Your daily coffee shop, favorite brunch spot, and go-to fancy dinner restaurant are not subject to any federal food safety standards. Average consumers are instead protected only by the food safety standards of their local regulatory agency, whatever those standards may be. For decades, the U.S. Food and Drug Administration’s (FDA) approach to food retail establishments has differed from its approach to all other areas within its purview. The lack of federal governance and half-hearted attempt to implement a command-and-control regulatory model have led to a breakdown in food retail regulation. Part I of this Note reviews the history of the FDA’s oversight of the food retail industry and presents the regulatory structure in three American cities. Part II examines the issues that arise from an incomplete federal governance model, including challenges addressing foodborne illness, consumer confusion due to systemic flaws, and implicit bias among food inspectors and consumers. Part III discusses the benefits of federalizing food retail safety and highlights important factors for a successful regulatory scheme.
Over the last several decades, American dining culture has shifted increasingly away from homes and into restaurants. Nowadays, Americans spend almost half of their food budget eating outside of their homes and consume nearly one-third of their total calories on meals in restaurants. Eating out is quicker and cooler, worth notifying one’s social media fans via a perfectly staged Instagram photo. However, more frequently eating out also increases the risk of exposure to food pathogens. Consumers have limited control over how their food is prepared, relying on restaurants to ensure the safety of their meals. Thus, until its sudden onset, foodborne illness is rarely a conscious concern.
The Centers for Disease Control and Prevention (CDC) estimates that 48 million foodborne illness cases occur in the United States every year. Of those instances, at least 128,000 Americans are hospitalized and 3,000 die after eating contaminated food. More than two-thirds of all cases were traced back to a restaurant setting. Given the lack of consumer control and a mix of factors contributing to foodborne illness, there is an increasing need for clear and consistent federal oversight in the food retail industry. Well-structured federal regulation would allow food retail establishments to assume an affirmative duty to ensure their food is safe and unadulterated.
The U.S. Food and Drug Administration (FDA) is the agency currently responsible for overseeing many components of the food industry that impact the average consumer. But, unlike with much of the regulated food industry, the FDA’s guidance for food retail establishments—including restaurants—is advisory rather than mandatory. For instance, every four years, the FDA issues a Model Food Code that provides recommendations on maintaining public health standards in the food retail industry but does not require regulatory agencies to adopt them. Because the FDA does not mandate these standards, states and cities can choose to adopt the Model Food Code’s guidance or decide their own food retail safety standards. Most states have opted for the former to varying degrees and sometimes choose to partner with the FDA through training and voluntary program audits.
In this Note, I argue for the strengthening of the FDA’s current approach to food retail regulation, specifically in dining establishments. The FDA’s oversight of food retail has not kept pace with modern legislation’s move toward standardized guidelines across the food industry. Instead, the FDA’s feeble attempt at a command-and-control model has crumbled into a system that is de facto deregulated and lacks uniformity. Below, I discuss a few of the issues resulting from the FDA’s ill-defined regulatory stance and advocate for a formalized move toward federally mandated food retail safety standards.
The FDA governs a wide range of food-related industries. On a high level, the FDA oversees the safety and purity of all foods sold in interstate commerce, with the exception of meat, poultry, and egg products. In this process, the FDA regularly inspects food-processing and storage facilities, dairy farms, and animal-feed processors to ensure supply-chain quality. As part of its inspection process, the FDA covers more than thirty thousand domestic food manufacturers and more than twenty thousand food warehouses.
The U.S. Congress has gradually broadened the remit of the FDA over the last several decades due to the rise of interstate transportation of food products. To address concerns around economic fraud and food safety, Congress established national uniformity in nutrition labeling as part of the Nutrition Labeling and Education Act in the 1990s. In 2006, Congress further expanded the FDA’s oversight via the National Uniformity for Food Act, which ratified food safety warning labeling requirements for any product traveling in interstate commerce. For instance, all foods containing raw or partially cooked ingredients had to be labeled as such. Most recently, the Food Safety Modernization Act changed the FDA’s governance model from a reactive approach to a proactive approach, requiring preventative control measures and science-based produce-safety standards for food facilities.
By contrast, regulation of food retail safety has not kept pace with other legislative movements that have expanded the purview of the FDA. Since the early twentieth century, the FDA has kept the food retail industry at arm’s length, citing the preservation of state interests and challenges of scale. This position faltered only once, after a 1974 report by the Government Accountability Office (GAO). The GAO conducted an independent audit of almost two hundred restaurants across nine cities and found major health violations in over 90 percent of the surveyed establishments. It attributed the poor quality of food retail safety to the ineffectiveness of the FDA’s non-mandatory guidelines, as well as the FDA’s limited monitoring of state programs. The GAO proposed three types of national uniform standards that would minimize inconsistency across state jurisdictions: criteria for restaurant sanitation, enforcement measures to correct unsanitary conditions, and penalties imposed for incurring health violations. Congress’s subsequent proposal laid out the measures for national uniform food retail safety standards that federal agents would regulate. Congress withdrew the proposal only three years later, citing criticism from the states.
Throughout its existence, the FDA has operated as an advisory rather than a regulatory governance agency for the food retail industry. The FDA issued its first proposed restaurant-sanitation regulations in 1934. Although early versions of the Model Food Code included recommendations to issue letter grades as a summary of a restaurant inspection, the FDA moved away from that recommendation by the late 1970s. After briefly entertaining the concept of a one hundred-point grading system, the FDA now simply recommends that local food inspection agencies treat the inspection report as a public record, but provides no further guidance on best practices for public disclosure.
Today, state and local agencies are responsible for monitoring over one million food retail establishments across the United States, covering full-service restaurants, fast food chains, delis, and coffee shops. In so doing, state and municipal agencies serve as the enforcers to ensure food retail establishments comply with locally adopted standards and minimize the incidence of foodborne illness. In the absence of federal requirements, state and local agencies have discretion to adopt any portion of the Model Food Code or ignore it altogether. As a baseline matter, fewer than half of the sixty-six state regulatory agencies that currently follow the Model Food Code adhere to the most recent version. The majority of agencies choose not to update their requirements with each publication, presumably because of the high costs of adoption.
Although the range of regulatory schema has evolved over the years, for the purposes of this Note I will discuss three models. On one end of the spectrum is the command-and-control model, in which the government regulates individual actors according to centralized recommendations. The philosophy underlying command-and-control suggests a reliance on a government entity with the appropriate expertise to provide guidance and recommendations. On the other end lies the market-competition model, in which there is no regulatory governance. Instead, individual actors are left to their own devices to achieve economic success. A true free-market paradigm cannot exist in a realm as critical to public health as the food industry, but I will refer to this schema as a nod to private actors who self-regulate to stay in business. Finally, outcomes-based regulation strikes a middle ground, in which the government tells regulated parties what they must accomplish, but allows parties to accomplish these goals by any means they choose (the classic example being carbon emissions). This schema works best when the expertise lies with a small group of individual actors, perhaps because of the relative complexity or incentive structure of an industry.
