Article Text

Download PDFPDF

Retailer compliance with tobacco control laws in New York City before and after raising the minimum legal purchase age to 21
  1. Diana Silver1,
  2. James Macinko2,
  3. Margaret Giorgio1,
  4. Jin Yung Bae3,
  5. Geronimo Jimenez3
  1. 1College of Global Public Health, New York University, New York, New York, USA
  2. 2Departments of Health Policy and Management and Community Health Sciences, Fielding School of Public Health, University of California, Los Angeles, California, USA
  3. 3Department of Nutrition, Food Studies and Public Health, Steinhardt School of Culture, Education and Human Development, New York University, New York, New York, USA
  1. Correspondence to Dr Diana Silver, College of Global Public Health, New York University, 411 Lafayette Street, New York, NY 10003 USA; diana.silver{at}


Objectives New York City (NYC) is the first large city to increase the legal minimum age for possessing tobacco products from 18 to 21 (Tobacco 21) and establish a minimum price law to reduce smoking rates among youth. However, retailer compliance with these regulations is unknown.

Methods Youthful investigators purchased cigarettes pre and post-Tobacco 21 implementation in 92 NYC neighbourhoods. Investigators recorded whether their ID was checked, the pack's purchase price, and observed compliance with additional regulations. Multivariable OLS and Poisson regression models assess pre and post Tobacco 21 compliance with ID checks and purchase prices, controlling for retailer type, location and compliance with other laws.

Results Retailer compliance with ID checks declined from 71% to 62% (p<0.004) between periods, and holding constant other factors, compliance with ID checks and sales at legal prices declined significantly after the laws changed. Compared to chain stores, independent retailers had significantly lower compliance rates (p<0.01).

Conclusions Several aspects of tobacco control appear to have deteriorated in NYC. Greater attention to monitoring retailer compliance with all tobacco regulations will be important for Tobacco 21 laws to be effective in reducing youth access to tobacco products.

  • Surveillance and monitoring
  • Priority/special populations
  • Public policy
  • Price

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Effective August 2014, New York City (NYC) raised the minimum legal age to possess tobacco products from 18 to 21, making it the largest US city to do so.1 At the same time, minimum price laws that prohibited discounting and required new signage at point of sale also went into effect.2 Although NYC's smoking rates have declined, progress has stalled in recent years.3 Eighty per cent of smokers in NYC report to have started smoking before age 21, and the transition from casual to regular smoking occurs around age 20.4 Moreover, people aged 18–20 are responsible for 90% of the cigarettes purchased on behalf of minors.5 The city's new laws are consistent with recommendations from the Institute of Medicine (IOM) which recently concluded that raising the minimum purchase age to at least 21 would likely result in reduced tobacco initiation among those of younger ages.6 The US Community Task Force also issued a statement supporting such laws.7 Retailer compliance with the multiple components of the new law is necessary for it to reduce youth access to and use of tobacco; however, NYC appropriated no new funds for its enforcement.

The Synar Amendment (effective 1996) strengthened retailer compliance with tobacco laws by requiring random inspections using youth decoys, prosecuting violators and mandating states maintain at least 80% compliance among retailers annually.5 ,8–11 However, some evidence suggests that using youth decoys accompanied by plainclothes inspectors may alert retailers to the ‘sting’ operation and overestimate compliance rates.12–14 Methods other than youth decoys could provide more robust and complete estimates of retailer compliance with the many existing laws designed to regulate tobacco availability.

In this study, we assess compliance with Tobacco 21 (T21) laws and minimum price laws after their passage in a large sample of NYC tobacco retailers, and compare compliance rates pre and post the passage of the laws.

