LOCAL GOVERNMENT LIABILITY FOR ERRONEOUS DATA: LAW AND POLICY IN A CHANGING ENVIRONMENT

R. Jerome Anderson
Consultant
Real Estate Information Systems Project
United States Agency for International Development

Alan R Stewart
Design Analyst
PlanGraphics, Inc.
San Antonio, Texas


1. INTRODUCTION

Kenneth Culp Davis, in a 1956 law review article on governmental tort liability,1 opined that there was a need for a "system of sovereign responsibility" in that field. Today, some thirty-eight years after the publication of Davis' article, there is an ever-growing need for a coherent "system of sovereign responsibility" in the field of local government liability for erroneous data from computer-based information systems, including geographic information systems (GIS). Indeed, the legal landscape is cluttered with various theories regarding liability, or non-liability, for governmental entities in the field of information dissemination. These theories range from first amendment concerns for free expression and wide distribution of data to tort theories of negligent or fraudulent misrepresentation, malpractice, and defamation. As governments engage in "proprietary" activities of "selling" information-based products and services, contract and possibly even Uniform Commercial Code (UCC) theories may be in issue. Thus, as governments face greater and greater use of their electronic information (whether they intend that use or not), these issues will become of increasing importance.

The purpose of this article is to explore the various theories under which units of local government may be sued for erroneous information which they may provide, to examine defenses which may be raised against those claims, and to suggest a framework under which governments may legitimately be held accountable for their errors while at the same time limiting their liability to a reasonable level. In addition to the issues listed in the preceding paragraph, tort claim legislation, as passed by the various states, will be examined to determine the extent to which those laws may be useful or not in protecting local governments from liability for erroneous information.

The scope of this article is limited to liability concerns of local governments for two specific reasons. First, there are literally thousands of local governments in the United States, and the sheer number of them requires that their potential liability for erroneous data be examined. Many of these local governments are developing GIS and other spatial information delivery systems, and will face liability issues as more and more demands for data are made on these systems by members of the public.

Second, local governments were chosen because it is at the local level that the bulk of parcel-related data is collected. Parcel-based systems are developed at the county or municipal level in the United States because that is where the data is captured, whether in the auditor's or assessor's office, or in the offices of clerks or recorders of deeds. Local governments need direction as to their potential liability exposure simply because they cannot avoid collecting and disseminating land-related data. In many respects, collecting and disseminating such data is their "business."

Four major sections follow this introduction. In the next section, five theories of liability will be discussed. The section after that presents a brief discussion of sovereign immunity, its judicial abrogation, and the response of the state legislatures in enacting tort claims legislation. Defenses to claims for liability will be presented in the following section. The final section will contain the authors' recommendations.

The recommendations and conclusions presented herein will obviously be subject to debate, criticism, and refinement as the research process continues. The ultimate test of the validity of the authors' recommendations will be the initiation of projects which advance the research identified herein, and the adoption of tort claim legislation along the lines suggested. In any event, it is important for local government officials and solicitors to understand the liability issues which will be raised as local governments become increasingly significant providers of spatial information.

2. THEORIES OF LIABILITY FOR DISTRIBUTION AND DISSEMINATION OF ERRONEOUS DATA BY LOCAL GOVERNMENTS

The question of liability for erroneous data or information is one which will become increasingly important, not only for local governments, but for those in the information industry generally. In 1992, the information industry was estimated to generate $12 billion in annual revenues, with online service providers generating $4.6 billion of that total.2 With a growing clientele that increasingly expects data from the computer to be "right," the potential for exposure to suit by disgruntled customers is growing. This is especially true in those instances in which clients of information providers have paid substantial sums to obtain information.3

Local governments are becoming increasingly aware of opportunities to provide information-based products and services, often for fees which cover more than the cost of meeting particular requests. Legislation in a number of states, such as Iowa,4 Connecticut,5 Oregon,6 and Alaska7 permit local governments to charge such fees. The extent to which governments actually "sell" their data, and generate significant fees from such sale, is an area to be pursued in further research. It appears, however, that based on discussions with attendees at conferences such as the Urban and Regional Information Systems Association Conference, and limited market studies the authors have conducted for clients in consulting engagements, that there is a growing amount of entrepreneurial activity on the part of local government information resource managers. Given the need for the data by the public, and the pressure on local governments to find new sources of revenue, it does not appear that this trend is likely to reverse.

If local governments are going to be in the business of providing data or information on an entrepreneurial basis, then it would appear that some analysis of their potential tort liability would be an important part of the business case or plan which should be developed in preparation for the business venture. Knowing the potential exposure they face would seem to be a necessary part of the equation as they decide whether or not to engage in that kind of activity. While exposure to tort liability may not be a determining factor, it may, in combination with other factors, influence the decision to engage in a business-type venture.

