The Financial Health of Political Parties in English Constituencies,

Transcription

1 The Financial Health of Political Parties in English Constituencies, RON JOHNSTON AND CHARLES PATTIE a This paper has been submitted for publication NOT TO BE CITED WITHOUT THE AUTHORS PERMISSION a Ron Johnston is a professor in the School of Geographical Sciences at the University of Bristol; Charles Pattie is a professor in the Department of Geography at the University of Sheffield.

2 The Financial Health of Political Parties in English Constituencies, Local political parties are important to several aspects of political life, but until recently no data were available on the financial situation of the local branches of England s main political parties. Using recently released data on the accounts for all constituency parties with a turnover of 25,000 per annum or more in 2005, this paper reviews the situation regarding their financial health. It finds very little evidence of substantial financial turnover for the vast majority of local Labour parties and an only slightly better situation for the Liberal Democrats: many more local Conservative parties meet the reporting threshold, but even so in almost half of all English constituencies none of the three parties reported a turnover of 25,000 or more. Political parties grassroots organisations play an important role in the political system. They provide connections between local communities and political elites; act as recruiting grounds for elected posts; raise funds for the central party; and, by no means least, provide the activists who engage in local political campaigning, particularly during elections (see Seyd, 1998). However, analyses of trends in local party membership suggest that these grassroots of the British party system are decaying: local parties are shrinking, and many of the remaining members are relatively old and inactive (see, for example, Whiteley and Seyd, 1998; Fisher, 2000; Seyd and Whiteley, 2004). At the same time, constituency parties have made growing use of new technologies, notably as campaign tools in recent elections: computers, desktop publishing facilities, fax machines, and increasingly and web resources are now widely employed (Denver et al., 2003). The health of local parties measured in a variety of ways is an important influence on British general election results, as well as an indicator of the general state of the country s democracy. Studies have shown, for example, that the more members a local (constituency) party has, the greater the intensity of its general election campaigns and the better its performance at the polls. (See, for example, Whiteley and Seyd, 2003; Fisher et al, 2006, 2007.) Others have shown that the amount spent on its constituency campaign as an indirect indicator of its intensity is also linked to a party s success (Johnston and Pattie, 2006, reviews this literature). All of the various indices of campaign intensity reach the same conclusions, therefore: the more intensive a party s local campaign, the better its performance. This is especially so if the party is out of government and challenging for power; in 1997 the Labour party benefited most from intensive constituency campaigns whereas in 2005 it was the Conservatives who did so (Pattie and Johnston, 2008). Intensive local campaigns require party members who can sustain them, both directly through their own voluntary labour and indirectly through money-raising activities which allow the deployment of fixed resources (such as constituency agents: Fisher et al., 2006b) as well as expenditure on posters, leaflets and other mobilising devices. Without both members and money it is unlikely that such campaigns will be viable, so 2

3 that, as Whiteley and Seyd (2002, 209) argue, if the decline in membership is uneven across the country then in places where they remain relatively strong local parties may gain a greater advantage in the contests for votes and seats. Furthermore, decline in party membership at the local if not the national level is undoubtedly linked to decline in financial resources; indeed, Fisher s (2000) study of a small sample of local parties shows a clear link between membership and income. The picture regarding the health of grassroots parties therefore suggests considerable decline in recent years, although little is known about their financial condition. Apart from Fisher s (2000) survey of 60 English constituencies, there is little available evidence regarding the financial health of local parties in Britain. In part, as discussed below, this has been an outcome of the regulatory environment, which has not until recently required local parties to publish their accounts. 1 However, recent legislation requires local party accounts to be returned to the Electoral Commission, if either income or expenditure exceeds a relatively low threshold ( 25,000) in any year. The Commission publishes these returns on its website and their availability for recent years allows systematic analysis to take place. This paper uses those data to provide a preliminary assessment of local parties financial health in England at the time of the 2005 general election. THE FINANCIAL REGULATION OF LOCAL POLITICAL PARTIES There has been little UK regulation of how political parties are funded nationally until the last decade. Constituency parties were subject to some regulation, to control how much their candidates spent during the brief official campaign immediately before a general election (usually between three and six weeks before polling day). The limits imposed in the late nineteenth century were changed under the Representation of the People Act 1918 and the Equal Franchise Act 1928 (Butler, 1963; Johnston, 1987; Ewing, 2007) and then by the Representation of the People Act They are now reviewed before every general election and the reported amounts spent are published. Candidates, through their agents, must also report on the sources of their income; these returns are not published and the vast majority inspected after the 1997 general election were very uninformative (Johnston, Pattie and MacAllister, 1999). The Political Parties, Elections and Referendums Act 2000 (hereafter PPERA: for a general introduction, see Ewing, 2007), in addition to the obligations on local parties and their candidates under the Representation of the People Act, requires that not only all registered political parties (the Registration of Political Parties Act 1998 provides the conditions for registration) but also all accounting units within those parties (such as a constituency party) make an annual return of their accounts. This only applies to accounting units whose gross income or total expenditure in a financial year exceeds 25,000 (Schedule 5, 6(1)). These returns are published on the Electoral Commission s website. 2 Under the 2000 Act the accounting units (AUs) must provide a statement of audited accounts that complies with such requirements as to its form and contents as may be prescribed by regulations made by the Commission (Clause 42 (2)); these can cover the accounting conventions to be used and what additional information should be provided in notes to the main accounts (Clause 41(3)). The requirements subsequently published by the Electoral Commission include: 3 3

