Official Report 423KB pdf
Item 2 is an evidence session on the committee’s inquiry into Scotland’s screen sector, focusing on research, statistics and value. This will be our final evidence session with stakeholders before we hear from Creative Scotland and the Cabinet Secretary for Culture, Tourism and External Affairs next Thursday. The committee then intends to publish its report before the summer recess.
I welcome our witnesses. We have Dr Michael Franklin of the Institute for Creative and Cultural Entrepreneurship at Goldsmiths college, the University of London; Andrew Barnes, associate director of Olsberg SPI; Alex Tosta, research manager in the research and statistics unit at the British Film Institute; and Inge Sorensen, lecturer in digital economy and culture at the University of Glasgow.
I will begin by asking Mr Barnes some specific questions, because I know that his consultancy, Olsberg SPI, was involved in preparing some of the data on which the screen unit collaborative proposal is based. Is that correct?
Yes.
On the underlying data about the screen sector in Scotland, what is your feeling about what you had to work with in preparing the data for the report?
The overall impression is that there is a lot of data out there, but we are perhaps lacking coherent data—a number of data sets that align and can be used in a single fashion. As you will note from the report that we wrote, we had to put in a range of findings for turnover, gross value added and full-time equivalent employee numbers on the basis that we could not be sure from the data that we would not double count. There are a variety of reasons for that. I do not know whether you want me to go into those at this stage or whether you have a follow-up question.
It is quite a technical issue for many of us, but some of the comments in your report leap out. One of those is about the risk of double counting, which you have addressed in some of your figures. Are you satisfied that the data does not contain any double counting?
We can be satisfied that the lower-end estimate does not contain any double counting, but the risk then is of undercounting. From the available data, we could not—at least not without a significant amount of granular research, which was beyond the scope of the project—identify precisely what production company did what in Scotland. The use of the Office for National Statistics-level statistics and Companies House filings on which companies are in which standard industrial classification code got us a certain amount of the way. The production spend data from Creative Scotland got us a certain amount of the way in the other direction. What we cannot tell with any precision is where those overlap in the middle and the degree of that overlap.
Has anyone ever questioned the data?
Not to my knowledge.
What implications do the concerns that you have pointed out have for the ambitious targets that the screen unit sets for increasing production spend?
The major challenge that you face as a nation in increasing production spend is to be able to identify the degree to which you have increased production spend and to accurately measure the impact of that at a future stage. We work across many countries starting at a much lower base level than that in Scotland in implementing screen support systems. We always tell those countries that including data collection provisions as part of an investment is required in order to be able to accurately determine the impact of that investment.
In Scotland, a certain degree of that exists, but it has to be asked how one puts in place a data collection methodology that makes it possible for the impact of the investments to be evaluated against the targets, and to track that back to the starting position. Unfortunately, the first part of that might not be possible, but the second part is absolutely critical.
Are you satisfied that we have a robust process in place in Scotland?
I have not seen any data on the process for the screen unit proposal, so I cannot answer that.
Okay. Before we move on, I want to ask a specific question about your data. You mentioned that you included cinemas with non-Scotland domiciled owners—I assume that you were referring to multiscreen cinemas. Therefore, that data went into the overall employment and investment figures for the screen sector in Scotland. For the purposes of our inquiry, the average multiscreen is not what we are talking about when we talk about boosting Scotland’s screen sector. Why was the data on such cinemas put in?
That was put in following discussion with our clients. It was felt that, without that data, the overall report would provide an underestimate of the impact of that part of the screen sector. Even if the beneficial owner of such a cinema is not domiciled in Scotland, the wages that it pays to Scottish residents and employees still have a downstream impact on the Scottish economy. It is a tricky balance to strike. That is why we disaggregated the information in the way that we did. We wanted to make it clear which bit related to Scotland-domiciled companies with a Scottish tax base and which bit related only to Scottish employees who would receive some benefit from that employment.
So Creative Scotland asked you to include that information; you were not initially inclined to do so.
I cannot remember which of our clients—Creative Scotland, Skills Development Scotland or Scottish Enterprise—asked us to do that. I would have to look back through my notes to find that out.
Thank you very much. I hand over to Claire Baker.
It might be helpful if the panel could outline what areas of data are to be prioritised. Mr Barnes has talked only about the economic impact, but we have had submissions on audience participation and the softer value of the sector. Where are there gaps in Scotland’s knowledge? In which areas do we need to increase the data?
