The business of philanthropy: Challenges in measuring the value of cultural giving

Corporationsareunderincreasingpressuretoprovideinfor-mation about their corporate social responsibility activities. However, there is limited work on how firms measure the value of their philanthropic giving. The value of a philanthropic act can be difficult to ascertain as it often relates to unique goods that do not have a market value. The paper examines the processes a sample of (mainly family-run) private corporations follow to value their donations to a not-for-profit cultural festival. The findings suggest that while giving to the festival is seen as a contribution to the local community, there are additional motivations for donating. Moreover, the firms do not attempt to formally assess the value of their act, which they believe is hard (if not impossible) to measure. Our case highlights a more complex system influencing philanthropic acts of private firms. However,unlikeourwork,thesestudiesdonotfocusonthereasonswhyorganisationsdonatetofestivalsandwhether or not they assess the impact of their giving. Our study also provides fresh insights into how businesses deal with the pressure to account for all of their activities, including their philanthropic giving (Miller & Power, 2013; Power, 2009).


INTRODUCTION
In an attempt to satisfy the expectations of stakeholders, corporations are providing increasing amounts of information on their corporate social responsibility (CSR) activities (Crane et al., 2014;Gelb & Strawser, 2001). Such information may include an analysis of the impact of their giving that relies on measures such as the social return on investment (SROI) (Arvidson et al., 2013;Moody et al., 2015), even though the suitability of such measures has been questioned in some studies (Cooney & Lynch-Cerullo, 2014;Yates & Marra, 2017). Additional concerns have been raised with respect to the objectivity of such analyses as they often create their own reality through the influence of their social context (Hines, 1988;Quattrone & Hopper, 2005). Moreover, determining the impact of giving to an event is particularly difficult when the event is 'unique' , such that some of the factors that drive the valuation are immeasurable (Callon & Muniesa, 2005;Karpik, 2010). Difficulties remain even when all factors can be measured as the value of the whole may be different than the values of its parts added up (Espeland and Lom, 2015;Hines, 1988;Pallesen, 2015). Although challenges of valuation are well understood, there is a paucity of research on the process adopted by firms when they are faced with the task of attributing value to a unique event, especially when the event is related to a philanthropic act and not to a core business activity.
This paper provides insights into the business practices (see, e.g., Miller & Power, 2013;Mennicken & Miller, 2012) followed when defining and attributing value in the context of a donation to a cultural festival. While there has been an increasing academic interest in the philanthropic acts of corporations (Muller et al., 2014;Saiia et al., 2003;Shabana et al., 2017;Wang & Qian, 2011), we still lack a thorough understanding of the methods corporations use when they assess the impact of their giving. Such an assessment is further complicated by the fact that not-for-profit organisations (NFPOs; the recipients of giving) still lack in terms of accountability and transparency (Cordery et al. 2019;Hyndman & McConville, 2016) and the information they provide to corporations tends to be limited.
Our focus is on businesses that donate to a cultural festival as cultural events are unique events with singular values.
The data are collected through a qualitative methodology and is based on a set of private corporations that financially support the Opera Estate festival run by a municipality in Northern Italy. All except one of the firms are family-run businesses. In contrast to expectations, none of the private corporations we study attempt to measure the impact of their donations to the festival and all of them see such impact to be 'unmeasurable' . Moreover, they argue that a valuation exercise would diminish the value of their giving and prefer to rely on their entrepreneurial instincts rather than a formal impact assessment when deciding to support the event. Nonetheless, they admit that they benefit from the informal feedback gathered from their network when deciding to renew their support, suggesting a communal form of accountability (Laughlin, 1996) or social accountability (O'Dwyer & Unerman, 2007) rather than a contractual agreement with enforceable terms (Dubnick, 1998). While the main driver behind their decision to donate is to 'give back' to the society, which is consistent with prior literature (Phillips & Jung, 2016a;Schervish, 2014), they expect that their support of the festival will also enhance their visibility and improve their reputation (Hogarth et al., 2018).
Moreover, some businesses highlight that their support of the cultural festival indirectly benefits the welfare of their employees who attend the festival events, which can lead to an improvement in the latter's productivity (see, e.g., Guiso et al., 2015;Zhao & Zhang, 2019;Gond et al., 2010). Taken together, our results suggest that the act of corporate giving is a multi-layered, complex act.
This study adds to the literature on corporate philanthropy in a number of ways. We challenge the view that corporations predominantly rely on SROI measures when assessing the impact of their giving. The private corporations we interview cast doubt on the SROI measures, which they feel cannot capture the true value of their giving, and do not undertake a formal impact assessment in relation to their giving. Their decision to support an event is influenced by their esteem and trust in relation to that event (Karpik, 2010;ter Bogt & Tillema 2016) as well as informal feedback mechanisms. Therefore, they adopt an unofficial, non-routine mode of information processing to support their decisions on corporate giving (Earl & Hopwood, 1980). Moreover, their philanthropic acts can be described as 'dispersed philanthropy' according to Bruch and Walter (2005). 'Strategic philanthropy' , which Bruch and Walter consider to be the optimal form, may not be attainable by the private corporations we examine as they are not under pressure from the market in terms of their giving. Finally, our findings imply that the scope of the 'market orientation' defined by Bruch and Walter is wider than originally thought. In particular, while the authors acknowledge the ability to attract and retain qualified employees as a driver for corporate philanthropy, they overlook the potential improvements in workforce productivity as an additional motivation, which is a very important one according to our results.
The remainder of the paper is divided into seven sections. The next section outlines the challenges of attributing value to unique events. Section 3 reviews the related studies on philanthropy and the prior literature on accounting for philanthropic acts. Section 4 introduces the research method. The findings are discussed in Section 5. Section 6 concludes the work by reflecting on the contribution of this study and by discussing future research directions.

