salt, high blood pressure, and performance-based regulation

REGULATORY POLICY FORUM

Salt, high blood pressure, andperformance-based regulation

Stephen D. SugarmanSchool of Law, University of California, Berkeley, CA, USA

AbstractPerformance-based regulation is a new approach to public health promotion. The aim of this article

is to explain how this idea might be applied to the public health goal of reducing salt consumption

as a way of reducing high blood pressure and thereby saving lives. Performance-based regulation is

compared with competing regulatory strategies.

Keywords: blood pressure, enterprise incentives, performance-based regulation, public health,

regulatory alternatives, salt.

This article demonstrates how the regulatory approach called “performance-based regu-lation” could be deployed to address high blood pressure, which as shown in Part I is ahuge public health problem. The major portion of the article explores the regulation ofsalt in light of the call by the American Medical Association for a 50% reduction in saltconsumption by the US public as a means to lower blood pressure. For public healthscholars, this brings to bear an innovative regulatory strategy on a problem that currentpolicies are failing to solve. For regulation scholars, this opens up for consideration alargely unexplored public health problem and provides the opportunity to address in anew context the competing merits not only of performance-based regulation and itsalternatives, but also of different sorts of performance-based schemes.

Part II presents and critiques existing approaches to the regulation of salt: requireddisclosures and “self-regulation.” Part III offers this article’s central proposal: usingperformance-based regulation to require food retailers (both supermarkets and restau-rants) to reduce sharply the overall amount of salt that passes through their cash reg-isters. Detailed attention is given to how such a scheme would be administered. Part IVcontrasts this proposal with other regulatory strategies for lowering US salt intake,including a traditional command and control regime, excise taxes, tort liability, andmanagement-based regulation. In addition, the close similarity between performance-based regulation and “cap and trade” schemes is exposed. Part V presents alternativeregulatory targets to show that performance-based regulation should not be viewed asa single solution, but rather as a broad approach that, as illustrated with respect to highblood pressure, can be implemented in a variety of ways. For example, rather thanregulating food retailers, regulation of health insurers is considered; and, in contrast tousing fines as incentives, paying bonuses is discussed. In all, fresh insights into regula-

Correspondence: Stephen Sugarman, School of Law, University of California, Berkeley, 327Boalt Hall, Berkeley, CA 94720-7200, USA. Email: [email protected]

Accepted for publication 13 February 2009.

Regulation & Governance (2009) 3, 84–102 doi:10.1111/j.1748-5991.2009.01048.x

© 2009 Blackwell Publishing Asia Pty Ltd

mailto:[email protected]

tory alternatives are presented through consideration of the potential application ofperformance-based regulation to a critical public health problem to which governmentis only beginning to turn its attention.

I. Public health introduction

High blood pressure contributes to a substantial amount of cardiovascular disease andearly death (Scientific Advisory Committee on Nutrition 2003; Dickinson & Havas2007).1 Although there are some doubters (Moore 1990; Taubes 1998), it is an article offaith among mainstream public health and medical communities that significantpopulation-based reductions in salt consumption (and thereby sodium reduction) indeveloped economies would sufficiently lower the blood pressure of a large number ofpeople to yield very large public health benefits (Intersalt Cooperative Research Group1988; Stamler et al. 1989; Elliot et al. 1996; Cutler et al. 1997; He & MacGregor 2002;Havas et al. 2007; Rose 2008).2 Indeed, the American Medical Association (AMA) hascalled for a 50% reduction in salt intake in the United States in hopes of saving up to150,000 lives annually (Dickinson & Havas 2007),3 and in the UK the government istrying to get people to reduce their salt consumption to about a teaspoon a day (FoodStandards Agency 2008a).

Salt enhances the flavor of many foods; some food processors rely on salt as apreservative; consumption of a small amount of salt is thought to be required for goodhealth; and in a majority of countries most people get the iodine they need for healthreasons through iodized salt (Food & Nutrition Board 2004). Yet people can enjoy safe,nutritious, well-flavored, and adequately iodized food through a diet that contains far lesssalt than the average person in higher income nations currently consumes. Indeed, theaverage American today may well be consuming seven times as much salt as is necessaryfor health (The Environmental Literacy Council undated).

Adding salt from the shaker, either in the course of home cooking or at the table, isnot a significant source of salt for most people. About 80% of the salt consumed byAmericans is already contained in restaurant foods and processed foods at the time theyare served or purchased (Havas et al. 2007). As a practical matter, for most people toreduce their salt intake, (i) they have to consume a different balance of food items, and (ii)food-makers must add less salt to their products. How might we bring those changesabout?

I propose using the strategy called performance-based regulation as a way to reducethe public health problems that experts say are caused by excessive salt consumption(Sugarman & Sandman 2007; Sugarman 2008).

II. Existing approaches to the regulation of salt in food: Disclosure andself-regulation

Before turning to performance-based regulation, I will focus briefly on existingapproaches.

Some people would consume less salt if they were more aware of the dangers ofsodium. Many people already try to avoid salty foods because they believe too much saltis risky. However, even those motivated to reduce their salt intake may find it difficult todo so because they are unaware of the amount of salt added to processed food andrestaurant food.

Performance-based regulation and salt S. D. Sugarman

© 2009 Blackwell Publishing Asia Pty Ltd 85

A typical traditional public heath strategy would be to better inform people, not justof the danger of high sodium intake, but also of the amount of salt contained in thespecific foods they are eating.

Indeed, regulatory initiatives like that have already been carried out to some extent. Inthe US, government-required labels on packaged foods indicate the contribution that aspecific food item makes to the recommended daily allowance of sodium (NutritionLabeling and Education Act of 1990, 21 U.S.C. § 343-1 (1990); 21 C.F.R. § 101.61 (2008)).Alas, these labels are often not easy for consumers to understand; the disclosures are oftenbased on counter-intuitive portion sizes and they are anchored in percentage terms thatare probably confusing to many buyers. Moreover, no such labeling requirements apply torestaurant food, fresh meat, poultry, and fish, and most prepared food sold in stores (21C.F.R. § 101.100 (2008); US Department of Health and Human Services 2000a).

To help consumers more intelligently shun high-sodium products, some suggestrequiring new labels on packaged food, prepared food, and restaurant menus (Dickinson& Havas 2007). Some favor a “traffic light” system, as has already been voluntarilyimplemented by some food sellers in the UK. Under this system, each food item wouldcarry a red, green or yellow label, letting people know whether its sodium content is high,low, or in between (Dickinson & Havas 2007; Food Standards Agency 2008b). Besidesimproving on the current US system of informing consumers, supporters of a traffic lightstrategy believe it would prod some food-makers to reduce the salt in the products theyproduce in order to win more favorable labels.

