The OECD Project – Opening Remarks
John Gordon
I am pleased to see that that the subject of culture statistics has not lost its allure. Many of you have travelled considerable distances and we appreciate you interest and willingness to share your collective experiences here.
Several who had hoped to be here are absent for a variety of reasons. Some are attending another meeting on culture being hosted by the European Commission. One delegate from Chile found her flight oversold and was unable find an alternate flight. One of my former colleagues from Canada found herself in a similar position but this time weather was the culprit, not overzealous ticket selling.
I know that for many of you this is but one of a series of meetings on this subject while for othersthis may be the beginning of something new. Certainly the subject matter,from an international point of viewgoes back more that 30 years.
Simon has referred to the UNESCO Framework for Culture Statistics developed in the 1980s and you will surly hear more about it in other presentations and discussions. Although the FCS was far from perfect, it did providethe underpinning for a lot of the work that happened subsequently.
The current project to revisit and revise the FCS will surely also have ramifications on future work, but for the moment,the review is still in relatively early stages and it may be prudent to await its completion before attempting to finalize a new framework.
It may not even be reasonable to believe that we will ever arrive at a single framework that will perfectly meet everyone’s needs, but I think it’s important to work collaborativelytowards that end. It may be that we will end up with a model that has a core and a series of expanding rings around it. Hopefully we will eventually get to the stage where can at least agree on a core area and then work at expanding this core to encompass additional areas of interest.
Already from what Simon has shown here, we can still see the shadows of the original FCS but obviously there will be significant changes as they hammer out the details. We look forward to a continuing cooperation with Simon and the rest of UNESCO. The final results of should be of great interest to all of us.
Marta provided some background on the Eurostat Leadership Group (or LEG) report of 2000. The project certainly had an effect on European statistical work and there are several present who worked on the project.Again I am certain that the intelligence gained from that work will also find its way into our discussions.
These are but two of many milestones; there is also a lot of other work being done by individual countries plus a multitude of observatories and other groups.
With all these active players, what is the role of the OECD in the field? A fair question. By the end of tomorrow, I hope that this workshop will be in a position to make recommendations to the OECD on precisely that subject.
======
Not long after I joined the OECD last May, I made a short presentationabout our project to an international conferenceand spoke of some the work to date. When I ask for questions at the end of the session, one of the first was from a veteran in this field. The question was short and to the point. John, are we going to start the process all over again?
The answer is, I hope, no.
There is a good foundation here, some of it laid by persons in this very room, and I believe in building on the strengths of those who have gone before.
We began our work by looking at the work done by UNESCO and Eurostat; we reviewed the work already undertaken by a variety of countries to learn from their experiences. Our initial mandate was to produce some comparable measures for a sample of five OECD countries in a relatively short timeframe.
I have lived through discussions and debates about defining culture and I knew that this was not the place to start over again if we were to meet our timetable. However, I also knew that we needed some sort of framework or the comparability target would certainly be missed
Based on collective experience of other researchers, it seemed that there were several principles that we should keep in mind.
First and foremost, if there is to be any hope of comparability over time or jurisdictions,a framework needs to be clear, comprehensive, and detailed.
And although experience has shown that the process involves many challenges, an OECD framework should, as much as possible, be compatible with international frameworks being used by others such as UNESCO, Eurostat, WIPO, etc.
It should also be comprehensive and should incorporate all the various aspects for culture:
- Industry
- Occupation
- Goods and services
- Government expenditures
- Consumer expenditures
Those of you who remember the material in our draft paper will recognize that these categories coincide with classification standards.
Using classification standards also fits into my belief of building on the work of others. A lot of surveys conducted by national statistical organizations are designed according to such standards and the resulting data are often published according to these standards. But as we all know, classification standards are not a magic solution to all the data problems in the culture sector.
For one thing, a lot of cultural activity is secondary as opposed to primary in nature. This is a significant factor in measuring economic activity. One only has to consider that design is perhaps the most crucial aspect of launching any new car model but the activities of such an automotive design centre rarely, if ever, can be isolated from the published industry statistics.
