--- title: "The Price Frontier: From Discovery to Administration — A CANOPTICON Research Brief" kind: "Build brief" topic: "McLuhan & media theory" date: "2026-09-01" url: "https://canopticon.com/p/the-price-frontier-from-discovery-to-administration-a-canopticon-resea" author: "Chris Telford" license: "CC BY 4.0" --- # The Price Frontier: From Discovery to Administration — A CANOPTICON Research Brief ## TL;DR - The "relative-price divergence" (cheap TVs/software, expensive housing/health/education) is real, well-established, and mostly explained by tradability + Baumol's cost disease — not by a single conspiracy; in Canada the technology "wins" have NOT cleanly offset the housing/food "losses," because the risers are non-substitutable necessities that dominate household budgets and have outrun wages over 2000–2026. - The Canadian housing explosion is genuinely multi-causal: the defensible version of the immigration claim (demand from a 2022–2024 temporary-resident surge meeting a structural supply shortfall) is supported by the Bank of Canada and CMHC, but the strong "immigration was engineered to inflate prices" claim is not — ultra-low 2009–2021 interest rates, chronic under-building, and investor/speculative demand carry at least as much weight, and money-laundering was explicitly found NOT to be a primary cause. - The "managed economy" frontier is real but narrower than the thesis implies: algorithmic price-setting demonstrably exists (RealPage rent-setting, Amazon repricing, surveillance pricing) and algorithms can learn tacit collusion in theory and lab settings — but this is decentralized administration of prices by private firms, not Hayekian central planning being replaced by a single computer. The genuinely original McLuhan angle is to treat the price itself as a medium whose "ground" is shifting from discovery to administration. ## Key Findings 1. **The divergence is real and the mechanism is boring (in a good way).** Mark Perry's "Chart of the Century" shows that from 2000–2024, US prices for TVs, toys, software, and clothing fell while hospital services, college tuition, and childcare rose far faster than the ~90% overall CPI rise. The dominant academic explanation is not "government bad / market good" (Perry's implied reading) but **tradability** (goods face global competition and automation; services don't) and **Baumol's cost disease** (labour-intensive services can't raise productivity, so their relative price rises as wages rise economy-wide). Both are well-established. 2. **The "wins" do not cleanly cancel the "losses" — composition matters.** The pro-market "time-price" framing (Tupy/Pooley/Perry) argues that because US wages rose faster than CPI, almost everything became more affordable in work-hours. But this aggregates away the lived problem: the things that got cheaper are discretionary and substitutable (you buy a TV rarely), while the things that got dearer are non-substitutable necessities that dominate budgets (shelter, food, care). You cannot eat a cheaper television. 3. **Canadian food inflation is supply-shock-led but amplified by a concentrated oligopoly.** The "big five" — Loblaw, Empire (Sobeys), Metro, Walmart and Costco — account for 75% of grocery sales, after a series of mergers since 1986 reduced the number of major grocers from eight to five; the bread price-fixing scandal (2001–2015) is a documented, admitted cartel. But the Competition Bureau found gross margins rose only "one to two percentage points since 2017," and the strong "greedflation" thesis (Weber) is genuinely contested — best understood as "sellers' inflation": concentration gives firms the ability to pass through and coordinate price hikes around cost-shock cover, not a sudden outbreak of greed. 4. **Canadian housing: multi-causal, immigration is one real factor among several.** The Bank of Canada explicitly states recent immigration "is adding more to housing demand than to housing supply in the context of structural imbalances." The federal government acknowledged the temporary-resident surge in 2024 and cut targets. But the same sources attribute large roles to a decade of ultra-low interest rates, chronic under-construction (CMHC: ~3.5M-unit gap by 2030), zoning, and investor demand. The conspiratorial "engineered scarcity" reading is not supported. 5. **Algorithmic price-management is here, but it is distributed, not centralized.** RealPage's rent-setting algorithm produced a DOJ antitrust case settled in November 2025; Amazon reprices constantly; the FTC's January 2025 surveillance-pricing study found firms use personal data to set individualized prices. The Calvano et al. (2020) result that Q-learning algorithms learn to charge supracompetitive prices "without communicating" is robust in the lab but contested as to real-world prevalence. 6. **The McLuhan frontier is the genuinely ownable ground.