Resumé

The study reported in this article compares local political activity of voluntary organizations in a Norwegian and an English city — Tromsø and Birmingham. The two cities display rather striking structural similarities with respect to the relative number of organizations as well as to organizational membership. The relative distribution of major types of organizations is also rather sinjilar, although there is a greater degree of organizational specialization in the English city.

With respect to political activity of the organizations the two cities are, however, quite different.
While two thirds of the organizations in Tromsø have been active in local political matters, this
holds true for less than 30 per cent of the organizations in Birmingham.

This difference is attributed to the finding that in the Norwegian case local government assists organizations with goods and services, while in the English case organizations have to depend on their own internal resources. An additional factor is that Norwegian organizations, when compared to their English counterparts, have greater access to political decision-making bodies.

Introduction

This paper compares the voluntary organisations of a Norwegian and an English city. It is particularly concerned with the local political activity of these organisations, and with explanations of variations in the frequency of such activity. The paper starts with an account of the number and type of voluntary organisations in the two cities, proceeds to an analysis of their internal organisational characteristics, then considers their rates of political activity, and" finally relates the characteristics of the organisations and their environment to variations in rates of political activity.

An elaborate justification for a study of voluntary organisations in contemporary
society is not necessary, for they have been picked out by a great
many of the major social and political theorists of industrial society for their

special importance. Known variously as voluntary associations, citizen organisations, secondary groups, or intermediary organisations, they play a major role in the theories of writers from the Tocqueville, Mill, Bryce, and Maitland, to Toennies, Durkheim, Simmel, Cole, Laski, Wright Mills, Parsons, Kornhauser and Dahl. These, and many other writers, have emphasised the capacity of groups to integrate individuals into society, to give a sense of security and belonging, to act as channels of communications between ordinary citizens and political leaders, to aggregate and articulate public opinion, to serve as training grounds for democracy, to force groups to search for political compromises in their competition with one another, to serve as centres of power outside the control of the state, and to act as agencies for the mobilisation of public opinion against arbitrary and undemocratic actions.l In keeping with their vital role as mediators between the individual and the state, voluntary organisations have more recently come to play a critical role in corporatist and neo-corporatist theory (Schmitter & Lembruch 1979, Rokkan 1966, Kvavik 1974 and 1976, Christensen & Egeberg 1979, Hernes & Selvig 1979), as well as in more empirical studies of national and local politics (Hutchenson & Shevin 1976, Gyford 1976, Byrne 1981).

In spite of all this, not much is actually known about voluntary organisations. To be more precise, empirical studies most usually fall neatly into two types: there are many sociological and behavioural studies of the people who join voluntary organisations, the joiners, and the effects this has on their attitudes and behaviour (Hausknecht 1962, Knoke 1981); there are also many studies by political scientists of particular pressure groups (farmers, doctors, antifluoridation groups, pro- and anti-common market groups, etc.) or of particular political issues involving pressure group campaigns.₂ But remarkably little is known about the total universe of voluntary organisations in modern society: how many there are, which interests they represent, how they work, what their organisational characteristics are, who runs them, what political role they play, if any, and with what success in what sorts of outcomes. Those who read and write the literature on voluntary organisations must now be weary of repeating these sorts of questions (Smith & Freedman 1972, Milovsky et al. 1980).

While the choice of voluntary organisations as a research topic needs no justification, the particular methods and research sites used in this particular study do. The data are taken from two quite independent surveys of voluntary organisations in a Norwegian city and an English city, namely Tromsø and Birmingham. It must be stated at the very start that a comparison of voluntary organisations in these two cities is by accident rather than anything resembling careful theoretical design. The authors happen to have completed quite independent studies in these two places and to have discovered after the event that they collected a similar body of data for similar theoretical reasons.

The data files resulting are compatible on a number of points of interest and importance concerning the role and significance of voluntary organisations in society and in community politics. Given the rather small amount of empirical research on community associations, therefore, it seemed sensible to make a modest amount of comparative mileage out of the two surveys. We should, however, explain the background of these studies before comparing their results.

Research Sites and Methods

On the face of it, a comparison of Tromsø and Birmingham does not look ai: all promising. Tromsø lies in the far north of Norway, some 400 kms inside the Arctic Circle. It is on the periphery of the periphery. By international standards it is not a large city, but with a population of 42,000 in 1972 it was the seventh largest in Norway. Tromsø was and is the administrative and political centre not just of its surrounding region, but of the whole of northern Norway. It is a major educational centre and has one of the nation's four universities, as well as a College of Education and other institutions of further and higher education. It is also an administrative centre with government offices and many of the commercial and service industries that go with a regional capital. When the survey of its voluntary organisations was carried out in 1972, Tromsø's main economic activities involved fishing and farming, boat building, canning and freezing, service industries, education, and transportation. With a good mixture of primary, secondary, and service industries the population is also mixed, with farmers and fishermen, white collar and professional, and public and private sector workers of all social classes.

The social and economic diversity of the population of Tromsø probably makes for a good breeding ground for voluntary associations. Even so, the city can scarcely compete with Birmingham in this respect. With a population of over a million, Birmingham is not just the second largest city in England, but one of the largest in Europe. At the time of the study it was a prosperous place which depended primarily on the motor industry, but which also had a wide range of other industries, both heavy and light. As the capital of a large metropolitican conurbation and a surrounding agricultural area containing several large cities and market towns, it is also a commercial and financial centre of major national importance. Its population, too, is mixed in terms of class, religion, race and nationality, for it has attracted workers from all over the United Kingdom and from the West Indies and Asia. The social and economic heterogeneity of the city combined with its size and importance as a centre of social, economic, and cultural life in Britain, suggests fertile ground

for a large, varied, and thriving array of voluntary organisations (Aiken 1970,
Danziger 1983, Clark 1971).

