Over the last two decades, Icelandic party politics has become increasingly unstable. Each of the four parties which have dominated the political scene in Iceland since the 1930s has experienced serious internal difficulties and substantial fluctuations in its electoral support. The established parties have faced increasing challenges from both local and national alternatives outside the four-party format, and on four different occasions in this period new parties have managed an electoral breakthrough of some significance.¹

Seen against this background, the election on 20 April 1991 was a relatively quiet affair.² Compared to the previous one in 1987, the election of 1991 was characterized above all by the recovery of the Independence Party (IP) after its split and worst electoral defeat ever experienced in 1987. The Independent Party's share of votes rose from 27.2 percent in 1987 to 38.6 percent in 1991, while the breakaway group, the Citizen's Party, which had garnered 10.9 percent of the votes in 1987, was reduced to electoral insignificance. As Table 1 reveals, no new party managed to win a seat in the Althingi in 1991, and there were only minor fluctuations in the support of the other established parties. Thus, the Progressive Party (PP) received exactly the same proportion of votes as in 1987 (18.9 percent), whereas the share of the Social Democratic Party (SDP) increased slightly (from 15.2 percent to 15.5 percent) as did that of the People's Alliance (PA) (from 13.4 percent to 14.4 percent). The Women's Alliance (WA), which first entered the Althingi in 1983, lost slightly (from 10.1 percent to 8.3 percent), but none the less established itself as the most serious challenger to the established four-party format since the 19305.

Apart from the five parties which won representation in the election, a number of parties competed unsuccessfully. Two of these can be regarded ad national parties of regional and rural protest, i.e. the National Party and the Home Rule Association, while the Liberals represented an unsuccessful

attempt by the Citizen's Party to broaden its electoral appeal by entering
an alliance with a number of independent politicians.

The Social Bases of the Parties

Until the twentieth century, Iceland was a peripheral region within the Danish kingdom. It obtained sovereignty in 1918, but a union with Denmark was maintained until 1944. A successful independence struggle was based on a strong sense of history among the population and its pronounced cultural homogeneity. Icelanders remain among the most homogeneous peoples of Europe, without any religious, ethnic or linguistic differences of significance.

As the politics of independence came to an end, a party system based primarily on socio-economic class gradually emerged between 1916 and 1930. This system was similar in form to the party systems in other Scandinaviancountries, that a Scandinavian type liberal party (cf. Berglund & Lindström 1978) was missing. The relationship of forces in the Icelandic party system, however, was very different from the Scandinavian pattern. Thus, the party furthest to the right, the IP, has always been the largest of the Icelandic parties electorally, while the Social Democrats have usually been the smallest party in the post-war period. The old farmers' party,

the PP, has also been much stronger than its Swedish and Norwegian counterparts, while the left-socialist PA - with roots in the Communist Party as well as some splinter groups from the SDP - was larger than the SDP in every election from 1942 until 1987.

For a number of reasons, the class bases of the Icelandic parties have been eroded to a far greater extent than has been the case in Scandinavia. The parties are generally weak organizations, with a non-principled approach to policy, but a strong emphasis on patronage and clientelism. Hence, they have also been characterized by a strong emphasis on officeseeking short-term electoral gain, rather than the cultivation of programmatic or particular social groups. As a rule, the coalition game is an open one, where minority governments are very rare, and every party is a potential coalition partner to each of the others.³

The IP is historically the most broadly based of the Icelandic parties (cf. Table 2). In addition to employers and the urban middle classes, which it traditionally has represented, it has also been a strong competitor for the agrarian and working-class vote (cf. Kristjansson 1979). The PP, despite its agrarian origins, likewise made an early bid for the urban vote in the inter-war period, and actually became the second largest party in the towns during the 19605, second only to the IP. The PA has its roots in the workingclassmovement, a major emphasis on nationalism in the post-war period similarly gave it a much broader electoral appeal. Finally, the SDP lost its organizational ties to the labor movement during the Second World War, leaving the party as the organizationally and electorally weakest of

the established parties, with difficulties in marking a place for itself in
Icelandic politics.

An opinion poll conducted less than a month before the 1991 election by the Social Science Research Institute at the University of Iceland reveals that class is at present a weak predictor of party choice in Iceland. The IP came out considerably weaker in the election than in the opinion poll, and the PA and the SDP stronger. None the less, the blurred social profile of the working-class parties is striking. Their joint share of working-class respondents is less than a quarter of the total, and their working-class support is very comparable to the support found among other groups, apart from the SDP's lack of support among farmers and fishermen. By comparison, the profiles of the PP and the IP are slightly clearer, inasmuch as the former enjoys disproportionate support among farmers and the latter among employers. Both, however, also enjoyed more support in all the other occupational categories than either of the old working-class parties.