** No one has cleanly articulated the price system as a *medium* undergoing a figure/ground reversal: from price-as-discovery (Hayek's decentralized signal) to price-as-administration (a centrally-computed instruction pushed at the consumer). That is the defensible CANOPTICON thesis. ## Details ### 1. The relative-price divergence: what the chart shows and what it hides Mark J. Perry (American Enterprise Institute) created the "Chart of the Century," tracking US BLS price changes by category since ~2000. Over January 2000 to end-2024, the overall CPI rose roughly 90% while average hourly wages rose about 123%. Within that, the split is stark: hospital services, college tuition, childcare, and (in recent years) food and housing rose far above the average line, while TVs, software, toys, clothing, cars, and furnishings fell below it — several (TVs especially, ~98% nominal decline after quality adjustment) collapsing in price. **What it actually shows.** The chart is a genuine and well-replicated description of relative-price divergence. Mike Konczal's reconstruction confirmed the broad pattern using 62 matchable BLS categories — though he noted the single steepest riser in his version was "delivery services," and that hedonic quality adjustment explains much of the TV collapse (a $300 TV in 2000 didn't become a $6 TV; it became a much better TV at a similar price, which the index records as deflation). **The contested interpretation.** Perry and popularizers (Marc Andreessen reposted it as "Why AI Won't Cause Unemployment") imply a simple story: blue lines = free market, red lines = government/regulation. This is the seductive-but-overstated reading. The mainstream economics explanation is two structural forces: - **Tradability.** Tradable goods (those that can be made abroad and shipped) face global competition and automation, which drives prices down; non-tradable services (a haircut, a hospital stay, a daycare spot) are produced and consumed locally and face no such discipline. The US BLS itself replicated New Zealand's tradable/non-tradable indexes and confirmed tradables show lower inflation than non-tradables. This is mostly orthogonal to regulation — a deregulated haircut is still non-tradable. - **Baumol's cost disease.** William Baumol and William Bowen (1960s) showed that in sectors where productivity can't easily rise (live music, teaching, nursing, medicine), wages still rise in line with the high-productivity economy, so unit costs — and relative prices — climb. The University of Chicago Booth IGM panel found 88% agreement (confidence-weighted) that "rising productivity in manufacturing leads the cost of labor-intensive services—such as education and health care—to rise." This is "probably not obvious to non-economists but has been demonstrated to be true." Health, education, and arts are the textbook cases — exactly the chart's red lines. Regulation, subsidy structures (third-party insurance payment, student loans inflating tuition), and market concentration are real secondary amplifiers, but the structural spine is tradability + Baumol. ### Does it net out? Canada and the US, 2000–2026 The user's intuition — that cheaper tech is cancelled by dearer housing/food — is half right, and the half that's wrong is important. **The optimistic case (real but limited).** The Human Progress / Gale Pooley "time-price" analysis argues that because US hourly wages rose ~123–131% against ~88–93% CPI, the "time price" of the average basket fell ~16%, meaning ~20% more abundance per hour worked. On this framing even housing and food became more affordable in work-hours. In Canada, the OECD reports real wages began recovering post-pandemic (Q1 2025 real wages +1.9% y/y), and Statistics Canada's distributional accounts show disposable income rising across quintiles. **Why it doesn't capture the lived experience.** Three problems: - **Composition and substitutability.** The cheapened goods are durables and discretionaries bought rarely; the dearer items are recurring necessities that dominate budgets. A household cannot reallocate a cheaper TV into rent. The basket's *weighted* experience is dominated by shelter, food, and transport, which is why "everything feels more expensive" even when an aggregate index says purchasing power rose. - **Distribution.** Baumol himself noted the cost-disease price increases "disproportionally affect the poor," because necessities are a larger budget share for low-income households. Statistics Canada's Q4 2025 data show the lowest-income quintile's net saving *worsened* as housing/utilities/insurance costs outpaced their income gains, while the top quintile pulled ahead on investment income — the income gap widened to 46.7 points. - **The Canadian wage record is weaker than the US.** The "time price" optimism rests on US wage data. Canadian real wage growth has been thinner; Canada's affordability deterioration (below) is sharper than the US, so the "wins offset losses" claim is materially weaker in the Canadian case the user actually cares about. **On official vs lived inflation (Billion Prices Project).