By Ricky Lawton
Background
There has been growing interest in recent decades in better understanding the contribution of cultural and heritage assets to the social welfare of the country. A growing body of evidence has been developed by the UK Department for Culture, Media and Sport (DCMS) on the value of cultural and heritage assets, both individually (for instance, the Natural History Museum and Tate Liverpool)1 and as groups of assets of the same type (through benefit transfer studies of multiple museums2, galleries3, theatres, historic buildings4 and even digital cultural services5). This evidence is based on Willingness to Pay (WTP) methodologies (Stated Preference surveys) which elicit a hypothetical price that visitors and non-visitors would be willing to pay to continue to enjoy the benefits which flow from cultural or heritage assets. It has been compiled as part of the Cultural and Heritage Capital programme6, headed by DCMS, and with ongoing research funded by UK Research and Innovation (UKRI).7
However, to move beyond piecemeal studies in order to understand the total value of the stock of cultural and heritage assets to the UK would require appropriate methods to develop ‘shadow prices’ (an estimated value placed on intangible costs and benefits) for valuing cultural assets consistent with the accounting methods used for other public goods, such as the natural environment. The approach to valuing assets for which there is no market price rests on a well-developed conceptual framework pioneered in natural capital accounting (and set out in another recent ESCoE Discussion Paper.8 In this Technical Report, we applied the framework to a cultural asset: Blenheim Palace, which is a World Heritage Site near Oxford, situated in a landscaped park created by the famous landscape gardener ‘Capability’ Brown. It was made designated a UNESCO World Heritage Site in 1987. The house and gardens are open to the public for an admission fee, though parts of the house and gardens are open to the public for free.
Aims
The aim of this study is to pilot an approach to estimate the value of a cultural asset, and where possible to disaggregate the services derived from it, in a way which can inform the potential for inclusion of such assets in the national accounts.
In national accounting, the ‘stock’ of cultural assets (for instance stately homes like Blenheim Palace and Gardens) provide a ‘flow’ of services through a combination of recreational, historical, cultural and educational services provided. This composes the cultural and heritage value of the site, which can be combined man-made cultural and heritage capital and semi-natural capital elements to understand the total value of the heritage asset to the nation.9
Stately homes like Blenheim Palace and Gardens constitute a ‘quasi-public’ good, because part of their value to the nation is already captured in the prices people pay for entry. However, this is only a partial value, because the entry fee may not be set at maximum willingness to pay to support the full suite of cultural heritage services it provides (as demonstrated in the voluntary donations that visitors are willing to pay even after having paid).
Methodology
A Stated Preference (SP) survey was run on a sample of 675 respondents, split between 382 participants who had visited Blenheim Palace and Gardens in some way in the past year (defined as Blenheim Visitors) and 293 who had not visited (defined as Blenheim Non-Visitors). The survey also captured self-reported on-site spend, travel behaviour and time use data, to enable triangulation of the main WTP results to other non-market valuation methods.
For the primary research question, two types of Stated Preference questions were asked:
- A contingent valuation question, which contained a willingness-to-pay question designed to elicit the welfare loss that visitors and non-visitors would experience if Blenheim Palace and gardens were to close to the public. It asked how much they would personally be willing to pay as a donation to avoid its closure. Respondents were asked if they would be willing to pay a top-up donation to keep the house and gardens open to the public in the face of increased operating and maintenance costs.10
- A Discrete Choice Experiment (DCE) to understand visitors’ preferences for different parts of the heritage site. In a DCE, or ‘conjoint’ exercise, respondents are asked over multiple choice questions their willingness-to-pay to support the site under different hypothetical closure scenarios, where different combinations of the house or gardens would be closed for a year. The DCE was designed to capture the relative importance of different parts of the site as a source of cultural and heritage value. This was framed as a one-off payment, linked to peoples’ visit either as a donation or an increase in ticket prices (payment vehicle varied between two survey sample groups).
