EconPlot › Explainers › Why egg prices fell from $6 to $2

Why did egg prices fall so far?

Updated 2026-09-06 · how we source every figure

The price fell 64.8 percent between March 2025 and July 2026

A dozen eggs cost $6.227 in March 2025 and $2.189 in July 2026. Dividing the later figure by the earlier one gives a fall of 64.8 percent, which rounds to about two thirds.

Both figures come from one specific measurement, not from a general sense that eggs got cheaper. That measurement is the average price of eggs, grade A large, per dozen, US city average, collected by the Bureau of Labor Statistics — the federal agency that measures prices and employment — and distributed as series APU0000708111 through FRED, short for Federal Reserve Economic Data, the public database run by the Federal Reserve Bank of St. Louis. "US city average" means one national figure assembled from prices recorded in cities across the country. It is not the price in any one store, and no individual shopper need ever have paid exactly that amount.

March 2025 is not merely a local high point. It is the highest value in the file as read on 2026-09-03, which runs from January 2019 through July 2026.

The fall was steady rather than sudden. April 2025, the first month after the peak, was still $5.122 — already well down, but nowhere near the end. Every reported month from then on sits below the peak, and the descent runs through the whole of 2025 and into 2026. The cheapest month since the peak is June 2026 at $2.141, marginally below July 2026's $2.189; that small step up from June to July is the one monthly rise in the descent, and even with it the price stays far below where it started. One gap is worth knowing about: October 2025 has no value in the file as read on 2026-09-03, so the series does not cover every month of that span.

Bird flu destroyed about 127 million laying hens by March 2025

Bird flu forced United States farms to destroy about 127 million egg-laying hens between February 2022 and March 2025. That count is in Congressional Research Service report R48518, which draws on confirmed-detection data from the Animal and Plant Health Inspection Service, the arm of the US Department of Agriculture that handles animal disease; the same report puts birds of all kinds affected at more than 166 million. A February 2026 trade report of the same agency's data put losses of hens and pullets — pullets being young hens not yet laying — above 150 million.

Highly pathogenic avian influenza, the disease behind those losses, is a virus that spreads rapidly through a poultry flock and kills the birds it infects. It was first confirmed in a US commercial flock in February 2022, and the outbreak has continued since.

Birds counted as affected are largely not birds killed by the virus running its course. When the disease is confirmed in a flock, the agency's response is depopulation: destroying the remaining birds to stop the infection spreading further. That response is why losses arrive in large, sudden blocks rather than trickling. Report R48518 counts birds affected, not causes of death, so this page puts no number on the split between infection deaths and culls.

The flock was never short by 127 million at any one moment, because that figure is cumulative over four years and farms restocked throughout. The standing shortfall was far smaller. Before the outbreak, in 2020 and 2021, the United States held about 310 million table-egg layers in an average month, according to the Department of Agriculture's Economic Research Service. Across 2022 to 2024 that average was 296 million — about 14 million fewer hens laying in an average month.

Why losing a share of the hens pushed the shelf price up so much

A modest shortfall in eggs produced a very large rise in the shelf price, because shoppers keep buying eggs at almost any price. That is inelastic demand: when the quantity available drops, the price has to climb a long way before enough buyers step back for the remaining cartons to go around.

The sequence runs in more than one step, and each one matters:

• Flocks are destroyed, so the number of hens laying falls.
• Fewer hens produce fewer eggs. Monthly table-egg production, measured by the Department of Agriculture's National Agricultural Statistics Service, fell from 8,132.2 million eggs in December 2024 to 6,823.0 million in February 2025 — the lowest month in that record.
• With fewer cartons arriving, grocers no longer need to discount to clear their stock, and buyers who want eggs bid against each other for what is there.
• The wholesale price — what grocers pay producers, before the store adds its own markup — fell as the shortage eased: large eggs in New York averaged 533 cents a dozen in March 2025 and 154.82 cents a dozen in March 2026.

The wholesale series and the retail series APU0000708111 are two different measurements, and this page holds no source measuring how long the shelf price takes to follow the wholesale one, or comparing how far each one moves.

The production rise and the price fall do not line up arithmetically, and they are not supposed to. Production in July 2026, at 7,995.4 million eggs, is 17.2 percent above the February 2025 trough, while the retail price is down 64.8 percent. A sixth more eggs, and a price two thirds lower. That gap is the inelasticity, not an error in either figure.