Providers of information may be held liable in tort under at least five theories. Those theories are negligent misrepresentation, fraudulent misrepresentation, strict liability, malpractice, and defamation. Intentional misrepresentation may also be a cause of action under applicable circumstances, and that tort is sometimes mentioned in municipal tort immunity statutes. Elements of these causes of action, and their potential applicability to municipal information providers, will be examined below.

2.1 Negligent Misrepresentation -- § 552, Restatement, Second, Torts

Perhaps the most likely basis for an action against a municipal data or information provider would be an action brought under § 552 of the Restatement, Second, Torts for negligent misrepresentation. Section 552 provides as follows:

(1) One who, in the course of his business, profession or employment, or in any other transaction in which he has a pecuniary interest, supplies false information for the guidance of others in their business transactions, is subject to liability for pecuniary loss caused to them by their justifiable reliance upon the information, if he fails to exercise reasonable care or competence in obtaining or communicating the information.

(2) Except as stated in Subsection (3), the liability stated in Subsection (1) is limited to loss suffered

(a) by the person or one of a limited group of persons for whose benefit and guidance he intends to supply the information or knows that the recipient intends to supply it; and

(b) through reliance upon it in a transaction that he intends the information to influence or knows that the recipient so intends or in a substantially similar transaction.

(3) The liability of one who is under a public duty to give the information extends to loss suffered by any of the class of persons for whose benefit the duty is created, in any of the transactions in which it is intended to protect them.

Subsection (1) imposes liability on local government employees, and upon local governments through the principle of respondeat superior, if the following conditions are met:

  1. information is provided;

  2. for a business purpose;

  3. in the course of business, profession, or employment, or as part of a transaction in which the information provider has a pecuniary interest;

  4. the recipient suffered a loss in reliance upon the information; and

  5. the local government did not exercise reasonable care or competence in obtaining or transmitting the information.

Section 552 (1) should apply those situations in which local governments seek to raise revenues, especially general revenues, through the sale of data. The governmental entity's defense, that of reasonable care or competence in obtaining or transmitting the information, requires an analysis of the design, development, operation, documentation, and output production, of the spatial data system. This requirement may put a significant burden on the agency which develops the GIS or other spatial data system.

A unit of local government could be exposed to liability even if it did not attempt to raise revenues from its spatial data system. This could occur if several units of local government were involved in a data sharing arrangement, in which each contributed various "layers" of data to a common GIS. This sharing of data in order to participate in a larger spatial data system than the government could afford to develop on its own could create a situation in which the government obtains a "pecuniary interest" under Subsection (1), thus triggering liability if the data which it contributed were erroneous and others in the consortium relied on that data to their detriment.

Additional concerns are raised under Subsection (2) in the event the entity to which a local government "sells" data or information products incorporates that data into its own information products and sells them to third parties. The language of § 552 (2) places on the governmental body which sells data the obligation to ascertain the uses of the data and the extent to which its customer intends to resell the data. If it can be reasonably expected that data will be resold, the potential liability exposure will extend to those who can be reasonably expected to-obtain the data through resale.

Finally, it should be noted that a unit of local government could potentially be held liable under § 552 for information disclosed pursuant to a request made under a state open records law. A response to a request under an open records law would be one arguably made "under a public duty," and thus would expose a governmental entity to liability if the requester suffered a loss because of erroneous data supplied in response to the request. Comments to this subsection of the Restatement seem to confirm this conclusion, without specifically addressing the open records law question. Even if a jurisdiction does not engage in a program of "selling" data, care should be taken to avoid negligent misrepresentation because liability could attach even when data is provided for personal or other non-commercial uses, if the data is provided under any public duty to do so.

2.2 Fraudulent Misrepresentation

Basic rules regarding fraudulent misrepresentation are presented in several sections of the Restatement. While a comprehensive treatment of all these sections is not possible in these pages, some of the more salient points should be briefly discussed.

Perhaps the most important aspect of the sections of the Restatement governing fraudulent misrepresentation are found in §§ 525 and 526. Section 525 provides as follows:

One who fraudulently makes a misrepresentation of fact, opinion, intention or law for the purpose of inducing another to act or to refrain from action in reliance upon it, is subject to liability to the other in deceit for pecuniary loss caused to him by his justifiable reliance upon the misrepresentation.

Section 526 provides as follows:

A misrepresentation is fraudulent if the maker

(a) knows or believes that the matter is not as he represents it to be,

(b) does not have the confidence in the accuracy of his representation that he states or implies, or

(c) knows that he does not have the basis for his representation that he states or implies.

While it is possible that a particular government employee may deliberately make a fraudulent misrepresentation for personal gain or otherwise, such a situation would probably be rare, and could be dealt with on a case by case basis under applicable legal standards. In that situation, the local unit of government which employs that person would probably escape liability because the statement would be made outside the scope of the person's employment.