The current food retail regulatory system is ill-defined and ambiguous because of the FDA’s decision to remain advisory. The FDA has all but set up a command-and-control regulatory framework but fails to take the final step of mandating any of its own guidelines, undoing the very foundation it has created. Many local agencies have thus incorporated elements of a command-and-control structure but have regressed into a market-competition regulatory model because of incomplete adoption. Three contrasting food retail safety systems help illustrate the range of regulation: New York City, Los Angeles, and Seattle.
A. New York City
As the most populous city in the United States, New York City has more than twenty-six thousand food retail establishments across the five boroughs. The New York City Department of Health and Mental Hygiene (DOHMH) employs approximately one hundred inspectors and strives for at least one unannounced onsite inspection at each food retail establishment per year. Of the test-case cities examined in this Note, New York City seems to have the lowest frequency of inspections, primarily because of the lack of available resourcing and the sheer number of establishments. In 2017, the DOHMH completed more than 40,700 inspections of full-service restaurants, bars, delis, and food trucks.
Since 2010, the DOHMH has issued letter grades at the end of each inspection. Inspectors evaluate an establishment’s practices, ranging from food handling to facility maintenance. Behaviors that are noncompliant with the city’s health code fall into three categories: public health violations, critical violations, and general violations. Public health violations trigger the greatest number of point violations and may spur an inspector to close a restaurant if it cannot immediately correct the violation. An example of a public health violation is the failure to keep foods at proper holding temperatures. The next category of critical violations includes activities that contribute to foodborne illness, such as failing to properly wash raw foods before serving. Lastly, a general violation pertains to activities such as facility maintenance, which may be peripherally related to foodborne illness. These are behaviors most easily observable by consumers, such as improper restroom facilities or inadequate food-ware sanitation. Violations across all categories are subject to a five-tier scale that indicates the severity of the condition and permits inspectors to assign more violation points at their discretion. Inspectors tally violation points at the end of a visit and convert the score into a letter grade: establishments with a score between zero and thirteen points receive an “A,” those with a score between fourteen and twenty-seven points receive a “B,” and those with twenty-eight points or more receive a “C.”
New York City’s food safety and inspection system functions closest to a classic command-and-control regulatory system in two major ways: the city (1) entrusts itself as the authority for instituting food safety, and (2) doles out pecuniary penalties for noncompliance. For any violation after the initial inspection, food retail establishments are required to pay fines associated with any observed violations during inspection. Fines can range anywhere from a few hundred to several thousand dollars. Food retail establishments can also be fined if they fail to prominently display their letter grades in a conspicuous part of the restaurant. These fines typically constitute a major part of DOHMH revenue.
As with any regulatory system, New York’s safety standards and inspection system has its critics. The most common complaint comes from restaurant owners, who argue that the idealized health code standards are not realistic in a fast-paced service environment. Others argue that the guidelines compromise the integrity of the food itself. A dish as simple as ice cream might trigger violation points if kept above forty-one degrees, at the expense of rendering it rock hard and difficult to serve. A few restauranteurs refuse to bend: renowned sushi chefs will brave the seven-point violation of preparing food without gloves to maintain pristine slicing technique during food service.
Given the massive scale of the New York City food retail system, local government has tackled the issue of food safety by implementing a quasi–command-and-control model. Although the system is not without its flaws, it is a paradigmatic example of how a large city manages the flow of millions of restaurant-goers on a daily basis.
B. Los Angeles
As the nation’s second-most populous city, Los Angeles boasts a robust restaurant scene, with more than twenty thousand establishments in the greater metro area. The Los Angeles Department of Public Health oversees food retail safety across the city and is home to one of the most longstanding restaurant-grading systems in the country. Restaurants are inspected anywhere from one to three times a year and follow a one hundred-point grading scale. Each incremental infraction subtracts from a restaurant’s score. Scores ranging from ninety to one hundred receive an “A,” eighty to eighty-nine receive a “B,” and seventy to seventy-nine receive a “C.” Restaurants that score fewer than seventy points receive a numerical score to be posted. Unlike New York, Los Angeles does not impose fines on restaurants for health violations. However, facilities that score below a seventy twice within a twelve-month period may be closed and subject to legal action.
Los Angeles food safety violations are classified as either critical risk factors or good retail practices. Critical risk factors directly contribute to foodborne illness. These factors include improper holding temperatures for foods and employees who are unable to properly wash their hands. As a threshold matter, Los Angeles lists only twenty-five discrete critical risk factors for inspection, substantially fewer than New York’s fifty-seven. Although some of the listed infractions overlap with those in New York, the defined violations on the Los Angeles inspection report contain less specificity than its counterpart. Within critical risk factors, inspectors further delineate between major and minor infractions. Good retail factors are defined as behaviors that help reduce foodborne illness, such as washing fruits and vegetables and maintaining functioning facilities. These behaviors appear to conflate New York’s critical and general violations into one category and result in smaller point deductions than New York’s critical risk factors.
Although Los Angeles’s food retail safety system is based on the California Food Code, its regulatory model is more akin to a market-competition paradigm rather than to a command-and-control paradigm. Outside of an inspector’s ability to close down an unhygienic restaurant, the Department of Public Health does not impose fines or other penal measures for offending establishments. Instead, Los Angeles relies primarily on the public disclosure of inspection letter grades to dictate consumer choice and spur changes in restaurant behavior. This has permitted some restaurants to remain open even after being cited for egregious health violations. The city has attempted to adjust for the regulatory failure of the market-competition model in recent years by requiring additional violation points for compounding health violations.
Although the food safety requirements of New York City and Los Angeles do overlap, the nation’s two largest cities also differ in their regulations in many ways. Most starkly, Los Angeles depends on the behaviors of its consumers to incentivize restaurants to clean up their act, whereas New York City uses a heavier hand to ensure restaurants comply with local food safety standards. Even within their standards, the two cities differ in assigning violation severity for similar offenses and grading cutoff scores for public disclosure.
Perhaps the most recent innovation of food retail regulatory models is Seattle-King County’s emoji safety system. The County’s food inspectors cover more than eleven thousand restaurants per year, providing at least two graded inspections and one ungraded educational visit. Inspectors rate their assessment based on critical (red) and non-critical (blue) violations. Critical violations are defined as food-handling practices whose violation may lead directly to foodborne illness. These practices include proper handwashing, cooking food at appropriate temperatures, and maintaining clean surfaces. Inspectors also evaluate low-risk factors, which are not incorporated into a restaurant’s rating. These include behaviors such as proper thawing methods and disposal of garbage. While Seattle’s critical violations seem roughly in line with the behaviors noted in New York and Los Angeles inspections, the city differs by excluding low-risk violations from inspection scores.
