ISF Paper 2004:4
The Norwegian Progress Party: Building Bridges across Old Cleavages
Tor Bjørklund
Department of Political Science, University of Oslo
P.O. Box 1097 Blindern, 0317 Oslo
e-mail: tor.bjorklund@stv.uio.no
Jo Saglie
Institute for Social Research
P.O. Box 3233 Elisenberg, 0208 Oslo
e-mail: jo.saglie@samfunnsforskning.no
Paper presented at the 12th Nasjonal fagkonferanse i statsvitenskap,
Tromsø, 7–9 January 2004.
The Norwegian Progress Party celebrated its 30th anniversary with its best local
elections ever. 13 municipalities elected Progress Party mayors. The purpose of this
paper is to explore this breakthrough by analyses of ecological data. We focus on
three questions. Firstly, the conflict between centre and periphery. The party has
been firmly rooted in the urban cultures, but many of its new mayors represent the
heartland of the Norwegian counter-cultures. Secondly, we find that support for the
Progress Party is independent of the level of unemployment in the municipalities.
The party does well, however, in municipalities with many immigrants, but this
connection is weaker than before. Thirdly, we find that incumbency is an asset for
the Progress Party – in contrast to most other parties. In conclusion, the Progress
Party appears to build bridges across the old cultural-religious and class cleavages.
The party is not a mediator that gives concessions to both sides, but its support
illustrates that that old cleavages are less salient.
1
The Norwegian Progress Party: Building Bridges across Old Cleavages1
A breakthrough at the local level
The Norwegian Progress Party celebrated its 30th anniversary with better local election
results than ever before. In fact it was a breakthrough for the Progress Party as a political
player in Norwegian municipalities. In 30 municipalities (of 434) the party gained more votes
than any other party; in 13 municipalities the Progress Party succeeded in seizing the position
as mayor of the municipality. The party leadership complained that the election result should
have resulted in more mayors; the past as a pariah among the political parties was not
completely buried. Nevertheless, the 2003 local elections turned out to be a watershed. The
Progress Party has participated in local elections since 1975, but until 2003 the party has only
gained the mayoral position twice. A long history of insignificance in local politics has ended.
The purpose of this paper is to explore this breakthrough, mainly by analyses of
ecological data. We focus on three questions. Firstly, the conflict between centre and
periphery. The Norwegian Progress Party has traditionally been firmly rooted in the country’s
geographical centre. Has the 2003 elections changed this pattern? Secondly, to what extent
does support for the Progress Party depend on unemployment and immigration in the
municipalities? Thirdly, it is well known that parties in government often lose votes. Has the
Progress Party’s status as an opposition party in most municipalities helped the party, as the
other parties got the blame for all the shortcomings in municipal services?
Perspectives on support for right-wing populist parties
Centre and periphery
According to some scholars the most fruitful soil for radical right parties is the periphery.
Hans-Georg Betz (1998) argues that these parties are a reaction against societal changes; they
appear in the transition between the industrial and the post-industrial society. The party’s
appeal is strong among the losers of modernisation, and there is an expected concentration of
1
An earlier version of the paper was presented at the EREPS (Extreme Right Electorates and Party Success)
annual meeting, University of Ghent, Ghent, Belgium 5. – 7. December 2003.
2
losers in the periphery. Time is lagging behind in those region; the societal problems multiply
with unemployment and a decreasing number of inhabitants. In short, the periphery is a
problem area.
At a more general level, a classical theory interprets right-wing extremism as a
reaction of pre-industrial strata against the modernising tendencies of industrial society, in
particular the marginalisation of the old middle class.2 The fear of being socially degraded is a
driving force for protest in the middle class.
Some political scientists have interpreted the Scandinavian Progress Parties in these
terms. In the updated edition of Seymour Martin Lipset’s Political Man (1981) Glistrup’s
Progress Party is discussed under the heading ‘Fascism and the Revolt against Modernity’.
The party is described as ‘a populist protest movement’ and according to Lipset the Danish
Progress Party was at that time over-represented in traditional parts of society, ‘in rural areas
and provincial, by the less educated, the self-employed, and by farmers.’3 Related to this
phenomenon was a reactionary populism with nostalgia for the past, a longing back to a preindustrial society before the harmonious unity was destroyed by big capital and the working
class.
This explanation, however, has obviously been wrong in the Norwegian case – and
also in the Danish case. The Norwegian Progress Party has always received most support
from young, urban men. From the party formation in 1973 to the end of the 20th century, the
party had its stronghold in populous areas – in the cities and their surrounding areas. The
explanation may be both economical and cultural. The economic explanation points to the fact
that the Norwegian periphery is no powerless area with few economical resources. The
periphery has been supported by state intervention and economic transfers. The long-standing
weak position for the Progress Party in the rural areas had been explained by the fact that the
party’s neo-liberal viewpoint is at odds with the political demands of the periphery. Free
market forces are regarded as a threat to the sparsely populated areas.
Another explanation points to the cultural cleavages between the centre and the parts
of the Norwegian periphery which have been characterised by the so-called ‘countercultures’.4 Firstly, a rural language movement confronted the urban bourgeoisie and the
Some of the data were made available by Statistics Norway and the Norwegian Social Science Data Services
(NSD), which assume no responsibility for our interpretation of the data. All election results from the 2003
elections are preliminary. The results may be slightly adjusted after the final count.
2
Hamilton (1982).
3
Lipset (1989: 499).
4
Rokkan (1967).
3
heritage from the Danish language. The language the movement advocated, ‘nynorsk’, was
based on dialects, especially from the southern and western region. Secondly, a religious
layman movement – mostly within the state church – defended Lutheran orthodoxy and
pietistic fundamentalism. The layman movement was often in conflict with the leadership of
the church. Thirdly, the temperance and prohibition movement was the most vivid people’s
movement in the beginning of the twentieth century. The temperance movement and the
religious layman movement were intertwined and had their strongholds in the southern and
western periphery. The peak of their power was perhaps in the 1919 prohibition referendum
when over 60 per cent of the voters supported banning alcoholic beverages containing more
than 12 per cent alcohol by volume.
