One more reading like this one and the light turns red

Where we are

The market-risk light reads YELLOW, as it has since September 25. It turns RED only when two things are true together: households are gloomy (the University of Michigan consumer survey below 55) and the money the world's four biggest central banks hold, counted in dollars, is shrinking. The first has been true since the August survey; the latest reading is 48.1, published September 25. The second is what moved this week.

Last week the money total was up 0.52% over twelve weeks. This week it is down 0.31%, and down 1.05% over four weeks. Down on both counts is the definition of 'shrinking', so this week's raw reading says shrinking. The light does not use a raw reading. Since September 25 the money input changes only when two weekly readings in a row agree, and this is the first. The reading the light actually uses is still 'flat', which it has been since September 21. Next Monday's reading settles it: a second 'shrinking' turns the light RED; a bounce back above zero over twelve weeks keeps it YELLOW.

The other reading that moved last week held this week. The overall cost and availability of borrowing reads 'neutral', heading tighter, for the second week in a row, which under our two-prints rule means it is now the confirmed state rather than a first signal.

The money: three months of the banks, one month of the dollar

The four-bank total stands at $24.79 trillion, down $77 billion over twelve weeks. Split that: the central banks' own balance sheets shrank by $176 billion, and currency moves added $99 billion by making the European, Japanese and Chinese balance sheets worth more in dollars than three months ago. The banks' own books set the direction; the exchange rate cushioned it.

Over the last four weeks the split flips. The total fell $262 billion, and $253 billion of that, 97%, is the dollar strengthening. The banks' own books moved by $9 billion. So the three-month drain is the central banks themselves; the one-month drop that pushed the reading past zero is the exchange rate. Both are true at once, and the second is the one that can reverse in a week.

The US-only measure, the Fed's balance sheet minus the Treasury's own account at the Fed minus cash parked overnight at the Fed, is $5.87 trillion, reading flat over twelve weeks: down $120 billion over that span, up $11 billion over four weeks. The Treasury's account stands at $880 billion, down from the $949 billion in last week's edition, and the overnight cash is at zero. This number explains the backdrop; it is not part of the light.

The trend this brief trusts most, the Fed, the European Central Bank and the Bank of Japan measured the same way at both ends, went from $17.53 trillion on July 20 to $17.35 trillion now, down 1.03%. The Fed is flat ($6.74 to $6.75 trillion). The Bank of Japan's leg is where it was twelve weeks ago, $3.96 trillion, though it fell from $4.10 trillion on the week: September's statement arrived on October 9, and the yen is at 157.8 to the dollar. The European Central Bank's leg is down from $6.83 to $6.64 trillion over the window. One caveat on that last figure: the ECB number on file is the one published September 30, describing September 25, and no newer weekly has appeared in twelve days. The ECB leg's move this week is the euro, not the bank.

China's statement on file is still August's: ¥49.857 trillion, down 0.7% on the month, the first decline after two rises, with the lending facilities that built the June and July expansion now running off. It is 72 days old and counts as current; September's is due between October 16 and November 5.

Borrowing: a small step back, inside a bigger step up

Riskier companies pay 3.15% more than the US government to borrow this week, down from 3.24% last week. That is still 0.44 points above where it was twelve weeks ago, so the tool keeps the reading at 'widening'; one week's easing does not undo a month's jump. Safer companies pay 0.82% extra, down from 0.86%, still calm.

The real cost of borrowing for ten years, the ten-year US government bond yield minus the inflation the market expects, is 2.87%: the bond at 5.22%, the inflation forecast at 2.35%. It was 2.88% last week, the first weekly reading since late August that did not go up. It was 2.33% on July 20.

The dollar, against a broad basket of currencies, is at 121.4, up from 120.3. Over four weeks it is up 2.27%; over twelve, 0.73%. In other words the whole of the three-month rise, and more, came in the last month, which is the same thing the money paragraph says from the other side. Expected stock-market volatility is 15.4, calm. The tool names three drivers for the 'neutral, tightening' reading, in this order: what riskier companies pay to borrow, the money total, and the real cost of borrowing.

The economy: same label, a thinner margin on one input

The economy label is unchanged: running hot, our label SUMMER, meaning growth is still going and prices are still rising. Under the growth rule adopted September 30, that is what every reading since December 4 has said, 78 in a row. Last week's edition explained the rule change; nothing about the rule moved this week.

Growth is a vote of three inputs, and two of three decide. Factory output: flat, down 0.1% over three months. Jobs, read from new claims for unemployment benefits: the four-week average is 198,000, 7.8% lower than twelve weeks ago, so jobs vote growing. Investment, business spending on computers, equipment and software after inflation: up 16.8% on a year earlier in the April to June figure, so it votes growing. Two of three say growing.

The margin worth watching is jobs. Claims vote 'growing' when the four-week average is more than 5% below where it was twelve weeks earlier. That gap has narrowed on each of the last three readings: 9.1%, 8.8%, 7.8%. Claims themselves are still falling; what is shrinking is how much faster they are falling than they were in the summer. If the gap closes below 5%, jobs vote flat, and with factories already flat the label would rest on investment alone. The next investment figure arrives with the next quarterly GDP release.

Prices: the Fed's preferred inflation measure, excluding food and energy, is 3.0% over a year in the August figure published September 30. The tool still reads inflation's direction as rising, because the shorter-term trends in that measure and in the consumer-price version both point up, while the market's five-year inflation forecast points down. Inflation remains the dominant risk.

The AI build-out, which is what the investment input mostly measures, reads 'booming' on the separate scoreboard: data-centre construction at $85 billion a year in August, chip imports at $5.5 billion in August, the four biggest builders' spending on equipment and buildings at $165 billion in April to June, and utility output (electricity and gas) up 6.2% on a year earlier in August.

What to watch

• Next Monday's money reading. A second 'shrinking' in a row turns the light RED. The deciding number is 0.31% below zero over twelve weeks, and the dollar moved the total by about 1% in the last month alone, so the exchange rate between now and then matters more than any central bank's weekly figure.
• The ECB gap. Twelve days without a new weekly figure means the European leg is moving on the euro alone. When the backlog lands it could move the total in either direction.
• Whether riskier companies' borrowing costs keep easing from 3.15%, or this week was a pause in the widening that began in late September.
• The jobs margin: 7.8% against a 5% line, narrowing three readings running.
• China's September statement, due October 16 to November 5, and the consumer survey's next final reading on October 30.
• Data quality: one source stale (the ECB weekly, 12 days), everything else current, the collector healthy, all thirteen weekly rows in each history recorded live. The dollar and exchange-rate figures are as of October 2, the latest the source has published.

The frame, not the trade

If the light turns RED next Monday, the rules written before the fact say what that means: the two inputs the light watches both read bad. They do not say sell anything. A colour is a summary of two numbers, and one of those numbers is 0.31% past its line, moved there mostly by the exchange rate. The light has turned RED twice in the last twelve weeks, for a week in early September and for a day on September 24 before the confirmation rule took effect, and each time what mattered was not the colour but which input flipped and why.

The useful question this week is the one the money paragraph forces: the central banks themselves have been shrinking for three months, and until this month the dollar hid it. Which holdings in your book were sized on the assumption that money stays plentiful? If the light goes red on a currency move, would you treat that differently from the light going red because a central bank cut its balance sheet? If the answer is yes, it may be worth writing down now what you would actually do in each case, while the reading is still one of two.

None of this is advice and none of it is a forecast. It is a week's data, the reasoning applied to it, and the rule that decides next week printed in advance.