ATTRIBUTING VALUE TO UNIQUE EVENTS
Organisations are under pressure to measure and account for their performances (Funk, 2015;ter Bogt, 2008). Offering an 'accurate' picture of performance improves transparency and serves as a means for an organisation to build or extend trust with its stakeholders (ter Bogt & Tillema, 2016). However, critics have observed that the information provided by organisations regarding their performances is often shaped also by internal accounting practices (Arrington & Francis, 1993;Miller, 1994) and that organisations to some extent create their own realities (Hines, 1988(Hines, , 1992 when generating and aggregating accounting data (Meyer, 1986). Furthermore, accounting decisions involve promises about the future, even though information used to predict future outcomes is often imprecise or incomplete (Mouritsen & Kreiner, 2016). Interestingly, Mouritsen and Kreiner (2016) indicate that the contractual agreement between two parties includes the prospect of forgiveness and forgetfulness, recognising the risk of falling short of expected performance. The agreement that is generated is valid only at a specific moment in time and space (Quattrone & Hopper, 2005), characterised by some type of magical adjustments (Meyer, 1986). These shortcomings undermine the objectivity of a valuation exercise, as the attribution of the value, ultimately, is an exercise of compromise (Annisette et al., 2017). Consequently, parties may end up relying on communal agreements that are less structured for the purposes of accountability (Laughlin, 1996), instead of contractual agreements with enforceable terms (Dubnick, 1998). The former has a more qualitative orientation and can be a more suitable form for promoting social accountability, which focuses on broader socially oriented outcomes and impacts rather than short-term financial objectives (O'Dwyer & Unerman, 2007). Contractual agreements and communal agreements can be linked with the hierarchical form and social form of accountability (Roberts, 1991;Roberts & Scapens, 1985), respectively. According to Power (1999Power ( , 2000, today's society is characterised by a desire to measure all sorts of activities. The ability and the need to identify value is particularly challenging for unique services and goods (Callon & Muniesa, 2005;Karpik, 2010) in the so-called 'nonmarket orders of worth' (Stark, 2011). Unique services and goods are identified as singularities as they are 'goods and services that are structured, uncertain and incommensurable' (Karpik, 2010, p. 10), without a clear market price. When a market exists for an asset, it is straightforward to identify the value attributed by the society to that asset. However, for many goods, services, or activities, such as an innovative product or an artistic piece, a market price might not be readily available. Despite the ubiquity of singularities there has not been much attention paid to how individuals value them (Ravasi et al., 2011;Stark, 2011). Karpik (2010) observes that some goods/services, such as art, are characterised by an 'aura' that may not be measured fully or correctly, and the 'aura' distinctly qualifies the activity. Karpik points out that attempts to quantify the 'aura' can reduce the meaningfulness of the measured activity/product. Nonetheless, he suggests that personal and impersonal 'devices' affect the final perception of the value, which is similar to Callon and Muniesa's (2005) argument that brands and friends provide what they call 'surrogate information' . The valuation process might be influenced by the interests (or passion) of the person making the judgement on value as well (Baxter et al., 2019), and it is based on a future promise or expectation (Mouritsen & Kreiner, 2016). Given the challenge of assigning a value to singularities, trust needs to be present to compensate for the lack of complete information (Karpik, 2010;Power, 1999;ter Bogt & Tillema, 2016). When an asset is unique, and the measurement is incomplete, trust becomes an essential element that guides the relationship between different parties engaged in the valuation exercise. Kornberger et al. (2015) suggest that only commensurable elements can be attributed an economic value. However, they argue that through time, non-commensurable factors acquire an 'historical objectification' that allows for the attribution of an economic value. For this to occur, two factors that are related to one another are important: visibility and quantification. An object becomes visible through a process of quantification (Espeland & Lom, 2015). However, quantification can become a source of misinterpretation due to any misunderstanding over the characteristics of the invisible elements that can be related to the above mentioned 'aura' (Espeland & Lom, 2015). Misinterpretations could modify the meaning and the value of the factor measured. Furthermore, having become visible, the measured factor acquires a new truth (Hines, 1988). Lucy and Marres (2015, p. 236) indicate that 'valuation is not only something that is "done", it is something that happens, or is happening' . This statement suggests that sometimes one does not notice the assessment process as it keeps taking place in the background.
Consequently, valuation is an on-going process, and some elements are constantly shifting. Reinecke (2015) indicates that quantification is a rather complex process as it 'translates the plurality of individual value judgements and interests into a one-dimensional numerical expression' , that creates a 'transformation' of the value in itself (p. 211). It is an exercise in 'attribution' of values. Furthermore, the valuation techniques used are susceptible to the social pressure (Quattrone & Hopper, 2005;Power, 1999) as both the society and the organisation's context define the extent and the nature of information that is required to make an assessment (Hopwood, 1985;Jones, 1992;Mihret, 2014;Miller, 1994 ).
Independent from which valuation technique is used, there is a tendency 'to emphasize that which is easiest to measure' (Espeland & Lom, 2015, p. 34;Pallesen, 2015); this inevitably leads to an inaccurate representation of the reality, which in turn negatively affects the valuation process. Espeland and Lom (2015) suggest that one should be selective in identifying the factors that are influential for the measurement. This implies that it is the assessor who decides what is value relevant and what is not. The final value that comes out of this process would not be fully accurate if all value-relevant factors have not been considered. Moreover, Karpik (2010) believes that it is impossible to measure everything, not because it would be an expensive exercise, but because of the risk of failing to account for the 'aura' when measuring an activity. However, he highlights that the decision maker uses personal judgement during the process when assessing the information gathered and attributing a value to the abstract dimensions.
The tension between the strand of literature that highlights an expectation to measure any event and the one that points out the challenges in the process of measurement is particularly interesting. Espeland and Lom's (2015) approach of selective measuring leaves us unsatisfied as this would create a new reality (Hines, 1988(Hines, , 1992, especially when parties receiving the measurement are not aware of the incompleteness of the assessment. On the other side, Karpik's definition of 'aura' captures our attention. Should we really attempt to measure the aura? Karpik's suggestion of other interactions supporting the definition of the value is in rather abstract terms with no measurement techniques related to it. How would organisations tackle the task of measuring the aura in the real world? Is the promise of a future return (Mouritsen & Kreiner, 2016) mitigating the need for a precise measurement? Does the corporate world limit its analysis of the aura to a few selected factors, as suggested by Espeland and Lom (2015), or do they take a full cost-benefit approach? If the latter view is the correct one, which techniques are businesses using to estimate the 'non-market-based values'? We investigate this theoretical tension by reflecting on the philanthropic giving of corporations to a cultural event and by analysing their approach to assessing the value of their donations. In order to understand the current practice and the rationales guiding it, we focus on a cultural event as valuing a cultural activity is particularly challenging (Donovan & O'Brien, 2016;Ellwood & Greenwood, 2016;Hone, 1997;Micallef & Peirson, 1997). The next section focuses on philanthropy in general and accounting for philanthropy in particular.

PHILANTHROPY
To understand the process of attributing value to a singularity, we focus on philanthropic giving to a cultural event. This is a well-defined activity that is not related to the core activities of the businesses in our sample. However, to assess the information provided by an organisation in defining the value of their giving, and consequently the process followed, we need to understand what philanthropy is and which factors affect the decision to give as discussed in the literature.
Additionally, we review the accounting literature on philanthropy, to understand the current state of knowledge in this area.

Philanthropy as an altruistic act
It is difficult to offer a unique definition of philanthropy as it has many facets (Gautier & Pache, 2015;Pharoah, 2016).
The word originates from 'philia' , which means love and care for others in ancient Greek (Phillips & Jung, 2016a). On the other hand, the modern meaning of philanthropy dates back to altruistic giving initiatives undertaken at the start of the twentieth century and is often associated more specifically with the giving of wealthy individuals such as Andrew Carnegie. The establishment of the 'Giving Pledge' by Bill Gates and Warren Buffett has given more weight to this modern meaning as it requires signatories to endow most of their wealth to philanthropic activities (Giving Pledge, 2019). There is also evidence that when a wealthy individual provides financial support to a cause, the cause benefits from that individual's social, cultural and symbolic capital as well. In return, and depending on the cause, the individual can expect a positive and accumulative impact on their capital either directly or indirectly (Gordon et al., 2016, p. 337).
It is also worth noting that there is a distinction between charitable giving and philanthropic giving, with philanthropy aiming to address the cause of a social problem, while the charitable approach mitigates the consequences of the same problem (Phillips & Jung, 2016a). Radbourne and Watkins (2015) provide further insights into the distinction between charitable and philanthropic giving. They observe that, differently from a charitable giving, a philanthropist is expected to have an ongoing engagement with the supported organisation at a certain level. Therefore, the ongoing relation involves more than a simple transfer of financial resources as it entails the existence (or development of) a relationship between the two parties (giver and recipient) that allows for further non-financial exchanges on the actions undertaken or to be undertaken.

