What motivates business to donate to politics? A framework and an empirical application

The secrecy, subtlety, and diversity of party-firm relations have proven challenging to synthesize. We introduce a new heuristic framework for the study business donors on the basis of pragmatic, partisan, and social motivations. We illustrate our framework with an analysis of donors to the UK Conservative Party from the property services and construction industries. While actors make it almost impossible to place particular events in our framework, in the aggregate the framework is empirically tractable. We triangulate a study of over twenty years of donations data, news on social events, and qualitative data to argue that the construction industry is motivated by a combination of pragmatic and social motivations, while the property services industry is best described by a combination of the social and partisan motivations.


Introduction
Much has been written about the motivations of political donors (Clark and Wilson 1961;Francia et al. 2003;McMenamin 2013a, b;Broockman and Malhotra 2020). There is a general acceptance that they encompass pragmatic, partisan and social motivations. However, most of this work focuses on the USA and provides little in the way of understanding the interaction between these different motivations. In this paper, we provide a comprehensive framework to study pragmatic, partisan, and social motivations, and their combinations. We show how they can be distinguished from each other empirically with a case study of the links between the property services and construction sectors and the Conservative Party in the UK to illustrate the framework in greater detail. We argue that a combination of pragmatic and social motivations is the dominant motivation for the construction sector and a combination of the partisan and social motivations drives donations from the property services sector to the Conservatives.

Theoretical framework: the three circles
We concentrate on donor motivation, as opposed to more general terms like interaction and relationship. Interaction is perhaps too behavioral a term: a meeting between a businessperson and a politician takes on a very different meaning if the motivation is a pragmatic pursuit of profit, a partisan commitment to a political project, or just a social occasion. The notion of a relationship is too general and complex. Motivation is more exact and allows us to focus on the donor, although we take politicians' incentives into account too. It is also important to point out that a focus on motivation facilitates a way into thinking about the meaning of interactions and the nature of relationships. Therefore, the below should be read as an outline for the way in which our framework can be a point of contact for, resonate with, and is inspired by several rich research traditions and researches.
The motivations that businesses have to donate to political parties/candidates can be thought of in terms of three motivations: pragmatic, partisan, and social. Pragmatic activity is interested activity, focused on increasing profits (McMenamin 2013a, b: 8). Partisan motivations exist when businesses support a particular political party without seeking a definite benefit for themselves. Social interactions between business and parties reflect either personal relationships between business leaders and party, or the desire to cultivate such relationships and/or a means to drum up much needed party funds via a process of holding events with invited guests (Francia et al. 2003: 67). Most of the existing literature can be located in this framework (Welch 1974(Welch , 1982Francia et al. 2003;Bond 2004;Della Porta 2004;Wilson and Grant 2010;Boas et al. 2014;Tomashevskiy 2015;Markus and Chamysh 2017;Harrigan 2017;Power 2020) and it provides a way of thinking about how the work of different scholars relates to each other. We review each in turn.
In the pure form of pragmatism, firms invest resources in political activity to win policy benefits that (they hope) will increase their profits. While business donations are an important part of the political system, political activity is a tiny proportion of overall business activity, and any investment of resources (in terms of time and money) in this area is unlikely to deliver a direct return. This raises the question of motivation: why bother? Policy-makers are time-poor, as well as information-poor. They cannot give a full hearing to all of the informationally-privileged business actors clamouring to lobby them. Moreover, politicians want to minimize the electoral costs of being perceived to make decisions for the good of businesses instead of voters. While business as a whole may be structurally advantaged (Lindblom 1977;Culpepper 2011) or informationally privileged (Bernhagen 2007), most business actors do not find it easy to present their information to whoever they choose, whenever they choose, and however they choose. So questions of gaining effective access remain important to these actors (Cooper and Boucher 2019). The value of access is well-articulated in the literature on Political Action Committee (PAC) motivations. Hall and Wayman (1990), for example, suggest that PACs are less interested in buying a specific policy outcome, but more in the access and time donations providesuch that they become a part of the creation of policy that will benefit them. This theory has been tested by a number of scholars who have found that PAC focus on the power of access is also affected by incumbency, electoral strength, and which party is in the majority (Cox and Magar 1999;Ansolabehere et al. 2003;Milyo et al. 2000;Barber 2016).