4 A standard reporting period (calendar years) and submission within three months of each year end; An income and expenditure account summarising for the year all incoming resources to the AU and all expenditure incurred by it, analysed in accordance with how they have been used it doesn t specify particular headings but provides categories which can be appropriately analysed ; A balance sheet indicating the AU s current assets (fixed and otherwise); Comparable data for the previous year (although recognising that this may not be possible in the first year for which returns are made); and A review of the unit s political activities during the year including its membership and of its financial activities. Exemplars for both the income and expenditure account and the balance sheet are provided, and in addition a separate Best Practice Guide is available identifying a number of common errors in AU returns such as the statement of accounts not being consistent with the data obtained by the Commission from the separate returns of donations. 4 FINANCIAL STRENGTH OF ENGLISH CONSTITUENCY PARTIES The data we use here taken from the Electoral Commission s website relate to all English constituencies for each of the country s three main political parties Conservative, Labour, Liberal Democrat for which returns were available and usable. 5 The annual accounts for 2005 provide the main focus; as a general election year, the amounts raised and spent then should provide a clear indication of the financial strength/weakness of constituency parties. In almost all cases, the 2005 accounts also gave comparative data for 2004 as required under the Electoral Commission s code of practice. Only data for constituency parties are examined here. In some places each of the three parties has an AU covering more than one Parliamentary constituency. These are excluded from consideration because it is not possible to identify how much was raised and spent in each of the constituencies covered. In the case of the Conservatives, this involved some 20 constituencies (in Bristol, Derby, Newcastle upon Tyne, North Tyneside, Nottingham and the London Borough of Wandsworth); for Labour there were city-wide returns for just six places (including Manchester and Sheffield), covering 19 constituencies; and for the Liberal Democrats there were twelve sets of grouped constituencies (several of them covering London boroughs) which combined 29 separate seats. In addition each of the three main parties had regional units (not necessarily covering the entire country) which made separate financial returns; these too are excluded from the current analyses. England had 529 Parliamentary constituencies in 2005, which would produce a total of 1,587 separate AUs if the Conservatives, Labour and the Liberal Democrats each had a functioning local constituency party with a turnover of at least 25,000 in every seat. There were only 377 analysable separate financial returns for 2005, however, which, even allowing for the joint AUs in some areas referred to above, means that in three-quarters of the cases either there was no constituency AU for at least one of the three parties, or such a unit existed but its financial turnover fell below the threshold, or, perhaps, such a local unit did exist but failed to make the necessary return. 4

5 Furthermore, the 377 units which did make a 2005 financial return were not evenly distributed across the three parties: 252 (almost exactly two-thirds) were Conservative constituency AUs, 63 were Liberal Democrat, and only 42 Labour. Thus financial data are available for local Labour parties in only 8 per cent of all English constituencies suggesting that in the vast majority of cases either the governing party had no functioning local unit or that it did but with a turnover of less than 25,000 in a general election year (and this for a party which focused so much of its campaigning on key seats and many more than 42 of them over the previous decade). Similarly, though slightly less surprisingly given their much more restricted electoral base, the Liberal Democrats produced a financial return for a constituency party in only 12 per cent of cases. Finally, even the Conservatives made financial returns from local units in only just under half of the country s constituencies. In almost all cases the Conservative and Liberal Democrat local units gave comparable data for 2004 in their returns (250 and 61 constituencies respectively); only 32 of the 42 Labour units provided data for 2004, however. How big were these constituency AUs? We use four measures, with data for as many units as were available, for both years: income, expenditure, assets, and members. These are summarised in Table 1, which gives the minimum and maximum values, plus the quartiles and the medians (these being preferred to the mean and standard deviation because of the skewness in many of the distributions). In terms of both income and expenditure, there is a clear difference between Conservative constituency AUs, on the one hand, and Labour and Liberal Democrat units on the other. In general terms, the income and expenditure of the median Conservative unit was twice the equivalent values for the median Labour AU in both years with each increasing its income and expenditure by c. 12,000 between 2004 and 2005, clearly indicating greater activity associated with the general election. The median Liberal Democrat constituency AU had slightly more income and expenditure than the median Labour unit, but still substantially less than the Conservatives. 6 When the range of values is taken into consideration the differences are even more stark: the upper quartile value (indicating that 25 per cent of units had values even higher) for both income and expenditure was larger for the Conservative local AUs than the maximum value for Labour, for example. In terms of cash flows, therefore, not only were there six times as many local Conservative as Labour constituency AUs with income and/or expenditure exceeding 25,000 in 2005 but in addition the average Conservative AU was much more financially healthy than that for either of the other two parties. (Note that the minimum values indicate that some AUs for each party made a return despite having a turnover of less than 25,000. Only three local Conservative parties Batley and Spen; Castle Point; Thanet North reported both income and expenditure below 25,000 in 2005, as did one Liberal Democrat unit Rochdale; five of the 42 Labour party AUs which made returns Sevenoaks, Stevenage, Tottenham, Tyneside North, and Wellingborough all had both income and expenditure below that threshold in general election year.) This clear difference in financial health is also apparent in the data on assets, especially in a comparison of the Conservative and Liberal Democrat AUs: whereas the former had median assets of 43,767 in 2005 (slightly down on 2004, indicating 5

