Sugar taxes have received much attention in the media over the last two years, as politicians and policy-makers world-wide have become aware of increasing medical evidence that sugar intake is associated with a wide range of health problems, notably obesity, heart disease, type 2 diabetes, cancer and tooth decay. These health problems are in turn causing significant increased health care costs in many countries. In the UK this has attracted particular attention because of the over-stretched nature of the NHS, which is already struggling with the burden of an ageing population and underfunding, to say nothing of Covid. The total costs of to the NHS and the UK economy as a whole have been estimated at around 27 billion, according to Public Health England (PHE), although this figure is now somewhat dated. This compares with the NHS total budget of 125 billion for 2018 . The single most wide-ranging problem that is addressed by a sugar tax is obesity; this is defined as having a body mass index (BMI) of over 30 . The prevalence of obesity has been increasing in many countries for decades, but is now increasing fastest in low and middleincome countries rather than richer countries. The US has the highest proportion of adults, at 35%, while in the UK the proportion is 25%, the highest in Western Europe, with an additional 37% of adults being classified as overweight (BMI over 25). Of additional concern is the increasing proportion of obese children; in the UK 10\% of 4-5 year-olds and 19\% of 10-11 year-olds are obese. A further demographic of concern is that obesity among both adults and children is most prevalent among lower-income households. Over the last ten years various governments worldwide have introduced some kind of tax on sugar-sweetened beverages (SSBs), including Mexico, France, Norway, Hungary and the UK. Denmark has actually done the reverse, and removed an existing tax. In some cases the tax on SSBs has been just one component of a more comprehensive public health program designed to reduce sugar consumption and obesity. Before examining the evidence of the effectiveness of these policies it is worth considering the arguments that have been put forward on both sides regarding the imposition of SSB taxes. In support of taxation a number of arguments have been proposed. The most important one from an economic standpoint is that consumption of SSBs imposes social costs that are negative externalities, an example of market failure. Non-SSB consuming taxpayers end up subsidising the health care of SSB consumers. Standard economic theory prescribes a tax in this situation in order to equate the private cost of consumption with the social cost. Another economic argument involves a different type of market failure, imperfect information. Consumers may be unaware both of the sugar content of various beverages and of the harmful health effects of consuming excess sugar. This has different policy implications to the previous problem, which will be discussed later. There is a further argument relating to sugar consumption and government intervention, which is not so commonly proposed, but involves aspects of behavioural economics. Evidence suggests that for products like sugar that involve visceral factors and the desire for immediate 1 gratification people use an excessively high discount rate for future costs, and may also discount hyperbolically as opposed to exponentially. This can cause time-inconsistent preferences, as people indulge their appetites and then regret it afterwards, maybe years later. Evidence also suggests that this kind of irrational discounting, not predicted by the standard neoclassical economic model, is particularly common among the young and the poor, who both tend to be high sugar consumers. A different type of argument in favour of sugar taxation concerns the generation of government revenue. When the tax was announced in 2016 the Treasury expected this to be around 500 million per year, although this has since been reduced to 275 million. It has been proposed that this revenue be used for subsidising activities that combat obesity. as school sports and breakfast clubs. This policy implication will also be discussed later. The main reason the estimate for tax revenue has been reduced is actually an indicator of the success of the tax measure, and relates to what many consider to be the most important argument in its favour. Ever since the tax was announced manufacturers of soft drinks have been reformulating their products in order to avoid or reduce it. The Treasury has estimated that 50% of manufacturers have already embarked on such sugar-reducing programs, including big retailers and brands like Tesco, Ribena, Fanta, Lucozade and PepsiCo. These programs have not been without controversy. Consumers have in some cases reacted unfavourably to changes in taste, as in the case of Ribena; others have criticised the use of artificial sweeteners like aspartame, that some studies have shown to have even more adverse