📈 economics

Price elasticity of demand for cigarettes in Spain: A segment-specific analysis (2005–2025)

This study analyzes Spanish cigarette sales from 2005 to 2025 to reveal that while tobacco taxation significantly reduces overall consumption with a short-run price elasticity of −0.800, demand is most responsive among high-price brands, highlighting the critical need for policies that narrow price gaps to prevent smokers from switching to cheaper alternatives.

Original authors: Gema Aonso Diego, Ángel García Pérez, Andrea Krotter

Published 2026-07-08

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Gema Aonso Diego, Ángel García Pérez, Andrea Krotter

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine the Spanish cigarette market as a massive, bustling marketplace where people buy packs of cigarettes every month. This study acts like a detective, looking at official sales records from 2005 to 2025 to answer one big question: How much does the price tag on a pack of cigarettes actually stop people from buying them?

Here is the story of what the researchers found, broken down into simple concepts.

1. The "Price Tag" Test

The researchers wanted to know: If the government raises the price of cigarettes (like turning up the volume on a radio), do people turn down the volume on their smoking?

They found that yes, people do buy less when prices go up, but not everyone reacts the same way.

The Overall Result: For every 10% increase in price, total cigarette sales dropped by about 8%.
The Long Game: This effect gets even stronger over time. If prices stay high, people eventually cut back even more (a drop of about 10.6% for a 10% price hike).

2. The Three "Cigarette Buckets"

The most interesting part of the study is that the researchers didn't just look at "cigarettes" as one big group. They sorted them into three buckets based on how expensive they are: Low-price, Medium-price, and High-price.

Think of it like buying coffee:

The High-Price Bucket (The Fancy Espresso): These are the expensive brands. The study found that people who buy these are very sensitive to price. If the price goes up, they are quick to stop buying or switch to something else. For a 10% price hike, sales in this bucket dropped by a massive 11.6%.
The Low- and Medium-Price Buckets (The Regular Coffee): These are the cheaper brands. The people buying these are stubborn. Even if the price goes up, they keep buying almost the same amount. For a 10% price hike, sales only dropped by 2% to 4%.

The Analogy: Imagine a heavy rainstorm (a price hike). The fancy umbrellas (high-price brands) are easily put away or replaced because people have other options. But the cheap, sturdy umbrellas (low-price brands) are the only ones some people can afford, so they keep using them even if they get a little more expensive.

3. The "Law" vs. The "Price"

The study also looked at two major laws passed in Spain (one in 2006 and a tougher one in 2011) that banned smoking in more places.

The Result: The laws helped reduce smoking, but the price hikes were the real heavy lifters.
The Metaphor: Think of the laws as a fence that makes it harder to smoke in public. But the price hike is like making the fuel for the car (cigarettes) so expensive that people simply decide not to drive at all. The study suggests that while the fence is helpful, the high cost of fuel is what really stops the car from moving.

4. The "Down-Trading" Trap

Here is the tricky part the researchers warn about. Because the cheap cigarettes are so "stubborn" (inelastic), if the government raises the price of all cigarettes equally, smokers might not quit. Instead, they might just switch from the expensive brand to the cheap brand.

The Scenario: A smoker who usually buys a "High-Price" pack sees the price go up. Instead of quitting, they might just buy a "Low-Price" pack instead.
The Consequence: The total number of cigarettes smoked doesn't drop as much as hoped because people are just trading down to the cheaper bucket.

5. What the Researchers Say We Should Do

Based on these findings, the authors suggest a specific strategy:

Don't just raise prices; close the gap. To stop people from simply switching to the cheap bucket, the price difference between the expensive and cheap brands needs to shrink. If the cheap brand becomes almost as expensive as the fancy one, people have no cheap escape route, and they are more likely to quit entirely.
Use the money wisely. The study suggests that the money the government collects from these taxes should be put back into helping people quit (like funding support programs), which creates a double win: fewer people smoke, and those who want to quit get help.

Summary

In short, this study tells us that raising cigarette prices works, especially for people who buy expensive brands. However, to really win the battle against smoking, we need to make sure the "cheap escape route" isn't too easy to use. If we make the cheap cigarettes expensive enough, people won't just switch brands; they'll actually stop buying.

Technical Summary: Price Elasticity of Demand for Cigarettes in Spain (2005–2025)

Problem Statement
While increasing tobacco prices through taxation is widely recognized as the most cost-effective intervention for reducing consumption, existing literature on the price sensitivity of cigarette demand in Spain suffers from significant limitations. Previous studies rely on outdated data, self-reported household surveys prone to recall bias, or hypothetical purchase tasks. Furthermore, there is a scarcity of research distinguishing how price elasticity varies across specific brand price segments (low, medium, and high). The existing evidence for Spain suggests relatively low sensitivity (elasticity estimates ranging from -0.10 to -0.54), but these findings may not reflect current market dynamics or the heterogeneous responses of different consumer segments.