There is one major advantage of studying two places as dissimilar as Tromsø and Birmingham, even if an accident of history brought them together in the first place. If the same, or similar, social and political patterns repeat themselves in places as diverse as Tromsø and Birmingham, there is reason to believe that these patterns may be fairly general. This is the 'most different systems' research strategy which has been recommended in one widely quoted text on comparative methodology (Przeworski & Teune 1970), and it constitutes the logic of the present comparison.

The Birmingham study was divided into three sections or phases. First, as complete a list as possible was drawn up of all the voluntary associations in the city. This proved to be a larger task than anticipated because there turned out to be more organisations than expected, but sifting of four major sources and many minor ones eventually produced a long though not a comprehensive list. A very brief mail questionnaire was then sent out to a sample of 576 organisations in order to find out which had been active in local politics (political activity was defined broadly) in the previous twelve months. Third, the secretaries of a sample of seventy politically active organisations were interviewed in depth about their organizations' activities, and a control group of 108 politically inactive organisations was sent a mail questionnaire. Further methodological details have already been published (Newton 1976).

In Tromsø a list of organisations was drawn up on the basis of information provided by the municipal authorities in the city. This list was supplemented by additional information from other sources, in particular from various well-informed persons in the city. Altogether 130 organisations were identified in this way. Although the aim was to draw up a list of the universe of voluntary organisations, the list is not complete; nor does it represent a pure random sample of the universe. As in Birmingham, it was decided to register only what may be termed major organisations, leaving aside specific branches or subunits of these organisations. For example, a sports club counts as one organisation, although it may have been subdivided into various branches, such as a youth branch, a skiing branch, or even geographically separated subunits. The same applies to the trade unions, where major associations or confederations were registered, leaving aside their industrial branches. Furthermore, pure social associations and clubs were not included in the sample, nor were the small religious organisations linked to the Norwegian State Church.

One exception was made to this procedure. Because of Tromsø's settlement pattern, and because the municipal boundaries are drawn very broadly (Tromsø municipality covers 2,000 square kilometres), it was decided to include the outlying branches of fishermen's associations which were located within the

local authority. Apart from being trade union branches, these organisations also act as multipurpose organisations for the local communities in which they operate. Since the data were originally collected in connection with a study of structure planning in the Tromsø municipality, it was important to include these outlying communities and their organisations on the list. With these limitations, the list contained as many major organisations as was possible to identify. Judged from information received from other sources after the survey had been carried out, the final list seems to give a fair picture of the organisational community in Tromsø, although the number of organisations registered is somewhat less than the total universe. A mail questionnaire was sent to all 130 organisations, and altogether 94 (72%) of them responded.

The Number and Type of Organisations

In order to study the organisations in some detail it was necessary, first of all, to draw up as complete a list as possible for the two cities. The lists give us a basic idea about the number and types of organisations in existence, information of some interest in itself, and also serves as a sampling frame for the more detailed second stage of the survey. Drawing up a list for Tromsø proved to be no great difficulty, because the size and compactness of the city mean that the group world is fairly well known to inhabitants, and pretty visible to the outside researcher after a little digging. The altogether different scale of things in Birmingham might have led to severe problems, but, fortunately, it seems the larger the community and the better organised the group world, the more likely it is to provide information about itself. Perhaps the first law of voluntary organisations is: the greater the concentration of organisations, the more information they publish about themselves, and the easier are to study.

As the figures in Table 1 show, 130 voluntary organisations were found in Tromsø, and 4,264 in Birmingham. These two figures are worlds apart, but when they are expressed on a per capita basis they produce the first striking similarity between the two cities. Tromsø has 3.1 organisations per thousand population while Birmingham has a figure of 3.9.³ While these figures are close, the difference between them is also consistent with the observation that larger cities tend to have proportionately more voluntary associations. In other words, as the scale of urban life increases, so also does the density of associational life. The main conclusion, however, is not that Birmingham has more groups, but that the two cities are similar.

The figures for the total number of organisations, and for numbers per thousand population, may also tell us something about the nature and quality of life in the modern city. It is commonly assumed that urban life has tended to erode primary and community relationships to the point where a large proportion of the population is alienated, isolated, and uprooted, with a few

attachments to each other or to the wider community. However, there is considerable evidence to suggest that primary social ties are not especially weak in large scale urban society, and that alienation and uprootedness do not prevail in the modern city (Fischer 1975, 1976, Berry & Kasarda 1977). Moreover, our figures for the two cities suggest an extraordinary large number of secondary associations, which is not at all consistent with a picture of individual isolation and social disintegration.

In general the organisations in Tromsø had a rather larger membership than those in Birmingham, an average of 534 compared with 407. This is explained, in part, by the greater degree of organisational specialisation in the English city, where there are small and highly specialised groups which cater for almost every social activity, pastime, and interest, from Alcoholics Anonymous and the Acrostic Society to the Zoological Society and the Zen Buddhists. In contrast, the organisations in Tromsø, as one would expect of a much smaller city, are of a rather more basic kind, and they appeal to a broader cross-section of the population. Consequently some of the more obscure interests catered for in Birmingham (such as the West Midland Water Polo Referees Association and the Birmingham Mouse Club) were absent in Tromsø. It has also been argued that a strong norm for co-operation in Norway results in a tendency for membership of voluntary associations to be concentrated in a relatively small number of organisations (Eckstein 1966).