Similarly, many artists find that the fruits of their passion are not sufficient to permit a reasonable standard of living and they find them selves working in other jobs that often take up more time and generate more revenue that their art. Once again, it can be difficult to identify the culture portion of their work in official statistics.
Many of these particular problems can be overcome if both primary and secondary activity is coded and reported but my experience has shown that coding of secondary activity does happen but is far from being the norm.
Sometimes the problems are not always immediately evident. Culture industries, as defined by classification standards are often much less homogeneous than many others, and so, as a result, small but possibly very culturally significant and unique pockets of activity can be missed when data is collected by surveys using relatively small samples.
Volunteer activity, a very significant factor in the culture sector, sometimes gets missed or even overlooked due a focus on economic measures.
Even if all the foregoing problems could be overcome, there is still the fact that culture is notusually well reflected in most current versions of classification standards. I expect that this workshop will provide some recommendations on dealing with these concerns.
======
Like most of you, waiting for the perfect situation is not arealistic option for the OECD. But rather than use the value-laden terms of definition and framework, we chose to talk about the scope of the current project with the intention to leave a formal framework adoption to a later date.
Because our initial direction was to work from official member country data sources, the availability of data was a factor in determining the scope.
Here is the list of sub-sectors that we ended up using. Libraries, museums and heritage sites and electronic games were put at the bottom because we were not able to obtain full comparability for these categories
Not surprisingly, the basic categories strongly resemble the original UNESCO FCS. We did exclude the two final categories of sport and environment since only some jurisdictions include sport in their culture statistics and none of our initial key countries included environment.
Although the LEG decided to include a sub-sector they called multimedia, we chose not to include it mostly for the practical reason that data were not easily available in most of our countries. It’s also an area where definitions vary considerably from one jurisdiction to another.
We did retain advertising and fashion although the LEG chose to exclude them.
======
As I alluded earlier, when dealing with international comparative measures of culture, one of the main challenges we faced, and one that seems to come out in almost every study, is that of comparability, or more precisely, the lack of true comparability.
Here is an updated version of a slide that was presented at the conference I referred to earlier. Aninitially impressive slide . . . until you begin to wonder about what that little extraneous mark is on the screen. I referred to it as that pesky little asterisk.
The story behind that asterisk is that when you begin to look at how the table was created you begin to discover that the concept and definitions of ‘culture’ are not all the same in the various countries. This is not to say that the table has no worth, I certainly don’t think that’s the case, but if you ask me how accurate it is, I can’t give you a definitive answer either.
I know that we, Helen in particular, worked very hard to make the numbers comparable, but as you know, when the details are not available in the original data, there is only so much you can do. Helen will talk a little more about this shortly
I’m not going to bore you buy reviewing all the various studies that have been published on culture statistics, but let me draw from one of the most recent. This comes from “The Economy of Culture in Europe” prepared for the European Commission and published last month.
“A strategic approach to the culture sector needs to be informed by the development of appropriate statistical tools and indicators at both national and European levels. . . In Europe, in the framework of statistical systems currently implemented, the statistical categorisations are not adapted to cultural activities and occupations. In addition, data generally provided by national statistical institutes either do not offer the level of details required, or are not available at all.
They use this to frame their first recommendation Establish a strong quantitative evidence base for policy makers.
Sounds familiar
Our first session this afternoon will look at the construction of indicators, predominately those relating to the social impact of culture. Although the OECD is more known for its activities in the economic aspects of societies, it also has an interest in the social side of the equation and our chief statistician referred to the wellbeing agenda in his earlier remarks.
Here is a quote from our draft paper which was inspired by a conversation with a fellow Canadian Paul Schafer. “Economic outcomes are not why most people become involved in culture and, therefore, economic indicators alone cannot be expected to provide exhaustive measures of the benefits cultural involvement brings to individuals and to the societies formed by these individuals.”
In the end, I’m not yet sure how the social side will be reflected in the suite of culture measures that will be retained by the OECD but I feel that it is absolutely necessary that we consider a place for them.