6 that the average party spent more than it raised in election year) the latter had five times less at only 8,029. Although the maximum value of assets for a Conservative local unit far exceeds that for either of the other two parties, however, the upper quartile for Labour units is greater than that for Conservative counterparts although note that data are only available for 30 constituencies, so we are dealing here with just seven constituency Labour parties compared to 60 for the Conservatives. In many cases, the constituency parties with substantial assets own properties (the Cambridge Labour AU had property valued at 300,000 in 2005, for example), from which they obtain considerable income (the Aylesbury Conservative AU reported rents of 194,000 in 2005). Despite the considerable wealth of a small number of local units, most have very few assets relative to the costs of their operations. The median Conservative AU among the 248 for which data are available had a ratio of expenditure to assets in 2005 of only 0.7, for example, with upper and lower quartiles of 1.5 and 0.3 respectively. Thus few Conservative constituency parties have sufficient assets (assuming that they are realisable) to sustain more than a year s annual expenditure; considerable and continuous fund-raising must be the norm. This is even more the case with the other two parties. For Labour, the median expenditure/asset ratio is 1.0 across the 36 local units (with upper and lower quartiles of 5.8 and 0.5), whereas for the 63 Liberal Democrat constituency AUs the situation was much worse, with a median ratio of only 0.2 and an inter-quartile range of Finally, most AUs (apart from those affiliated to the Labour party) provided information on their membership, at least for Given the very sparse data for Labour, the only viable comparison is between the Conservatives and Liberal Democrats. In general as shown by the inter-quartile range and median data Conservative constituency AUs had twice as many members as their Liberal Democrat counterparts, and the largest was some five times bigger (Cambridgeshire Northeast and Harrogate & Knaresborough respectively). Membership increased slightly in the Liberal Democrat AUs between 2004 and 2005, however, whereas in the Conservative units it tended to decline. Interestingly, there was only a weak correlation between the number of members and annual income in 2005: r 2 values of only 0.26 for the Conservatives and 0.16 for the Liberal Democrats. THE GEOGRAPHY OF LOCAL PARTY FINANCES Which constituencies have party AUs with sufficient income and/or expenditure to require them to return their annual accounts to the Electoral Commission? Two possible answers to that question are addressed here. The first links financial health to electoral strength, arguing that the more popular a party is in a constituency as measured by either the total number of votes it attracted at the last general election in 2001 or its percentage share of the votes cast then the better able it should be to raise money locally and sustain an AU with an annual turnover of 25,000 or more. The second looks at the marginality of seats, arguing that rational political action would involve local parties being more active (perhaps stimulated and assisted by the national party organisation) and thus raising and spending more money in those seats that might be won or lost at the next election. 6