Gender, by contrast, seems to be an increasingly important variable in accounting for variations in party support in Iceland. According to an opinion poll conducted a week before the election, the WA enjoyed the support of 18 percent of female respondents, but only 2 percent of male respondents. The corresponding figures in the 1987 election study were 22 percent among women and 5 percent among males (cf. Hardarson & Kristinsson 1987). The WA is in particular strongest among relatively welleducated public sector employees and age cohorts between 25 and 45 years. Women in Iceland are clearly more left-wing than before, and the WA has to a marked extent won votes at the expense of the PA. Among the Icelandic parties, however, it is above all the IP which has become underrepresented among women. Thus, the poll indicated that the IP enjoyed the support of 48 percent of male respondents, but merely 33 percent of the female respondents.⁴

The regional factor is also of great significance in the Icelandic political landscape. Constituents, particularly those outside the South-West, generally that their MPs will support the claims of individuals, firms and particular areas for state support. Regional claims of this kind go handin-hand a relatively interventionist state, and various particularistic services take up a sizeable proportion of many a politician's time. An electoral system which gives överrepresentation to the regions outside the South-West has strengthened this tendency.⁵

None of the parties explicitly reject regional politics of this kind, but their involvement varies, as does their regional support. As Table 3 indicates, the PP is the champion of the regions par excellence. The PA also won a greater proportion of the regional vote than it did in the South-West.

The other parties are much stronger in the South-West than in the regions. The WA, for example, is generally less involved in clientelistic practices than the other parties, and the Social Democrats tend to be more sceptical about support for the regions than are other parties. The IP. by comparison, is slightly schizophrenic in this respect. While ideologically it constitutes the party of non-intervention and liberalization in Icelandic politics, many of its regional MPs and supporters expect it to provide services to particular regional interests similar to those which, for example, the PP does for its clients. Since the electoral system creates overrepresentation the regional interests within the parliamentary party. this is by no means an insignificant element in the IP.

Election Issues

No single policy issue dominated the 1991 election. In fact the policy positions of the parties were rather unclear in many respects, leaving local factors, personalities and the government record in the forefront of the campaign. The election term had been eventful with regard to government coalition formations. A coalition government of the IP, the PP and the SDP, formed after the election in 1987, broke up in the autumn of 1988. and the PA replaced the IP in coalition with the other two parties. This government of the left was reinforced a year later through the addition of the Citizen's Party, while the IP and the WA remained in opposition.

During the campaign, the governing coalition parties emphasized the achievements of the government, each of the parties quite naturally stressingthe areas where they had held ministerial responsibilities. The left-wing coalition formed in 1988 had received very unfavorable results in the opinion polls taken throughout 1989. During 1990 its fortunes improved. however, reflecting among other things the conclusion of a moderate wage agreement between employers and the unions. This agreement was

instrumental in bringing down inflation in 1990 to 7 percent, the lowest figure in decades. Throughout the entire campaign, therefore, all of the governing parties made a major point of the success in the right against inflation during their stay in power.

The IP nevertheless entered the campaign with strong support according to public opinion polls conducted by the Social Science Research Institute. The party did not recover immediately after the internal split in 1987. In fact, it was only upon leaving the coalition government in the second half of 1988 that it recovered much of its prior popularity, with popular support growing from less than 30 percent to between 40 percent and 50 percent support. In the same period support for the WA fell markedly, from a high of nearly 30 percent in the middle of 1988, to less than 15 percent a year later. During the reign of the initial IP-PP-SDP coalition, the WA managed to become the leading opposition party, relying on a diffuse protest element in the population in addition to its core of feminist supporters. Once the government of the left took over in 1988, however, the IP became the main opposition party, while the WA experienced difficulties in presenting a clear profile on its own.

The IP entered the election campaign determined to hold its own from the opinion polls. It criticized the tax increases made by the government, but on the whole was cautious in the politics it put forward. A leadership contest had taken place within the party at its conference in March, only six weeks before the election. At this point Reykjavik Mayor D. Oddsson replaced the previous leader, Th. Palsson. Palsson's leadership had been considered weak, and many felt that Oddsson might be able to provide the strong leadership so long lacking in the IP for electoral and governmental purposes. Even so, both Oddsson's leadership qualities and the vagueness of some of the party's policies featured prominently in the criticism directed at the party during the campaign.