health effects than sugar (Hoffman, Ronan and Haspula, 2018). As for arguments against the sugar tax, these are sometimes led by libertarians who resent the encroachment of the 'nanny state'; they claim we should all be free to make our own choices, even if these turn out to be bad ones. However, this argument does not address the arguments described earlier related to market failure. A second argument against the tax is that is regressive. This only considers the cost of the tax but fails to consider its benefits. As we shall see, some empirical studies in Mexico and the US indicate that the benefits are highly progressive, and that the benefits to poorer households far outweigh the costs. It should also be pointed out that, even if the overall effects were regressive, this is not a valid economic argument. Economic theory in this case proposes that the regressive nature is necessary to eliminate market failure and achieve an efficient allocation of resources. Any effect on perceived fairness should be mitigated by redistribution policies of an overall macroeconomic nature to have the desired social effect. A third argument against the tax is that it will reduce jobs. This is actually true for any change in government policy, in a short-run and narrow context. In this case some manufacturing and retail jobs would undoubtedly be lost in the short-run. However, the resulting reallocation of resources, and increased demand for substitutes like bottled water, will increase jobs in other areas. The strongest argument against the tax is that it is not effective in terms of its overall objective of reducing obesity and related diseases, and in turn reducing health care costs. This 2 will be discussed later in terms of empirical evidence. Before doing this we shall examine the nature of the tax in the UK. The tax is referred to as the Soft Drinks Industry Levy (SDIL), and came into force in April 2018. It involves two bands: the lower band is 18p per litre on soft drinks with more than 5 grams of sugar per 100ml; the higher band is 24p per litre on drinks with more than 8 grams per 100ml. The lower band corresponds to a tax rate of between 5 and 25%, a large range corresponding to the fact that drinks are priced differently according to brand and volume of bottle/can. Most affected will be energy drinks, since these are highest in sugar, and then carbonated drinks or sodas. There are a number of exemptions from the notably milk (which is marginal anyway at 5gm/100ml ), milkshakes, yoghurt drinks and fruit juice. The narrow scope of the tax is again controversial, and this aspect will be discussed later in relation to policy imnlirations and ronslucions Let us consider first how people said they would react to the tax. A Mintel poll of 2,000 consumers in 2018 found that only 47% said they would reduce consumption. This figure was higher among younger people and in London compared with Scotland, so demographics are a significant factor. Another interesting finding was that a much higher proportion of people appeared willing to respond to 'nudges': 75\%, said they would reduce consumption of unhealthy food/drink if nutritional information on product packaging was easier to understand; 56% said they would cut down if advertising of unhealthy products was regulated more strictly. One problem with interpretation here is that the questions posed were somewhat vague. Just how much would people be prepared to reduce consumption? It is not possible to estimate price elasticities based on this kind of information. At this point in time it is too early to assess the impact of the SDIL in the UK, but a number of empirical studies have been conducted in other countries, estimating effects of policies and modelling their future effects. We can now consider this empirical evidence, but will see that its interpretation is controversial, largely because of methodological issues. One of the most prominent examples of a tax on SSBs has been Mexico, where a 10% tax was imposed in January 2014. It was a part of a more general anti-obesity policy, involving an 8% tax on non-essential energy-dense foods (EDFs) and restrictions on food marketing aimed at children, in response to the high adult obesity rate of 32%. Sales of SSBs actually increased in 2014 by about 6% and in 2015 by 7%. This finding involves the most important methodological issue: it does not take into account other factors affecting SSB consumption, such as population growth, economic growth, and seasonal factors like weather. If these other factors are taken into account we can obtain a truer picture of the counterfactuals: what would have happened without the tax. On this basis it has been estimated that consumption fell by about 6\% in 2014 and a further 10\% in 2015 (Lancet Diabetes and Endocrinology, 2017). Consumption was estimated to fall by a larger proportion in low-income households, up to 17% by the end of 2014 . In addition it was estimated that there was a 4% rise in consumption of untaxed