It is more likely that fraudulent misrepresentation could occur in the GIS environment if a unit of local government determined that it must sell data in order to raise revenues, and in the course of selling the data deliberately misrepresented the quality of the data. If, in an aggressive cost recovery program, the government's employees were pressured to sell data, regardless of the quality of the data and regardless of whether the data was appropriate for the purpose for which the customer needed or wanted the data, any of the three tests found in § 526 (a), (b), or (c) might be implicated. In this situation both the employee and the municipal employer would probably be liable for the damage caused by the fraudulent misrepresentation.

While it may not be likely that a particular government or agency would deliberately try to aggressively market data which is not suitable for the needs of its customers, it is not unlikely that there will be isolated instances in which data is sold under circumstances in which the seller may not "have confidence in the accuracy of his representation[s]" regarding the data This is so because of the nature of GIS databases. Data may be obtained from multiple sources, at multiple scales, with differing degrees of accuracy and currency. When data from those multiple assembled to meet a particular request, it is not unlikely that the particular combination of data could not be relied upon for the purposes for which the customer sought the data. If the governmental seller were aware of the limitations of the data, but represented the data as being suitable for the customer's purposes, liability under §§ 525 and 526 of the Restatement could be possible.

2.3 Strict Liability

In some cases, publishers or creators of data may be held strictly liable for the data products they create. While it is difficult to foresee an instance in which such liability might arise in the GIS context, the possibility of a claim being made on the basis of strict liability should at least be addressed.

Section 552C of the Restatement, Second, Torts provides, in pertinent part, as follows:

(1) One who, in a sale, rental or exchange transaction with another, makes a misrepresentation of a material fact for the purpose of inducing the other to act or to refrain from acting in reliance upon it, is subject to liability to the other for pecuniary loss caused to him by his justifiable reliance upon the misrepresentation, even though it is not made fraudulently or negligently.

The comment to Subsection (1), at b. Relationship to action for restitution or breach of warranty [], indicates that this subsection may impose strict liability because it eliminates the requirement of intent or negligence. Therefore, a claimant could bring an action under this section of the Restatement for damages caused by erroneous data without the necessity of alleging or proving negligence or intent. While this might make it easier for the plaintiff to prove the case, this Section of the Restatement does provide some measure of protection for the defendant, in that Subsection (2) limits damages to the difference in value between what the plaintiff received and what the plaintiff paid. This would amount, for practical purposes, to a refund of the purchase price at most. Claims for damages or loss of bargain are not permitted under Subsection (2) of § 552C.

Another possible basis for strict liability under the Restatement is found under § 402A. In a rather famous case, Brocklesby v. United States,9 the publisher of an instrument approach chart was found to be strictly liable under § 402A for defects found in the chart. Because the publisher's procedures required its employees to investigate the source material it obtained from the federal government, and make corrections as necessary, the court found that the publisher could have prevented the injuries resulting from the defects, had it followed its procedures. Because the product it created was "defective [and] unreasonably dangerous to the user," 10 the court imposed strict liability on the publisher.

The extent to which this case has precedential value for units of local governments is not known, but it does raise the question of the extent to which those responsible for creating and maintaining GIS databases must fully investigate and document the quality of the data which they place in their systems. If an investigation reveals defects in the data, then it would seem prudent for GIS administrators to exclude such defective data from any products or services they provide, in order to avoid strict liability for damages which may be caused by such defects.

2.4 Malpractice

Students of societal change driven by developments in information technology have noted that there is a growing need for specialists who can find, analyze, and assemble data from disparate sources to meet specific client requests.11 Cornish has described the new environment as one in which "knowledge scientists" will package "statistics, manipulated data, law, company information, economic projections and some documents.''l2 While it may be some time before county and municipal governments in the United States offer such comprehensive services, those jurisdictions seeking to recover costs for development of their spatial data systems may well find themselves offering fairly sophisticated search and retrieval services, as well as data analysis, macro programming, and cartographic production services. As these services expand, county and municipal clerks and GIS technicians may find themselves moving along the continuum toward becoming "knowledge scientists."

As this occurs, a claim of malpractice may be raised if the information supplied in response to a particular request is deficient in some way. Indeed, as Tarter has stated, ". . . myth of machine infallibility seems to create a demand for a higher standard of quality for machine readable data than for traditionally distributed information.''13 The field of computer malpractice is developing, with at least two courts holding that computer consultants are professionals for purpose of malpractice liability.l4 Suits against librarians for malpractice have also been anticipated.l5 One cannot predict how soon such an action could be brought against a municipal or county information provider, but as the field of "information malpractice" develops, it is not unlikely that a malpractice claim could be brought against a governmental entity for failing to use the requisite skill in meeting requests for information, especially if the governmental entity has held itself out as being expert in providing, for example, complex spatial data analyses upon request and for payment of significant fees.