Tables 4-6 reveal how voters responded during the campaign when asked how much importance the next government should attach to different issues. In these tables figures indicate the percentage among each party's supporters who thought the next government should attach great importance the issue in question. Concerning taxation, for example (cf. Table 4), IP supporters were clearly the ones most opposed to tax increases of any kind, while PA supporters are the ones least opposed to tax increases. A more egalitarian taxation system is strongly favored by the WA and to a lesser extent the PA supporters. The middle ground concerning taxation is occupied by the SDP and PP supporters.

With regard to equality and social welfare (Table 5), IP supporters are
the ones least in favor of the proposed measures. WA supporters, on the
other hand, are consistently more supportive of equality and social welfare.

The remaining three parties fall in between the two poles, the PA most
similar to the WA and the PP being most similar to the IP.

On economic efficiency and growth (Table 6), IP and SDP supporters appear more enthusiastic than the others. WA supporters, by contrast, seem less concerned with economic efficiency and more with environmental protection than the others. Interestingly enough, however, the supporters of the PP are the ones least preoccupied with economic growth, and they are also less in favor of changing the system of fisheries management, which most likely would lead to rationalization unfavorable to regional balance. The PP, it should be noted, was instrumental in moulding the present system of fisheries management.

Relations with Europe

Perhaps the most peculiar factor in the election of 1991 was nevertheless

the question of membership of the European Community. Iceland is a member of EFTA, and as such has taken part in the negotiations on the European Economic Area (EEA), but the question of EC membership has hitherto not been on the agenda in Icelandic politics. Yet it became one of the most discussed issues in the last weeks before the election. Three of the parties in Iceland - the WA, the PA and the PP - have already decided that they are against membership of the European Community. The IP and the SDP on the other hand, reserve judgement on the issue, but do not preclude membership at some future point. Opinion polls conducted before the campaign started indicated that voters were rather favorably disposed towards an application for membership, although many remained undecided(cf. 1991).

Nationalism in the post-war period has been a major cleavage in Icelandic politics. It has been a strong motivating force in disputes concerning the US military base, NATO membership, extensions of the fishing limits and foreign economic policy. The parties most generally willing to raise the nationalistic banner against economic openness and close cooperation with Western states have been the PP and the PA. At times, these parties have undoubtedly derived electoral gains from their nationalistic emphases. The WA also seems to belong to this current of Icelandic nationalism.

As early as the autumn of 1990 it seemed clear to many commentators that the PP leadership might somehow be interested in making the question of EC membership an election issue. The only problem was that no other party, nor in fact any interest organization, had proposed such a move. The PP none the less made a determined effort during the last weeks before the election to make EC membership an issue, thereby suggesting that the Independence Party and the Social Democrats were not to be trusted in the matter. IP and SDP leaders refused to take this bait, however, and D.

Oddsson stated that when the time came to decide on EC membership, the
issue would be resolved in a referendum.

PP leader and Prime Minister Hermannsson's reply to this idea came three weeks before the election when he suggested that the forthcoming election should be considered a referendum on the issue. The Social Democrats were outraged by this statement, coming from the Prime Minister a government of which they were a part. The IP also reacted strongly, and the quarrel around this issue became one of the dominant themes of what remained of the campaign. The IP and SDP's main emphasis was that they had no intention of joining the EC in the coming term, if every, and that the issue in any case was not on the agenda until a number of other issues had been resolved, including the EEA and the future policies for fishing within the European Community. Even so, the other parties continued with their campaign against EC membership.

The impact of this issue on the election results was probably not very significant. The PP did not improve its position in the opinion polls during the last weeks of the campaign, and in the election it received support comparable to that registered in the opinion polls throughout the winter of 1990-91. The IP lost considerable support during the last month of the campaign,₆ while both the SDP and the PA improved their position even although they were on different sides in the EC dispute. Support for the WA hardly changed at all during the last months of the campaign, although it is the strongest opponent to European integration of all parties, even opposing Iceland's participation in the EEA negotiations. If the issue had an impact in the election, in short, its effect was probably above all to strengthen the PA.

But while the question of EC membership may not have had a great
impact on the election results, the elections may have influenced the issue
of EC membership significantly. Table 7 gives the distribution of attitudes

among respondents to an Icelandic application for membership of the EC
in October 1990 on the one hand and May 1991 on the other.

