Legal rights in data

Practical Law UK Articles 5-504-1074 (Approx. 18 pages)

Legal rights in data

by Richard Kemp, Paul Hinton and Paul Garland, Kemp Little LLP

Data law and the data-centric world

Towards an analytical model for the data-centric world

It is now appropriate to speak of the data-centric world. In the commercial world and across many sectors, businesses are paying ever more attention to their own and others' data as a way of adding value to the organisation and conferring a competitive advantage. This, in turn, is focusing attention on legal rights in data across many business sectors, where there are an increasing number of disputes.

In legal terms, data is funny stuff, being effectively inert. It is more precise (and helpful) to speak of legal rights in relation to data, rather than legal rights in data. Those rights are intellectual property (IP) rights (copyright, database right, confidentiality, patents and trade marks) and contract rights (data rights). Each IP right has its own rules, and applying those rules to data leads to a complex, multi-layered analysis where the law is unsettled and uncertain.

This means that data is an area where contract is very often king, and so most businesses regulate access to data by means of a series of agreements. In addition to these data rights, different rights and obligations arise:

In the software, communications and platform infrastructure on which the data resides.
In the database architecture.
Through different types of regulation, for example, competition law, privacy and sector-specific rules.

Legal aspects of the data-centric world: towards an analytical model

Scope of this article

Central to this model are the rights that subsist in relation to data, which form the fourth layer of the model (labelled Data in the figure above) and are the subject of this article.

Accordingly, this paper focuses on data rights.

The article provides an overview of sectors in which data commercialisation plays a key role and then considers these rights individually. Finally, it discusses a number of practical data licensing hotspots.

Patent rights are not considered further here; the processes and systems by which commercialisable data is produced may themselves be patentable, for example, as software patents, and are outside this article's scope.

Other areas not addressed here, other than peripherally, are:

The legal aspects of database software and data architecture.
Information management and security.
The duties that are imposed on, and the related rights that may arise for, various classes of persons under statute and regulation (for example, competition law as it affects data businesses and regulation applying to specific sectors).
Data protection law in relation to personal data.

Business context

In August 2010, it was reported that UBS was being sued in the UK courts for allegedly copying articles from oil and gas research published by Energy Intelligence Group Inc. In November 2010, BGC Capital Partners launched proceedings in New York against Tullett Prebon claiming damages for unjust enrichment and unfair competition when Tullett Prebon had allegedly used US Treasury data produced by BGC in breach of contract.

These recent cases each involve assessments of the legal rights in the data concerned and highlight the topical nature and relevance of these issues. In order to put why these assessments matter into a broader business context it is worthwhile quickly overviewing a number of sectors where the commercialisation of data plays an important role. These are:

Market data in the financial services industry.
Fares data in the air transport industry.
Clinical outcomes data in healthcare.
Metadata in the internet and music industries.

Financial services industry: market data

Market data in the financial services industry continues to be the crucible from which much of the current case law on data rights emanates (see Legal rights in data). Current economic conditions are increasing pressures in the market place between data providers (who wish to extend revenues) and data users (who wish to cut costs). These pressures have led to an intensification of the legal skirmishing around who can do what with financial data, and that in turn focuses attention on underlying data rights.

By "market data", as used here, we mean computerised data relating to quote (pre-trade or price discovery) and trade (post-trade or price) data for various types of financial instruments, whether for a particular trade or transaction, as an index or otherwise aggregated. Market data is generally used to refer to real-time (virtually instantaneous) or delayed data, whereas reference data is more historic or static, although market data is increasingly used for all types, be it real-time, delayed, end of day, historic or static. Market data is generally considered separately from the technology and telecoms platform and infrastructure through which the data is generated and disseminated.

At its simplest, the market data represented by, for example, an on-exchange equity trade will be a message sent in a standard way and in a standard format recording the key elements of the trade, such as:

Identity of the buyer.
Seller and security.
Number of shares.
Price per share.
Time of bargain.

Market data is an area where there is plentiful information about market shape, spend and players, and as such it provides a good business context for legal analysis. A report from consultants Burton-Taylor estimated the global spend on market data (often quoted as being a typical bank's second largest expenditure item after headcount and payroll) to have been $22.7 billion in 2009.

The report shows, by geographical region (Americas, EMEA (Europe, the Middle East and Africa) and Asia) that spend across the three largest segments (investment management, equity sales and trading and fixed income) accounts for around 60% of the total of nine segments, and that in each of the three geographical regions the two largest data providers (Thomson Reuters and Bloomberg) account for over 50%.

The radical changes that the financial world is currently undergoing continue to increase the significance of market data. These changes are driven by technology and regulation. Technology is fuelling a race to ever-higher trading speeds through a combination of direct market access, algorithmic and high-frequency trading and lower latency, and therefore putting a high premium on the most recent pre and post-trade data.

Regulation, as it relates to trading data, creates competition between trading platforms and this competition encourages growth in both the sources and fragmentation of data, putting a premium on comprehensiveness and accuracy of data. Selling and buying the most timely, accurate and comprehensive data confers a competitive advantage throughout the financial-services world.