The counter-cultures offered a barrier against class politics, and consequently also a
barrier against electoral success for Labour and the Conservatives. In the strongholds of the
counter-cultures, the three centre parties, i.e. the first political party in Norway, the Liberals,
and its offsprings – the Christian People’s Party and the Farmers party (later Centre Party) –
received massive support. The counter-cultures have also been regarded as a barrier against
the Progress Party. The counter-cultures and the Progress Party, however, have one thing in
common: they are all anti-establishment movements. However, the Progress Party clearly
belongs to the urban side of the cultural cleavages. The Progress Party argues for cancelling
the obligatory test in both forms of Norwegian in the primary and the secondary school; only
the pupil’s main form of language should be the subject of written tests. As ‘nynorsk’ is the
second form of language for most Norwegian pupils, this would be a blow to the defenders of
nynorsk. Undoubtedly, the Progress Party has from its first day strongly argued against high
alcohol taxes and various forms of regulations in drinking and selling alcohol. The party has
never been especially concerned with religious matters.
The counter-cultures have lost some of their political importance during the last
decades. Moreover, the Progress Party has been transformed since the party split in 1994,
when the libertarian faction left the party. Ideological neo-liberalism has been de-emphasised,
and the party has adopted a ‘catch-all’ approach. Accordingly, the party’s regional profile has
become less pronounced during the 1990s. Progress Party votes are now spread more evenly
across the country. Therefore, a question can be raised if the Progress Party after 30 years is
adapting to the voting pattern expected from the international literature on right-wing
populism. Does the newly acquired position in parts of the periphery fit with what some
observers regard as the ‘natural’ stronghold of the party?
4
Unemployment and non-western immigrants
The ‘modernisation loser thesis’ may be examined from another angle. The unemployed are
the clearest examples of losers in the transition to a post-industrial society. The question here
is whether unemployment affects the electoral strength of the Progress Party. We are
interested in the macro level – the rate of unemployment in the municipality. Unemployment
may easily have consequences also for the employed and groups outside the working force.
Fear of unemployment can be spread. Consequently, the rate of unemployment in a
community may increase support for radical right parties such as the Progress Party.
According to literature about radical right parties, their support is dependent on
unemployment and immigration.5 The combination of both high unemployment and many
non-western immigrants is expected to give good conditions for electoral breakthroughs.
Some have regarded immigration and unemployment as intimately connected. One million
immigrants in the labour force suggest that the immigrants have taken the jobs from one
million unemployed natives. Jean Marie Le Pen, the chairman of National Front, has made
that argument. But on the other hand he makes the case that unemployed immigrants represent
an economic burden for society. In this way, immigrants appear as scapegoats for both the
employed and the unemployed.
Empirical research has nevertheless produced mixed results. Knigge (1998) has
studied the relationship between the rate of unemployment and the support for the radical
right parties in Europe. According to her findings, in years when unemployment decrease the
support for radical right parties seems to increase. Other analyses have concluded differently.
Jackman and Volpert (1996) have analysed the success of right radical parties in 16 West
European countries. The strength of the radical right alternative is related to unemployment,
the threshold for representation defined by the electoral system and the number of political
parties participating in the election.6 Their conclusion is the opposite of Knigge’s: the level of
the unemployment correlates positively with electoral results for the radical right. Givens
(2002) has studied the regional results for radical right parties in the various counties of
Austria, France and Germany. The electoral support is analysed according to regional
variations in unemployment, the share of immigrants and the turnout. Two periods are
separated, before and after the breakthrough for the radical right alternative. The crucial year
is 1986. After 1986 unemployment gives a positive significant outcome, i.e. high
unemployment coincides with electoral success for the Austrian and French parties (but not
5
See for example Betz (1994).
5
the German ones). In the period before 1986 the conclusion is, however, the opposite: a
negative correlation between unemployment and election results in Austria and Germany (but
not in France). Finally, Golder (2003a) emphasizes that it is important to distinguish between
different kinds of extreme right parties. He found that immigration had a positive effect on
populist parties irrespective of the unemployment level. However, unemployment only
mattered when there were many immigrants in a country. In addition, Golder pointed to the
importance of electoral institutions.
On the basis of this research, it is difficult to predict how unemployment and
immigration in Norwegian municipalities – where unemployment is comparatively low and
the share of non-western immigrants is low, but rising.
Lack of local office as an electoral asset
A third perspective on the breakthrough of the Progress Party in Norwegian local politics
points to the fact that the party has usually been kept out of local office. The party’s ‘pariah
status’ may, however, have been an electoral asset. In general, incumbency is an electoral
liability in Europe.7 Governing parties gets the blame for all kinds of problems, and tend to
lose votes. Most of the research on the electoral effects of incumbency has studied national
elections, but a similar effect might be found at the local level.
Norwegian local government has usually been characterised by a high degree of
consensus and accommodation. Most municipalities are governed according to the ‘alderman
model’, where the executive committee is elected by proportional representation by and from
the local council. If consensus is dominant, the voters can hardly assign blame and punish the
local office-holders. Nevertheless, Pål Martinussen (2002) has shown that many Norwegian
municipalities experience partisan cooperation based on political agreements. It is thus
possible to identify a local ‘cabinet’. Moreover, Martinussen (forthcoming) found that officeholding parties lost votes also in Norwegian local elections. He did not carry out any partyspecific analyses. However, there may be reason to expect incumbency to be a greater liability
for the Progress Party than for other parties. As an anti-establishment party it may be
punished harder than other parties, if it takes on the responsibilities of local government
office. Thus, our last research question is whether the Progress Party has been punished by the
voters where they have been in office, and rewarded in municipalities where they have been
excluded from positions of power.
6
Jackman and Volpert’s results have been criticised by Golder (2003b).
6
Building bridges across cultural and religious cleavages
Concomitant with the surge in local elections, the geographical profile of the Progress Party
has changed. Support for the party has spread from its old strongholds in the centre towards
the periphery. In the 2003 local elections the Progress Party achieved its best results in
middle-sized municipalities (population 10, – 59,999 inhabitants) (Table 1). However, the
steepest increase is found in the small municipalities which are located in the sparsely
populated areas (Figure 1, Table 1). In Oslo, on the other hand, support for the Progress Party
has declined during the last decade. How can the new Progress Party support in the periphery
be explained? By raising the question, we raise a new one: in which part of the periphery does
the party gain most votes?