The pragmatic motivation essentially seeks an exchange between the business and political actors. A discrete exchange is explicit and simultaneous, but these remain uncommon in rich democracies (see Fisher 2009;Power 2020), as politicians are more likely to become mired in scandal by engaging in them. By contrast, a reciprocal exchange (Heidenheimer 2002: 774) is one in which each actor's part of the exchange is separately performed, with the terms left unstated and uncertain (Molm 2000: 261-2). They often involve more (and smaller) payments; reciprocal exchanges make it hard to associate a specific payment with a specific policy decision. They, therefore, reduce political costs sufficiently to allow politicians to accept, and seek, useful funding from business.
In a basic sense, the sale of access is a discrete exchange of cash for the chance to meet decision-makers or those close to decision-makers. In a much more important sense, that of a lucrative benefit to contributing businesses, the sale of access is a reciprocal exchange. Considering the different circumstances that influence levels of access-most fundraising events provide, at their most useful, only a brief, and low-quality opportunity to lobby. The 'language of access may serve to symbolically launder the money going from' business to policy (Hall and Wayman 1990: 800), but, ironically, it also may serve to taint meetings between businesspeople and politicians, where there is no opportunity to receive a hearing and state a case. On many social occasions, there will be too many people, and too little time, to even mention an important and sensitive issue, which is of particular concern to a firm. However, many access opportunities are consistent with reciprocal exchanges between businesses and parties. Such events serve to build and maintain contacts that might, on a later occasion, increase the chances of a real lobbying opportunity and the chances that such access achieves its aim of influence. High-quality access, on the other hand, consists of an often brief, and often secret, one-to-one meeting or communication with a decision-maker, as well as some real expectation that the 'pitch' will be considered favorably.
In practice, the distinction between discrete and reciprocal exchanges can be subtle. Indeed, businesspeople that mistake discrete for reciprocal exchanges can be disappointed. A ticket purchased as a lobbying opportunity can turn out to be a social occasion, at which lobbying is neither possible nor welcome. Instead, the rationale is that a sequence of such payments and meetings can develop in such a manner, in which a politician feels under an obligation to try to reciprocate when lobbied in the future (Bond 2007). Moreover, such meetings contain elements of both a reciprocal and a discrete exchange. There has been a discrete exchange of cash for the in-itself unimportant opportunity to share a social occasion with some politicians, and a reciprocal exchange of cash for the important opportunity to receive a policy benefit. The notion of reciprocal exchange also suggests there is a hard-to-observe continuum between asocial political investment and networking for business profit.
The pure partisan motivation resonates with the spatial theory of electoral competition (Downs 1957), which assumes that votes are cast on the basis of a selfinterested evaluation of parties' policies. Similarly, businesses can decide whether to donate to parties based on an evaluation of which party's policies would best serve them. Spatial theory describes a pure partisan motivation because there is no profit motive, and therefore no pragmatism, and no particular interest in attending social events. Pragmatism revolves around private goods, while partisanship is a focus on perceived public goods-one expresses a preference for government based on a particular set of values and assumptions (Francia et al. 2003: 67). In terms of businesses, this often manifests as support for a free-market ideology, but can also support other views of government and business, such as a developmental state.
Nevertheless, partisanship and pragmatism are not mutually exclusive, and the pragmatic-partisan interaction makes for a form of conditional investment. An offer of information, money, or other resources by business to a particular party may conflict with the partisan leanings of certain business actors. Its commitment to support a general policy position may preclude a pragmatic exchange in the hope of a profitable narrowly-targeted policy good. This could also work in reverse: parties may not want to offer access or policy goods to supporters of their opponents, regardless of the information, money, or other resources they could gain in return. Undoubtedly, co-dependence can mean that the unwavering commitments of election campaigns are replaced with a can-do attitude when a new government is formed. Nonetheless, pragmatic-partisan commitments are meaningful and can be observed in political finance in many countries (particularly where the party system engenders a larger effective number of parties), where business donates to both sides, but is more generous and consistent in giving to some parties than others (McMenamin 2008;Power 2020).