drinks, mainly bottled water, a substitute product, in 2014. At this point another methodological issue arises. it can be claimed that the reduction in SSB consumption is merely a proximate objective. We need to examine the consequences of the policy in terms of its ultimate objective, that is health outcome. A study in the British Medical Journal (BMJ) in 2016 reports that this translates into a fall in consumption of only about 12ml per day, or two teaspoonfuls. This represents a very low reduction in caloric intake, in the range of 10-20 calories per day. However, even this small reduction translates into a significant health outcome on a national basis. The Lancet study also estimated that the tax would prevent 190,000 new cases of diabetes over the next ten years, with a resulting saving of 785 million of health care costs. We shall return to this issue shortly in terms of the UK experience. A similar tax was introduced in France in 2012, but applied to drinks with artificially sweeteners as well as sugared drinks. The tax increased prices by about 3.5%, and was followed by a fall in consumption of about the same proportion over the next five mont Since then sales have rebounded somewhat, so that overall soft drinks sales were higher in 2015 than in 2011 before the tax was imposed. This illustrates another methodological issue: short-run and long-run effects may be different, as consumers take some time to change their habits. Although there have been many other examples of such taxes being used, we will quote just one more before considering modelling studies based on hypothetical situations. This is a smaller-scale situation, relating to Berkeley in the US. A tax of 25%, one cent per ounce, was imposed on specific SSBs in 2015, with 47% of this being passed on to consumers in the form of higher prices. Different studies have produced different estimates of the consequent drop in consumption, varying from 10% to 26%. One study estimated that consumption of SSBs fell by 21% among low-income households in the four months following the tax. A modelling study for the US as a whole was reported in the American Journal of Preventive Medicine in 2015. This was based on a tax of $0.01 per ounce on all SSBs lasting for 10 years. It was estimated that such a policy would cost $51 million to administer in the first year, and would reduce consumption of SSBs by about 20%, but the resulting reduction in average adult BMI would only be 0.08 by the end of the second year. However, this is sufficient to have large health outcomes: a gain of 871,000 quality-adjusted life years (QALYs), a savings of $54.9 bilion in healthcare costs, as well as a generation of $12.5 billion in tax revenue over the whole 10-year period 2015-2025. As already stated, it is too soon to have real empirical data relating to the UK, so we will refer instead to another modelling study reported in the Lancet Diabetes and Endocrinology (2016). This study used nationally representative data from the National Diet and Nutrition Survey rolling programme and scaled up the predictions to estimate national health outcomes of a policy of reducing sugar in all soft drinks, including fruit juice, by 40% over a five-year period. The intermediate consequences do not appear impressive: an average calorie reduction about 38 calories per day, and an average weight reduction of about 1.2kg per adult. This translates into a reduction in adult obesity from about 28% to about 26%. However, this corresponds to a reduction of 1 million obese adults, and a further reduction of 0.5 million overweight adults; this would reduce the incidence of type 2 diabetes by between 274,000 and 309,000 over the following two decades. What conclusions can we draw at this stage regarding sugar taxes and demand? These can be ummarised below: 1. Policies should be evaluated based on ultimate outcomes not proximate ones marginal benefits in terms of reduced consumption, calorie intake and BMI can translate into significant benefits in terms of ultimate health outcomes and cost savings. 2. Taxes on SSBs should form only one component, and probably not the main one, of any government policy aimed at improving the overall health of a country. The UK policy has often been criticised for being too narrow in scope. First of all there appears to be no real justification for exempting fruit juices, with a sugar content of 9gms/100ml. A broader tax policy would also include non-essential EDFs, which are basically junk food. The problem here is one of defining the term 'non-essential'; an increasing number of nutritionists now claim that saturated form an essential part of the diet, containing key nutrients, even though they are energy-dense. 