Fares data in the air transport industry

The global air transport industry (a $600 billion revenue sector supporting 30 million jobs) depends on data about air fares being made available to the widest worldwide audience. This can be through individual airlines' websites or through aggregation and distribution directly (or indirectly through Global Distribution Services) to intermediaries like Expedia and travel agents, and onwards to the individual or corporate passenger.

The whole data distribution mechanism works on the basis of commonly formatted and automated messages by means of a series of automated rules that are agreed through airline-industry standards bodies who are managed under the auspices of the International Air Transport Association. The data ownership and licensing issues that arise in relation to air fares data are complex, in part due to the tensions between the ownership of the underlying fares data and the open standards that predicate the form in which it is generated and disseminated, and in part due to the multiplicity of actors at each stage of the chain.

Healthcare: clinical outcomes data

In all branches of medicine, the computerisation, de-personalisation and aggregation of information about health outcomes create a huge, growing resource and knowledge base for use in future preventive medicine and treatment. For the data to be most useful it needs to be created according to common methods and defined processes and be searchable and usable in standard ways based on a common IT architecture. As regards the national cardiac collection, for example, rights might potentially arise in:

The clinical and other data at the various stages of creation (for example, after the procedure, when initially aggregated at that hospital and then as subsequently aggregated across a number of hospitals) and in its various forms:
- unencrypted/encrypted;
- anonymised or personal;
- raw or cleaned; and
- structured according to one or more taxonomies.
The non-data inputs, for example:
- database methodologies;
- taxonomies and architecture;
- dataset definition and content; and
- querying structures and mechanisms.
Non-data outputs, for example:
- reports;
- queries; and
- derived materials.
The IT platform level, especially the database software in which the database will reside.

The NHS structure removes many difficulties that would have arisen had the various actors been in the private sector because the Secretary of State or Department of Health provides unification as to ownership of these rights. Even so, as the data is so valuable to pharmaceutical companies (for treatment development) and IT companies (for predictive analytics and other software tools, and so on), the terms on which the data is made available are of critical importance for the future.

One could extend consideration of public sector data from the NHS to other government departments (it is now only a question of time before these large databases about private citizens can be joined up) and start to speak about the government's data estate in the same way as their property estate, especially in these more fiscally challenged times.

Metadata

In the internet world, metadata is most often included in web pages (as meta elements placed as tags in an HTML or XHTML document, for example) in order to enable search engines to categorise them correctly. Metadata is therefore key to accessing and using internet resources.

Equally, the recorded music industry, whose business patterns are changing fundamentally as a result of digitisation, is seeing the development of databases containing tens of millions of tracks. A track is the basic product unit of the music industry and the various rights in a particular track (the copyrights of the composer, author, publisher and record company and performers' rights) will be categorised in that track's metadata. Again, the metadata is key to accessing and using the tracks in the database, in this case for clearing (that is, obtaining necessary permissions for) the rights concerned when a track is downloaded to be played publicly.

In each case, whoever controls the metadata may be able to control access to the underlying resource (internet web page or music track) and so, in the market place, the metadata can become an asset capable of making money. This again raises questions about:

What rights arise in the metadata asset.
Who owns them.
How they can be commercialised.

Legal rights in data

Data is funny stuff in legal terms. Legally inert in itself, it is more accurate to speak of the different kinds of rights that may arise in relation to data, rather than simply of data rights. Data and information may, therefore:

Be a literary work and attract copyright protection.
Have the attributes of a database to attract database right (and also, confusingly, database copyright).
Have the quality of confidentiality to enable enforcement as a confidence.
In some circumstances, give rise to trade mark or passing off rights.

Copyright, database right, confidentiality and trade marks are enforceable as IP rights against the world (as a right in rem) in accordance with its own rules and requirements. This makes the data legal analysis multi-layered from the outset, as requirements can differ widely from right to right. For example:

Copyright is a formal right protecting a particular expression.
Database right has its own requirements.
Copyright and database right protection for databases can be mutually exclusive.

The analytical complexity is compounded against the backdrop of increasingly global data businesses as IP rights are primarily national rights, so the rules and requirements will vary not only from right-to-right but, for each right, from country-to-country.

This layered nature of the IP rights analysis in relation to data, together with the current unsettled and uncertain state of the law in the UK, means that data remains an area where contract is king in the business world. And contracts of course confer enforceable rights (rights in personam) on the parties to them (subject to competition law and other statute and public policy rules that cannot be excluded) but not against the whole world (rights in rem).

Generally in relation to data, therefore, the good news for IP rights (as rights in rem) is that they can be enforced against the whole world. The bad news is that they confer rights of uncertain scope which are expensive to enforce. For contracts rights (as right in personam), meanwhile, the good news is that they confer strong rights and the bad news that this is only against the contracting parties themselves and not against the whole world.

What follows is an analysis of these rights in relation to data.

Copyright

Initial practical points

Copyright does what it says on the tin: it protects an original work against copying without licence or permission. That is to say, it is a formal remedy, protecting expression but not the underlying idea.