At the conclusion of each visit, inspected restaurants receive an emoji that symbolizes a rating of “Excellent,” “Good,” “Okay,” or “Needs to Improve.” Restaurants that are rated “Excellent” display signage featuring a beaming smiley face and a description noting that the restaurant has had few or no critical violations over the course of its last four inspections. A “Good” rating, a regular smiley, denotes some critical violations in the same period. A restaurant that receives an “Okay” rating features a face with a small, begrudging smile, signifying the restaurant has received many critical violations over the last four inspections. A restaurant that “Needs to Improve” either has been shut down or has required multiple inspections to remedy its food safety practices.
In addition to its simple disclosure method, Seattle’s rating system is unique in incorporating food safety over time rather than basing food safety solely on the last inspection date. A restaurant’s rating is calculated by averaging the total points of critical violations over the restaurant’s four most recent routine inspections. Since restaurants are typically inspected between one and three times per year, the rating generally reflects a contemporaneous view of the establishment’s practices. Moreover, ratings are assigned on a curve to account for inspector variation across Seattle-King County. Approximately half of restaurants in a zip code will receive “Excellent” ratings, 40 percent “Good,” and 10 percent “Okay.” A “Needs to Improve” restaurant is exempt from the curve because of the additional remedial actions required to cure its violations. Given the curve, there is no uniform cutoff between an “Excellent” rating and a “Good” rating, meaning two identically performing restaurants could receive two different ratings simply by virtue of location. A fines schedule associated with health violations is not immediately available on the Seattle Department of Health website, indicating the primary penalty for a poor inspection is a disappointed smiley face.
Although Seattle still bears signs of a command-and-control regulatory model, its system also leans heavily toward market-competition regulation. The County spent over two years developing its inspection and rating system, soliciting feedback from municipal government agents, restaurant owners, and academic experts. The ratings are designed to be easily understood by consumers, who can base their dining choices on a relatively small slice of data. Furthermore, the curve distribution promotes market competition, as restaurants will be naturally incentivized to edge out their competitors. The County’s reliance on a broad range of stakeholders has created an inspection system that has addressed certain concerns identified in other cities’ systems by relying on a cumulative view rather than a point-in-time inspection, staying away from harsh consumer-facing letter grades, and minimizing grade skew. Despite these benefits, many restaurant owners contest the curve, arguing that it adds more, not less, subjectivity to the rating system.
Seattle represents a novel, contemporary approach to food safety. The city has opted for simplicity in some areas, such as the emoji rating, but has also created some areas for confusion, such as the rating system curve. Given its emphasis on public disclosure, the city’s food retail safety system is more similar to that of Los Angeles than that of New York City.
Our test-case cities reveal only a few of the possible permutations that emerge out of advisory food safety guidelines. Although each system has successful elements, common issues also arise from the lack of national standards. The FDA’s half-hearted attempt at a command-and-control model has given rise to market competition in several municipalities, leading to inconsistent regulatory enforcement and adoption. This Section discusses several fundamental challenges the status quo does not adequately address. First, the lack of uniformity frustrates the industry’s overall goal of reducing the incidence of foodborne illness and creates unnecessary inconsistency. Next, the adoptive requirements set out by the FDA result in a high degree of noise and arbitrariness in public disclosure, leading to consumer confusion. Finally, the FDA’s current regulatory structure fails to account for cultural nuances in ethnic cuisine, allowing implicit bias to creep into inspection results and consumer opinion.
A. Limited Correlation to Foodborne Illness
According to the CDC’s most recent longitudinal study, restaurants remain by far the most prominent source for foodborne illness in the United States. The number of foodborne illness outbreaks remained steady over the CDC’s study period from 2009 to 2015. Furthermore, the outbreaks with the largest number of cases reported and the most severe outcomes (i.e., the highest number of hospitalizations and deaths) involved exposures across multiple states. The continuing threat of foodborne illness in the food retail industry illustrates the need for the regulatory system to adequately address the prevalence of foodborne illness outbreaks. This can be challenging because of the difficulty in accurately identifying a specific occurrence of foodborne illness. Incubation periods for food pathogens can last up to three or four days, leading consumers to commonly misattribute the cause of illness to their most recent meal. Nonetheless, a well-defined federal food retail safety paradigm could diminish the threat of severe outbreaks better than the localized efforts of state and municipal agencies.
Given the origin issue, studies examining links between inspection regimes and decreased occurrences of foodborne illness reveal mixed results. Most studies have focused on the implications of public disclosure for consumer choice and restaurant behavior. A widely cited study examining the impact of grade disclosure found a 20 percent decrease in foodborne-illness-related hospitalizations following the introduction of restaurant inspection grades in Los Angeles. Although this decrease tracked with a general national decline in foodborne illness during the same period, researchers postulated that heightened public scrutiny around hygiene and inspection results could have motivated restaurants to adopt more sanitary food practices. Market-driven motivation to beat out competitors or to attract customers could be an argument against federal regulation, but subsequent studies have not been as quick to draw a direct correlation.
Relative to the efficacy of restaurant grades in Los Angeles, the efficacy of restaurant grades in New York City has not been as well established. Although a DOHMH report observed improvements in restaurant sanitary conditions in the two years after the start of mandatory grade disclosure, another study found that instances of food poisoning remained stable during the same period. A subsequent study examining calls to 311, a municipal comment box for health concerns, revealed no significant drop in complaints of foodborne illness during the initial years following public disclosure. Moreover, researchers have been consistently unable to establish a definitive correlation between public disclosure in New York City and norovirus, one of the most prominent foodborne pathogens. The conflicting results between New York City and Los Angeles seem perplexing, given both cities implement similar grading and posting requirements. If anything, New York City’s extensive grading rubric provides more opportunities for an inspector to notice a food safety violation. The contrasting data highlight the need to find synergies across municipalities, which can be challenging without a centralizing governing body.
Finally, although there have not yet been studies evaluating the efficacy of the new Seattle system, additional studies outside of this Note’s test-case cities have generally been unable to establish any linkage between restaurant inspection and foodborne illness. The cities in these additional studies all adhere to some version of the Model Food Code, indicating that simply following the recommended guidelines is insufficient to consistently decrease illness outbreaks. Although convoluting factors such as population size or lack of public disclosure might have also contributed to these findings, it does not appear that the FDA’s current adoptive approach is effective in reducing the occurrence of foodborne illness across the nation. Allowing local agencies to cherry-pick food retail practices frustrates the FDA’s predominant goal of ensuring the safety and purity of foods in interstate commerce.
An examination of this Note’s test-case cities and additional studies demonstrates that local food safety standards have not been strongly linked to decreases in foodborne illness. Because this is the primary goal of any food safety regulatory model, it is alarming that local agencies continue in their status quo. Even as an advisory agency, the FDA must look to the effects of implementation and work to better understand why food retail safety standards have had limited success.
B. Inconsistent Implementation Paradigms
The noncompulsory nature of the Model Food Code has created inconsistencies in local implementation that are both unnecessary and wasteful. The sixty-six state agencies that regulate food retail across the country vary broadly in their adoption of FDA guidelines. A number of states adhere to outdated versions of the Model Food Code for no other purpose than to avoid the administrative burdens associated with updating their practices every four years. From a local agency’s perspective, there may be high transaction costs associated with constantly keeping up with FDA requirements. Some agencies, including the California Department of Public Health and New York State Department of Health, have declined to adopt the Model Food Code altogether.