3. Nudge policies may be more effective than taxes at changing behaviour. This is supported by opinion polls like the Mintel one reported earlier, where people claim that more user-friendly information on packaging and restrictions on advertising aimed at children would significantly affect their consumption. Other nudge policies have recently been proposed in the UK by a health select committee of MPs, including restrictions on sweets being sold at supermarket tills, and bans on the use of cartoon characters in the advertising of junk food. In a broader context nudge policies also relate to the provision of healthy school food, cycling lanes on roads and an environment more conducive to taking exercise, for example with less pollution. 4. Revenue from taxes should not be ring-fenced for particular uses, however desirable they may appear to be in a narrow context. The temptation to do this stems from the mental accounting psychological bias related to lack of fungibility of income and spending categories. People tend to want to match spending with income categories. However, this may not result in an efficient allocation of resources. Economic theory proposes that any tax revenue, regardless of source, should be spent in an optimal manner, meaning in a way that achieves the greatest marginal social benefit. 5. Comprehensive policy programmes may take time to achieve meaningful results. A good example of this has been drink-driving laws in many countries. When these were initially introduced decades ago, there was considerable opposition, mainly from the libertarian lobby. Gradually social norms have changed, and may now be more important than any market norm. Thus social acceptability, or lack of it, may more important than the threat of fines, losing one's license, or even going to prison, in terms of affecting driving behaviour. In this context, signalling and herding are of key importance. When more people have better nutrition, lower BMI, and take more exercise, reference points change and herding effects will shift social norms accordingly. 1. Outline the methodological issues involved in terms of interpreting evidence from empirical studies. (25\%) 2. Explain why the reformulation of products caused by a tax may be problematical. (15\%) 3. Outline the shortcomings of the current SDIL in the UK. (20\%) 4. A 2-litre bottle of a typical SSB currently costs about $2.30 in the US. Using the information above related to the 2015 study in the American Journal of Preventive Medicine, calculate the PED for SSBs in the US, assuming all the tax is passed on to the consumer in the form of a higher price. (30\%) 5. Assuming health care costs per person in the US are the same as for the UK, estimate the effect of such a tax on total health care costs in the UK. (10\%) Sugar taxes have received much attention in the media over the last two years, as politicians and policy-makers world-wide have become aware of increasing medical evidence that sugar intake is associated with a wide range of health problems, notably obesity, heart disease, type 2 diabetes, cancer and tooth decay. These health problems are in turn causing significant increased health care costs in many countries. In the UK this has attracted particular attention because of the over-stretched nature of the NHS, which is already struggling with the burden of an ageing population and underfunding, to say nothing of Covid. The total costs of to the NHS and the UK economy as a whole have been estimated at around 27 billion, according to Public Health England (PHE), although this figure is now somewhat dated. This compares with the NHS total budget of 125 billion for 2018 . The single most wide-ranging problem that is addressed by a sugar tax is obesity; this is defined as having a body mass index (BMI) of over 30 . The prevalence of obesity has been increasing in many countries for decades, but is now increasing fastest in low and middleincome countries rather than richer countries. The US has the highest proportion of adults, at 35%, while in the UK the proportion is 25%, the highest in Western Europe, with an additional 37% of adults being classified as overweight (BMI over 25). Of additional concern is the increasing proportion of obese children; in the UK 10\% of 4-5 year-olds and 19\% of 10-11 year-olds are obese. A further demographic of concern is that obesity among both adults and children is most prevalent among lower-income households. Over the last ten years various governments worldwide have introduced some kind of tax on sugar-sweetened beverages (SSBs), including Mexico, France, Norway, Hungary and the UK. Denmark has actually done the reverse, and removed an existing tax. In some cases the tax on SSBs has been just one component of a more comprehensive public health program designed to reduce sugar consumption and obesity. Before examining the evidence of the effectiveness of these policies it is worth considering the arguments that have been put forward on both sides regarding the imposition of SSB taxes. In support of taxation a number of arguments have been proposed. The most important one from an economic standpoint is that consumption of SSBs imposes social costs that are negative externalities, an example of market failure. Non-SSB consuming taxpayers end