** The MIT Billion Prices Project (Cavallo & Rigobon, started 2008) scrapes millions of online prices daily. Its central finding cuts *against* the "official CPI is a lie" intuition: in countries with honest statistics, online price indexes closely track official CPI (and can anticipate turning points months early). The dramatic divergence case was Argentina, where the BPP showed real inflation roughly double the government's manipulated figure 2007–2015. So: official Canadian/US CPI is broadly accurate as an *average*; the "everything is more expensive" feeling is not measurement fraud but the composition/distribution problem above, plus the salience of frequently-purchased items (groceries, gas, rent) over rarely-purchased cheapening ones. ### 2. Food and housing specifically **Food.** Canadian grocery inflation 2020–2026 was led by genuine global supply shocks (pandemic supply-chain breakage, the Ukraine war, climate-driven crop losses, energy/input costs) — the Centre for Future Work found carbon pricing contributed only ~0.15% to food inflation, debunking that political talking point. But Canada's grocery sector is a documented oligopoly: per the federal government's 2026 food security strategy, the "big five" — Loblaw, Empire (Sobeys), Metro, Walmart and Costco — account for 75% of grocery sales, following a series of mergers since 1986 that reduced the number of major grocers from eight to five. The **bread price-fixing scandal** is the hard evidence of coordination — a 2001–2015 industry-wide scheme; Canada Bread was fined $50M (2023, the largest price-fixing fine in Canadian history) and Loblaw/George Weston settled a class action for $500M (approved May 2025). The Competition Bureau's June 2023 study, "Canada Needs More Grocery Competition," concluded the sector "is not as competitive as it could be" and found gross margins rose "by one to two percentage points since 2017" — modest, not the excess some alleged. This maps onto the **"greedflation" / "sellers' inflation" debate.** Isabella Weber (UMass Amherst) and Evan Wasner argue (2023, *Review of Keynesian Economics*) that the COVID inflation was "predominantly a sellers' inflation" — firms with market power hike prices when they expect competitors to do the same, using "sector-wide cost shocks and supply bottlenecks" as implicit coordination cover. Critics (Justin Wolfers and others) note that since mid-2022 corporate profits fell while prices kept rising, which embarrasses a pure-greed story. The honest synthesis: "greedflation" as *subjective greed* is wrong (Weber herself rejects the label); "sellers' inflation" as *structural pass-through-plus-coordination enabled by concentration* is the defensible version and is partly corroborated (IMF, ECB found profits a large share of European inflation 2021–2022). This is genuinely contested terrain — flag it as such. ### 3. Canadian housing — pressure-testing "engineered demand" The user's claim: a $1M 2008 home becoming $3–4M is primarily because immigration manufactured demand to drive prices up. Let me separate the defensible core from the overstated shell. **The price facts.** The Canadian property bubble ran roughly 2002–2022; the Dallas Fed rated Canadian real estate "exuberant" from 2003; some cities saw up to 337% price growth 2003–2018. Toronto rose 21.7% in the single year 2016. By 2022 UBS ranked Toronto the world's most bubble-risk city. So the user's order-of-magnitude (a 3–4x rise over ~15 years in hot markets) is broadly accurate for detached homes in Toronto/Vancouver, though the national average rise is smaller. **Factor-by-factor weight (mainstream evidence):** - **Interest rates (2009–2021) — very large.** A decade-plus of ultra-low rates inflated borrowing capacity and asset prices; the 2022+ rate shock (BoC policy rate to 5.0%) then crushed affordability and flatlined/reversed prices, demonstrating rates' causal force. The National Bank affordability monitor shows mortgage-payment-to-income hit crisis levels (>50% nationally; far higher in Vancouver/Toronto) before easing as rates fell to 2.25% by end-2025. This is arguably the single biggest driver of the *level*. - **Supply shortfall / zoning — very large and structural.** CMHC's September 2023 update maintained that Canada needs ~3.5 million additional units beyond current trends by 2030 to restore affordability (benchmarked to "levels last seen around 2004"); about 60% of that gap is in Ontario and British Columbia, "because housing supply hasn't kept up with demand over the past 20 years in some of the largest urban centres." CMHC's June 2025 reframing escalated this to up to 4.8 million new homes over the next decade — 430,000–480,000 starts/year by 2035, roughly double the ~245,000 actual in 2024. The Parliamentary Budget Officer's competing estimate is far smaller (~1.3 million units, benchmarked to 2019 affordability rather than 2004) — a genuine expert disagreement on *magnitude*, not direction. - **Immigration / population — real, recent, and acknowledged.