On the other hand, there are certain health risks that are universal. The test-case cities in this Note focus on similar behaviors that pose a high risk for spreading foodborne illness: improper holding or cooking temperatures, unsanitary employee behaviors, etc. But whether these same behaviors trigger a two-point violation or a twenty-eight-point violation is a question of interpretation. Given there is agreement on many general standards for food retail safety begs for a more efficient approach to monitor and correct risk. Permitting state and local agencies to tailor these fundamental safety requirements wastes time and tax dollars. This results in a lack of uniformity at the macro level, at which jurisdictions structure their idiosyncratic food retail safety systems, and the micro level, at which individual subjectivity from food inspectors further exacerbates differences in enforcement.
At the macro level, the implementation of specific local standards creates unnecessary inconsistency across jurisdictions. For instance, although New York City and Los Angeles both use letter grades for public disclosure, inspectors’ grading methods differ substantially. New York City’s system tallies up the total number of violation points, whereas Los Angeles deducts from a perfect score of one hundred. Critical risk factor violations in Los Angeles receive two- and four-point deductions for minor and major violations respectively. By comparison, critical violations in New York City range from two to twenty-eight points, depending on severity. This means a restaurant found in violation of improper cooking time would have four points deducted in Los Angeles but would accrue anywhere from ten to twenty-eight points in New York. Yet, the cutoff between an “A” and “B” grade in Los Angeles and New York are approximately the same, at ten and thirteen points respectively. This nominal difference does not account for the incremental violation points that a New York restaurant can incur during inspection. Consequently, for two cities that employ the same letter grades, a “B” or a “C” rating can indicate very different levels of risk. At a minimum, federal requirements should reconcile the discrepancies and promote consistent messaging across jurisdictions.
The lack of consistency in food retail system implementation is made worse by the inevitable variability found at the micro level. Individual food inspectors exercise considerable discretion in reporting health violations in a restaurant and ensuring public health and safety. Despite this, the Model Food Code merely requires that a food inspector have the “knowledge, skills and ability to adequately perform the required duties,” requiring no minimum education or certification qualifications. This leads to jurisdictions with vastly different requirements. For example, compare Los Angeles’s minimum requirement, a certification as an Environmental Health Specialist, to New York City’s, a science background at the college level in addition to specialized inspector training.
Considerable variation within a single jurisdiction also calls for more trickle-down consistency. A 2009 audit of the New York City food retail safety system revealed that while the average inspection score was around twenty-five points, some inspectors recorded average scores of fifteen points while others gave average scores of fifty points. The relative complexity of the New York City inspection criteria could account in part for such disparate variability, although the wide range of scores is still cause for concern. For example, two perfectly reasonable inspectors may disagree over whether food is not in “good condition” or how to assess “improperly constructed surfaces,” since both terms as described in the New York Food Code are ambiguously defined. This may lead a consumer to question whether an inspection score is truly indicative of a restaurant’s food safety practices. Given the significant noise in the system, regulatory agencies should work to align standards and reduce discrepancies where possible to account for inevitable human subjectivity during implementation. Ultimately, a uniform national regulatory system would lead to more consistent implementation paradigms at both a macro and micro level.
C. Ratings Gamesmanship and Consumer Confusion
As more developed food retail markets move towards some form of public disclosure, consistent messaging about inspection results can educate consumers and allow them to make more informed choices. But the lack of federal regulation has resulted in ineffective information disclosure to consumers, leading consumers to question the legitimacy of publicly available information about restaurants.
One of the primary concerns with current public disclosure requirements is the gamesmanship involved in achieving the “optimal score.” After the introduction of Los Angeles letter grades, the distribution of restaurant scores changed from a standard bell curve to a sudden spike of “A” grades. The only variable that had changed was the public disclosure requirement, since the actual food retail safety standards remained the same. It is possible that the grade posting requirement actually motivated restaurant owners to pay closer attention to food retail safety standards and improve sanitation. However, if that were the case, it would seem more likely for the bell curve to shift further to the right, rather than create a concentration of scores just at the cutoff for the highest grade. Alternatively, perhaps restaurant owners invested only the minimum effort required to achieve an “A” but saw little value in correcting minor violations that would not threaten their “A” grade. Then again, the spike could also be attributed to inspectors who ignored a minor violation or who no longer felt the need to delineate between restaurants that scored in the low to mid-nineties numerical range.
Similarly, after New York City’s introduction of grade disclosure, the distribution of restaurants at the cutoff between “A” and “B” grades shifted notably. Prior to grading, roughly the same number of restaurants received thirteen versus fourteen violation points during an inspection. However, after grading was introduced, the number of restaurants that received thirteen points increased to more than double the number that received fourteen points. The proportion of “A” restaurants in New York City has ballooned to over 90 percent. A higher proportion of “A” grades should indicate a higher level of food retail safety, but the data showing relatively steady reports of foodborne illness refute this claim. Instead, the grade skew seems to be more a product of individual malleability rather than lasting corrective behavior. A recent example of the Michelin restaurant Per Se demonstrates the manipulation of the grading system. After receiving a whopping forty-two points during a restaurant inspection, Per Se appealed to an administrative judge. Despite the egregious number of reported offenses, the judge merely waived the bulk of the violation points and allowed the restaurant to continue operating.
In New York City, the focus on achieving the “A” grade has led to an unfortunate shift in resource allocation. Food retail establishments can request a second inspection if they do not achieve an “A” on their first try, which has led to numerous appeals from establishments on the cusp. As such, inspectors are unable to spend as much time monitoring the worst-scoring restaurants. Before mandatory public disclosure, inspectors spent up to 15 percent of their time covering “C” grade restaurants. After restaurants began posting grades, inspector time dedicated to “C” restaurants dropped to less than 5 percent. The public disclosure campaign inadvertently shifted its focus from protecting and educating consumers to catering to restaurant owners, adding to the inundation of “A” grades across the city.
The Seattle Department of Public Health has made concerted efforts to address the issue of grade skew by using a curve system. As a result, roughly half of the restaurants within a zip code will receive an “Excellent” rating, regardless of their numerical scores. The city reasoned that since most consumers dine locally, they would benefit the most by receiving relational ratings. Furthermore, since inspectors typically cover a specific geographic area, the curve would help smooth out variability between inspectors.
Of our test-case cities, Seattle comes the closest to addressing the major concerns of public disclosure, but it too is not without its faults. Its ratings of “Excellent,” “Good,” or “Okay” essentially track with the three letter grades used by New York City and Los Angeles. But total reliance on relational subjectivity can mean an “Excellent” rating in one area signifies something completely different when compared to another part of the city. The fundamental challenge with a pure curve system is its assumption that, on average, any given part of the county is representative of the whole. However, differences in economic development across the county may result in areas that contain establishments of overall poorer quality. Based on Seattle’s curve system, it would be difficult to discern whether an “Excellent” restaurant is objectively excellent or simply better than its surrounding peers.