It may also be that slightly higher rates of social and political participation in Norway help to explain its marginally greater proportion of groups and their larger membership. According to surveys about 81 percent of Norwegians between the ages of 16 and 79 belong to one or more voluntary organisations, compared with a figure of 61 percent of electors in England and Wales.⁴ Once again, however, the striking thing about the two figures is not so much the difference between them, but the fact that both countries sustain a high level of involvement in groups.

Knowing the number of organisations in the city and their average membership size enables us to calculate the total number of organisational membership in each place. This is not the same as the total number of people who are members of organisations, a figure often calculated from samples of individual joiners, but a figure derived from multiplying the total number of organisations by the average size of their membership. In Tromsø the 130 organisations have an average membership of 534 members, which gives a total of 69,420 memberships. This is 165 percent of the total population of the city, which means that the average citizen belongs to 1.65 organisations. The equivalent figures of Birmingham are 4,246x407 = 1,735,448, or 172 percent of the total population. In both cities, therefore, the average number of the population (including all small children who belong to no organisations at all) belong to just short of two voluntary associations. The figures are close to those found in national surveys of organisational membership in the two countries. Once again the most striking aspect of the figures is how close they are in two such dissimilar cities.

Organisations can be classified according to their main activity, a fairly straightforward exercise in the majority of cases where the titles are enough for accurate classification. Sports organisations, churches, professional business, and trade unions, educational, youth, and technical and scientific associations are usually identified easily enough from their titles, although there are some cases where the name alone is not enough, and a little more probing is necessary. There are also some cases which overlap two or more categories — a youth club which is mainly concerned with running football clubs, for example, or a technical and scientific organisation which is also involved with

education — but in most cases the primary activity and membership of the
group is clear either from the name it gives itself, or from reference sources
which lists organisations according to their main activities.

The largest single category by far in both cities is made up of sports clubs and associations, cf. Table 2. These make up a third of the total in Tromsø and fully a half in Birmingham. No doubt the sports differ a little in the two cities, with football clubs in both, but cricket clubs in one and ski clubs in the other, but the two figures underline the importance of sport in the life of the modern community. Although most clubs seem to cater for sports activists rather than spectators, it seems that most cater for male sport, so perhaps our previous comment should be modified to underline the importance of sport in the life of males in modern society. The next largest category in both Tromsø and Birmingham involves activities which are likely to include as many women as men, however, and this is made up by the churches and welfare organisations. In both cities these kinds or organisations make up about a fifth of the total. After this the table has a long, thin tale consisting of a list of rather small categories of various kinds. The proportions involved are much the same in both cities.

The one figure which is glaringly different concerns trade unions, which constitute only 1 percent of the total in Birmingham but 23 percent in Tromsø. The disparity is so great compared with all the other figures, that it is likely to be caused by the difference in counting procedures, rather than the differences in the real world. In Birmingham each trade union was counted as one organisation, no matter how many branches it had. The largest union in the city had several hundred branches, but it was treated as a single voluntary organisation. In Tromsø, a substantial number of farmer and fishermen organisations for the surrounding villages were, as pointed out before, included in the total for the city. This has increased the number of trade unions substantially, and it has also affected the relative distribution of organisations in Tromsø across the various types. It is likely, therefore, that if the counting procedure followed in Birmingham had been applied in Tromsø, the two sets of figures in Table 2 would have been almost identical.

The Organisational Characteristics of Voluntary Organisations

Table 3 looks more closely at the organisational characteristics of the organisations. One of the most important of these is whether any given group can afford to pay for the services of a full-time administrator. With someone of this kind to manage the day-to-day affairs of the organisation it will tend to be more effective and efficient, but without it will have to rely entirely on voluntary and probably part-time help. An organisation which can raise the necessary funds to pay a full-time organiser has a considerable advantage,

therefore; it has a better chance to maintain its membership figures and collect its membership fees so that it can continue with the same scale of operations from one year to the next, and it can accumulate expertise in running voluntary activities and so has a good chance of accomplishing more than groups run on an amateur basis.

Quite probably the general image of voluntary associations is of a rather poorly managed set of rather disorganised activities relying upon the wellmeaning but rather ineffective efforts of a small army of middle-class ladies and retired men. The evidence suggests a different picture. About a quarter of the associations are run by full-time paid staff — 22 percent in Tromsø, and 28 percent in Birmingham, to be exact. Judging by the interviews conducted with these organisers, and with those who do the same work on an unpaid and often part-time basis, a large proportion of the organisations are run by capable administrators who are experienced in the running of both voluntary and other organisations. In general the group would seems to have attained a fairly high level of organisational development.