Still, even the most ambitious disclosure strategy would probably not suffice to bringabout reductions in salt consumption to the degree that advocates believe is desirablefrom a public health perspective. An analogy might be made to tobacco products. Ageneral awareness of the dangers of smoking has indeed caused many people to eitherquit or not take up smoking in the first place. But today’s sharply lowered smoking ratesin the developed countries are also importantly the result of a range of deliberate publichealth policies beyond education and disclosure, such as increased tobacco taxation,controls on where people are permitted to smoke, and “counter-ad” campaigns thatdisparage Big Tobacco (Chaloupka & Warner 2000; US Department of Health andHuman Services 2000a; Chaloupka et al. 2001; Pierce et al. 2005).

Some would argue that experience from tobacco control shows that government mustgo beyond adopting a stronger disclosure scheme and additionally launch a “socialmarketing” campaign designed to convince more and more people to reduce their saltintake (Social Marketing Institute 2008; Weinreich Communications 2008).

At a minimum it is fair to say that we do not know the extent to which a well-conceived educational crusade about the dangers of excessive salt consumption wouldactually cause those who most need to change their eating behavior to shift meaningfullyaway from high-salt products. Experiments with approaches in various countries maysoon provide some answers (Dickinson & Havas 2007; BBC News 2008).4

Still, we know from years of tobacco control efforts that even a combination ofaggressive policies has not yet succeeded in bringing smoking prevalence rates in the USmeaningfully below 20% (CDC 2007), making it implausible that the nation will get evenclose to the national Public Health Service goal for 2010 of adult smoking rates of only12%. On this basis, it seems highly unlikely that the AMA’s objectives with respect toblood pressure could come anywhere near being achieved merely through the use ofdisclosure strategies regarding salt.

S. D. Sugarman Performance-based regulation and salt

© 2009 Blackwell Publishing Asia Pty Ltd86

Given the limited prospects of a disclosure strategy, some advocates such as the AMAseem to favor a mandatory across-the-board reduction in the amount of salt contained inall restaurant and processed food (Dickinson & Havas 2007). So far, however, these callsfor action appear primarily to be admonishments to industry backed up with vaguethreats of potential regulation of an undisclosed sort if the goal is not met. Absent thespecification of those threats, this comes down to relying on the “social responsibility” ofenterprises.

To be sure, some health gains can be achieved from pressing business. In the UK,under the Food and Drink Federation’s Project Neptune scheme, firms pledged to reducesalt (as well as fat and sugar) in their products. They have done so, thereby helping tocontribute to an overall UK salt reduction of about 5% over five years (Food and DrinkFederation undated). Still, corporate self-regulation goes only so far and by itself isunlikely to achieve anything like the AMA’s goal of a 50% reduction in salt in the US diet(Reich 2007).

III. A new way to regulate salt consumption: Performance-basedregulation

Performance-based regulation is not a new idea. It has been tried in some settings andproposed in even more, including the regulation of power plant emissions, buildingsafety, the safety of nuclear energy plants, the achievement of elementary and secondaryschool pupils, the safety of transportation systems, workplace safety, and so on(Coglianese et al. 2003; May 2003). Its potential application to public health problemsthat turn on consumer behavior (like smoking and drinking) has received attentionprimarily in earlier articles of mine (e.g. Sugarman 2009).

The essential features of performance-based regulation are (i) requiring outcomes,(ii) measuring whether the required outcomes are met, and (iii) imposing penalties ifthey are not. Abstractly, this is a very appealing approach, both because it focuses on thedesired social objective and it leaves the regulated parties to figure out the best way ofachieving that objective.

In practice, performance-based regulation can face serious problems, however.Adopting performance goals that can be accurately measured at a reasonable cost oftenputs those designing the scheme in something of a dilemma. To make the measurementside work may mean specifying outcomes that are only rough proxies for the genuinesocial objective; or insisting on the true social objectives may mean relying on imperfectmeasures of compliance. Moreover, even if the true social objectives are well specified andeasily measured, the regime may fail to attend to the risk that undesired side effects willoccur along with the desired outcomes. Getting the penalties right for non-compliancewith the outcome targets is also fraught with dangers. Penalties that are too high may becharacterized as sufficiently unfair as to politically undermine the entire scheme; toopuny penalties may cause the regulated parties simply to pay up instead of trying toachieve the desired social goal. An additional challenge is consistent and effective enforce-ment of the penalties by those managing the regulatory scheme on the ground; moreover,conventional government regulators have to re-conceive their roles (Coglianese et al.2003; May 2003).

Nonetheless, in response to critiques of various traditional regulatory approaches, acommitment to try to move regulation in the performance-based direction was evident



in both the Clinton and Bush II administrations (e.g. National Performance Review1995). As with other policy issues it is a matter of “compared with what” so that theplusses and minuses of performance-based regulation need to be weighed against regu-latory alternatives, as will be explored below.

The potential application of performance-based regulation to the problem of highblood pressure and to salt is new here. This is a particularly interesting problem to explorebecause it so nicely reveals how differing performance-based regulatory schemes mightbe utilized, illustrating that the social objectives of performance-based regulation can beset at very different levels of generality and the potentially regulated parties can occupyvery different positions in society (May 2003).

A. Performance-based regulation of food-makersWhen it comes to salt and high blood pressure, one might initially envision aimingperformance-based regulation at companies that either produce the food sold in retailmarkets or prepare the food served in restaurants. For example, food-makers could begiven outcome targets cast in terms of reduced salt levels, say, a 5% reduction every yearfor a decade until the AMA’s 50% target is reached. Substantial fees (or fines) would beimposed on companies failing to reach their targets. PepsiCo, for example, would have tocut in half the salt in its potato chips and similar products; Kraft would have to cut in halfthe salt in its cheese, crackers, hot dogs and the like; Pizza Hut would have to halve the saltin the pizzas it serves, and so on.

Further reflection suggests that a uniform reduction in the salt content of everyproduct is probably not the right regulatory strategy. Some products are already reduced-salt items for which further reductions are not practicable. Also, the same proportion ofsalt reduction will yield very different taste changes in different products.

A 50% reduction in salt consumption would surely more desirably be achieved by acombination of (i) reducing the salt in certain foods (indeed, by more than 50% in somecases); and (ii) at the same time, keeping the salt in other foods at or near what it currentlyis, but reducing consumption of those foods. In other words, eating reduced-salt versionsof what you currently consume (e.g. less-salty salami and less-salty Big Macs) is not theonly way to reduce salt intake. You could also eat less of the salty food you now eat,without changing the amount of salt in any portion. This would be accompanied by ashift in your consumption, so that you also would be eating more food that contains nosalt or less salt. Most likely, you would do some of both.

