Countries are facing the blight of illicit tobacco trade (ITT). In 2017, about 11.6% of the overall global cigarette market were considered illicit with low and middle-income countries were more at risk of illicit trade compared to high income countries (Euromonitor International, 2019). ITT is both a global health and economic problem. It does not only increase access to harmful tobacco products especially among children and poor population, but also causes significant losses in government revenues.
Methodology
The scale of ITT in a particular country is driven by different demand and supply-side factors. Tbe former refers to preference of smokers for cheaper or specific tobacco produce. While the latter refers to activities of illegal and legal enterprises to increase profit, sales, and market shares exacerbated by the presence of corruption and organized crime, and weak government institutions (OECD, 2008). The multi-dimensional nature of ITT prompts governments to measure holistically the enabling factors to help in designing appropriate policy responses.
We constructed an ITT index, which measures the enabling factors of ITT in 160 economies. It is important to note that the index does not score a country’s effectiveness in combating illicit tobacco trade. Rather, the index examines their structural and institutional capability to combat ITT. It focuses on general governance, tobacco policies and systems and effectiveness of governance that contribute to the political and regulatory environment that reflects the country’s potential to address different kinds of illicit tobacco trade.
Figure 1 is the theoretical framework we used in constructing the ITT. It shows the enabling factors of ITT. We identified three domains that could lead to higher supply of illicit tobacco and eventually higher tobacco consumption. We conducted an extensive literature search to identify these different drivers of ITT then categorized them into three: general governance, supply and demand tobacco control policies, and trade and customs practices.
Figure 1. Theoretical Framework
The validity of the index depends the quality of the underlying variables. To ensure transparency and replicability, we only included variables with well-documented methodology. For each domain, we measured the internal consistency in the set of individual indicators (using Cronbach’s alpha, which is considered a good measure of scale reliability (OECD, 2008). The alpha coefficients for the three domains were more than 0.80 suggesting that the items have relatively high internal consistency. Not all countries have values for the variables used in constructing an index. We have addressed missing data using group mean imputation (OECD, 2008). By group mean imputation, we replaced missing data with the mean value of income grouping (that is, based GNI per capita) which the missing record belongs.
Governance
General governance is often discussed as one of the primary drivers of ITT. Empirical studies have suggested that ineffective governance, corruption, instability of government, presence of organized crimes, weak law enforcement, presence of informal channels, and ineffective tax administration exacerbate ITT (Goel, 2008; Meriman, Yurekli, & Chaloupka, 2000; Chionis & Chalkia, 2016). In constructing the scores under governance, we have identified the following sub-domains and their corresponding variables and data source (see Table 1).
Table 1. Governance indicators
Indicator | Operational definition | Data source |
---|---|---|
Intellectual property | • A Likert scale on the extent is intellectual property protection in the country (lowest: 1; highest: 7) • A Likert scale on the extent are property rights, including financial assets, protected (lowest: 1; highest: 7) | World Economic Forum’s Global Competitiveness Survey 2017-2018 (World Economic Forum, 2020) |
Corruption | The authors’ generated the following variables: • % of firms experiencing at least one bribe payment request • % of public transactions where a gift or informal payment was requested • % of firms expected to give gifts in meetings with tax officials • % of firms expected to give gifts to secure government contract • % of firms expected to give gifts to get an operating license • % of firms expected to give gifts to get an import license • % of firms expected to give gifts to public officials “to get things done” • % of firms identifying the courts system as a major constraint | Analysis of World Bank’s World Enterprise Survey (various years) (World Bank, 2020) |
Rule of law | • A Likert scale on the extent do government officials show favoritism to well-connected firms and individuals when deciding upon policies and contracts (lowest: 1; highest: 7) • A Likert scale on the independence of judicial system from influences of the government, individuals, or companies (lowest: 1; highest: 7) | World Economic Forum’s Global Competitiveness Survey 2017-2018 (World Economic Forum, 2020) |
Organized crime | • A Likert scale on the extent of organized crime (mafia-oriented racketeering, extortion) impose costs on businesses (lowest: 1; highest: 7) • A Likert scale on the reliability of police services (lowest: 1; highest: 7) | World Economic Forum’s Global Competitiveness Survey 2017-2018 (World Economic Forum, 2020) |
Government effectiveness | • A Likert scale on the extent of public trust in politicians (lowest: 1; highest: 7) • A Likert scale on the extent of burden of government regulation (lowest: 1; highest: 7) • A Likert scale on the extent of efficiency of legal framework in settling disputes (lowest: 1; highest: 7) • A Likert scale on the extent of transparency of government policymaking, 1-7 (best) | World Economic Forum’s Global Competitiveness Survey 2017-2018 (World Economic Forum, 2020) |
Informality | The authors’ generated the following variables:
• % of firms competing against unregistered or informal firms • % of firms formally registered when they started operations in the country • Number of years firm operated without formal registration • % of firms identifying practices of competitors in the informal sector as a major constraint | Analysis of World Bank’s World Enterprise Survey (various years) (World Bank, 2020) |
Tax administration | The authors’ derived the rescaled score (lowest: 0 to highest: 100) using the following questions:
• % of firms identifying tax rates as a major constraint • % of firms identifying tax administration as a major constraint | Analysis of World Bank’s World Enterprise Survey (various years) (World Bank, 2020) |
Tobacco control policies
Higher taxes and smoke-free environment free regulations are some example of tobacco control policies. The association between price and tax structure and ITT remains contentious. However, empirical studies suggest that tobacco control policies appear to be important factors in reducing ITT. In general, ITT is higher in countries with lower cigarette prices and lower tax rates (Joossens & Raw, 2012; Nguyen & Nguyen, 2020; Ross, Vellios, Batmunkh, Enkhtsogt, & Rossouw, 2020; Chisha, Janneh, & Ross, 2020; Flippidis, Chang, Blackmore, & Laverty, 2020). Table 2 shows the sub-domains and their corresponding variables and data source under tobacco control policies.