There are a number of preliminary practical points about copyright to note:

When considering computerised data, don't forget that simple literary work copyright will subsist in related documentation such as written statements or descriptions (whether in hard or soft copy or as a web page), and technical, functional and user specifications and the like.
Among others, the well-known US case of Lowry v Legg Mason 271 F Supp 2d 737 (D Md 2003), where Legg Mason, the large US-based investment management company, was ordered in 2004 to pay $19.7 million in damages for unlawful coping of stock market analysis, underscores the need to guard against literary work copyright infringement through in-house awareness, education and communication.
Although outside the scope of this piece, computer software has for a generation been subsumed under literary copyright in English law.
Moral rights generally apply to literary work excluding software.
Since the end of 2003, a number of important copyright permitted acts (including research for private study) have been limited to non-commercial purposes, so copying third-party internet materials at work is likely to infringe.
The most commonly missed copyright point is still the basic one that copyright in works created in employment are owned by the employer automatically, by operation of law, but when created by a contractor will need to be assigned in writing for copyright to be owned by the engaging company.
The development of copyright law over the last hundred years has led to copyright being a particularly supple and flexible right, capable of assignment (legal transfer) and hence licensing by being sliced up by restricted act, by time, by geography and so on.
Future copyright may also be presently assigned.

Background: the old (pre-1998) law

Before 1998, when Council Directive 96/9/EC on the legal protection of databases (Database Directive) was implemented under English law (see Key elements), the copyright works relevant to data were literary copyright, which protected a table of data, and compilation copyright, which protected a collection of tables. Tables as literary copyright works were subject to the (low) standards of originality (effectively, not copied from elsewhere) and the degree of skill and labour required (roughly, what is worth copying is worth protecting). A collection of tables as a compilation attracted a further thin sliver of copyright in the aggregation over and above the copyright (if any) in any single table.

Therefore, taking as a market data example the on-exchange equity trade referred to above (see Financial services industry: market data), it is possible that pre-1998 copyright could have subsisted in:

The unpopulated message format as a literary work.
The populated message as a literary work.
A set of populated messages as a compilation.

A further example, this time from the recorded music industry, is label copy, consisting of the form or table setting out the particulars of a sound recording owned by a record company such as:

Title.
Track name.
Duration.
Composer.
Author.
Publisher.
Featured artist/musicians.
Record company.

In this example, copyright could have subsisted in:

The uncompleted form.
The completed form.
A set of completed label copy forms as a compilation.

Current law

It is worth mentioning the pre-1998 position to highlight the three areas of changes made on 1 January 1998 by the Database Directive:

A definition of database was introduced as a searchable collection of systematically or methodically arranged works, data or other materials.
A new right, database right, was introduced into English law (see Database right).
Copyright protection for databases was changed in three ways:
- literary work copyright for a table or compilation was expressly removed for databases;
- databases were included in the list of literary works in which copyright subsists; and
- the standard of originality for database copyright was changed from the traditional "not copied from elsewhere" standard to a requirement that the author's own intellectual creation had gone into selecting or arranging the database's contents.

Database copyright and Football Dataco

The first UK judgment to consider the new author's intellectual creation standard for database copyright was Football Dataco Ltd v Brittens Pools Ltd [2010] EWHC 841 (Ch), where Floyd J held that the fixture lists concerned were protected by database copyright, but no other type of copyright, and not by database right (see Key elements). The judge adopted a four-step test to determine whether database copyright subsisted:

"Identify the data which is collected and arranged in the database;

Analyse the work which goes into the creation of the database by collecting and arranging the data so identified, to isolate that work which is properly regarded as selection and arrangement;

Ask whether the work of selection and arrangement was the author's own intellectual creation and in particular whether it involved the author's judgment, taste or discretion;

Finally one should ask whether the work is quantitatively sufficient to attract copyright protection." (Football Dataco at paragraph 91.)

Applying these steps and having identified the data, Floyd J found that:

There was a relevant selection in the fixture lists.
The authors had applied judgment and discretion.
The work was quantitatively sufficient.

He accordingly found that the fixture lists were protected by database copyright.

He also held that the fixture lists failed to attract database right for essentially the same reasons as British Horseracing Board v William Hill Organisation (Case C-203/02) [2005] ECR I-10415 (BHB case) and Fixtures Marketing v Oy Veikkaus Ab (Case C-46/02) [2004] ECR I-10365; Fixtures Marketing Ltd v Svenska Spel AB (Case C-388/02) [2004] ECR I-10497 and Fixtures Marketing Ltd v Organismos Prognostikon Agonon Podosfairou AE (Case C-444/02) [2004] ECR I-10549 (Fixtures Marketing cases) (see BHB and the Fixtures Marketing cases).

However, the judge added:

"It seems to me that the only way in which the Fixture Lists could conceivably attract copyright is by virtue of the collection and arrangement of the data contained in them, that is to say as a database. It follows that I do not see any scope for the subsistence of copyright by any other route." (Football Dataco at paragraph 100.)