Who are the givers?
A number of studies offer a profile of givers. While the majority of them focus on individuals as givers (Bekkers & Wiepking, 2011;List, 2011;Schervish & Havens, 2003;Schervish & Szántó, 2006) relatively few concentrate on corporations (Cho, 2009;Feliu & Botero, 2016;Porter & Kramer, 2002;Valor & Zasuwa, 2017;Zolotoy et al., 2019). One reason for this might be that non-profit activities and organisations are primarily funded by individuals (Cordery & Smith, 2015;Havens et al., 2006). Of course, findings from studies on individual donors might extend to corporate donors as it is individuals who own and manage corporations. In the context of corporate giving, a distinction can be made between direct donations and donations made through a foundation related to the business, such as a family-run foundation, or community foundation (Leat, 2016). Overall, key determinants of the variation in giving are identified as the geographical location, religious affiliation and context, and the specific social circumstances (Bekkers, 2016;Pharoah, 2016). Studies that profile givers have some limitations, however, as they account for a few select factors against the complexity of the reality (Bekkers, 2016). In fact, the decision to give originates from a complex combination of individuals' private and public spheres (Pharoah, 2016).
Evidence also suggests that most donors tend to give only after they are asked for a donation (Phillips & Jung, 2016a). Consequently, giving is generally an induced action, and organisations seeking donations should pay attention to the art of approaching and convincing potential donors. In particular, there is a shift away from the traditional marketing approach (Rentschler et al., 2002). Nowadays, it is recognised that initial conversations with potential donors should not centre around financial needs of the organisation, giving the two parties a chance to get to know each other before the donor (or philanthropist) makes a financial commitment (Radbourne & Watkins, 2015). This implies a relationship that gradually builds up and results in a longer-lasting and more effective engagement between the recipient and the giver. The long-term relationship is economical since the organisation would incur costs each time it sought a new donor (Rentschler et al., 2002). Finally, it is common for philanthropists/donors to have an affinity with the cause, event, or organisation to which they provide support (Bruch & Walter, 2005;Radbourne & Watkins, 2015).

Why do givers donate?
While 'doing good to others' is an obvious driver, there are more reasons that lead to generosity. Bestowers get satisfaction from their giving (Schervish, 2014;Waters, 2016). Moreover, the act of giving normally provides positive returns (and certainly not a negative one) as stakeholders take note of the altruistic actions of givers (Patten, 2007). Seifert et al. (2004) could not clearly identify a positive return, but notice that no negative impact incurs. The authors further note that corporations that engage in philanthropy typically have slack resources. It is also suggested that tax benefits can be a driver for giving (Feliu & Botero, 2016), but according to Pharoah (2016), tax relief is far from being the sole rationale for giving. Nonetheless, there is some criticism on corporations using foundations as a means for tax avoidance (Smith et al., 2016).
Desire for legitimisation and visibility are noted as additional factors that motivate corporations (Brammer & Millington, 2006;Wang & Qian, 2011) as well as individuals (Gordon et al., 2016) to donate. For example, supporting a training course in a community not only helps to legitimise but also develops the pool of potential workforce (Pollard, 1960). When a philanthropic act targets the local community, it is expected to improve the relationship between the organisation and the community, which in turn helps legitimisation and visibility. Regular and wellorganised philanthropic acts can also help the organisation attract talent from afar as individuals would be keener to relocate to a well-supported community (Porter & Kramer, 2002). All of these have an impact on the perceived reputation of the corporation (Hogarth et al., 2018).
While Moir and Taffler (2004) find that it is legitimisation and marketing (rather than altruism) that drives contributions towards arts, findings in Muller et al. (2014) support a less pragmatic motivation for giving. The latter identify empathy as an essential element driving corporate philanthropy, which conforms to the word's original meaning in ancient Greek, that is the care for others.
Finally, a well-planned legitimisation action can bring multiple benefits. Raiborn et al. (2003) suggest that philanthropic activities increasingly support the overall performance of the organisation by improving the internal environment. Additionally, corporate philanthropy influences external communication as well as employees' perceptions of the organisation (Gond et al., 2010;Zhao & Zhang, 2019). On the one hand, employees can be expected to be more productive when they have a stronger belief in the organisation's strategy (Guiso et al. 2015;Zhao & Zhang, 2019).
On the other hand, Chen et al. (2008) find no relationship between the commitment of employees and the organisational philanthropic strategy. These contradictory findings on the impact that philanthropic actions have on employees highlight a need for further research on the interrelationships between corporate giving, legitimisation and corporate strategy.
The corporate philanthropy model matrix Bruch and Walter (2005) identify four main forms of philanthropy depending on the way the organisation shares its 'competence' (expertise) and the impact the donation has on the organisation's 'market' (external stakeholders). The first form, 'dispersed philanthropy' , has a low level of engagement with both competence and the market, signalling no clear strategy for the donation. When the corporation primarily focuses on transferring its competence and neglects the market, this delivers 'constricted philanthropy' . While a pure focus on transferring competence is an inefficient form of philanthropy according to Bruch and Walter (2005), Phillips and Jung (2016b) consider competence the most valuable asset that a corporation can share. When the level of engagement with the market and competence are high and low, respectively, we reach the third form named 'peripheral philanthropy' . Bruch and Walter argue that 'strategic philanthropy' , which is the fourth form, is the optimal form of philanthropy as the corporation not only shares its competence acquired through internal stakeholders but also ensures that the donation has a positive impact on the external environment, benefiting external stakeholders. They also suggest that corporations should advertise the positive impact their donations have on the market. Advertising is seen as an act of legitimisation under the guise of a marketing strategy. However, if the marketing effect predominates, the giving could not be classified as a donation (Seifert et al., 2004). While the analysis in Bruch and Walter is based on large, international public corporations, it is unclear if their framework is valid for private corporations and, in particular, family-run businesses, with no dedicated philanthropy department.

The less-bright side of corporate philanthropy
Although most academic research focuses on positive implications of philanthropy (see Hebb & MacKinnon, 2016), some authors highlight a less-bright (or even a darker) side of corporate giving. There is evidence that some corporations embrace philanthropy to move attention away from organisational shortfalls (Du, 2015;Chen et al., 2008).
According to an even more cynical point of view, sometimes a philanthropist may not be driven by a desire to eradicate a problem such as income inequality but may give to retain the status quo of the organisation, as the corporate vision is not concerned with the distribution of wealth (Cordery et al., 2019;Gordon et al., 2016;Smith et al., 2016;Thümler, 2016). Philanthropy can be a way to influence policy as well (Gordon et al., 2016). Smith et al. (2016) discuss that, within the current paradigm, philanthropists often fail to fully understand the need for their giving and the exact problem their giving will help resolve. Furthermore, the authors argue that the current approach usually fails to implement an integrated solution that involves the local society and the relevant political bodies. Finally, the giving might engender a culture where the recipient becomes dependent on external funds, while the root causes of the problem remain unresolved (Smith et al., 2016;Thümler, 2016).
Corporations increasingly view their philanthropy as an investment. Under this view, a corporation's focus is on maximising the return it gets on its donations (Hebb and MacKinnon, 2016), and the positive impact of the donation on the society reduces to a side effect (Thümler, 2016). This has been labelled as the 'darker side' of philanthropy (Smith et al. 2016) and, according to Thümler, there is no robust solution offered so far by governments or the society to address this darker side.
To address issues related to corporate giving, there is a need for better coordination between corporate givers and state institutions, particular at a time when public sector funds are curtailed (Gordon et al., 2016;Hebb & MacKinnon, 2016;Schuyt, 2010). There is also a growing expectation on corporations to support society. Schuyt (2010) notes that interactions among the government, markets and philanthropists are essential to promote a healthy society. In particular, each of the three mechanisms should make a sufficient contribution to address specific problems in the society.
Finally, Silber (2012) suggests that philanthropists may give with 'anger' when they feel it should be the public sector's duty to support the cause they are donating to.