2.5 Defamation

Defamation actions, like fraudulent or intentional misrepresentation cases, will hopefully occur only rarely, if at all, in cases involving governmental spatial data. It is possible, however, that given the proper set of facts, an action could be brought against a municipal defendant for defamation. Court opinions indicate that in matters of purely private concern, as opposed to matters of public importance, punitive damages may be awarded against defendants who defame plaintiffs. In a leading case on this issue,16 the Supreme Court of the United States permitted a plaintiff who alleged defamation to recover both compensatory and punitive damages from a credit reporting agency which had erroneously reported that the plaintiff had filed for bankruptcy.

Local governments face potential liability in analogous situations. Possible claims could arise from improper listing of property for tax sale, erroneous reporting of unpaid taxes, or an incorrect statement as to judgments entered against an individual or business. To the extent that local governments collect and publish data on business or financial matters of their citizens, they may make errors and be exposed to claims of this type. As geographic information systems become more comprehensive in their scope, and truly become multi purpose cadastres containing financial data regarding properties and persons, the potential for this kind of liability will increase.

3. LOCAL GOVERNMENT IMMUNITY FROM TORT CLAIMS

There is an extensive literature dealing with the subject of local government immunity from tort claims.17 An exhaustive review of that literature is far beyond the scope of this paper. For purposes of this analysis, it may be sufficient to indicate that local government immunity from tort claims is generally considered to be a doctrine created at common law.l8 Judges generally believed that they had the power to change that doctrine in their decisions19 as changing circumstances led them to remove immunity from units of local government and allow tort plaintiffs to recover damages for their injuries. Virtually all of the states, with the exception of New York and Hawaii, have enacted some sort of legislation dealing with local government tort liability.20

The laws enacted by the various state legislatures generally fell into one of two categories, with some exceptions that may be noted below. Those categories are "open-ended tort liability acts," which impose broad liability and then grant immunity in certain specific areas; and "closed-ended" acts, which grant immunity generally, and then indicate the specific actions which may give rise to liability.2l Three states -- Arkansas, Vermont, and North Carolina -- provide for immunity except to the extent that local units of government carry insurance to protect them against their torts. 22

Some of the acts maintain old distinctions between governmental and proprietary actions;23 other states specifically state that the distinction between such actions are abolished;24 still other states provide that immunity is granted for both kinds of actions.25

The tort immunity statutes were analyzed to determine how they might be applied to claims which may be brought against local governments for erroneous information or defamation. Three categories seemed to be apparent from this analysis. The first category is complete immunity from suit for actions which might arise from erroneous data. This category includes statutes which grant general immunity or which grant immunity for the kinds of actions which might arise from erroneous data. The second category includes statutes which do not grant immunity to local governments. The third category includes those statutes which grant some immunity to local governments, but which do not grant immunity with respect to proprietary functions. The results of this analysis are presented in Table 1.


Table 1
Local Government Tort Claim Immunity in the Fifty States

Complete		No Immunity			No Immunity for            
Immunity						Proprietary Functions      
20 States		22 States			8 States                 


As can be seen from Table 1, twenty-two of the states withhold immunity from suit for tort actions generally, and another eight states withhold immunity for actions which are determined to be "proprietary" in nature. Assuming that the provision of spatial data in a cost recovery program is a "proprietary" activity, local governments in up to sixty percent of the states would not be able to raise a governmental immunity defense in a tort action based on erroneous data. This leaves a significant number of jurisdictions limited to traditional defenses against such claims.

4. DEFENSES AGAINST TORT CLAIMS

Because § 552 of the Restatement, Second, Torts indicates that one which provides data under a duty to do so may be liable for damages caused by errors in that data, any public body which maintains spatial data is potentially subject to liability, even if the agency does not provide data or information services on a fee basis as part of a cost recovery program or business-like enterprise. For this reason, local government solicitors should be alert to the defenses which they may raise in the event they need to defend a suit, whether the action is brought by a fee-paying client or a requester under a state open records law.

Four different defenses will be briefly noted in this section. First will be defenses based on contracts. Local government tort immunity statutes will be discussed next. The third defense is based on exercising the proper care in developing and maintaining spatial databases. The concluding defense is based on the First Amendment to the United States Constitution.

4.1 Contractual Defenses

While contracts may not be appropriate in situations in which a requester has sought data or information under a state open records law, they should be used in any situation in which data or services are provided in the context of a cost recovery program or other entrepreneurial enterprise. These contracts should specifically provide that the purchaser of the data or service waives any claims for loss or damage caused by any errors in the data, its manipulation or analysis, or in the manner of the presentation of the data. A covenant not to sue should also be included.