The combination of partisan and social motivations resonates with the literature on party identification (Campbell et al. 1960;Butler and Stokes 1969). One reason for this is that parties can provide a cue to the complex policy environment, so that businesses can save time and effort. Party ideologies or brands can provide a trustworthy shortcut that does not require a constant updating of positions in the policy space. Just as politicians do not have enough information about business and may be prepared to defer to expert practitioners; business does not have enough information about politics and may be prepared to defer to politicians. Socialization may also underpin the partisan-social motivation. Businesspeople spend a lot of time with other businesspeople and will influence each other politically to support the dominant choice of the business community. If business were conducted in isolation from other businesspeople, the political choices of businesses may be more diverse. In relation to individual voters, the classic socialization literature placed a great emphasis on the learning of a political identity and values during childhood (Campbell et al. 1960;Achen and Bartels 2016). Once established, such an identification can be highly resistant to the policy manoeuvres of parties. This applies less obviously to businesses, but, we might expect that, if your mother is a businessperson you are more likely to be a businessperson too and to learn her political values and loyalties as well. These social links between parties and businesses can, therefore, maintain identification when preferences are much less aligned (see Power 2020: 172-173).
Finally, we have a combination of all three. In this triple motivation, we should see an interlocking and reinforcing of motivations. The combination of the political, economic, and social is an important theme in political sociology (Mills 1959: 7-9;Shipman et al. 2018), but here we focus only on the motivations of politically active firms, not the wider and more complex systems of which the firms are part. Firms, largely, pursue increased profits. In doing so, they can simultaneously strengthen the social network of their leaders and their firm's partisan relationship with a party or parties.
For example, the rise in inequality in the US in recent decades has been attributed to 'organizational politics' in the sense of a determined and sustained selfinterested business campaign combining money, ideology, and networking (Hacker and Pierson 2010: 175-179). The 'Inner Circle' posits this very specific mechanism (Useem 1984). A group of businesspeople with multiple directorships develop the ability to scan the business and political environments, so that they can engage personally and collectively with parties on behalf of the narrow interests of particular firms and on behalf of a general preference for a particular party. These social ties can help to overcome collective action problems, and facilitate the support of a preferred political party (Bond 2004). Figure 1 shows this synthesis in the form of a Venn diagram. The space denoting a social motivation without a pragmatic or partisan motivation is left blank, as it is unpolitical. That said, it can be a genuine motivation-just as we all attend certain events simply because we can afford to, and we like the idea of going.
The above ideas are, of course, not startlingly original. We have taken inspiration from ideas about donor motivations, and social science more generally, to propose a heuristic that is more comprehensive than existing approaches, but retains the coherence of a single overall framework. We focus narrowly on motivation, but our framework also provides a point of contact for a range of thicker social science theories, which are highly relevant to lobbying and political finance. Next, we propose a strategy to engage with our framework empirically.

Operational hypotheses
It is difficult to measure motivations, as there are strong incentives for the participants to hide and obfuscate their behavior. However, by triangulating different indicators, we can make reasonable inferences. The best indicator is donation data, although this needs to be tracked over time and is most informative if there are changes in government. If it is combined with measures of social interaction and a political party's professed ideology and policies, we can make reasonable assessments of the motivation the business sector in general or a particular business sector has. The coding of at least one party as a business party or party at least somewhat more oriented toward business is helpful to distinguish the partisan motivation from social and pragmatic motivations.
In the following, we outline the observable implications of our theory for donations data, measures of social interaction, and party discourse. All of these implications are probabilistic associations between data patterns. To be placed in a category, the evidence should fulfil all or most of the indicators for the fundamental motivations (pragmatic, partisan, social). It should not fulfil any or many of the indicators for fundamental motivations excluded from the definition. Table 1 summarizes our triangulation strategy. For example, where there are donations by more donors when the party is in government and/or is likely to be in government in the foreseeable future we assume pure pragmatism. It does not require social interaction; neither does it require a party with an ideology or policies that are attractive to business. We will compare the property services and construction sectors to all other business sectors to evaluate whether they are more likely to fit into any of the categories than other types of British businesses.