7 Table 2 presents data relevant to the first of these answers. For each party, the same summary measures as in Table 1 are provided, comparing the constituencies for which a return of accounts was made in 2005 with those for which there was no return. For the Conservatives and Liberal Democrats, there is a very clear difference between the two groups of constituencies; those whose AUs submitted their annual accounts were much larger than those which did not, in terms of votes won at the previous general elections. The median Conservative local party which had a turnover of 25,000 or more won twice as many votes as one which did not, for example, whereas the comparable difference for the Liberal Democrats was more than three times. There were similar, though slightly smaller, differences in their percentage share of the votes cast. Whereas it was in the constituencies where they were more successful electorally that the local Conservative and Liberal Democrat AUs were able to raise and/or spend more than the threshold of 25,000 in 2005, this was not also the case with the relatively small number of Labour local units. Although there are differences below the median at both the minimum and the lower quartile the constituencies where the unit made a return were smaller in terms of Labour electoral support than those where it did not at and above the median there was little difference. Among the constituencies where Labour performed well at the 2001 general election the likelihood of the local AU making a return of its accounts was unrelated to the size of that performance. Turning to the issue of marginality, Table 3 identifies the number of constituencies in which each party s AU made a financial return, according to which pair of parties came first and second there in 2001, which won then, and the margin of victory. Thus, for example, the first block shows that there was a local Conservative AU with a turnover in excess of 25,000 in most of the seats which the party won in 2001, whether Labour or the Liberal Democrats came second, and whatever the margin of the Conservative victory. Among the seats where the Conservatives came second, on the other hand, there was much more likely to be a local Conservative AU large enough to make a financial return in most of the seats won by the Liberal Democrats (save the eight very safe seats, where there was a large Conservative presence according to this measure in only three), but only in the more marginal seats won by Labour. Of the 167 very safe Labour seats with the Conservatives in second place, the latter had a local party which made a financial return in only 18. The Conservatives had an active local party, it seems, both in the constituencies they held after an election which was close to their nadir for two centuries in terms of seats won, and in those they needed to win if they were to form a government again (many of which they held before 1997). A similar pattern of presences and absences is shown for the Liberal Democrats. A local AU made a 2005 financial return in most of the seats that the party won in 2001 with the Conservatives in second place, especially those in the two most marginal categories, and also in most of the seats where the Liberal Democrats came second to the Conservatives in close contests. Almost all of the small number of local Labour AUs which made a financial return were in constituencies won by Labour with the Conservatives in second place, but with no significant concentration in the more marginal contests. 7

8 These three distributions mean that there were relatively few constituencies in England in 2005 in which more than one of the main political parties had a local AU with a turnover above the 25,000 threshold. Indeed, only six had all three making a return, just one of which (Guildford) was an ultra-marginal after the 2001 election, with another (Shrewsbury & Atcham) a marginal. A further 22 had both of the country s two largest parties (Conservative and Labour) making a return: 19 of them returned a Labour MP in 2001, and twelve of them were either ultra-marginal or marginal contests between those two parties. Just one constituency (Cambridge) had both Labour and the Liberal Democrats but not the Conservatives making a return. Finally, there were 42 constituencies where both Conservatives and the Liberal Democrats made a return. In total, almost half of England s constituencies (248 of 529) had no local political party whose AU had either an income or an expenditure of 25,000 or more in 2005 a general election year. A further 209 had a single locally-active party with that volume of turnover, leaving 65 with two such AUs (in all but one case one of those was the Conservative party) and, as already noted, just six where all three were active. INCOME SOURCES AND EXPENDITURE PATTERNS The Electoral Commission provides a format for annual accounts to be delivered in, but many AUs did not employ it in 2005 (in part if not, in some cases, in total). As a consequence, although it is possible in the majority of cases to categorise both income and expenditure into a small number of categories, the exceptions mean that no truly comparable data sets can be compiled. For present purposes, a small number of categories for each has been identified and summary statistics derived, as presented in Tables 4-5. In these, because of those difficulties, the extreme values are of relatively little import, but the median and the upper and lower quartiles give an overall impression of variations in both income sources and patterns of expenditure. Four main income sources have been identified members subscriptions, donations, fund-raising and appeals. In a number of cases it was unclear which category some items should be allocated to, and there were other difficulties (whether the income from fund-raising was net or gross, for example; gross is the default here). Nevertheless, general patterns were clearly discerned, and are shown in Table 4 as the percentage of the constituency AUs income in each year derived from these four categories. The first three columns identify a clear difference among the three parties: the Conservatives raise much more of their income from members subscriptions than either of their competitors although this difference was much smaller (especially at the median) in 2005 than 2004 as local Conservative parties tapped other sources of income, notably donations, undoubtedly to sustain their general election campaign activities. 7 Even in a non-election year, however, subscriptions provide at most onethird of their incomes in the majority of cases, whereas for the Liberal Democrats the upper quartile value indicates that they formed only 3 per cent of total income. The Conservatives also gained much larger shares of their local incomes from fund-raising activities than either of the other two party AUs, especially in 2004 when the median shows 21 per cent of income from this source for them as against only 6 per cent for Labour and nothing for the Liberal Democrats. Only the Conservatives gained 8