** This is where the user's claim has a defensible core. Canada's non-permanent resident population went from <1% (2000) to 6.8–7.6% (2024, ~3.1M people), with temporary residents driving ~60% of 2023's record population growth. The Bank of Canada (Dec 2023 staff note) found immigration "is adding more to housing demand than to housing supply in the context of structural imbalances." The government explicitly acknowledged the surge and in October 2024 announced the first-ever caps on temporary residents (target: below 5% by 2027), stating it expects to "alleviate some pressure in the housing market," reducing the supply gap by ~670,000 units by 2027. **But:** the IRCC's own research review notes high prices "persisted in Canada for decades, even during periods of low immigration," and that immigration also *adds* construction labour. The post-2022 surge intensified an existing crisis; it did not create it. Crucially, the timing defeats the strong claim — the bulk of the price explosion (2002–2021) *predates* the 2022–2024 temporary-resident surge. So immigration is a real recent *amplifier of demand*, but it cannot be the *primary* cause of a price run that was already 15+ years old when the surge began. - **Foreign capital / money laundering — real but explicitly secondary.** The BC Cullen Commission (final report June 2022, 1,800+ pages) found money laundering in BC is "enormous" and that real estate is "highly vulnerable" — but it explicitly concluded money laundering is **NOT the cause of housing unaffordability.** An earlier expert panel estimated laundering contributed roughly a 5% increase to BC house prices. So foreign/illicit capital is a real distortion at the margin, not a primary driver. - **Domestic investor/speculative demand — large and rising.** By 2020, 21% of all Toronto housing and 56% of Toronto condos were investor-owned; in Vancouver ~33% of all housing and ~48% of condos. Investors increasingly crowd out first-time buyers. This is a major demand-side force largely independent of immigration. **Verdict on the user's claim.** The *defensible* version: "the post-2022 immigration/temporary-resident surge added significant demand to an already supply-starved market, the government admitted this, and it pushed prices and rents higher at the margin." True and well-sourced. The *overstated/conspiratorial* version: "immigration pipelines were manufactured primarily to drive up housing prices." Not supported — it inverts cause and effect (immigration policy was driven by labour-market and demographic goals, with housing pressure an acknowledged side-effect, not the aim), ignores that the price run predates the surge by 15 years, and underweights interest rates, chronic under-building, and investors, which the evidence treats as at least co-equal drivers. Honest framing: immigration is one real demand lever among several; the binding long-run constraint is supply. ### 4. The "managed economy" frontier The user's central thesis: as we manage attention with algorithms, are we increasingly managing the financial/broader economy with algorithms? Mapping the real state: **(a) Dynamic / personalized retail pricing.** Amazon was found (Profitero, 2013) to make more than 2.5 million price changes per day — versus ~50,000 *per month* for Best Buy or Walmart at the time. (Caveat: this is a 2013 figure for Amazon's own first-party pricing engine, widely cited since but now over a decade old; cite it as such.) The frontier has moved to **personalized/surveillance pricing**: the FTC's surveillance-pricing study (press release January 17, 2025) found that "details like a person's precise location or browser history can be frequently used to target individual consumers with different prices for the same goods and services," based on 6(b) orders to Mastercard, Accenture, PROS, Bloomreach, Revionics and McKinsey & Co. Then-Chair Lina Khan: retailers set prices "from a person's location and demographics, down to their mouse movements on a webpage." New York's Algorithmic Pricing Disclosure Act (effective July 8, 2025) now requires the disclosure "This price was set by an algorithm using your personal data." **(b) Algorithmic rent-setting — RealPage.