(Reprinted from HKCER Letters, Vol. 74 May-Aug 2003)
Cleanliness and Punishment
Severe Acute Respiratory Syndrome (SARS) has taught us the importance of environmental hygiene in a way that no publicity campaign can. Yet, old habits die hard. A person who ignores sound hygiene practices can put himself as well as the wider community at risk. When education is not sufficient to convert people's habits, some form of punishment may be necessary if we want to maintain a healthy and clean environment for all.
Economists identify two factors that determine the effectiveness of punishment in deterring undesirable acts. First, perpetrators of these acts must have a reasonable chance of getting caught. Second, the level of punishment for those who get caught must be sufficiently severe that it hurts. Up to some point, these two factors are substitutable with one another. To stop people from littering the streets, for example, the government may step up its enforcement efforts or raise the amount of the fine. How to choose the right mix between these two approaches is a matter that warrants some discussion.
The penalty is too high
Enforcement efforts requires real resources, while fines bring in revenue. So it may appear that raising the fine is a more cost-effective method of punishment. But this method may backfire if the level of the fine is perceived to be incommensurate with the gravity of the crime. Our heightened awareness of public health notwithstanding, littering and spitting are relatively minor offences after all. One does not gain a great deal by littering or spitting on the street. A fine of $600 is surely sufficient to deter people from committing these acts provided there is a sufficient certainty that perpetrators will be caught. For a middle-to-low income person in Hong Kong, $600 is about ten percent of his monthly earnings. Raising the fine to $1,500 means that this person will lose one quarter of a month's salary if he is caught littering the streets.
One immediate implication of the hefty penalty is that law enforcement by frontline officials will be made a lot more difficult. At the existing fine of $600, verbal abuse at law enforcement agents when they are carrying out their duties are common, and cases of physical assault have been reported. After the fine is increased by 150 percent, the situation will be even more difficult to manage. Given the current campaign to clean up the city, there is no doubt that frontline officials will not hesitate to enforce the law. But no campaign can last forever. When the enthusiasm wanes, it is not clear whether the sympathy will lie on the law enforcement agent or on the poor person who is slapped with a fine he can ill afford to pay. The will of these frontline officials remains to be tested.
In a 2000 study by Cass Sunstein, David Schkade, and Daniel Kahneman, published in the Journal of Legal Studies, the researchers conducted experiments with University of Chicago law students to find out people's moral intuition about optimal deterrence. They find that a majority of subjects reject the probability-penalty tradeoff proposed by economists. In playing the role of jurors, these subjects are reluctant to impose heavier penalty on a crime even if they are aware that the probability of detecting the crime is low. Even if an economist may wish that people would think in more "rational" terms, the reality is that penalties that are incommensurate with the crime run against widely held moral intuition. An attempt to impose too stiff a penalty by policy makers may not only induce hostility on the part of the offenders and on-lookers, but may also invite passive resistance on the part of law enforcement officials. After all, letting off an offender with a verbal warning would make everybody's life a lot easier.
The expected punishment is too low
Do not be mistaken: I am not suggesting that punishment for littering or spitting on the street is too high. Quite the contrary, punishment is too low even at a fine of $1,500, when the probability of facing prosecution is factored into account.
A comparison of littering offenses with parking violations will be illuminating. The amount of the fine for illegal parking is $320. To a driver who belongs to the middle-to-upper level income group, the fine probably amounts to less than two percent of his monthly income. Yet, most drivers will think twice before they park illegally, and by most standards the problem of illegal parking in Hong Kong is reasonably under control. This is because drivers face a significant chance of paying the fine if they commit a parking violation. In the year 2001, the police issued a total of 695,118 fixed penalty notices for parking offenses. This is more than 1,900 parking tickets a day. Traffic violations involving moving vehicles are much more difficult to detect than are parking violations. But even in that category, the number of fixed penalty notices issued was substantial, at 432,669. In comparison, 28,720 summons were issued against littering offenders in that same year. Since June 2002, summons were replaced by fixed penalty notices, and about 18,000 notices have been served to date. This amounts to a daily average to less than 50 fixed penalty notices against spitting and littering.
We do not know how frequent these two types of offenses are committed, but there is reason to believe that public cleanliness offenses are no less common in the population than are parking offenses. If we assume that the two types of acts are equally common, this would mean that a littering offender faces a probability of prosecution that is merely 1/38 as large as that for a parking offender. Even when the government stepped up its enforcement against littering offenses in its current cleanliness campaign, issuing on average 94 fixed penalty notices per day, the chance of getting caught for littering or spitting in public areas is still minuscule compared to the chance of getting caught for illegal parking.
Because of the low initial level of enforcement, doubling the enforcement efforts may not be sufficient to achieve the desired deterrence effect even though the fine has also been raised to $1,500. I would suggest that the government be even more aggressive in pursuing the offenders¡Xperhaps increasing its enforcement efforts five fold in the first few months, and then easing it off to a two to three fold increase subsequently. One advantage of vigorous enforcement is that justice is seen to be done. A heavier fine with a lower probability of prosecution can achieve the same deterrence effect, but people may resent random enforcement for being arbitrary. Another advantage of vigorous enforcement is that it creates many examples that people can relate to. It is one thing to learn on television that the fine for spitting on public places has been raised; it is another to learn from personal experience that one's friend or acquaintance has just been slapped with a fine. The vividness of the experience helps reinforce the deterrence effect.
