Much has changed in the past two decades. The 1990 Oil Pollution Act (OPA), passed a year after the Exxon Valdez spilled more than 10 million gallons of crude into Prince William Sound, states that a company cannot ship oil into the United States until it presents an acceptable plan to prevent spills; it must also have a detailed containment and cleanup plan in case of an oil spill and all vessels entering U.S. waters must eventually be double-hulled. Since then, the number and volume of spills in U.S. waters has declined considerably. For example, the Coast Guard reports the number of spills to have dropped from about 700 to 400 annually, and the volume of oil spilled reduced from about 5 million gallons to 600,000 gallons annually since OPA was enacted.

But those numbers do not tell the whole story. Not all spills are large and many are not even accidental: vessel operators have been known to clean their bilges out near a port in order to save money, and some spills simply occur through faulty or negligent transfer operations.

Aside from technological mandates such as double-hulled tankers, how effective are the various approaches—monitoring, enforcement, penalties—in deterring oil spills, and what is the best mix?

Assessing data on compliance and enforcement is not an easy task. A reported increase in enforcement activities might indicate more frequent spills, but it could also reflect better monitoring and detection, or more vigorous prosecution. Empirical studies must be carefully designed to sort out the effect that these variables have on actual spill frequency versus spill detection.

Monitoring oil transfer operations has been found effective in reducing oil spill volumes: the crew of a tanker apparently takes more care when the Coast Guard is watching. Such monitoring might also have a general deterrent effect on all vessels that transfer oil: if their captains believe they might be monitored in the future, they probably train their crews and check their equipment more thoroughly, even if they are never actually monitored. Random port patrols looking for oil sheens have a similar influence because they raise the probability of detection for all vessels entering that port. However, compliance inspections themselves have not been found to be as effective as the other two mechanisms.

Alternative Approaches
Because government monitoring is expensive, three alternatives have been tested: targeted monitoring for vessels thought likely to be out of compliance or likely to spill oil; differential penalties based on prior compliance history, with higher penalties for frequent violators; and mandatory self-reporting, with higher penalties for vessel operators who do not voluntarily report their spills.

Targeted monitoring. In the early 1980s, the Coast Guard began classifying ships as low risk (to be monitored only occasionally) and high risk (always monitored). This two-tiered enforcement policy has been found to be effective in reducing the cost of enforcement without having a negative effect on the environment.

Differential penalties. A 2000 study by Weber and Crew found penalties ranging from $.003 to $73.35 per liter, and estimated that increasing the fine for large spills from $1 to $2 a gallon decreased spillage by 50 percent.  They concluded that the current penalty policy—relatively high per-gallon fines for small spills and very low per-gallon fines for large spills—undermined deterrence. Their results parallel mine, that the Coast Guard’s statutory maximum penalty of $5,000 was too small relative to the optimal penalty required. Under OPA, the potential penalties considerably increased, up to $1,000 per barrel of oil discharged.

Self-reporting. To increase deterrence and lower the cost of government monitoring, vessel operators are told they must report any spill, and if the government detects a spill that was not voluntarily reported, the penalty is higher and may include a criminal sanction. Firms found to be out of compliance are more likely to self-report violations in subsequent periods. This suggests that firms try to regain credibility with the government so that they will be taken off a target list.

Firm reputation. Information that a firm has been sanctioned for violating environmental laws may be of interest to shareholders or lenders if the monetary sanction reduces the expected value of the firm and therefore its share price or bond rating. It may also give lenders and insurers pause about risking more capital on that particular firm. Other costs might include future debarment from government contracts, targeted enforcement by EPA, and lost sales to "green" consumers. Several studies looking at bad environmental news, such as oil or chemical spills or the announcement of civil or criminal enforcement actions, have demonstrated a negative stock price effect; however, the evidence is mixed as to whether or not this price effect simply reflects the expected cost of penalties and cleanup as opposed to any additional reputation penalty.

Policy Implications
Despite OPA's success in reducing oil spills, costs are still significant. A recent Coast Guard study estimated the total cost of removal and damages from oil spilled since 1990 to be $1.5 billion. If the government's goal is to improve the environment at the least cost to society, then firms that are the most likely to cause significant harm need to be identified along with those most likely to be responsive to enforcement activities as well as compliance assistance. This kind of empirical evidence can help government agencies plan targeted enforcement measures. Additional evidence on the cost of enforcement and compliance must be gathered, however, to conduct a cost-benefit analysis.
 
In terms of sanctions, the evidence to date shows little deterrent effect from fines that are only a few thousand dollars. To have any real effect, significantly larger fines and/or targeting individuals instead of firms may be appropriate.

Finally, community pressure and social norms can be important factors in compliance. External market pressures may exert some influence on firm behavior and help prevent oil spills from occurring. Being known as a polluter may induce firms to take precautions, lest consumers and shareholders exact their own form of punishment.