This observation is reinforced by the finding in both cities that about a fifth of organisations are of the second order. First-order organisations are those with a membership of individuals, and second-order organisations have a membership of first-order organisations. An amateur football club, with a membership of individual footballers, will belong to a League which organises matches between members of the League.. In its turn the Football League may well belong to a Sports Confederation which covers the full range of amateur sport from archery to yachting. It is possible to distinguish between first, second, third, fourth, fifth and possibly more levels of organisation in the group world, but for simplicity's sake a distinction is drawn in this study between organisations which have membership of individuals and those which have a membership of other organisations.s

Almost all interest groups are brought under some sort of second order or umbrella organisation (trade unions, and business and commercial organisations in particular, but professional associations are also organised to a lesser extent in this way) and so also are many of the cause groups, particularly charities and welfare organisations, and churches. In fact, for every major category listed in Table 2, there is likely to be an elaborate hierarchy of umbrella organisations which co-ordinate the activities of first-order organisations on a regional, national, or even international basis. The group world is indeed elaborate and complex.

In the preceding paragraphs we have used the distinction between cause groups and interest groups which is virtually universal in the literature.⁶ Organisations which represent people in their occupational capacities are interest groups, and they consist of trade unions, professional associations, and commercial and business associations. Cause groups are all the other organisations. There does seem to be a marked difference between the two cities in this respect, for almost half the organisations in Tromsø but only a quarter in Birmingham were interest groups. The cities are pretty comparable so far as business and professional associations are concerned, and the disparity is the result of the very large number of trade unions located in Tromsø, as already discussed.

A fourth important characteristic of voluntary organisations is the extent to which they are established in the community in general and in the political system in particular. Established groups are perceived by the community in general as representing legitimate and responsible interest in a proper manner.⁷ Professional associations of lawyers and doctors are most generally well established, compared with, say, militant left or right-wing groups of students, which are not. Groups can move from being rather poorly to rather well established in a few years, as ecology groups have done in the last decade, for example.

Five indicators of degree of establishment were used in the Birmingham study, namely whether the organisation had among its members (1) an elected, or (2) a co-opted member of the city council or its committees, or (3) an MP, or (4) any other member of another public body in the city, or whether (5) the organisation had been consulted by local officials in the city in the previous twelve months. In the Tromsø study two indicators were used, namely whether the organisation included among its membership a member of the city council, or a co-opted member of a standing committee. (Members of standing committees need not be members of the city council.) Given the more elaborate measure in Birmingham, one would expect a higher proportion to have an establishment score of one or more. So the figures in Table 3 show. In Tromsø 47 percent of the groups were classified as established (though this is a matter of degree, rather than an either/or concept), compared with 60 percent in

Birmingham. In both cases the high proportion of established groups suggests that a great many voluntary organisations are well accepted and well connected with local officials, and are likely to be consulted about political matters of interest to them. There are, in other words, close and extensive links between political officials and a large section of the group world. This leads us to the local political activity of voluntary organisations.

The Local Political Activity of Organisations

Voluntary organisations are generally thought to play a crucial role in modern democracies, particularly at the community level. They aggregate and articulate public opinion; they provide an organised base for various interests; they provide organised centres of power outside the state; they act as channels of communication between political leaders and citizens; some of them fight for the poorly organised and the unorganised (such as children or old people); they act as 'schools for democracy' in which citizens learn the skills of organisation, committee work, persuasion, and compromise; they provide organised networks of people who can be mobilised politically should they feel it necessary, and thus act as a long-stop against some forms or arbitrary government action; and they provide collective political resources for individuals who have few private political resources. It is not surprising that citizen organisations are the mainstay of pluralist theory which has dominated the study of both American politics and of local politics in the USA and in many other parts of the western world.B What is surprising is that so little empirical research has been done on the political activity of voluntary organisations.

The organisers of voluntary organisations in the two cities were questioned carefully about their political activity, broadly defined, in order to get an idea of the proportion involved in pressure group and political activity at the local level in a twelve-month period. The responses produced the first large and consistent set of contrasts between the organisations in the two cities: 63 percent of them had been politically active in the preceding year in Tromsø, compared with only 29 percent in Birmingham. The difference is substantial, and the Tromsø figure differs also from the figures arrived at in a similar study conducted by Selle and Svaasand in a sample of rural municipalities in western Norway, where they found a rate of participation of 21 percent (Selle & Svaasand 1982). Part of the discrepancy between the two Norwegian figures may be accounted for by differences in the counting procedure applied in the two studies. Selle and Svaasand included branches of major organisations as well as social and religious clubs in their sample, while — as noted before — the Tromsø study included only the major organisations.

However, variations in the counting procedures are unlikely to account
entirely for the difference between the two Norwegian figures. A supplementary

explanation would relate to differences in the communities studied. Selle and Svaasand drew their sample from smaller rural communities, ar.d according to the findings of Nokken, there are clear differences in the frequencies of such contacts between larger urban and smaller rural municipalities in Norway. While 75 percent of the councillors and administrators in the larger urban municipalities reported that they had been contacted by voluntary organisations, the figure for smaller rural municipalities was just above 50 percent. Furthermore, she found that the volume of local government activity — measured as per capita expenditure figures — was positively and strongly related to the frequency of organisational participation. She concludes that population size, density, social complexity and level of local service provision is positively and significantly related to political activity by voluntary organisations (Nokken 1976). This would account for some of the difference between Selle and Svaasand's findings for small rural communities, and the results reported here for Tromsø.

It should be added that the Tromsø figure is rather similar to the figure
found by Bratbak for a medium-sized North-Norwegian municipality, using
a similarly broad definition of political activity (Bratbak 1979).