Football Dataco highlights the key difficulty about copyright in relation to data: the hurdles for database copyright are higher than for traditional literary copyright but a database is specifically excluded from literary copyright as a table or compilation. Therefore, a database may not qualify for copyright at all (by failing to get over the database copyright hurdles) but still be a database in the copyright sense, and so be incapable of attracting copyright by any other route.

On 9th December 2010, the UK Court of Appeal settled a number of questions in the case to be referred to the European Court of Justice, as follows:

"1. In Article 3(1) of Directive 96/9/EC on the legal protection of databases what is meant by "databases which, by reason of the selection or arrangement of their contents, constitute the author's own intellectual creation" and in particular:

(a) should the intellectual effort and skill of creating data be excluded;

(b) does "selection or arrangement" include adding important significance to a pre-existing item of data (as in fixing the date of a football match);

(c) does "author's own intellectual creation" require more than significant labour and skill from the author, if so what?

2. Does the Directive preclude national rights in the nature of copyright in databases other than those provided for by the Directive?" (Football Dataco and others v Yahoo! UK and others [2010] EWCA Civ 1380.)

Practical suggestions

We suggest that market participants wishing to build up their rights consider the following steps:

Claim and assert copyright (in contracts, by copyright notices, on websites and in other relevant documentation) in relation to data existing in, generated by or derived from, their systems and operations.
Make sure that the copyright position is considered as whole, taking into account literary copyright in all documents ancillary to or associated with the data.
In relevant cases, include language in methodologies and technical (and other relevant) specifications that articulate that relevant resulting datasets and databases and the data within them are the products of the company's own intellectual creation, following Football Dataco to maximise the chances of database copyright arising.
Take advantage of the suppleness and flexibility of copyright as to slicing (see Initial practical points).
Take effective contractual assignments of present and future copyright in all relevant contracts.

Database right

Key elements

Database right was introduced as a new right into English law on 1 January 1998 by the Database Directive, implementing EU legislation (see Background: the old (pre-1998) law). The key elements of database right are:

Subsistence: database right subsists in any database "if there has been a substantial investment in the obtaining, verifying or presentation of the contents of that database" where, at the time it was made, the maker was:
- a national of an EEA member state or habitually resident in the EEA; or
- a company or partnership with its central administration, registered office or principal place of business within the EEA.
Ownership: the first owner of database right is the maker, who is the person who takes the initiative in and assumes the risk of obtaining, verifying, or presenting its contents.
Duration: the first generation of a database right is of 15 years duration from the end of the year when the database was completed.
Infringement: database right is infringed if a person without the owner's consent "extracts or re-utilises all or a substantial part of the contents of the database" or carries out "repeated and systematic extraction or re-utilisation of insubstantial parts of the contents of a database". Extraction is defined as the "permanent or temporary transfer [of the contents] to another medium by any means or in any form" and re-utilisation is defined as "making … available to the public by any means".

BHB and the Fixtures Marketing cases

The scope of database right was narrowed by the judgments of the ECJ in a series of fixtures marketing cases in November 2004, which included the important BHB case. The effect of these cases is to make it more difficult to claim database right, broadly because:

It is more difficult to show that the right subsists.
Where the right does subsist, it is more difficult to show infringement.

In holding what was meant by obtaining, the ECJ decided that investing in creating the underlying materials themselves in the database was irrelevant for database right; only collecting or aggregating them to create the database was relevant for database right:

"[t]he expression 'investment in … the obtaining … of the contents' of a database in … [the Database Directive] must be understood to refer to the resources used to seek out existing independent materials and collect them in the database. It does not cover the resources used for the creation of materials which make up the contents of a database."

This part of the BHB and Fixtures Marketing cases raises particular hurdles, particularly for real time data providers and database developers where the database's content is effectively made at the same time as the database itself. In the world of market (particularly trade) data, this is particularly the case for on-book or on-exchange transactions where the data is effectively generated by the platform itself (that is, the operator of the exchange or other relevant platform).

For OTC (over the counter) and other off-book transactions, there would appear to be a better case to argue that the data is produced by the broker, member or customer, and that the platform operator's expense is genuinely in obtaining, verifying or presenting the data.

However, it is not the case that database right is without value. For example, it is generally believed that database right subsists in the taxonomy or classification systems for securities identifiers.

Where the right does subsist, it is more difficult to show infringement. This is effectively because the bar has been raised on what constitutes substantial and insubstantial in relation to the unauthorised extraction and re-utilisation of a database's contents.

Finally, on future rights, Regulation 23 of the Copyright and Database Right Regulations (SI 1997/3032) equates database right to copyright for the purposes of section 91 of the Copyright Designs and Patents Act 1988 and future copyright. Future database right is therefore like future copyright.

Practical suggestions

As a practical matter, market participants wishing to build up their rights should consider:

Continuing to claim and assert database right in and to data in relevant systems.
Amending processes to be able to demonstrate a substantial obtaining, verifying and presenting investment at the database as well as the contents stage.
Amending internal and externally available documentation (including possibly technical and other relevant specifications) to show this.
Taking appropriate (or amending current) contractual and noticed acknowledgements and/or commitments from contracting parties.

