The ‘ Pots and Pans ’ protests and requirements for responsiveness of the authorities

This paper examines under what conditions it is justifiable that the government takes into account the demands of protesters and whether the terms of procedural-equality in protest participation were met in the ‘Pots and Pans’ protests in Iceland in 2008–09. The protests were triggered by a financial melt-down in Iceland and did not come to an end until almost all the main demands of the protesters had been met. The main conclusion is that due to the seriousness of the issues which triggered the protests and that those issues were of national concern, together with the large numbers of protesters and wide support for their demands as well as extensive public discussion about the issues of the protests, they were a prime example of a situation when the authorities should consider taking the demands of protesters into account. Furthermore, in this paper it is established that giving in to the demands of the protesters was within the terms of procedural-equality between the protesters and those who did not participate – adding to the justification that, in this case, it was justifiable to defer to the protesters’ demands.


Introduction
This paper analyses whether the authorities were justified in considering the demands of those who actively engaged in the so-called 'Pots and Pans' protests in Iceland in 2008-09 and did so without violating the terms of procedural-equality in participation between those who protested and those who abstained.The 'Pots and Pans' protests were triggered by the collapse of Iceland's three major banks in the fall of 2008 in the wake of the global credit crunch.Protests of this scale and length of time had not taken place in the history of Icelandic democracy, a country with almost no tradition of protesting (e.g.Bernburg 2016) and are today, together with the financial collapse, referred to as landmark events in Iceland.The protests ended when the authorities gave into the main demands of the protesters and resigned.Thus, the 'Pots and Pans' protests can, in those terms, be considered to have had an outcome desired by those who had actively engaged in the protests.However, of importance is whether the outcome of the protests was within the principles of democracy in the sense that it was justifiable that the authorities took into account the protesters' demands and did so without violating the conditions of procedural-equality in participation -and these are the main research questions in this paper.Beetham (2003) argues that there are five requirements for when political representatives can be expected to take into account protesters' demands instead of their own decisional autonomy, as follows: 1) the protest issue(s) should be of major importance, 2) the issue(s) should be of national concern, 3) a large number of citizens should participate in the protests, 4) the demand(s) of protesters should be supported by the general public and 5) extensive public discussion and scrutiny of the protest issue(s) should have taken place.In this paper, the question as to whether these five requirements apply to the 'Pots and Pans' protests in Iceland is examined.
In addition to Beetham's five conditions, the requirement of procedural-equality between those who take part in and those who abstain from the protest is added as an important justification for the notion that the authorities are within democratic principles to consider protesters' demands.Based on the idea that political participation is a way for citizens to make their wishes and grievances known to the authorities and to make governments accountable and politicians responsive, Teorell et al. (2007) differentiate between political equality as outcome-oriented equality and as procedural-oriented equality.The requirement of outcome-oriented equality is that the political involvement of citizens should have the purpose of producing the most desirable result.Proceduraloriented equality, hereafter referred to as procedural-equality, refers to the notion that all citizens should be treated equally regardless of whether they take part in trying to influence the authorities.When it comes to political protest, procedural-equality means first that the authorities should not give into, or take into consideration, the demands of protesters at the cost of those who abstain from the protest.Second, proceduralequality means that the terms of participation should be just -that those who have the motivation to participate should not be inhibited from doing so due to lack of resources for participation.This paper proceeds as follows.First is a discussion about Beetham's

STJÓRNMÁL
& STJÓRNSÝSLA five conditions and whether they were met in the 'Pots and Pans' protests in Iceland in 2008-09.After this is a brief discussion of why people engage in political protests and in more detail the importance of procedural-equality in protest-participation.Lastly, an analysis of whether the principle of procedural-equality was violated in the 'Pots and Pans' protests is presented, with a discussion at the end.Beetham (2003) raises the question: after the unprecedented mass demonstrations in the UK against going to war in Iraq what the proper relationship between political representatives and organized public opinion should be and whether governments can simply ignore, at will, citizens' demands that are channelled through mass demonstrations on a large scale.He bases this on the assumption that the mode of decision making in contemporary democracies is changing, moving from the aristocratic principle to the democratic principle.In the aristocratic principle, representatives are not bound by their voters or the electorate at large, and representatives have no obligation to consult with citizens at all.The democratic principle is about the right of every citizen to associate with others to influence their representatives in addition to voting for their preferred party or candidate every few years.As mentioned, Beetham argues that there are five conditions that apply to the issues or circumstances of political protests where political representatives can be expected to defer, or at least take into consideration, the demands of protesters, and these are as follows (p.604):