Thus, there are methodological differences between the studies in Tromsø and Birmingham, but such differences can hardly account for the large disparity in the rate of participation. It is difficult to avoid the conclusion that there is a major difference between the two countries. In part this may be explained by the fact that local government has a special significance in Tromsø, which is separated by over a thousand kilometres of (often snow covered) mountains from its capital city. Probably more important, rates of political participation and general community involvement in Norway are high by the standards of most of the wealthier nations of the western world, and this seems to be reflected in high levels of group participation (Kjellberg & Hansen 1979).

In short, the political system of Norway seems to be more open, participatory, and democratic than that of the United Kingdom where, as some writers have pointed out, local government tends to be relatively more closed and secretive.⁹ Consequently, widespread group involvement in local government affairs may be somewhat more tolerated, or openly welcomed, in Norway than in the UK.

Not only are a greater proportion of groups politically active in a twelvemonth period in Norway, but these groups seem to be active on more issues. In Birmingham the politically active groups were involved in an average of 1.4 separate issues in the preceding year, whereas in Norway the average was 3.5 issues. Eight out of ten of the politically active groups in Birmingham were involved in one or two issues, compared with half the Tromsø groups which were involved in three or more issues.

In addition the modus operandi of groups in the two countries appears to be significantly different. Whereas voluntary organisations in Birmingham tend to approach local government bureaucrats and appointed officials in the first instance if they want something done about a public matter in the city, groups in Tromsø tend to use political channels rather than administrative ones. In the British city, sixty percent of the groups approaches a local government officer or department to get something done, whereas in Tromsø sixty percent approached an elected politician or a political party or appealed to the electorate via the mass media.¹⁰ This suggests that issues are rather more likely to be treated as political matters in Norway, compared with Birmingham where groups are inclined, initially at least, to use routine, bureaucratic and departmental channels, and to turn to political actors and organisations only if they fail initially and decide to pursue the matter. The Norwegian pattern appears to be closer to the American one than to the English, at least at the local level (Eisinger 1972).

A willingness to use political means to achieve their ends may also help to explain why voluntary organisations in Tromsø are more likely to co-operate among themselves when trying to influence the course of public affairs. A third of them combined with other voluntary organisations in order to achieve their political goals, compared with only a fifth in Birmingham. Such a readiness to build coalitions and to act in concert with other interests and organisations is said by many theorists to be a typical feature of modern pluralist democracy and one which has the important consequences of spreading and diffusing political conflict (Banfield 1961, Dahl 1963, Polsby 1963). It also has the effect of forcing compromises between groups which must, perforce, look around for allies in order to achieve their political goals. However, such coalition building is typical of only a minority of groups in the two cities, and to this extent some considerable doubt is thrown upon the pluralist picture of community politics as a process of coalition-building, wheeler-dealing, log-rolling, and

compromise. Although Tromsø seems to be somewhat closer to the pluralist pattern, in both cities most groups act in isolation of one another and try to secure their own goals without the aid (handicap?) of outside allies. Allies usually exact their price, which is exactly the point the pluralists wish to make, and if you can get your own way without paying this price, then why bother to search for friends and allies? Harold Laski once observed that the formula for success in politics was to form yourself into a pressure group of one.

Organisational and other Determinants of Local Political Participation

As observed, the most significant difference between voluntary organisations in Birmingham and Tromsø relates to their involvement in local politics. In the previous section some suggestions were made as to the causes of these differences. This final section of the paper will look at this question in some more detail. More precisely, it will examine the extent to which there are systematic differences in organisational resources between the politically active and inactive organisations in the two cities, and whether any variation may be detected between the two cities in this respect (Galaskiewicz 1981, Solvang 1977, Selle & Svaäsand 1982). From the aggregate figures we have seen that the similarities between Tromsø and Birmingham with regard to organisational characteristics do not repeat themselves when we look at the political activities of the organisations. This suggests that the relationships between such characteristics and political activity may differ in the two cities.

In order to examine these issues we have settled for a simple causal analysis, where the total number of variables included has been restricted to four. As dependent variable we use 'political activity in the last 12 month period' (as previously defined). It is treated as a dummy-variable distinguishing between active (1) and inactive (0). The three explanatory — or independent — variables are the organisational resources of the groups, as follows: (A) membership size; (b) full-time paid organisers/officers; (c) degree of establishment.

In order to standardise and simplify the comparison across the cities, these variables have also been recoded and treated as dummy-variables. For the two last variables we have used the same distinction as in Table 3, so that organisations with full-time officers and established organisations are given a score of 1. Size of membership has also been dichotomised into a dummyvariable. The distinction between what we term 'large' (1) and 'small' (0) organisations is, however, somewhat arbitrary, and also differs slightly between the two cities. This is because the original coding in the two surveys makes it impossible to retrieve the exact membership figures of each organisation. We do not, however, expect this to distort the comparability of the results. In

Tromsø, organisations with more than 200 members are coded as 'large', whereas
'larger' organisations in Birmingham have more than 300 members.

In Figure 1, the causal model is presented. The hypothesised relationships between the variables are indicated by plus and minus signs. We expect all these 'explanatory' variables to be positively related to political activity. As regards the internal relationships between the 'explanatory' variables, we expect size and the presence of full-time officers to be positively related to degree of establishment. Furthermore, we expect size and full-time officers to be positively related.