Seattle’s system demonstrates that consumer education can be distilled into a simpler message, but should clearly tie its messaging to empirically proven findings regarding foodborne pathogens and risky food handling practices. Although it is effective up to a certain point, market-competition regulation must be reined in to provide an accurate depiction of food retail safety.
While public disclosure of restaurant inspection results is important for consumer education and decision-making, it can also create adverse incentives to artificially inflate the consumer-facing rating. When combined with the political gamesmanship—such as the example of Per Se, or the relative distribution of smiley emojis in Seattle—restaurant ratings may make customers skeptical of the legitimacy of such grades in communicating an accurate message around food safety. Centralized oversight would prevent local agencies from getting stuck in the weeds and losing track of the ultimate objective of its food retail safety program. Furthermore, a clear, transparent set of standards would minimize consumer confusion and bring credibility to public disclosure.
D. Implicit Bias Among Food Inspectors
The emergence of market-competition regulation out of a failed command-and-control model has also negatively impacted diverse restaurateurs, particularly those who focus on ethnic cuisine. There are two primary issues: First, disparate regulatory agencies have historically been slow to incorporate feedback from a diverse range of restaurant owners, including those that serve ethnic cuisine. Second, the discretion left to inspectors creates an opportunity for implicit bias and preconceived notions about ethnic cuisine to creep into inspection results.
The definition of an “ethnic” restaurant can be ambiguous, just as the definition of American cuisine can evade clearly-defined boundaries. Nonetheless, even though French and Italian cuisine are arguably as foreign to Americans as Asian and Latin-American food, the latter two are more likely to be considered “ethnic cuisines.” The ethnic restaurant is often associated with the concept of authenticity, the notion that an establishment is genuine only if it adheres to certain cultural expectations and customs. Coupled with the elevated demands of authenticity is the suspicion that ethnic restaurants are also less sanitary than an all-American establishment. Conceptually, authenticity is an inherently subjective standard that reflects the opinions of the majority. This raises issues when contextualizing food retail safety standards, which tend to reflect a Westernized view of food preparation and handling. As a result, the imposition of food retail safety standards absent consideration of cultural nuances can adversely impact non-European cuisines.
Peking roast duck is illustrative of the cultural battle between ethnic restaurants and regulatory agencies. Historically, roast ducks are common in Chinese cuisine and restaurants often hang the ducks by restaurant windows to attract customers. Even though the ducks were typically held in the cooking area away from customers, health officials in New York and Los Angeles independently underwent periods of repeatedly fining Chinese establishments for violations of food holding temperature regulations. Restaurant owners fought back, arguing that high temperatures dried out the meat and detracted from the authenticity of the dish. After years of back and forth, Chinese restaurant owners eventually prevailed in both cities. In California, the then-governor Jerry Brown signed legislation that exempted Peking ducks from state holding temperature requirements. In New York City, the DOHMH revised guidelines to permit restaurants to hang ducks for up to a certain period of time or to take additional precautionary measures.
The reactionary model of amending local food codes only in response public outrage prevents regulatory authorities from properly addressing critical cultural considerations. Combined with inspector discretion, this creates the potential for implicit bias when inspectors visit restaurants in which they are unfamiliar with food preparation customs or if they experience communication challenges with the owner. Due to these perceptions, ethnic restaurants are at a heightened risk of being cited more frequently for heath code violations.
First, ethnic restaurants typically receive more health inspections than non-ethnic restaurants. It is unclear how much of this dynamic creates a self-fulfilling prophecy where unsanitary restaurants tend to get flagged for more frequent visits, which in turn creates more opportunities for inspectors to notice violations. A comparative study across five major cities in the United States confirmed that ethnic restaurants, on balance, received significantly more major health code violations than non-ethnic restaurants. Similarly, a 2018 study found that Asian restaurants in New York City performed 13.4 percent worse in letter grades year over year, driven by an increase in the average violation score from 19.8 to 21.7 points.
Existing research points to other convoluting factors that contribute to an increased number of health violations in ethnic restaurants. In addition to different food preparation and handling customs, owners of ethnic restaurants are more likely to be recent immigrants or first-generation Americans. Limited language skills and cultural customs may diverge from American norms, making owners of ethnic restaurants easy targets for health citations. In addition, municipal food codes may be offered only in English and contain dense and incomprehensible guidelines. Finally, a lack of training and clear understanding of regulatory norms are common barriers that may contribute to a disconnect between owners of ethnic restaurants and food inspectors.
The market-competition regulation prevalent in many jurisdictions also puts ethnic restaurants at a disadvantage compared to non-ethnic restaurants. Because many agencies do not have the resources to publish their guidelines in languages other than English, it is the owners of ethnic restaurants who suffer the loss. Once branded with a poor inspection score, owners struggle to retain customers and correct their public perception. Furthermore, reliance on a single cultural perspective and potentially outdated food retail guidance only exacerbates the impact on minority communities.
Owners of ethnic restaurants have the unique challenge of preserving the “authenticity” of their cuisine while conforming to the food retail safety standards that may conflict with their cultural practices. Without a uniform set of standards that account for cultural differences, they will continue to be disproportionately impacted by the idiosyncrasies of local regulatory systems. Centralized oversight will help ensure these restaurant owners are appropriately protected from human error during regular food safety inspections.
E. Implicit Bias Among Consumers
Similar to the bias concerns found with inspectors discussed earlier, incorporating consumer input into the food retail inspection system adds another layer of implicit bias into the regulatory oversight of food retail establishments. Due to resource constraints, local jurisdictions have increasingly turned to crowdsourced data to address the scalability challenge of active restaurant monitoring. Inspection authorities now frequently use aggregate Yelp reviews or 311 calls for recurring reports of foodborne illness to identify potential high-risk establishments and flag restaurants for inspection.
At first glance, the incorporation of consumer feedback seems to address some scalability issues by providing sufficient data points to create a more representative picture. One study examining the correlation between Yelp reviews and the incidence of inspection violations found that a restaurant’s average star rating was negatively correlated with a restaurant’s health code rating. This meant that for each additional star added to a restaurant’s average rating, the risk of a poor health code rating dropped by approximately 36 percent. In other words, the higher a restaurant was rated by consumers, the less likely it was to receive a poor health inspection score by inspectors.
On the other hand, the incorporation of consumer input assumes the reported data accurately represents actual conditions. But like food inspectors, diners are susceptible to their own perceptions of cleanliness. First, consumers’ perceptions of restaurants do not always correlate with inspection scores, suggesting that a highly rated Yelp restaurant could still have poor sanitation practices. Second, research shows that consumers are more likely to leave a review when they have a negative experience as opposed to a positive one, creating opinion skew. And finally, some ethnic minority groups are less likely to make consumer-related complaints, raising questions of representative sampling. All of these factors suggest that restaurant reviews are not always indicative of a restaurant’s actual food safety practices.