Results
- Willingness-to-pay estimates for the continued public access to the site
- Visitors to Blenheim Palace and gardens positively value the heritage site beyond what they already paid in entry fees to use its services. Visitors would be willing to pay a one-off donation of £9.26 on average per person, with a lower bound of £7.84 (the lower bound figure based on the 95% confidence interval is recommended to offset risks of over-estimation due to hypothetical bias in surveys such as this).
- Blenheim non-visitors in the general public would be prepared to donate at £3.85 (lower bound £2.97).
To put these figures into context, the lower bound visitor WTP of a top-up donation of £7.84 represents around a quarter of the current full-price entry ticket to Blenheim Palace and gardens (£32), which suggests that stated WTP ‘shadow price’ is not an over-estimate of the value of the site to people’s welfare.
2. Discrete Choice Experiments: Change in WTP for closure of different parts of the house and gardens.
For the DCE, two different modes of payment were tested on a split sample of respondents.
First, a donation was used (n=171 completed this DCE).11 Participants were shown multiple ‘choice sets’ (the combination of attributes and their open/closed status along with a donation amount), Naturally it was expected that a lower donation value would be associated with the highest utility, holding other factors constant, because people are able optimise their levels of cultural services for a lower price.
However, results show that visitor preferences are the opposite of this expected implied relationship, with participants on average selecting choice sets which had the higher donation price. In other words, visitors appear to gain more utility from paying a higher donation, regardless of which parts of the house and gardens were open/closed in the DCE. As such, higher donation amounts appear to exert dominance over respondent preferences.
A second DCE was designed and tested on an independent sample of 144 visitors to explore whether this counterintuitive finding reflects the donation payment mechanism. A ticket price was used instead of donation, with the price attribute varying around the standard ticket price of £32.12
In this second DCE, results show that utility distributions for the eleven parts of the site are consistent with expectations (i.e., people prefer situations where more parts of the site were open than closed), at least for respondents whose survey answers suggest they are less ‘attached’ to Blenheim Palace (in that they said they were less likely to visit again in the future, had visited less frequently or had not paid to access the house). However, for those with affinity to Blenheim, higher utilities are still observed for higher ticket amounts, again suggesting that even these respondents gain utility from giving more of their money to the site.
Taken together, the DCE results do not provide meaningful WTP estimates for the closure of different parts of Blenheim House and Gardens. When overlaid onto the cultural services model proposed in the accompanying conceptual paper13: the higher utilities observed for higher donation/ticket prices would counterintuitively mean that more preferred parts of the house/gardens (i.e., those parts of the house with higher utilities) would have lower WTP values. Further research is required to test the working hypothesis that different parts of a heritage asset provide different types of cultural services, and that the provision of one instead of another may help to explain why people are willing to pay for different services provided by a heritage site.
Discussion: Triangulation of non-market valuation methods
A further contribution of this study to the ongoing discussions around accounting of cultural and heritage asset stocks and flows is to apply a range of methods to triangulate what values could be attributed to service flows stemming from Blenheim Palace and Gardens by comparing the estimates elicited through the Stated Preference exercise with estimates derived from:
- ‘Revealed Preference’ measures which take prices paid for associated services as a proxy of the non-market value of the site, in this case people’s actual prices paid for entry, in the shop etc, and travel costs;
- ‘Travel cost’ and ‘Time-use value’ Measures based on the time involved in visiting the site.
Unsurprisingly perhaps, given the range of valuation methods, we get a range of results as outlined in the Summary table below. This likely represents both the different dimensions of the economic value of culture and heritage obtained through the different methods and potentially the different biases that arise in applying the different methods.
In all cases, we consider the extent to which these non-market valuation estimates can be counted together, and alongside accounting estimates of value taken from the Blenheim Economic Impact Assessment report, 2021/2022 (total revenue of £15m in 2021/2022).