The analysis of the model is based on samples of organisations in Tromsø and Birmingham. The samples are not random in a purely statistical sense, as noted before. The data from Birmingham are drawn from two separate samples, one (a statistical sub-sample) including the politically active organisations, and the other the inactive organisations. Furthermore, the two subsamples were deliberately made approximately the same size. Caution should therefore be taken not to use these data as estimates of the universe with regard to the distribution of values on each of the four variables included in the analysis. However, our concern here is not with distributions (which we already have considered in the previous sections), but rather with the relationships between the variables in the model. For this purpose a pure random sample of observations is not necessary. Rather, we assume the causal model itself to be stochastic, which permits us to employ traditional procedures for significance testing.

The relationships have been estimated by regression analysis based on the
following structural equations:

In Table 5, the estimated path-coefficients for both cities are reported.

With one exception, namely the direct effect of size on political activity, all coefficients for Birmingham are significant. Size, has, however, a positive indirect effect on political activity for the Birmingham organisations, which makes the total effect of the variable positive — although hardly significant. In both cities, the strongest direct effect — and total causal effect — on political activity comes from degree of establishment. As noted before, this variable may be regarded as another aspect of political activity of voluntary organisations, and this finding indicates that these two modes of participation represent supplementary, rather than alternative ways of exerting influence on local decisions.

In the case of Tromsø, most coefficients are in the hypothesised direction, though none are significant. It is also interesting to note that neither in Tromsø, nor in Birmingham, does size of membership have any effect on an organisation's political activity. In other words, contrary to expectations, size in itself does not seem to represent any significant political resource for an organisation. Size may, however, constitute a basis for the mobilisation of other — and more relevant — political resources (such as money and full-time officers), thus having an indirect effect on political activity as observed in Birmingham.ll

Let us now turn to the question of how the differences in the relationships between Birmingham and Tromsø are to be accounted for. Before suggesting more substantial explanations, it should be stressed that the differences may be of a purely statistical — or rather stochastic — nature. As noted above, the estimated coefficients are, with two minor exceptions, in the same direction in both cities, the only differences being in their magnitude and level of significance. A larger number of observations in Tromsø may very well make the coefficients significant.

A more likely explanation relates to differences in access to the political
decision-making bodies, where such access is much more resource-demanding
in Birmingham than in Tromsø. In other words, to the extent that the local

political system is more open and encourages more participation, then political activity is likely to be less difficult and costly than in Birmingham. It may also be that a smaller city with a more accessible and less complicated form of local government is more likely to engage the activity of community organisations.

However, political resources or accessibility to decision-makers are a necessary, not a sufficient condition of political activity. In addition, there is the whole matter of what, if anything, is to be gained from such activity. In other words, to what extent do local organisations in the community depend on municipal services in order to carry out their activities, or are they independent in this respect? Although we have no firm evidence on which to base a judgement, scattered information indicates that Norwegian voluntary organisations are more dependent on local government services than English ones. If the amount of money spent by local authorities is an indicator, then it is significant that local government in Britain spent 17 percent of GNP in 1976, while in Norway the figure was almost 21 percent in 1974 (Newton et al. 1980). On a more immediate level, almost all football grounds, swimming pools, and gymnasiums in Norway are owned by its municipalities, which have also built community centres (usually with the financial support of central government) to house the activities and the organisational requirement of local voluntary groups. This is less true in Britain, where such physical facilities are more likely to be in private hands. This explanation reverses the normal logic which argues that the higher the level of political inputs (individual and group participation) the higher the level of government outputs is likely to be. The reverse may also hold: the more governments do, the more individuals and groups are encouraged to participate to gain influence over exactly what is done and for whom. As one recent American study concludes: 'groups relying on external (government) funds are more likely to make political demands' (Lewis & Clarke 1980). We do not wish to enter into a fruitless chicken-and-the-egg discussion about which comes first, inputs or outputs, but simply to argue the possibility that a high level of one will tend to reinforce a high level in the other.¹²

However, a simple distinction between inputs (group activity) and outputs (government activity) also breaks down in the context of local organisations, where a good deal of activity on the part of some local voluntary organisations contributes to government outputs. In a study of budgeting in Oslo, Brofoss observed that several voluntary organisations, in particular social welfare organisations, acted as implementing agents for the municipal authorities. This applies in particular to services like children's day nurseries, old-age homes, various other types of social work, and certain cultural activities to some extent — activities which in principle could be carried out by public bodies. Thus, rather than acting as pressure groups vis-å-vis local authorities, a relationship-

of interdependence — which Brofoss (1980) characterises as symbiotic — has developed between these organisations and the municipal authorities. Such relationships may be more frequent — in relative terms — in smaller municipalities, where the administrative apparatus is less developed than in cities like Oslo. Thus, voluntary organisations are not just legitimate political actors, they also perform functions as policy instruments for public authorities.

Such interrelationships between local authorities and voluntary organisations are certainly present in Britain, but differences between the two countries may reflect ideological preferences about the role of the state and the separation of public and private spheres of activity. The 'mixed administration' — to use Hernes' (1978) expression — which characterises the Norwegian social democratic state is less significant in the British liberal state. While the relationship between public authorities and voluntary organisations in Norway is characterised by co-operation and interdependence, political activity among voluntary organisations in Birmingham is closer to the traditional style of pressure-group politics, though conceivably less so at the local than at the national level (Berger 1981). For these reasons, what are ordinarily termed organisational resources are less important for local political activity among voluntary associations in Norway than in Britain. The differences are a matter of degree, rather than of kind, but they seem to be of some significance, none the less.