up subsidising the health care of SSB consumers. Standard economic theory prescribes a tax in this situation in order to equate the private cost of consumption with the social cost. Another economic argument involves a different type of market failure, imperfect information. Consumers may be unaware both of the sugar content of various beverages and of the harmful health effects of consuming excess sugar. This has different policy implications to the previous problem, which will be discussed later. There is a further argument relating to sugar consumption and government intervention, which is not so commonly proposed, but involves aspects of behavioural economics. Evidence suggests that for products like sugar that involve visceral factors and the desire for immediate 1 gratification people use an excessively high discount rate for future costs, and may also discount hyperbolically as opposed to exponentially. This can cause time-inconsistent preferences, as people indulge their appetites and then regret it afterwards, maybe years later. Evidence also suggests that this kind of irrational discounting, not predicted by the standard neoclassical economic model, is particularly common among the young and the poor, who both tend to be high sugar consumers. A different type of argument in favour of sugar taxation concerns the generation of government revenue. When the tax was announced in 2016 the Treasury expected this to be around 500 million per year, although this has since been reduced to 275 million. It has been proposed that this revenue be used for subsidising activities that combat obesity. as school sports and breakfast clubs. This policy implication will also be discussed later. The main reason the estimate for tax revenue has been reduced is actually an indicator of the success of the tax measure, and relates to what many consider to be the most important argument in its favour. Ever since the tax was announced manufacturers of soft drinks have been reformulating their products in order to avoid or reduce it. The Treasury has estimated that 50% of manufacturers have already embarked on such sugar-reducing programs, including big retailers and brands like Tesco, Ribena, Fanta, Lucozade and PepsiCo. These programs have not been without controversy. Consumers have in some cases reacted unfavourably to changes in taste, as in the case of Ribena; others have criticised the use of artificial sweeteners like aspartame, that some studies have shown to have even more adverse health effects than sugar (Hoffman, Ronan and Haspula, 2018). As for arguments against the sugar tax, these are sometimes led by libertarians who resent the encroachment of the 'nanny state'; they claim we should all be free to make our own choices, even if these turn out to be bad ones. However, this argument does not address the arguments described earlier related to market failure. A second argument against the tax is that is regressive. This only considers the cost of the tax but fails to consider its benefits. As we shall see, some empirical studies in Mexico and the US indicate that the benefits are highly progressive, and that the benefits to poorer households far outweigh the costs. It should also be pointed out that, even if the overall effects were regressive, this is not a valid economic argument. Economic theory in this case proposes that the regressive nature is necessary to eliminate market failure and achieve an efficient allocation of resources. Any effect on perceived fairness should be mitigated by redistribution policies of an overall macroeconomic nature to have the desired social effect. A third argument against the tax is that it will reduce jobs. This is actually true for any change in government policy, in a short-run and narrow context. In this case some manufacturing and retail jobs would undoubtedly be lost in the short-run. However, the resulting reallocation of resources, and increased demand for substitutes like bottled water, will increase jobs in other areas. The strongest argument against the tax is that it is not effective in terms of its overall objective of reducing obesity and related diseases, and in turn reducing health care costs. This 2 will be discussed later in terms of empirical evidence. Before doing this we shall examine the nature of the tax in the UK. The tax is referred to as the Soft Drinks Industry Levy (SDIL), and came into force in April 2018. It involves two bands: the lower band is 18p per litre on soft drinks with more than 5 grams of sugar per 100ml; the higher band is 24p per litre on drinks with more than 8 grams per 100ml. The lower band corresponds to a tax rate of between 5 and 25%, a large range corresponding to the fact that drinks are priced differently according to brand and volume of bottle/can. Most affected will be energy drinks, since these are highest in sugar, and then carbonated drinks or sodas. There are a number of exemptions from the notably milk (which is marginal anyway at 5gm/100ml ), milkshakes, yoghurt