Module 3: Consumer and luxury economics

Lesson video · Consumer and luxury economics (3:56)

View study sheet (PDF) View SQL cheat sheet (PDF)

Teaching this module? A facilitator guide adds a preparation checklist, timing cues, discussion guidance, slides, and a printable PDF.

This session uses auction and resale transaction data to teach two of the central metrics in luxury goods economics: pricing power and sell-through. Rather than relying on stated survey preference or brand-tracking indices, students work directly with what collectors and buyers actually paid at auction, across watches, jewelry, handbags, fine art, and other categories, drawn from more than 850 vendors with history to 1949. The session builds toward a single question that runs through the field: when demand for a brand strengthens, does it show up first in what people are willing to pay over the pre-sale estimate, in how reliably lots find a buyer, or both.

Target course(s) and level

Consumer behavior, luxury brand management, or marketing economics course. Suitable for undergraduate marketing and economics majors and for MBA students in a luxury or consumer goods concentration.

Learning objectives

By the end of this session, students will be able to:

  1. Define pricing power in the context of auction markets and compute it as a ratio of realized price to pre-sale high estimate.
  2. Define sell-through (clearance rate) as the share of offered lots that find a buyer, and explain why unsold lots must be counted, not discarded.
  3. Compare pricing power and sell-through across two or more brands and articulate what each metric captures that the other does not.
  4. Explain why auction data reflects revealed preference (what buyers actually paid) rather than stated preference (what buyers say they value).
  5. Identify a limitation of using auction data alone to characterize a brand's overall consumer demand.

Prerequisites

An introductory course in microeconomics or consumer behavior. No SQL or coding background is required; all queries are run using the sandbox's point-and-click interface with pre-written SQL provided by the instructor.

Materials and access needed

  • Sandbox access at sandbox.altfndata.com, self-registered with a work or school email, auto-approved.
  • Projector or screen share for the instructor demo.
  • The coverage browser and data dictionary tabs in the sandbox, used to orient students to available brands and vendors before the demo.
  • No instructor class API key is needed; this module is entirely no-code.

Session outline (75 minutes)

  • 0 to 15 min: Introduce pricing power and sell-through as concepts, using the Van Cleef and Arpels example (pricing power near 1.36x) as an anchor.
  • 15 to 20 min: Sandbox orientation for students who have not used it before.
  • 20 to 40 min: Guided demo, compute pricing power for two brands and compare.
  • 40 to 55 min: Guided demo, compute sell-through for the same two brands and compare.
  • 55 to 70 min: Small-group discussion, students interpret what the two metrics together suggest about each brand's demand.
  • 70 to 75 min: Wrap-up and homework assignment.

In-class demo (sandbox-first, no code)

  1. Open sandbox.altfndata.com, sign in, and select the jewelry data table from the SQL editor dropdown.
  2. Run the pricing power query below for Van Cleef and Arpels and record the resulting ratio.
  3. Change the designer filter to a second brand (for example, a competitor in the same category) and re-run the query. Record the second ratio and compare.
  4. Explain in plain terms what a ratio above 1.0 versus below 1.0 means for each brand.
  5. Switch to the sell-through query and run it for the same first brand. Note that this query requires counting both sold and unsold lots, so the WHERE clause does not filter on status.
  6. Re-run the sell-through query for the second brand and compare the two clearance rates.
  7. Open the pre-built charts tab and build a simple bar chart comparing pricing power for both brands side by side.
  8. Discuss as a class: does the brand with higher pricing power also have higher sell-through, or do the two metrics diverge?

Datasets and queries used

Dataset: jewelry data (documented fields: designer, model, item_title, sale_date, usd_price_decimal, sale_estimates_high_usd_price, status, vendor, stock_ticker).

Pricing power query:

SELECT
  designer,
  APPROX_PERCENTILE(usd_price_decimal, 0.5) AS median_realized,
  APPROX_PERCENTILE(sale_estimates_high_usd_price, 0.5) AS median_high_estimate,
  APPROX_PERCENTILE(usd_price_decimal, 0.5) / APPROX_PERCENTILE(sale_estimates_high_usd_price, 0.5) AS pricing_power
FROM all_jewels_gems_data
WHERE designer LIKE '%Van Cleef%'
  AND status = 'sold'
GROUP BY designer;

Sell-through (clearance rate) query, counting all offered lots regardless of outcome:

SELECT
  designer,
  COUNT(*) AS lots_offered,
  SUM(CASE WHEN status = 'sold' THEN 1 ELSE 0 END) AS lots_sold,
  CAST(SUM(CASE WHEN status = 'sold' THEN 1 ELSE 0 END) AS DOUBLE) / COUNT(*) AS sell_through_rate
FROM all_jewels_gems_data
WHERE designer LIKE '%Van Cleef%'
GROUP BY designer;

Second-brand comparison, same two queries with the designer filter changed, for example to a competing maison in the jewelry data or a comparable brand in the handbags data.

Discussion questions

  1. What does it mean economically for a brand's pricing power to sit above 1.0, and what does it suggest about how the auction house is setting its pre-sale estimates?
  2. Why must unsold lots be included in a sell-through calculation, and what would happen to the metric if they were silently dropped?
  3. Can a brand have strong pricing power but weak sell-through? What would that combination suggest about its collector base?
  4. How does revealed preference (what a buyer actually paid at auction) differ from stated preference (what a buyer says they would pay in a survey), and why does that distinction matter for a luxury brand's strategy team?
  5. What might explain a brand having very different pricing power in one category, such as watches, than in another, such as handbags?
  6. Auction data captures secondary market activity. What does secondary market demand tell you, and not tell you, about a brand's primary retail demand?
  7. If you worked in brand strategy for a luxury house, how would you use pricing power and sell-through together to decide where to focus marketing spend?
  8. What other data would you want alongside auction data to build a complete picture of a brand's consumer demand?

