetzagentur | SMARD.de, realised generation, Germany: lignite (filter 1223), photovoltaics (4068), natural gas (4071), daily totals 1–28 Oct 2025 and 2026, retrieved 29 Oct 2026, 16:24 CET, summed and rounded to 0.1 GWh by Inside Desk · CC BY 4.0 · CSV

Who stepped back. Gas-fired plants generated 4,223.2 GWh over the same 28 days, against 4,852.7 GWh in 2025¹. Desk arithmetic: about 13% less. That fits the price of the fuel. The Dutch TTF benchmark settled at €97.35 per megawatt hour on Wednesday and is on course for a monthly rise of about 34%, Investing.com reports². A gas plant that has to buy at those prices is often the last one anyone wants to run.

Continue in reading view →

Every month the European Commission asks managers and households across the EU how things look. Thursday’s October results carry two lines worth reading together.

The first is about prices. Managers’ selling-price expectations “increased considerably” in all four business sectors: industry by 3.4 points, retail by 2.8, construction by 2.4 and services by 2.1, all further above their long-term averages¹. Consumers’ sense of how much prices rose over the past twelve months jumped by 3.9 points, while their expectations for the next twelve months stayed “unchanged at a high level”¹.

The second is about households. EU consumer confidence fell by 1.2 points, the second fall in a row, because people judged their household’s past and expected finances more harshly and planned fewer major purchases¹. Retail confidence dropped by 1.7 points¹. ING’s economists put it plainly: consumers are “shying further away from major purchases”². The overall sentiment indicator fell 0.6 points to 96.9 in both the EU and the euro area; in Germany it edged down by 0.7¹.

There is a brighter line too. Employment expectations rose, to 98.9 in the EU and 98.2 in the euro area¹. Firms are more inclined to hire than they were in September.

On Thursday the European Central Bank left its rates unchanged; the deposit rate stays at 2.50%³. Its statement noted: “Energy prices have continued to rise”³. For you, that means no new push from Frankfurt this time on what banks pay on savings or charge on new loans.

What can you do with this? Surveys measure intentions, not prices. Germany’s flash inflation figure arrives on Friday and will show what actually happened. …

Continue in reading view →

Chronic traumatic encephalopathy, CTE, can only be diagnosed after death, and only in brains that families choose to donate¹. That has always been the problem with the numbers. Donors are not a random sample: families of players who had symptoms are more likely to give¹. A study published this week in JAMA Network Open tries to correct for exactly that bias¹.

Researchers at Boston University, Mass General Brigham and the Concussion & CTE Foundation studied 387 deceased former NFL players whose brains were donated; 362 of them, 94%, had some degree of CTE pathology¹. They then went to the dull records: National Death Index death certificates for all 2,362 NFL players who died between 2008 and 2023, including how many died of neurodegenerative disease or had dementia¹. Players with cognitive symptoms turned out to be about five times more likely to be donated¹. Adjusting for that, the estimate for all deceased players is 74.5%¹. An August study by the same group could only say somewhere between 25% and 98%¹.

For comparison, CTE has been seen in under 1% of thousands of donors to community brain banks in the United States, Australia, Austria and Brazil¹. The authors add a fair caution: the estimate covers players who have died, and does not predict how many living former players or active players are affected¹. Not everyone with CTE pathology develops symptoms, particularly in early-stage disease¹.

This is occupational health in its plainest form: a job, a repeated exposure, and a measured outcome. Earlier work found NFL players develop dementia at four times the rate of the general population¹. Matt Hasselbeck, an 18-year NFL veteran who supports the research, said he was looking at the findings “like getting a scouting report”¹.

The researchers’ advice to former players worried about their brain health is concrete: see a clinician who specialises in neurodegenerative conditions, and work on the risk factors that can be changed¹. …

Continue in reading view →

At least 14 people were killed when a Russian jet-powered drone hit a crowded platform at the main railway station in Dnipro on Thursday, local officials said, according to the BBC¹. Dnipropetrovsk regional head Oleksandr Hanzha said at least 14 others were wounded, including an 11-year-old boy¹; six were in critical condition². Ukrzaliznytsia, the state railway, said a passenger train bound for Zaporizhzhia was standing at one of the platforms¹. Russia’s military had not commented; Moscow denies deliberately targeting civilians and says its strikes on railways aim at infrastructure used to move troops and equipment¹.

An elevated threat level had been announced in Dnipro before the strike, said the railway’s head, Oleksandr Pertsovsky. “But unfortunately there were still a large number of people on the platform,” he said¹. That sentence is the whole problem of a working railway in wartime: trains run because people need them, and people wait on platforms because that is where trains stop.

The railway keeps a ledger. Of 521 locomotives hit since the start of the full-scale invasion, almost 60% were hit this year, Ukrzaliznytsia told AFP, and attacks on railway facilities rose from 1,206 in 2025 to 1,700 so far this year, the BBC reports¹. Desk arithmetic: that is about 300 locomotives this year, and the 2026 count of facility attacks is already about 40% above last year’s full total.

The same day brought other strikes: two 15-year-old girls were killed in a missile attack on central Chernihiv, regional head Vyacheslav Chaus said¹, and one man was killed in the Rivne region². Overnight Moscow’s mayor said 613 Ukrainian drones had been launched towards the capital region, most shot down, while it hosted the Russia–Africa summit¹.

Zelensky called for further US, European and G7 sanctions¹. For you, the useful thing to follow is not the rhetoric but the railway’s counts: they are compiled by the people who repair the track, and they tell you whether protection of stations, trains and passengers is keeping up.

The Lancet Countdown on health and climate change published its tenth annual indicator report this week: 146 experts from 71 institutions, 54 indicators². The headline everyone will quote is heat. Heat was linked to an estimated 540,800 deaths a year between 2016 and 2023, and to a record 710,000 in 2023 alone³. Desk arithmetic: from the Countdown’s own data file, the 2016–23 yearly average of 540,811 is 35.8% above the 2006–15 average of 398,371¹. That is the “nearly 36% since Paris” you will read elsewhere³.

The more interesting number is the one the model subtracts. The Countdown runs the same years twice: once with the temperatures that happened, once with temperatures as they would have been without human-caused warming¹. In that second world, heat deaths were about 179,260 a year in 2006–15 and 170,875 a year in 2016–23¹. Flat, slightly down. Desk arithmetic: the share attributed to climate change rose from about 219,100 a year to about 369,900, an increase of roughly 69%¹. In 2023, 71.5% of the estimated heat deaths were attributed to climate change¹.

Continue in reading view →