«Jonathan Haskel Imperial College Business School, CEPR, Ceriba, IZA and UK-IRC Keywords: copyright, IP, innovation, knowledge, investment, ...»
www.wipo.int/copyright/en/publications/pdf/copyright_pub_893.pdf There are a number of issues with this approach, both in general and particularly when trying to measure investment in artistic assets protected by IPRs. First, and most important, is that industries as defined by the SIC do not accurately measure upstream activity. For example, the measured output of banks includes the software written by financial service firms.
Measurement of just the software industry would miss this. Additionally, since the ownership rights of IPRs tend to be complex and cross industry boundaries, it is very difficult to identify the upstream from data categorised by the SIC. Whilst some cases are clear, there are a number on the edges, hence the discussion of ‘core’, ‘partial’, ‘interdependent’ and ‘nondedicated support’ industries (WIPO, 2003).
Second, and related to the first point, is that this approach captures industry activity that is not in fact upstream asset creation. For instance, using the example of television, industry output will include the production of short-lived goods such as news, which is not an asset in the National Accounts framework. Such issues are the root of the WIPO definitions discussed e.g. industries whose activities contribute to asset creation, with the remainder of output being other goods and services that do not function as assets. Of course in the case of television industry, output will also include downstream use (broadcast) as well as upstream creation.
Third, such an approach would require detailed data on the sources of revenue and the rights associated with each payment. It would also require a very careful treatment of international flows. The level of detail required would make obtaining such data and subsequent estimation difficult and problematic.
Fourth, whilst this method can be used to say something about the proportion of output or employment that is in some way linked to ‘creative output’, it cannot be used for an accurate analysis of the value or volume of artistic asset creation. Instead we need to consider the sector from a broader viewpoint than that provided by the SIC.
c. Licence payments: capital compensation An alternative method is to use data on the revenues that accrue to copyright owners through royalties or licence fee payments, often referred to as capital compensation or in the SNA terminology, ‘Operating Surplus’. The method exploits the competitive equilibrium relationship (see memo item at the bottom right of Figure 1), where it is assumed that at the margin the owner is prepared to invest up to the expected net present value (NPV) of future
revenues generated by the asset. Provided ownership is retained, those future revenues are largely received in the form of royalties or licence fees. If ownership is transferred (i.e. the asset is sold), the NPV of remaining future revenues will be reflected in the transaction cost, P(N)N. Data on such revenues, correctly cumulated according to the rights they allow the payee and the length of period to which they refer, and correctly discounted to retrieve the NPV, are a potential source for estimating the value of the artistic investment. Such data is partly held by Collecting Societies, who provide a centralised payments receiving house for particular artistic assets, and distribute those payments to the asset owners i.e. to the holders of IPRs.
Licence fees are paid by downstream users and flow to the owners of assets in the upstream.
Therefore fees and royalties are some component of downstream input costs and output, as shown in (4). The ONS exploit this relationship in their estimates for Books and Music, by assuming that a constant proportion of (partial) downstream output equates to the payments made for the use of original assets. They further assume that the sum of these payments (that flow as income to the upstream sector) are a proxy for the value of annual investment in asset creation. Deriving such a relationship relies on numerous assumptions regarding steady-state conditions, as discussed in more detail later.
More generally, in the case of licence fees it is important to think about the nature of the transaction and its purpose. This brings us to an important distinction emphasised by the OECD (2010). The OECD Handbook on IP measurement (OECD, 2010) states on page 15
“If the acquisition of a copy with a licence to use is purchased with regular payments over a multi-year contract and the licensee is judged to have acquired economic ownership of the copy, then it should be regarded as the acquisition of an asset. If regular payments are made for a licence to use without a long-term contract, then the payments are treated as payments for a service. If there is a large initial payment followed by a series of smaller payments in succeeding years, the initial payment is recorded as gross fixed capital formation and the succeeding payments as payments for a service”.
How does this fit into our framework? If we are estimating investment in the creation of originals as the expected NPV of revenues generated from the commercialisation of the asset, it is appropriate to use data on all multi-period fees. These payments are normally thought of as fees for the short-term use of the asset e.g. payments made by cinemas to a studio for the right to project a film original for say 3 months, and it is these that provide the return to the asset owner. In that case it is correct to think of those payments as rentals for services that show up in the data as intermediate consumption for the using firm, who are simply purchasing a short-term good or service. 7 The difficulty arises where those payments are for rights that last longer than 12 months. In practice, in the National Accounts this is treated as an investment by the using sector, since the user has acquired economic ownership of (some portion of) the asset. Consider the case of software, where there is investment by the creator, say Microsoft, and also investment by the user provided the software has a service life of at least one year. The producer is investing in the creation of an asset which it can commercialise and generate future revenues.
The user is also investing in a piece of capital to enhance its productive potential. This treatment, applied in the capitalisation of software, is not accepted by all, since some believe
that it is double-counting. We understand that concern, but summarise our view as follows:
- if the payment for rights are exclusive and permanent then that is simply transfer of ownership i.e. negative GFCF for the seller, and positive GFCF for the buyer
- if the payment is for long-lived or permanent rights, but they are not exclusive, then the owner has not ‘lost’ capital. But the buyer has acquired a good that will contribute to output for more than one year. This is investment in use by the buyer For artistic originals, the purchase of long-term exclusive asset rights to use in the production of other goods and services is rare in most cases, but where a licence is purchased for more than one year, additional investment can be allocated to the using industry. This is also consistent with our view on embedded originals which we shall discuss later. At present the UK National Accounts include no estimates for investment in the use of artistic originals.
