Tobacco industry pricing strategies in response to excise tax policies: a systematic review

Objective To explore what is known about the tobacco industry’s (TI) price-based responses to tobacco excise tax policies and whether these vary by country income group using a systematic review. Data sources Studies assessing TI pricing tactics were identified via searches of five online databases using a combination of search keywords. Study selection Inclusion criteria were applied by two reviewers independently who screened all search results (titles and abstracts) for possible inclusion. They identified 37 publications that reported TI pricing tactics. Data extraction Study details were tabulated, and information was extracted on the country income group, population characteristics, excise tax structure, and pricing strategies. Data synthesis Of the 37 publications identified, 22 were conducted in high-income countries, while 15 covered low-income and middle-income countries (LMICs). Major pricing strategies employed were: differentially shifting taxes between products (35 studies); launching new brands/products as pathways for downtrading (six studies), product promotions and different prices for the same products for different customers (six studies); price smoothing (two studies); and changing product attributes such as length/size of cigarettes or production processes (three studies). Conclusions While there is limited evidence to fully ascertain industry responses to tax increases, this review suggests that the TI widely uses a multitude of sophisticated pricing strategies across different settings around the world with the intention of undermining tax policies, thereby increasing tobacco consumption and maximising their profits. There is a need for further research in this area especially in LMICs so that effective policy responses can be developed.


INTRODUCTION
Excise taxes on tobacco products are the most powerful and cost-effective tobacco control measure, since the resulting higher prices both reduce consumption and expand government revenues. [1][2][3] Higher prices discourage smoking among current smokers and prevent initiation, especially by the price sensitive young. 4 The WHO's Framework Convention on Tobacco Control (FCTC) recognises the importance of this policy in Article 6 and calls on governments to implement tax/price measures to reduce tobacco use. 5 However, the success of taxes is largely dependent on the extent to which the industry passes on tax increases and uses strategies to mitigate their impact. 6 7 An examination of internal tobacco industry (TI) documents revealed their knowledge of the influence of price on cigarette demand and consumer behaviour. [8][9][10][11][12] Furthermore, the oligopolistic nature of tobacco markets with limited competition gives them significant pricing power, especially since demand for their products is relatively inelastic, resulting in inordinate profitability. 8 13 14 This profit and pricing power therefore gives the TI options when it comes to responding to tax increases.
However, studies exploring industry price-based responses to tax increases are a relatively recent area of academic enquiry and have tended to focus on particular countries, notably high-income countries (HICs). As such, little is known about overall behaviours or trends especially across country income groups. Since pricing tactics impact the effectiveness of tobacco taxes, they are important for all countries to understand, particularly for low-income and middle-income countries (LMICs). Global tobacco consumption has become increasing skewed towards these countries where, with a few exceptions, cigarette affordability has generally increased, [15][16][17][18][19] giving the TI even greater potential to use price to undermine tax increases.
Given the emerging nature of research in this area, it is not yet clear if the current evidence on industry pricing strategies has identified tactics that are consistently used across differing countries, country income groups, or other contexts. For example, industry pricing will likely vary as it shifts from demand maximisation to profit maximisation. 13 20 Furthermore, evidence suggests that cigarette price differentials (difference between the cheapest and most expensive brands) are generally larger in LMICs than in many HICs, which allows the industry greater pricing freedom. 21 22 There are therefore significant uncertainties in regards to the industry's response to tax and associated pricing policies.
Consequently, this paper aims to undertake a systematic review of the literature on the TI's responses to tax increases in order to explore the range of pricing strategies they adopt to undermine their impact and specifically whether these vary by country/income-group/overtime/across price segments/products. Understanding these strategies is essential for developing appropriate countermeasures and hence the most effective tobacco taxation policies. The paper does not intend to examine wider industry pricing strategies, such as their arguments against higher taxation, including the threat of illicit trade, as these have been examined in detail elsewhere including via systematic review. [23][24][25][26]

METHODS
A systematic review methodology was adopted, following the Preferred Reporting Items for Review Systematic Reviews and Meta-Analyses (PRISMA) reporting guidelines. 27 A full protocol for this systematic review was developed and is registered with the PROSPERO international prospective register of systematic reviews (ID: CRD42021234336) at http://www.crd.york.ac.uk/PROSPERO.