Case analysis
We first consider the extent to which the UK is a least or most likely case; second the extent to which the related property services and construction sectors constitute a least or most likely case; and third whether our selected country-sector case presents any particular challenges in drawing inferences. The UK's political finance regime is relatively transparent, permissive, and provides scant public funding (Fisher 2009;Koss 2011;Piccio and van Biezen 2015;Webb and Keith 2017;Power 2020;Scarrow et al. 2017;Boatright 2015). The transparency of the British political finance system means that discrete exchanges are risky and relatively rare and that reciprocal exchanges are somewhat safer and more common. The UK's two main national parties compete on a left-right basis, which has made the Conservatives, at least until recently (McMenamin 2020), the natural party of business and encourages partisan motivations. The nature of planning creates valuable private goods for the construction and property services industries and consequently an incentive for pragmatism and a more controversial relationship with politics than most other business sectors in many countries, including well-governed ones like the UK (See Fellner and Mittelstaedt 2019 for Austria; McMenamin 2013b for Ireland; Brown and Loosemore 2015 for Australia; Sohail and Cavill 2008 for the USA). We do not think there is any reason why the property sector should have stronger or weaker social ties with parties than any other sector. There is a greater than average potential for partisanship between the construction and property services and the Conservative Party, since the role of the state in housing continues to be contested and maps on to the left-right dimension that continues to structure so many party systems.
The UK's transparent and permissive system of political finance means that reported donations are a relatively clean measure of business calculations and, therefore, make it potentially an excellent case for observation (McMenamin 2011(McMenamin , 2020. Moreover, the change of government in 2010 provides a crucial opportunity to test for pragmatism. We combine donations data with other evidence, especially in relation to the social and partisan motivations. We next introduce these data sources.

Data
We study all donations reported by all units of the Conservative Party since the first recorded donation on 12 February 2001 and 25 June 2020. We classify all companies and limited liability partnerships as businesses. The data was cleaned to reduce over-counting of donors due to typographical errors, spelling, punctuation, honorific variations, and donations by different sections or brands of the same business. A prominent example of the latter is JC Bamford Excavators, JCB Sales, JCB Research, etc. Nonetheless, there is still some small over-counting of the number of distinct donors. The sectoral classification is UKSIC from the UK Companies House as of August 2019 and we single out property services and construction. There were 6561 donations from 1873 businesses. In our regression analyses, we only study donations valued at £7500 and over due to a change in the reporting threshold in 2010. We count the number of distinct donors on a given day. This is intended to gauge donor support for the Conservatives. The donations data is summarized in Table 2.
It is harder to measure social motivations, but newspaper reports of fundraising events can provide some useful information. In order to roughly measure the social motivation, we search in newspapers for mentions of donors attending Conservative party dinners. We included all 'UK National Newspapers' in the Nexis database between 1 January 2001 and 30 November 2020. After some experimentation, we settled on the following word search.
Property OR Construction AND Conservative AND business AND dinner AND donation OR donor AND NOT Republican AND NOT Australia 1 and Conservative AND business AND dinner AND donation OR donor AND NOT Republican AND NOT Australia.
We took a random sample of 120 weeks and checked the validity of the article counts according to whether each article contained at least one sentence about donations and/or donors to the UK Conservative Party. We found that a satisfactory 77 per cent were valid for business as a whole. 2 We needed to separate property services and construction, so proceeded to code all hits for property and/or construction manually. Crucially, mentions of property development were assigned to construction, in line with the Standard Industrial Classification. We also noted whether each article referred to the date of the dinner or dinners. Thirty-six per cent had an unclear chronology, 21 per cent had occurred in the previous year, 24 per cent in the previous month, and 45 per cent on the previous day. Given the presence, but inexactitude, of the temporal references, we counted article hits by quarter. Table 3 presents summary statistics. 1 There were a number of stories that appeared covering politics in both Australia and the USA in our initial searches, so we eventually settled on removing articles which mentioned the word 'Republican' and 'Australia', which fixed this issue. 2 Counts of dinners including all business are therefore multiplied by 0.77 to avoid overcounting relative to property services and construction.