B. Performance-based regulation of retailersThis suggests that, with respect to salt, it might be wisest to apply performance-basedregulation to retailers, both stores and restaurants. In the US context, Wal-Mart, Kroger,Costco, Safeway, 7–11, and other retail food chains would be required to cut the totalamount of salt contained in the food they sell. Fast-food retailers like Burger King andsit-down restaurant chains like Applebee’s would similarly have a reduced-salt target.

Most importantly, imposing the regulation on retailers would allow them to balancea variety of complementary salt-reducing strategies in the most efficient manner. WhenMcDonald’s is told to reduce the amount of salt in its foods, we do not care whether thereduction comes, for example, from cutting in half the amount of salt added to its Frenchfries or from shifting half of its French fries sales to sales of less-salty or salt-free foods likeunsalted baked potatoes, fresh salads with salt-free dressing, or even to sodium-free



drinks. Or, instead of reducing the amount of salt in its French fries, we would be just ashappy (salt-wise) if McDonald’s cut its French fries portions in half in a way that cutstheir consumption in half.

Focusing on large retailers would help make both setting individual firm targets andmonitoring the achievement of those targets an administratively manageable task. Tar-geting large retailers would sweep in a large majority of the processed and restaurant foodnow sold (Martinez 2007). By also including in the scheme the large wholesalers whoprovide a significant share of the food sold via small restaurants and markets, the regimewould probably cover a large share of what is left, and still keep the number of regulatedfirms to a modest number.

Cash register data already collect the necessary details about what the target firms sell,and there are already reliable data collected concerning the amount of salt in most ofthose items. Once agreement is reached on the best way to determine salt content of foodsnot currently monitored (a difficult, but surely not insurmountable, task), salt and salesdata could be readily merged to display how much salt each regulated firm is now selling.Cynics might imagine that specialized, salty-foods delis would be created outside thescheme, but even if this happened to some degree, it would create vivid new targets forinformation campaigns and additional regulation.

This is, in a nutshell, how performance-based regulation might be structured. To besure, additional complex administrative details would have to be worked out. Forexample, simply telling Wal-Mart that, over the course of 10 years, it has to cut in half theamount of salt in the total volume of food it sells is probably not quite the right way toset Wal-Mart’s target. After all, during that period Wal-Mart might increase its marketshare of the food sold in the relevant market, and Wal-Mart may increase its sales simplybecause of population growth or because of a relative increase in the share of food thatpeople buy in markets as compared with restaurants. Changes like this should be takeninto account in the way the performance-based target is set.

One solution that should be rather easy to administer is first to randomly identify, say,$100,000 in Wal-Mart food sales at the start of the regime and then determine that thosesales contain x-grams of salt. Next, Wal-Mart’s regulatory target might be to cut thatamount of salt in half by the end of the period. Because of inflation, however, we wouldprobably want to measure the end-of-the-period food basket in terms of 2009 dollars.

If some food retailers – say Whole Foods – already sell a significantly less salty foodbasket than average, the regulatory regime would probably want to take this into accountas well. Hence, a nationwide salt-level target for 10 years out that is 50% of today’s levelmight require a smaller salt-reduction by Whole Foods than by other retailers – say 7–11– that sell a saltier food basket today.

I have so far arbitrarily set a 50% reduction as the overall national target because thatwas proposed by the AMA. However, it is by no means clear that this is the right target.Nor is it clear at what pace the 50% target ought to be achieved. Should it be a tenth ofthe way toward the 50% reduction goal for each of the 10 years? Should the rate ofreduction start out faster or slower? Should there be some period at the start for theregulated companies to have no required reductions at all while they decide on andimplement their strategies? Maybe by the end of the third year they should have achievedan 8% reduction, with a 6% further reduction required in each of the next seven years. Ileave these details aside for now, confident that reasonable solutions to these adminis-trative matters could be worked out.5



The crucial point to emphasize is that Wal-Mart and its fellow retailers would be incharge of achieving the social objective, and they could do so in a variety of ways. Theycould press some of their suppliers to reduce the salt in their products, thereby changingthe nature of processed foods that Wal-Mart now sells. They could press for smallerportion or package size for some salty products. In order to prompt higher sales ofproducts with less salt, they could stock fewer high-salt products and instead feature moresalt-free (or low-salt) products on the shelves, and change where in the store saltierproducts are displayed. They could engage in customer education to the extent theythought it effective. They might develop their own traffic light system for package label-ing. They could also use price strategies to shift buying practices in ways that mostefficiently allowed Wal-Mart to meet its goal. Some combination of these approacheswould most likely be tried, and creative firms like Wal-Mart could probably developsalt-reduction strategies that are not now readily imaginable. The upshot would likelyinclude shifting the US diet away from processed foods and toward healthier, naturalfoods.

I concede that it is possible that a single-minded focus on lowered salt could possiblyshift consumption to foods that are high in sugar and fat, and hence be unhealthy in otherways. This is the familiar “unintended consequences” point and suggests that any schemedesigned to reduce salt consumption must, at a minimum, consider whether a compan-ion feature should also aim at containing, or better yet, reducing fat and sugar consump-tion, something I put aside for the present.

Wal-Mart might try to reach its salt-reduction target in ways that differed sharplyfrom those tried by other retailers. Over time, competition among these sellers mightreveal the best current strategy for meeting the enterprise’s reduced salt target andremaining most profitable. To be sure, some companies may simply jump on whatappears to be the most successful bandwagon invented by others. Still, industry leadersare likely to adopt innovative approaches that give them at least a short-term financialedge. These leaders are likely to work with both public health professionals and expertswithin firms that make processed foods to design strategies for reducing Americans’consumption of salt while maintaining the pleasure of eating. Put differently, the mostdynamic firms may organize their efforts around learning communities created topromote the effective reduction of salt (APICS Learning Communities).

This regime needs teeth to have real bite, and so it is critical to specify what happensif the regulated firms do not meet their goals. The general approach of performance-based regulation is to impose fees that reflect the social costs that occur from salt intakebeyond the target. It would be highly contentious as to what those social costs are. Theunderlying idea, however, is to give the regulated companies a strong financial incentiveto engage in salt-reduction activities. Getting the penalty level exactly right would not beeasy. Given that uncertainty, other things being equal, it is probably better to err on thehigh side as a technology-forcing strategy. In any event, if a sensibly staffed regulatoryagency were put in charge of the regime, I am again confident that it could forge areasonably workable solution.

Some might worry that food retailers would focus their attention on reducing saltconsumption in only certain stores or customers; for example, getting wealthier people tocut back sharply while leaving inner-city poor people no better off in terms of theirhealth. This is equivalent to “hot spot” concerns raised about the use of performance-based regulation to deal with pollution from power plants. The worry there is that public



utilities will meet their overall targets by lowering emissions in plants located in higherincome areas, leaving in place high pollution levels in urban poor communities. Whether,as an empirical matter, this would be a real concern with respect to salt reduction is notclear. Perhaps, to the contrary, Wal-Mart might find it easier either to focus its efforts onlow-income communities where the prospects for the biggest reductions in salt use lie, orto adopt a uniform policy across all of its stores. But if there turns out to be a “hot spot”problem with salt reduction, this could be countered by complicating the regulatoryscheme somewhat and imposing the performance-based targets at the store level.