Search strategy
We aimed to identify all independent empirical studies published in academic journals on industry pricing strategies in response to excise taxes. We therefore developed a comprehensive search strategy with the assistance of a reference librarian. Systematic searches were undertaken in five electronic databases (MEDLINE, Web of science, Embase, Scopus, and ScienceDirect), which were then verified by the main findings of the same searches in PubMed Central and Google Scholar. The following search string was applied with minor adaptations for the requirements of each database: ("Tobacco" OR "cigar*" OR "smok*") AND ("Tobacco company" OR '"transnational tobacco company" OR "transnational*" OR "TTC" OR "TTCs" OR "industr*" OR "manufacturer*") AND ("taxation" OR "tax" OR "excise" OR "price" OR "pricing") AND ("strateg*" OR "tactic*" OR "polic*" OR "intervention" OR "influence") Titles and abstracts of potentially relevant studies were independently screened for relevance against the eligibility criteria by one of two reviewers who also both screened a 10% sample of the other's exclusion decisions to ensure consistency. Following preliminary review, full texts of all studies that were found to be possibly relevant were then obtained and assessed independently by two reviewers according to the eligibility criteria. Any disagreements during the entire selection process were resolved through examination of the full manuscript, discussion and, where necessary, consultation with the third author.

Eligibility criteria
All searches were conducted between April and September 2020, and the search was not restricted by study design or year of publication. Similarly, no language or geographic restrictions were used. We included original studies that are primarily quantitative and/or qualitative peer-reviewed research articles featuring industry pricing tactics in response to tax policies. We excluded duplicate reports of the same study, brief reports that were only abstracts and conference abstracts. Unless price was mentioned, articles that focused on other TI strategies (such as lobbying/ marketing/advertising, etc) were also excluded.

Data extraction
Data from each study were extracted by one reviewer and verified by at least one other member of the author team. Studies were tabulated by country and country income groups: high, upper middle, lower middle, and low income using the World Bank's (WB) classifications with the year of the study also recorded. 28 We identified not just the year of publication but the year(s) the data covered. We sought additional data from the time of the different studies to contextualise our findings: information on tobacco taxes, including tax structure, excise tax as a proportion of price, and reliance on specific tax were taken from the WHO reports on Global Tobacco Epidemic, and gross domestic product (GDP) per capita at purchasing power parity (PPP) was taken from The WB's statistics. 29 All studies were reviewed in depth by all the authors who met regularly to review and discuss results and categorisation of industry pricing strategies until final classification and evidence tabulation had been agreed. Strategies not explicitly examined were reported as 'not examined (N/E)' but where examined and found not to occur, were recorded as 'absent'.

Data synthesis
We did not formally assess the methodological quality of reviewed articles because the goal of this review was not to assess the robustness of the evidence but rather to examine the breadth of strategies seemingly employed by the TI in different countries. However, we remained mindful of study quality when reading each paper in case any featured an obvious weakness or error that we wanted to highlight (none did). A narrative review is conducted and reported as wide heterogeneity in study design/settings/populations/indicators/outcome measures assessed among the included studies precluded a metaanalysis. 30 31 However, where possible, quantitative results are compared between studies.

Study selection
An initial search of the five databases located 6205 potential publications (which was then verified with the searches in Google Scholar and PubMed). A further 13 abstracts that were not in the search results, but which potentially met the inclusion criteria, were identified through forward and backward snowballing. The records were stored in a reference management system (Endnote), which facilitated ease of screening for duplicates and organisation of the final set of abstracts leaving 2632 documents. Of these, 2566 were excluded based on title and abstract screening. The inclusion and exclusion criteria were then applied to the remaining 66 publications, bringing the number of papers to 37 ( figure 1 shows the PRISMA study selection flow chart).