I can tell you how the BFI would approach such an issue. First, we would identify the outcome that we wanted to achieve from any research or data collection. We would use that to identify the key and critical questions that needed to be answered, and we would follow that up by identifying what data was already available and what data needed to be accessed. It depends on what your priority is for data collection and any additional research that needs to be carried out. I do not know whether—
I am sorry to interrupt. I do not know whether the panel has had a chance to look in detail at the screen unit proposals—a fairly lengthy and quite technical document has been published. Are the intended outcomes clear enough? As you said, that would lead on to what information would need to be collected to reach that stage. Do you agree with the priorities that have been identified? Do you think that they are the right ones?
Because I look at data on a United Kingdom level, I know that one of the limitations with any data is that there is not sufficient granularity to go down to four-nation level. From what I saw in the documentation, that is one of the priorities. There would need to be further investigation of how that data could be obtained.
Those are really good points. The interrelation between the data sets is extremely important. As has been pointed out, at the UK level there are certain things that every funder of films across the UK would like to know. There are also specific questions that need to be asked to provide the necessary granularity at the Scottish level.
09:15However, any such system should be set up in such a way as to be part and parcel of the general business, as someone said. Although there should be people with expertise in statistics and people who do research, the data should be part and parcel of what happens, such as the allocation of funding or the operation of the exhibitions sector.
I am talking about an approach whereby all the data feed through as part and parcel of what is going on, as opposed to one in which every three or four years people have to think, “Are we doing it right?” There should be a continuous, on-going process, and it should be linked to what goes on at the BFI—and also at European level. The role of teamwork, especially in ensuring that data can be mapped across different areas, is important.
In your submission you mention the work of the University of Edinburgh. Are close enough collaborations in place, or do collaborations need to be more developed?
They absolutely need to be more developed in both sectors—of course I say that, as an academic. Three things that are happening right now are really important—you will have seen that in the creative industries sector review at Westminster. There are proposals to make links with bids for Arts and Humanities Research Council funding. An interesting proposal from the University of Glasgow, which Dr Sorensen can talk about, is through to the final round.
There is also a bid from the University of Edinburgh, which is to do with data-oriented creative industries analysis, and although film is not particularly mentioned in the bid, a great link could be made in that regard. There is expertise here that should be used and there is a great confluence of interests that could be maximised, especially in the context of the work that Professor Speed is doing on blockchain, for example. It would be a win-win for everybody.
It is important to define the term “screen sector”. It was interesting that the committee mentioned Netflix in its interim report. The collaborative proposal is quite media-centric and focused on film and TV, but the screen sector is many things—games, virtual reality, mobile and web content—and it is important that those things are also measured in a Scottish context.
I appreciate why the proposal needs to be film and TV-centric just now, but if we are to future proof the new screen unit and the development of the Scottish industry, it will be important to take a holistic view of what “screen” is.
Coherence across the four nations is particularly important. My company is in the process of finalising a revised economic impact study of the UK’s screen sector tax reliefs. At the outset of the process, one of the aims was to try to identify, to the greatest extent possible, on a national and regional basis, the breakdown of UK-wide production spend and impact. The data did not allow for that. Although Northern Ireland Screen, Creative Scotland and the Welsh agencies collect data, they all do so in a slightly different way, which leads to a lack of coherence.
If we are to be able to identify not just the impact in Scotland as a nation but how Scotland compares to other nations in the UK that are identified in the screen sector proposal, it is important to have a degree of coherence across UK-wide data gathering that allows for such comparisons to be drawn.
Skills Development Scotland told us:
“Skills Development Scotland, in partnership with Creative Scotland, has funded and commissioned a research programme to look at exactly what is going on within the company base of the screen sector and, much more significantly, what is going on within Scotland’s freelance workforce.”—[Official Report, Culture, Tourism, Europe and External Relations Committee, 8 March 2018; c 21.]
What do the witnesses hope will come from that work? Other people told us that they think that an annual survey should be carried out. What are your views on that?
Interesting things are going on in the freelance workforce survey space at the moment. Creative Skillset has recently started to redo research. In our report, we suggested that it had not done some for a while, which was correct at the time. That work might have some value. To avoid the double use of resources, it would be helpful to ensure that any work that is done aligns with what Creative Skillset is doing at the UK-wide level. Historically, it has broken things down into nations and regions.