9 substantial sums from specific appeals and then only in a minority of constituencies, even in Whereas subscriptions, fund-raising events and specific appeals all provided substantial proportions of their income for many Conservative AUs, donations were the main source for both Labour and Liberal Democrats; they also provided a substantial proportion of Conservative AU incomes, especially in election year. To some extent, these variations may reflect inter-party differences in interpretations of the main income categories; nevertheless the Conservatives clearly depend much more on financial support from their local members to support constituency parties both directly, through subscriptions, and indirectly, through fund-raising events and appeals. Turning to expenditure, three main categories are identified in Table 5: salaries (mainly for agents and secretaries); printing (mainly leaflets and other literature); and elections. Again, in many cases it was unclear how to treat some reported expenditure items in the accounts relative to these three main categories (some separated postage from stationery whereas others didn t, for example, and in a substantial number of cases the total amount spent on electioneering especially in 2005 was difficult to discern). As before, therefore, it is the general pattern as shown by the inter-quartile range which provides the best indicator of similarities and differences across the three parties. One clear difference stands out across the three parties; on average Conservative constituency AUs spent much larger proportions of their money on salaries than either of their competitors, with Labour spending nothing in this category in the majority of the constituencies where it made a separate financial return. Even in a general election year, salaries was the largest expenditure item for the median Conservative constituency AU. With regard to printing, on the other hand, the Liberal Democrats stand out as spending much more than their opponents in the average constituency, perhaps not surprisingly given the continuous campaigning strategy to which they are committed in the seats that they hold and those they hope to win at the next election (Cutts, 2006). Labour AUs on average reported spending more of their money on election campaigning than either of their opponents, notably in 2005 but also in the preceding non-election year. Few Labour AUs spend more than 25,000 in any financial year, however, and where they do the money is more likely to be targeted on election campaigns than is the case with the Liberal Democrats and, especially, the Conservatives. The latter are generally richer and more likely to have other, ongoing commitments, notably salary payments. They may still spend substantially on the constituency election campaign but, since that amount is legally constrained (see below), this additional spending in years with a general election is rarely more than 25 per cent of total expenditure, compared to 36 per cent for the Liberal Democrats and 58 per cent for Labour. The general election campaign was the most important activity for the constituency AUs in 2005; indeed, many noted this in the overview statements accompanying their financial returns, with comments in some cases on how much was spent, what it was spent on, and the outcome. The Somerton & Frome constituency Conservative AU, 9

10 for example, reported that it generated more election literature, knocked on more doors and erected many more posters than at previous elections. For Labour the Eltham party reported that the purchase of new printers enabled it to reduce the costs of leaflet production; leaflets cost the Tyneside North party 10,214.78, with other expenditure including 750 on an election address, 1, to a delivery firm, 1,000 on canvassing, and a further on leaflets. The Cheadle Liberal Democrat AU reported spending 11,543 on constituency newspapers in 2005, and that for Cheltenham returned expenditure of 15,479 on printing and campaigning and 13,291 on their election agent s account. The implication is that the constituency AUs which made financial returns for 2005 should also have been those which spent most during the official general election period, as reported in the mandatory separate returns to the Electoral Commission. This expenditure is limited according to a formula based on a threshold sum plus a variable amount to reflect the number of registered electors and whether the constituency was categorised as borough (i.e. urban) or county (rural); the minimum that could be spent in any constituency was 9,690 and the maximum 14,692, with a median of 11, Table 6 compares the frequency distributions for spending by each party in those constituencies where its local unit made a financial return for 2005 with those where it did not. In each case the median reported campaign expenditure was substantially greater in the constituencies whose local parties returned their accounts to the Electoral Commission than in those where they did not. The difference was especially large for the Liberal Democrats, for whom the median campaign expenditure by parties submitting accounts was more than three times greater than in other seats, and the upper quartile was also more than twice as large. There was also a substantial difference across the inter-quartile range for both Labour and the Conservatives. For all three, therefore, having a financially strong local party clearly enabled much more to be spent on the general election campaign. (There were, or course, also variations according to seat marginality: the average Liberal Democrat AU in an ultra-marginal seat spent 10,780, for example, whereas one in a very safe seat spent only 9,745.) CONCLUSIONS The clear conclusion to be drawn from the data presented here is that the majority of constituency political parties were not financially very healthy in England at the time of the 2005 general election. Indeed, in about three-quarters of the cases there was no local party which either had an income or incurred expenditure of more than 25,000 during that year. The situation was especially bad for Labour, with only 42 constituency parties making returns (out of a possible 529), and little better with the Liberal Democrats (62 returns); there were many more local Conservative parties whose turnover exceeded the legal threshold for making a return to the Electoral Commission, but even so those appeared in less than half of all constituencies. This widespread absence of constituency organisations making financial returns suggests that the three main political parties are extremely weak in many parts of the country. An annual turnover of 25,000 is not a high threshold especially for a general election year in which the local party could legally spend on average about 11,000 during the official campaign period itself and as much as it wished with some constraints in the previous four months. This does not necessarily imply that 10