Figure 1 and Table 1 about here
In table 2 we have listed the 13 municipalities that elected Progress Party mayors in
2003. We use the results from four Norwegian referendums – the two prohibition referendums
(1919 and 1926) and the two European Union referendums (1972 and 1994) to measure the
support for ‘periphery positions’ in the municipalities. Our ‘centre-periphery index’ is the
mean of the four referendums, more precisely the per vent voting for prohibition plus the per
cent voting against Norwegian membership in the European Union. The index varies between
22 and 89 per cent, the higher the value the more typical is the peripheral position. The
periphery index and the per cent of the pupils learning ‘nynorsk’ tell us something about the
cultural position of the municipality. In addition, the strength of the counter-cultures is usually
accompanied by a strong showing for the three centre parties.
Table 2 about here
The Progress Party has now grown roots in two quite different part of the country. On
the one hand, in the heartland of the counter-cultures. Eight Progress Party mayors were
elected in the ‘Bible Belt’ at the western coast. Four of these municipalities are situated in the
7
See e.g. Rose and Mackie (1983), Narud and Valen (forthcoming).
7
Sunnmøre region in Møre and Romsdal County – an area with 100 per cent ‘nynorsk’ and
around 80 per cent as the mean value in the four referendums. The western periphery has also
been a stronghold for the three centre parties. During the last 50 years, however, they have
declined substantially in this region while the Progress Party has made considerable inroads.
On the other hand, the Progress Party has gained mayoral positions in the county
Vestfold by the Oslo Fjord – a region without ‘nynorsk’ in schools and with a tradition for
having anti-prohibition attitudes, as well as clearly being supportive of Norwegian
membership in the European Union. The same characteristic applies to the municipality
Ullensaker in the south-eastern county Akershus. The last of the Progress Party mayors does
not come from any of these two regions; Nordreisa is a part of the northern periphery.
The declining political importance of the counter-cultures
The rise of the Progress Party in the heartland of the counter-cultures is partly a result of the
decline of the counter-cultures. More specifically, it is a consequence of the diminishing role
of the cultural-religious dimension in politics. Alcohol, religion and language were earlier
political issues of first order. In the inter-war period (1918–1940) several governments had to
leave power as a result of dispute around such questions. The importance of the culturalreligious issues has declined as the emergence of the industrial society brought class politics
into focus. The Norwegian historian Jens Arup Seip (1980) wrote that teetotalism and the
language question during the 1950s were in a process of fading away as distinguishing marks
between the political parties. From 1950s up to recent time this process has continued.
Electoral research has also described this decline. According to Berglund (2002: 202),
language has not influenced party identification since the 1970s. Alcohol and religion have
not lost their impact, but these issues are mainly important for the Christian People’s Party.
Language is no longer a burning political issue. One camp in the dispute, those who
argued for an amalgamation of ‘nynorsk’ and ‘bokmål’, has nearly disappeared. The fight for
‘samnorsk’, a combined language, is a lost cause. In the 1950s important politicians in Labour
supported this alternative. Today two official forms of language are broadly accepted, but
those who defend ‘nynorsk’ are a declining group. In primary school, around 15 per cent of
the pupils had ‘nynorsk’ as their main language in 2000. The share has steadily declined since
the peak in 1944 with 34 per cent.8
8
Grepstad (1998).
8
The Progress Party is definitely not a defender of ‘nynorsk’, but as the language
question is not a salient issue the Progress Party does not appear as an enemy of ‘nynorsk’.
For many the language question is a non-issue. Today the Progress Party has a mayor in
Ørsta, the municipality where the inventor of ‘nynorsk’, Ivar Aasen, was born. The dialects in
the region around Ørsta do not deviate much from ‘nynorsk’ and in this region ‘nynorsk’ is
the official language in school and administration. The newly elected mayor from the
Progress Party stated in newspaper interview that he himself defended ‘nynorsk’; this
language was an important part of the region’s identity.9
Obviously, the Progress Party support for liberalisation in alcohol politics may appear
as a dilemma for defenders of the counter-cultures. However, the party’s policy corresponds
to a general trend of liberalisation. An increasing number of municipalities have granted
licences for the sale and serving of alcohol. While 25 per cent of the population lived in ‘dry’
municipalities in 1962, this applied to only 1 per cent in 1991.10 The last of the ‘dry’
municipalities – Audnedal – permitted sale of beer in 2002. Support for liberal policies has
increased,11 and the share of the population who does not drink alcohol has substantially
declined (1969: 23 %, 2001: 11 %).12
Ørsta also has a strong tradition for the prohibition movement. In the 1919 referendum
a part of Ørsta (Vartdal) voted 100 per cent for prohibition with a turnout as high as 98 per
cent. Still, the teetotalism cause has many supporters in Ørsta and beer is not freely sold in the
municipality. Beer is only sold in special outlets; the ‘beer monopoly’. Indeed, a beer
monopoly is in contradiction with the Progress Party’s policy. In a newspaper interview the
mayor admitted that the beer monopoly ought to be abolished, but it was not his first priority.
The beer monopoly had not at all been an issue during the election campaign in Ørsta.13
For the religious layman movement the Progress Party seems to be an acceptable
partner. On the plus side it can be noted that the Progress Party more clearly than any other
parties have defended Israel in the Middle East crisis.
To sum up, the Progress Party has seized mayor positions in municipalities with the
highest and the lowest score on the periphery index – in the heartland of the counter-cultures
9
Aftenposten 26. October 2003.
Saglie (1996: 318).
11
Saglie (1996).
12
Source: The Norwegian Electoral Programme.
13
Aftenposten 26. October 2003.
10
9
and in a region where the counter-cultures have no historical roots.14 Fifty years ago the
mayors from Sunnmøre were always recruited from the centre parties and the mayors from
the county Vestfold often from the Conservatives. Today mayors from the Progress Party
unite the two regions. The party appears to build bridges across an old cultural-religious
cleavage. The Progress party is not a mediator who gives concessions to both sides, but its
new geographical profile illustrates that the old cleavages are outdated or at least not salient
anymore. In addition, these results indicate that the party and its electoral appeal have been
transformed.