** This is the strongest real-world case. The DOJ and eight states sued RealPage (August 2024) alleging its revenue-management software let competing landlords share confidential data and coordinate rents — the original complaint called it "a modern way to violate a century-old law." The DOJ filed its proposed settlement on November 24, 2025 (no admission of liability; appointment of a monitor; restrictions on using competitors' nonpublic data); Antitrust Division chief Gail Slater said "RealPage was replacing competition with coordination, and renters paid the price." Greystar, the largest US landlord, paid $50M to settle a class action plus $7M to nine states; landlord defendants collectively paid over $200M in class settlements. Cities (San Francisco, Philadelphia, Minneapolis, Berkeley) banned the practice; New York legislated against it. **Canadian parallel:** I found no equivalent enforcement action — a genuine open question and a possible CANOPTICON instrument target (is RealPage-style software in use in Canadian rental markets, where the same landlord-concentration dynamics exist?). **(c) Algorithmic collusion theory.** Calvano, Calzolari, Denicolò & Pastorello (*American Economic Review*, 2020) showed experimentally that Q-learning pricing algorithms "consistently learn to charge supracompetitive prices, without communicating with one another," sustained by "collusive strategies with a finite phase of punishment followed by a gradual return to cooperation." Assad et al. found empirical support in German retail gas stations (algorithmic adoption associated with higher margins). This is real and important — but heavily contested as to (i) whether lab results generalize, (ii) whether the behaviour is "genuine" collusion or spurious, and (iii) how regulators can even detect it. The regulatory response (OECD, FTC, DOJ) is nascent. **(d) High-frequency / algorithmic trading.** HFT accounts for roughly 50–60% of US equity trading volume (down from a ~70% peak in 2009–2010) and ~35–43% in Europe; broader algorithmic trading was estimated at ~85% of US market volume (2012). Markets are already, overwhelmingly, machine-mediated at the transaction layer. This is the most complete example of "the economy run by algorithms" — but it is execution/liquidity provision, not allocation of real resources. **(e) Central-bank models / nowcasting.** Central banks increasingly use nowcasting and machine-learning models to read the economy in real time and set policy — a form of algorithmic *macro-management*. (I was unable to deepen this with fresh sourcing within budget; treat as a known but lightly-evidenced strand here, and a candidate for follow-up.) **(f) The theoretical question — Hayek vs. the People's Republic of Walmart.** Hayek's "The Use of Knowledge in Society" (1945) framed the price system as a decentralized information mechanism: no central board could aggregate the dispersed, local knowledge that prices automatically summarize. The revival argument — Phillips & Rozworski's *The People's Republic of Walmart* (2019) — holds that Walmart and Amazon already run vast internally *planned* economies using computation, suggesting big-tech-scale compute makes central planning newly feasible, reviving the 1920s–30s socialist calculation debate. **Pressure-test:** the strongest critique (well-argued by reviewers) is that Walmart/Amazon solve *logistics and distribution of final goods* — inventory, routing, scheduling — which Mises himself conceded a planned economy could do; they do **not** solve the actual calculation problem, which is the *allocation of capital/production goods* in the absence of factor markets. So the "computation defeats Hayek" claim is seductive but overstated: what's demonstrably happening is private firms administering prices *within* a market system, not a central computer replacing the market's allocative function. The price is increasingly *administered* rather than *discovered* — but administration is distributed across many private algorithms, not centralized. ### 5. The McLuhan / media-theory angle This is where CANOPTICON can own genuinely original ground, because the economics literature does not think in these terms. **The price as medium.** McLuhan's "the medium is the message": the *content* (the specific dollar figure) is the meat the burglar carries to distract the watchdog of the mind; the *structural change* (how the price is formed and delivered) is the real message. A price discovered through decentralized markets is a different *medium* than a price computed by an algorithm and pushed at you — even when the number is identical. The shift from discovery to administration is a medium-change that the "it's just a price" framing renders invisible. **Figure/ground and the new ground.