Donating to culture and art
Because NFPOs rely on donations for their existence, they can fall under the influence of their donors. This is true for NFPOs that organise cultural activities as well. According to Browar (2002), cultural organisations need to be conscious of large philanthropists' desire to assess the impact of their donations and, thus, provide them with relevant information (Mermiri, 2011). This is particularly important at a time when an increasing number of NFPOs are competing for donations. Due to this competition, corporations are forced to act strategically in terms of deciding where to donate (Bulut & Yumrukaya, 2009;Bruch & Walter, 2005). The new generation of large philanthropists is characterised by their business approach when deciding to give to cultural organisations as well as other organisations (Browar, 2002). Radbourne and Watkins (2015) highlight the importance of building a relationship with potential large donors that is based on esteem. This may involve building personal rapports (Radbourne & Watkins, 2015;Bulut & Yumrukaya, 2009) with those corporate employees who manage the philanthropic activities of their organisation (Daellenbach et al., 2013).
It can be argued that certain forms of art such as ballet and opera are for the elite classes and supporting these forms would not benefit the society as a whole. Milner and Hartnell (2015) disagree and argue that supporting any form of art is for the benefit of the society. According to Zamagni (2010), the decision to support art should not be made based on the assessment of potential returns as such support is an investment in the society, whose importance has been somewhat forgotten over time.

Accounting for philanthropy
From the accounting perspective, philanthropy affects performance and, thus, entails a return (Bruch & Walter, 2005;Kramer & Pfitzer, 2016;Porter & Kramer, 2011). Porter and Kramer (2011) highlight that corporations should work towards the 'creation of shared value' (CSV) and philanthropy should be part of their strategy. They suggest that engagement with social needs increases the value of the corporation, which can be quantified as a return, and benefits the whole system. Corporations' engagement with social needs is considered to be part of a natural evolution of the CSR (Crane et al., 2014;Bosch-Badia et al., 2013). However, the CSV concept has been criticised for focusing too much on multinational businesses and ignoring (i) the imbalance of power between the parties involved in the act of philanthropy, and (ii) the differences between western and non-western cultures (Voltan et al., 2017). It has been argued that the CSV can only exist in a utopia where corporations give social needs the same level of attention they give to their core business (Crane et al., 2014). Nonetheless, Bruch and Walter (2005) argue that philanthropy is a relevant part of the business, and, as such, high-level managers should be in charge of it to ensure its effectiveness. Kramer and Pfitzer (2016) further argue that corporations should collaborate with the government and competitors to resolve social problems, and, thus, create shared value. They also note that tackling social issues is essential to business growth.
Finally, the CSV concept is criticised for failing to offer a practical method to measure the value created through CSR activities (Crane et al., 2014;Voltan et al., 2017). Assessing the value of an investment made in the society is highly challenging. With regard to the philanthropic giving, the recipient of the resources is usually a charitable organisation, which would normally provide information on the impact of the investment it receives. However, it is unclear whether such information is sufficient for corporations to estimate the shared value created by their giving.

Lack of transparency
Evidence suggests that charities are facing issues concerning accountability and transparency (Cordery et al., 2019;Hyndman & McConville, 2016). However, their reporting standards have shown improvement over time (Connolly & Hyndman, 2013). In particular, there is a widespread effort to align with best practices and to enhance reputation through positive reviews via organisations that assess the performance of charities such as the GuideStar (Connolly & Hyndman, 2013;Harris & Neely, 2018). Transparency is interpreted as a sign of good governance and professional management, and transparent organisations attract more resources from donors (Harris & Neely, 2018

Assessment of impact
The method of SROI was developed by the Roberts Enterprise Development Fund (REDF) in order to help philanthropists and foundations assess the impact of their giving (Arvidson et al., 2013;Moody et al., 2015). The SROI is a cost-benefit analysis that focuses on non-financial information. According to Raiborn et al. (2003), cost-benefit analysis is the right approach for corporations to assess the value of philanthropic activities. However, they note that, because firms naturally produce more information on core activities than philanthropic activities, the cost-benefit analysis of the latter could be less precise.
Other limitations of the SROI method have been noted in the literature as well (Moody et al., 2015). Implementing the SROI method is a complex exercise, which requires considerable resources. This calls into question whether it is worth implementing this method in the first place (Cooney & Lynch-Cerullo, 2014). According to Yates and Marra (2017), the SROI simplifies the factors that it assesses, providing an 'approximation of impact' (Cooney & Lynch-Cerullo, 2014, p. 391;von Schnurbein, 2016). There is also concern about the sensitivity of results to data inputs (Yates & Marra, 2017). It is acknowledged that small variations in inputs can lead to substantial differences in results (Moody et al., 2015). Another critique is the SROI's heavy focus on monetising the impact of giving (Cooney & Lynch-Cerullo, 2014 (Phillips & Jung, 2016b) and to investigate the ways they are adopted by organisations. Maas and Liket (2011) suggest that large corporations are increasingly monitoring the impact of their donations.
However, their work is based on the constituents of the Dow Jones Sustainability Indices, and Cho et al. (2012) find that some of these constituents embellish information regarding their philanthropic activities. This raises a general concern about window dressing by organisations regarding their donations. Von Schnurbein (2016) highlights that assumptions made to justify the value of a donation can be questionable, which is consistent with the concerns raised in Cho et al. (2012).
In summary, while we have an understanding of the tools organisations can use to assess the impact of their giving, we know much less about whether or not organisations make use of these tools or other methods. This gap in our knowledge raises further questions. In their analyses, do organisations focus on measuring the value of their giving (Espeland & Lom, 2015;Herbohn, 2005;Kornberger et al., 2015;Maas & Liket, 2011;Raiborn et al., 2003) or do they deem such value to be unmeasurable and focus on the 'aura' of their giving instead (Karpik, 2010)? If organisations attempt to measure the value of their giving, do they limit their analysis to what is material only (Social Value UK, 2019c)? We address these questions by studying a sample of private corporations that donate to a cultural festival. There are a number of studies that assess the impact of cultural festivals (BOP, 2011(BOP, , 2016Fringe World, 2017;Fondazione Fitzcarraldo, 2012;Guerzoni, 2008Guerzoni, , 2009LDA, 2008;Maughan & Bianchini, 2004;Powell, n.d.;UN, 2010).
However, unlike our work, these studies do not focus on the reasons why organisations donate to festivals and whether or not they assess the impact of their giving. Our study also provides fresh insights into how businesses deal with the pressure to account for all of their activities, including their philanthropic giving (Miller & Power, 2013;Power, 2009).