Of course, with a fine set at $1,500, vigorous enforcement of the law is made more difficult that it otherwise would be if the fine were maintained at a more modest $600. Frontline law enforcement officers have to confront angrier offenders. More importantly, people who receive the $1,500 fixed penalty notice are more likely to contest it in courts, creating a burden on the most expensive part of our justice system. It turns out that a large fine may not always be the most effective way to bring about deterrence.
Differential rewards are important too
In addition to cracking down on littering and spitting in public places, the government has also launched a "blitz operation" against unhygienic food premises. Again, the penalties can be quite severe. Under the Demerit Points System, repeat offenders of public health regulations can have their licenses suspended or cancelled. Understandably, stiff penalties like these are seldom invoked. In 2001, for example, there were two cases of license cancellation. Heavy penalties applied sparingly can hardly deter.
Punishments are useful for enforcing minimum standards. When it comes to health and safety, however, minimum standards are not enough. Restaurants and other food premises should be given incentives to perform better than the minimum standard that the law requires. The problem is that consumers have no way to differentiate the best-practice restaurants from the merely satisfactory ones. Since restaurants cannot credibly convey their hygiene record to the public (every restaurant wants to claim that its record is superb), consumers cannot reward the extra efforts that restaurant owners put forth to improve their hygiene above the minimum level. The lack of information explains why many restaurants have fancy entrances and dirty restrooms¡Xand one can only imagine what the kitchens look like.
An analogy with academic standards may be instructive. Failure in an examination can be quite destructive for a student, and therefore the worst performers have an incentive to study to avoid failure. But precisely because the negative consequences of failure can be quite large, the standard for passing an examination is usually set at a relatively low level. If a teacher can only give a pass or fail grade, the average and the best students have no incentive to work hard. This analogy suggests a straightforward solution that many teachers already know: give the pass students finer grades (A, B, C, D's) to reflect their differential performance. I see no reason why this simple solution cannot be adopted to provide incentives to restaurants.
The Food and Environmental Hygiene Department already conducts regular and surprise health inspections on all restaurants and food premises in Hong Kong. It has a complete record the violations committed by each establishment. The only matter is to make this record widely and conveniently known to consumers. The government is considering to publish a "blacklist" of the worst offenders. But such a measure does not go far enough. Very few restaurants will make it to the blacklist, so the measure provides no incentive to reward performance above the minimum standard. Furthermore, few consumers will consult the blacklist before choosing a restaurant. A far more effective measure is to introduce a "grade card" system, in which restaurants and food premises are assigned different "grades" depending on the outcome of the health inspection. Restaurants should be required to display the "grade card" prominently at the entrance. For example, restaurants with a perfect record will display a green card, those with a few minor violations will display a yellow card, and those barely above the minimum will display a red card. The government's role is to educate the public about the criteria of awarding the different color codes. Say, for instance, the color codes may reflect a point system, and the government should let the public know what violations corresponds to how many points. The basic idea is to provide consumers with information that is easily accessible and understandable, so that consumers' choice will provide the market discipline to reward performance.
Economists have long supported policies that seek to increase the amount of information available to consumers. When consumers are well informed, the invisible hand of the market provides a system of incentives that is more flexible and fine-tuned than incentives based on the blunt arm of government regulations. A recent study by Ginger Jin of University of Maryland and Philip Leslie of Stanford University, published in the Quarterly Journal of Economics, has added some spice to this dictum.
In 1997 the Los Angeles County government passed an ordinance requiring restaurants to publicly display grade cards resulting from hygiene inspections. Prof. Jin and Prof. Leslie studied the effects of this policy. They had three key findings. First, the revenue of A-grade restaurants increased by 5.7 percent following the introduction of mandatory disclosure; the revenue of B-grade restaurants increased by 0.7 percent; and the revenue of C-grade restaurants decreased by 1.0 percent. This suggests that consumers do make use of the information about hygiene conditions to reward good performers by patronizing these restaurants more often. Second, average inspection score among all restaurants increased by 5.3 percent after the implementation of the mandatory grade card system. This indicates that mandatory disclosure does provide the incentives for restaurants to improve their hygiene conditions if the results of their efforts can be credibly conveyed to consumers. Third, the researchers estimated that mandatory disclosure caused a 20 percent reduction in food-borne illness hospitalizations in Los Angeles County compared to the rest of California. This suggests that the rise in average hygiene score reflects a real improvement in hygiene conditions, not a nominal improvement due to more lenient ratings.
The Jin and Leslie study underlies the importance of information and disclosure. Consumers know how to respond to information, and they are a better judge of their own interests than any benevolent government. Coincidentally, this is also a lesson we learn dearly in the SARS episode.
Professor Wing Suen is Professor of Economics at School of Economics and Finance, The University of Hong Kong.