Regardless of which measures are retained, I think that it is important, at least from the point of view of the OECD member countries that the measures meet at least three criteria:
- They must be measurable. That is to say that the underlying data required to produce the measure must be available or, if not presently available, there must be a practical methodology available to obtain the required data.
- Comparisons at the international level must be meaningful, and of course, the measures must be truly comparable.
- These comparisons should be useful to policy makers at the national level.
On this last point I would allow a certain amount of leeway because experience has shown that policy makers have not always realized the utility of some proposedmeasures when they were being discussed in the abstract, but once they began to have access to them, they soon discovered their value.
Much of the body of work already in place comes close to meeting these criteria, and all of these criteria are not necessarily applicable for all work on the measurement of the culture sector, but if international evaluation is part of the equation, we always seem to come back to the subject of comparability.
One of the keys to comparability is clarity.
If I were asked to highlight the most important lesson that I have learned about surveys, data and statistics over my career it would be this -- clarity of purpose.
The data gatherer must both clearly understand what the question is that needs to be answered and what the data are that are required to provide the answer. But this not sufficient in and of itself.
The data provider must also clearly understand what data are being requested.
And there is a crucial third element;both parties must share the same understanding. If any of these elements is missing, the outcome of the exercise will surely disappoint at least one of the participants.
We achieve the one aspect of this equation by having a clearly defined frame of reference; the shared understanding is achieved by having clear definitions that mean the same thing to both parties. When running an industry survey, this means asking the questions using terminology familiar to the respondent. When trying to assemble data from secondary sources such as national statistical offices, the “industry jargon” is usually national classification standards.
To overcome differences among the variety of national standards, international standards were created. This is a laudable approach but in our case, although it does point us in the right direction, it does not completely solve our problems. In the first place, the premise that all national standards are compatible with international standards is still far from true, certainly not at the detailed level where most culture classes are found. And secondly, most standards were originally developed at a time when the service sector was a much less importantpart of societies and, as I mentioned earlier, culture, which is predominately found in the service sector, is not usually well reflected in the standards
Many have called for revisions to these standards -- and they are right -- but the revision cycles are extremely long and the process complex. The OECD works a lot with these standards and this is perhaps an area where the organization can play a role.
However, most of us have stakeholders or clients that cannot wait for these processes to come to fruition. One approach is to create new standards that do serve culture well. This is an approach that Australia followed. The Canadian province of Québec has also taken a similar approach and one of the presentations this afternoon will give you an insight into their process.
Many of us end up working, to a greater or lesser extent, with data collected by some other agency or group and we do not have the luxury of applying our own standards a priori. One of the presentations this afternoon will show how the UK dealt with this problem by developing allocation factors to extract the culture portion from mixed classes.
As part of our discussions in this workshop, I would like to explore the possibility of developing generic methodologies for creating these required allocation factors. I say generic because the intent would be to develop a methodology that would be equally valid in different jurisdictions.
We will also hear about work done in the field of trade statistics and the trials and tribulations of pulling together data on government expenditures on culture.
Although the revision cycles of international standards are long, chances do come along eventually and such is the case with ISCO. The ILO is not undertaking a full revision but rather what they call and update. Nevertheless there is a window of opportunity open before us and we have been asked to provide recommendations to the ILO on suggested changes to those occupations within the culture sector. We will spend some time tomorrow working on such a recommendation.
Finally, if you have succeeded in partitioning in all your mixed classes, the next step may be to move towards a full satellite account for culture. Finland has taken this step and its account comes into being next January 1. We will hearabout the account from both a development perspective as well as from a policy user’s perspective.
Chile also has the taken this step; Columbia and Mexico have indicated that they are planning to follow suit. The work on satellite accounts in Latin America has been coordinated and supported by the intergovernmental organization Convenio Andrés Bello. A representative from the organization,along with colleagues from Chile and Mexico had hoped to attend this workshop but were unable to do so for a variety of reasons. Both Chile and the Convenio Andrés Bello have, however offered theirsupport if the workshop recommends the setting up of an expert group on satellite accounts.