Local counsel will need to carefully study the law regarding such clauses in each jurisdiction. Courts are often reluctant to enforce such clauses, especially if the parties to the contract do not have equal bargaining power. If the courts find that there is a lack of consideration for the waiver and covenant not to sue, the clauses will probably not be enforced. It may be possible to reduce the cost of the data or service to such a level that the clause could be enforced, but if that is done, the price would probably be so low that any cost recovery goals would not be met. Even though such clauses may not be enforced, they should probably be retained for whatever benefit they might have in litigation.

4.2 Defenses Based on State Tort Claim Acts

State tort claim acts should be carefully analyzed to determine the extent to which they offer protection. In those jurisdictions in which immunity is granted for the kinds of tort claims discussed in this article, solicitors should be cognizant of the procedural rules which must be followed in order to properly raise the defense. Failure to do so may result in loss of the defense, with possible liability as the result.29

If the state does not grant immunity from liability resulting from erroneous information, the applicable tort claim legislation should be reviewed to determine if a damage limitation may apply. Many of the tort claim laws which do not grant immunity in most cases limit the amounts which plaintiffs may recover against local government defendants. If the damage award has the possibility of exceeding these limits, these statutes will afford at least some protection.

4.3 Exercising the Proper Standards of Care

Subsection 552(1) of the Restatement, Second, Torts provides that liability will be imposed for loss caused by erroneous data if the provider of that data "fails to exercise reasonable care or competence in obtaining or communicating the information." Thus, an action in tort for damages for erroneous information may possibly be successfully defended by showing that the information provider did in fact use reasonable care or competence in collecting the information, designing the system, and preparing the output from the system. This raises the issue of standards. Standards must be defined so that conduct alleged to be negligent may be judged against criteria which are measurable so that an objective determination as to the actor's conduct may be made.

In the design and development of a spatial data system there are a number of areas in which standards may be applied. Croswell and Archer have identified six areas in which standards are or should be developed:

Unfortunately, the GIS field has not matured sufficiently to permit standards in these areas to be developed to the point that they could be utilized in determining whether or not a particular spatial data system was properly designed and developed. While various agencies and committees within the federal government have published materials regarding metadata (data about data) and cadastral data content,31 it is not known if adherence to these standards would be sufficient to defend a claim of negligence. The same would be true with adherence to national map accuracy standards.32 It appears that the field will need to mature further before defined standards in these areas may be accepted by courts as standards against which to judge negligent conduct.

The issue of whether or not a local government acted in a reasonable or competent manner is further clouded by the fact that users or requesters may play a part in shaping the request for information products, and may give direction to the information provider, thus potentially limiting the provider's control over the final product. Actions by the requester may raise to the status of contributory negligence, or they may be so overt as to remove all liability from the information provider. The specific facts of each situation will control, to a large extent, the answer to the question as to whether the governmental information provider acted in a reasonable manner.

The question of acting in a reasonable or competent manner may be even more confused if the information provider is a multi-participant consortium of government, utility, and academic bodies. Large consortiums are common in spatial database development. The response to a specific request may involve data from several different providers, with employees of still other participants actually interpreting the request, writing the necessary code, and producing the output. Distinctions must be made as to why a particular product or service did not meet the need for which it was created. Data gathering, data conversion, database design, query design, and the format or content of the specific output must all be considered. Errors might be found in some or all of those processes. Determining which caused a particular loss may become a very tedious process.

The standards to be applied may also vary depending on whether the particular output generated by the information provider is defined as a "product" or a "service." This is important because a service is generally held to a simple negligence standard, while a product may be subject to a strict liability standard.33 Furthermore, if the product is determined to be a good, then Uniform Commercial Code Standards may be implicated.34

The current situation of undefined standards may work to the benefit of information providers, if they can show that their conduct, even though it may have caused a loss, was nevertheless "reasonable" or "competent" when compared with the actions of others in similar situations. On the other hand, judges and juries may be inclined to find unreasonable conduct if it can be shown that the information provider failed to adhere to even some of the rudimentary standards which now exist in the field. Obviously, designers and managers of spatial data systems should do their utmost to insure the integrity of their systems, and communicate the limitations of their data and software to potential customers who come to them with requests for information-based products and services. Full and complete disclosure of the limitations of both data and software, with an explanation of how those limitations will affect the resulting products or services, may prove to be one of the most effective defenses.

4.4 First Amendment Defenses

It may be possible, in certain circumstances, for a local government to successfully defend an action for damages based on erroneous information by asserting a federal Constitutional defense based on the First Amendment. While this defense may not be applicable in all situations, if the facts warrant it, the defense should be raised.