Confidentiality

General

The UK rules on the protection of confidentiality are likely to operate to provide a remedy in circumstances that are relevant to data. Unlike copyright and database right, which are formal remedies protecting information in the way in which it is displayed, the law of confidence can protect the substance of the information itself.

The series of cases starting with Prince Albert v Strange [1849] EWHC Ch J20 shows that, where the structure of information in aggregated form is not publicly available, then, even though underlying information may itself be in the public domain, the law of confidence will intervene to protect the information as aggregated. For example, in the wireline cases, which began with Exchange Telegraph Co Ltd v Gregory & Co [1896] 1 QB147, a number of which concerned exchange information, the information traded was essentially in the public domain but the UK court held that what was protectable in confidence terms was the structure of the information in its aggregated form.

Practical suggestions

As a practical matter, and somewhat counter-intuitively, confidentiality frequently emerges as offering the best protection outside contract (see Contract). Consideration should be given in an organisation's system of contracts and website and other notices to ensuring that data is expressly stated to be confidential. This will be likely to involve reviewing contractual definitions of confidential information to be sufficiently broad as to cover computerised data.

It will be necessary to show that the aggregated database information concerned (and preferably trade data within an individual trade report) is not freely publicly available, which in turn highlights the need for general confidentiality provisions in website and other notices and contractual terms and conditions. If it is impractical to do this for all data, then consider doing this for real-time and near real-time data on the basis that historic data inevitably holds less value and gets more widely disseminated.

Trade marks

General

Although none of the case law relates directly to the UK, trade mark law is one of the most developed areas in relation to data. This is strange, one might think, as the essential function of a trade mark is to protect the origin of goods and services and data itself is normally considered separately from the name of the service under which it is marketed.

However, the cases, which mainly relate to the index world, are instructive and are overviewed here. They emphasise the nature of IP rights as national in nature and scope and that a course of conduct which may be infringing in one country may be unimpeachable in another.

A second central point is that all trade mark systems recognise the need to limit the protection a registered trade mark gives its proprietor and that a trade mark owner will generally not be able to prevent a third party in honest use of its marks or using them in a descriptive sense.

The following cases show just how delicate it can be to strike the balance between enforcing a trade mark as a hallmark of origin and permitting honest or descriptive use.

US cases: futures

In the earliest of a series of cases, Standard & Poor's Corp v Commodity Exchange Inc 683 F 2d 704, 708 (2nd Cir 1982), Commodity Exchange (Comex) attempted to publish the value of the S&P 500 Composite Stock Price Index (S&P 500) as the Comex 500 Stock Index and to create, issue and trade futures contracts based on that index, that is, contracts to be settled in cash for the difference between the value of the S&P 500 at the time of the contract and at a stipulated future date.

Standard & Poor's (S&P) claimed under the US tort of misappropriation to the effect that its investment in the S&P 500 had been misappropriated by Comex. The Illinois District Court found for S&P, holding that Comex "intended improperly to misappropriate the property of the plaintiff and trade directly on … the S&P Index".

The following year, in Board of Trade of the City of Chicago v Dow Jones & Co 98 Ill 2d 109 (1983), the Chicago Board of Trade (CBOT) attempted to create a futures contract based on an index identical to the Dow Jones Industrial Average (DJIA). Dow Jones claimed successfully in the Illinois Supreme Court for misappropriation of its investment in the DJIA.

The S&P and CBOT cases are generally cited as authority in the US for the proposition that an index creator has a protectable property interest in a financial index and in its related trade marks where futures contracts based on the index are priced at the difference between contract and index values and settled in cash.

US cases: options

In Golden Nugget Inc v American Stock Exchange Inc 828 F 2d 586 (9th Cir 1987), the American Stock Exchange (ASE), without Golden Nugget's consent, issued, listed and traded options, that is, the right but not the obligation to buy (for call options) or sell (for put options) a specific amount of Golden Nugget's stock at a stipulated future date. Golden Nugget claimed that the ASE had misappropriated its property, engaged in unfair competition and infringed the Golden Nugget trade mark. The court found for the ASE, holding that Golden Nugget, having sold its shares to the public, had given up "any proprietary rights in the shares that would allow it to control the manner or means of resale".

US cases: exchange-traded funds

In NASDAQ Stock Market Inc v Archipelago Holdings LLC 336 F Supp 2d 294 (SDNY 2004), NASDAQ had created the NASDAQ-100 index and QQQ, an exchange-traded fund (ETF) that tracked the NASDAQ-100.

An ETF is a fund which mimics or tracks an index and whose shares can be traded on a stock exchange. The correlation between the fund and the index is achieved through establishing the fund as shares of the companies used to calculate the index, and weighted in the same way as in the index. Performance of the ETF shares therefore mimics or tracks the performance of the index.