Table 2. Tobacco control policy indicators
Indicator | Operational definition | Data source |
---|---|---|
Tobacco tax burden | % share of tobacco taxes to retail price | World Health Organization (2018) |
Price of cigarette | The average price is calculated based on prices of three brands of cigarettes known to be most sold in the country | World Health Organization (2018) |
Tobacco control policies | Compliance score of regulations on smoke-free environments (lowest: 0; highest: 10) | World Health Organization (2018) |
Price dispersion | % share of cheapest brand price in premium brand price | World Health Organization (2018) |
Affordability | % share of GDP per capita required to purchase 2000 cigarettes of the most sold brand | World Health Organization (2018) |
Trade and customs practices
Limited capacity of customs in terms of technology, tools and manpower to track trade and distribution, and weak customs governance facilitate smuggling and illegal trade of tobacco products (McLinden & Durrani, 2013; Holden, 2017; Little, Ross, Bakhturidze, & Kachkachishvili, 2020). Table 4 shows the corresponding variables and data source under trade and customs practices.
Table 3. Trade and customs practices
Indicator | Operational definition | Data source |
---|---|---|
Efficiency of custom processes | The efficiency of customs and border management clearance (lowest: 1; highest: 5) | World Bank Logistics Performance Survey (2018) (Arvis, Ojali, Wiederer, Raj, & Dairabayeva, 2018) |
Custom infrastructure | The quality of trade and transport infrastructure (lowest: 1; highest: 5) | World Bank Logistics Performance Survey (2018) (Arvis, Ojali, Wiederer, Raj, & Dairabayeva, 2018) |
Quality of logistics service | The competence and quality of logistics services (lowest: 1; highest: 5) | World Bank Logistics Performance Survey (2018) (Arvis, Ojali, Wiederer, Raj, & Dairabayeva, 2018) |
Capacity to track and trace | The ability to track and trace consignments (lowest: 1; highest: 5) | World Bank Logistics Performance Survey (2018) (Arvis, Ojali, Wiederer, Raj, & Dairabayeva, 2018) |
Data standardization and aggregation
We conducted data standardization using Min-Max method by subtracting the minimum value and dividing by the range of the indicator values. This method transforms the indicators previously measured in different scales into normalized indices with identical range of values between 0 and 1 (OECD, 2008). This method is done by subtracting the minimum value of the indicator and dividing by the range of the indicator values:
where
$I_{ij}$ is the standardized value of the ith indicator for the jth country;
$x_{ij}$ is the actual value of the ith indicator for the jth country;
$min(x_i)$ is the minimum value of the ith indicator across all country;
$max(x_i)$ is the maximum value of the ith indicator across all country
Indicators under each domain were then aggregated to obtain a single measurement score per domain. Under the normative assumption that each domain is equally important in the assessment of enabling factors a whole, equal weighting is applied in aggregating the individual indicators under each domain. The overall score is then computed as the geometric mean of the three-dimension indices. This approach in creating a composite score using equal weighting and geometric aggregation is similar to the methodology used by the Human Development Index (HDI). The higher the ITT score (that is, 0=lowest; 1=highest) the less vulnerable to illicit tobacco trade. There are various aggregation methods discussed elsewhere including their advantages and disadvantages (Talukder, Hipel, & vanLoon, 2017).
We examined the validity of the ITT index. We conducted bivariate analysis with the share of illicit tobacco trade to tobacco consumption. We obtained data on illicit tobacco trade and smoking prevalence from Euromonitor International and World Health Organization. The analysis of ITT with the share of illicit tobacco trade. Only countries with available data were included in the analysis.
We triangulate the ITT scores with the size of the tobacco market in a particular country. Here, we adjusted the ITT score with smoking prevalence of the country using the formula [adjusted ITT=(1-(prevalence/100)) x ITT overall score], which should yield to a more realistic score. By applying this formula, countries with high prevalence of smoking will lead to lower adjusted ITT score, meaning more vulnerable to ITT.