Characteristics of the included studies
The sample of 37 articles were all published between 1996 and September 2020. Twenty-two were conducted in HICs, 15 in LMICs, and none covered a low-income country (table 1). All 37 focused on pricing strategies for factory-made (FM) cigarettes, while five also briefly discussed roll-your-own tobacco (RYO).   *Excise tax proportion of price (%) quoted for the previous year due to data availability. †Data for GDP per capita, PPP at the time of study (current international $) are presented for 2019 as 2020 data were unavailable. ‡Cigarettes in the USA are taxed at both the federal and state levels, so taxes will vary within country.

Review
The majority of studies used commercial pricing data obtained either from governmental sources or independent market research databases (eg, Euromonitor/Nielsen), while a few used data from surveys of retailers or consumers (eg, ITC survey). We identified six broad pricing strategies that the TI employs: differentially shifting taxes; price smoothing; shrinkflation; changing product attributes or production processes; price discrimination and price related promotions; and introducing new brands/ segments/products as pathways for downtrading (see table 2 for definitions). While these tactics seem to vary in relation to the income level of countries, we did not find any correlation with the type of excise taxes or taxation structures.

TIs pricing strategies
Differential shifting of taxes The most frequently employed strategy identified through this review was the differential shifting of taxes. The overall pattern found in HICs is overshifting (seen in 15 out of 22 studies) with studies from the USA, 32-40 UK, [41][42][43][44][45] Ireland, 46 Czech Republic, 12 New Zealand 47 48 and Ukraine 49 highlighting this behaviour. The TI in these countries persistently raised tobacco prices by overshifting tax increases on tobacco products at the higher end of the market but undershifted on products in lowest segments to keep their prices cheap. This resulted in an increasing price gap between premium and budget cigarettes as well as the range of prices available within each price segment. A few exceptions to this overall trend were studies where shifting of taxes was: not examined 50 51 ; examined but found to be both over and undershifted 39 47 ; and undershifted (two from the USA, one from Europe) . 37 52 53 Both these US studies used product level consumer data: namely Nielsen Homescan data for 2006-2007, 52 and cigarette pricing data from chain supermarkets and drug stores between 2001 and 2011, 37 and both found that tax increases were only partially passed onto consumers. The study from Europe used Euromonitor passport data covering 23 countries between 2006 and 2017, and found that tax increases were selectively undershifted in the cheap cigarette price segment, whereas fully shifted in the expensive category. 53 The evidence obtained also suggests that a low and static tax environment enabled overshifting of taxes and allowed the industry to adopt strategic approaches for expanding their markets and generating considerable growth in total revenues. For example, a study from Taiwan shows that between 2011 and 2016 taxes levied on tobacco products remained unchanged, which prompted all the major companies to use aggressive pricing and segmentation strategies in order to maintain customer loyalty while making profits. 54 However, the industry's behaviour of overshifting is remarkably similar even in countries where prices of the cheapest segments are relatively high such as the UK and USA.
In contrast to HICs, the predominant pattern in LMICs (11 out of 15 studies) over the years covered (2000-2019) was undershifting. This was observed in 8/10 countries (South Africa, 55 Mexico, 56 57 Indonesia, 58 59 Turkey, 60 Thailand, 61 Bangladesh, 62 Pakistan 63 and Mauritius 64 ). The study authors frequently highlighted that in these countries undershifting does not signify the TI's lack of scope to overshift but its deliberate choice not to as the primary purpose in these economies seems to have been market expansion as opposed to profit maximisation. Hence, tax increases are mostly absorbed for brands across all price categories. 64 The pattern varied slightly in China, which is an unusual market given its state-owned tobacco monopoly and its role in setting both prices and taxes. Between 2009 and 2015, the monopoly decided against increasing cigarette prices, which led to reduced profit margins for the industry with no changes in the rates of cigarette consumption in light of constant prices. 65 Similarly, there is evidence of the pattern changing overtime in South Africa: from overshifting of taxes to undershifting. Between 1990 and 2005, the industry increased its price by more than the increase in taxes in order to increase revenues, which resulted in a positive impact on tobacco control as it reduced cigarette consumption. This prompted the industry to change focus from increasing prices to maintaining market Table 2 Description of pricing strategies

Pricing strategy Description
Differential shifting of taxes Taxes on tobacco products may be shifted to different extents: ► Overshifting takes place when the TI increases the price of products above that required by the tax increase. The burden of the tax increase (and more) falls entirely on the consumers rather than the producers. ► Undershifting occurs when the industry absorbs (to some extent) tax increases, thus delaying/preventing the intended tobacco price rises. In this scenario, the cost of the tax increase is at least partially borne by the producers.