11 the local parties in many constituencies were either moribund or unable to mount a viable campaign (based largely on voluntary labour and relatively small sums to meet campaigning costs), but it does suggest that there is little depth to the party grassroots in very many places especially for Labour and the Liberal Democrats. The lack of returns from constituency parties in some places is puzzling, because it is out of line with other evidence regarding political activity there, especially with regard to the general election campaign. Data returned to the Electoral Commission on individual donations to local parties, for example, suggest that more of them than made returns should have had a turnover of 25,000 or more. Labour constituency parties received more than 2million in donations from trades unions between , for example (as discussed in Johnston and Pattie, 2007); between July 2004 and June 2005, inclusive, 269 separate constituency parties received 797 separate donations (of over 1,000 each) totalling 1,450,000. (Comparable data for the Conservatives show 730 donations to 222 different constituency parties, totalling nearly 3million; for the Liberal Democrats the figures were 285 donations, 117 constituencies, and an income of 800,000.) One check regarding the pattern of spending is provided by data on constituency agents. The number of these, especially permanent agents, has declined considerably in recent years, as Denver et al. (2006) have described, and in many parts of the country either agents and other paid organisers are shared by two or more constituencies (with many appointed and paid for by the national and/or regional organisation in order to implement centrally-determined campaigning strategies; reflecting what Denver and Hands, 2002, call post-fordism or flexible specialisation ) or they are voluntary workers. Nevertheless, some constituency parties do still have their own agents. Of the 64 Conservative constituency parties in England which responded to Fisher et al. s (2007) survey of the 2005 general election, for example, 59 had a full-time agent; compared with this, of the 279 Labour constituency parties that responded only 10 had a full-time agent with a further 235 having a voluntary agent only, and for the 180 Liberal Democrat constituency parties the respective figures were 22 and Interestingly, of the 59 Conservative constituency parties with a full-time agent, five did not make a financial return for 2005; for Labour, this was the case with five of the ten; and for the Liberal Democrats for six of the 22. In some cases, therefore, either the salaries were low, or the employment was for part of the year only, or the payment came from other sources than the local party. The implication to be drawn from these comparative data is that some at least of the expenditure at the constituency scale, especially before and during general election campaigns, does not involve the constituency party organisations. Indeed, the accounts returned by a number of constituency parties provide strong hints that this is the case. For example, several Conservative constituency parties in Gloucestershire contributed to an organisation called Gloucestershire Conservatives (which did not make a separate financial return), according to a quota scheme; Cotswold constituency contributed 114,000 in 2004, for example, and 100,500 in 2005, as well as 115,400 to the central party s target seat support fund in And the Shropshire Conservative Group, an association of two constituency parties, incurred expenditure on behalf of both until being wound up on 31 May 2005, when the Shrewsbury & Atcham constituency association withdrew; its Campaign Director and 11

12 four clerical staff were made redundant. The Group s accounts are appended to those of The Wrekin constituency. It spent 59,000 on salaries in 2004 and 47,000 in 2005; its total expenditure was 135, in the former year and 162, in the latter which included 16, reported as election expenses in 2004 and 37, in 2005 plus 33, was spent on political literature in There is no indication of the relative proportions spent in Shrewsbury & Atcham and in The Wrekin. 10 For Labour, the accounts for the Derby South constituency party indicated that it gave 30,819 to the Derby Labour Party Campaign Fund in 2005, which may have obtained income from other sources (and there was no separate return by it as an AU), and other AUs similarly indicate that further units were operating outside the financial reporting requirements. 11 There is less evidence for this with the Liberal Democrats, and outwith the constituency AUs which did make returns there is little evidence of well-funded activities: the East Midlands Region Liberal Democrats, for example, reported income of only 21,701 in 2005 (but 103,873 in 2004) against expenditure of 22,092 (2004, 100,626 two-thirds of which was spent on the European Parliament election campaign), recording that the major item of expenditure in the year [2005] was the cost of support to held, target and starred seats the reported sum was 8, In other words, apart from the constituency parties and other local organisations which make returns to the electoral commission as AUs, there is a range of active bodies (especially in the run-up to a general election) associated with the political parties which appear not to fall under the PPERA requirements to make financial returns. 13 Notwithstanding their existence, however, the general picture presented here (and assuming that all AUs which should have made a return did so) is one of weak financial health for English political parties at the constituency level, with Labour in an especially parlous situation, although there are also few substantial constituency Liberal Democrat parties beyond the seats that they hold and a small number they have expectations of winning (mainly from the Conservatives). Among the local parties which made returns, Conservative AUs were generally larger (in terms of both members and income) and wealthier than their Labour and Liberal Democrat counterparts although few of them have substantial assets and those which do (as with the other two parties) mainly benefit from properties which generate rental income. The AUs that made returns were mainly in constituencies where the party performed well at general elections (either winning or being in close contention); no party had many active AUs in parts of the country where their electoral prospects were poor undoubtedly thereby contributing to a cycle of small, poorly-funded parties unable to mount successful election campaigns (for local as well as the national government), which further exacerbated their weakness in attracting members and income with which to tackle their problem. As well as having more financially substantial AUs, the Conservatives also differed from the other parties in other ways, most of which probably reflect the greater wealth and incomes of their members. Thus Conservative AUs raised a great deal of their income on average from subscriptions, appeals and fund-raising events, whereas both Labour and the Liberal Democrats relied very much more on donations. The Conservatives also spent more on salaries, indicating a more substantial local 12