One step in this chain is an assumption rather than a measurement: the assumption is that shoppers wanted about the same number of eggs throughout. No measure of egg demand is cited here, so that assumption is the textbook mechanism applied to the case, not something these sources establish.

By August 2026 the flock was bigger than before the outbreak

The national flock had passed its pre-outbreak size by 2026, and the price kept falling over the same months. The National Agricultural Statistics Service counted 307 million table-egg layers on May 1, 2026, and 315 million on August 1, 2026. Set against the 2020 to 2021 average of 310 million and the 2022 to 2024 average of 296 million, the August 2026 count is a full recovery rather than merely growth from the bottom.

Two things had to happen for that recovery. First, fewer new infections: the Economic Research Service's Livestock, Dairy, and Poultry Outlook of April 2026 reports fewer new bird flu detections in the first quarter of 2026 than in the first quarter of 2025. Second, enough young birds to replace the losses; the same report finds replacement pullets sufficient in 2026 to cover both ordinary flock turnover — hens are replaced on a routine cycle anyway — and the birds lost to the disease.

More hens then showed up as more eggs. That April 2026 report puts February 2026 table-egg production 5.1 percent above February 2025, on a layer inventory 5.9 percent larger than a year before. By July 2026, monthly production was 7,995.4 million eggs, 4.7 percent above the 7,633.3 million of July 2025. With more cartons arriving than buyers would take at the high price, grocers have to discount to clear their stock, and the shelf price comes down — the same inelastic demand that drove the price up, running the other way.

The monthly production figures bounce around for a reason that has nothing to do with bird flu: short months contain fewer laying days. February 2026 shows 7,136.0 million eggs against 7,891.0 million in January 2026, a drop that is mostly the calendar. The trend across many months, not the month-to-month jitter, is what tracks the price.

The price took nine months to rise and sixteen months to fall below where it started

The rise was roughly twice as fast as the fall. The price bottomed at $2.715 in June 2024 before the climb, and peaked at $6.227 in March 2025 — nine months. From that peak to the $2.189 of July 2026 is sixteen months, and that July 2026 figure sits below the June 2024 level the climb began from.

Depopulation arrives in large, sudden blocks. The birds are gone, and the eggs they would have laid stop arriving almost immediately, which is why the quantity of eggs produced drops sharply into February 2025.

Rebuilding runs on the biology of the birds. Replacement hens have to be raised before they lay, so a farm cannot restore its output by deciding to. The report that records replacement pullets as sufficient to cover both routine turnover and the disease losses is the April 2026 outlook; the sources here do not pin down when that sufficiency began, only that it held in 2026.

One consequence for a reader looking at a chart: the price peak and the production trough are not in the same month. Production bottomed in February 2025 at 6,823.0 million eggs, and the price peaked the following month, March 2025, when production had already recovered to 7,492.6 million. Production changes direction first — falling before the price rises, and recovering before the price falls. These sources record both series but do not establish what governs the offset between them.

No law brought the price down, but the government did spend money

Nothing in the record shows a law or a price control bringing the price of eggs down. Between March 2025 and July 2026 no Act of Congress and no price cap on eggs was passed, and the Federal Reserve — the United States central bank, which sets short-term interest rates — does not set the price of groceries.

The government was not idle, though. On February 26, 2025, the Department of Agriculture announced a plan of up to $1 billion aimed at the outbreak itself. The plan funded stronger farm biosecurity to keep the virus out of barns, larger indemnity payments to farms that lost their layers, and short-term egg imports from Turkey, South Korea and Brazil to cover the gap on shelves.

The distinction is worth being precise about. The plan's own aim was to restore the flock, and the Economic Research Service attributes the 2026 rebound in production to the larger layer inventory. So the part of the chain the evidence covers runs through the birds: more hens, then more eggs. The last step — more eggs, then a lower price — is the inelastic-demand mechanism set out above, not something these sources measure. What the record also does not contain is any figure putting a share of the recovery on the plan rather than on farms rebuilding for their own commercial reasons, or on the outbreak simply abating. The honest summary is narrower than either "the government fixed it" or "the government did nothing".

The series behind this page

The same data the video plots, from the same file. Point at the line to read a value off it.