Archipelago hosted and facilitated the trading of QQQ shares on its ArcaEx exchange without a licence from NASDAQ. NASDAQ claimed that by operating a market for trading QQQ shares, Archipelago was misappropriating NASDAQ's IP in the QQQ shares. The New York District Court rejected the claim, holding that regardless of any IP interest NASDAQ had in the QQQ fund, no such interest was infringed when an exchange provided a market place for the trading of shares NASDAQ had sold to the public. The court said:

"Once an index creator or its licensee has created a fund of securities tracking the index and has sold shares in the fund to the public, those who engage in secondary trading of the fund's shares, or provide a market place for such trading, do not infringe a property interest of the creator of the index and the fund."

These US cases came full circle with the July 2010 judgment in Chicago Board Options Exchange Inc and others v International Securities Exchange LLC (and others) 06CH24798, Cook County, Illinois Circuit Court, Chancery Division (Chicago). Here, the Chicago Board Options Exchange (CBOE) held the exclusive licence from Dow Jones and S&P to offer options based on the DJIA and the S&P 500 respectively. The International Securities Exchange (ISE) was looking, without a licence from Dow Jones or S&P, to offer standardised index options based on the DJIA and S&P 500, which would be cleared by Options Clearing Corp (OCC), the sole clearing agency for standardised index options. CBOE, Dow Jones and S&P launched proceedings to restrain the ISE and OCC.

Dow Jones and S&P alleged that their substantial investments of "resources, skill, judgment, creativity and efforts required to develop and maintain their indexes" had conferred on them proprietary interests in the DJIA and S&P 500 which gave them the exclusive right to authorise the "creation, issuance, listing, trading, clearing, and settlement of financial products, including index options, that are based on the underlying indexes".

They sought summary judgment on two bases:

The ISE's proposed actions would misappropriate Dow Jones' and S&P's proprietary interests in their indexes as well as CBOE's exclusive rights under its licences.
The ISE's proposed actions constituted unfair competition under Illinois common law.

In its analysis, the court reiterated its view that the plaintiffs' claims were not based on copying published index values from websites and other sources. Rather, it was the connection of ISE's proposed options to and association with the DJIA and S&P 500 that would allow ISE to exploit the plaintiffs' "research efforts, skills, expertise, reputation and goodwill" embodied in the indexes. Significantly, the plaintiffs did not complain of any copying or dissemination of the index values themselves and they appeared to concede that they might have no rights in them. The court clearly agreed with this analysis, referring to the index values as being "a matter of basic market fact".

The court considered a number of previous cases, including CBOT referred to above (see US cases: futures), which it stated was "on all fours" with this case, and held that the plaintiffs were entitled to protection of their rights in the indexes from the ISE's proposed use.

Cases from the rest of the world

In SSE Infonext v FXI (2006) Judgment of the Shanghai Pudong Court, the Shanghai Stock Exchange (SSE) had signed a one year agreement in May 2005 with FXI (a joint venture company between Xinhua Financial and FTSE International) giving FXI access to the SSE's real-time data for FXI's own use but prohibiting it from using that data to develop derivative products. FXI licensed the Singapore Stock Exchange (SGX) to develop the SGX/FTSE Xinhua China A 50 Index Future, which commenced trading in September 2006.

FXI did not provide SSE real-time data to the SGX but derived the China A50 Index Future from stocks including 38 real-time SSE stocks provided by SSE Infonext. SSE Infonext claimed against FXI for breach of contract and IP rights. The 2006 judgment of the Shanghai Pudong New Area District People's Court found for the plaintiff, holding that the May 2005 agreement had been breached by FXI and imposing a fine. The judgment was upheld on appeal.

In Commerzbank v Deutsche Börse AG, Federal Supreme Court, 2009 I ZR 42/ 07 DAX; OLG Frankfurt a M (Lexetius.com/2009, 2817), Deutsche Börse AG (DB) published the DAX (the German share index) and DivDAX (an index of the 15 companies included within the DAX with the highest dividend yield). DB owned the registered trade marks for DAX and DivDAX. Commerzbank issued option warrants linked to the DAX's value which it noted as "related to DAX®" also stating that "DAX® is a registered trademark of Deutsche Börse". Commerzbank separately issued option warrants linked to the DivDAX's value, describing those warrants as "Unlimited DivDAX® Index Certificate".

In 2006, Commerzbank terminated its licence agreement with DB following a price rise and applied to the German court for a declaration that it could use DB's indices as underlyings and refer to the index names without the need for DB's consent. DB opposed that application, seeking a ruling that the use of its marks was infringing.

At first instance, the Frankfurt District Court in its judgment of 26 July 2006 found in favour of DB, stating that Commerzbank imitated and unfairly exploited DB's reputation in both indices where it linked the value of warrants to the value of the index. Commerzbank appealed and the Frankfurt Appeal Court, in its judgment of 13 February 2007, dismissed the appeal in relation to DivDAX but allowed it for DAX. Both parties appealed to the Federal Supreme Court, which considered Commerzbank's use of DAX and DivDAX separately, holding, in relation to DAX, that the Court of Appeal had ruled correctly that there had been no infringement in using the mark "DAX" in relation to the option warrants issued by Commerzbank.