Introducing new brands/segments/ products
The industry introduces new and cheaper factory made (FM) and roll-your-own products (RYO), including cheaper variants of existing products and potentially new price segments, to increase the opportunities for smokers to down-trade instead of quitting. This strategy helps retain customers who no longer want, or are unable, to pay for higher priced products. Simultaneously, they retain higher priced offerings that allow the industry to profit from those who are willing to pay higher prices for 'luxury' brands/products.

Price discrimination and price related promotions
Selling the same product at different prices to different customers, often through targeted price-related promotions, can preserve affordability of products across all income groups following a tax increase. This helps to prevent price-sensitive users from quitting or reducing consumption, ensures potential new customers are not deterred by high prices, but allows the industry to take advantage of those less sensitive to price.

Price smoothing
The industry prevents any sudden jumps/increases in price the consumer would face following tax rises by smoothing that increase throughout the year by employing smaller, more frequent increase in prices. This minimises the public health impact of tax increases by ensuring that smokers never face a sudden, large, potentially quit inducing price increase.

Shrinkflation
Varying the numbers of cigarettes per pack. The industry reduces the number of sticks in a pack from (for example) 20 to 19, 18 or even 17 sticks, or weight of tobacco per pack, to disguise price rises and prevent purchase costs of a packet of tobacco being tipped over certain psychological levels. The higher cost per cigarette does not become immediately obvious to most smokers.
Changing product attributes or production processes Complex tobacco tax structures that levy different tax rates based on different characteristics (eg, length, weight, price, or product type) mean the tobacco industry can exploit different tax classifications by changing physical product attributes or production methods to achieve classifications with lower tax rates.
Review size, resulting in lower rate of decrease in tobacco consumption since then. 66 Two LMICs (Mauritius 64 and Ukraine 49 ) showed evidence of both overshifting and undershifting of taxes. In Mauritius, excise policies have been in line with the international best practices, 67 and nominal taxes were increased six times between 2008 and 2018. However, the TI still managed to undermine the policy by adopting a multifaceted pricing strategy of overshifting on premium and popular brands while substantially undershifting on economy cigarettes. 64 Thus, this selective overshifting and undershifting increased price gaps and product segmentation, a scenario that was also observed in Bangladesh and Mexico. 20 57 Similarly, the study from Ukraine discovered that smaller tax increases were absorbed in 2007-2008, while larger surges in tax rates forced the tobacco manufacturers to raise their prices in order to maintain profits in the subsequent year. 49

Introducing new brands/segments/products as downtrade pathways
Existing evidence indicates that where multiple products are available in a market, the increase in taxes on one product (usually cigarettes) can encourage smokers to downtrade to cheaper alternatives. 42 51 61 68 Six identified studies (three each from HICs and LMICs) showed that tobacco manufacturers introduced new brands/segments/products as cheaper substitutes, presumably to make sure cigarette smokers do not quit as taxes cause prices to rise. In the UK, 42 45 Spain, 51 and Thailand, 61 the TI launched new economy products and brands of both FM and RYO. Similarly in other countries, new RYO products and new variants of previous FM brands or new subeconomy brands were launched in the subeconomy sector aimed at women and younger people. 45 61 Moreover, in the UK both FM and RYO brands were downgraded from a higher to a lower segment category by cutting their price. 45 A study examined the supply-side factors in Bangladesh's cigarette market between 2006 and 2017, and noted patterns of change in consumers' and producers' behaviour. 62 The widening price differential among brands led to brand substitution from higher price to low-price cigarettes. Furthermore, income growth and shifting preferences led to product substitution from bidis to FM even though the price of bidis remained unchanged during the study. The TI, therefore, created two different pathways to aid its profitability. By offering low-priced cigarettes, they generated new sales by encouraging a shift from bidis (which are generally independently made and very low priced). Low-priced FM also offered premium FM smokers a cheaper option instead of quitting, thereby helping the industry to maintain sales volumes.