IV. Comparing performance-based regulation of retailers withregulatory alternatives

Performance-based regulation is not the only way governments could act to move beyonddisclosure and self-regulation strategies. In this part, various alternatives are comparedand contrasted. To be emphasized at the outset are critical dimensions of those compari-sons. These include (i) whether the scheme demands “inputs” or “outcomes;” (ii) thenature of the regulatory body – for example, an agency or courts; (iii) whether theregulation is aimed at the entire industry or firm-by-firm; (iv) how consumer behavior ismeant to be shaped; for example, whether responses to changes in price are central orwhether other or additional mechanisms are envisioned; (v) the nature of the penaltiesimposed on non-compliant firms; and (vi) the administrative complexity of the scheme.

A. Cap and tradePerformance-based regulation is actually quite similar to a “cap and trade” scheme. Thatregulatory approach, currently in fashion as a way to deal with climate change, usessomewhat different language to do largely the same thing that performance-based regu-lation would do – demand outcomes from private industry. If a “cap and trade” schemewere applied to salt, instead of assigning companies regulatory “targets” (as performance-based regulation does), firms would be required to have government-issued “permits” forthe salt contained in their food. The total quantity of salt permits available in the societywould be reduced each year until they amount to, say, 50% of the current level, which isparallel to a reduced target over time in a performance-based regulation scheme. Ifgovernment were to award salt permits to existing retailers, say, based on the total volumeof food they currently sell, the schemes would be very similar indeed.

Under the “trade” part of the “cap and trade” approach, companies would be allowedto buy and sell permits, a feature designed to achieve the social goal in the most efficientmanner, making “cap and trade” more of an industry-wide scheme than the regulation offirms one by one. Notice that performance-based regulation could be constructed in thisway too, so that were Costco, for example, to reduce the salt in its food by even more thanits performance-based target requires, it would be allowed to sell its excess compliance toSafeway, for example. (One argument for including a “trade” feature in the performance-based regulatory scheme is that this facilitates new firms entering the food market sellingproducts containing some salt, while at the same time assigning them a salt target ofzero.) The two approaches become even more alike if the penalties under the “cap andtrade” plan (for selling without a permit) were set equal to the fees imposed for non-compliance under the performance-based regulation plan.

Sometimes it is proposed that the necessary permits in a “cap and trade” schemeshould be auctioned off by government rather than freely awarded to existing firms.



Making performance-based regulation analogous to that approach would imply giving allfirms an initial target of zero salt and then requiring them to buy from government theallowance they need, with the total allowance available for purchase declining each yearuntil the overall social objective of a 50% reduction in salt is reached.

In short, by tinkering with the details, performance-based regulation and “cap andtrade” regulation can be made pretty much functionally the same. This reveals thatessentially common approaches have seemingly been accidentally framed arounddifferent labels.

How does performance-based regulation compare with other regulatory strategies?While the analysis so far has emphasized similarities, the following comparisons willfocus on key differences.

B. Command and controlUnder a traditional “command and control” regulatory regime, the relevant governmentagency would probably deal with the salt problem by sorting food products into catego-ries and having public officials decide how much salt would be permitted to be added toeach food category (Havas et al. 2007).6 Assume that financial penalties, analogous to thefees imposed under performance-based regulation, were imposed on those who addmore salt than is allowed. Such a scheme would be complex and contentious, and perhapsdifficult to oversee as all (or at least many) existing and new food products would have tobe assigned to a salt-reduction category and monitored for compliance. Among otherthings this is likely to dramatically increase the number of regulated firms as comparedwith imposing performance-based regulation on large retailers as proposed above. More-over, in the US context one can foresee a litigation nightmare over what foods go in whichcategories and how much salt reduction is required for each of them.

Notice further that, unless people keep eating the same array and quantities of food(an unlikely result were the salt in many foods sharply reduced), regulators would havelittle idea whether the salt reductions they demand of food producers would actually yielda halving of the amount of salt that people consume.

In any event, a key difference to emphasize between an “input”-oriented commandand control scheme and performance-based regulation is that the latter relies uponprivate firms rather than regulatory officials to decide how to achieve society’s overallsalt-reduction target. The hope under performance-based regulation is that, by havingfood retailers in charge, consumer preferences will be maximized, subject to the publichealth constraint of an overall reduction in salt consumption. Government regulators arelikely to be much worse at blending those two goals.

C. Excise taxesNext compare performance-based regulation with a regulatory approach that taxedthe amount of salt contained in food (Wilson 2004). This strategy might well beimplemented at the food-maker level. For example, product lines like Nabisco andFrito-Lay, who add salt to the food they manufacture, might be taxed on the sale oftheir products to retailers (or wholesale distributors). Restaurants like Wendy’s wouldbe taxed on the food they make and serve. Alternatively, it might be possible to imposethe salt tax even further back up the food chain – on sales by salt companies such asMorton Salt and Cargill to their food-maker customers (although this would require



giving additional attention to imported salt and imported food already containing salt)– or further down the chain on retailers when they sell products with salt in them toconsumers.

Regardless of where the tax is imposed, under this approach food-makers would havetwo main choices: they could keep the salt content of their products as it is and see theprice of what they make go up, or reduce (or even eliminate) salt from their products asa way to reduce (or eliminate) the tax burden. Were the tax high enough, then forproducts where salt reductions are reasonably easy to make without causing significantchanges in taste, food-makers would likely be forced by competition to reduce the saltthey currently add (perhaps substituting other flavor enhancers). For other products,food-makers might well feel the need to keep salt levels as they are now, and thoseproducts would become more costly to consumers. Market responses to higher priceswould cause at least some drop in consumption of saltier products, thereby also contrib-uting to a reduction in salt in consumers’ diets.

That said, we can now better compare performance-based regulation with a system ofimposing excise taxes on salt. Notice that an excise tax scheme, like the command andcontrol approach and unlike performance-based regulation, has no overall salt-reductiontarget as part of the plan. The legislature (or regulatory agency) adopting the salt taxcould, of course, manipulate the level of the tax over time to try to achieve the sort of saltreduction target demanded by performance-based regulation. But performance-basedregulation embraces a specified public health goal up front.