Table 7 indicates that support for EC membership has decreased dramatically connection with the campaign. Yet the main pattern at both points in time is similar: An application for membership in May enjoyed greater support in the South-West (24 percent) than in the Regions (13 percent); greater support among IP and SDP supporters (31 percent) than in the other parties (12 percent); and greater support among the younger voters (24 percent of those under 40) than among the older ones (16 percent of those over 40).

With several of the parties driving hard against membership and none to defend it, the reduced support for membership application is perhaps not surprising. But it shows that if the impact of the issue on the results was not a great one, this was either because voters attached little importance to the issue or because the IP and the SDP successfully refuted the claim that they were aiming at membership.

A New Government

A new coalition government of the IP and the SDP took over only ten days after the election. This is an unusually short time for coalition formation in Iceland, a process which often takes about two months, and rarely less than a month. The SDP was in a pivotal position after the election, with the power to decide between a government of the left in coalition with the PP, PA and possibly the WA - which gave clearer signs than before of wanting to join a government coalition - or a government with the IP. The SDP felt that some of its key issues, such as the EEA, rationalization within the agricultural and fishing sectors and the development of power-intensive industry in Iceland, would be better protected in a coalition with the IP than in a new leftist government, despite doubts about the new government among some of its supporters. Unlike so often before, the possibility of an IP-PP coalition hardly entered the picture after the election, which in part, at least, may be accounted for by the latter party's growing reservations about openness towards Europe and the liberalization this entails.

Only once before has a two-party coalition of the IP and the SDP been formed in Iceland, but it remained in power for twelve years, between 1959 and 1971. This government was instrumental in bringing Iceland into the group of free-trading nations by liberalizing foreign trade in 1960, by acceding to GATT in 1967, and by pursuing membership of EFTA in 1970. It also took the first steps in developing power-intensive industry in Iceland.

To an unusually large extent the present government has an electoral
basis among the urban population in the South-West of Iceland. The voters

of the governing parties tend to be the ones most interested in liberalization, rationalization, openness towards Europe and the development of powerintensiveindustry. government is likely to tread carefully in these matters, however, not only because of the opposition it may encounter from the other parties, but also because of reservations within the governing parties themselves, especially the IP. None the less, it seems quite likely that the election of 1991 and the subsequent coalition formation may have set the tone for political debates in Iceland during the 19905.

---

NOTES

1. These were the Union of Liberals and Leftists (8.9 percent of votes cast in 1971), the Social Democratic Alliance (7.3 percent of votes cast in 1983), the Citizen's Party (10.9 percent of votes cast in 1987), and the WA, formed in 1983. Of these four, only the WA still exists today.

2. Unless it is dissolved due to extraordinary circumstances, elections to the Althingi, the Icelandic parliament, occur every four years.

3. It should be noted, however, that the issue of US military bases in Iceland more or less excluded cooperation between the IP and the PA (or its predecessor) between 1947 and 1978, since the PA in that period made the removal of the base a condition for its participation in any government.

4. This opinion poll was also conducted by the Social Science Research Institute.

5. While 64 percent of the 1991 electorate resides in the South-West, its share of MPs is a mere 46 percent. Disporportionalities in the representation of parties, however, are for the most part corrected by a somewhat complicated electoral system. Under this system 50 seats are allocated according to the Largest Remainder formula for proportional representation on the basis of constituency results. The remaining 13 seats are accorded to the parties on the basis of nationwide results, using d'Hondt's formula, so as to reduce disproportionalities in the representation of parties while maintaining överrepresentation of the peripheral regions.

6. The share of the IP fell from 48 percent in the March poll of the Social Science Research Institute to 40 percent in a poll conducted only a week before the election in April.

REFERENCES

Berglund, S. & Lindström, U. 1978. The Scandinavian Party System(s). Lund: Studentlitteratur.

Hardarson, O. Th. & Kristinsson, G. H. 1987. 'The Icelandic Parliamentary Election of 1987".
Electoral Studies 6, 219-234.

Kristinsson, G. H. 1991. 'Iceland' in Wallace. H , ed. The Wider Western Europe. London:
Pinter Publishers.

Kristjansson, S. 1979. 'The Electrical Basis of the Icelandic Independence Party, 1929-44".
Scandinavian Political Studies 2 (NS). 31-52.

Olafsson, S. 1991. 'Verkefni næstu rikisstjörnar'. Reykjavik: The Social Science Research
Institute.