Changing product attributes or production processes
In an attempt to exploit the different approaches/levels to taxing different products, the industry may reclassify its products, change their physical attributes or reduce its production levels, so they may fall within lower tax brackets/categories. Three studies (from the UK, 42 the USA, and Indonesia 69 ) indicate that the TI modified its products in order to avoid higher taxes and the resulting impact on sales/profits. However, evidence of this pricing strategy was far less than other tactics, suggesting that either it is not popular or has not been widely explored. Examples include the UK, where cigars were subject to relatively lower axation than cigarettes and exempt from some of the restrictions covering all of the European Union (EU). Consequently, the TI took advantage of these loopholes in legislation by launching economy cigars and low-priced cigarette-like filtered cigarillo products from 2010, including under existing cigarette brand names from 2020. 45 70 Similarly, in the USA from 2009 to 2013, tax on RYO tobacco was more than that for pipe tobacco so the industry relabelled the former as pipe tobacco reducing its tax liability. 50 71 As a result, sales of RYO tobacco fell while sales of pipe tobacco increased.
The review illustrates that a complex taxation system such as a tiered system with differential tax rates for each type of product further favours this tactic. For example, the studies covering Indonesia highlighted its elaborate multitiered tax system that varied by type of product (cigarette/kretek), mode of production and manufacturing facility, and favoured smaller scale production facilities. Furthermore, up to a 2009 rule change, tobacco companies had a tax incentive to split production between large numbers of small-scale producers. 59 69 Price discrimination and price-related promotions Evidence from six recent studies, all from HICs (covering the USA and the UK), highlight how the industry has developed targeted promotional tactics to offer pathways to smokers to reduce their costs. 35 39 It targets a particular category of consumers and uses price discrimination to lower the price of certain products by using coupons, cartons, gifts, or through differential pricing by geographic location/store type (eg, using retailer rebates). This strategy is beneficial for the industry as it maximises profits by minimising the effect of tax increases on demand and is disproportionately more impactful for the poor or price-sensitive consumers who are more likely to take advantage of price promotions. 72 In the USA, smokers are able to avoid the full impact of taxes by buying cartons or purchasing from lower tax jurisdictions such as Native American reservations. 32 36 Tobacco companies have closely monitored 'border' or 'tax-differential' stores in the country, where they provide extensive promotions of their products to consumers in order to modify shopping patterns and increase sales. 39 Correspondingly, in the UK between 2002 and 2014, tobacco purchases from supermarkets increased in popularity as the prices were cheaper than those in convenience stores. 45

Price smoothing and shrinkflation
Two studies from the UK identify the use of price smoothing and shrinkflation strategies by the TI. In response to a tax increase every year, the industry was found to initially chose to cut its profits and not (fully) pass on the tax to consumers, thus preventing any quit-inducing large surge in prices but then incrementally increase prices throughout the year in order to slowly pass on the tax increase and thus increase profits. 45 The extent to which the industry undershifts and the duration over which it increases price was most marked in the lowest price segments and more so in FM than in RYO tobacco. 45 The TI also masked the increase in prices by keeping pack purchase prices similar to before the tax increase but reduced the size of a standard FM pack (eg, from 20 to 17-19 sticks) and pouch of RYO tobacco (eg, from 12.5 g to 10 g). 42 45 In the UK, this occurred most markedly in value and subvalue FM categories, where the real pack prices of the cheapest FM and RYO products remained static between 2012 and 2017, despite the rise in price per stick/stick equivalent in all segments. 45