Calculation of such a measure is difficult since it requires detailed data on all transactions to identify which payments were one-off or across multiple years, and whether or not they involved any transfer of economic ownership.
d. Proxy of upstream output using aggregate data As shown in Figure 1, it is possible to infer a proxy for P(N)N using data on aggregate industry sales, where the aggregate industry is some mix of up and downstream activity. That
Capital compensation, or operating surplus, is a component of value-added, but it is generated in the industry that is leasing the capital i.e. in the industry that owns the capital. In this example, film studios. Rental payments in the downstream cinema are not part of the value-added of cinemas, but they are a part of value-added for a more widely defined “film industry”.
P(G)G = P(N)N + P(Y)Y Where P(G)G are gross revenues, P(N)N are upstream revenues and P(Y)Y are downstream revenues, then it is in principle possible to take some proportion of P(G)G as representative of P(N)N. This is a variant of method b).
3.2 Example of Model using Film The following text provides a more detailed description of various components of output and inputs, which have been or can be used to study the output of the creative sector. These include those terms necessary for estimating investment, using the four broad approaches set out above. For simplicity and consistency we continue to use the example of Film Originals to help describe each term 1: Upstream input costs [Σ P(X)X] e.g. Costs of Movie Production These are the costs of factor and intermediate inputs to asset creation and are the basis for one of the two primary methods for measurement of GFCF in Artistic Originals, as recommended by both Eurostat and the OECD. It is also the basis of the method used by ONS to measure GFCF in Television Originals by UK Public Service Broadcasters (PSBs), and in Film Originals for a sample of UK production companies and funding bodies.. As mentioned in the discussion of our framework, data on input costs will not necessarily equate with the revenues data without some additional estimation or allowance for a monopolists mark-up, μ.
Put another way, use of data on upstream input costs alone implicitly assumes that μ=1.
Since, in practice, detailed data on capital compensation and intermediate inputs in the upstream sector(s) are rarely available, data on labour costs are sometimes multiplied by some factor (γ) to account for additional overheads. This is the method used by ONS in the estimation of own-account software GFCF (Chamberlin et al, 2007).
2: NPV of ΣP(R)R [Downstream rental payments] e.g. Rental payments made by cinema to owner of new movie This the basis of the other primary method for estimating GFCF in artistic originals as recommended by Eurostat and the OECD. As the owner of a unique IPR, upstream annual revenues are generated from royalties and licence fees paid by downstream users. The sum of those payments over the lifetime of the asset can be used to estimate the value of the asset by making use of the equilibrium condition set out in equation (2), which states that the value of investment equates with the NPV of licence fees and royalty payments that accrue to the asset owner. This is the basis of the method used by Soloveichik (2010) for Books and Music.
Note that this method is equivalent to direct measurement of P(N)N, provided data on rentals, their timing and the length of time to which they refer are sufficiently detailed, and that such an estimate will implicitly include μ, the mark-up received by the innovator due to its market power as the owner of a unique asset. Potential sources for part of such data are the transactions mediated by collecting societies and rights management agencies.
It is worth noting that the ONS also use a variant of this approach in calculations for Books and Music. As we discuss later, by applying steady-state conditions it can be shown that aggregate cross-sectional royalties can be used to approximate annual investment. To calculate aggregate cross-sectional royalties the ONS apply a royalty rate, λ, to downstream sales of copies of originals.
3: P(N)N [Upstream output] e.g. Sales of film production companies This is the output of the upstream or innovation sector, as defined in equation 1, that is, the revenues received by the asset owner but based either on data for direct market transactions or industry sales aggregates. Such revenues could be received in a number of ways. First the upstream could receive a rental for the one –off or short term use of the asset, for example the right to project a film at a cinema for one month.8 Second, it may receive a rental for a longer period, perhaps for the right to manufacture the DVD for the next five years. Third it may receive payment for the outright purchase for a component of the asset rights, such as a oneoff payment to use the movie logo on merchandise for perpetuity. Therefore accurate valuation requires that each payment is treated correctly and not simply summed. Any estimate of P(N)N based on revenues will implicitly include μ, the innovators mark-up over the costs of production reflecting its market power as a holder of a copyright protected asset.
Data on direct market transactions for artistic originals are hard to find and industry revenues provide no information on: the type of payment received (i.e. asset purchase or rental); the length of time those payments cover (i.e. more or less than 1 year); or the rights those payments provide access to (i.e. all rights or some rights, and whether or not they are exclusive). Since the upstream and downstream sectors are often present in the same industry/firm and the initial investment often takes place on the firm’s own-account, it is In practice, in film the owner (studio) receives the rental as a percentage of revenues generated by the cinema, rather than as a flat fee (Soloveichik, 2010).
extremely difficult to split industry output into those that revenues that accrue to the asset and those from sales of final output. 9 4: P(Y)Y [Downstream output] e.g. sales of cinemas Continuing with our example of Film, the innovator is the owner of the final asset. In practice this may be a studio which has outsourced production to a production company, but funds and owns the final asset. The using sector is made up of a collection of industries and subindustries including chiefly cinema projection but also DVDs, TV broadcasting; merchandise production etc. Accurate measurement of P(Y)Y would require data on the revenues generated from the commercialisation of artistic assets in all those industries, and in all countries, but only where they relate to UK-owned artistic assets. A component of those revenues, P(Y)Y, is the rental payments made for the use of IP, P(R)R, in this example the projection of the film or sale of some merchandise.