13 presence than the other two, for whom volunteer labour was more important to sustain the party. As a consequence of these more permanent commitments, the Conservatives spent less, in relative terms, on election campaigns while still being able to match their competitors in absolute terms. The picture of English local party financial health presented here is not a very positive one, therefore. The nature of politics and political activity is changing rapidly, of course, and it may be that financially viable constituency parties are no longer necessary to the country s overall political health. Although local campaigning has gained in importance in the last two decades, this has been very much focused on a small number of marginal constituencies and it is there especially for the Conservatives and Liberal Democrats that apparently viable and active local parties continue to exist, with much assistance from the parties central institutions. These provide the organisational matrix within which more centralised, and centrallyfunded, election campaigns are focused and it may be that more local members raising and spending money are not particularly relevant to national goals. Nevertheless, even with modern campaigning methods it is hard to envisage a party system operating without a substantial local presence, and the data presented here suggest that this does not exist in many parts of the country. To that extent, the continuation of a viable, vibrant local political scene looks to be in jeopardy. NOTES 1 Indeed, until the last decade local (i.e. constituency) Conservative parties were independent of the national party structure and made only voluntary contributions to it if any: Pattie and Johnston, 1996, 1997) 2 At 3 At 4 At foraccountingunits_ E N S W.PDF. 5 These data are also briefly analysed by Ewing (2007, 84-85). 6 Fisher s (2000) data for a small sample of constituencies show even greater between-party differences, with Conservative mean income in 1997 four times that of the average local Labour party and 5.6 times that of the average Liberal Democrat. There were similar differences in expenditure: the mean for Conservative parties was 3.5 times that for Labour parties and 4.5 that for Liberals Democrats. The differences in membership were much smaller, however: in 1997 the average local Conservative 1.4 times larger than the average Labour party, and 3.8 times the average Liberal Democrat. 7 Differences between the parties in how members join and pay their subscriptions may account for some of these differences, but there is no evidence in the accounts of amounts being transferred to Aus from central funds from membership subscription income. 8 These data exclude Staffordshire South where the election was postponed because of the death of a candidate. It was held seven weeks later, when the funding limits for by-elections were applied. 9 We are extremely grateful to Justin Fisher for making these data available to us. 10 The Conservatives West Midlands Region Accounting Unit reported an income of 96,938 in 2005 ( 93,285 of which came via donations), and spending 93,137 on the general election campaign. 11 For example, the Trade Union and Labour Party Liaison organisation: North West Region reported expenditure of 40,500 in 2005 to North West Labour Party to meet its three stated objectives of maintaining and servicing the National Trade Union and Labour Party Liaison Committee, building the profile of the trades unions within the party, and the development of training and organisation programmes for lay activists at regional and local level with particular reference to the 2005 General Election. There was no separate return for the North West Labour Party. 13

14 12 The East of England Regional Party spent 177,143 in 29005, however, including 11,470 on a regional administrator/campaigns organiser, 12,139 on local party support, and 146,554 on campaign expenditure. 13 It may also be the case that Aus are being assisted without any transfer of money. For example, much publicity was given during 2007 to donations being given to the Conservative party centrally by Bearwood Corporate Services Ltd. (a company owned by Lord Ashcroft, a former party treasurer) that were being deployed to support campaigning in marginal seats that the Conservatives would need to win at the next general election if they were to get a majority in the House of Commons. From January 2003 to September 2007, as well as making cash donations of 1,297, to either Conservative Central Office or a constituency AU, Bearcroft also made donations in kind worth 1,288,565.62: in , Bearcroft s financial donations totalled 334,136 (of which 275,000 was donated in June- August 2007) whereas its donations in kind all to Conservative Central Office totalled 1,251,005 (of which 806,384 was donated in the first nine months of 2007). It would appear, therefore, that much local activity is now being centrally financed, although only the release of AU accounts for later years will allow this to be checked. REFERENCES Butler, D. E. (1963) The Electoral System in Britain since 1918 (Oxford: The Clarendon Press) Cutts, D. J. (2006) Where we work we win : examining Liberal Democrat campaigning at the sub-national level using a case study approach. Journal of Elections, Public Opinion and Parties 2: Denver, D. and Hands, G. (2002) Post-Fordism in the constituencies. In D. Farrell and R. Schmitt-Beck, editors, Do political campaigns matter? Campaign effects in elections and referendums. London: Routledge, Denver, D., Hands, G., Fisher, J. and MacAllister, I. (2003) Constituency campaigning in Britain Party Politics, 9, Ewing, K. D. (2007) The Cost of Democracy: Party Funding in Modern British Politics. (Oxford and Portland OR: Hart Publishing) Fisher, J. (2000) Small kingdoms and crumbling organisations: examining the variation in constituency party membership and resources. British Elections and Parties Review, 10, Fisher, J., Denver, D. and Hands, G. (2006a) Party membership and campaign activity in Britain: the impact of electoral performance. Party Politics, 12, Fisher, J., Denver, D. and Hands, G. (2006) Unsung heroes: constituency electoral agents in British general elections. British Journal of Politics and International Relations, 8, Fisher, J., Denver, D., Fieldhouse, E., Cutts, D. and Russell, D. (2007) Constituency campaigning in 2005: ever more centralization? In D. Wring, J. Green, R. Mortimore 14