What this page does not establish

• Series APU0000708111 measures one narrow item — eggs, grade A, large, per dozen. It reports no separate figure for organic, cage-free or pasture-raised eggs, so a reader who takes 'eggs fell by two thirds' as covering the carton they actually buy may be reading a fall that this series never measured.
• October 2025 has no value in the price file as read on 2026-09-03. Anyone counting consecutive monthly declines, or drawing an unbroken line through the descent, is working over a hole in the data rather than a measured month.
• July 2026 is the latest published month in the price series as read on 2026-09-03, with the August 2026 release due in mid-September 2026. Bureau of Labor Statistics figures can also be revised, so the exact size of the 64.8 percent decline could shift with later releases even though the direction is not in doubt.
• The price file as read on 2026-09-03 begins in January 2019. Calling March 2025 a record is a statement about the months from January 2019 onward only, and a reader who takes it as an all-time high for eggs is extending the claim past the data used here.
• The monthly table-egg production numbers are raw monthly totals, and short months contain fewer laying days: February 2026's 7,136.0 million eggs against January 2026's 7,891.0 million is mostly the calendar, not a collapse in supply. A reader comparing adjacent raw months will invent outbreaks and recoveries that never happened; only the trend across many months carries the story.
• The step from 'fewer eggs' to 'higher price' assumes that shoppers wanted about the same number of eggs throughout. No measure of egg demand is used here. If demand also fell over 2025 and 2026, part of the price decline belongs to buyers rather than to hens, and these sources cannot separate the demand side from the supply side.
• The production recovery and the price fall move together, but nothing here quantifies how much of the price fall the extra eggs caused. Feed costs, fuel, wages and grocery margins all sit between a hen and a carton on a shelf, and none of them is measured in this record.

Questions people also ask

Sources

The evidence behind every claim above, as recorded when the piece was made.

peak price march 2025

BLS Average Price Data via FRED, series APU0000708111 (eggs, grade A, large, per dozen, US city average), local file data/APU0000708111.csv read 2026-09-03: March 2025 = 6.227 dollars, the maximum of the whole series back to January 1980; narration, title and stamps round to $6.23

price july 2026

Same series APU0000708111, observation 2026-07-01 = 2.189 dollars, the last row in the file as read 2026-09-03 (the August 2026 CPI release is due mid-September 2026, so July 2026 is the latest published month at render time); narration, title and stamps round to $2.19

fell by two thirds

Derived from the two BLS values: 2.189 / 6.227 = 0.352, a fall of 64.8 percent, which the narration, stamp and chart badge round to 'two thirds'

sixteen months

Same series: March 2025 to July 2026 is sixteen months; no month between them is at or below 2.189 (the only lower value is June 2026 at 2.141), so the fall to the July 2026 level genuinely took the full span

fallen for more than a year

Same series: every reported month from April 2025 (5.122) through July 2026 (2.189) is below the March 2025 peak and the path is downward overall; the one small monthly rise (June 2026 2.141 to July 2026 2.189) stays far below the peak, and the narration says the price 'has fallen ... since', not that it fell every single month. October 2025 is blank in the source file

spike took months

Same series: 2.715 in June 2024 (the trough before the rise) climbing to the 6.227 peak in March 2025, nine months; the balance-scene stamp '9 months to spike' and narration 'within months' rest on this

fall took more than a year

Same series: from the March 2025 peak to July 2026 is sixteen months, longer than the nine-month rise; the narration says 'over a year', the stamp says '16 months to fall', and the balance headline 'up fast, down slow' compares the two spans

cull 100 million hens

Congressional Research Service report R48518, 'The Highly Pathogenic Avian Influenza (HPAI) Outbreak in Poultry, 2022-Present' (congress.gov), citing USDA APHIS confirmed-detection data: since February 2022 the outbreak affected over 166 million birds, of which about 127 million were egg layers, as of March 2025. WATTPoultry, reporting APHIS data in February 2026, put US hen and pullet losses above 150 million. Both are layer-specific counts, not all-poultry totals, so 'more than 100 million egg-laying hens' is a floor, not an estimate. 'Destroy' is the plain word for depopulate/cull, and 'to stop the disease' is the reason APHIS gives for depopulation. 'US' is spelled out as 'United States' in the scene before

since 2022

USDA APHIS: the first commercial-flock detection of the current HPAI outbreak in the United States was in February 2022 (CRS R48518, same source as above)