Income and social indicators
Does this mean that the Progress Party gradually adapt to the expectations of the literature – a
party for the modernisation losers? The mean income in the municipality can tell us
something about its economic situation. If the Progress Party recruits modernisation losers
one can expect a low income level in the districts where the party succeeded. The average
income levels in the 13 municipalities with a mayor from the Progress Party are reported in
Table 2. However, they do not deviate from the grand mean with a low figure. Contrary to the
expectation the figure is more often above than under the grand mean.
Table 3 corroborates this result: there is a strong correlation between the mean income
and Progress Party vote in the municipalities. In addition, the table shows that the increase in
Progress Party vote from 1999 to 2003 was especially strong in municipalities with a high
average income. The party’s breakthrough in the periphery appears to be limited to the
affluent parts of the periphery.
Table 3 about here
In addition to income we use a measure of the general standard of living in the
municipality. Statistics Norway has constructed a ‘living standard index’ based on various
14
In order to illustrate let us single out two small municipalities (with just under 5.000 inhabitants), one – Tjøme
– situated outside the counter-cultural area, another – Stranda – in the heartland of the counter-cultures. In the
1919 prohibition referendum 20 per cent voted for prohibition in Tjøme compared to 95 per cent in Stranda. In
the 1972 referendum about Norwegian membership in the European Union 38 per cent supported membership
among the voters in Tjøme; the corresponding figure in Stranda was 71 per cent. Tjøme has been a Conservative
stronghold. In the parliamentary elections during the 1940s, 1950s and 1960s the Conservatives in Tjøme
sometimes got a little more than half of the votes, and usually around 45 per cent. In the same period the results
in Stranda were always below 10 per cent. A similar contrast can be seen in the support for the centre parties.
During the first 25 years after the World War II the centre parties usually received less than 15 per cent of the
votes in Tjøme compared to between 58 and 74 per cent in Stranda.
10
indicators from the municipalities: the share living on social security, the share of disabled
persons, crime and unemployment rates, mortality, the educational attainment and income.
Some parts of the northern periphery and some urban neighbourhoods have the lowest score
on living standard. In the 13 municipalities with Progress Party mayor (Table 2) the overall
picture is a living standard more or less as a mirror of the whole country, or a little better.
Two municipalities (Austevoll and Stranda) have a remarkably high position on the living
standard index. Table 3 shows that there is no correlation between Progress Party vote and the
index. The social indicators of the municipalities give no clear-cut picture, but the
communities with Progress Party mayors are definitely not hit by serious social problems. In
short, the Progress Party vote in 2003 does neither fit the traditional Norwegian pattern nor
the expectations from the literature.
Unemployment and non-western immigrants
The surge of support for European radical right parties in the second half of the 1980s was
partly caused by the question of immigration. The issue was put on the political agenda as a
consequence of an influx of refugees and asylum seekers. There was one movement from
south to north, from the third world, and another from east to west followed by the collapse of
communism.
The immigrant issue emerged more than ten years after the launching of the Progress
Party. In the first years of the party’s history the immigrant question was a non-issue. Except
for the Sami people the Norwegian society had a homogeneous population. Figure 2 describe
the share of non-western immigrants in Norway from 1973 to 2003.15 From a rather
insignificant number of immigrants the share has constantly increased and passed 5.0 per cent
(5.2), or approximately 220.000 persons in 2003. In 1973 the share was 0.3 per cent and
somewhat just 50.000 persons. Consequently, in the course of the lifetime of the Progress
Party the number has been more than quadrupled. In the course of this period, the immigrant
question has evolved from a non-issue to a more or less salient issue in recent election
campaigns.
15
Immigrants are defined as persons with parents born abroad. First generation immigrants are those who
themselves immigrated; second generation immigrants are children from the first generation, i.e. they are born in
Norway. Non-western immigrants are originally from Asia (except Israel), Africa, Latin-America, ExYugoslavia, Turkey and republics from the earlier Soviet Union.
11
Figure 2 about here
In contrast to the number of immigrants the support for the Progress Party has varied
with ups and downs over the years. Its second breakthrough after the launching of the party in
1973 was in the 1987 local elections. The party then passed the ten per cent threshold (13.4
per cent), and for the first time the immigration question was on the agenda in an election
campaign.16 The election thereby introduced a new phase in the party’s history: that of an
anti-immigrant party. The reason was simple. During the 1987 election campaign the number
of asylum seekers and refugees exploded. It multiplied from a low level but the growth in
ratio was higher than in other Western European countries.17 This fact actualises the question
about the presence of immigrants as an impetus for voting the Progress Party. Is there a
connection between the number of immigrants and support for the party? That is, not in the
sense that immigrants vote for the Progress Party, but rather that a high number of immigrants
are a reminder of problems connected with the immigration, and then lead to support for an
anti-immigration party.
In Figure 2 we follow the support for the Progress Party during the last 30 years
combined with the level of unemployment in the same period. The ups and downs of the
Progress Party are not systematically connected to the rate of unemployment. Compared to
other European countries the unemployment rate in Norway has been rather modest in most of
these years. But it has been regarded as a serious problem especially after it started to increase
from 1987 and when it peaked in 1993. Thus, the successful 1987 election took place before
the increase in unemployment. As previously mentioned the immigration issue seems to be a
plausible candidate for explanation.
In addition to information about the rate of unemployment on the nation-wide level we
also have data on the rate of unemployment in the various municipalities. In the 1980s the
municipalities in the periphery were hit harder by unemployment than the populous areas and
cities. In the 1990s this pattern disappeared and unemployment was equally spread along the
urban-rural axis.
We have conducted a regression analysis with the share of non-western immigrants
and unemployment rate in the various municipalities as independent variables and support for
the Progress Party in the elections of 1987, 1995, 1997 and 2003 as the dependent variable
16
Bjørklund (1988).
12
(Table 4).18 In 2003 the share of non-western immigrants still gives a statistically significant
coefficient but the figure has declined somewhat. The reason is the party’s movement from
centre to periphery; Oslo – with a large immigrant population – is no longer a stronghold for
the Progress Party.
