spb/wp9_uqo Public

1# PAPER_REVIEW — Critical assessment before the scholarly upgrade (2026-08-05)23Scope: `paper/` (v1.1, 26 pp., 32 references). Assessment only — no results questioned.45## 1. Core contribution: is it clearly stated?67The empirical contribution is clear and genuinely stated in one sentence: *a national-scale8decomposition of Canadian house prices into structure vs. location using 1,153 absorbed FSA9fixed effects on 140,931 listings*. Three supporting claims (elasticity ≈ 0.55, bathrooms10≈ +11%, bedrooms ≈ 0; OOS R² 0.764; premia span ×9) are quantified and reproducible.1112**Weaknesses in the contribution statement:**13- The introduction motivates in three bullets ("first-order question", "granular data",14  "transparent benchmark") but never states *what is new relative to specific prior work*.15  No named paper is identified as the closest antecedent, so the gap is asserted, not shown.16- "Among the most geographically comprehensive single-equation hedonic exercises for17  Canada" is unverifiable as written because the Canadian empirical hedonic record18  (Des Rosiers/Thériault's Quebec program, Toronto and Vancouver studies) is never cited.19- The headline result — location ≈ 30 pp of R² — is never connected to the macro20  literature quantifying the land/location share of housing value (Davis–Heathcote;21  Knoll–Schularick–Steger), which is the natural external benchmark for its magnitude.2223## 2. Literature review: gaps by subfield2425Current coverage (good): Rosen framework and second-stage identification; semi-log26functional form; spatial econometrics vs. fixed effects debate; three amenity27capitalization classics; brief AVM motivation.2829**Missing subfields (each currently has zero citations):**30311. **Urban spatial structure / monocentric city.** The paper estimates an urban price32   gradient (§ Results 5.6) with no reference to Alonso–Muth–Mills, nor to modern33   quantitative treatments (Ahlfeldt–Redding–Sturm–Wolf; Duranton–Puga). The gradient34   section cites only Combes et al. (2015).352. **Land value and the location share.** Davis & Heathcote (2007), Knoll, Schularick &36   Steger (2017), Albouy's work on urban land values. Directly benchmark the 30-pp claim.373. **House-price indices.** The hedonic/repeat-sales index literature (Bailey–Muth–Nourse;38   Case–Shiller; Hill's survey) — the paper claims relevance for "house-price indices" in39   the first paragraph without citing any index paper.404. **Housing submarkets.** Goodman & Thibodeau; Bourassa et al. on submarket definitions —41   the FSA-fixed-effect design is a submarket argument and should engage this strand.425. **Spatial dependence in hedonics (empirics).** Dubin (1988); Basu & Thibodeau (1998);43   and the Canadian spatio-temporal work of Dubé & Legros. Moran's I is used but only44   Moran (1950) is cited.456. **Capitalization theory.** Tiebout (1956), Oates (1969), Roback (1982) — the FSA46   premium is interpreted as capitalized amenities with no reference to why amenities47   capitalize.487. **List price vs. transaction price.** The central data caveat (list prices) has no49   citation — Genesove & Mayer (2001) on seller behaviour, Han & Strange (2015) on the50   listing/search microstructure.518. **AVM / ML valuation empirics.** One citation (Mullainathan & Spiess 2017). Missing the52   applied mass-appraisal accuracy literature and the ML-vs-hedonic horse-race studies the53   OOS section implicitly competes with.549. **Canadian housing.** One CMHC report. Missing the Canadian hedonic tradition55   (Des Rosiers and coauthors on Quebec; Toronto/Vancouver-specific studies) and the56   supply-side explanations for Vancouver/Toronto price levels (Saiz-style constraints;57   superstar-city dynamics of Gyourko–Mayer–Sinai).5810. **High-dimensional FE methodology.** Correia (2017) is cited, but the intellectual59    origin (Abowd–Kramarz–Margolis) and the standard computational references60    (Guimarães & Portugal) are not.6111. **Sample selection / trimming practice.** The 1% trim and its robustness variants have62    no methodological anchor.6364## 3. Weak argumentation / unsupported claims6566- "Competitive with commercial automated valuation models" (§ Robustness, Conclusion) —67  no AVM accuracy benchmark is cited; needs at least one accuracy-standard reference68  (e.g., IAAO mass-appraisal standards or a published Zestimate-type accuracy figure).69- "Credible national estimates for Canada are scarce" — unsupported; requires citing what70  *does* exist to establish scarcity.71- Half-bath negative coefficient interpretation ("older or more compartmentalised floor72  plans") is speculative and unflagged — should be presented as conjecture or supported.73- "This is the housing counterpart of the within estimator" — fine, but the AKM/FE lineage74  should be cited where the claim is made.75- The bedrooms≈0 result is called "a textbook hedonic finding" without a supporting76  citation (Sirmans et al. tabulate mixed signs — an easy, honest anchor).77- Urban gradient "classic urban theory predicts" — no Alonso/Muth/Mills citation.7879## 4. Underdeveloped sections8081- **Introduction** (~1 page): thin on stakes (housing = largest household asset class;82  assessment/taxation machinery; index construction) and has no explicit numbered83  contribution list.84- **Related work** (~1.5 pages): organized adequately by theme but shallow within themes85  (2–4 citations each) and never says *where the gap is* at the end of each strand.86- **Discussion: absent.** Results flow directly into robustness and a short conclusion.87  There is no section interpreting magnitudes against prior literature (is 0.55 a typical88  elasticity? is an 11% bathroom premium high? is ×9 premia dispersion large vs. the US?),89  no engagement with findings that could contradict (e.g., submarket papers arguing90  *coefficients*, not just intercepts, vary by location — which the paper's own91  heterogeneity section partially confirms), and no consolidated Limitations & Future Work92  (currently squeezed into two conclusion paragraphs).93- **Methodology**: several choices uncited (trimming, FSA pooling threshold, kNN Moran94  implementation choice, quantile regression estimator details).9596## 5. Positioning: current vs. ideal9798**Current:** a well-executed measurement paper that reads as a technical report — it99presents itself mainly as "we ran a big transparent regression."100101**Ideal:** a contribution at the intersection of (i) the classic structure-vs-location102decomposition, (ii) the land-share macro literature, and (iii) the AVM benchmarking103debate — i.e., "the neighbourhood is the single largest priced component of Canadian104housing; we quantify it with a design that is transparent enough to audit and accurate105enough to rival black-box AVMs." The framing should name its closest antecedents (US106land-share estimates; submarket hedonics; ML-valuation horse races) and state exactly107what is new: national scope for Canada, listing-level granularity, absorbed-FE scale, and108a validated accuracy claim.109110## 6. Consequences for the upgrade (implemented in Phases 2–3)1111121. Add ~28 verified references across the 11 gaps above (target ≈ 60 total).1132. Restructure Related Work into thematic subsections, each ending with the gap.1143. Add a proper **Discussion** section: magnitudes vs. literature, agreements/115   disagreements, implications (assessment, taxation, indices, AVM), limitations & future116   work.1174. Deepen the introduction (stakes, numbered contributions, roadmap); strengthen the118   conclusion without overselling.1195. Ground every methodological choice in a citation; keep every number, figure and table120   untouched.121