The freelance data is always a particularly difficult area. Although a lot of Scottish freelancers work in the UK’s film production and television production sectors, the question of where they ordinarily work is key. That goes back to the question of the definition of what you are trying to achieve as a policy objective. Is it having Scottish workers, workers who are based in Scotland, or workers who ordinarily work in Scotland? Those could be three different things.
The question of how we achieve granularity in the data to allow us to identify the Scottish freelance workforce, however we choose to define it, is key. To be honest, I have never seen anyone answer that question, and I do not know that there is a straightforward answer to it.
From reading the evidence, it seems that there is quite a lot of missing data on freelance workers. We have read in the evidence that we have received that they have not seen that it has been up to them to take part in some of the other reviews that have been done. Do you think that the review in question will be able to identify gaps in the skills requirements of companies in the sector?
It should be able to do so to some degree, at least at a high level. If you are looking at which departments in a production sector have gaps, the line producers and producers will be able to identify where they have difficulties hiring. To get a breakdown to exact grades so that we know exactly what level of seniority is being looked for might prove to be a bit more difficult, because the industry is very fluid. People in it move around a lot for work, and what is true one day might not be entirely true the next day if a different production comes up and takes somebody from Scotland to somewhere else in the world.
As I have said, it is a tricky challenge. I noticed that there was something in one proposal about using the workforce databases. That is always a challenge, because the most senior members of crew do not like to go on to them, as they see doing so as a sign that they cannot get work in any other way. The industry is such a word-of-mouth one that they would probably reject that approach. Finding a way to identify where the gaps are at a high level should be able to be done, but I suspect that it is only when people start to work and put courses together that the really granular stuff will become more apparent.
Should there be an annual survey to be able to build data?
That would probably be helpful, but there will also have to be a certain degree of qualitative rather than purely quantitative data gathering as part of that.
Before I move on to a specific question about production spend, I want to ask about a frustration that we have come across a number of times from the industries. There is a sense that wheels are being reinvented that do not have to be. Can anyone cite examples of public body screen unit equivalents elsewhere that are effectively marshalling or co-ordinating the data collection that is required to grow and sustain industries?
As members know, I work in the research and statistics unit at the BFI. We have an extensive database that helps us to provide what is in essence a compendium of all statistics across a range of areas in film, which is in our statistical yearbook. Within that, we have statistics on not just the value chain of box office through distribution et cetera but on audiences, education and film economy, which relates to the GVA of film in the UK and levels of employment. We also look at public investment, which goes beyond the tax relief to see who are the other public funders of film in the UK.
The issue for the committee is that I provide data at the UK level, but it is obvious that the committee wants data at a Scottish level. A lot of the time, such data is not easily available. As the committee is about to go on to the issue of production, I will use that as an example. We track all productions of films in the UK, but it is difficult to get data to identify where the spending in the UK is being done. The nearest that we get to that is getting data on the location of shoots. For example, we will know that a production is shot in Scotland, but we will not know for how long or how much money the production is spending in Scotland. That is just one of the issues.
To go back to what I said originally, quite a comprehensive range of data is already freely available at UK level. I know that you want it at a granular level, but we are, unfortunately, limited by the data that is available. To go back to production spend, another limitation is that budgets for a production on the ground do not provide much data. A production might want to do a month’s shooting in Scotland, but the budget will just be for the shoot as a whole and will not say where it is taking place.
I presume that that problem is not unique to the UK and that we would find that colleagues in the US, for example, where productions are shot across multiple states, would have similar problems when trying to break down that data.
I do not know much detail about what goes on in other countries regarding production; I know about that only in a general way. However, it is pretty much the way that you describe. When a large US studio comes to the UK to shoot, it will be able to say what its UK budget is and what its spend is elsewhere, but if it has done some shooting in Scotland, it will not be able to say how much was done in Scotland and how much that cost.
On Mr Greer’s point about the United States, because of the different way in which film incentives work in the US—we finished some work in the state of Georgia about a year ago on this—there tends to be good data on production spend on a state-by-state basis because it is the states that incentivise the production. The same applies in Australia and in Canada, where the provinces provide incentives. As a result, there is audited spend data that provides good detail. In some provinces, such as British Columbia, it is even possible to break down the data into parts of the province, because there are additional incentives that provide uplifts for economically deprived areas. That is relatively common in those sorts of state-level jurisdictions in federal countries.