Several other factors may contribute to a consumer’s limited ability to discern the sanitary standards of a food retail establishment. The consumer dining experience is one of many components that shape a restaurant’s operations, albeit an important one. As a threshold matter, many instances of foodborne illness go unreported. Furthermore, those who do report often have trouble attributing their sickness to the correct establishment. On the flip side, the mere absence of reported foodborne illness does not indicate an absence of risky food safety practices. Most food safety violations occur in the back of the house, where consumers are unable to observe. Finally, the lack of consumer training leads to the risk of false positives, ultimately resulting in more transaction costs on the part of regulatory agencies.
The use of consumer big data can also cast unfavorable light on ethnic restaurants. Food retail regulators have regularly employed the Yelp terms “Vietnamese,” “Thai,” “Japanese,” and “Chinese” as specific predictors for establishments that may have poor food safety practices. A study of food complaints in New York City and Seattle revealed that complaints and reviews flagging food-safety issues were disproportionately made against Asian food restaurants. Importantly, even after controlling for actual health inspection scores, researchers found that Asian establishments in New York City were 42 percent more likely to receive food poisoning complaints compared to non-Asian establishments. Similarly, Seattle Asian establishments also emerged as targets for Yelp terms such as “vomit,” “sick,” and “diarrhea” even after controlling for inspection scores. This suggests that ethnic restaurants are targets for more negatively rated consumer reviews, regardless of their actual inspection score or cleanliness. The skew of negative consumer reviews in turn influences how regulatory agencies refine their inspection strategy, putting ethnic restaurants in the hot seat.
Looking at existing data, incorporating consumer input into food safety governance has created more problems than it solves. Studies show that consumer feedback introduces more bias into an already flawed regulatory system. Regulators must recognize the need for a proactive approach to minimize bias from inspectors and customers alike, and that disparate systems as they exist today will not permit this change to happen.
The FDA’s current approach to food retail safety has required significant investment on the part of the federal government without the benefit of true adoption and enforcement. Despite the Agency’s efforts to publish a new Model Food Code every four years, a state or local agency’s full discretion to design their own food retail safety system defeats the purpose of a federal agency. Without more federal regulation, the food retail industry is unlikely to effectively address the ongoing threat of foodborne illness. The resulting inconsistency and consumer confusion only make it harder for regulatory agencies to accomplish their goal of reducing the occurrence of foodborne illness among consumers.
Based on the existing issues with national food retail safety regulation, the federal government has several options in reshaping its regulatory framework. The default option is to continue as is and allow the food retail system to function as a semi-regulated industry. An effect of the status quo is the downward slide of cities from semi-regulation to free market, driving restaurants to adjust their behaviors to promote the greatest amount of business rather than industry-established safety practices. Although there is an argument that unsanitary restaurants are unlikely to succeed in the long run, lack of guidance in the food retail industry is at odds with the federal government’s oversight of many other aspects of the food industry. The prevailing incidents of foodborne illness suggests that food retail establishments need more regulation, not less. Furthermore, instances of unconscious bias may disproportionately affect restaurants serving ethnic cuisine, creating an uneven playing field in the free market.
The second option is to adopt a command-and-control model of regulation, reinstating the government as the authority to determine the appropriate food safety standards on a national level. The federal government already has much of what would be required for a command-and-control system. A central regulatory agency exists and already expends considerable effort in establishing guidelines for food safety. The incremental step to establish a command-and-control system would be to mandate adoption and implement an enforcement strategy.
Finally, an outcomes-based regulation model would enable restaurants to satisfy health and safety objectives in a manner of their choosing. This would effectively eliminate state and local agencies as the middlemen, leaving implementation to individual parties. Although an outcomes-based approach could address some concerns of cultural differences in ethnic restaurants, the number of food retail establishments across the nation would likely make outcomes monitoring unwieldy. The food retail industry is not the ideal environment for an outcomes-based model, which relies on a few players in the industry who are well-versed in their business. The sheer volume of individual parties would require some central oversight to ensure food retail establishments were all meeting their safety goals.
Due to the minimal efforts necessary to shift to a true command-and-control model, I recommend that the FDA take a stronger position on the requirements for food retail establishments and advocate for greater federal regulation. Given the number of individual restauranteurs and establishments in the food industry, a command-and-control model may fare better than outcomes-based regulation, and certainly better than the existing market-competition model. As an initial milestone, implementing mandatory food safety guidelines would be one step closer to communicating the latest relevant information to inspectors, restauranteurs, and consumers. State and local agencies could still oversee the execution of such guidelines, but under closer guidance by the FDA. As agencies would no longer be free to tailor their safety standards, federalization would help facilitate a consistent approach to food retail safety across the country.
Considering the legislative trend towards allocating more responsibility to the FDA, expanding the Agency’s powers to cover more of the food supply chain does not seem an outsized goal. The FDA already invests considerable effort into publishing a Model Food Code once every four years and issues updates on a biannual basis. Furthermore, a number of states currently follow some version of the Model Food Code, suggesting complete federalization would require only incremental changes for state and local agencies. Given the FDA’s efforts to keep the Model Food Code updated with the latest research, there is little reason for states to continue to adhere to outdated federal guidelines. Adoption of national standards would cement the baseline command-and-control regime for food retail safety, rather than leaving guidance and implementation entirely up to local agencies.
The primary criticism of a stronger command-and-control system is the lack of resourcing available to execute a federal plan. Although the concept of federal standards for food retail establishments has limited policy-based downsides, the implementation barriers are well-established and challenging to resolve. The 1975 GAO report estimated that direct federal regulation of the food retail industry would cost $156 million annually. In today’s value, that cost would range approximately between $730 million to $760 million. Although the FDA already oversees a significant remit with its existing budget, the incremental cost may be worth the benefit to American consumers writ large.
Despite existing resource constraints, the FDA can begin by making incremental changes, such as mandating Model Food Code standards for all regulatory agencies, before adopting a widescale implementation model. It will take time for state and local agencies to adopt such standards if they have not previously, and agencies that already comply should continue to do so. Like the Supplemental Nutrition Assistance Program, which leaves implementation to state and local authorities, the FDA should provide oversight and conduct occasional inspections to ensure widespread compliance in the initial stages of federalization. Once federal standards have been fully adopted, the FDA can explore additional avenues to expand its ability to conduct the actual inspections, if needed. Complete federalization is still in the distant future, but as with the expansion of FDA oversight in other areas, an incremental approach in the food retail industry seems most prudent.
Another criticism of the command-and-control model is the overreaching paternalism of the federal government. The legislative history of the FDA reveals substantial pushback from states at the notion of federal food retail regulation. However, as illustrated by the CDC’s report on foodborne illness, outbreaks are increasingly crossing state lines. A federal system for food retail safety would be better equipped to handle these interstate outbreaks. The current proposal for command-and-control would not exclude states from providing their input, but rather would incorporate their concerns into the federal model.