Table 4 about here
The unemployment rate has a statistically significant coefficient only in 1987, and then
with a negative correlation, i.e. high unemployment goes together with low support for the
Progress Party and vice versa. A hypothesis can be formulated which fits to this pattern: low
unemployment contributes to the success of the radical right parties. The absence of a
persistent unemployment gives room for experimenting and a chance to new party
alternatives. Prosperous times open possibilities to test the previously untested. However, in
1995, 1997 and 2003 there is no significant correlation between unemployment and support
for the Progress Party. The connection between unemployment and the support for the
Progress Party has moved to a zero correlation.19 According to Table 3, the Progress Party
increased its support slightly more in municipalities with low unemployment. On the other
hand, on the micro level it is documented that the Progress Party recruits somewhat better
among the unemployed than among the employed.20
The explained variance for support for the Progress party, based on share of nonwestern immigrants and unemployment rate in the various municipalities, has declined
substantially from 1987 to 2003 (from 28 to 7 per cent). Still, the number of non-western
immigrants has a significant effect on support for the Progress Party. In addition, the steadily
increasing number of non-western immigrants (Figure 2) has contributed to putting the
immigration question on the political agenda.
17
Betz (1994: 77).
The analysis is based on the county council elections in 1987, 1995, 2003 and the 1997 parliamentary
elections.
19
We also included interaction terms, but these analyses produced an irregular pattern. There was a positive
interaction effect in 1987, no significant interaction effects in 1995 and 1997, and a negative interaction effect in
2003. In other words, the combination of high unemployment and many immigrants does not seem to generate
support for the Progress Party.
20
Bjørklund and Goul Andersen (2002).
18
13
Since the second half of the 1980s the Progress Party has been able to claim ‘issue
ownership’ to the opposition to the official immigration policy.21 The position as a critical
voice towards the official immigration policy has been strengthened during the last few years.
Evidence in this regard can be provided from election studies of 1997, 2001 and 2003. Voters
were asked to name the ‘best party’ – or the party they had most confidence in – with regard
to immigration policy. In 1997 and 2001 one third were most trustful towards the Progress
Party, in 2003 the share had increased to 39 per cent (see Table 5). Among those who had
confidence in Progress Party’s immigration policy about one third voted for the Progress
Party in 1993, while about half did so in 2003.
Table 5 about here
In spite of this development the number of non-western immigrants in the 13
municipalities that elected Progress Party mayors is rather low and definitely below the
national average. On the other hand, the party has lost ground in Oslo – where more than 40
per cent of the immigrants live. As shown in Table 3, the increase in Progress Party vote from
1999 to 2003 is not at all related to the share of non-western immigrants.
It may be argued that analyses which include all Norwegian municipalities present a
misleading picture. Most municipalities have less than three per cent immigrants, and during
the last years Oslo’s special status has become even clearer. As much as 16.4 per cent of the
inhabitants in this city were non-western immigrants in 2003. The impact of immigration on
voting behaviour is presumably larger in Oslo than elsewhere in Norway. Therefore, we take
a closer look at districts within the city of Oslo.
Neighbourhoods in Oslo
Oslo is an ethnically segregated city, where the number of immigrants varies sharply from one
district to another. At first, immigrants settled in old working class areas downtown.
Gradually they moved to the satellite towns. As the prices downtown increased the apartments
in the satellite towns seemed to appear more favourable. A division between those parts of the
city with a high share of immigrants and those with low share is in effect a division between
east and west, the old border between working-class areas in the east and middle-class areas
21
Both studies of party rhetoric (e.g. Hagelund 2003) and voter preferences (e.g. Narud and Valen 2001,
Bjørklund and Goul Andersen 2002, Aardal 2003) confirm the significance of the immigration issue for the
Progress Party.
14
in the west. The division was formed by the industrial society. Accompanied by a social
segregation was a political segregation: the socialist parties had their stronghold in east and
the bourgeois parties in west.22 According to the 2003 population statistics around one third of
the inhabitants in some parts of eastern Oslo are non-western immigrants. On the other
extreme, there are only three or four per cent immigrants in some neighbourhoods in western
Oslo.
Table 6 about here
The Progress Party, however, cuts across the old division between east and west (see
Table 6). The party’s strongest areas are the satellite towns in the east. Here we find the ‘new’
working class, the leftist parties are rather strong and the immigrants are many. The Progress
Party’s weakest results are found in the old working class areas downtown where the leftist
parties still have their strongest foothold. In Old Oslo, for example, the support is clearly
under the mean in spite of the fact that the number of immigrants are among the highest. In
these districts the Socialist Left has become the largest party, larger than Labour, the party
which had a dominant position in these quarters in the last century up to the 1960s. One
explanation may be that a part of the new radical middle class has moved into the old working
class area, and they have no affinity to the Progress Party.
In a middle position we find the up-scale residential areas in the west. The Progress
Party got a fairly good result in these upper middle class districts with few immigrants and
weak support for the leftist or socialist parties. Consequently, the Progress Partly represents a
break of the old division between east and west; a pattern formed by the industrial society.
Again, the Progress Party has built a bridge across an old cleavage, this time the class conflict
from the industrial society. Oslo is still a politically divided city – with Conservative and
Labour strongholds – but the Progress Party has a foot in both camps.
Incumbency: asset or liability?
As mentioned above, incumbent parties tend to lose votes in Norwegian local elections. We
assumed that incumbency might be an even greater liability for the Progress Party than for
22
See e.g. Øidne (1970).
15
other parties. However, the municipality of Os – the only municipality that elected a Progress
Party mayor in 1999 – seems to disprove this hypothesis. The Progress Party vote in Os
increased from 36.6 in 1999 to 45.7 % in 2003. Government office was clearly not a liability
for Progress Party mayor Terje Søviknes.
The mayor of Os may, however, be an exception. Søviknes is a political celebrity. In
1999 he was elected deputy leader of the national party organisation, and he was on his way
to a national political career. He resigned from all national offices in 2001 because of a sex
scandal, but kept his position as mayor of Os. As the only municipality governed by the
Progress Party, Os received much attention from the national media as a showcase for
implementation of Progress Party policies. The Progress Party’s success in Os might therefore
not be a representative case.
Os was the only municipality with a Progress Party mayor prior to the election, but 20
municipalities had deputy mayors from the party. This gives us a broader basis for assessing
the impact of incumbency on the Progress Party vote. Table 7 contains the average gains or
losses of votes in 2003, by party and incumbency. Municipalities where a party did not field
candidates in 2003 are excluded from the calculations.23 We have operationalised
‘incumbency’ in two ways: as the party of the mayor and as the party of the mayor or deputy
mayor. Mayors and deputy mayors are elected by the local council (or in some cases directly
by the voters) for a four-year term. Accordingly, our data on incumbency are based on the
elections of mayors that took place in 1999.24 Two cities – Oslo and Bergen – practise a
parliamentary system. Here the positions of mayor and deputy mayor carry less political
weight. Consequently, we used the more politically relevant positions as leader and members
of the City Government in these two municipalities.