A lot of countries do this well and do a lot of work on it. We have recently worked in Australia, which has very good data, and we have worked in Canada, the United States and most European countries. They all have good data to a degree, but it depends on their particular aims—they always collect data based on what they are trying to prove. They could always put more money into data gathering and analysis, but that is kind of like asking, “How long is a piece of string?” The question is always how much value that would add.
Maybe it would be helpful to look at the Scandinavian countries. There seem to be two needs for statistics here. One is the need to drive the industry and tell us what the Scottish screen sector is really like; and the other, specifically in this context, is the need to drive policy. The Scandinavian countries produce film law every four years that decides what the sector should do and gives direction and budgets to the screen funders. In that process, there is collaboration between the Government, which sets down the law, and the screen agencies, stakeholders and the industry. In that context, data becomes very relevant and interesting in deciding what happens in the next four years. That could be a relevant example for the committee to consider.
That is a good example. Is there an element of trust that makes that process easier? If the production companies have a level of trust in the public agencies and in the Government, is there more of a collaborative approach and a willingness to share data than there may be in countries where there is a more challenging relationship between production companies and the state?
09:30
That is a really interesting way of looking at it. You could also look at it in a completely different way. I know the most about what happens in the Danish context, where the film law is set every four years, and everybody has a stake. There is a consultation between stakeholders, the various industry bodies and the policy makers. People know that it is coming up and they know how to inform the policy-making process. That generates trust, although I am sure that it also generates distrust. It is also a form of quality control, because the industry and stakeholders can say that there are issues facing the industry that need to be discussed and that need to be integrated into the policy for the next four years.
There is therefore a degree of collaboration but also a degree of quality control. Complaints or issues can come up in that process. It can also drive important structural changes to the industry. For example, the Danish film law has quite a big budget allocation linked to it and the policy puts a priority on creating non-format content—web content and VR. Budget is then allocated to that priority. Sweden has 50:50 gender targets that are enshrined in policy.
You can make policy decisions that match Government priorities and industry and stakeholder priorities. It seems to be quite a proactive and good way of organising screen policy. So far, in a sense, Creative Scotland has decided what its priorities would be for the screen sector. I am sure that that takes a lot of energy, time and resources, whereas if the priorities are a given and it is the agency’s role to best deliver that particular film policy, that might create a better and more transparent process.
That is a really useful example—thank you.
Mr Barnes made an interesting point about how you can always spend more on gathering data but it is a “How long is a piece of string?” question. There is a level of uncertainty about how data will be used and to what benefit. There is an element of risk, because we do not know the ultimate benefits if we go down that route. I think that that explains why, historically, there has been a bit of a lack of wanting to participate or to collect data within Scotland. The question is: what will the benefits be?
There is on-going uncertainty about that, which requires work to be done. The European Audiovisual Observatory deals with data from all the different areas across Europe and finds great variety in the resources that are applied. For example, if we look at the BFI and the applied thinking of CNC in France, which put huge amounts of resources into it, we can see the applications. It is about taking the best practice, exactly as Dr Sorenson said, and applying that, within the idea that some of it will be uncertain and experimental.
Richard Lochhead has a supplementary.
The discussion about the issue of trust began to address my question, which is about how the data is made robust. Clearly, the reasons why we want data include ensuring that companies fulfil their obligations in relation to filming and spending in Scotland. In response to getting public support, companies have to show that they have spent money in Scotland. How do you ensure that the data is robust, especially with companies from outside Scotland that are filming here?
Are you asking whether the statistics, the data and the research are robust and valid and therefore of good quality and can be trusted, or are you asking about the process of—
I am asking about the process of how the data is collected, who collects it and who ensures that it is checked.
I will have to refer back to the BFI practice in responding to that. We produce all our statistics and research for the public and industry good. Because we produce official statistics, we follow the statistics code of practice under which all stats must be for the public good. By following the code, we develop a sense of quality and trust in the statistics. That feeds into a circle in which companies are more willing to give their data to us in general.
I have noticed a large appetite in the film sector for more data. Whenever we have user engagement, we find that the sector wants more data and is willing to share it. However, there is not a natural ingrained process in film for sharing data. The best example of how data is shared in my job is through the certification process, through which a film can be officially certified as British and so tax relief can be claimed. Companies do that in part because they see a benefit for the production of the film, but also because the process has been established for many years and has been officially reviewed. That is backed up by a team that has gone out to the industry and provided guidance on how it can provide data.