The donations data mentioned above can be used to make inferences about the strength of the partisan motivation. In addition to counting numbers of newspaper articles to evaluate the social motivation, we also conducted qualitative triangulation for each of the three motivations. For example, it is generally assumed that the Conservatives have a closer ideological link with the property and construction sectors than other parties in the UK (Hay 1992;Davies 2013;Farrall et al. 2016). To further draw out ideological links we consider this in terms of speeches and manifesto content over our period of study. We conducted a close-reading of Conservative manifesto commitments related to property, house-building and ownership, and construction to consider the way in which these policies were presented (e.g., as a long-standing ideological commitment). We did the same with speeches, first reading those delivered by successive Secretaries of State for Housing, Communities and Local Government and Levelling Up, Housing and Communities. Alongside this, successive Conservative Prime Ministers and their speeches to Housing and Construction Associations. In many political systems, like the UK, one party is especially associated with the business sector; in others parties compete to be seen as the political champion of business; and in others the concept is more problematic, because all or none, aspire to the title of 'party of business'.
There, of course, other ways of studying motivations that have different advantages and disadvantages than the evidence we gather and analyze. While important insights about behavior and motivation can be gained from surveys of businesspeople (Allern et al. 2020;Crepaz et al. 2021), this method is less useful than for many other subjects, due to concerns about admitting to corrupt or other questionable behavior. Surveys can also provide information on potential social ties between business leaders and politics. For the really famous business leaders, some of this information is available from open sources. In addition to the problems of non-response and less-than-frank responses, survey data tends to concentrate on the very biggest firms, which, notwithstanding their economic and political importance, represent a small proportion of donors and diners in the UK. We do not claim our sources are always superior, but only that they help us make a useful contribution to a literature comprised of multiple methods and sources. The replication data and code is available at: https:// doi. org/ 10. 7910/ DVN/ HMILK2

Empirical analysis
For ease of presentation, we will first analyze the evidence by data source and then plug our detailed conclusions into our triangulation framework for the property and construction sectors.

Donations
We restrict ourselves to donations of £7,500 and over to maintain consistency across the two decades. There were almost 100 donors from the property services sector and exactly fifty from construction that met this threshold. These donors averaged between 3.5 and 4 donations. As Fig. 2 shows, both were among the sectors with the most active donors, along with manufacturing, wholesale, and finance. We conduct negative binomial regressions of the number daily donors for the property services and construction sectors, but first, by way of a benchmark, donors from all other sectors. Our independent variables are whether the donation was made during an Election Campaign, the number Days after (an) Election, whether the Conservatives (were) in Government, and the Conservatives' Poll Lead over Labor. The poll data is from ICM's voting intention question. We include a trend to control for time effects and the standard errors are autocorrelation and heteroscedasticity consistent. We provide descriptive statistics on these data in the Appendix. Table 4 shows the regression results.
For donors outside of our sectors of interest, there have been more donations to the Conservatives when in government than when in opposition. Also, as the party becomes more popular, there is an increase in donations. There is a jump in donations during the formal electoral campaign and a gradual increase in donations as the parliamentary term moves toward an election. We analyze marginal effects to estimate the size of the effects (details in Appendix). For the other sectors, the shift to government led to a fifty per cent increase in the number of donations. Similarly, a move from 5 per cent behind Labour to five per cent ahead also led to a fifty per cent increase. This suggests a substantial, but hardly overwhelming, pragmatic motivation.
Surprisingly, property services donations were greater during the Conservatives' decade in opposition than their decade in government. Also, there is no significant association between the popularity of the party and property donations. There is a big increase in donations during the short period of the official electoral campaign and a gradual increase through the parliamentary calendar as a mandatory general election approaches. This evidence suggests that partisanship dominates, but provides no information on social motivations.
By contrast, there have been many more construction donations while the Conservatives have been in government than when they were in opposition. The number of construction donations is also much higher when the party is popular. Construction donations increase during election campaigns at a very similar rate to property donations. However, there is a much steeper increase in construction donations over the parliamentary term. Again we analyze marginal effects and again the details are in the Appendix. There is a massive, three hundred per cent, increase in daily donations from the construction sector when the Conservatives are in government. This is six times the size of the effect for other sectors. The increase associated with a ten per cent swing in the opinion polls is very similar at approximately fifty per cent. Taken together these suggest a powerful pragmatic motivation in the construction sector.