Study data and methods

Data collection

We selected a random sample of 20 subway stops in each of 4 NYC boroughs (Manhattan, Bronx, Brooklyn, Queens) and 12 dense retail areas in Staten Island (n=92). At each stop, youthful, racially diverse female field investigators located the three nearest, most visible convenience stores, and two national chain stores, from which to purchase packs of Marlboro Gold cigarettes. For data collection before the new laws, all investigators were ages 18–21; after the laws were made effective, investigators ages 21–22 were employed for the study. Each stop was visited twice: pre-T21, March–April 2014 (421 packs purchased), and post, September–December 2014 (407 packs purchased). Three-tenths of the investigators in the first data collection period were white, and two-fifths in the second period; four of the investigators participated in both rounds, establishing some consistency across periods. Between data collection waves, the national pharmacy chain CVS stopped selling tobacco products, potentially contributing to a slight decline (4%) in purchases completed in wave 2. The study methods and protocol, including age appearance verification procedures, are described in more detail elsewhere.15

Analytic approach

We begin by calculating descriptive statistics regarding compliance with ID checks and legal sales prices, and calculate average sales prices in the preperiods and postperiods by store type and borough. For the postperiod, we also present descriptive statistics of compliance with requirements that a sign stating the legal age for purchase and another depicting the required NYS/NYC tax stamp were posted, and construct an index for the postperiod indicating compliance with all four of these laws.

We examine the effect of the passage of these laws on compliance with ID checks in two models by estimating prevalence ratios using multivariable Poisson regression models that control for retailer type and borough, and, in sequential models, a binary indicator of sale below the minimum legal price and a continuous measure of total purchase price. We use this same approach to examine compliance with minimum sales price laws, and include compliance with ID check along with retailer type and borough in the model. Last, to examine the effect of the new laws on the average total purchase price, we use ordinary least square (OLS) regression controlling for retailer type, city borough and an indicator of compliance with ID check. For all models, we use a binary indicator for the post-law period such that a coefficient >1 (or a positive coefficient in OLS models) would indicate that compliance (or price) has increased post-law. All models include robust SEs to adjust for potential non-independence of observations due to time and area sampled. Analyses were conducted using Stata V.13.1.16


Table 1 presents the per cent of retailers complying with ID check, minimum sales price law, and the posting of two signs at the point of sale, by retailer type and city borough in the pre and postperiods. Compliance with ID checks declined from 71% to 61.4%, and compliance with sales at the legal price declined from 96.9% to 92.1%, with average prices declining from $12.47 per pack to $12.27 per pack. Between periods, ID checks declined in independent (62.9% vs 53.1%) and chain store retailers (86% vs 80%) although chain store retailers had higher rates of compliance overall. Compliance with legal sales prices declined among independent retailers (95.9% vs 88.9%) as did average prices ($12.47 vs $12.15), while compliance and prices increased slightly in chain stores (98.7% vs 99.2%, $12.47 vs $12.53). Across boroughs, patterns of compliance and average prices across periods were less consistent. In the postperiod, 87.1% of retailers displayed the new legal age signage, and 69.6% displayed the new tax stamp sign. Only 46.5% or retailers complied with all four laws. In multivariable Poisson regression models using the postperiod sample only, retailers in full compliance with the other three measures (tax sign posted, legal age sign posted, and pack sold at or above minimum price) were 42% more likely (CI 1.17 to 1.73; p<0.01) to have checked for ID than retailers who did not (results not shown).

Table 1

Descriptive Statistics of pre–post compliance with NYC minimum age sales and minimum purchase price law provisions by retailer type and city borough

Table 2 presents results of multivariable models assessing the impacts of the T21 law on compliance with ID checks, sale below minimum prices, and average prices. In model 1, compliance with ID check (outcome 1) declined by 10% when compared with the pre-T21 period, other factors constant. Chain retailers were about 38% more likely (p<0.01) to have conducted an ID check overall compared to independent ones, holding city borough and pack price constant. Compliance with the minimum price law was associated with ID checks (p<0.01). In model 2, these relationships remain consistent while higher prices were associated with greater compliance with ID check. As shown in model 3, the likelihood of purchasing cigarettes at or above the minimum legal price (outcome 2) decreased by 4% after T21 (p<0.01) and chain store retailers were 4% more likely than independent ones, on average, to sell cigarettes at or above the legal minimum price, other factors constant (p<0.01). Stores that checked ID were about 8% (p<0.01) more likely than those that did not to sell cigarettes at legal prices, all else constant. Last, our OLS model estimating the impact of T21 on total cigarette pack price (outcome 3), demonstrates that mean prices declined by approximately 18 cents after T21, while retailers that checked ID charged about 25 cents more compared with those who did not (p<0.01), other factors constant (model 4). In all analyses, some significant differences were seen across boroughs, with compliance generally highest in Staten Island.