METHODOLOGY
As discussed earlier, there is an expectation on corporations to engage with philanthropy and assess the impact of their giving. Our focus is on private corporations as they have lower pressure with respect to transparency than public corporations and, thus, the assessment of their giving would unlikely be subject to window dressing (Cho et al., 2012).
Moreover, we use qualitative methods, which presents an opportunity to integrate findings from existing studies that mainly use quantitative methods and target public corporations (see, e.g., Chen et al., 2008;Seifert et al., 2004). The research method used here is inspired by grounded theory (Corbin & Strauss, 1990;Glaser & Strauss, 1967). It aims to understand the social and psychological process underpinning the accounting action that the organisation engages with, along with the technologies, in the definition and determination of the value and impact of its giving. We follow a constructivist approach (Thornberg & Charmaz, 2013).
Our analysis is based on a case study approach (Yin, 2003). The case study is chosen based on its suitability to address the research questions. Additionally, the research context is familiar to the interviewer as she had lived in the surrounding area; the familiarity with the context is expected to facilitate access to the data, as interviewees might be more willing to engage with the researcher (Miller & Glassner, 2004). The aim of the case study is to identify the practice in use (Scapens, 1990) Miller & Glassner, 2004). Furthermore, an active interview style has been adopted (Holstein & Gubrium, 2004). The interviewer was aware of the shortfalls of the interview approach (Corbin & Struass, 2008;Holstein & Gubrium, 2004) and made all efforts to contain any negative effects. The researcher tried to refine her sensitivity during the interviews, as well as when analysing the data (Corbin & Strauss, 2008). We have also used some secondary data, which include official documents and information provided on the festival's official website and other online sources. Document analysis provides an additional source of information that is externally available, sourced through the festival's official website. We had studied the literature on festivals and accountability before the collection of data. Therefore, we had some background knowledge and had built expectations prior to the interviews (Corbin & Strauss, 2008).

The case organisation
A specific cultural festival, the Veneto Opera Estate Festival, was chosen as the focus of this study. This festival is of regional significance and takes place in the Veneto region, which is in north-east Italy. While the heart of the festival is in Bassano del Grappa, 36 local councils hosted events in collaboration with the festival in 2018 (OE, 2018b). All municipalities liaise with the management team of OE for programming and advertisement.
The OE is a publicly-run organisation, and the Bassano del Grappa municipality manages all its services. Corporations wishing to support the festival go through a formal but straightforward application process.
We interviewed all but one of the private corporations that support the festival. The businesses that support the OE trade internationally, and, some, are among world leaders in their sector. The businesses that were interviewed are: AGB (medium size -family owned, international business), specialising in hardware systems for doors, windows and shutters; Distilleria Nardini (medium size -family owned, international business), the longest family-held producer of acquavite, grappa and liqueurs; ETRA (large -local business), managing water and waste systems along with other public services; FAMI (large size -family owned, international business), producing industrial furnishings; MEVIS (large size -family owned, international business), specialises in springs; Pasta Giglio (medium size -family owned, international business), producing pasta; and PENGO (medium size -family owned, international business), specialises in the distribution of household objects. These companies started supporting the OE Festival at different points in time. One business has been a long-time sponsor since mid-1980s, four of them have been sponsoring the festival for several years or longer, and two of them have started their sponsorship in the same year when we interviewed them. Table 1 provides further information with regard to the size and turnover of the businesses.

Data collection and analysis
The festival manager, the councillor in charge of culture, the external adviser, and all companies were contacted directly, and all agreed to be interviewed. The interviewees were informed about the aim of the research. Additionally, they were informed that the interview content would be anonymised, wherever possible. All interviewees, but one, agreed the interview to be digitally recorded, and all interviews were transcribed. The interviewed managers are in charge of the sponsorship scheme within their company, and, although a single voice from each organisation was heard, they are the ones with insights with regard to the decision to support the festival and the assessment of the value of their support (Llewellyn & Northcott, 2007).
Interviews were undertaken in 2013, 2015 and 2018 (see Table 2). In 2013, the festival manager (coded: FM) and cultural councillor (coded: CON) were interviewed face to face. In 2018, we conducted a follow-up interview with the festival manager over the phone and also interviewed the external adviser (coded: GZ) via Skype. These interviews provided essential information on the development of the relationship between the festival and the corporate world.
There were three rounds of interviews with the corporate managers: all face to face at company headquarters in 2013, face to face and over Skype/phone in 2015 and over Skype in 2018 (for the coding of these interviews, see Table 2).
During the first round of interviews, the interview questions focused on gathering information on (1) the relationship the organisation has with the festival, (2) the rationale for the business to support the festival, (3) the relevance of information received from the festival organisation and (4) the analysis that was undertaken to assess and record the impact of the contribution. Three interviews took place with one of the business owners while the remaining interviews were with the chief manager or communication manager. The interviews were transcribed by the researcher as soon as possible after the interview took place. This process allowed for higher sensitivity towards the material as the researcher 'relived' the interview. Once the interviews had been transcribed a manual coding was carried out. At the first level of analysis, a free coding was undertaken (Corbine & Strauss, 2008). This was followed by a thematic analysis of the interviews to identify categories and nodes within the data, which helped us build the themes that are covered in the next section. This analysis highlighted a need for more information on (5)  During the second round, two interviews were conducted face to face, two interviews took place over Skype/phone, while two interviews could not be finalised. For the third round of interviews, all six businesses were contacted with only one agreeing to be interviewed. We also contacted two businesses that just became a sponsor. While both agreed to be interviewed, only one of these interviews took place. The new sponsor interviewed was asked the same questions as the others. Additionally, this final round of interviews allowed us to explore the impact of a new tax credit scheme on the decision to support the festival. As per the previous rounds, the interviews were transcribed by the researcher and manually coded (Corbin & Strauss, 2008). The material was assessed against the ones collected in the previous rounds. In total 16 interviews were undertaken, 12 of which with seven businesses, two with the festival manager, one with the external adviser, and one with the cultural councillor. The analysis identified emerging themes that were coded and aggregated into nodes. The coding highlighted factors of relevance such as similarities and variations between the different interviewees that are reflected in the sections below. Of note is that all corporations are private and, furthermore, with the exception of one they are all family-run enterprises. They do not have a specialist department that deals with philanthropic giving, and there are no foundations associated with them. The decisions related to philanthropy are taken by owners.