Table 7 about here
As expected, Table 7 shows that office is a liability for most parties. The results for the
Socialist Left Party and the Centre Party were nevertheless marginally better when the parties
were in office. And contrary to expectations, local government office was clearly an asset for
23
Joint lists between two or more parties are not counted as party lists, and mayors who represent joint lists are
not counted as party representatives.
24
We collected newer data for three municipalities: the parliamentary systems of Oslo and Bergen (where the
composition of the City Government may change during the electoral period) and the municipality of Re, which
was created in 2001 by a merger of two municipalities.
16
the Progress Party. The party gained ground across the country, but more so in municipalities
where the party had the mayor or deputy mayor.25
Why was local office an asset for the Progress Party? One explanation may be that
these kinds of positions may improve the party’s image. In opposition, the party may be
labelled ‘irresponsible’ and ‘extremist’. In office, the party appear as a more responsible and
acceptable alternative. This implies that a vote for the Progress Party vote is more than a
political protest – it expresses support for policies or candidates. Another implication is that
local Progress Party politicians in visible positions can improve the party vote. In other words,
the party may be less dependent on the party chairman and his appearances in the national
media than some commentators assume. A final implication is that if the 2003 results can be
generalised, the election of 13 Progress Party mayors in 2003 may produce an even better
result for the party in the 2007 elections.
Bridges across two old cleavages
To sum up, the Progress Party has built bridges across two old cleavages: the culturalreligious from the pre-industrial society and the class conflict from the industrial society. The
party is not a bridge-builder in the sense that it is driven by a conscious wish to erase political
cleavages. Rather, the party’s presence reflects that something new has happened. Old
cleavages that used to shape voting patterns have been diluted. But what are the new
cleavages?
Unemployment seems not to be a dividing line. In the 13 municipalities with Progress
Party mayors the unemployment rate varies both below and above the natural average. During
the last years the unemployment rate has been rather low in Norway and it does not vary in
any systematic way by the size of the municipality.26 Nor can the Progress Party’s new
success in parts of the periphery be connected to a powerless periphery with multiple social
problems. The party’s best results are found in affluent municipalities. Consequently, the
theory of a party attracted by modernisation losers – or a party emerging as a protest from the
marginalised old middle class – does not receive any support.
25
This is not a product of the party’s success in Os. When Os is excluded from the calculations, the average
Progress Party gain in municipalities where the party held office drops from 6.0 to 5.9.
26
As already mentioned that was the case in the 1980s and in 1987 support for the Progress Party was related to
low rate of unemployment. In the municipalities with high unemployment, situated in the periphery, the Progress
Party gained few votes.
17
The strength of the Progress Party in different regions and districts reflects the decline
of traditional cleavage politics. According to Hans-Georg Betz (1994) the forces behind this
trend are individualisation and fragmentation following the transition to post-industrial
capitalism. The decreasing importance of old cleavages has various consequences. A room is
left for both issues and candidates.
The party founder Anders Lange detested the established ‘particracy’ and its formal,
bureaucratic organisations. In contrast, his successor Carl I. Hagen has consciously built up a
strong and well-organized party organisation. While other Norwegian parties lose members,
the Progress Party manages to recruit new members.27 The party’s success in municipalities
where it has been in office indicates that local party work and popular local candidates have
been of importance. The party has apparently managed to establish itself as a credible political
alternative in several municipalities.
With regard to issue voting, immigration has clearly had an impact on the party’s
success. From the second half of the 1980s immigration surfaced as a salient issue. We have
documented that the share with confidence in the Progress Party’s immigration policy has
increased during the last years. In addition, among those with confidence in the Progress party
an increasing share point at the Progress Party as the preferred electoral alternative. But the
Progress Party is not a single-issue party.
The start in 1973 was a tax revolt, then gradually the Progress Party advocated promarket solutions and in the 1970s it became a forerunner for neo-liberalism which step by
step has spread from the right to the left on the political party spectrum.28 Now the critique of
the lavish welfare state from the 1970s has disappeared. The Progress Party appears as a
defender of welfare policies, i.e. welfare to ‘our own’ inhabitants (‘welfare chauvinism’).
More money to the health sector and better care for the elderly are their new slogans. At the
same time the Progress Party wants to cut taxes, but less forcefully so than in the 1970s.
Normally this calculation does not add up; the old adage says that you cannot increase public
spending and cut taxes at the same time. But with the Norwegian oil resources available, what
is ordinary considered impossible may now appear possible. The Progress Party is at the
moment the clearest spokesman for the alternative: spending rather than saving the oil money.
27
28
Heidar and Saglie (2002).
Goul Andersen and Bjørklund (2000).
18
Literature
Aardal, Bernt (2003), ‘Ideologi og stemmegivning’ in Bernt Aardal, ed.: Velgere i villrede....
Oslo: Damm.
Berglund, Frode (2002), Partiidentifikasjon og politisk endring. PhD dissertation. Oslo:
Institute for Social Research.
Betz, Hans-Georg (1994), Radical Right-Wing Populism in Western Europe. New York: St.
Martin’s Press.
Betz, Hans-Georg (1998), ‘Introduction’ in Hans-Georg Betz and Stefan Immerfall, eds: The
New Politics of the Right. Neo-Populist Parties and Movements in Established Democracies,
New York: St. Martin’s Press.
Bjørklund, Tor (1988). ‘The 1987 Norwegian Local Elections: A Protest Election with a
Swing to the Right’, Scandinavian Political Studies, 11: 211–234.
Bjørklund, Tor and Jørgen Goul Andersen (2002), ‘Anti-Immigration Parties in Denmark and
Norway: The Progress Parties and the Danish People’s Party’ in Martin Schain, Aristide
Zolberg and Patrick Hossay, eds: Shadows over Europe: The Development and Impact of the
Extreme Right in Western Europe, New York: Palgrave Macmillan.