DISCUSSION
The primary focus of this study was to assess the extent of literature that exists on TI's price-based responses to taxation and Review hence to understand the overall pricing strategies employed worldwide. We can offer no insights into the (relative) success of these countertaxation policies, either in terms of their impact on tax revenue or public health, or if this varies by context, since this was not explored. Overall, the findings strongly suggest that the industry uses a variety of strategies and employs differential tactics for different price tiers of their products, presumably to balance enhancing their revenues/profits with maintaining/ boosting the volume of their markets. A large number of studies included in the review were carried out in the USA (11 studies) or the UK (five), and apart from the tactic of differential shifting tax increases, most of the other strategies identified were either more common in (changing product attributes/production processes) or were only found to be present in HICs (price discrimination and promotions, price smoothing, and shrinkflation). This may signify the absence of these strategies in LMICs but may also be a result of their not being considered by the research from these jurisdictions. Hence, the present review highlights the need for more research in this area, most especially in LMICs. [73][74][75] Although the predominant pattern in HICs was of overshifting as compared with undershifting in LMICs, not all studies included in the review conformed to this trend; a few demonstrated the presence of both the patterns simultaneously for different cigarette segments, 20 39 47 49 64 whereas others showed the opposite trend. 52 53 Therefore, current evidence suggests rate of tax pass-through varies over time and place, as the industry is able to price discriminate by country/income group depending on the tax rate/structure and likely also with the maturity of the market and the extent to which wider tobacco control measures are in place. Another factor that could explain differences is the role of state-owned tobacco monopolies (in markets like China, Vietnam, and Thailand), which face a different set of incentives to privately owned tobacco companies. 76 77 The TI's conduct in terms of tobacco tax pass-through can be understood in the context of the global tobacco market, as they balance driving consumption in newer markets with expanding profit margins in mature ones. 55 78 In HICs, the TI focuses on driving up value rather than volume as a profit maximising model, whereas in many LMICs, the industry pursues the volume expansion model by keeping prices cheap to drive up sales and establish use. 13 However, in many countries the industry uses a mix of these models, such as in the UK and the USA, where an initial response to tax increases is absorption of taxes on products across all price categories, although premium brands are only undershifted for a short time and are in fact progressively overshifted throughout the rest of the tax-year. 20 45 Though overshifting is good for reducing tobacco consumption, it represents extra profits for the TI and hence a missed opportunity for governments to have raised taxes even more and therefore collected additional revenues. Certain external factors such as increasing affordability as a result of excise taxes failing to match inflation/growing income levels also play a pivotal role as the industry uses this to its advantage by overshifting taxes as evidenced in a few LMICs, such as Bangladesh and China. 18 20 This finding is in concordance with another study that showed affordability elasticity as a useful parameter to predict sensitivity of consumers to tobacco tax and price policy changes. 79 There is also variation with respect to the presence of other tobacco control measures. For example, if legislation is expected to decrease demand, then manufacturers may decide to undershift tax increases at that point in order to reduce the impact of the wider legislature. 1 A few studies from the review emphasised the popularity of cheaper domestic brands of cigarettes among price sensitive (lower socioeconomic groups and youth) and heavy smokers, particularly in the USA and Mexico. 35 36 56 80 This finding is in line with other studies on pricing tactics more generally, such as from Canada, where discount and native brands were found to be more popular among youth smokers with relatively less spending money and higher cigarette consumption. 81 This clear preference for economy brands illustrates why the TI looks to undershift at the lower end of the market.
One interesting point of variation not covered in the highlighted studies is the potential for different companies to respond to taxes in different ways or extents. 82 We know various companies favour different tax structures (eg, because of their brand portfolios) 83-86 so these differences might extend to countertaxation strategies too.
The review also illustrates wide disproportions between industry's claims and actions; the TI aggressively opposes taxation by suggesting it is the cause of the illicit trade (owing to price increases), yet these allegations are inconsistent with their pricing behaviour. 23 An analysis of the tax and price changes in the UK found that around two-thirds of the increase was due to taxes, while one-third was imposed by the TI. 43 Furthermore, illicit trade levels are often considerably lower in HICs where prices/taxes on cigarettes are relatively high as opposed to some LMICs, suggesting that price alone does not explain the level of illegal trade. Although illicit is not a focus of our study, the review did not highlight any evidence of increases in taxes influencing the illicit trade, a finding consistent with studies from other LMICs. 87 This suggests there remains scope for further tax increases in most countries, which should be relatively large and unexpected, as the industry seems better able to adjust its prices to undermine the impact of smaller and/or predictable tax increases. 43 45 78 Our study also highlighted that in some countries, there may be changes in observed industry pricing strategies over time due to changes in market dynamics. In the case of South Africa, there was a switch from overshifting to undershifting in 2010, which coincided with increased competition from domestic low-cost producers and a rapid increase in the illicit trade. 55 88 While this might be a unique occurrence, it could also indicate a wider trend and hence needs to be further explored by following industry pricing behaviours over time. Finally, the emergence of heated tobacco products (HTPs) and other nicotine products in recent years have complicated the nature of the tobacco market and may thus have changed industry pricing strategies. Therefore, further research on industry pricing strategies over this full range of products needs to be prioritised.