Methodology
This study utilizes a comprehensive, longitudinal dataset provided by the Spanish Ministry of Finance (Comisionado para el Mercado de Tabacos), covering monthly sales data for conventional cigarettes from January 2005 to December 2025. The analysis excludes the Canary Islands, Ceuta, and Melilla due to distinct fiscal policies.

Data Processing: Financial variables were adjusted for inflation using the Consumer Price Index (base year 2025). Variables (consumption, price, and income) were transformed into natural logarithms to facilitate elasticity interpretation.
Segmentation: Cigarette brands were classified into low-, medium-, and high-price segments based on nominal price terciles. Brands were assigned to the segment they occupied most frequently over the study period.
Econometric Model: The authors estimated a log-linear dynamic demand model. The specification includes:
- Dependent Variable: Log of monthly cigarette sales.
- Key Independent Variables: Log of cigarette prices and log of wages (proxy for income).
- Controls: Dummy variables for two major tobacco control laws (Act 28/2005 and Act 42/2010), a COVID-19 lockdown dummy, a linear time trend, and monthly fixed effects.
- Dynamic Component: A lagged dependent variable ( $Q_{t-1}$ ) was included to capture habit persistence.
Estimation Techniques:
- Short-run elasticities were derived directly from the price coefficients.
- Long-run elasticities were calculated using the formula $\beta / (1 - \rho)$ , where $\beta$ is the short-run price coefficient and $\rho$ is the lagged consumption coefficient.
- Newey–West standard errors were applied to correct for heteroskedasticity and serial correlation.
- Instrumental variable (IV) estimation (Two-Stage Least Squares) using specific excise tax as an instrument was employed to address potential price endogeneity; Durbin–Wu–Hausman tests confirmed the exogeneity of prices in the baseline models.
- Segment-specific differences were tested using Seemingly Unrelated Regression (SUR) and Wald tests.

Key Results

Overall Elasticity: The study estimates an overall short-run price elasticity of demand of -0.800. This indicates that a 10% increase in price leads to an approximate 8% reduction in consumption. The long-run elasticity is estimated at -1.06, suggesting that demand becomes more elastic over time as consumers adjust their habits.
Segment Heterogeneity: Significant heterogeneity exists across price segments:
- High-price brands: Exhibit the highest responsiveness with a short-run elasticity of -1.163.
- Low-price brands: Show an elasticity of -0.350.
- Medium-price brands: Show the lowest responsiveness with an elasticity of -0.220.
- Statistical tests confirm that the high-price segment is significantly more elastic than both the low and medium segments, while the difference between low and medium segments is not statistically significant.
Policy Impact:
- Act 28/2005: Associated with a significant reduction in sales in the preliminary analysis, but this effect was attenuated when price and other covariates were included in the full model.
- Act 42/2010: Showed a strong, significant negative association with sales in preliminary models. However, in the full model controlling for price, the direct effect of the legislation was not statistically significant, suggesting that concurrent price increases (taxation) may be the primary driver of reduced consumption.
- COVID-19: The lockdown period (Q2 2020) was associated with a temporary decrease in purchases.
Income and Persistence: Higher wages were associated with increased consumption, particularly in medium and high-price segments. Consumption persistence (habit formation) was strongest in the low-price segment (coefficient 0.708), indicating higher addiction levels or fewer substitution options for this group.

Significance and Contributions
The paper claims to provide the first segment-specific analysis of cigarette price elasticity in Spain using official administrative sales data over a 20-year period. Its primary contributions include:

Methodological Rigor: By utilizing tax-paid sales data rather than self-reported surveys, the study avoids under-reporting biases, potentially explaining why the estimated elasticity (-0.800) is higher than previous Spanish estimates (which ranged from -0.10 to -0.54) and closer to broader European ranges.
Segment-Specific Insights: The findings challenge the assumption of uniform price sensitivity, revealing that smokers of high-price brands are significantly more responsive to price hikes than those in lower-priced segments.
Policy Implications: The authors argue that while taxation is the most effective tool, the inelasticity of low- and medium-price segments suggests a risk of "down-trading" (smokers switching to cheaper brands) if price gaps are not managed. Consequently, the paper advocates for tax policies that minimize price differentials between segments to prevent substitution and maximize public health impact. It further suggests that tax revenues should be reinvested in cessation programs to support vulnerable populations and enhance equity.

Limitations Acknowledged by the Authors
The authors note several constraints: the use of aggregated data prevents analysis of individual sociodemographic heterogeneity; the exclusion of specific autonomous regions (Canary Islands, Ceuta, Melilla); the inability to estimate cross-price elasticities between different tobacco products (e.g., e-cigarettes, heated tobacco); and the potential influence of illicit trade, though Spain's levels are relatively low. Additionally, the short pre-intervention period for Act 28/2005 limits the precision of trend identification for that specific policy.