** Classically, the market is the *ground* and individual prices are *figures* moving against it. The novel claim: the pricing algorithm is becoming the new *ground* — the environment within which all prices form — while the market recedes into background myth. We still narrate "the market set this price," but increasingly an algorithm did. This is precisely McLuhan's figure/ground reversal: the service (instant optimized prices, "personalization") is the figure we notice; the disservice (loss of price-discovery, individualized extraction, tacit coordination) is the subliminal ground. **The tetrad of the pricing algorithm** (a buildable CANOPTICON artifact): - *Enhances:* speed, granularity, personalization of price; firms' ability to extract consumer surplus. - *Obsolesces:* the posted/uniform price; haggling; the price as a shared public signal; arguably price-discovery itself. - *Retrieves:* the bazaar — individualized, negotiated, opaque, person-specific pricing that the fixed price-tag (a 19th-century innovation) had abolished. (This is the sharpest McLuhan move: surveillance pricing is the *return of the haggle*, but with the asymmetry inverted — the seller now knows everything about the buyer.) - *Reverses into:* algorithmic collusion / a privately-administered command economy — the "free market" pushed to its computational extreme flips into something resembling central planning, but corporate and opaque rather than public and accountable. **Attention economy → allocation economy.** Dan Shipper (Every, "The Knowledge Economy Is Over. Welcome to the Allocation Economy," January 18, 2024) coined "allocation economy" to describe how AI turns workers from makers into managers/allocators of model output: "You won't be judged on how much you know, but instead on how well you can allocate and manage the resources to get work done." CANOPTICON can productively *extend and repurpose* this term: where Shipper means humans allocating AI labour, the deeper political-economy move is that *allocation itself is being delegated to algorithms* — the same engagement-maximizing logic that governs what we *see* (attention economy) is migrating to govern what we *pay and get* (an allocation economy in the resource-distribution sense). The parallel is exact and ownable: recommendation algorithms shape attention; pricing/matching algorithms shape allocation. Note the term's provenance honestly — Shipper's original framing is about labour, not resource allocation, so this is an extension, not a citation. ### 6. Synthesis — understanding + frontier **The mechanical understanding (so the dynamic is clear):** Relative prices diverge because tradable goods get cheap (global competition + automation) while non-tradable, labour-intensive services get dear (Baumol). Necessities (shelter, care, food) are disproportionately non-tradable and Baumol-afflicted, so they rise relative to wages and dominate budgets — which is why aggregate "purchasing power rose" coexists with "everything feels unaffordable." On top of this structural spine sit amplifiers: concentration/market power (groceries), monetary policy (rates inflating then deflating asset prices), supply constraints (housing), and demand surges (immigration, investors). Price *formation* is now increasingly mediated by algorithms — at the transaction layer (HFT), the retail layer (dynamic/surveillance pricing), and the rent layer (RealPage) — shifting prices from discovered signals toward administered instructions. **Three frontier thesis/instrument candidates:** **Candidate A — "The Return of the Haggle: Surveillance Pricing as Medium-Change" (STRONGEST).** - *Novelty:* High. Framing personalized/surveillance pricing as a McLuhan figure/ground reversal — the retrieval of the pre-fixed-price bazaar with inverted information asymmetry — is original and defensible. Nobody owns this framing. - *Already done:* The economics (FTC study, RealPage) and the McLuhan theory exist separately; the synthesis does not. - *Public data/evidence:* FTC 2025 surveillance-pricing study; RealPage filings; NY disclosure law; Calvano et al. Strong. - *Buildable instrument:* A Cloudflare-native tool that detects/documents price variation by user signal (location, device, history) across Canadian e-commerce — empirically testing for surveillance pricing in Canada, where (unlike the US) there is little enforcement or even documentation. High defensibility; genuinely novel data. **Candidate B — "RealPage North: Algorithmic Rent-Setting in Canada" (STRONG, EMPIRICAL).** - *Novelty:* Medium-high. The US story is well-covered; the *Canadian* question — is RealPage/YieldStar-style software setting rents in Toronto/Vancouver/Hamilton, and what would that mean given Canadian landlord concentration? — is under-investigated. - *Already done:* US litigation extensively; Canadian angle largely absent. - *Public data/evidence:* US DOJ case as template; Canadian rental data; corporate landlord disclosures. Moderate (the core empirical question requires original investigation — ideal for an instrument). - *Defensibility:* High if framed as investigation, not assertion. Flag clearly that you have not yet confirmed Canadian usage. **Candidate C — "Price Without Discovery: The Hayek Inversion" (INTELLECTUALLY DEEPEST, HIGHEST RISK).