The court commented that Article 6 of Council Directive 89/104/EEC to approximate the laws of the member states relating to trade marks (Trade Mark Directive) and the relevant German trade mark legislation served to:

"reconcile the interests of trade mark protection and free movement of goods/services in such a way that the trademark right can play its role as an essential element of a system where there is genuine competition."

What was critical to the Supreme Court was that Commerzbank's use of the DAX mark was:

"restricted to the extent that is required to use the reference value [and] the Court of Appeal has established that the reference to the share index DAX is factual and informative in the form appertaining to the case without giving the impression that the security is issued by [DB] or that trade relations exist between the parties."

Based on these principles, the use remained within the limits of "common decency".

In relation to DivDAX, the court ruled that the Court of Appeal had also been correct in finding that the way in which Commerzbank had used the DivDAX mark was an infringement as it:

"offends against common decency within the meaning of Section 23 No 2 Trademark Law. Commerzbank unfairly violates the legitimate interests of [DB] as the registered trademark owner if it uses the logo protected by trademark law as a designation of origin for its security."

Although decided principally by reference to German trade mark legislation, this decision follows the same reasoning that would likely be applied by other European courts (based on the Trade Mark Directive) and is broadly comparable with the US decisions above.

English and EU law

Article 6(1)(b) of the Trade Mark Directive (enacted into all EU member states' law) provides that:

"a trade mark shall not entitle the proprietor to prohibit a third party from using in the course of trade … indications concerning the kind, quality, quantity, intended purpose, value … or other characteristics of goods or services … provided he uses them in accordance with honest practices in industrial or commercial matters."

In the UK, this finds its way into Section 10(6) of the Trade Marks Act 1994, which provides that:

"… the use of a registered trade mark by any person [shall not be prevented] for the purpose of identifying goods or services as those of the proprietor or a licensee.

But any such use otherwise than in accordance with honest practices in industrial or commercial matters shall be treated as infringing the registered trade mark if the use without due cause takes unfair advantage of, or is detrimental to, the distinctive character or repute of the trade mark."

Users of Index trade marks need to be careful that any use, without licence, is in a legitimate, descriptive sense and (if in the EU) in accordance with that country's version of honest practices. Drawing on futures, options and ETFs utilising branded indexes, the above cases highlight the sorts of issues that will be taken into account.

As a final point on trade marks and data, in many cases the need is for reliable, accurate data for the production of downstream instruments like futures, options and ETFs and, with the increasing importance of fast trading/low latency, the desirability of a service contract from the index creator or disseminator for data integrity purposes can, as a practical matter, easily trump the preference to avoid the need for a trade mark licence.

Contract

General

Finally, whether or not copyright, database right, rights of confidence or trade mark rights subsist in relation to any data, it is possible to impose obligations and confer rights relating to that data by agreement between contracting parties. The BHB litigation (see BHB and the Fixtures Marketing cases) spawned significant follow-on cases. In one of these, Attheraces v The British Horse Racing Board [2005] EWHC 3015 (Ch), it was argued that a data owner could not charge for data in the absence of IP rights. The judge, Etherton J, expressly rejected this saying:

"I agree with [the data supplier] that it is entitled, in principle, to impose a charge for use of its … data by, and for the benefit of, [users], whether or not [it] has IP rights in respect of the data, and, in particular, database rights under the Databases Directive and the Databases Regulations or copyright, and irrespective of the extent of any such rights.

[The data supplier] has, in the data, a valuable commodity, for which it is entitled to charge. There is no authority to the contrary, including the [BHB] case." (Attheraces at paragraph 285.)

At a time when the scope and extent of IP rights in relation to data is uncertain, this powerful dictum of Etherton J underscores that, in practice, contract is king. Contract's kingdom is, however, limited to the in personam world of rights and duties between contracting parties (extended to a limited extent by third party beneficiary rules), and it is complicated by contract IP (dressing up in an agreement as IP rights that depend on the contractual nexus between the parties for enforceability).

Common issues in data agreements

Whether characterised as access, service, supply or licence agreements, contracts for the provision of data across industry sectors tend to generate similar sorts of issues in practice. Without intending to provide an exhaustive list, four of the common ones here are set out below.

Competition law may (and frequently will) add further complications to a legal analysis. The golden rule, whether as a data provider or data user, is to provide expressly in the contract for the particular rights you want and for the particular obligations you want your contractual counterparty to accept, and not leave it to implied terms or rules of construction, particularly bearing in mind that the background contract and IP law is so unsettled.