Homework assignment

Each student selects three brands within a single category (all three in watches, or all three in handbags, for consistency) and computes pricing power and sell-through for each using the sandbox SQL editor. The deliverable is a two-to-three page brief written as if for a brand strategy team, ranking the three brands by pricing power and by sell-through, explaining any disagreement between the two rankings, and recommending which brand shows the strongest evidence of collector demand. The brief must include the exported query results as an appendix and must not draw conclusions from recent quarter-over-quarter price changes, since recent-period record counts are not yet stable enough to read as a trend. Grading criteria: correct computation of both metrics (30 percent), quality of the comparative ranking and its justification (35 percent), appropriate handling of the sold-versus-unsold distinction (20 percent), and clarity of the written brief (15 percent).

Going deeper

Key terms

  • Pricing power: the ratio of median realized price to median pre-sale high estimate for a brand's sold lots; a value above 1.0 means buyers paid over the auction house's forecast.
  • Sell-through (clearance rate): the share of offered lots, sold and unsold combined, that found a buyer.
  • Revealed preference: what a buyer's actual purchase demonstrates about willingness to pay, as opposed to a stated preference expressed in a survey or interview.
  • Pre-sale estimate: the auction house's forecast of a lot's likely price range, captured in sale_estimates_high_usd_price, set before bidding begins.
  • Secondary market: the resale and auction market where a good changes hands after its original retail sale, as distinct from a brand's primary retail channel.
  • Collector base: the group of buyers actively competing for a brand's lots at auction, whose behavior pricing power and sell-through both attempt to describe from different angles.
  • Clearance: informal shorthand some auction professionals use for sell-through, worth knowing since students may encounter it in outside reading.

Common pitfalls

  1. Filtering the sell-through query to status = 'sold', which removes the unsold lots from the denominator and produces a clearance rate of 100 percent regardless of the brand's real performance.
  2. Treating one exceptional sale, a single record price, as proof that a brand's overall demand has shifted, rather than as one data point within a much larger set of lots.
  3. Comparing pricing power for a brand in one category against a different brand in another category, say a watch brand against a handbag brand, as though the ratio means the same thing across categories with very different estimate-setting conventions.
  4. Under-specifying the designer filter and quietly working with a small, unrepresentative slice of a brand's lots without checking the row count first.
  5. Reading the last one or two quarters of a brand's activity as a demand trend, when recent periods are still having records added and are not yet stable enough to compare period over period.
  6. Assuming pricing power and sell-through must move together. A brand can have strong pricing power, buyers pay well over estimate when they do buy, while still clearing only a modest share of what is offered, and the two metrics should be read side by side, not collapsed into one.

Additional queries to explore

Pricing power and sell-through computed together in a single query for one brand, with a guard against missing estimates, so students see both metrics built from the same base query:

SELECT
  designer,
  COUNT(*) AS lots_offered,
  SUM(CASE WHEN status = 'sold' THEN 1 ELSE 0 END) AS lots_sold,
  CAST(SUM(CASE WHEN status = 'sold' THEN 1 ELSE 0 END) AS DOUBLE) / COUNT(*) AS sell_through_rate,
  APPROX_PERCENTILE(
    CASE WHEN status = 'sold' AND sale_estimates_high_usd_price > 0
         THEN usd_price_decimal / sale_estimates_high_usd_price END, 0.5
  ) AS pricing_power
FROM all_jewels_gems_data
WHERE designer LIKE '%Van Cleef%'
GROUP BY designer;

Sell-through broken out by vendor for a single brand, to see whether one auction house clears a higher share of its lots than another for the same maison:

SELECT
  vendor,
  COUNT(*) AS lots_offered,
  SUM(CASE WHEN status = 'sold' THEN 1 ELSE 0 END) AS lots_sold,
  CAST(SUM(CASE WHEN status = 'sold' THEN 1 ELSE 0 END) AS DOUBLE) / COUNT(*) AS sell_through_rate
FROM all_jewels_gems_data
WHERE designer LIKE '%Van Cleef%'
GROUP BY vendor
ORDER BY lots_offered DESC;

The same combined pricing power and sell-through query run against the fine art data, to see whether the metrics read differently outside jewelry:

SELECT
  designer,
  COUNT(*) AS lots_offered,
  SUM(CASE WHEN status = 'sold' THEN 1 ELSE 0 END) AS lots_sold,
  CAST(SUM(CASE WHEN status = 'sold' THEN 1 ELSE 0 END) AS DOUBLE) / COUNT(*) AS sell_through_rate
FROM all_fine_art_data
WHERE designer LIKE '%Picasso%'
GROUP BY designer;

Extension activities

  • Extend the three-brand comparison from the homework assignment to a fourth or fifth brand within the same category, and discuss whether the ranking by pricing power and the ranking by sell-through stay consistent as the group grows.
  • Apply the same two metrics to a second category entirely, such as fine art or automobiles, and discuss whether the concepts of pricing power and sell-through translate as cleanly outside luxury accessories.
  • Have students bring in one piece of outside information about a brand, such as a recent price increase announcement or a retail expansion, and discuss informally whether the sandbox data is consistent with that outside signal.

Connections to other modules

Module 2, data science and SQL, is where the pricing power and sell-through queries used here are first constructed and explained step by step. Module 4, investments and equity research, takes a brand's pricing power and connects it to that brand's public equity, using the same underlying logic developed in this module. Module 11, auction theory and behavioral economics, offers a deeper account of why pricing power and sell-through can diverge for the same brand.