We need to have a mix of all those aspects. You could look at the situation in Scotland but, as Inge Sorensen has mentioned, it would be useful to see what other countries do, too. The Danish example is a good one of on-going user engagement. I have found that to be vital for developing trust in any part of the data and statistics process.
The standard way to ensure that people have met their obligations is to have the data audited. Auditors tend to be required in every incentive system and for public funding where there are such spend requirements. They have a duty under law in signing off data, and leveraging that duty tends to be the approach that most jurisdictions take.
You mentioned the states in America and the provinces of Canada. Have you looked at how rigid they are about making sure that companies that say that they are spending the money in Georgia, British Columbia or wherever are doing so? How does that compare with how we test the data here?
I cannot speak to how you test that data here. I have never looked at that issue in detail, so I refer you to Creative Scotland.
Other jurisdictions tend to be very robust in determining where the spend happens, although there have been cases in the UK where audited spend statements have been proven to be incorrect and people have ended up in jail as a result. We also know from conversations with producers in Ontario that the volume of checking that is required has led to a lag in how long it takes them to get their funding back from the provincial tax credit system.
That all speaks to systems that are robust, that take a lot of care in how the money is returned to productions that have availed themselves of tax credit systems and other selective funding systems and where producers tend to see putting in place that audit requirement as a cost of doing business.
Two things are going on here. The first, which is what we are talking about, is incentives for film, mainly through Creative Scotland or the UK Government. Secondly, as you are aware, a separate Ofcom review on the regulation of out-of-London spend is taking place. The issue of robustness applies to both those elements, does it not?
It should. Again, I cannot speak to how Ofcom collects its data, because I have never looked into that in detail.
Mairi Gougeon asked whether there should be a yearly update of data. Surely a real-time model should be applicable instead of a yearly update?
What is meant by “real-time” in this context? The questions that we were sent in advance referred to previous evidence to the committee, and when I worked through that evidence, I could not find the term. What is meant by that particular term and what does it propose?
I was thinking about other sectors. Before I became a parliamentarian, I worked in an electronics company. Data was collected daily. I am not suggesting that that should be done because it would be difficult for the wider sector. However, it would be useful to have weekly or monthly updates, and it would be helpful in relation to data collection for further analysis.
One of the things that we have found when working around the world is that production companies—I will speak about production companies in this case—anticipate that if they have real-time data about what is being produced, what the market is doing, and what consumers are interested in, they will be able to make films or television programmes that fulfil those particular demands. The challenge for the sector is that the lead time on production is so great that it can take three years for a film or a television programme to go from idea to final content—it even takes a while to build a physical facility such as a studio. By then, the data that was available at the start is no longer relevant to the market into which the content is released.
The question that I would ask about real-time data is how it will be used. It goes back to that “How long is a piece of string?” question. Is there value in asking people to collect that data? Is it based on the particular needs of the market and the people who are administering the product? It is not something that we have ever identified as a major concern.
That is helpful. I wanted to test that area because of my previous experience outside Parliament.
A few moments ago, Mr Tosta made a point about the industry wanting more data. Are the requests and specifications for the data clear enough? How much flexibility is there around that, in the context of the changing nature of the sector and the changing data that could be required?
From the documentation that we were sent, it looks as though the film industry is asking for quite standard data. You would naturally have to go back constantly to do user engagement and talk to the industry to make sure that the data adapts to the changes in the industry.
A prime example of that is that, 10 years ago, my team was mainly looking at the value chain and public investment. That data collection is now extended and goes further into audiences and education. That is because of constant user engagement.
As we have mentioned definitions, for me, flexibility is all about the parameters of the statistics and the data. It is about the timeliness of the real-time data, as well as accuracy. You have to be responsive to what the industry wants in relation to the data that you collect and what you want to collect data on, but you have to bear in mind that you cannot make such changes immediately in statistics; it does not happen that quickly.
In a way, people in the film industry appreciate that. They ask a question at the development stage of a production, but that question will change by the time they want to release the film. You have to be constantly on top of that.
09:45Consistency is also part of ensuring quality. I may sound as though I am going round the houses, but I am thinking about the stats delivery process. I know that the committee wants data on employment as well as skills and production spend. Such data can be provided consistently and, with user engagement, the data can be adapted to be more suitable for the film industry in year 5, because there have been changes between year 1 and year 5. I hope that that is clear enough.