One example of the use of this defense is found in the New York case of Daniel v. Dow Jones & Co.. Inc.35 The plaintiff in Daniel brought an action against the publisher of an on-line news and information retrieval service based on negligent misrepresentation. The alleged misrepresentation was the failure of the defendant to indicate that the price of certain shares of oil company stock were in Canadian dollars. The plaintiff assumed the price was in United States dollars, and, based on that assumption, suffered a loss when he invested in those shares.

In deciding the case, the Court relied upon two theories. One centered around the required "special relationship" necessary to maintain an action for negligent misrepresentation. The court held that there was no such relationship in this case, and followed the rule that the publisher of newspapers are generally not liable on negligence theories for inaccurate statements absent that special relationship.

The Court also discussed the First Amendment defense, and held that it too barred this claim. Relying on Time, Inc.. v. Hill,36 the Court held that "The First Amendment precludes the imposition of liability for non defamatory, negligently untruthful news."37 Furthermore, the court stated that

...the Supreme Court held that a negligence standard opened the press to the 'intolerable burden' of trying to satisfy a fact finder that its acts had been reasonable in determining the accuracy of the news items portrayed. A free press must operate without the imposition of such 'intolerable burdens' on reporting. The First Amendment guarantees that society will be able to receive news limited, if at all, only by the most carefully imposed, narrowly constructed, finely drawn, restraints in areas such as copyright, defamation, unfair competition, incitement and obscenity.38

Thus, the news organization successfully defended the negligence claim on the basis of the First Amendment.

The extent to which this defense would be open to local governments is not known. It does appear possible, however, that a local government which provided on-line access to spatial information databases could utilize this defense if it were found that the database contained erroneous data and that the errors were the result of negligence. In that situation it would be arguable that the local government was in a situation analogous to that of the publisher and the customer was simply a subscriber. If that were the case, the local government may be able to argue that it should have the same rights as other publishers, and rely on First Amendment protection to shield it from liability for mere negligence.

5. RECOMMENDATIONS

5.1 Suggestions for Further Research

The foregoing analysis suggests that there are a number of areas for further research. Most of these research areas are pragmatic in nature, and involve a study of current activity in the field, rather than in-depth legal analysis, although some of that may be needed as well. The authors' suggestions for further research are as follows:

  1. Determine the amount of data sales which are actually taking place, and the extent to which sales are likely to continue. While it appears intuitively, and anecdotal evidence confirms, that the number of jurisdictions selling products from spatial databases and the volume of such sales are growing, there is little empirical evidence to support these conclusions. If the market is in fact growing, then risk management activities should be given a high priority. If the market is shrinking, then perhaps protection from liability is not as important.

  2. An analysis should be made to determine if there is a correlation between sales of data and the presence of tort claims acts with a broad grant of immunity. If there is such a correlation, an evaluation should be made to determine if the presence of the favorable tort claims acts led to the increased level of sales. If there is a positive correlation between favorable tort legislation and sales, the passage of such legislation in other jurisdictions could be seen as a catalyst. On the other hand, if there is no such correlation, then changes in the tort claims laws may not be needed and effort need not be expended to secure those changes.

  3. Studies should also be conducted to determine the number and type of claims which are being brought against information providers, and what, if any, defenses are successful against those claims. To the extent that lessons can be learned from an analysis of litigation, local governments can benefit and devise strategies to protect themselves based on the successful actions of other information providers. Local governments could structure their information-based activities to reflect those of the successful information providers.

  4. Research should be conducted to determine the effectiveness of the various kinds of state tort claim laws. Do "open-ended" laws cause significant hardships for local governments? Do "closed-ended" laws unduly burden plaintiffs, and prohibit them from recovering damages to which they are rightfully entitled? To what extent do these laws contribute to a proper allocation of risk between those who are injured and those who presumably can pay to compensate for injury? Has our experience under these laws given us some understanding of what kinds of laws produce what kinds of results, so that changes may be made where warranted?

  5. Additional research should be performed in the area of standards for the design and development of spatial data systems. Standards should also be developed regarding the collection of the data to be stored in these systems. Standards for documentation of data files, or metadata standards, should also be developed.

Answers to these questions will not help to resolve debates over which policy should be preferred. Undoubtedly there will be those who wish to defend local governments from claims, as there will be those who believe that the governments are in the better position to bear the losses which are suffered as a result of erroneous data. But while this research will not end the debate, such research will at least give policy-makers and legislators the factual support they need as they consider policy and legal changes to respond to the continuing changes in the field of government electronic information dissemination.

5.2 Recommendations for Policy and Legal Change

It is the authors' position that local governments should be free to engage in the business of selling products and services based on their spatial databases and other information resources. It is their belief that such activity can be healthy, both for the governmental units, as sellers, and for the consuming members of the public. Such sales can actually create a "win-win" situation, in which the public obtains data which it would not otherwise have, and the government obtains a revenue stream with which it can develop and improve its information resources. The assumption is, of course, that without that revenue stream governments would be in a less financially stable position and would not be able to provide additional products and services. The authors further assume that those who benefit most from the provision of data should contribute to the cost of that data, and that therefore "user fees" for information products are not inappropriate.