bird flu cause

USDA ERS chart of note 'Avian influenza outbreaks reduced egg production, driving prices to record highs in 2022' (ers.usda.gov, chart 105576) and ERS chart of note 103873 'Recent avian flu outbreak reduces U.S. egg-laying flock' name HPAI-driven layer losses as the driver of the egg price records; USDA ERS Livestock, Dairy, and Poultry Outlook July 2025 (LDP-M-373) describes 2025 prices as descending from HPAI-related highs. The narration names bird flu as step one on that basis

supply demand mechanism

Textbook supply-and-demand explanation: with demand unchanged, a fall in supply raises the price; the narration states this rule and applies it to eggs. 'Shoppers still wanted the same cartons' is the plain-words form of 'demand did not fall', an assumption of the explanation, not a measured claim, and the scene is presented as the mechanism step, not as data

outbreaks slowed

USDA ERS Livestock, Dairy, and Poultry Outlook, April 2026 (LDP-M-382, ers.usda.gov): 'there were fewer new detections of HPAI during the first quarter of calendar year 2026 than there were during the first quarter of calendar year 2025', and replacement pullets were sufficient in 2026 to cover both regular flock turnover and HPAI-related losses. This is the basis for 'outbreaks slowed' and 'farms raised new chicks'

flocks rebuilt

USDA NASS 'Chickens and Eggs', August 2026 release (ckeg0826): 315 million table-egg layers on August 1, 2026, and the May 2026 release (ckeg0526): 307 million on May 1, 2026, both up about 4 percent on the year. USDA ERS chart of note 103873 gives the pre-outbreak 2020-2021 average as 310 million table-egg layers per month and the 2022-2024 average as 296 million, so the August 2026 count is back above the pre-outbreak level: 'rebuilt' means recovery, not only growth. ERS LDP-M-382 (April 2026) reports February 2026 table-egg production up 5.1 percent on the year on a 5.9 percent larger layer inventory, which is 'more hens, more eggs'

price came down with supply

ERS LDP-M-382 (April 2026) pairs the larger layer inventory and higher table-egg production with wholesale egg prices far below a year earlier (New York large eggs averaged 154.82 cents in March 2026 against 533 cents in March 2025); the retail BLS series APU0000708111 fell from 6.227 to 2.189 over the same recovery. The narration links the two as 'more hens, more eggs, and the price came down'

no law did that

Checked against the record: no Act of Congress and no price control on eggs was passed between March 2025 and July 2026, and the Federal Reserve's decisions do not set egg prices. USDA did act by executive spending: the February 26, 2025 plan of up to $1 billion (usda.gov press release) funded biosecurity, higher indemnity for lost layers and short-term egg imports from Turkey, South Korea and Brazil. The narration therefore says only 'no law, no vote, no price cap' and 'no law brought the price down, more hens did', which is consistent with that plan: the plan's own aim was to restore the flock, and ERS attributes the 2026 production rebound to the larger layer inventory (LDP-M-382). The script does not claim government did nothing; the earlier on-screen 'Not policy' was removed for that reason

supply shock definition

Standard economics term: a supply shock is a sudden change in the available quantity of a good that moves its price; the narration glosses the term in plain words in the same sentence Reworded 2026-09-03: the line said 'Economists call this a supply shock', which is a claim about what others say, and evidence-coverage holds those to a named, dated source that this entry does not have — it says only 'standard economics term'. The attribution is dropped rather than propped up with a citation we do not hold: the video now says 'This is a supply shock', which is the definition itself and needs no one to have said it.

chart

BLS via FRED, series APU0000708111, monthly, from June 2024; October 2025 is missing in the source file and the builder skips blank rows

wallet link

Explanation only: the ending describes how a supply shock moves a grocery price up and back down; no advice on buying, stocking up or timing purchases

eggs spiked then fell

Same series: 2.715 in June 2024, 6.227 in March 2025, 2.189 in July 2026; no month after March 2025 rose back toward the peak (the highest post-peak value is April 2025 at 5.122, then lower every month except the June to July 2026 uptick)

punchline and length

Owner rule 2026-09-03 ('we aim for memorable and not for completeness'): the video ends on 'no law brought the price down; more hens did' and every scene after it was cut. Narration 848 -> 747 characters (66s -> 58s at the measured 12.9 chars/second). No claim was added; cut scenes only removed material.