Political protests and responsiveness of the authorities
1) The issue(s) should be of major importance.A subjective criterion of importance could be if the issue motivates people to take action through protesting.
2) The issue(s) should be of national importance and not merely represent a local or sectional interest.
3) The campaign demonstration should involve large numbers.4) The mobilization should be supported by a clear majority in public opinion polls, preferably over time.5) The issues should have been discussed extensively in a public debate.
If these five criteria are met, this provides grounds, according to Beetham, for elected representatives to take public opinion into account in their decision making.It might be argued that Beetham's view is more in line with decision making in majoritarian democracies and is less useful as an analytical concept in other types of democracies, such as proportional and/or consensual democracies.However, I consider these five criteria to be a useful starting point for analysing under what conditions it is justified, or even required, that the authorities take account of public opinion.To these five requirements I add the importance of the notion that the conditions of procedural-equality between protesters and abstainers should be met if the authorities are to take into account public opinion.

STJÓRNMÁL & STJÓRNSÝSLA
The 'Pots and Pans' protests and requirements for responsiveness of the authorities

The 'Pots and Pans' protests and responsiveness of the Icelandic authorities
The global credit crunch in 2008 hit the Icelandic economy hard, starting with the collapse of the country's three major banks in October.Prior to the economic recession, the banking sector had grown to a size that was beyond the capability of the Icelandic state to support.An example of the over-sized banking sector is that in 2007, the balance sheet of the three failed banks was nine-fold Iceland's annual GDP.All three banks had big operations abroad but were insured, regulated and formally supervised by the Icelandic state (Danielsson & Zoega 2009).
In the fall of 2008, the seriousness of the situation soon started to reveal itself.For example, on 9 October, the British authorities used an anti-terror law to take control of the Icelandic banks in the UK (Bloomberg 2008, Mason 2009), an act which was highly controversial but has never been legally contested.Landsbankinn, one of the failed Icelandic banks, had offered the so-called Icesave deposits in the UK and in The Netherlands in the years before the economic crash.In the UK alone, Icesave depositors numbered approximately 300.000, which is a size similar to the Icelandic population.Following the crash, refunding the customers of Icesave abroad was undertaken first by the British and the Dutch authorities.The terms of the Icelandic state's refunding of the money became a matter of a diplomatic dispute between the Icelandic authorities and authorities in the UK and The Netherlands (Icenews 2011).The matter was finally settled by the EFTA court's ruling in February 2013 in favour of the Icelandic authorities in regard to the terms of the reimbursement (EFTA Surveillance Authority n.d.).Önnudóttir and Harðarson (2011) point out examples of the direct consequences of the recession: the currency restriction imposed in October 2008, rising unemployment, inflation and the need for assistance from the International Monetary Fund (IMF).As early as October 2008, protesters started to gather outside the parliament, and their main demands were that the government should resign, an early election should be held, and that the board of the Central Bank, together with the CEO and board of directors of the Icelandic Supervisory Authority, should resign.The protesters also demanded a revision of the Icelandic Constitution, and this demand can be taken as a sign of how the protest organizers managed to frame the protests as a need for democratic reform (Bernburg 2016).The over-sized failed banks, the financial melt-down, the currency restriction and the British authorities' use of an anti-terror law to freeze the assets of the Icelandic banks in their country are clear demonstrations that the issues that triggered the 'Pots and Pans' protests were both major and national ones.Thus, Beetham's first two requirements, that the protest issue(s) should be both major and concern the whole nation, are met.Even if the demands of the protesters can be narrowed down to the fact that they were demanding resignations, new elections and a revision of the Constitution, it is of importance what triggered those demands -and those triggers were both major and national concerns.
Indicators about the scale of the protests and support can be found in the 2009 Icelandic National Election Study (n.d.) (ICENES).Respondents were asked if they had taken part in a protest after the bank collapse, and 16.8% of them said that they had