If governments are going to engage in such activity, however, they must be prepared to accept the responsibility which attends commercial enterprise. It would seem to be unfair to require a private information vendor to subject itself to litigation over data quality and other issues, while not requiring a public sector data provider to meet the same requirements. This suggests that tort claim laws should be amended to provide that local governments could face liability if users suffer losses from erroneous data created or supplied by local governments.

Rather than rely on the "governmental/proprietary" test which has led to much confusion, uncertainty, and conflicting results when applied by courts,39 the authors suggest that the approach taken by Connecticut in its political subdivision tort immunity statute be considered. That statute provides, in pertinent part, as follows:

Except as otherwise provided by law, a political subdivision of the state shall be liable for damages to person or property caused by: .. (B) negligence in the performance of functions from which the political subdivision derives a special corporate profit or pecuniary benefit[.]40

This test is appealing for two reasons. First, if a government provides information in response to a statutory requirement, such as a request under an open records law, immunity would apply because in that case the political subdivision would receive no "special corporate profit or pecuniary benefit." It would simply receive the fee imposed by the open records law, which is typically the cost of reproduction or duplication only. This minimal fee should not be considered a "special profit or pecuniary benefit."

Second, if the political subdivision wishes to enter the market as an information provider, and supply information or services or provide other entrepreneurial functions, it will be held accountable for its errors as would any other information provider. This will place the local government on the same footing as other information providers, and would remove the advantage of governmental immunity, thus leveling the playing field as between public and private information providers.

The field of government-provided information services is yet in its infancy. As local governments attempt to ''reinvent''41 themselves and find new sources of revenue to meet an ever-expanding list of demands for services, more and more governments may attempt to enter the information services market. Liability issues are but one aspect of the legal framework which should be established for that activity with some forethought and analytical rigor. At least some of the issues to consider in that regard have been identified in this paper.

6. REFERENCES

1 Kenneth Culp Davis, Tort Liability of Governmental Units, 40 MINN. L. REV. 751, at 813 ( 1956).

2 Blodwen Tarter, Information Liability: New Interpretations for the Electronic Age, Xl COMPUTER/LAW J. 481, at 483 (1992).

3 Id, at 484.

4 I.C.A. § 22.2.

5 C.G.S.A. § 7-148h.

6 ORS 268.310.

7 House Bill No. 405, Sixteenth Legislature, Second Session.

8 This part of the test raises significant questions as to what constitutes a "business" in the public sector. For purposes of this discussion, it will be assumed that this condition will be satisfied if the governmental entity imposes a charge for data in excess of the "cost of duplication" requirement set by many state open records laws. If the entity charges a reproduction fee, presumably liability would arise under Subsection (3) for any data provided under an open record law or other legal mandate.

9 767 F.2d 1288 (9th Cir. 1985).

l0 767 F.2d at 1296

11 Blodwen Tarter, Information Liability: New Interpretations for the Electronic Age, XI COMPUTER/LAW J. 481, at 491-93 (1992); Graham Cornish, Document Supply in the New Information Environment, 23 (3) JOURNAL OF LIBRARIANSHIP AND INFORMATION SCIENCE 125 (September, 199l).

12 Cornish, Op. Cit., at 126 and 127.

13 Tarter, op. cit., at 484.

14 Data Processing Services, Inc. v. L.H. Smith Oil Corp., 492 N.E.2d 314 (Ind. Ct. App. 1986); Diversified Graphics, Ltd. v. Groves, 868 F.2d 293 (8th Cir. 1989); Note, Computer Malpractice: Two Alternatives to the Traditional "Professional Negligence " Standard, XI COMPUTER/LAW J. 323 (1992).

15 Tarter, op. cit., at 491~93; William Z. Nasri, Professional Liability, 7 (4) JOURNAL OF LIBRARY ADMINISTRATION 141 (1986).

16 Dun & Bradstreet v. Greenmoss Builders, 472 U.S. 749 (1985).

17 Among the numerous articles, cases, and annotations which could be cited are the following: Davis, Tort Liability of Governmental Units, supra note l; Fleming James, Jr., Tort Liability of Governmental Units and their Officers, 22 U. CHI. L. REV. 610 (1955); 57 AM. JUR. 2D Municipal, County, School, and State Tort Liability (1988); Haney v. City of Lexington, 386 S.W. 2d 738, 10 A.L.R 3d 1362 (1964); Note, Municipal Immunity in Ohio--How Much Wrong Can a Municipality Do?, 15 U. TOL. L. REV. 1559 (1984); Comment, Governmental Immunity in Illinois: The Molitor Decision and the Legislative Reaction, 54 NW. U.L. REV. 588 (1986); Comment, Legislative Waiver of Governmental Immunity, X LOY. L. REV. 105 (1959-1960).