In an effort to streamline food retail safety standards, the FDA will need to reconcile its current Model Food Code with some of the more paradigmatic regulatory regimes in the United States. A command-and-control regulatory system would require the FDA to clearly define a standard set of inspection requirements and corresponding pecuniary consequences to structure its regime and resolve discrepancies between jurisdictions. Furthermore, mandating a set of identifiable health risks would provide a minimum standard of quality for restaurants and eliminate duplication of offenses across jurisdictions. For example, to provide for some local flexibility, the Agency could consider adopting a severity scale for violations, like the New York model. Conversely, individual inspector discretion would need to be tightly cabined within the parameters of the food retail safety standards. Mandatory uniform safety guidelines would give weight to widespread adoption and compliance.
Uniform food retail safety standards should also feature two important components discussed in this Note: public disclosure and cross-cultural integration. First, given the trend of cities moving towards more public disclosure, the FDA should consider returning to its previous position and mandating public disclosure of health inspection results. The notion that cities should disclose the results of health inspections is prevalent in some of America’s largest cities. By contrast, other jurisdictions only feature outdated online repositories of restaurant inspection results, leaving it to the consumer to seek out public health information. Explicit public disclosure requirements would reduce the information asymmetry between food retail establishments and consumers and incentivize proactive behavior from both parties. Whether the disclosure takes the form of letter grades or smiley faces is immaterial. What matters is that it is clear public disclosure.
Federal standards for food retail safety must also account for changes in the American demographic and incorporate cultural considerations. An obvious step is to ensure that all food safety guidelines be made available in languages other than English. Federal efforts to standardize training resources in this respect will create further efficiencies across state and local jurisdictions. Next, the FDA should partner with restaurant coalitions to identify cuisines that may cultivate different food preparation practices and strive to reach a middle ground in which ethnic restaurants can comply with food safety standards without losing authenticity. This collaboration would be in line with the Agency’s move towards more proactive regulation and promote buy-in from restaurants as well. Finally, the Agency should include sensitivity training in training qualifications for food inspectors. Inspectors with comprehensive knowledge of a wide range of cuisines and food practices can better partner with local restaurant owners to evaluate the safety of their establishments.
The food retail industry moves quickly. Thousands of restaurants open and close across the United States each year. As dining out becomes ubiquitous in American life, it is imperative that those with the most knowledge can address prevalent public health risks. Given the existing efforts expended by federal, state, and local counterparts, it is evident that reducing the occurrence of foodborne illness remains a top priority. It is time for the FDA to establish a definitive stance on food retail regulation and adopt mandatory safety inspection and disclosure requirements as a federal standard. A partnership with state and local agencies as well as restaurant owners will allow the FDA to foster a safer environment for American consumers and adapt to the evolving food retail industry.
Ping Liu: J.D. 2020, University of California, Berkeley School of Law. I would like to thank Professor Sugarman for inspiring this work, and my classmates for their feedback and encouragement. Many thanks to the editors of the California Law Review for guiding me through the publishing process and for the countless hours they contributed to this Note.
. See CDC Estimates of Foodborne Illness in the United States, Ctrs. for Disease Control & Prevention (Nov. 5, 2018), https://www.cdc.gov/foodborneburden/2011-foodborne-estimates.html [https://perma.cc/5VMR-PW5F].
. See L. Hannah Gould et al., Surveillance for Foodborne Disease Outbreaks—United States, 1998–2008, Ctrs. for Disease Control & Prevention Morbidity & Mortality Wkly. Rep., June 28, 2013, at 1, 5 (tracking foodborne illness from 1998 to 2008).
. See Richard A. Merrill & Jeffrey K. Francer, Organizing Federal Food Safety Regulation, 31 Seton Hall L. Rev. 61, 82–84, 92 (2000) (noting that due to largely political decisions dating back to the 1940s, the United States Department of Agriculture (USDA) oversees the safety of meat, poultry, and egg products).
. See Background on the FDA Food Safety Modernization Act (FSMA), U.S. Food & Drug Admin., https://www.fda.gov/Food/GuidanceRegulation/FSMA/ucm239907.htm [https://perma.cc/7NK8-KZSN] [hereinafter FSMA Legislation].
. See Model Food Code, supra note 5, at iii (outlining historical timeline of the FDA as an advisory body); see also Food Service Sanitation, 42 Fed. Reg. 15,428 (Jan. 27, 1977) (to be codified at 21 C.F.R. pt. 940) (withdrawing proposal for federal standards due to state pushback).
. See Food Service Sanitation, 42 Fed. Reg. at 15,428 (“The primary objections to the proposal were that it abridged a long-term understanding between the States and the Federal government regarding the regulation of the food service industry, and that the proposal inadequately addressed the special circumstances involved in the operation of interstate conveyances and their sources of food.”).
. U.S. Food & Drug Admin., Adoption of the FDA Food Code by State and Territorial Agencies Responsible for the Oversight of Restaurants and Retail Food Stores 5–6, 8 (2017) [hereinafter Adoption Report].
. Priya Krishna, New York Restaurant Inspections, By the Numbers, N.Y. Times (June 5, 2018), https://www.nytimes.com/2018/06/05/dining/new-york-restaurant-inspections-by-the-numbers.html [https://perma.cc/3QXV-B4TS].
. See Priya Krishna, The Life of a Restaurant Inspector: Rising Grades, Fainting Owners, N.Y. Times (June 5, 2018), https://www.nytimes.com/2018/06/05/dining/restaurant-health-inspector.html [https://perma.cc/9QF3-M62F].
. Marcia Murphy, Office of the N.Y.C. Comptroller, New York City Budget Brief: New York City Fine Revenues Update (2017), https://comptroller.nyc.gov/wp-content/uploads/documents/New-York-City-Fine-Revenues-Update.pdf [https://perma.cc/6DHA-CAUN] (reporting $30 million in restaurant fines in 2016, a leading source of revenue for the DOHMH).
. Glenn Collins, Even the A Students Sometimes Break Health Rules, N.Y. Times (Feb. 28, 2012), https://www.nytimes.com/2012/02/29/dining/new-york-city-restaurants-skirt-inspections-finer-points.html?pagewanted=1&_r=2 [https://perma.cc/7STT-J4MW] (noting also that sushi chefs take care to wash their hands approximately forty times over the course of a typical dinner service).
. Compare NYC Inspection Worksheet, supra note 33, with L.A. Inspection Report, supra note 45 (noting Los Angeles’s requirement of “proper hot and cold holding temperatures” and New York’s prohibition on a “[h]ot food item that has been cooked and refrigerated is being held for service without first being reheated to 165°F or above within 2 hours”).
. Stephanie K. Baer, What That Restaurant Letter Grade Isn’t Telling You About Health and Cleanliness, San Gabriel Valley Trib. (June 25, 2015), https://www.sgvtribune.com/2015/06/25/what-that-restaurant-letter-grade-isnt-telling-you-about-health-and-cleanliness [https://perma.cc/F82X-AV6Y] (exposing a restaurant that was permitted to remain open after inspectors found a lack of running water in the establishment).