Conclusion

This paper finds quite striking similarities in the voluntary organisations of two very dissimilar cities. The number of organisation per thousand population, the average membership size of each organisation, the average group membership of each citizen, the number of organisations of different types, and the organisational characteristics of the group world in the two cities are alike — sometimes closely alike. While it would clearly be unwise to base firm conclusions on the comparison, particularly when methodological difficulties and differences are taken into account, the parallels between the two studies raise the distinct possibility of uniformities in the nature of the group world, even in places are diverse as Tromsø and Birmingham.

The rate of local political activity on the part of voluntary organisations was different in the two countries, however. In Tromsø 63 percent of the organisations had been active in local politics (activity being broadly defined) in the previous twelve months, compared with 29 percent in Birmingham (where local political activity was also broadly defined). In part this may be due to the fact that Tromsø is a thousand mountainous kilometres from its capital city, and hence rather more dependent upon its own local resources than Birmingham, where the tendency may be for groups to explore contact

with national government in London, rather than local contacts with the city council. In part, it is also likely to be explained by the fact that local government does more in Tromsø, and so generates a higher level of political interest and activity on the part of local groups. Mainly, however, the differences in local political activity seem to be explained in terms of the more open and participatory local political culture and structure in Norway, as compared with England.

This interpretation is, in turn, supported in the last section of the paper which finds that organisational resources are more important as determinants of local political activity in Birmingham. Where rates of political activity are generally rather high, there is little to distinguish groups which are politically active and politically inactive, but where rates are much lower, organisational resources, such as size, paid staff, and degree of establishment in the community, may account for a good deal of the variance between active and inactive groups. This observation, however, leads us back to some general questions about the relationship between voluntary organisations and the local political system, and between corporatist and pressure group styles of local political activity.l3 Unfortunately, however, our data do not allow us to explore these questions, which will, therefore, have to wait for further cross-national comparisons. Meanwhile, our results suggest that such a comparison may produce some fascinating similarities and contrasts.

---

NOTES

1. The literature, as this list of names suggests, is voluminous, but is comprehensively summarised in Constance Smith & Anne Freedman (1972).

2. The work is now so extensive that it has generated its own secondary literature on pressure or interest groups.

3. A comparison of the Birmingham figure with those for six other localities in England suggests that the Birmingham figure is reasonably reliable, and that other localities have a similar number of organisations — see Stephen Hatch (1980).

4. Survey of the Level of Living, Oslo, Central Bureau of Statistics, 1981, and Committee on the Management of Local Government, 'The Local Government Elector', London, HMSO, 1967, p. 114. Almond & Verba (1965, 249) give a figure of 47 percent for the US, but the source quoted above uses a sample double the Civic Culture size, and goes into the question of group membership in greater detail.

5. Selle & Svaasand (1982) call them 'peak organisations'.

6. Moe (1981) bases a whole theory of groups on the distinction between interest and cause groups.

7. The concept of the degree of establishment is similar to that of access, as used by Truman (1967), and that of helpfulness, as used by Dearlove (1971).

8. Robert Dahl (1982, 166) states that 'In a political system as large as a country, a plurality of relatively independent organizations is necessary not only for mutual control but also for the democratic process'.

---

9. According to Sharpe (1973, 8) 'British urban government ... is often closed to the point of absurdity and virtually the whole of the decision-making process is usually fenced off from the public and the press'.

10. The national pattern of pressure groups contacts is therefore repeated at the local-government level in each country. On pressure group contacts at the national level see Samuel Beer (1965), and Christensen & Egeberg (1979).

11. This finding is consistent with what Moe (1980) calls the 'by-product' theory of large pressure groups.

12. As Eckstein (1960) has observed, the development of the welfare state has turned groups into pressure groups. On the relationship between the growth of the welfare state and voluntary welfare organisations in Britain, see F.J. Gladstone (1979).

13. There is some disagreement among those who have written about local interest groups in England as to whether they show 'corporatist' or 'pressure group' styles and modes of operation — see Roger King (1983), who suggests an interpretation which leans towards the pressure group model, and Peter Saunders (1979) who argues for a more corporatist interpretation.

REFERENCES

Aiken, M.T. 1970. 'The Distribution of Community Power: Structural Bases and Social Consequences',
in M.T. Aiken and Paul Mött (eds.), The Structure of Community Power.
New York: Random House.

Almond, G. & Verba, S. 1965. The Civic Culture, Boston: Little Brown.

Banfield, E.G. 1961. Political Influence. New York: The Free Press.
Beer, S. 1965. Modern British Politics. London: Faber and Faber.

Berger, (ed.) 1981. Organising Interests in Western Europe. Cambridge: Cambridge University
Press.

Berry, B.J.L. & Kasarda, J. 1977. Contemporary Urban Ecology. New York: Macmillan.

Bratbak, B. 1979. Organiserte Interesser i Lokal Politikk: Litt om Deltakingsmønster og Kontaktflåter,
mimeo, University of Bergen.

Brofoss, K.E. 1980. 'lnteresseorganisasjonene og budsjetteringsprosessen i Oslo Kommune',
in F. Kjellberg (ed.), Den Kommunale Virksomhel. Oslo: Universitetsforlaget.

Byrne, T. 1981. Local Government in Britain. Harmondsworth, Middlesex: Penguin Books.

Christensen, T. & Egeberg, M. 1979. 'Organized Group-Government Relations in Norway:
On the Structural Selection of Participants, Problems, Solutions and Choice Oppurtunities',
Scandinavian Political Studies, 2, New Series, pp. 239-259.