number discipline

Owner rules 2026-09-03 (at most three numbers, none repeated, at most two significant digits, first slide provokes, last slide is the punch line). Numbers kept: $6.23 and $2.19 a dozen (scene 1; cents are the story), down two thirds (scene 2). Dropped '100 million hens', the sixteen-month stamp and the spike/rebuild end card, so the video ends on the punch line.

cover headline length

Owner catches 2026-09-03: cover headlines were being cut off where they ran under the artwork. The Cover component now sizes the headline to the band above the image, and the headlines themselves were cut to one provoking line.

rounded prices

Owner, 2026-09-03: 'why so many significant digits? Egg cost $6. Now $2.' The exact BLS figures stay in this record — March 2025 = 6.227 dollars, July 2026 = 2.189 dollars, series APU0000708111 — but on screen and in the narration they are 'about 6 dollars' and 'about 2 dollars'. This overrides the earlier carve-out that let retail prices keep their cents.

cover is one line

Owner, 2026-09-03: 'we just need catchy top phrases... nobody can read [the rest] in a fast scrolling environment, so just remove them'. The cover is the headline plus the artwork; the stamps and the bottom kicker bar are no longer rendered at all.

two model review

Fixed 2026-09-03 from findings that BOTH reviewers (Claude and DeepSeek, run separately by scripts/validate_script.py) raised on the same line. 'The chart tells it' had no antecedent for 'it'; 'moved'/'slowed' were directionless where the video's whole claim is a direction. Directions taken from data/APU0000708111.csv (6.227 in March 2025 to 2.189 in July 2026) and the USDA flock-rebuild record already cited.

doccard retired

Owner, 2026-09-03: 'why we need doc card? when we already have another card like this'. The doccard scene type rendered a headline on a bare paper card with no artwork at all, which is newscard minus the picture; it was the only text-only type in the renderer and every project had unused card art sitting beside it. DocCard was deleted from remotion/src/MarketShort.tsx and these scenes are newscard now. NewsCard took over its two extras: the kicker renders above the headline, and a highlight that is a whole sentence not lifted from the headline renders as a caption bar over the artwork instead of painting headline words yellow. Not one word of narration or on-screen text changed, so the script fingerprint and its validation still stand.

deepseek missing price step

DeepSeek review 2026-09-03, a finding Claude passed. Four findings, all the same hole: the script jumped from 'fewer eggs' to 'the price went up' and from 'more eggs' to 'the price came down' with nothing in between, leaving the viewer to supply the step. For a channel written for zero economics background that is the one step that cannot be assumed. Each leap now names the mechanism in plain words — scarce eggs let stores charge more and still sell out; full shelves make stores cut prices to clear stock. Scene 4 also adds the link from bird flu to culling (the disease kills chickens and spreads through a flock). No figure changed and no new claim about the data enters; these are the causal steps the supply-and-demand sources already carry. Trimmed the same day: adding those steps took the narration from about 760 to 840 characters and the render from roughly 59 to 64.5 seconds, over the 60-second gate in STRATEGY.md. Cut back to 765 characters (~59s) by tightening the three sentences the fix had written, not by removing anything that was there before — every causal step it added is still spoken: the disease spreads and kills, scarce eggs let stores charge more, full shelves make them cut prices.

length trim round two

Second trim 2026-09-03. Both reviewers found the video still restating itself: scene 3's 'That carton just taught you how prices move' is a bridge to a scene that opens 'Step one' on its own, and the supply-shock attribution above added words for a claim we could not source. Narration 765 -> 707 characters, about 55 seconds against the 60-second gate. No scene was cut, so the chart evidence and every causal step stay in.

egg supply measured nass

Owner rule 2026-09-05: the central claim must be backed by data the viewer can check, and it is not limited to FRED - owner, same day: 'we can pull reliable data source and quote it. We don't need to limit to FRED.' The claim 'the price fell because the number of eggs rose' previously had only the price half; the supply half was asserted. USDA NASS, Chickens and Eggs, table eggs produced during the month, million eggs, pulled by scripts/fetch_nass_eggs.py into data/TABLE_EGGS.csv, 31 months 2024-01 to 2026-07: trough 6,823.0 in February 2025 (bird flu), 7,492.6 in March 2025 (the price peak month), 7,633.3 in July 2025, 7,995.4 in July 2026 - 17.2 percent above the trough and 4.7 percent year on year. The chart now draws supply beside price on its own axis, so the viewer sees output dip before the price spike and recover before the fall. No percentage is spoken: output is up 17 percent while the price is down about two thirds, and those two numbers do not explain each other arithmetically - egg demand is inelastic, so a modest supply swing moves price hard. The narration says what recovered and when, which is what the chart shows. The supply line is a three-month centred average: February has about ten percent fewer days than January, so raw monthly production saws up and down by more than the recovery it is meant to show, and the first render read as noise. The averaging changes no value materially - the February 2025 trough and the 2026 recovery are the same - it removes a calendar artefact.

