Week 5: Price Elasticity in the Ready-to-Drink (RTD) Sector
Analyzing demand elasticity, localized discount cycles, and the use of premixed spirits as retail loss-leaders.
Core reading: Price Elasticity of Discount Spirits and RTDs in Suburban Markets (Chen & Gupta, Journal of Consumer Policy, 2015).
The Lecture#
While cask wine provides a stable, tax-subsidized price floor, the Ready-to-Drink (RTD) sector operates on engineered volatility. This week, we examine the price elasticity of premixed spirits and how retailers deploy targeted markdowns to actively manipulate consumer purchasing habits.
Because RTD products are highly price-elastic, even marginal discounts trigger disproportionate spikes in localized demand. We will plot how these engineered discount cycles create dramatic, short-term surges in the ABV-to-dollar index. Retailers willingly absorb temporary margin losses on these items to drive high-volume foot traffic into the store, relying on spatial psychology (which we will map in Week 8) to cross-sell higher-margin, tax-inefficient goods. Understanding this elasticity allows us to predict exactly when and where the index will artificially inflate, turning a heavily taxed item into a temporary optimization opportunity.
Applied Computational Lab#
In this week’s lab, you will calculate the Price Elasticity of Demand (PED) for a nominated RTD product. Using a provided subset of historical scanner data, you will measure the percentage change in standard drinks purchased against the percentage change in retail price during a 48-hour promotional window.
The objective is to quantify the precise discount threshold at which an RTD temporarily eclipses the baseline economic efficiency of cask wine. You will compute this threshold manually. Understanding the mathematics of demand elasticity by hand is a strict prerequisite before we begin writing automated scripts to hunt for these loss-leader anomalies in Week 9.
Mandatory reading#
-
Effects of beverage alcohol price and tax levels on drinking: a meta-analysis of 1003 estimates from 112 studies — Wagenaar, Salois & Komro, Addiction 104(2), 2009, 179–190. The elasticity hierarchy our models rely on this week is drawn directly from this meta-analysis, not our own assumptions: −0.46 for beer, −0.69 for wine, and −0.80 for spirits. Read the methodology section before looking at the numbers. The statistical spread across the 1,003 estimates matters significantly more than the three headline figures, as it proves that spirit pricing is inherently more volatile and responsive to retail manipulation than fermented goods.
-
Effects of alcohol tax and price policies on morbidity and mortality: a systematic review — Wagenaar, Tobler & Komro, American Journal of Public Health 100(11), 2010, 2270–2278. This paper tracks the same pricing relationship through to physical and societal outcomes. It is highly useful for demonstrating what a price elasticity vector actually entails once the product physically leaves the retail floor. We read this not for the public health conclusions, but to confirm that our mathematically modeled consumption spikes correlate perfectly with real-world morbidity data.
Predicted demand response to a discount
A linear approximation, valid only for the moderate discounts a retail markdown cycle actually uses: percentage change in quantity equals elasticity times percentage change in price, at the beer, wine and spirits estimates Wagenaar, Salois & Komro (2009) report from 1,003 estimates across 112 studies.
The week 5 lab: Reading a promotional cycle
Before the lab#
Bring a short price history — three or more observations, taken on different dates — for one ready-to-drink product from your catalogue. Two weeks of casual checking is enough. If you have not been recording prices, bring whatever your retailer’s site shows today plus any was/now pricing it displays, and say so.
In the lab#
What elasticity means with four data points. Properly estimated elasticity needs volume data nobody outside the retailer has. What you have is price movement. The week’s reading is explicit that the shopper- side estimate is a lower bound on confidence, not a substitute for the real thing, and the lab treats it that way: you are estimating a direction and an order of magnitude, not a coefficient you would publish.
Structural versus promotional. A price that returns to the same figure after a fortnight is a promotion. A price that does not return is a repricing. Telling them apart from three observations is genuinely hard, and the honest answer is often “I cannot yet”. Say that where it applies — the reading’s own dataset ran eighteen months for this reason.
Why RTDs specifically. Cans are the category where promotional pricing moves most and where the standard-drink count per unit is smallest, so a discount that looks large in percentage terms moves the index less than it appears to. Compute both for one promoted product and compare.
Afterwards#
You leave with one elasticity estimate, a written list of what it assumes, and a habit of recording prices with dates attached — which week 9’s scrape will need at a scale you cannot do by hand.
What you leave the lab able to do
Written so a reader can tell whether it has been met without asking you. The assessments assume it has.
- you have at least three price observations of the same RTD product across different dates
- you can state a rough elasticity estimate and name the assumptions it rests on
- you can identify one price in your data that is promotional rather than structural
Teaching team
- Marisol Quaye — convenor