Strengths and limitations
We have followed PRISMA reporting standards for systematic reviews. We made extensive attempts to obtain published literature via searches in the five main databases by employing a wide variety of search terms and by including a wide range of study designs in order to avoid overlooking evidence from weaker studies. This review systematically assessed all the available evidence on industry pricing strategies to undermine tobacco excise control interventions/policies, which are relevant to both HICs and LMICs. While the help of a university librarian was sought to pilot and revise the search strategies used herein, there is a possibility that not all relevant studies have been located. There could be other studies that offer some insights on pricing tactics of the industry that have been missed as tax/pricing responses were not their primary focus or they did not report this in their abstracts. Furthermore, we did not search the grey Review literature and thus may have missed relevant technical reports. However, we believe such missed investigations will be low in number, and hence that we have provided a comprehensive overview of the industry pricing approaches. This is, to our knowledge, the first attempt to provide a broad overview of the global evidence relating to industry efforts to influence excise levels.
The main limitations of this review relate to the quality of included studies as articles were not evaluated for the rigour of their methodology, but rather just for their discussion of industry's pricing tactics. There is a chance therefore that we may be reporting results obtained from studies of low quality. However, we note that all of the studies were published in peer-reviewed journals, and we found no indications to suggest quality issues during our review process. Nevertheless, methodological differences between the studies might have contributed to the individual findings among the included studies. We are also limited by the evidence available on pricing strategies from LMICs due to the low volume of studies that we found and the relative lack of studies in particular from low-income countries, making it difficult to assess the full response of industry to tax increases in certain jurisdictions or to evaluate the trends across different strategies.

CONCLUSION
This systematic review of the global literature on industry's price-based responses to taxation highlights that the TI use multifaceted strategies to undermine tobacco excise policies all over the world, thereby increasing tobacco consumption and the associated harms. It is not yet clear the extent to which such tactics vary by country or by other local context, as the review also revealed a relative paucity of literature looking at the details of these pricing practices, most especially in LMICs. Further research is therefore needed in this area as these pricing strategies must be routinely monitored and understood so that effective tobacco excise policies can be developed. Given observed industry behaviour, it seems like there is still ample room to increase tobacco taxation in most countries.

What this paper adds
What is already known on this subject ⇒ Studies covering tobacco industry price-based responses to tobacco excise tax policies exist, but there has been no attempt to collate or collectively assess this information, including how it varies by country/context.

What this paper adds
⇒ This paper systematically reviews current evidence of tobacco industry pricing strategies to undermine tobacco tax policies. ⇒ The review determines that tobacco industry tends to employ a variety of pricing tactics to mitigate the positive tobacco control effects of tax increases and that the six most commonly identified strategies in the existing literature both from high-income and low-income and middle income countries are: differential shifting of tax increases; shrinkflation; price smoothing; changing product attributes and production processes; introducing new brands/segments/ products as pathways of downtrade for consumers; and price discrimination and price-related promotions. ⇒ The review highlights the paucity of research on tobacco industry pricing efforts to undermine taxation especially from low-income and middle-income countries.