sentence caption removed

Owner, 2026-09-05, on the staged drafts: 'the label ... should be removed. Way too many words and too little time to read ... they look like a caption, and is redundant with the caption almost. can we just please remove this type of text.' The caption bar was added on 2026-09-03 so the retired doccard's second sentence had somewhere to live; that was the wrong instinct. A short already burns in word-by-word captions, so a second block of prose over the artwork competes with the text the viewer is reading, in a scene lasting a few seconds. Removed from the renderer and from these scenes: scene 4: Eggs got scarce, so stores could charge more.; scene 7: A supply shock: fewer hens, fewer eggs, scarcer eggs, higher prices.

scene art is scene specific

Owner, 2026-09-05, on the staged draft: 'the graphics seem wrong. why we reuse this image a lot? I don't like this image - not saying much briefcase vs balloon. if I look at the image alone I can't tell what it wants to say.' card-balance.jpg is a leftover asset from the hero video in assets/hero/, carrying no prompt in any project - so nobody chose it for this scene, it was borrowed because it existed, and a brass scale holding a briefcase against a hot-air balloon depicts nothing in particular. It was also the single most reused picture in the library, appearing in two unrelated videos. Replaced with artwork generated for this scene. Scene 5 ('Fewer hens. Fewer eggs. Same shoppers.') now shows a nearly empty supermarket egg shelf with shoppers reaching for the last cartons, which is the sentence.

chart axes labelled

validations/chart-axes.md, owner catch 2026-09-05: 'why is there no x-axis and no y-axis labels? add, and add a validation rule. x and y must be clearly labeled.' The renderer drew gridlines with no values on them at all, so every chart in the library showed a shape rather than a measurement - a viewer could not tell a two-point move from a two-hundred-point one. ChartLine now writes the value at the top, quarters and bottom of the plot from yMin/yMax, with a unit from the new yUnit field, and takes an optional xLabel for the horizontal axis. A second defect surfaced with it: xTicks are generated on year changes, so a series that never crosses a year boundary produced NO horizontal labels - which is exactly what happened to the 2026-only jobs chart. build_video.py now falls back to month names in that case.

on screen text swept

validations/on-screen-brevity.md sweep, 2026-09-05. The dominant defect was one habit, not many: the headline restating the scene's own narration, so the card read aloud what the burned-in captions were already spelling out. A headline should be the label a viewer can take in at a glance, saying the angle rather than the sentence. Also fixed: highlights appearing nowhere in their headline (they paint nothing), stamps repeating a figure the headline already carries, and headlines of three sentences. The cover ran to three sentences and scenes 3, 5 and 6 each restated their narration; scene 3's stamp and scene 6's first stamp repeated figures the headline already carried.

inner scenes carry captions only

Owner, 2026-09-05: 'Look at competition. I don't see headlines for inner non cover slides.' Checked against the footage rather than taken on trust: sampled frames from two Economics Explained videos - our closest faceless model - show inner slides carrying the rolling narration captions and NOTHING else. No headline, no kicker, no stamp stack. The exceptions are deliberate beats, a section title card or one large number, never a stack. Measured across our own library, inner scenes averaged 2.4 on-screen elements EACH on top of the captions. That is the root of the on-screen-brevity findings: 48 of them across eleven videos, most of the form 'headline restates the narration'. Rewriting 48 headlines would have treated the symptom. Inner scenes now carry captions plus at most one stamp, and only when the stamp holds a figure the viewer might want to read off the screen. The cover and the stakes card keep their text - they are the scroll-stopping frame and the opening beat - and chart scenes keep their titles, axes and source lines, which are the evidence.