That is helpful. It takes me back to Dr Sorenson’s point about the Scandinavian model and the four-year cycle. Would it be useful to implement that kind of flexibility and joint approach here, bearing in mind Mr Tosta’s points?
To return to Mairi Gougeon’s question about skills and the freelance base, there is a need for a better, more granular understanding of how freelancers work in Scotland, how much they earn and what industries they work across—a screenwriter will often be engaged in a variety of genres and functions throughout the year, perhaps writing for a corporate job one week and then working on a feature film the next.
There is often a skill drain from Scotland to London and towards other industries. We need to find out how freelancers who engage in the sector work, how they make their money and what we can do to keep them creative and working in the screen sector in Scotland. We can then determine how we drive future development in Scotland. That could feed into trying to identify future trends, such as whether people are starting to work and develop content for Netflix more and are doing less for the BBC and so on.
If we can see where the industry is moving and what people are doing, that will enable us to identify the skills needs and consider what it takes for people to continue being creative in Scotland, such as tax incentives and so on. That could feed into bigger statistics, although not the work that Alex Tosta is doing. Does that answer your question? I veered a bit off course.
It sounds as though we need to move to a big data approach, potentially without the clarity of what that data should be and what is required. There are also some questions about whether the screen unit will have the technology, capability and skills to gather and collate that information.
It has to be delivered retrospectively, as Alex Tosta said. You might need to engage in a census that would then feed into a bigger data set around that.
You have just described the parameters of a data system. I think that you were talking about how to reach a starting point and that goes back to understanding the primary purpose and the fundamental questions that you want answered. I may sound as though I am repeating myself, but that is fundamental from the perspective of research, data and statistics. Once you have clearly defined those things, you can decide on an approach and methodology to get the data.
It is important to have user engagement running in parallel with that, so that the film industry is involved in shaping its own industry and having a say in the statistics that are used to describe the industry so that they are appropriate. Whether to start off by running an annual census or by taking an audit of the available Scotland-specific data is up for discussion, which is probably better done outside the committee.
There are certain issues that certain methodologies will not help. Inge Sorensen mentioned freelancers. We know from the work that Creative Skillset has done that it is difficult to get data on freelancers, because of the whole definition of freelancers and the behaviour and activity of that type of employee and employer. That is just one example. For me, it is all about setting out parameters first, based on the desired outcome and the questions that you want to be answered. Once you have those sorted, you can then discuss methodologies, real-time data and what data we will look at in the future.
Research and statistics are, by their nature, retrospective, as Inge Sorensen mentioned. That is because we have to collect data at a certain point in time and then report on it. There is no escaping that. I know that companies such as Netflix say that they can record what their viewers are watching or what their directors are making immediately, but they have a very different set-up from how most statistics are collected in any industry. It is important for the committee to know about that in its further considerations on a screen unit for Scotland.
Thank you. Two more members want to ask questions, so it would be helpful if people could keep their answers succinct.
We have talked about skills, but I would like to look at how employment fits into the process. We have a growing cultural sector, we have more people working in the arts and culture, and the economy is bigger and better, but when it comes to collecting employment data we still seem to have some gaps in the process. Is that because there is not enough investment? Does that have an impact on the whole process? How can the gap in the employment sector data be filled?
I seem to be the first to answer all the questions. At the BFI, when we report on employment in the industry at a UK level, we get our data from the Office for National Statistics. The data is collected from its various business surveys and population surveys. I mention business because I look at the whole of the GVA and the number of companies. One issue that we face when trying to provide data at a lower level, such as at the level of the four nations or at regional level, is that often the data has been suppressed, because that information might disclose either an individual or a company. That is one data issue that I have in relation to Scotland. I was checking employment figures yesterday and found that they had been suppressed for parts of Scotland, so that hinders me from providing a UK picture.
It would be difficult to fill that gap; the official statistics will always be suppressed because of the rules and regulations around disclosure control. There would need to be further investigation into whether you should do additional research or additional data collection. The rest of the panel may have something to add to that.
We would probably approach it from the ground up, at company level. We worked on that kind of project recently in the Republic of Ireland and we had to approach it from the ground up, looking at individual companies and identifying the number of people working for them through a survey. We did the same thing in the current UK screen sector tax reliefs work for the games sector, because of the particular issues around SIC codes in that industry. I cannot see another way of doing it, other than through a primary survey.
Does the investment behind it have an impact, or not to the same degree?