The excise tax scheme would generate government revenues since clearly not allproducts would become salt-free. This would provide government with income thatwould have to be used for some political purpose (note the analogy to the generation ofrevenue that would occur under a “cap and trade” scheme if the government were to sellsalt permits). Put differently, an excise tax is applied from the get-go, whereas underperformance-based regulation, the fee is imposed only if a firm fails to meet its target.This latter feature of the performance-based approach might be thought of as providingan exemption from tax up to the amount of the allowable salt level. The result is that,under performance-based regulation, if all companies meet their goals, there would be norevenues paid into the government (analogous to a “cap and trade” plan in which saltpermits are handed out). In an era of resistance to new taxes, this could be a strongpolitical plus.

Notice how the tax approach relies for its public health gains on a combination ofactions by food-makers and consumers that occur in response to the tax. Performance-based regulation, by contrast, counts more directly on retailers to drive behavioralchanges as a way to avoid penalties. As a result, companies like Wal-Mart would takecharge of the problem, instead of simply passing on to customers the taxes that have beenimposed on their suppliers.

Whether it would be better for large retailers to be directly on the hook may, in theend, depend on the real world of food marketing. If large retailers have real market power(which is probably the case), then it might well be desirable to put them at the center ofthe regulatory scheme. Otherwise it might be that the choice between performance-basedregulation, an excise tax, and a “cap and trade” scheme turns most importantly on whichframing is politically most attractive.

Note, further, that a still different way to deploy performance-based regulation wouldbe to aim it at salt-producing companies like Morton and Cargill, requiring them to sell



sharply reduced amounts of their product over time – an approach that surely would beattractive from the administrative perspective since so few firms now account for most ofthe salt sold in the US. Putting aside what could well be serious concerns about saltsmuggling, imported salt, and imported foods containing salt, this approach wouldpresumably cause a rise in the price of salt with consequences similar to an excise tax.Notice again the contrast with performance-based regulation aimed at retailers. Restrict-ing the amount of salt that Morton and Cargill can sell leaves it up to the market to playout how that translates into a changed pattern of food consumption. My proposal countson the creativity and initiative of Wal-Mart and the like to manage the best way to meetconsumer preferences constrained by an ever-reduced quantity of salt consumption.Obviously, those who don’t trust Wal-Mart to do anything in furtherance of the publicinterest would be skeptical about the social appeal of this strategy.

D. Tort liabilityA brief comparison between performance-based regulation and tort liability may also behelpful. Traditionally, tort liability is based on fault (proof of negligence). This is also true,as a practical matter, for claims about the design and labeling of products under “productliability” law (Restatement (Third) of Torts, Product Liability § 2 (1998)). That makes tortlaw most analogous to command and control regulation, except that it is the judicialsystem (including juries in the US), rather than administrative agencies or legislatures,that decide what specific safety precautions product sellers should take (or should havetaken).

In other words, were conventional tort law applied to our problem, it would be bestunderstood as an “input” control scheme in which the judicial system decides how muchsalt each product would be allowed to have before it is deemed “defective,” therebysubjecting its seller to liability. Although, under existing product liability law, food pro-ducers would probably be freed from liability so long as they provided an adequatewarning as to the amount of salt in their products, it is at least imaginable that juriescould be deployed to decide whether certain products are “too salty.” Whether jurieswould be any good at making that sort of decision is highly debatable, however.

But this is not the only sort of tort law that one could imagine. Instead, governmentcould adopt a regime of true “strict liability,” in which no proof of negligence (or “defect”)is required. Such a regime would be more like performance-based regulation because inboth schemes the covered firms are left to decide how to improve outcomes and are heldresponsible if they do not. However, notice that strict tort liability in this respect is evenmore like an excise tax (imposed by and paid to victims) than it is like performance-basedregulation. This is because, as with the excise tax, the firm would bear the costs of allnegative outcomes. By contrast, as already noted, performance-based regulation wouldhold firms to a standard that required them to substantially reduce, but not eliminate, theharms their products cause, and so they would not face penalties if they met their targets.In any event, it should be emphasized that applying true strict liability in tort to ourproblem would be as sharp a change from current law as would be applying performance-based regulation to salt.

Moreover, in the real world, as an administrative matter, applying tort liability of anytype to this problem would be extremely problematic because of the need to prove thatthe individual claimant suffered a specific harm of which salt was in fact the cause.Performance-based regulation is free from this problem of individual causal matching.



E. Management-based regulationOne more regulatory alternative to which performance-based regulation can be com-pared is management-based regulation, although this label has been applied to more thanone strategy (Coglianese & Lazer 2003). For these purposes, performance-based andmanagement-based regulation can be understood to share the common feature of requir-ing the regulated firms to adopt a plan designed to achieve a socially designated goal. Forexample, large retailers like Safeway and McDonald’s would have to submit to the regu-lators their plans to reduce the salt consumption of their consumers by, say, 50% in 10years. In this respect, both approaches are outcome-oriented and rely on enterprises tofigure out how best to achieve the desired target.

However, the typical management-based approach does not penalize firms for failingto achieve their goals. Rather, the only actual requirement they face is to adopt a plan (or,in a stronger version of the scheme, obtain regulatory approval of a plan reasonablydesigned to meet their goals). Hence, in this respect management-based regulation ismore of a process-oriented scheme. It would be attractive to business because, in effect,it requires only a good faith effort on their part. For the same reason it puts less pressureon industry because it does not insist on actual accomplishments.

In sum, considering all of the alternatives canvassed here, if a specified reductionin the amount of salt consumed by Americans is the desired social outcome,then performance-based regulation of retailers has some comparatively appealingfeatures.

V. Even bolder ways of using performance-based regulation

A. Performance-based regulation using reduced blood pressure as a targetImagine that my proposed plan of using performance-based regulation to enlist theefforts of retailers to reduce people’s salt consumption were put into effect and wassuccessful. Assume that by the end of 10 years, food sold by US retailers, in the aggregate,contains half the salt that it now contains (subject to adjustments of the sort alreadynoted). Assume further that this translates into the American people reducing their saltconsumption by 50%, thereby meeting the AMA goal.

However, a question remains: who individually would have reduced their salt intake?To the extent that those who already have relatively low blood pressure reduced their saltintake, there is reason to be skeptical about whether they would enjoy any importanthealth benefits (Moore 1990; Taubes 1998).

For those who now have decidedly high blood pressure, a large overall reduction insalt intake nationwide would probably lead to substantial salt reduction and in turn bloodpressure reduction for at least some of them. This should translate into substantial publichealth gains for that group.

Yet people with decidedly high blood pressure account for only a modest portion ofthe US public – less than 30%. What about everyone else? Here we are thrust into acontentious issue. Some think that those in this large in-between group would benefitlittle from salt reduction in terms of reducing their blood pressure (Moore 1990;Midgley et al. 1996; Taubes 1998). Even if we assume they did, some think that little orno health benefit would come from any such blood pressure reduction (Moore 1990).Others, by contrast, think a huge number of lives would be lengthened significantly byachieving even a modest lowering of national average blood pressure readings (Intersalt



Cooperative Research Group 1988; Stamler et al. 1989; Elliot et al. 1996; Cutler et al.1997; He & MacGregor 2002; Havas et al. 2007).