Clark, T.N. 1971. 'Community Structure and Community Decision-Making, Budget Expenditures,
and Urban Renewal in 51 cities', in CM. Bonjean, et al. (eds.), Community Politics, New York,
Macmillan, pp. 293-313.

Dahl, R.A. 1963. Pluralist Democracy in the United States. New Haven: Yale University Press.

Dahl, R.A., 1982. Dilemmas of Pluralist Democracy: Autonomy Versus Control. New Haven:
Yale University Press.

Danziger, J.N. 1983. 'Group Influence in American County Polities', Local Government
Studies, 9.

Dearlove, J. 1971. 'Councillors and Interest Groups in Kensingston and Chelsea', British Journal
of Political Science, 1, pp. 129-153.

Eckstein, H. 1960. Pressure Group Politics, London: Allen and Unwin.

Eckstein, H. 1966. Division and Cohesion in Democracy: A Study of Norway. Princeton: Princeton
University Press.

Eisinger, P.K. 1972. 'The Pattern of Citizen Contacts with Urban Officials', in H. Hahn (ed.)
People and Politics in Urban Society. Beverly Hills: Sage.

Fischer, C.S. 1975. 'The City and Political Psychology', American Political Science Review, 69,
pp. 559-571.

Fischer, C.S. 1976. The Urban Experience, New York: Harcourt, Brace and Jovanovich.

Galaskiewicz, J.P. 1981. 'Interest Group Politics from a Comparative Perspective', Urban Affairs
Quarterly, 16, pp. 259-280.

Gladstone, F.J. 1979. Voluntary Action in a Changing World. London: Bedford Square Press.

Hatch, S. 1980. Outside the State: Voluntary Organisations in Three English Towns. London:
Croom Helm.

Hausknecht, M. 1962. The Joiners. New York: Bedminister Press.

Hernes, G. 1978. 'Makt, blandingsøkonomi og blandingsadministrasjon', in G. Hernes (ed.).
Eorhandlingsøkonomi og Blandingsadministrasjon. Oslo: Universitetsforlaget.

Hernes, G. & Selvig, A. 1979. 'Local Corporatism in Norway', Scandinavian Political Studies,
2, New Series, pp. 261-275.

Hutchenson, J.D. & Shevin, J. 1976. Citizen Groups in Local Politics: A Bibliographic Review.
Santa Barbara: Clio Books.

King, R. 1983. 'The Political Practice of the Local Capitalist Associations', in R. King (ed.).
Capital and Politics. London: Routledge and Kegan Paul.

Kjellberg, F. & Hansen, T. 1979. Det Kommunale Hamskifte. Oslo: Gyldendal.

Knoke, D. 1981. 'Commitment and Detachment in Voluntary Associations', American Sociological
Review, 46, pp. 141-158.

Kvavik, R.B. 1974. 'Interest Groups in a 'Cooperative' Political System: The Case of Norway',
in M.O. Heisler (ed.), Politics in Europe. New York: Davis McKay.

Kvavik, R.B. 1976. 'Patterns of Interest Group Behaviour in Norwegian polities', Paper Congress
of the International Political Science Association, Edinburgh.

Lewis, D.A. & Clarke, S.E. 1980. 'Money, Maintenance and Mission: The Effects of External
Funding on Poor People's Community Organizations', Center for Urban Affairs, Northwestern
University, mimeo.

Milovsky, C. et al. 1980. 'Survey Approches to Community Organization Research', Yale University.
Center for Applied Social Research.

Moe, T.M. 1980. 'A Calculus of Group Membership', American Journal of Political Science, 4.
Moe, T.M. 1981. Toward a Broader View of Interest Grups', Journal of Politics, 43, pp. 531-543.

Newton, K. 1976. Second City Politics. Oxford: Clarendon Press.
Newton, K. et al. 1980. Balancing the Books. Beverly Hills: Sage.

Nokken, F. 1976. Kommunalpolitikken og det lokale organisasjons samfunn, mimeo, Institute
of Political Science, Oslo.

Polsby, N.W. 1963. Community Power and Political Theory. New Haven: Yale University Press.
Przeworski, A. & Teune, H. 1970. The Logic of Comparative Social Inquiry. New York: Wiley.

Rokkan, S., 1966. 'Norway: Numerical Democracy and Corporate Pluralism', in R.A. Dahl
(ed.), Political Opposition in Western Democracies. New Haven: Yale University Press.

Saunders, P., 1979. Urban Politics: A Sociological Interpretation. London: Hutchinson.

Schmitter, P.C. & Lehmbruch, G. (eds.) 1979. Trends Towards Corporatist Intermediation. Beverly
Hills: Sage.

Selle, P. & Svaasand L. 1982. Interest Articulation among Voluntary Associations. Paper,
European Consortium for Political Research, Aarhus.

Sharpe, L.J. 1973. 'American Democracy Reconsidered: Part I', British Journal of Political
Science, 3.

Smith, C. & Freedman, A. 1972. Voluntary Organisations. Cambridge, Mass.: Harvard
University Press.

Sol vang, B.K. 1977. Organisational Resources and Methods of Participation: A Report from the
Social Sector of a Norwegian Town. Paper, European Consortium of Political Research,
Berlin.

Survey of the Level of Living, 1981, Oslo, Central Bureau of Statistics.

The Local Government. Elector, 1967, London, Committee on the Management of Local
Government, HMSO.

Truman, D. 1967. The Governmental Process. New York: Alfred Knopf.