THE OPERA ESTATE FESTIVAL
In 2016, OE has been classified as the third most important interdisciplinary festival in Italy (Tich, 2018) by the Italian Ministry of Culture, which provides financial support to the festival. The Opera Estate receives funding from various sources. The Italian Ministry of Culture is their biggest funder and supports opera, theatre and dance performances.
Donations from the corporate world (including foundations) and ticket sales are the next largest sources of funds, with similar amount of funding obtained from the two sources. This is followed by funding obtained from regional and other smaller institutions.
As the festival manager mentioned: The statement reflects the feeling discussed by Silber (2012): philanthropists give with some resentment, as they consider the government to underinvest in socially relevant issues. The interviewed manager argues that the public sector has to support the businesses. The expectation that the government duly contributes aligns with Schuyt's (2010) observation that philanthropy should not dominate in the society. For the interviewee, the fact that their giving qualifies for a tax credit signals that they are making a noteworthy contribution to their external environment. In fact, the government publicly announces the names of the 'mecenati' (patrons) that qualify for the Art Bonus (Art Bonus, 2019).
The Art Bonus tax credit helped a corporation (the latest sponsor) to finalise its decision to donate to the festival: it [the tax credit] has been the push to say "let's do it, as there is this possibility let's take advantage of it". It was since some years we were thinking of it' . (X.-O) The change in the tax credit allowed some donors to transfer their 'gain' back to the OE.
It is worth noting that while corporations could sponsor a specific event and were allowed to place banners at their sponsored event in the past, this has not been possible since early 2000s, and the funds are given on an unrestricted basis. Currently, donors are identified as 'friends' of the festival and their names are collectively placed on the advertising material, for example at the end of the festival programme (OE, 2018b). There is a transition in the way the collaboration is taking place: While the government recognises the importance of private funding to cultural events, OE as well is exploring new ways of engaging with corporate donors. This report proved particularly important during the economic downturn to justify and to persuade some funders, particularly the public sector, to keep funding the event. However, the festival management recognised the need for more focused care towards the corporate world, and, in 2016, it engaged an external adviser with a special view on collaboration. This adviser describes his role as being a social weaver (Tessitore Sociale, 2018). The festival acknowledges that he is helping them to evolve in their approach.

The changing relationship between the festival and the corporate world
The adviser's vision is that all members of society should play an active role towards the well-being of the community/society (see also Tessitore Sociale, 2018). This manager had been surprised by the range of activities offered by the OE. The driver behind the decision to sponsor the festival was not only the tax credit but also the discovery that the OE is much more than a summer event. The year before sponsoring the OE, they had been invited to support a spin-off activity of OE called 'Dance Well' , which primarily targeted Parkinson sufferers and their family members (Dance Well, 2018 It is evident that the esteem factor towards OE has played an essential role in the decision to become a sponsor; the trust in the organisation as well as in the person sitting at the interface between the two worlds has been the triggering factor. This supports prior research that highlights trust as an important factor in guiding cultural relations (Karpik, 2010;Laughlin, 1996;ter Bogt & Tillema, 2016). Although the decision to become a sponsor is based on different factors, the esteem element emerges as the triggering element towards the final decision. This is, consequently, a dimension that NFPOs, such as the OE, need to carefully nurture. Some changes are taking place: OE points out that they make an effort to use an appropriate language in their communications with their corporate donors, asking professional advice before sending out such communications when necessary.
The following sub-sections analyses the corporations' views on their giving. It is divided into three parts: the reasons for corporate giving (section 5.2), the way corporations account for the information related to their giving (section 5.3), and the issues surrounding the measurement of the value of giving (section 5.4).

Giving to society while accounting for the workforce
The literature review highlighted that philanthropic giving is driven by the willingness to contribute to the good of society (e.g., Schervish & Havens, 2003 This highlights how strongly patronage is part of the family and business culture. There is passion for doing good and sharing part of the richness with the larger community. All interviewees deem the moral aspect as a key driver of their giving: it is the 'right thing to do'; interestingly this moral affiliation had been missed by earlier studies (Feliu & Botero, 2016). However, our analysis reveals that another factor that has considerable relevance when deciding to give is the potential impact of giving on the workforce: First [we do it], to help out the community. To enter into the social dimension and help the area where the company is located. Second, to give the possibility to the workers that work for our organisation, to take part in cultural events they would otherwise not have been able to attend. (6.-M) we prefer, as a company policy, to support cultural activities. We strongly believe in culture' , and 'we saw the opportunity to offer our collaborators [i.e. workforce] a possibility to get exposure to cultural Thus, caring, be it for the society or the workforce, is also a driving factor behind the giving (Schervish, 2014), corroborated throughout the interviews. With the exception of two, all managers see the event as an opportunity for their workforce to participate in shows that they might otherwise not attend. Employees interested in attending an event have an opportunity to get free tickets. These seem to contradict the expectation that corporate philanthropy is not affecting the employees, as some studies had speculated (Chen et al., 2008). Participation in these events is seen as a valuable activity that supports the development and morale of staff (as expected by Muller et al., 2014). In particular, three businesses consider the morale of their workforce to be a factor they need to pay close attention to. Interviewee 6.-M recognises the workforce commitment, satisfaction and motivation as crucial for ensuring the success. Philanthropic giving seems to enhance the perception towards the company (Gond et al., 2010;Zhao & Zhang, 2019) and the productivity of its staff (Guiso et al., 2015;Zhao & Zhang, 2019), and consequently is helpful in building the success of the business. This is very well expressed in the following quote: anything that helps to give enjoyment/fun 2 , in the way of getting away from routine activities to an activity that stimulates. We have to remember that all these [that we do is] to make sure that our staff feels well; when they feel well, they work better, and they are also giving a better contribution, and the business grows. (3.-O) This is further articulated by there is always the aim to enhance the cultural level of our workers. I think that if I facilitate workers to approach forms of art different from the ones they usually see, the company will benefit out of it. [. . . ] if I get used to seeing nice things, one that is badly done will jump out, I will immediately notice it. (X.-O) The success of the company is the result of the sum of multiple factors that is made possible when everyone is involved in the process striving towards perfection. It is interesting to notice how art is perceived as helping to enhance productivity. Philanthropy is used by almost all corporations as a way to benefit their employees, who are their key internal stakeholders. This is an important finding, which has not received much attention in the literature.
Currently, five organisations receive tickets for their workforce to access the festival events. It is interesting to notice that 3.-O often referred to the workforce as the factor that 'makes the business' . The workforce and the retention of talent seem to be of paramount importance, and it seems that the businesses realise that they have an essential capital in which they invest by keeping employees engaged and motivated.
An additional factor that seems to add motivation to become a sponsor of the festival is the national worker contract. For the past few years, the specific sector contract requires the employer to provide a defined value of welfare benefits to each worker. The access to the festival events provides one of the possible contributions towards this legal requirement. It is worth noticing that it is not simply the legal framework for tax relief that is influencing the donation but also the legal requirement for workers' welfare.

Enhancing visibility and building reputation
The focus of giving is not only on the local society and the existing workforce as it extends to the potential workforce: we would expect to benefit from an increased reputation, [and increase in] the knowledge of the brand [by the local population] which will allow the company to have a better reputation also when it is looking for the best workers in the area. (5.-M) The sponsorship of the festival is expected to help the business attract talented workers as the contribution to the festival enhances the image of the sponsoring organisation in the local population. Porter and Kramer (2002) expect that giving to the community helps to attract workers as they will be keener to relocate to a developed neighbourhood.
However, in contrast to Porter and Kramer, in this study, the business wants to attract the younger generation living in the catchment area. Organisations in this study seem to compete for the best local talent. This information needs to be read in light of local characteristics -in this region, individuals prefer to find a job that will allow them to continue living in the area where the extended family resides.
There is an indirect and positive expectation that cultural affinity will make the business seem more relevant to the local population: Maybe you do not know what it [the logo, the business] is but you start to associate it with culture.
Therefore, being a supporter of the festival increases the firm's brand value and makes it more appealing as a workplace to future generations of workers. The association with the brand/logo instils a cultural reputation and sends out a signal of togetherness with the community. This is consistent with the view that branding is a source of information and judgement (Callon & Muniesa, 2005;Karpik, 2010). It emerges that the interviewed companies indirectly assess the impact that the branding has on the community.
In more general terms, the decision to contribute to the festival is justified as follows: we look for activities [to sponsor] that are integrated with the local context, which will bring benefits to the business as well as to the population. (2.

