Notícias FX

Four PMIs in Seventy-Five Minutes. The One Your Filter Carries Was the Least Informative of the Four.

Four flash purchasing-managers surveys landed between 07:15 and 08:30 UTC today - France, Germany, the euro area, Britain.

Seventy-five minutes, and a filter padding fifteen minutes either side of each entry was flat from 07:00 to 08:45, 105 minutes for four events. The entry most systems carry was the least informative of the four. The 08:00 euro-area aggregate missed on manufacturing by 0.3 points. The French print 45 minutes earlier missed by 1.2 and went into contraction at 49.3; the German print 30 minutes earlier beat by 0.7 at 54.2. The aggregate is smaller in magnitude than either component and carries the opposite sign to the larger economy - a 4.9-point level gap and a 1.9-point surprise gap averaged into a 0.3-point miss, 30 to 45 minutes late. Ten fields, seven negative surprises and three positive. Sizing by surprise magnitude fails on the same morning: the biggest beat was Britain's composite at plus 1.3 and sterling fell, with cable near 1.3380. Two publishers label the same British figure pair differently, composite on one and services on the other. And two clock changes eight days apart compress the European-to-American event gap from 315 minutes to 255 for one week only, which is the worst failure pattern because it heals before you investigate it.

The Day Was Over by 10:30. The 12:30 Beat and the 14:00 Cut Added Six Pips at Most.

Dollar-yen traded 156.0 pips on 22 October, 158.25 to a low one publisher puts at 156.69.

A different publisher's European-session field had the low at 156.75 by 10:30 UTC - 150.0 pips, 96.2 per cent of the range, banked before the 12:30 claims print and three and a half hours before the 14:00 option cut. The ceiling on what both scheduled windows added is the 6.0 pips between those two lows, 3.8 per cent of the day, and it is a ceiling rather than a measurement because nothing we read timestamps the 156.69. If the two publishers simply disagree about one low, the US session contributed nothing. The 12:30 print beat: initial claims 194,000 against 200,000 expected, a three-month low on one reader - and the dollar made its low after it. The same release's continuing claims rose 25,000 to 1.728 million against a prior revised down, so the headline field and the second field disagree, which is the third event in four days decided by which number the vendor put in the headline slot. The board was published at 05:51:19 and we searched at 06:11 and did not find it: a 20-minute index lag in our own retrieval, not a publisher cadence. And yesterday's distance test fails: 157.50 carried $1.7bn five pips from spot at publication and expired 81.0 pips above the day's low. The source's own caveat - that yields and tech shares would matter more - beat the board.

The 73 Per Cent Held and the One Reachable Strike Gave Back 16 Pips. Neither Was the Day's Move.

Yesterday this desk published that 73.01 per cent of a $4.446bn dollar-yen option board sat at prices nothing visible would reach, and that 158.00 was the only strike where flow should be more than noise.

Both halves score. The day ran 157.84 to 158.39, 55.0 pips: 156.50 finished 134.0 pips below the low, 159.50 finished 111.0 above the high, 160.25 finished 186.0 above it, and $3.246bn expired unreached - 73.01 per cent, to the decimal. The 158.00 strike was reached, by 16.0 pips at about 12:10 UTC, and the pair closed back above it at 158.12 to 158.18. So the reachability test passed and bought nothing: 16 pips is 29.1 per cent of the day's range and it reverted, which means a binary rule fired correctly on a quantity that is not binary. The harder finding is the window. The low was set an hour and fifty minutes before the 14:00 cut as the ten-year fell to the 5.07 to 5.08 area, and the day's other high came four hours and forty minutes after it, on a wire report of Gulf embassy evacuations that the State Department then narrowed to Saudi Arabia alone. The day's two highs sit eighteen hours and four pips apart, both about 111 pips under the 159.50 strike. Both catalysts were outside the expiry window, one on each side. The suggestion is that a board with its mass two to three session ranges away belongs in a schedule as a veto, not a trigger.

We Look at 06:00. The Page We Come For Is Stamped 06:21.

This slot fires at 06:00 UTC.