** - *Novelty:* High conceptually, but crowded by the *People's Republic of Walmart* discourse. - *Already done:* The socialist-calculation revival is active; the McLuhan overlay is not. - *The trap:* It is easy to overstate this into "central planning is back / Hayek was refuted." The evidence does NOT support that — administration is distributed across private firms, and the calculation problem (capital allocation) remains unsolved by Walmart-style logistics. Treat as essay/theory, not empirical claim. - *Defensibility:* Defensible only if it explicitly resists the overstatement — the honest thesis is "prices are increasingly administered rather than discovered, by many private algorithms, which is neither free-market discovery nor public central planning but a third opaque thing." That nuance is the contribution. **Seductive dead ends / overstated claims to avoid:** - "Official inflation is fake / a lie." The Billion Prices Project evidence says no (for honest-statistics countries). The real story is composition and salience, which is more interesting and defensible. - "Immigration was engineered to inflate house prices." Inverts cause/effect; defeated by timing (price run predates the surge). Use the defensible demand-amplifier version. - "Money laundering caused the housing crisis." The Cullen Commission explicitly says it did not. - "Pricing algorithms are already colluding to run the economy." Lab-true (Calvano), real-world-contested. State the uncertainty. - "Big-tech compute has solved central planning / refuted Hayek." Overstated; conflates logistics with capital allocation. - "Cheaper tech offset dearer housing, so people are actually better off." True only in aggregate work-hours and weaker in Canada than the US; ignores distribution and necessity-weighting. ## Recommendations **Stage 1 — Publish the framing essay (now).** Lead with Candidate A ("The Return of the Haggle"). It needs no new data, synthesizes well-sourced material (FTC 2025, RealPage, Calvano, Baumol, Hayek) under an original McLuhan lens, and pressure-tests the immigration and greedflation claims honestly — which is CANOPTICON's differentiator. Anchor it on the price-as-medium / figure-ground-reversal argument. **Stage 2 — Build the detection instrument (next).** A Cloudflare-native tool that probes Canadian e-commerce/rental listings for price variation keyed to user signals. *Benchmark that would justify escalation:* if you can document statistically significant price differences by location/device/history on even one major Canadian retailer or rental platform, you have an original, publishable empirical finding and a strong instrument — escalate to a full investigation. *Threshold to abandon:* if probes show uniform pricing across signals, pivot to documenting the *capability* and regulatory gap rather than claiming active practice. **Stage 3 — Investigate the Canadian RealPage question (parallel).** FOI/Competition Bureau inquiries, landlord-tech vendor research, rental-listing analysis. *Benchmark:* confirmed use of revenue-management pricing software by a major Canadian landlord = a significant story warranting standalone treatment. **What would change these recommendations:** - If a Canadian regulator (Competition Bureau) opens an algorithmic-pricing case, Candidate B becomes urgent and lead-worthy. - If new academic work resolves the algorithmic-collusion "genuine vs spurious" debate toward real-world prevalence, Candidate C gains empirical legs and can move from essay to investigation. - If StatsCan/BoC publish a clean decomposition attributing housing price growth to factors, update the Section 3 weights accordingly. ## Caveats - **The greedflation/sellers'-inflation debate is genuinely unresolved.** I have presented the defensible synthesis; reasonable economists disagree on magnitude. Do not present "sellers' inflation" as settled. - **Housing factor-weights are contested in magnitude, not direction.** CMHC (~3.5M gap, rising to up to 4.8M over the next decade in its 2025 reframing) and the PBO (~1.3M gap) disagree substantially because they target different affordability baselines (2004 vs 2019). Present ranges. - **The Amazon "2.5M price changes/day" figure is from 2013** and refers to first-party pricing; cite it as a historical benchmark, not a current statistic. - **Algorithmic collusion is lab-demonstrated, not field-established.** Calvano et al. is robust experimentally; real-world prevalence and detectability remain open. - **The "allocation economy" term is Dan Shipper's (2024) and originally describes human management of AI labour**, not algorithmic resource allocation. CANOPTICON's usage is an extension; attribute honestly. - **Central-bank algorithmic management (4e) is under-sourced here** due to research-budget limits; verify before building on it. - **The McLuhan framings are interpretive, not empirical.** Their value is as lenses that generate testable instrument concepts, not as claims of fact. - **"Time-price" abundance optimism rests on US wage data;** the Canadian real-wage record is weaker, so the "wins offset losses" claim is materially softer for Canada than the cited US figures suggest.