Social interaction
The property industry was mentioned in seventeen per cent of reports of dinners with business, which was almost double the sector's share of Gross Value Added (GVA) in 2010, the middle of our sample. 3 Mentions of construction (and property development) featured in almost twelve per cent of reports, also close to double its share of GVA. Although this is not an exact benchmark, it does suggest that interaction between the two sectors and the party was relatively substantial. We conduct negative binomial regressions of the number of quarterly national newspaper reports of dinners for the property and construction sectors, but first, by way of a benchmark, reports about dinners just mentioning business. This is a different benchmark to that used in donations. It is not meaningful to separate property and construction dinners from dinners attended by businesspeople in general. We employ three of the four independent variables from the donation regressions, but measured at a quarterly, not a daily, interval. We have not included the Election Campaign variable because fundraising dinners are not held during election campaigns. Table 5 provides the regression results.
There is a large increase in the number of reports about dinners for business in general when the Conservatives are in government. The number of reports also increases as the Conservatives become more popular relative to Labour. Finally, there are more reports as the parliamentary term moves toward an election. For the property services sector the increase in reports associated with government is over twice that for general business. The increase associated with the party's lead over Labour is over three times that business in general. There is not a significant relationship between the parliamentary term and the number of reports. Attendance at dinners by representatives of the property services sector would seem to be more pragmatic than attendance by businesspeople in general. For the construction sector, there is also an increase in reports when the Conservatives are in government and relatively popular. Again, the effects are much bigger than for other sectors. The increase associated with government is considerably smaller than for property, while the increase associated with intended votes is considerably larger. As with property services, we can conclude that attendance at dinners by representatives of the construction sector seems to be more pragmatic than attendance from other sectors.
Some of the increase in reports associated with government reflects greater media interest. However, it is also likely that the increase reflects greater social activity, by way of more events and larger attendances. The donations data suggest that more donors were interested in engaging with the party and many will have donated so that they could attend dinners. Similarly, the association between the popularity of the party and increased reporting may reflect greater journalistic interest, although this is less compelling than the norm of paying closer attention to the government. This caveat should not apply to the contrasts between the dynamics of the different sectors. It is not clear why journalists would increase their reporting of the property services and construction sectors relative to other sectors when the Tories entered government, unless there was something to report in terms of more engagement or more significant engagement. So, while this media data is much weaker evidence than the much harder and more accurate donation data, it can help us make inferences about changes over time and contrasts between sectors.

Qualitative triangulation
The construction sector has a greater interest in public and private goods from the political systems, as it depends on the development and construction of new projects. The property services sector manages existing buildings and property, as well as trading new and previously-owned buildings. That said, house completions count for forty to sixty per cent of house sales in a given year, so the property services sector also has an interest in ongoing development and construction.
Housing can be a valence issue or an ideological issue for Conservative partisans. Of course, everybody believes that a decent standard of housing should be provided to all. In that sense, it is a valence issue in which voters need to decide which party will most competently deliver the end that all agree on. Conservatives can argue that their practical approach to housing is more effective than Labor's supposed ideological fixation on public ownership. Conservatives can also present it as an ideological issue and often trumpet their commitment to private home ownership. This fits in nicely with a more general Conservative discourse of self-reliance and free markets.
Moreover, the link between Conservatism specifically and property owning, is one with a rich history. Indeed, the term 'property owning democracy' itself was coined in the early 1900s 'by a British Conservative [Noel Skelton], who hoped to adapt Conservatism to the arrival of a mass working-class electorate by proposing the diffusion of individual property ownership as an ideological alternative to the collective ownership defended by socialists' (Jackson 2012: 37). This mantle was then taken on by politicians such as Anthony Eden, Harold MacMillan, Winston Churchill and Margaret Thatcher (where Right-To-Buy became a central tenet of the party's attempts to transform the UK in the 1980s) (Francis 2012;Davies 2013).
Housing ministers and other senior Conservatives emphasized both the valence and ideological angles in conference speeches throughout the 2010s. For example, in 2016 Sajid Javid drew on the tradition of framing housing as an issue where the Conservatives take a practical approach.
Harold MacMillan put it best more than 90 years ago: 'Housing is not a question of conservatism or socialism', he said, 'it's a question of humanity'. Tackling this housing shortfall isn't about political expediency. It's a moral duty.