15 and S. Atkinson, editors, Political Communications: the General Election Campaign of (London: Palgrave Macmillan), Johnston, R. J. (1987) Money and Votes: Constituency Campaign Spending and Election Results (London: Croom Helm) Johnston, R. and Pattie, C. (2006) Putting voters in their place: geography and elections in Great Britain. Oxford: Oxford University Press. Johnston, R. J. and Pattie, C. J. (2007) Funding local political parties in England and Wales: donations and constituency campaigns. British Journal of Politics and International Relations 9: Johnston, R. J., Pattie, C. J. and MacAllister, I. D. (1999) The funding of constituency party general election campaigns in Great Britain, Environment and Planning C: Government and Policy 17: Pattie, C. and Johnston, R. (2008) Still talking, but is anyone listening? The changing face of constituency campaigning in Britain, Party Politics, 14, Pattie, C. and Johnston, R. (1996) Paying their way: local associations, the constituency quota scheme and Conservative party finance. Political Studies, 44, Pattie, C. and Johnston, R. (1997) Funding the national party: changing geographies of local fund-raising for the British Conservative Party, 1984/85 to 1993/94. Political Geography 16, Seyd, P. (1998) In praise of party. Parliamentary Affairs, 51, Seyd, P. and Whiteley, P. (2004) British party members: an overview. Party Politics, 10, Whiteley, P. and Seyd, P. (1998) The dynamics of party activism in Britain: a spiral of demobilization? British Journal of Political Science, 28, Whiteley, P. and Seyd, P. (2002) High-intensity participation: the dynamics of party activism in Britain. Ann Arbor: University of Michigan Press. Whiteley, P. and Seyd, P. (2003) Party election campaigning in Britain: the Labour party. Party Politics, 9,

16 TABLE 1. SUMMARY DATA FOR CONSTITUENCY POLITICAL PARTIES WHICH RETURNED THEIR ACCOUNTS TO THE ELECTORAL COMMISSION, Conservative Income ( ) Expenditure ( ) Minimum 15,123 18,983 10,584 15,755 Lower Quartile 36,003 42,690 34,499 42,452 Median 52,995 63,514 51,101 62,720 Upper Quartile 74,850 91,608 72,134 90,242 Maximum 360, , , ,192 N Assets ( ) Members Minimum -37,187-25, Lower Quartile 19,971 20, Median 46,642 43, Upper Quartile 98, ,872 1, Maximum 1,400,477 1,401,383 2,881 2,829 N Labour Income ( ) Expenditure ( ) Minimum 4,004 10,543 3,096 12,914 Lower Quartile 16,203 24,752 10,492 28,022 Median 22,627 34,152 21,662 35,393 Upper Quartile 43,193 45,290 35,084 51,239 Maximum 76,728 89,422 58,029 79,831 N Assets ( ) Members Minimum 3,575 1, Lower Quartile 10,723 14,785 Median 48,055 35, Upper Quartile 200, ,609 Maximum 794, , ,217 N Liberal Democrat Income ( ) Expenditure ( ) Minimum 10,226 9,157 3,187 17,414 Lower Quartile 20,605 30,108 21,061 33,639 Median 28,877 40,086 29,435 42,343 Upper Quartile 41,305 53,734 40,996 57,372 Maximum 80, ,129 87, ,839 N Assets ( ) Members Minimum -5,973-13, Lower Quartile 3,926 3, Median 9,504 8, Upper Quartile 17,644 15, Maximum 492, , N

17 TABLE 2. THE RESULT OF THE 2001 GENERAL ELECTION IN THOSE CONSTITUENCIES WHERE THE LOCAL POLITICAL PARTIES RETURNED THEIR ACCOUNTS TO THE ELECTORAL COMMISSION, Result Number of Votes % of Votes Return No Return Return No Return Conservative Minimum 5,459 1, Lower Quartile 16,089 6, Median 19,347 9, Upper Quartile 22,133 14, Maximum 27,911 18, Labour Minimum 6,558 3, Lower Quartile 17,880 12, Median 19,746 18, Upper Quartile 21,724 21, Maximum 26,194 28, Liberal Democrat Minimum 3,065 1, Lower Quartile 13,776 4, Median 18,517 5, Upper Quartile 22, Maximum 32,282 29,