. Stephanie K. Baer, That ‘A’ Grade at Your Favorite LA Restaurant Will Be More Meaningful Soon. Here’s Why, San Gabriel Valley Trib. (June 10, 2016), https://www.sgvtribune.com/2016/06/10/that-a-grade-at-your-favorite-la-restaurant-will-be-more-meaningful-soon-heres-why [https://perma.cc/6WPY-BXXK].
. Carley Thomson, With the New Year Comes a New Way to Rate Food Safety: A Q&A with Becky Elias and Damarys Espinoza, Pub. Health Insider (Dec. 19, 2016), https://publichealthinsider.com/2016/12/19/with-the-new-year-comes-a-new-way-to-rate-food-safety-a-qa-with-becky-elias-and-damarys-espinoza [https://perma.cc/Z3XN-HLGJ].
. Jesus Hidalgo, How Good is “Good”? Digging into Those Food Safety Emojis, Seattle Wkly. (Feb. 21, 2017), https://www.seattleweekly.com/news/how-good-is-good-digging-into-those-food-safety-emojis/ [https://perma.cc/YVS5-G5DS].
. Daniel Dewey-Mattia et al., Surveillance for Foodborne Disease Outbreaks — United States, 2009–2015, Ctrs. for Disease Control & Prevention Morbidity & Mortality Wkly. Rep., July 27, 2018, at 1, 3 [hereinafter CDC Report].
. See Kristen M. Altenburger & Daniel E Ho, When Algorithms Import Private Bias into Public Enforcement: The Promise and Limitations of Statistical Debiasing Solutions, 175 J. Institutional & Theoretical Econ. 98 (2018).
. See, e.g., Thomas Farley, N.Y.C. Dep’t of Health & Mental Hygiene, Restaurant Letter Grading: The First Year (2011); Katie Filion & Douglas A. Powell, The Use of Restaurant Inspection Disclosure Systems as a Means of Communicating Food Safety Information, 20 J. Foodservice 287 (2009); Petrona Lee & Craig W. Hedberg, Understanding the Relationships Between Inspection Results and Risk of Foodborne Illness in Restaurants, 13 Foodborne Pathogens & Disease 582 (2016); Wong et al., supra note 1.
. See, e.g., Lee & Hedberg, supra note 65 (finding food grades in Illinois were not significantly correlated with foodborne illness outbreaks); Filion & Powell, supra note 65, at 287–97 (finding that mandatory disclosure did not necessarily incentivize consumers to change their behavior given confusion around inspection criteria).
. See id.; Miguel A. Cruz et al., An Assessment of the Ability of Routine Restaurant Inspections to Predict Food-Borne Outbreaks in Miami–Dade County, Florida, 91 Am. J. Pub. Health 821, 822 (2001); Timothy F. Jones et al., Restaurant Inspection Scores and Foodborne Disease, 10 Emerging Infectious Diseases 688, 690 (2004); Valerie A. Yeager et al., Relationship Between Food Safety and Critical Violations on Restaurant Inspections: An Empirical Investigation of Bacterial Pathogen Content, 75 J. Envtl. Health 68 (2013).
. Compare Environmental Health Specialist II Job Listing, Cty. of L.A. Pub. Health, https://www.governmentjobs.com/careers/lacounty/jobs/1240253/environmental-health-specialist-ii [https://perma.cc/2K8Q-VPG5], with Krishna, supra note 36.
. William C. Thompson, Jr., Office of the Comptroller, City of N.Y., Audit Report on the Department of Health and Mental Hygiene Oversight of the Correction of Health Code Violations at Restaurants 13 (2009) [hereinafter 2009 Audit].
. Betty Hallock, Per Se Health Code Violations Overturned by Judge – At Least Most of Them, L.A. Times (Apr. 4, 2014), https://www.latimes.com/food/dailydish/la-dd-per-se-health-code-violations-overturned-by-judge-20140404-story.html [https://perma.cc/U3X6-9PCZ]; Hugh Merwin, Per Se’s Terrible Health Grade Says More About the DOH Than It Does About Per Se, Grub Street (Mar. 7, 2014), http://www.grubstreet.com/2014/03/per-se-health-department-inspection.html [https://perma.cc/SZD2-CJLG].
. Bethany Jean Clement, New Food-Safety Emojis: Grading Restaurants on a Curve Raises Concerns, Seattle Times (Jan. 19, 2017), https://www.seattletimes.com/life/food-drink/king-countys-new-food-safety-emojis-restaurants-are-graded-on-a-curve [https://perma.cc/KV27-WYXP].
. Francis Lam, Cuisines Mastered as Acquired Tastes, N.Y. Times (May 29, 2012), https://www.nytimes.com/2012/05/30/dining/masters-of-a-cuisine-by-calling-notroots.html [https://perma.cc/99GC-FU3Y].
. See Eveline Chao, The Roast Duck Bureaucracy, Open City (Mar. 11, 2014), https://opencitymag.aaww.org/the-roast-duck-bureaucracy [https://perma.cc/QDN4-SEWT]; Lehman et al., supra note 90, at 2602.
. Ben Heubl, New York’s Asian Restaurants Fare Poorly in Health Inspections, Nikkei Asian Rev. (July 8, 2018), https://asia.nikkei.com/Spotlight/Datawatch/New-York-s-Asian-restaurants-fare-poorly-in-health-inspections [https://perma.cc/R7VJ-UDD5].
. See Cassandra Harrison et al., Using Online Reviews by Restaurant Patrons to Identify Unreported Cases of Foodborne Illness — New York City, 2012–2013, Ctrs. for Disease Control & Prevention Morbidity & Mortality Wkly. Rep., May 23, 2014, at 441, 444.
. See generally Eric T. Anderson & Duncan I. Simester, Reviews Without a Purchase: Low Ratings, Loyal Customers, and Deception, L1 J. Marketing Res. 249, 249 (2014) (observing that approximately 5 percent of product reviews for a retailer’s website were largely posted by individuals with no record of ever having purchased an item and were mostly negative).
. Jun Seok Kang et al., Where Not to Eat? Improving Public Policy by Predicting Hygiene Inspections Using Online Reviews, Proc. 2013 Conf. on Empirical Methods Nat. Language Processing 1443, 1445–46 (2013).
. See Sugarman, supra note 27, at 97 (proposing outcomes-based regulation to reduce the number of calories, amount of added sugar, and sodium in American foods, given “America’s twenty largest retailers alone account for more than 60 percent of the food we purchase in traditional food stores . . . .”).
. See U.S. Food & Drug Admin., FDA at a Glance (2019), https://www.fda.gov/media/131874/download (The FDA budget for fiscal year 2019 came in at $5.7 billion, of which 18.8 percent was dedicated to the food industry. This indicates that over $1 billion dollars is already allocated towards food safety regulation).
. Understanding SNAP, the Supplemental Nutrition Assistance Program, Feeding America, https://www.feedingamerica.org/take-action/advocate/federal-hunger-relief-programs/ [https://perma.cc/6Y4B-9666].