18 57 AM.. JUR. 2D Municipal, County, School, and State Tort Liability § 19 (1988).

NCGIA Initiative 16 Paper

19 Id.

20 New York takes the position that the Court of Claims Act by implication makes units of local government liable for tort claims because that act renders the state so liable. See Court of Claims Act § 8 and note 58 thereunder. Hawaii has not enacted a tort claims act applicable to units of local government. While there have been several cases which have granted very limited immunity from suit in specific areas local governments in Hawaii are generally subject to liability for their torts. Telephone conversation with Larry Foster, Associate Dean of the University of Hawaii Law School, 21 September 1994.

21 57 AM. JUR. 2D Municipal. County, School, and State Tort Liability § 129 (1988).

22 ARK. CODE ANN. § 21-9-301 (1993); VT. STAT. ANN. tit. 29, § 1403 (1993); N.C. GEN. STAT. § 160A 485 (1993).

23 OHIO REV. CODE ANN § 2744.02 (Anderson 1992).

24 Harrigan v. City of Reno, 475 P.2d 94, at 95 (Nev. 1970); § 41-4-2B NMSA 1978 (1989 Repl.).

25 See, e.g. TENN. CODE ANN. 29-20-205(6) (1980).

26 The states granting complete immunity are as follows: Arkansas, California, Colorado, Delaware, Idaho, Illinois, Indiana, Louisiana, Maine, Nebraska, New Jersey, New Mexico, North Carolina, Oklahoma, Pennsylvania, South Dakota, Tennessee, Vermont, West Virginia, and Wyoming.

27 The states withholding general immunity from local governments are as follows: Alabama, Alaska, Arizona, Connecticut, Florida, Hawaii, Iowa, Kansas, Kentucky, Maryland, Massachusetts, Minnesota, Mississippi, Montana, Nevada, New Hampshire, New York North Dakota, Oregon, Rhode Island, South Carolina, and Washington.

28 The states withholding immunity from proprietary activities of local governments are as follows: Georgia, Michigan, Missouri, Ohio, Texas, Utah, Virginia, and Wisconsin.

29 A Pennsylvania case affords an example of the result if the defense is not pled properly. Petticord v. Joyce, 516 Pa. 35, 531 A.2d 1383 (1987) (defense raised for the first time on appeal; use of the defense denied). See generally Forsythe, Pleading the Defense of Local Agency Immunity in Pennsylvania: The Requirements of Rule 1030 New Matter and the Preliminary Objection, 27 DUQ. L. REV. 661 (1989).

30 Croswell and Ahner, Computing Standards and GIS: A Tutorial, PROCEEDINGS OF THE ANNUAL CONFERENCE OF THE URBAN AND REGIONAL INFORMATION SYSTEMS ASSOCIATION (VOL. 11 ) 88 (1990). See also STANDARDS COMMITTEE, URBAN AND REGIONAL INFORMATION SYSTEMS ASSOCIATION, CONCEPTS, ISSUES, AND DEVELOPMENTS ON GIS-RELATED STANDARDS: ANNOTATED BIBLIOGRAPHY (1994).

31 FEDERAL GEOGRAPHIC DATA COMMITTEE, CONTENT STANDARDS FOR SPATIAL METADATA (1993); FEDERAL GEOGRAPHIC DATA COMMITTEE, CADASTRAL DATA CONTENT STANDARD, VERSION 1.0 (1994).

32 ACSM-ASPRS Geographic Information Management System Committee, Multi-Purpose Geographic Database Guidelines for Local Governments, ACSM BULLETIN 42-50 (August 1989).

33 Lurie and Weiss, Computer Assisted Mistakes: Changing Standards of Professional Liability, 11 SOFTWARE LAW JOURNAL 283 (1988).

34 Article 2 of the Uniform Commercial Code applies to "goods." UCC § 2-102.

35 520 N.Y.S.2d 334 (N.Y. City Civ. Ct. 1987).

36 385 U.S. 374 (1967).

37 520 N.Y.S.2d at 339.

38 Id.

39 57 AM. JUR. 2D Municipal, County, School, and State Tort Liability § 87 (1988); Davis, supra note 1 at § 7. The Restatement, Second, Torts has rejected the governmental/proprietary distinction. See § 895C and comment e. thereto.

40 CONN. GEN. STAT. § 52-557n (a)(1)(B).

41 D. OSBOURNE AND T. GAEBLER, REINVENTING GOVERNMENT: HOW THE ENTREPRENEURIAL SPIRIT IS TRANSFORMING THE PUBLIC SECTOR (1992).