Even if we assume the latter view is correct, a separate point is that many who wouldotherwise have done things like take medication, exercise more, or lose weight in order toreduce their blood pressure might not do those things if they lowered their blood pressurethrough salt reduction. These people might have reduced their blood pressure by roughlythe same amount whether or not they reduced the salt in their diets. In terms of bloodpressure alone. In terms of blood pressure alone these people might be no better off. Still,many would probably think it better to reduce blood pressure by reducing salt intake thanby taking medication.

The general point is that using a measure like reduced national salt consumption asa performance-based target imposed on retailers may not necessarily be the best publichealth target, and adopting a more fine-tuned target might conceivably be more effective.

B. Performance-based regulation of health care providersThis line of analysis suggests thinking about the problem in very different terms. Insteadof imposing performance-based regulation on food retailers, suppose we were to imposeit on health care plans (or insurers). The idea is that, rather than impacting bloodpressure indirectly via salt reduction as the outcome target, we could focus directly on alowered blood pressure target.

If large health care plans – such as Blue Cross, Kaiser Permanente, and Health Net –were required to reduce the blood pressure of their members, we might not use anacross-the-board reduction in blood pressure as the overall target. Rather, we might wantmuch of the reduction to come from those with relatively higher blood pressure. Hence,the target might be phrased as reducing the number of patients with high blood pressureby 5% each year or reducing the blood pressure of those in the above average bloodpressure group by 5% each year. Or perhaps different targets could be set for those at veryhigh versus moderately elevated blood pressure levels. Also, analogous to the point madeearlier about Whole Foods and salt, health care plans that have already engaged in effectivedisease prevention efforts might be given appropriate credit for this in setting their targets.

Obviously, it would be wrong to expect doctors or insurers to reduce the bloodpressure of everyone in their group to a healthy level. But, as with my proposal ofapplying performance-based regulation to food retailers, the targets could aim for apartial reduction.

Under performance-based regulation, the health care plan could employ a variety ofbehavior-changing strategies. Some of these might well be aimed at salt consumption. Forexample, health care professionals could advise patients to eat fewer products that arehigh in salt and to eat smaller portions of such products, to eat reduced-salt products, toshop generally at stores selling healthier products, and so on. While many health careprofessionals might protest that they cannot control their patients’ eating behavior, thisis only partly true. After all, in general, they are fairly good at controlling the drug-takingbehavior of their patients. The point is that if health care plans had a financial stake inreducing salt consumption by their patients, I am confident that they would developinventive measures to be deployed toward achieving that goal.

But of course if the health care plans’ regulatory target was reduced blood pressure intheir patients, then encouraging reduced salt consumption is not the only approach thoseplans might try. They might instead focus on having patients exercise more, lose weight,



reduce stress, or take blood pressure-lowering medication. This raises the question ofwhether health care groups would adopt a strategy that may be cheapest and mosteffective for them, but which has riskier side effects for their patients, such as prescribingmedication to those who could readily lower their blood pressure via behavioral changes.Yet this risk should not be exaggerated. One check against this perverse response is thatif health care plans push their patients to do things that are ultimately bad for their health,this could hurt the health care group down the line.

A separate problem is that if health care plans are the target of a performance-basedregulation plan, they might not be eager to accept patients with high blood pressure, orperhaps more precisely, those who are unlikely to lower their blood pressure despiteefforts by the health care plan. This could cause several health care plans to try to screenout some patients at the front end or drive them out along the way. To discourage thatpractice, there would have to be some regulatory controls that made this sort of gamingstrategy largely ineffective. That, of course, would add to the administrative burden ofthe plan.

It should also be noted that a performance-based regulation scheme aimed athealth care plans need not be enforced by imposing fees for failure to achieve theiroutcome goals. That financial incentive was proposed in the context of the applyingperformance-based regulation to food retailers on the ground that, since those com-panies profit from the sale of salt, it is only fair that they take responsibility for reduc-ing the negative social consequences of salt consumption. Yet, indeed for eitherapproach, one could instead envision a system of rewards. Imagine that health careplans could earn cash payments from the government if they lowered the blood pres-sure of their patients. Public expenditure could be justified by the reduced burden thesenow healthier patients would later impose on the Medicare and Medicaid systems.(One could imagine a more complex scheme in which both carrots and sticks areemployed. Regulated firms not meeting their targets could be fined and the money usedto reward those who do.)

If we assume a system of financial rewards, participating health care plans might passmost of those rewards on to the doctors in their networks based on physicians’ individualsuccess with their patients. One concern about any such reward scheme is that doctorsmight just ignore the rewards and thereby make the whole program meaningless.

Finally, there is clearly a risk that some health care plans or individual doctors wouldcheat and some of those would get away with it. For example, in order to obtain thereward, some might over-report the patient’s blood pressure at the start or under-reportit later (or both). This implies that the agency running the regulatory scheme would haveto employ auditing strategies designed to discourage such fraud. In the end, perhaps thesevarious administrative problems would be so substantial to make this performance-basedstrategy impractical to operate.

Were it tried out, however, it is worth noting that health plans and doctors mightchoose to pass some of their financial rewards on to their patients as a way of winningtheir cooperation in changing their behavior.

This line of thinking might suggest to some that rather than involving health careproviders at all, an even simpler performance-based regulatory scheme would have thegovernment directly providing financial rewards to individuals who bring their bloodpressure down from high to healthy levels. Yet one has to be skeptical about howsuccessful people will be in achieving this health improvement by acting on their own.



Indeed, the fundamental understanding behind public health policy is that simplyrelying on individual citizen-level behavioral changes will not suffice. Otherwise, as away of preventing dental caries, for example, it would seemingly be enough to offerfluoride toothpaste in the marketplace and not bother with putting fluoride in thewater supply.

C. Performance-based regulation using healthy people as a targetUpon even further consideration, it becomes clear that merely reducing salt consump-tion or merely reducing blood pressure, even in patients with high blood pressure, isnot really the health outcome that matters most. Ultimately, what we want is for peopleto have longer and healthier lives, especially when that can be achieved at a reasonablecost and without unreasonable risk. Hence, one could imagine taking performance-based regulation one step further and base penalties or rewards on those end-pointoutcomes.

For example, the regulatory regime could judge retailers, not on how much theyreduce the salt in their products, but on whether a group of people assigned to them liveslonger and healthier. Maybe Wal-Mart would have to take responsibility for everyone inTexas and Arkansas. Costco might get those in Washington and Oregon, while 7–11 mightget those in Colorado, and so on. Or maybe, instead of retailers, food-makers like PepsiCoand Kraft are the ones who should be given these obligations.