drinks and fruit juice. The narrow scope of the tax is again controversial, and this aspect will be discussed later in relation to policy imnlirations and ronslucions Let us consider first how people said they would react to the tax. A Mintel poll of 2,000 consumers in 2018 found that only 47% said they would reduce consumption. This figure was higher among younger people and in London compared with Scotland, so demographics are a significant factor. Another interesting finding was that a much higher proportion of people appeared willing to respond to 'nudges': 75\%, said they would reduce consumption of unhealthy food/drink if nutritional information on product packaging was easier to understand; 56% said they would cut down if advertising of unhealthy products was regulated more strictly. One problem with interpretation here is that the questions posed were somewhat vague. Just how much would people be prepared to reduce consumption? It is not possible to estimate price elasticities based on this kind of information. At this point in time it is too early to assess the impact of the SDIL in the UK, but a number of empirical studies have been conducted in other countries, estimating effects of policies and modelling their future effects. We can now consider this empirical evidence, but will see that its interpretation is controversial, largely because of methodological issues. One of the most prominent examples of a tax on SSBs has been Mexico, where a 10% tax was imposed in January 2014. It was a part of a more general anti-obesity policy, involving an 8% tax on non-essential energy-dense foods (EDFs) and restrictions on food marketing aimed at children, in response to the high adult obesity rate of 32%. Sales of SSBs actually increased in 2014 by about 6% and in 2015 by 7%. This finding involves the most important methodological issue: it does not take into account other factors affecting SSB consumption, such as population growth, economic growth, and seasonal factors like weather. If these other factors are taken into account we can obtain a truer picture of the counterfactuals: what would have happened without the tax. On this basis it has been estimated that consumption fell by about 6\% in 2014 and a further 10\% in 2015 (Lancet Diabetes and Endocrinology, 2017). Consumption was estimated to fall by a larger proportion in low-income households, up to 17% by the end of 2014 . In addition it was estimated that there was a 4% rise in consumption of untaxed drinks, mainly bottled water, a substitute product, in 2014. At this point another methodological issue arises. it can be claimed that the reduction in SSB consumption is merely a proximate objective. We need to examine the consequences of the policy in terms of its ultimate objective, that is health outcome. A study in the British Medical Journal (BMJ) in 2016 reports that this translates into a fall in consumption of only about 12ml per day, or two teaspoonfuls. This represents a very low reduction in caloric intake, in the range of 10-20 calories per day. However, even this small reduction translates into a significant health outcome on a national basis. The Lancet study also estimated that the tax would prevent 190,000 new cases of diabetes over the next ten years, with a resulting saving of 785 million of health care costs. We shall return to this issue shortly in terms of the UK experience. A similar tax was introduced in France in 2012, but applied to drinks with artificially sweeteners as well as sugared drinks. The tax increased prices by about 3.5%, and was followed by a fall in consumption of about the same proportion over the next five mont Since then sales have rebounded somewhat, so that overall soft drinks sales were higher in 2015 than in 2011 before the tax was imposed. This illustrates another methodological issue: short-run and long-run effects may be different, as consumers take some time to change their habits. Although there have been many other examples of such taxes being used, we will quote just one more before considering modelling studies based on hypothetical situations. This is a smaller-scale situation, relating to Berkeley in the US. A tax of 25%, one cent per ounce, was imposed on specific SSBs in 2015, with 47% of this being passed on to consumers in the form of higher prices. Different studies have produced different estimates of the consequent drop in consumption, varying from 10% to 26%. One study estimated that consumption of SSBs fell by 21% among low-income households in the four months following the tax. A modelling study for the US as a whole was reported in the American Journal of Preventive Medicine in 2015. This was based on a tax of $0.01 per ounce on all SSBs lasting for 10 years. It was estimated that such a policy would cost $51 million to administer in the first year, and would reduce consumption of SSBs by about 20%, but the resulting reduction in average adult BMI would only be 0.08 by the end of the second year. However, this