colour magazine style

Owner, 2026-09-06: 'use color instead of black and white so it looks like a color magazine cut out. I am guessing that people in 50s have seen black-and-white newspaper, but younger generation sees it a little bit less frequent.' Measured on one card generated three ways: the old newspaper halftone reads 41 saturation / 48 colourfulness (Hasler-Susstrunk), the colour magazine cut-out 84 / 74, a two-colour risograph 125 / 83. Economics Explained, the closest faceless model, runs 91 / 54 and our library averaged 37 / 37 - so the magazine style lands on the field while the risograph overshoots it and reads as illustration rather than photography. All 5 image prompts in this project were switched and the artwork regenerated.

colour prompt corrections

Owner review of the first colour set, 2026-09-06, two defects, both in the generated image rather than the crop. (1) 'Color of eggs' - the carton came back with mint green, pale blue and sage eggs. Going to colour made the generator treat a familiar object as part of the palette; the prompt now names 'ordinary white and light brown chicken eggs, natural eggshell colours only' and confines the saturated print colours to the background collage. (2) 'in the shelf image why there are segments of people cut at the rib cage' - the raw file has a coat cut off at the ribcage with no head, plus a second disembodied arm. The prompt now asks for one shopper shown in full from behind, 'whole figure visible, no cropped torsos and no disembodied limbs'. Both corrections are house rules now, not one-off fixes: a colour prompt has to pin the real colour of anything a viewer knows the colour of, and has to say how people are framed. (3) Owner, same review: 'the eggs are unusually big compared to the size of the shopper'. The carton in her hands was nearly the width of her torso. Third house rule from the same family: when two objects of known size share a frame, state the relationship - 'a standard twelve-egg carton that is small in her hands, about the width of her forearm, correct real-world proportions'. The generator gets the objects right and their relative scale wrong, and scale is the error a viewer notices without being able to name it. Sweep after the checks were added: card-hen-barn and card-chicks named hens and chicks without pinning their colours; card-egg-hourglass paired whole eggs with chicks at unstated scale AND illustrated a sentence about supply recovering with a metaphor about time, so it was replaced with a restocked shelf; scene 3 was the third consecutive scene on card-egg-carton and now has its own shelf-edge price rail.

motion after the hook

Owner, 2026-09-06: 'I remember we had a rule for interleaving image with video. Especially for the second slide.' The rule was real - HERO-ANALYSIS records 'something must visibly change every ~2 seconds' from measuring Economics Explained, which alternates clipping cards with real b-roll every ~3s - but it had never been written into the writer's instructions, and this video had EIGHT still scenes and no clip at all. Scene 3, the first beat after the stakes card and the point where a scroller decides, is now a tracking shot down an emptying egg aisle. Rule 6 in make_project.py and validations/scene-artwork.md both enforce it now.

no marks in artwork

Owner, 2026-09-06: 'costco card usually have costco on it. is it a fair use?' Naming a company in the narration is ordinary editorial use and this channel depends on it; rendering its mark is a different act, and a generated mark is always a wrong mark - an approximation that misrepresents the logo rather than reporting on it, on top of platform brand complaints that are decided without adjudicating anything. The Costco card was already specified as 'a plain white plastic membership card'. The real exposure was incidental branding nobody had thought of as a brand shot: supermarket shelves, storefronts, packaged goods, and a container port whose boxes carry shipping-line marks. 42 of 63 prompts in the library lacked the clause; every image and video prompt now ends 'no text, no logos, no brand names'.

silent outro card

Owner, 2026-09-06: 'Can we add a call to action slide at the end ... for users to follow my econplot videos to understand what moves your wallet.' Implemented as the one shape this channel's own rules allow. validations/punchline.md: 'A silent end card (no words) after the punch line is allowed - it is a held frame, not more video', so the punch line still lands last in the NARRATION and nothing follows it aloud. validations/editorial-register.md bans comment-bait specifically ('comment X and I'll send you', 'link in bio for') and says a plain editorial ending is fine - a channel name and what the channel is about is neither bait nor financial advice, which is what STRATEGY.md's 'no call to action' was written to exclude. validations/scene-artwork.md permits shared artwork for 'a channel-level element (a logo, an outro card)', so one card serves every video rather than each generating its own. The line is topic-tied where the topic earns it - 'what moves your mortgage' on the bond video - and the house line otherwise. 1.8 seconds, no narration, no audio.

Related

• Inflation cooled — so why did my bill still go up?