STJÓRNMÁL
& STJÓRNSÝSLA done so.When asked if they had supported or opposed the protests, 69.9% said they had supported the protests (very much/tended to support).Both the protest participation and support expressed for the protests are in line with what Bernburg (2016) finds in his research on participation in the 'Pots and Pans' protests.Both ICENES data and Bernburg's research show that Beetham's (2003) requirements, that the protests should include a large number of participants and be widely supported, are met.The length of the protests is here of importance as well.Bernburg (2015) describes how the protests evolved, with the first protests organized immediately following the bank collapse in early October.Shortly after that, a series public meetings (both indoors and outdoors) started, where social critics, intellectuals and activists gave speeches about the situation.Önnudóttir and Harðarson (2011) describe how the government early on seemed intent on ignoring the protests; but as the weeks passed, the protests intensified in size and noise.After almost four months of protesting, and some minor clashes between the protesters and the police, the government gave in and resigned at the end of January 2009.Shortly after that followed the resignations of the heads of the Central Bank and the Icelandic Financial Supervisory Authority, and an early election was scheduled and took place April 25, 2009.
The time that passed from the start of the protests (early October 2008) until the government resigned (end of January 2009) indicates that there was space for extensive public discussion, and scrutiny of the demands of the protesters took place.Since then, there has been an ongoing debate and discussion about what went wrong in the Icelandic political and financial system (e.g.Bergmann 2014;Mixa 2009;Þórlindsson & Jónsson 2009;Special Investigation Commission 2010;Önnudóttir & Harðarson 2011) and about the consequences of the recession (e.g.Önnudóttir et al. 2016;Einarsdóttir 2010;Oddsson 2010).While it is not clear whether and what role the Constitution played in the perceived failings of the political system to prevent or mitigate the economic collapse, the protests put the issue of its revision on the agenda.Early in 2011, the parliament appointed 25 members to a Constitutional Council 1 , and in mid-summer 2011 the Constitutional Council handed over a draft to the national parliament for a new constitution (Constitutional Council 2011).The parliament took the revisions under consideration but did not present a new draft before the 2013 national election -and the revision of the Constitution is still an open question.
It is clear from this short overview of the financial melt-down in Iceland that according to the circumstances of the 'Pots and Pans' protests, Beetham's (2003) five conditions that provide grounds for political representatives to consider and/or defer to public opinion are fulfilled.The protests were triggered by events that were both major ones and national ones.The protests involved large numbers and were supported by a clear majority of the Icelandic population.The events that triggered the protests were at the time and have since then been discussed extensively in public debate.The Icelandic government did give in and resign, but it took almost four months of protesting that escalated both in size and noise.
Beetham's five requirements indirectly touch upon the principle of equality, which

STJÓRNMÁL & STJÓRNSÝSLA
The 'Pots and Pans' protests and requirements for responsiveness of the authorities relates to the notion that the authorities should guarantee that the demands of the protesters are not given into at the cost of those who abstain.However, I do not consider those five criteria to be sufficient, and procedural-equality in political participation should be more explicitly stated.Adding the requirement of procedural-equality in protest participation is based on whether the active public, those who protest, are representative of the public at large (Teorell et al. 2007).Procedural-equality between those who participate and those who abstain is an important justification, in addition to Beetham's five requirements, so that the authorities act according to the democratic principle when they account the demands of protesters and do not do so at the cost of those who abstain.Moreover, procedural-equality also means that participation in protests should not be repressed due to lack of resources people have for protest participation if they have the motivation to take part.

Protest participation and procedural-equality
Considerable research has been devoted to why people take part in protests (e.g.Teorell 2006), possible feedback effects for participants in protests (e.g.Opp & Kittel 2009) and different motivations to protest depending on the issue of the protests (Norris et al. 2005).One piece of the puzzle is why people invest time and other resources in participating in demonstrations as compared to low-cost activities such as signing a petition or boycotting.This is even more puzzling given the fact that the rewards are often scarce and that even if the protesters demands are met, the results are often far from their expectations (Opp & Kittel 2009).
When examining what factors can explain why people engage in political protest, a distinction must be made between macro-level factors and micro-level factors.On the individual (micro) level, it has been established that the active membership of various organizations increases the likelihood of taking part in political protests.Factors such as modernization, resources and opportunities to protest (macro) have been shown to explain cross-country differences in protest activity (Roller & Wessels 1996).The interplay between micro-and macro-level factors is of importance, or, as Roller and Wessels (1996) point out, individual-level activity is within the context and the flexibility of political system to protest.Furthermore, the issue of the protest is of importance.Norris et.al. (2005) find in a study on protests in Belgium that different types of issues mobilize different ideological groups of protesters.Those who they label as new-left are more likely to participate in protests against globalization and racism and the old-lefts to participate in protests concerning social security.
In distinguishing between three models of democracy -participatory, responsive and deliberative democracy -they can be found to differ in their emphasis on the amount and type of citizen participation (Teorell 2006).The responsive model requires a minimum of participation of citizens, such as by voting every few years.The participatory model emphasizes participation beyond voting, and the deliberative model requires involvement in political discussion about policy decisions.Whether and to what extent citizens value different types of democracy is often discussed in terms of how policy