Common issues in data agreements include:

Scope of licence. If the data provision agreement is structured as a licence, make sure (as licensor) that the agreement contains statements of the usual rules of construction regarding reservation of rights (for example, "all rights not expressly licensed are expressly reserved") and that the licence is personal to the licensee. Also expressly detail in each licence that is being conferred, for example:
- whether it is non-exclusive/sole/exclusive;
- whether there are any geographical or product restrictions on the data;
- whether or not the licence duration is coterminous with the agreement;
- whether the data is for internal only use or (more likely) for onward dissemination and, if so, how and on what terms;
- whether the licensee may or may not sub-license; and
- the treatment of derived and commingled data.
From the licensee's point of view, again make sure the contract covers expressly all the rights you need and for as long as you need them. Consider whether, although characterised as a licence (that is, a permission to do what IP law would stop you doing without the consent) it is effectively an access, service or supply agreement and, if so, does dressing it up as a licence put you in a worse position?
Derived data. Across many industry sectors, but particularly in the financial market data area, questions frequently arise about the extent to which a user or licensee may use the input data to create derived data (an obvious example is to take equity index data and create a chart or graph from it) and then who owns what rights in the derived data, particularly whether the supplier of the input data has a property interest in the derived data.
The issue of derived data is complicated by carrying a certain amount of baggage from copyright law, where derivative work is a defined term under US copyright law (17 USC § 101) but not English copyright law (section 17 of the Copyright, Designs and Patents Act 1988 refers to adaptation).
Again, the practical answer is to provide in the contract (or a policy document incorporated into the contract) expressly for what happens in relation to derived data. In the absence of express terms, it will be necessary to assess each of the rights that may apply (in the index/chart example this will be copyright, database right and confidentiality) and how each may affect the position. Then, following Investors Compensation Scheme v West Bromwich [1997] UKHL 28, it is necessary to adopt a purposive approach and construe the relationship first, and the words second.
Commingling. Commingling data is like deriving data, but with the user taking input data from more than one supplier or licensor and creating something different with the commingled data (the analogy would be red, blue and yellow water being piped into a swimming pool, which turns green). Here, again, the answer is for each contract to cover expressly what is to happen, although that will present practical difficulties. For the supplier or licensor who wants to claim an interest in the downstream (that is, commingled) data in the absence of an express entitlement, the problem is that copyright, as a formal remedy, is unlikely to help and database right is of limited value, as we have seen. Confidentiality may, however, provide some assistance.
Post-term use. The most frequently encountered problem area is what happens on termination of the contract as regards post-term use where the contract is silent. At one end of the spectrum, can the user or licensee continue to use the data supplied up until termination in the same way as when the contract was in force, or, at the other end of the spectrum, does it have to expunge or purge all the data from its systems?
Again, this should be covered expressly in the contract. This is because the relevant background law is not without difficulty when the contract is silent. This makes it much more problematic to foresee in any particular case what the outcome of any claim would be. Relevant areas of background law include:
- the construction of the contract (see the West Bromwich case);
- the implication of terms; and
- the applicability of section 3(2)(b) of the Unfair Contract Terms Act 1977.

Practical suggestions

A number of suggestions relating to contracts have been made above in relation to copyright, database right and confidentiality. Essentially, consideration should be given in contracts to:

Data being defined in all appropriate contracts to cover all data that is relevant.
Confidential information being defined to cover data.
Taking appropriate assignments of present and (where possible) future rights in the contracts concerned.
For data creators and providers:
- taking express acknowledgements from contract counterparties to the effect that relevant rights subsist in the data and are owned by the data creator or provider;
- expressly requiring the counterparty to undertake to take, allow or suffer no act inconsistent with rights of the data creator or provider under the agreement(s) concerned; and
- dovetailing all relevant contractual pieces to ensure that there are no gaps.

An unsettled road ahead for rights in the data-centric world

In today's commercial world, with its accent on cost control as much as revenue growth, the debate about the value of timely, accurate and comprehensive data is becoming more heated in the financial services and other sectors, where data expenses represent a sizeable input. Data users are under market pressure to access the most valuable data as cheaply as possible while data providers seek to maximise their data revenues. The combination of data's increasing centrality and a tougher business environment is leading to growing pressures as data users and providers battle in the market place to command the best data at the best prices. Despite legal complexities, these pressures are leading to greater skirmishing and beginning to play themselves out in more direct and open legal confrontation.

These pressures are also leading to greater legal analysis as to what rights subsist in relation to data in any given case. Just as these pressures being to mount, however, the state of the law on data rights is getting more, not less, uncertain and unsettled, both as to IP (where in the UK, as shown above, copyright, database right, patents, confidentiality and trade marks all have important limitations as to their effectiveness in the data-centric world) and as to contract (where competition law constraints are starting to impact the enforceability of terms previously considered to be industry standard).

IP type rights offer the benefit of enforceability in rem against the whole world, but in 2011 they are characterised by the following real difficulties:

The analysis of each right depends on its own rules, making the analysis multi-layered.
Within any single country the scope and extent of each right are uncertain.
In the international context, legal outcomes may differ between countries. For example, for the same fact pattern, copyright or trade mark rights may be enforceable in one country but not another.
In a contractual context, what may look like IP rights are in fact contractual rights and obligations.

As already highlighted, this means that contract continues to be of paramount importance: contract is indeed king in the data-centric world.

End of Document

Also Found In

Resource ID 5-504-1074

Published on 25-Jan-2011

Resource Type Articles

Jurisdictions

England
Wales