Givens, Terri E. (2002), ‘The Role of Socioeconomic Variables in the Success of Radical
Right Parties’ in Martin Schain, Aristide Zolberg and Patrick Hossay, eds: Shadows over
Europe: The Development and Impact of the Extreme Right in Western Europe, New York:
Palgrave Macmillan.
Golder, Matt (2003a), ‘Explaining Variation in the Success of Extreme Right Parties in
Western Europe’, Comparative Political Studies, 36: 432–466.
Golder, Matt (2003b), ‘Electoral Institutions, Unemployment and Extreme Right Parties: A
Correction’, British Journal of Political Science, 33: 525–534.
Goul Andersen, Jørgen and Tor Bjørklund (2000), ‘Radical right-wing populism in
Scandinavia: from tax revolt to neo-liberalism and xenophobia’ in Paul Hainsworth, ed.: The
Politics of the Extreme Right. From the Margins to the Mainstream, London and New York:
Pinter.
Grepstad, Ottar (1998), Nynorsk faktabok 1998, Oslo: Nynorsk Forum.
Hagelund, Anniken (2003), ‘A Matter of Decency? The Progress Party in Norwegian
Immigration Politics’, Journal of Ethnic and Migration Studies, 29: 47–65.
Hamilton, Richard F. (1982), Who voted for Hitler? Princeton: Princeton University Press.
Heidar, Knut and Jo Saglie (2002), Hva skjer med partiene? Oslo: Gyldendal Akademisk.
Jackman Robert W. and Karin Volpert (1996), ‘Conditions Favouring Parties of the Extreme
Right in Western Europe’, British Journal of Political Science, 26: 501–521.
19
Knigge, Pia (1998), ‘The Ecological Correlates of Right-wing Extremism in Western Europe’,
European Journal of Political Research, 34: 249–279.
Lipset, Seymour Martin (1981), Political Man. The Social Bases of Politics. Expanded
Edition, Baltimore: John Hopkins University Press.
Martinussen, Pål E. (2002), ‘In Search of the Government in Local Government: Coalition
Agreements and Office Payoffs in Norway’, Scandinavian Political Studies, 25: 139–171.
Martinussen, Pål E. (forthcoming), ‘Government Performance and Political Accountability at
Sub-national Level: the Electoral fate of Local Incumbents in Norway, Scandinavian Political
Studies.
Narud, Hanne Marthe and Henry Valen (2001), ‘Partikonkurranse og sakseierskap’, Norsk
statsvitenskapelig tidsskrift, 17: 395–425.
Narud, Hanne Marthe and Henry Valen (forthcoming), ‘Coalition Membership and Electoral
Performance in Western Europe’ in Kaare Strøm, Wolfgang C. Müller and Torbjörn
Bergman, eds: Coalition Governance in Parliamentary Democracies, Oxford: Oxford
University Press.
Rokkan, Stein (1967), ‘Geography, Religion and Social Class: Cross-Cutting Cleavages in
Norwegian Politics’ in Seymour Martin Lipset and Stein Rokkan, eds: Party Systems and
Voter Alignments, New York: Free Press.
Rose, Richard and Thomas T. Mackie (1983), ‘Incumbency in Government: Asset or
Liability’ in Hans Daalder and Peter Mair, eds: Western European Party Systems, Beverly
Hills: Sage.
Saglie, Jo (1996), ‘Attitude Change and Policy Decisions: The Case of Norwegian Alcohol
Policy’, Scandinavian Political Studies, 19: 309–327.
Seip, Jens Arup (1980), Dyd og nødvendighet. Høyres historie gjennom 100 år. 1880–1980,
Oslo: Gyldendal .
Øidne, Gabriel (1970), ‘Sosial og politisk struktur i Oslo’, Tidsskrift for samfunnsforskning,
11: 125–158, 249–284.
20
Table 1. Support for the Progress Party in local elections 1987–2003, by the size of the
municipalities measured by inhabitants.
Municipality size/
inhabitants
- 2.499
2.500 - 4.999
5.000 - 9.999
10.000 – 19.999
20.000 – 59.999
60.000+
Oslo
Election result
Progress Party
1987 1995 1999 2003 diff. 0387
5,0 4,5
7,7 10,9
+5,9
7,0 6,4
9,9 17,0
+10,0
8,6 8,1 11,1 15,9
+7,3
11,6 11,3 14,1 20,2
+8,6
13,1 13,0 14,1 20,5
+7,4
15,7 13,7 14,9 16,6
+0,9
18,4 20,2 16,6 16,5
-1,9
12,3 12,0 13,4 17,9
+5,6
(N)
(120)
(127)
(90)
(58)
(33)
(6)
(1)
Source: Statistics Norway/The Norwegian Social Science Data Services
Table 2. Social, political and cultural characteristics of the 13 municipalities with mayor
from the Progress Party.
PP Inhabivote
tants
2003
Vestfold
Nøtterøy
Tjøme
Tønsberg
Akershus
Ullensaker
Hordaland
Os
Austevoll
Møre og R.
Sula
Stranda
Skodje
Ørsta
Vestnes
Fræna
Troms
Nordreisa
Country
Peri% nyphery
norsk
index
in
(1919- primary
1994) school
Vote
for
centre
parties
1953
Vote
for
centre
parties
2003
Mean
income
(NOK)
Social % nonindex* western
immigrants
(2003)
Unemployment
rate
(2003)
33.8 19.927
33.2 4.541
30.9 35.656
27.3
26.0
34.4
0
0
0
11.3
13.5
11.3
7.3
6.4
4.3
257.500
244.100
247.200
6.1
7.1
7.0
3.7
2.2
3.3
3.4
4.1
3.8
26.5 22.931
38.3
0
33.2
12.4
271.000
3.2
4.7
3.3
45.7 14.328
33.2 4.439
55.2
66.9
99.8
100.0
49.2
58.6
15,8
22.4
246.000
277.400
4.6
2.7
2.3
1.5
2.8
3.1
38.5 7.304
36.2 4.705
29.1 3.590
24.9 10.269
24.8 6.442
24.5 8.937
72.4
78.2
80.9
85.4
79.8
85.0
99.8
100.0
99.8
100.0
100.0
73.9
44.5
72.3
70.3
65.3
56.9
67.1
18.7
31.6
31.3
33.1
27.4
35.5
222.900
225.700
222.000
211.500
214.900
221.000
4.4
1.7
4.4
5.3
4.6
6.3
2.1
3.4
2.1
2.8
0.9
1.3
5.8
1.4
4.2
3.7
3.3
3.9
32.7
72.9
0
22.3
23.4
201.500
8.6
1.7
5.0
52.9
15,0
29.5
18.6
218.048#
5.6#
5.2
4.2
16.4
4.726
* The lower the figure, the better are the social conditions.