Of course, it seems more straightforward to demand that PepsiCo reduce the salt inthe potato chips it makes, than to demand that PepsiCo improves the health of the peopleof the state of New York. Nonetheless, performance-based regulation need not require atight fit between altering the regulated party’s product and achieving its regulatory target.Put differently, it is not necessarily a defense that the regulated enterprise might be unableto comply with the regime by acting alone. Companies would simply have to figure outwho could join with them to make a difference and then enlist the others in the cause.PepsiCo might engage the doctors, teachers, and television stations in the geographic areafor which it is responsible.

It would be crucial, of course, to convince the public (and legislators) that theproducts that PepsiCo makes are sufficiently harmful to health that imposing the obli-gation on PepsiCo to improve public health outcomes is only fair. I concede that it ispolitically far more likely that advocates would be successful in requiring Wal-Mart orPepsiCo to reduce the salt in food they sell than requiring them to achieve general healthgains in an identified population. Still, the point is that there is nothing in the nature ofperformance-based regulation that precludes adopting more sweeping targets that arefurther along the road to the ultimate social objective.

Moreover, remember that performance-based regulation need not be anchored inpenalties for non-compliance, but might instead be based on financial rewards for theachievement of societal goals. So paying bonuses to PepsiCo for improved public healthin the population assigned to it might be politically more popular than penalizing it forfailing to do so – and clearly the bonus scheme would be more acceptable to PepsiCo.From the governmental perspective, such payments could be justified on cost–benefitgrounds – at least if the cost of paying for improved patient health now is less than thepresent value of saved medical costs down the line.

Notice that this sort of payment for performance is very different from actual ongoingexperiments in the medical services delivery world in which “pay for performance” has to



do with paying health care workers for providing specified medical interventions – likegiving aspirin to patients who arrive at emergency rooms complaining of chest pain.Those are rewards for inputs, not outcomes (Centers for Medicare and Medicaid Services2005).

Conclusion

To sum up, performance-based regulation is perhaps best understood as a generalstrategy designed to financially entice identified (typically non-governmental) actorsto take responsibility for achieving socially desirable outcomes. Like most publichealth interventions, it is motivated by a degree of paternalism rooted in the beliefthat all too many people are just unable on their own to take steps that are needed toserve their own best interests. Yet, with performance-based regulation, the public healthgoal is sought to be achieved through mechanisms that otherwise maximize consumerpreference.

As illustrated here, there often are many options as to how to set the regulatorytarget, and there are many parties on whom the performance-based regime might beimposed. Inevitably, a number of trade-offs are involved in deciding what specificperformance-based regulatory regime to adopt, including political acceptability, admin-istrative ease, assured public health outcomes, and so on. To be sure, it is unclearwhether performance-based regulation of even the most modest sort (e.g. forcingretailers to reduce the amount of salt they sell) is politically practicable. At the least, thisrepresents a new way of thinking about public health regulation that, I believe, deservesfurther analysis and evaluation.

Whether we should have confidence that private actors faced with new financialincentives will indeed effectively act in ways that promote the public health is sure to becontroversial. Yet having a government agency decide how much salt should be in everyfood product will surely be more so. And simply relying on providing better informationand voluntary action by the business community will, I fear, leave our society far short ofour public health goals (Sugarman & Pramming 2008).

Acknowledgment

The author thanks Christine Fujita, Berkeley Law 2009, for research assistance.

Notes

1 The risks of developing cardiovascular disease (CVD) increase progressively the higher blood

pressure levels rise over 115/75 mm Hg. About half of deaths related to CVD are attributed to

hypertension (defined as systolic blood pressure �140 mm Hg and/or diastolic blood pressure

�90 mm Hg or taking antihypertensive medication).

2 According to a report for the AMA: “Substantial public health benefits accrue from small

reductions in the population blood pressure distribution, achievable with long-term modest

reduction in sodium intake. A 1265-mg/d lower lifetime intake translates into an approximately

5-mm Hg smaller rise in SBP as individuals age from 25 to 55 years. This corresponds to a 20%

lower prevalence of hypertension and a reduction in mortality rates of 9% for CHD, 14% for

stroke, and 7% for death from all causes and would save 150,000 lives annually.”



3 The US Department of Health and Human Services has also set a goal that at least 65% of the

population should consume less than 2,400 mg/d of sodium a day (US Department of Health

and Human Services 2000b).

4 In Finland, the government requires foods that exceed specified limits to be labeled as “high

salt.” In New Zealand and Australia, the National Heart Foundations have worked with food-

makers to reformulate products and have created a labeling program, which identifies foods

that meet strict standards for sodium.

5 For a far more detailed performance-based regulation scheme as applied to childhood obesity,

see Sugarman & Sandman (2007).

6 For example, in the United Kingdom, the Food Standards Agency has divided processed foods

into about 70 categories (e.g. breads, canned vegetables, soups). For each category, it has

calculated the percentage of dietary sodium derived from each, and set voluntary reduction

targets.

References

APICS (Advancing Productivity, Innovation, and Competitive Success). Learning Communities.

[Last accessed 10 Mar 2009.] Available from URL: http://www.apics.org/Resources/

LearningCommunities/

BBC News (2008) Salt Intake “Is Starting to Fall”. [Last accessed 10 Mar 2009.] Available from URL:

http://news.bbc.co.uk/1/hi/health/7518820.stm.

CDC (2007) Cigarette Smoking among Adults – United States, 2006. Morbidity and Mortality

Weekly Report 56, 1157–1161.

Centers for Medicare and Medicaid Services (2005) Medicare “Pay for Performance (P4P)” Initia-

tives. [Last accessed 10 Mar 2009.] Available from URL: http://www.cms.hhs.gov/apps/media/

press/release.asp?counter=1343.

Chaloupka FJ, Warner KE (2000) The Economics of Smoking. In: Culyer AJ, Newhouse JP (eds)

The Handbook of Health Economics, pp. 1539–1627. Elsevier, Amsterdam.

Chaloupka FJ, Wakefield M, Czart C (2001) Taxing Tobacco: The Impact of Tobacco Taxes on

Cigarette Smoking and other Tobacco Use. In: Rabin RL, Sugarman SD (eds) Regulating

Tobacco, pp. 39–71. Oxford University Press, Oxford.

Coglianese C, Lazer D (2003) Management-based Regulation: Prescribing Private Management to

Achieve Public Goals. Law & Society Review 37, 691–730.

Coglianese C, Nash J, Olmstead T (2003) Performance-based Regulation: Prospects and Limita-

tions in Health, Safety and Environmental Protection. Administrative Law Review 55, 705–730.

Cutler JA, Follmann D, Allender PS (1997) Randomized Trials of Sodium Reduction: An Overview.