The most reliable estimate of total aggregate value for Blenheim Palace and Gardens in the summary table above may be the combination of actual revenue and willingness to pay among visitors as a top-up donation to support the site, and non-visitors as a donation for the same outcome (combined total £46million per year). This is a more conservative estimate than some of the higher estimates obtained through travel cost methods. It also has conceptual consistency with the WTP question eliciting over and above any prices paid for entry. It is designed to capture the ‘surplus’ welfare value (over and above the prices already paid) of maintaining public access to the site for oneself and others, which aligns with the conceptual model described in our companion paper.
Conclusion
While expenditure on cultural/heritage engagement is captured in current statistics, as are some of the associated heritage assets, this nevertheless leaves significant assets and the associated capital services unaccounted for. Just as with other missing capitals (like natural capital, social capital etc.) enhanced measurement is desirable for decision-makers using official statistics so that decisions can be based on a full evidence base including the sustainability of the use of economically important assets.14
However, market prices do not exist for many significant culture and heritage assets and alternative methods for estimating (shadow) prices are needed as explored in this empirical paper. There has been a move toward incorporating ‘omitted’ capital service flows at market prices (or an exchange value alternative) into the measurement of GDP.
We conclude that non-market valuation methods may be needed as a supplement where markets or exchange values for cultural and heritage assets do not exist, including stated and revealed preference methods as applied in this study.
The Technical Report can be downloaded here.
ESCoE blogs are published to further debate. Any views expressed are solely those of the author(s) and so cannot be taken to represent those of the ESCoE, its partner institutions or the Office for National Statistics.
1H. Bakhshi et al., ‘Measuring Economic Value in Cultural Institutions’, Cultural Value Project (London, UK: Arts and Humanities Research Council, 2015), https://ahrc.ukri.org/documents/project-reports-and-reviews/measuringeconomicvalue/.
2D. Fujiwara et al., ‘The Economic Value of Culture: A Benefit Transfer Study’ (London, UK: Department for Digital Culture Media and Sport, 2018).
3R. N. Lawton et al., ‘Regional Galleries and Theatres Benefit Transfer Report’ (Arts Council England, 2021), https://www.artscouncil.org.uk/sites/default/files/download-file/Arts%20Council%20England%20-%20Regional%20Galleries%20and%20Theatres%20Benefit%20Transfer%20Report.pdf.
4 Ricky Lawton et al., ‘The Economic Value of Heritage in England: A Benefit Transfer Study’, City, Culture and Society, Recent developments in urban heritage valuation: Concepts, methods and policy application, 26 (1 September 2021): 100417, https://doi.org/10.1016/j.ccs.2021.100417.
5 M. Arber et al., ‘Measuring the Economic Value of the Digital Offer of Galleries and Museums: An Exploratory Use of Contingent Valuation Techniques’ (London, UK: Arts Council England, 2023), https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/
file/1165079/Digital_Offer_Report_September_26_2022.pdf.
6 https://www.gov.uk/guidance/culture-and-heritage-capital-portal
7https://www.ukri.org/news/new-projects-to-measure-value-of-culture-and-heritage-to-society/
8https://www.escoe.ac.uk/projects/cultural-capital-as-an-intangible-asset/
9The meaning of ‘cultural value’ is specified through a non-comprehensive list of sub-components as detailed by the Burra Charter.: Australia ICOMOS, ‘The Australia ICOMOS Charter for the Conservation of Places of Cultural Significance (the Burra Charter): Guidelines to the Burra Charter : Cultural Significance and Conservation Policy.’ (Sydney, New South Wales, Australia: ICOMOS, 1987).
10Visitors were reminded this donation would be on top of any ticket price, membership, or other payments they had made to support the site in the last year.
11Specifically, the WTP amount that they had selected in the CV question was piped into the DCE exercise, and varied by subtractions of 5%, 10%, 15%, 20%, 25%.
12As in split sample 1, the price attribute varied by 5%, 10%, 15%, 20%, 25%.
13https://www.escoe.ac.uk/projects/cultural-capital-as-an-intangible-asset/
14Carl Obst and Michael Vardon, ‘Recording Environmental Assets in the National Accounts’, Oxford Review of Economic Policy 30, no. 1 (2014): 126–44.