-O)
Opera Estate provides some visibility to sponsoring organisations as the logos of sponsors are featured in the festival leaflets. There is, in the above quote, a clear signal that visibility is important (Brammer & Millington, 2006). Being part of an event that has social and cultural value is perceived as valuable. This association is interesting and unexpected as culture is not a core activity of this corporation.
Similarly, another interviewee highlights that the donation provided the opportunity to say 'we are there too' . To be 'there too' means to belong to a group of corporations that care for the local community and culture, and which have been selected to be associated with the long-established event. This can be further extended to the visibility that the festival offers to the specific companies. This is a return expected to be obtained by all funders. Visibility is based on the positioning of products. Through the festival, the patrons are exposed to the brands and logo of the supporting corporations. It is important to note that this visibility does not necessarily affect the sales of the organisation, as most companies do not have festival patrons as customers.

How to account for the impact of donations
With the exception of one manager, the companies do not express an interest in receiving additional information on the event's performances suggesting they prefer a lean provision of documentation. This contradicts the increasing pressure towards transparency, where increasing the amount of information that flows from one organisation to others is assumed to improve trust and legitimacy (Cordery et al., 2019;Hyndman & McConville, 2016;Power, 1999). We now explore this issue further.

Networking
The managers interviewed suggest that the information made available to them by the festival is satisfactory. According to the general manager of the festival, all donors receive a financial report on the festival performance as well as details on the attendance by the workforce of sponsors, based on the corporate free tickets submitted to access the event. The interviewed managers highlight that they could gather additional information when attending the events, as the Opera Estate manager and her team is always in attendance and are easy to approach. An informal way of gathering information seems to be more meaningful to these managers (Hyndman & McConville, 2016;Uddin & Belal, 2019), which helps avoid a ritualistic action that would create additional paperwork. According to Pitkänen & Lukka (2011), the dominance of informal feedback depends on the nature of the issue and organisational preferences, and informal feedback is vital in any organisation. By relying on informal feedback from the festival administration, the firms are essentially following an unofficial, non-routine mode of information processing rather than an official, routine mode such as a management accounting system (Earl & Hopwood, 1980).
Managers prefer informal feedback from the festival manager and her team who are recognised as holders of key insights due to their expertise and professionalism. As the corporations do not require detailed feedback from the recipient of their act of giving, the findings confirm that trust (further highlighted in Section 5.2.3) is an essential guiding element when the available information is incomplete, as might occur in cultural organisations (Karpik, 2010;ter Bogt & Tillema, 2016). Furthermore, the fact that the businesses prefer receiving informal feedback directly from festival organisers rather than spending time and effort on scrutinising the formal report, agree with the view that charities should engage more with storytelling (Connolly et al., 2017).
However, three businesses indicated that they conduct their own analysis based on their workforce's attendance to the festival events: Although, the majority of organisations seem to gather feedback from their workforce via face-to-face discussions.
However, one respondent indicated that the statistical information is trivial to his final judgement. Managers pay attention to the feedback from their workers through informal channels such as a conversation. Interestingly, none of the managers feels the need to introduce a more formal feedback system for workers to report on their experiences.
Furthermore, they do not consider this information as a proxy of the value of their donation. The positive impact of their donation on their workforce is not perceived as sufficient information to quantify the value of their giving, since it is far from fully capturing the whole impact of their donation. Rather than collecting formal feedback from employees and OE, these companies seem to prefer to rely on their own judgement.

Esteem reflects trust
The interviewees were asked to further discuss the way they assessed the impact of their contribution, as this insight is lacking in the existing literature. Surprisingly, there is not much interest in such an exercise: we did not measure it, and this wasn't the intent either. (1.

-M)
Philanthropy is considered useful for society, workers and brand value. However, there is no intention to assess the achievements against a particular expectation. As one interviewee put it: when choosing an event to support the problem always is.. whether or not we are sure the money will be used appropriately. This is the doubt that always afflicts us; how the funds will be used. The experience of this company owner shows that the business relies on the quality of the festival to deliver what has been 'agreed' , that is the promise of a future performance (Mouritsen & Kreiner, 2016). Through the years, the festival has provided feedback on quality and also on how effectively the resources were used. The trust between the company and the festival has increased over time, substituting the need for detailed accounting information. This is similar to what was evidenced in ter Bogt and Tillema's (2016) study of the relationship between local authorities and theatres.
In the case of unique events, trust rather than other forms of accountability is expected to guide relationships (Karpik, 2010). This also agrees with the communal form of accountability highlighted by Laughlin (1996) as the interviewed firms were not interested in establishing contractual agreements enforceable through judicial or quasi-judicial actions (Dubnick, 1998 Therefore, it is crucial for festival managers to foster trust and esteem when engaging with donors. This is consistent with the idea of downward accountability (see, e.g., O'Dwyer and Unerman, 2010) whereby the recipient of giving interacts and mutually learns with givers, rather than simply providing formal reports.
Overall, this section highlights that donors benefit from the informal communications they have with the festival managers and have a high esteem for the festival. But, how do they evaluate the information they collect in order to decide whether it is worth giving to OE? The specific type of the organisation (i.e. family ownership) allows more flexibility as well as insights into the top management when making a personal judgement. Therefore, there is no need to engage in complex valuation exercises.

The immeasurability of the value of donation
The difficulty of measuring the value of the donation also relates to the fact that they bring a slow development, and if well managed they can bring very successful results. However, as said before, a result that is not easy, or, better said, is not measurable in economic terms and is not While donations are among the costs of a business, it is evident that organisations do not find it useful to add further costs by attempting an evaluation exercise that brings limited benefits. The cost-benefit analysis is not considered to be relevant, and this stands in contrast with the suggestion by Raiborn et al. (2003) and the growth in the number of SROI-related tools available for use. The accountability is limited to informal data, which is at odds with the expectation that more extended analysis is or should be undertaken (Herbohn, 2005;Maas & Liket, 2011;Raibon et al., 2003;Power, 1999), although it corroborates the feeling of a damaging impact when subjective assessments are made (Moody et al., 2015;von Schnurbein, 2016). In fact, while at first the corporations suggest that there is no need for the measurement, further discussion suggested that the statement is driven by their belief that an objective measure cannot be obtained.