Table 2

Changes in ID check compliance, total purchase price, sale below minimum price and packs sold without New York State/City tax stamp, pre and post-T21 (n=827)


Our study reveals that compliance with minimum purchase age laws appears to have decreased after raising the legal age to 21 from 18, and was at a rate substantively below the 80% minimum level of compliance required of US states by the federal government during both periods.9 ,17 Compliance with minimum price laws also declined indicating that poor compliance was not solely a result of a lag in integrating the new laws into practice. In our sample, compliance across laws clustered: retailers complying with other tobacco regulations (such as minimum price and signage) were much more likely to comply with required ID checks.

Failure to comply with minimum age, minimum purchase price and other laws was highly concentrated among independent retailers. This was the case in preperiods and postperiods, but this difference grew after raising the minimum age for purchase. Non-compliance with the minimum legal price laws was almost completely confined to independent retailers.

There are several possible explanations for these findings. First, although the T21 law went into effect 9 months after enactment, retailer awareness of the new law may have lagged—especially among small, independent retailers. Minimum price laws have a higher degree of automaticity than minimum legal age laws, which may result in higher compliance. Second, inspection for ID checks are performed for all retailers only about once a year, lessening the perceived risk of violations.18 ,19 Last, enforcement of the laws examined here involves five different state and city agencies, each with different protocols and resources for inspection, prosecution and follow-up, which could limit their effectiveness.

This study has several limitations. Our sample, while large, was not a random sample of all retailers, but represents retailers in easily accessible, retail-dense environments, so estimates cannot be generalised to the city as a whole. The sample contains a larger portion of chain stores than a random sample would yield: in NYC approximately one-tenth of retailers are chain stores, while in our sample 3/10 are chain stores. We note, however, that the volume of cigarettes sold by these chain retailers may be quite large. In addition, some stores in each wave of data collection may have varied, although we use robust SEs to adjust for neighbourhood clustering of retailer types. Although they were instructed to dress in sweatshirts, sneakers and without make-up, our investigators were of legal age for purchase. However, in NYC, retailers were not able to defend themselves against violations of legal sales age laws of age 18 if the person appeared to be at least 25 years of age;18 when the legal sales age was raised to 21, the age in this provision was raised to 30.20

Without additional attention to enforcement, localities may not be able to realise the potential of the regulatory tools they employ. Methods such as ours may be helpful to enhance ongoing monitoring of retailer compliance with tobacco control laws.

What this paper adds

  • This study provides the first look at the early implementation of New York City’s increase in the minimum legal age of purchase for tobacco from age 18 to 21.

  • In the 92 retail-dense neighbourhoods sampled, compliance by retailers with ID checks for age declined in the period after the change in the law, while prices for cigarettes declined.

  • Compliance with other tobacco sales regulations by retailers was associated with higher rates of compliance with ID checks.


The authors thankfully acknowledge the field investigators who gathered the data for this study.



  • Contributors DS conceived of and oversaw the study and drafted the manuscript. JM provided guidance to the analysis and interpretation of results, revised drafts of the manuscript and helped conceptualise the study. MG conducted the analyses and contributed to the draft. JYB oversaw a component of the data collection and aided in the interpretation of the findings. GJ directed a component of the data collection and edited and assisted the preparation of the manuscript.

  • Funding This study was funded by the Institute for Human Development and Social Change (IHDSC), at New York University.

  • Competing interests None declared.

  • Ethics approval This study was approved by New York University's University Committee on Activities Involving Human Subjects in January, 2014.

  • Provenance and peer review Not commissioned; internally peer reviewed.