# The mean in 435 municipalities.
Source: Statistics Norway/The Norwegian Social Science Data Services
21
Table 3. Correlations between Progress Party vote in the 2003 county council elections and
social characteristics of the municipalities.
Mean income
Progress Party vote 2003
Difference PP vote 03–99
0.49 *
0.22 *
Social index# % non-western
immigrants
(2003)
0.04
0.26 *
–0.08
0.01
Unemployment rate
(2003)
0.01
–0.09
N=434
* Significant at the .01 level.
# The lower the figure, the better are the social conditions.
Source: Statistics Norway/The Norwegian Social Science Data Services
Table 4. Determinants of Progress Party vote in the 1987, 1995, 1997 and 2003 elections.
Unstandardised regression coefficients.
The share of non-western immigrants
Unemployment rate
R2
(N)
1987
4.10 *
–0.51 *
28%
(435)
1995
1997
1.31 *
1.18 *
–0.04
–0.06
14%
11%
(435)
(435)
2003
1.04 *
–0.00
7%
(435)
* Significant at the .01 level.
Source: Statistics Norway/The Norwegian Social Science Data Services
Table 5. The share with confidence in the Progress Party concerning immigration policy and
the party vote among those with confidence in the party (1997–2003).
Confidence in the Progress Party’ s immigration policy
Party preference among those with confidence in PP. %
Socialist Left
Labour
Centre parties
Conservatives
Progress Party
Other
1997
34 %
2001
33 %
2003
39 %
(1477)
(3079)
(482)
2
30
20
14
32
2
100
6
11
12
22
44
5
100
7
18
11
16
47
2
101
(508)
(1020)
(186)
Questions: 1997: ‘Then the question about how many immigrants should be allowed into Norway. Which party do you feel has the best
policy?’ 2001: ‘Then the question about Norwegian immigration policy. Which party is closest to your own view on this issue?’ 2003:
‘Which party do you have most confidence in, when it comes to immigration policy?’
Source: Norwegian Electoral Programme 1997, MMI Exit Poll 2001, Research International Exit Poll 2003.
22
Table 6. Share of non-western immigrants, votes for the Progress Party and the leftist parties
(Labour, Socialist Left, Red Electoral Alliance) in selected districts in Oslo (2003), ranked
according to the share of non-western immigrants.
Søndre Nordstrand
Stovner
Romsås
Old Oslo
.
Ullern
Grefsen-Kjelsås
Nordstrand
Vinderen
Oslo
Non-western
immigrants %
34.7
33.6
33.6
31.6
.
5.5
4.7
3.9
3.4
16.4
Progress Party %
2003
19.2
26.1
21.4
10.9
.
17.2
14.0
19.1
14.4
16.5
Leftist parties %
2003
55.9
55.6
59.7
66.1
.
26.7
47.1
36.1
26.0
48.6
Søndre Nordstrand: Klemetsrud skole, Prinsdal skole, Hallagerbakken skole, Rosenholm skole, Mortensrud skole, Bjørndal skole, Toppåsen
skole.
Stovner: Stig, Høybråten, Vestli, Stovner videregående skole, Stovner skole.
Romsås: Tiurleiken skole, Nordtvet, Ammerud, Rødtvet.
Gamle Oslo: Hersleb, Kampen, Vålerenga, Gamlebyen
Ullern: Smestad, Lysejordet, Øreaker, Bestum, Ullern videregående skole.
Grefsen-Kjelsås: Kjelsås, Grefsen, Skeidhuset, Korsvoll, Tåsen, Sogn videregående skole, Ullevål skole.
Nordstrand: Ekeberg skole, Bekkelaget skole, Nedre Bekkelaget, Nordseter, Nordstrand, Munkerud, Ljan, Lambertseter skole, Lambertseter
videregående skole.
Vinderen: Vinderen, Slemdal, Hovseter, Midtstuen, Voksen, Huseby.
Table 7. Average gains/losses of votes in the 2003 municipal elections, by party and
incumbency1
Mayor2
Progress
Conservative
Liberal
Christian
People’s
Centre
Labour
Socialist
Left
Out of office
In office
Out of office
In office
Out of office
In office
Out of office
In office
Out of office
In office
Out of office
In office
Out of office
In office
(N)
4
–2.5
–4.7
(289)
(86)
-4
–3.1
–3.6
–0.1
0.0
–0.3
–1.2
(286)
(42)
(296)
(92)
(267)
(163)
-4
Mayor and/or
deputy mayor3
3.7
6.0
–2.4
–4.0
–0.8
–1.7
–2.9
–3.7
–0.3
0.2
–0.2
–1.1
2.6
2.9
(N)
(301)
(20)
(227)
(148)
(261)
(23)
(224)
(104)
(242)
(146)
(208)
(222)
(299)
(29)
1. For each party, we have excluded municipalities where the party did not field a list in 2003.
2. In Oslo and Bergen: Chief Commissioner (head of City Government).
3. In Oslo and Bergen: Member of City Government.
4. Too few cases for statistical analysis.
Source: Statistics Norway/The Norwegian Social Science Data Services
23
22
Figure 1 Support for the Progress Party in the country as a whole, in
Oslo, and in municipalities with a population below 5000, 1973-2003.
Per cent
20
17,9
18
15,3
16
14
12,3
14,6
13,5
13,0
12,0
12
10
8
6 5,0
4,5
4
2
7,0
6,3
1,4
1,9
6,3
3,7
2,5
0
1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003
Country
Oslo
Population < 5000
24
Figure 2 Support for the Progress Party, unemployment rates, and share
of non-western immigrants, 1973-2003. Per cent
18,0
16,0
PP-votes
14,0
12,0
10,0
8,0
6,0
Non-western
immigrants
4,0
2,0
Unemployment
0,0
1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003
25