American Journal of Clinical Nutrition 65 (Suppl.), S643–S651.

Dickinson BD, Havas S (2007) Reducing the Population Burden of Cardiovascular Disease by

Reducing Sodium Intake: A Report of the Council on Science and Public Health, American

Medical Association. Archives of Internal Medicine 167, 1460–1468.

Elliot P, Stamler J, Nichols R, Dyer AR, Stamler R, Kesteloot H et al. (1996) Intersalt Revisited:

Further Analysis of 24 Hour Sodium Excretions and Blood Pressure Within and Across Popu-

lations. British Medical Journal 312, 1249–1253.

Food and Drink Federation (undated) Salt Reduction: The Facts. [Last accessed 10 Mar 2009.]

Available from URL: https://www.fdf.org.uk/resources/salt-thefactsfinal1.pdf.

Food Standards Agency (2008a) Salt. [Last accessed 10 Mar 2009.] Available from URL: http://

www.salt.gov.uk/.

Food Standards Agency (2008b) Traffic Light Labelling. [Last accessed 10 Mar 2009.] Available from

URL: http://www.eatwell.gov.uk/foodlabels/trafficlights/.

Havas S, Dickinson BD, Wilson M (2007) The Urgent Need to Reduce Sodium Consumption.

Journal of the American Medical Association 298, 1439–1441.



http://www.apics.org/Resources

http://news.bbc.co.uk/1/hi/health/7518820.stm

http://www.cms.hhs.gov/apps/media

https://www.fdf.org.uk/resources/salt-thefactsfinal1.pdf

http://www.salt.gov.uk

http://www.salt.gov.uk

http://www.eatwell.gov.uk/foodlabels/trafficlights

He FJ, MacGregor GA (2002) Effects of Modest Salt Reduction on Blood Pressure: A Meta-analysis

of Randomized Trials: Implications for Public Health. Journal of Human Hypertension 16,

761–770.

Institute of Medicine, Food and Nutrition Board (2004) Dietary Reference Intakes for Water,

Potassium, Sodium, Chloride, and Sulfate. National Academies Press, Washington, DC.

Intersalt Cooperative Research Group (1988) Intersalt: An International Study of Electrolyte Excre-

tion and Blood Pressure: Results for 24 Hour Urinary Sodium and Potassium Excretion. British

Medical Journal 297, 319–328.

Martinez SW (2007) The U.S. Food Marketing System: Recent Developments, 1997–2006, ERR-42. US

Dept. of Agriculture, Econ. Res. Serv.

May PJ (2003) Performance-based Regulation and Regulatory Regimes: The Saga of Leaky Build-

ings. Law & Policy 25, 381–401.

Midgley JP, Matthew AG, Greenwood CMT, Logan AG (1996) Effect of Reduced Dietary Sodium

on Blood Pressure: A Meta-Analysis. Journal of the American Medical Association 275, 1590–

1597.

Moore TJ (1990) Overkill. The Washingtonian 64–67, 194–204.

National Performance Review (1995) Reinventing Environmental Regulation. [Last accessed 10 Mar

2009.] Available from URL: http://govinfo.library.unt.edu/npr/library/rsreport/251a.html.

Pierce JP, White MM, Gilpin EA (2005) Adolescent Smoking Decline During California’s Tobacco

Control Programme. Tobacco Control 14, 207–212.

Reich RB (2007) Supercapitalism: The Transformation of Business, Democracy, and Everyday Life.

Knopf, New York.

Restatement (Third) of Torts, Product Liability § 2 (1998).

Rose G (2008) Rose’s Strategy of Preventive Medicine. Oxford University Press, Oxford.

Scientific Advisory Committee on Nutrition (2003) Salt and Health. [Last accessed 10 Mar 2009.]

Available from URL: http://www.food.gov.uk/multimedia/pdfs/saltandhealth0503.pdf.

Social Marketing Institute (2008). Social Marketing. [Last accessed 10 Mar 2009.] Available from

URL: http://www.social-marketing.org/sm.html.

Stamler J, Rose G, Stamler R, Elliot P, Dyer A, Marmot M (1989) INTERSALT Study Findings.

Public Health and Medical Care Implications. Hypertension 14, 570–577.

Sugarman S (2009) No More Business as Usual: Enticing Companies to Sharply Lower the Public

Health Costs of the Products They Sell. Public Health 123, 275–279.

Sugarman SD (2008) Should Tobacco Companies Be Required to Reduce Their Number of Customers?

[Last accessed 10 Mar 2009.] Available from URL: http://www.alternet.org/story/73783/.

Sugarman SD, Pramming S (2008) Head to Head, Should We Use Regulation to Demand Improved

Public Health Outcomes from Industry? Yes and No. British Medical Journal 337, a1750.

[Last accessed 10 Mar 2009.] Available from URL: http://www.bmj.com/cgi/content/full/337/

oct02_2/a1750.

Sugarman SD, Sandman N (2007) Fighting Childhood Obesity Through Performance-based Regu-

lation of the Food Industry. Duke Law Journal 56, 1403–1490.

Taubes G (1998) The (Political) Science of Salt. Science 281, 898–907.

The Environmental Literacy Council (undated) Salt. [Last accessed 2 Jan 2009.] Available from

URL: http://www.enviroliteracy.org/article.php/1184.html.

US Department of Health and Human Services (2000a) Reducing Tobacco Use: A Report of the

Surgeon General. US Department of Health and Human Services, Centers for Disease Control

and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office

on Smoking and Health, Atlanta, GA.

US Department of Health and Human Services (2000b) Healthy People 2010, 2nd edn. With

Understanding and Improving Health and Objectives for Improving Health, 2 vols. US Govern-

ment Printing Office, Washington, DC.



http://govinfo.library.unt.edu/npr/library/rsreport/251a.html

http://www.food.gov.uk/multimedia/pdfs/saltandhealth0503.pdf

http://www.social-marketing.org/sm.html

http://www.alternet.org/story/73783

http://www.bmj.com/cgi/content/full/337

http://www.enviroliteracy.org/article.php/1184.html

Weinreich Communications (2008). Homepage. [Last accessed 10 Mar 2009.] Available from URL:

http://www.social-marketing.com.

Wilson N (2004) Salt Tax Could Reduce Population’s Salt Intake. British Medical Journal 329: 918.

[Last accessed 10 Mar 2009.] Available from URL: http://www.bmj.com/cgi/content/extract/

329/7471/918-c.

Laws cited

21 C.F.R. § 101.61 (2008).

21 C.F.R. § 101.100 (2008).

Nutrition Labeling and Education Act of 1990, 21 U.S.C. § 343-1 (1990).



http://www.social-marketing.com

http://www.bmj.com/cgi/content/extract

salt, high blood pressure, and performance-based regulation

Documents