Prime Minister Boris Johnson asserted the ideological dimension in this speech to the 2020 Conservative Party conference.
We need to unleash the urge not just to build, but to own ...We will fix the long-term problems of this country not by endlessly expanding the state, but by giving power back to the people -the fundamental life-affirming power of home ownership, the power to decide what colour to paint your own front door.
The abstract commitment to private ownership does little to reduce conflict between existing private owners and aspiring private owners. Homeowners often resist more building in their area in order to preserve the character of the neighborhood and the value of their homes. Aspiring private owners would prefer more building to bring prices down. The Conservatives' increasingly aged members are mostly in the former group, while many of their voters, and indeed marginal voters, are in the latter category. The construction industry generally makes money by constructing more buildings, but the land banks of any given developer are more valuable the rarer planning permission to build is. Property service firms can also generally seek to make profits on a larger volume of transactions, but some firms may focus on a smaller number of high-value deals. This tension seems to have been well expressed during the Daily Telegraph's 'Hands off our lands' campaign. The government had proposed changes to planning permission to free up more land to address housing shortages and reduce the price of property. The normally loyal Telegraph, and a collection of rural and heritage groups, denounced the policy as having been made at the behest of property developers.
In the 2010s, the Conservatives had both a Business Forum and a Property Forum (White and Saner 2010; Blake 2011; Gillespie 2011). We are not aware of the existence of a specialized forum for any other sector, so this is an indicator of the closeness of the property industry to the Conservative Party (although the forum is now defunct). The forum presumably was, to some extent, a means of discussing general policy issues, such as the reforms to planning which ignited the 'Hands off our lands' campaign. Nevertheless, it appears to have been primarily a networking opportunity for pragmatic businesspeople. The Property Forum would set up 'breakfast meetings' between donors and party elites at which planning and property issues were discussed. At the time the attendance fee was reported as £2,500 and the forum itself was said to raise approximately £150,000 a year for the Conservative Party (Mackenzie 2011). This close relationship has been investigated since the 2010 election (Bloodworth 2014;Stone 2017;Christophers 2018;Jarvis 2018;Walker 2021). And, as Michael Slade the former chair of the Conservative Property Forum, admitted, You do run the thin line of someone saying: I'm only doing this to have access and influence, but that was what politics was always about. It is a little unfair but there must be 20 per cent truth in it. (Pickard 2011). Table 6 summarizes the empirical evidence on the property services sector. Donations suggest that most property services donors are not pragmatically motivated, thereby ruling out pragmatic, pragmatic-partisan, and triple motivations. The Conservatives appear potentially attractive to the property services sector (and often play up a link in speeches and election manifestoes), so a partisan motivation seems likely. There is a substantial social element between the sector and the party, thereby undermining the case for pure partisanship. The combination of a partisan and social motivation indicates partisan-social as a most convincing explanation. There is, however, a counter indication because there are more reports of dinners with the property sector when the Conservatives are popular and in government, perhaps implying a pragmatic rationale to these social occasions-and a potential triple motivation. Nonetheless, the increase in reports at least partially reflects an increase in the salience of the Conservatives and their donors for journalists. Table 7 offers a summary of the evidence on the construction sector. In this instance the construction sector combines pragmatic and social motivations. There is a large increase in donations when the Conservatives are in government and when they are popular and donations are frequent compared to other sectors. There is substantial reporting of attendance at dinners, which are strongly associated with when the Table 6 Empirical analysis of property services  Conservatives are in government and are popular. The next most convincing category is a triple motivation, combining pragmatic, social, and partisan. However, if the construction sector reflected the triple motivation there would be a substantial level of donations and dinners when in opposition. By contrast, the data show a massive increase in both when the Tories enter government, suggesting pragmatic donors seek to further their interests by networking with a newly powerful political party.

Construction
Our finding that the construction industry is primarily motivated around pragmatic-social motivations fits quite neatly into theories about party-firm relations and interpretations of British politics and policy-making. In the context of low corruption and a vigilant media, quid pro quo exchanges are too risky for businesses and, much more so, for politicians. Instead, pragmatic businesses manoeuvre to increase their chances of a less explicitly tangible benefit. Even a weak social interaction establishes, maintains, and bolsters the probability that politicians will feel an obligation to reciprocate donations with an opportunity to lobby or even a decision. This reciprocity makes it possible to deny a connection between a payment and a benefit and social activities have the added benefit of providing a cover story for pragmatism.