5.4.1
The entrepreneurial instinct -the personal judgement CSR is perceived as central to the organisation, but not the need for reporting on it. This seems to differ from the current trend of increasing CSR reporting (Gelb & Strawser, 2001), and the expectation that organisations assess the impact of their engagement with matters that are socially relevant (Kramer & Pfitzer, 2016;Porter & Kramer, 2011).
The above comments highlight that measuring the value of the donation would only have brought additional costs without any other perceived benefits. As discussed in the following, the critical issue is that the measure would lose its worth and would become meaningless as a result of the exercise of being measured. For the interviewed managers, there are aspects of a donation that could not be accounted for: They do not see the relevance of carrying out an exercise that would be imprecise and expensive. As one interviewee points out, some organisations might provide some information regarding the value of contributions, but there is the need to be aware that the value cannot be entirely accurate: I could make an evaluation. I worked for XX, a big company, some years ago. We made economic evaluations of these types of activities. However, we already knew before doing such valuation that the figure would not be exact, it was [just] a number, a guess.
There is an arbitrary attribution of imprecise information that might be made public, and once it becomes public it might be considered a true and precise value by the external world (Hines, 1988(Hines, , 1992. There is a misjudgement or exclusion of at least the imperceptible value of the donation (Espeland & Lom, 2015), which leads to mis-assessment: there is a certain part that cannot be measured. This manager provides an insightful metaphor, the factors that 'add flavour' to the activity are abstract. Trying to 'categorise' information would limit its value and a cost-benefit analysis would compromise the real value of the activity.
The 'aura' of the activity would be lost when trying to measure a singularity (Karpik, 2010). In fact, the ability to assess the value in abstract terms (rather than attaching a hard number to it) is a distinctive feature of an entrepreneur, according to the interviewee. When valuing singularities, the entrepreneurial instinct is a fundamental feature that dismisses ineffective estimates. The entrepreneurial instinct appears to weigh heavily on the personal judgement. It can direct a decision against the general expectations, leading a manager to invest in new areas. Thompson and Tuden (1959) and Earl and Hopwood (1980) discuss that as cause and effect relationships (the donation amount and the impact of giving in our case) and objectives become uncertain, judgements and inspirations start to play a larger role in decision making, explaining the weight the interviewee places on entrepreneurial instinct.
There is a general difficulty in precisely measuring the impact of any action, and this difficulty is even more intense

CONCLUDING REMARKS
In this study, we examine a sample of private corporations that give to a cultural festival. Cultural festivals are unique events à la Karpik (2010) and present significant challenges to businesses in terms of determining the value of their giving (Donovan & O'Brien, 2016;Ellwood & Greenwood, 2016;Hone, 1997;Micallef & Peirson, 1997). Our focus is on understanding why private corporations support a cultural activity and the processes they follow in determining the value of their giving. We conducted semi-structured interviews with the companies, all except one of which are familyrun businesses. The interviewees unanimously agree that it is hard if not impossible to measure the impact of their giving, which casts doubt on the usefulness of SROI measures (see Section 3.2.2). In fact, none of them undertake the task of impact measurement in relation to their giving, which they consider to be infeasible. They also reject the idea of carrying out a simplified assessment (Espeland & Lom, 2015). Overall, they argue that the effort put into the task of impact measurement would not be justifiable, not least because the outcome could never accurately represent the real impact of their giving. This argument lends support to some of the criticisms made against SROI measures (Cooney & Lynch-Cerullo, 2014;von Schnurbein, 2016;Yates & Marra, 2017). It also suggests that the problems with the use of an SROI measure are not limited to the growth in the number of SROI measures available (Phillips & Jung, 2016b) but are more generally related to their viability. Finally, while the interviewees are not interested in undertaking a formal impact measurement task, they are happy to receive feedback in a 'story-telling' format (Connolly et al., 2017;Isserman & Markusen, 2013). The latter is informally produced by the organisers of the cultural festival upon their request (see also Uddin & Belal, 2019;Hyndman & McConville, 2016). They also gather informal feedback from their workforce attending the festival events. Overall, interviewees exhibit a clear preference for unofficial, non-routine methods (see Earl & Hopwood, 1980) to collect information on their philanthropic acts.
The study also highlights the factors that motivate these businesses to give to the festival and that influence the managers' decision to give: They view their support as an opportunity to give back to the society, as an act of care (Phillips & Jung, 2016a;Schervish, 2014) that provides satisfaction for 'doing good' (Schervish, 2014;Waters, 2016).
Furthermore, they argue that, since their workforce attends activities organised by the festival, their giving would have a positive impact on the well-being and productivity of their workforce (Gond et al., 2010;Guiso et al., 2015;Zhao & Zhang, 2019). According to Gordon et al. (2016), such a positive impact would lead to an improvement in the performance of a company. Finally, the interviewees highlight that supporting the festival leads to an enhanced visibility, which is beneficial to their business in terms of improving reputation (Hogarth et al., 2018) and increasing their ability to recruit talented individuals. Seifert et al. (2004) examine corporate philanthropy in Fortune 1000 companies and find no significant effect on profits from corporate generosity. In contrast, our findings suggest that while a direct link between the amount of giving and profits may not exist, corporate givers still benefit from enhanced visibility, reputation and ability to attract talent, whose values are difficult to capture within the framework studied by Seifert et al. (2004). This further questions whether an attempt to measure the impact of philanthropic giving is worthwhile as it is clear that what is measured becomes the new reality (Hines, 1992(Hines, , 1998. Clearly, public corporations are under pressure from stakeholders to assess the impact of their giving. Such pressure is less for private corporations and especially for family-run ones, which are the main focus of this study. This explains why interviewees do not feel the need (or pressure) to undertake an exercise that they do not deem to be worthwhile.
The OE festival has been changing the way it engages with corporate philanthropists, encouraging an inclusive and friendly dialogue rather than a formal one with a pure focus on sponsorship. This is evidence of a new type of relationship emerging between givers and recipients (Browar, 2002;Radbourne & Watking, 2015) with a move towards a more communal form of accountability (Laughlin, 1996). Overall, esteem and trust (ter Bogt & Tillema 2016;Karpik, 2010) lie at the heart of the relationship between the private corporations and the organisers of the festival. In fact, our findings suggest that these two traits may pre-exist between the parties before a financial commitment is made by the giver. In general, informal communications or collaborations on other matters lay foundations of esteem and trust prior to a formal sponsorship agreement, which is in line with the process commended by Radbourne and Watking (2015). It also presents a case of downward accountability (see, e.g., O'Dwyer & Unerman, 2010) as the festival organisers are keen to interact and mutually learn with the firms that support the festival. Overall, the relationship between the parties is mainly governed by communal forms rather than contractual forms of accountability (Laughlin, 1996;Roberts, 1991;Roberts & Scapens, 1985).
Our findings can also be interpreted according to the matrix of corporate philanthropy developed by Bruch and Walter (2005). The philanthropic acts of the corporations we examine fit well within the 'dispersed philanthropy' category as their engagement with the competence orientation and market orientation are both low. While their philanthropic acts have potential to evolve towards 'constricted philanthropy' , it is unclear if they will ever attain 'strategic philanthropy' as they do not face much pressure from external stakeholders in terms of their decisions related to philanthropy. Their current engagement with the market orientation is limited to achieving higher attractiveness as an employer through their philanthropic acts and enhancing the welfare of their existing workforce. While Bruch and Walter (2005) explicitly recognise the former within the market orientation, there is not much mention of the latter. Therefore, our findings suggest that the market orientation can be expanded to explicitly account for philanthropic acts that aim to improve the welfare of current employees. This would also be consistent with the studies that document the positive impact of corporate philanthropy on the workforce productivity (Gond et al., 2010;Guiso et al., 2015;Zhao & Zhang, 2019).
The case study in this paper is built around a cultural festival supported by family-run firms based in Italy where the patronage of culture and art dates back to the Renaissance (Zamagni, 2010). This raises a number of interesting questions for further research. Does the fact that organisations that give to culture and art are identified as 'patrons' rather than 'philanthropists' imply such organisations have different return expectations with respect to their giving?
Do family-run private corporations in other parts of the world have similar views on supporting cultural activities and assessing the impact of their giving? We would welcome further studies that address these questions to verify whether our results are corroborated in their contexts.