The option expiry board it exists to read hits the wire at 06:01 and carries a page stamp of 06:21, so we are between one and twenty-one minutes early every trading day, and this morning there is no 19 October board to read at all. The generalisation: a publication hour is a property of the source, not of your schedule. Measured this morning, in UTC: the live expiry board at 06:21, the same house's European opening news at 05:45, a second expiry page about 01:23, a third ladder about 09:55, Tokyo broker notes at 22:09, 22:49, 23:15 and 03:16, Tokyo wire previews at 22:55 and 02:51, Tokyo desk notes from 00:03, a broker report at 23:22, positioning at 19:30 Friday with its best reader at 22:16, a government yield series at 20:16, a European spread series stamped to the 17:35 close, a European agency at 09:00 and US labour statistics at 12:30. Five of the eight Japanese-session pages are published after the Tokyo slot has already fired, and the pre-US slot fires sixty minutes before the US labour release. Two of today's best sources had never been cited here, purely because they publish at 02:51 and 03:16. And the table breaks twice in the next fortnight: Europe leaves summer time on 25 October, the United States on 1 November, seven days apart, so any filter keyed to a UTC hour is wrong twice in opposite directions.

We Carried 130 Basis Points for Five Runs. It Is a Thirteen-Point Range.

For five runs we carried the French-German ten-year spread at 130.3 basis points as though it were a level.

It was 140.6 on 2 October, 127.4 yesterday and between 135.5 and 138 this morning: a 13.2-point range in four sessions, with our figure 2.9 above the low. The German leg moved 2.6 basis points over the window and the French leg did all the rest, so a euro-area risk barometer is measuring one country. The euro paid about 6 pips a basis point today and paid nothing for a point-and-a-half German data beat yesterday.

The Governor Speaks This Afternoon. His Own Bank's Calendar Does Not Carry It.

The Bank of Japan's governor addresses the National Securities Convention in Tokyo this afternoon, and the only publisher carrying it gives the time three ways in one paragraph — 2:35pm Tokyo, 0635 GMT and 0235 US Eastern — two of which agree and one of which is an hour out. The Bank's own release schedule for this week does not carry the speech, and its monetary-policy speeches page stops at 10 September. An October hike reads roughly one in four on a single unnamed estimate, against a 1.25 per cent policy rate and a 29–30 October meeting confirmed at the Bank itself. The one properly documented event this morning is a ten-year auction of about 2,600 billion yen, announced by the ministry on 29 September.

Richmond Is 2, 3 or 5. Euro Confidence Cannot Agree on the Prior.

Three and a half hours before it prints, the Richmond Fed index is forecast at 2, 3 or 5 depending on your feed.

Euro area consumer confidence is worse — one calendar rounds the decimal off and turns a full point of expected deterioration into no change. Both claims we published yesterday rested on one feed.

Your Calendar Had Goolsbee an Hour Early Because It Read a Central Clock as Eastern

The Chicago Fed publishes two of today's three American events in Central Time.

One widely-read Monday round-up stamped both of them Eastern — and both arrived in your schedule exactly sixty minutes early. The Treasury bill auctions on the same page are correct, which is the control.

We Published a Prediction With a Ten-Minute Window. Nothing We Read Measures Ten Minutes.

Last night this desk published its first falsifiable prediction with a threshold and a deadline: no more than twenty pips in the ten minutes after Japan's 23:50 trade balance.

We cannot mark it. Our sources publish hourly, the Tokyo morning spanned forty pips, and the answer sits in the twenty-to-forty band we cannot resolve. Six and a half hours on we still could not read what Japan printed. And the ONS spent twenty-one minutes this morning serving two pages that disagreed about whether UK CPI existed — one saying "not yet published", the other carrying 3.1%.

Three Readers Want Empire State Between 14.0 and 15.0. The New York Fed Will Not Tell You Which Minute It Prints.

Empire State prints at 12:30 UTC and three readers give three consensus figures — 14.75, 15.0 and 14.0 — against an August prior of 20.6 that the New York Fed itself calls its highest reading in more than four years.

We publish the band and settle the prior at the issuing institution. And the New York Fed does not promise a release minute: its own note says "at or shortly after 8:30 a.m.", so a symmetric news window is the wrong shape. Plus yesterday's UK cut closed: total pay printed +3.9%, exactly on consensus.

The UK Jobs Number Was on the Wire at 06:01. At 06:26 the ONS Still Said It Was Not Published.

The ONS released the UK labour market report at 06:00 UTC and FXStreet had it on the wire ninety-six seconds later: unemployment 4.9% against a 5.0% consensus, unchanged in the three months to July.

At 06:26 the ONS's own release page still said "This release is not yet published" and Trading Economics' Actual field was still empty. This desk's rule — go to the issuing agency, not an aggregator — is right about authority and wrong about latency. Plus today's windows, the 17:00 auction clock now second-sourced, and a Fed hike at 92.4%.