I have not seen any evidence of it having an impact.
Do you have nothing collated that would identify that as an issue?
No.
I want to voluntarily declare an interest, in that I worked in the television production sector for 13 years before I came to the Parliament, so this is a subject that is close to my heart.
I want to pick up on a point that Dr Sorensen made about definitions of the sector. When I was in television production, we would get very excited about the overnights if we had a couple of hundred thousand viewers, but the reality today is that people across the length and breadth of the country are making content that is achieving millions of views overnight, and they are monetising that.
Given that the sector has evolved from traditional film and television production into the online world, web production, content in the charitable or not-for-profit sector, VR and gaming production and advertising, how do we best collect data on all those other aspects of the production sector, which are the ones that we probably talk about the least, and who should be responsible for collecting it?
That is a big question. One way to look at the issue is that, currently, most screen work is very focused on stimulating and funding production—the screen unit will be, too—but there is not much data on distribution, because it is hard to get, as companies do not want to disclose it. If we look at distribution data as much as production data, it might provide different optics on the sector. The way forward is to define what the screen sector is and then collate the data from the different agencies that have information about inward investment, productions that are funded here and broadcast productions. I do not have a quick fix, but certainly it would be helpful to have more data on distribution and where things are seen and distributed to. I do not know whether that is even possible. Alex Tosta might know more.
The digital landscape has created a new extension to the screen sector. When you start getting into it, film is quite nebulous. It appears to be structured, but there are elements that do not have much structure. That sometimes makes it difficult to gain any information about it, beyond official data collection or the way in which the BFI looks at production—we actually employ someone to track production in the UK. There is a general difficulty in understanding what the new digital area is, let alone starting to track it, what it is doing and what it contributes to the economy overall.
At the BFI, we have started with an approach that is in a way basic, although it is still quite complex. That involves following productions and developments in high-end television, children’s television, animation television and video games that are going through the certification process and trying to become officially British. That is just a starting point. In a way, the sector has traditionally not been in the world of data sharing or providing data to see what is going on. It appears that companies either want to keep within themselves—as with Netflix, which people can watch anywhere and anyhow—or, in the gaming world, from what I have gleaned from the little that I know about it, they kind of know what is going on and there is a lot of network infrastructure.
Overall, it is difficult. At the moment, we have a structured and almost traditional approach to collecting data on those new areas, but they are not traditional, and they work slightly differently. In a way, the data gathering has not really adapted enough to fit them. It is difficult to suggest anything without looking further into those areas, doing more on-the-ground work and getting more involved. I know that Ukie, as the trade body, speaks a lot to the games sector and does a lot of events just to get people more involved, even with certification. That may be one approach involving more on-the-ground work. That is just a small suggestion.
Does the panel have any views on the gathering of data on the gender pay disparity in the industry? We talk about that a lot in other industries, but we perhaps do not talk about it very much in the screen sector. Is any qualitative or quantitative data produced on whether, for example, women freelancers are paid the same rates as their male counterparts, or on the representation of women in the industry and whether specific roles are dominated by one sex or the other? How much data is produced on that and where could we do better in that respect?
10:00
There is a report on diversity in Scotland. I was a producer before I came back into academia 10 years ago. The problem is not so much with gender pay; it is about how to keep women in the business once they start families. That issue is not unique to Scotland; it is a pan-world thing. It is not a particularly family-friendly industry, and that is certainly why I left. That is just an observation.
I would flag up that there are many initiatives that are addressing the issue, and it is really important for the screen unit to engage with those. There is an initiative on parenting in film, which is called Raising Films, and another called Women in Film and Television. Both of those specific UK initiatives are doing really good work. At international level, there is the Annenberg inclusion rider on the issue of pay equality, which can be put into contracts. At budgetary level, when public funding is applied for and received, the budget will have a breakdown, certainly at a high level, of the named-cast pay. The funders have that information, and certainly the commercial bodies will have it. It is absolutely vital to foreground the importance of those issues.
The second part of your question was about representation on screen. I would like to flag up the great work that is being done by Mr Tosta’s colleagues in the BFI. On Monday, Mr Anderson from the BFI did a presentation about using its filmography database to allocate where different genders and ethnicities are being represented and underrepresented in the filmography of British film. Scotland absolutely needs to engage with that work as a partner.
I thank all our witnesses for coming to give evidence.
10:02 Meeting suspended.Previous
InterestsNext
Annual Report