is sufficient to have large health outcomes: a gain of 871,000 quality-adjusted life years (QALYs), a savings of $54.9 bilion in healthcare costs, as well as a generation of $12.5 billion in tax revenue over the whole 10-year period 2015-2025. As already stated, it is too soon to have real empirical data relating to the UK, so we will refer instead to another modelling study reported in the Lancet Diabetes and Endocrinology (2016). This study used nationally representative data from the National Diet and Nutrition Survey rolling programme and scaled up the predictions to estimate national health outcomes of a policy of reducing sugar in all soft drinks, including fruit juice, by 40% over a five-year period. The intermediate consequences do not appear impressive: an average calorie reduction about 38 calories per day, and an average weight reduction of about 1.2kg per adult. This translates into a reduction in adult obesity from about 28% to about 26%. However, this corresponds to a reduction of 1 million obese adults, and a further reduction of 0.5 million overweight adults; this would reduce the incidence of type 2 diabetes by between 274,000 and 309,000 over the following two decades. What conclusions can we draw at this stage regarding sugar taxes and demand? These can be ummarised below: 1. Policies should be evaluated based on ultimate outcomes not proximate ones marginal benefits in terms of reduced consumption, calorie intake and BMI can translate into significant benefits in terms of ultimate health outcomes and cost savings. 2. Taxes on SSBs should form only one component, and probably not the main one, of any government policy aimed at improving the overall health of a country. The UK policy has often been criticised for being too narrow in scope. First of all there appears to be no real justification for exempting fruit juices, with a sugar content of 9gms/100ml. A broader tax policy would also include non-essential EDFs, which are basically junk food. The problem here is one of defining the term 'non-essential'; an increasing number of nutritionists now claim that saturated form an essential part of the diet, containing key nutrients, even though they are energy-dense. 3. Nudge policies may be more effective than taxes at changing behaviour. This is supported by opinion polls like the Mintel one reported earlier, where people claim that more user-friendly information on packaging and restrictions on advertising aimed at children would significantly affect their consumption. Other nudge policies have recently been proposed in the UK by a health select committee of MPs, including restrictions on sweets being sold at supermarket tills, and bans on the use of cartoon characters in the advertising of junk food. In a broader context nudge policies also relate to the provision of healthy school food, cycling lanes on roads and an environment more conducive to taking exercise, for example with less pollution. 4. Revenue from taxes should not be ring-fenced for particular uses, however desirable they may appear to be in a narrow context. The temptation to do this stems from the mental accounting psychological bias related to lack of fungibility of income and spending categories. People tend to want to match spending with income categories. However, this may not result in an efficient allocation of resources. Economic theory proposes that any tax revenue, regardless of source, should be spent in an optimal manner, meaning in a way that achieves the greatest marginal social benefit. 5. Comprehensive policy programmes may take time to achieve meaningful results. A good example of this has been drink-driving laws in many countries. When these were initially introduced decades ago, there was considerable opposition, mainly from the libertarian lobby. Gradually social norms have changed, and may now be more important than any market norm. Thus social acceptability, or lack of it, may more important than the threat of fines, losing one's license, or even going to prison, in terms of affecting driving behaviour. In this context, signalling and herding are of key importance. When more people have better nutrition, lower BMI, and take more exercise, reference points change and herding effects will shift social norms accordingly. 1. Outline the methodological issues involved in terms of interpreting evidence from empirical studies. (25\%) 2. Explain why the reformulation of products caused by a tax may be problematical. (15\%) 3. Outline the shortcomings of the current SDIL in the UK. (20\%) 4. A 2-litre bottle of a typical SSB currently costs about $2.30 in the US. Using the information above related to the 2015 study in the American Journal of Preventive Medicine, calculate the PED for SSBs in the US, assuming all the tax is passed on to the consumer in the form of a higher price. (30\%) 5. Assuming health care costs per person in the US are the same as for the UK, estimate the effect of such a tax on total health care costs in the UK. (10\%)