When social and pragmatic motivations are combined participants can claim, often correctly and sometimes disingenuously, that the occasion was purely social. Partisan motivations can provide another cover story. Participants can claim that payments, discussions and representations only related to public goods, not any private benefits for businesses. Just as motivations can actually be less innocent than they might be, party-firm interactions can be more innocent than some participants would wish. Sometimes businesses are naively cynical and expect that a payment will serve a pragmatic interest and are disappointed to be rebuffed by politicians who have taken their money. We believe few businesspeople will make this mistake more than once.
Our finding of partisan-social motivations for property services industry does make sense when compared to construction. The incentives for pragmatism and the salience of public policy are both lower for those trading ownership and use of property than they are for those developing and contributing to construction projects. Nonetheless, this distinction is subtle and we would not have been surprised had property services also fallen into the pragmatic-social category. While the property services firms often overlap with the construction sector, they also overlap with financial services. Notwithstanding the Tories' abandonment of finance in Brexit negotiations, this sector was traditionally socially linked and politically supportive of the Conservative party and our finding in relation to property services may make sense in that context. Like property services, the finance sector features a relatively high number of donations per donor. Of course, there may also be some idiosyncratic reason, that we failed to capture here, why property service donations did not increase as much after the return to power in 2010.

Conclusions
We present conclusions at three interlinked levels: the inferred motivations of the property and construction sectors, our theoretical framework, and the potential for generalization to other countries.
Firms from the property services and construction sectors donate more frequently and appear to attend more dinners than their equivalents from other sectors. There is also a generally positive relationship between Conservative ideology and their businesses and, to a lesser extent, a positive relationship between Conservative policy and the sectors. However, a more detailed analysis suggests different motivations between the two complementary sectors. The property services sector does not display a pragmatic pattern of donation. They are partisans of the Conservative party and their support is not merely a reflection of dispassionate calculation that the party's policies are closest to their preferences, but is underpinned and reinforced by an ongoing social dimension. So, they combine a partisan and social motivation. The construction sector does show a strongly pragmatic pattern of donation along with a strong social element, thereby placing it in the pragmatic-social category. The increase in donations associated with the power and popularity of the Conservatives is too steep to categorize the construction donors as partisans and to put them in the triple motivation category.
We have presented a new and comprehensive heuristic of the motivations of businesses. We think this framework is progressive (Lakatos 1978: 33): it is more general than previous proposals and we have demonstrated that it is possible to distinguish the different motivations from each other, even if it is difficult to do. The terms of our framework can be applied to donor motivations in any capitalist democracy with permissive political finance. Systematic comparative case studies using a triangulation of methods of sources in each unit should be possible and fruitful.
We suspect that the convenient ambiguity we find in the UK is associated with a context of low-corruption and high transparency. Since exchanges cannot be conducted in plain sight or total secrecy, this ambiguity is also highly functional. When challenged, politicians maintain that paying for access is not a form of bribery and that those seeking access may be partisan supporters and/or those looking for a very particular sort of social occasion or connection. This conjecture suggests, then, that similar behavior is to be found in other permissive, low-corruption contexts. If this is correct, a much more concerning question is: to what extent do existing party-firm networks of reciprocal obligation contribute to the emergence of grand corruption in established democracies? this policy issue is not a substantial confounder of for the big increase in donations while the Conservatives were in government. The debate on whether the Code for Sustainable Homes should be upgraded from a voluntary code to compulsion began   Fig. 2 does not appear to be consistent with the code being a major driver for donations. There is a big increase in the year before the ministerial statement, but this is likely to reflect the impending general election (see Power 2020 on the cyclical nature of UK donations). Over the half decade of debate (post-2010) on building regulations, construction donations were fewer than after 2015, when the construction industry had achieved a legislative victory (see Tables 8, 9, 10 and Fig. 3). Lowess smoothed, bandwidth = 0.25.
Funding Open Access funding provided by the IReL Consortium.

Conflict of interest
The authors have no conflict of interest to report.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.