STJÓRNMÁL & STJÓRNSÝSLA
decisions are arrived at and whose interest should be kept in mind when making those decisions -and are, as such, outcome-oriented.For political decisions to be fair and just, the decisions taken, or the outcomes, must be within the terms of outcome-oriented equality.This means that the 'best' decision is taken, a decision that is fair in terms of both how it is arrived at and in terms of whom or what groups it concerns.
However, as Teorell (2006) points out, to understand why people participate in politics, including protests, in addition to outcome-oriented equality, procedural-equality is of importance as well.Procedural-equality means that the terms of participation should be fair.Focusing on protest participation, the principle of procedural-equality concerns whether the terms of participation are equal between protesters and abstainers.Procedural-equality in the context of participation in protests is defined by Teorell et al. (2007) as the notion that every citizen should have the same opportunity to be politically involved if he or she chooses to be so.If people have the motivation to participate but their participation is repressed due to lack of resources or capacitating factors they can draw from for their participation, this violates the terms of procedural-equality.This also means that if people are politically involved because of their motivations and/ or ambition to be so, and not only because they have the capability to participate, their participation adheres to terms of fairness in political participation or, in other words, to terms of procedural-equality in participation.Teorell's (2006) distinction between the resources people have for participation and their motivation to participate is of importance (see also Teorell et al. 2007).Resources, also referred to as capacitating factors, are about the factors enabling participation, while motivation is about whether people have the ambition to or want to participate.Teorell divides resources for participation into physical, human and social capital.Physical capital refers to material assets, such as an individual's income, wealth and even free time.Human capital is about human skills and enabling factors such as education, political knowledge and political efficacy.The more educated one is, the greater knowledge one has about politics; and the greater sense of political efficacy one has, the more resources one has in terms of human capital for political participation.Social capital refers to how involved people are in social networks; and the more involved one is, the more likely one is to be engaged in politics.It is acknowledged that these three types of capital can and do overlap, but the distinction is useful in the sense that it draws attention to the fact that citizens can have low resources in one type of capital, but they can make up for this by having higher resources in other types of capital.
Motivation to participate in protests is in its simplest form is whether or not people have a desire to participate.According to Teorell et al. (2007), the motivation to act is a function of each individual's perceived stake in the issue at hand.This can, for example, be his or her dissatisfaction with how the political system works.Motivation can also be driven by factors such as ideological attachment and a civic duty to take part.Teorell et al. (2007) make the point that information and information cost are important factors for motivation to act and that political interest, frequency of discussing politics with others and media exposure to political content are all factors that decrease information

STJÓRNMÁL & STJÓRNSÝSLA
The 'Pots and Pans' protests and requirements for responsiveness of the authorities cost.However it can also be argued that information and information cost are resources for participation, not only incentives.These can reflect both a motivation and a resource for participation in the sense that the more informed one is, the less time one has to invest in becoming informed and is, as such, a resource for participation -and the information itself can feed into a motivation to take part in protests.Also, as pointed out by Teorell (2006), resources can have an impact on incentives to participate in terms of the idea that those who are low on resources feel less motivated, because it is too costly for them.Or, it could be the other way around: those who are motivated to participate do invest in resources for participation (e.g.time and information).However, of analytical importance for the terms procedural-equality in participation is that all citizens have equal resources to be politically involved regardless of whether or not they participate.
While it is also of importance to understand the relation between resources and incentives, that is set aside in this paper for future consideration.In this paper, the focus is on comparing participants and abstainers in the 'Pots and Pans' protests in terms of procedural-equality -as a crucial requirement and an addition to Beetham's (2003) five criteria discussed earlier in this paper -as an indicator that the authorities were justified in giving in to the main demands of the protest participants.The terms of proceduralequality mean that it is of importance that those who have the motivation to act are not inhibited due to their lack of resources.Moreover, protest participants should not be able to impose their will against the abstainers' wishes simply because they have more resources to protest.Even more important is that if abstainers from the protests have a stronger motivation to participate when compared to participants but are lower on resources for participation, this violates the terms of procedural-equality in participation -and this part I test directly here and hypothesize the following: H1: If people had the motivation to participate in the 'Pots and Pans' protests, low resources for participation did not prevent them from participating.

Procedural-equality in participation in the 'Pots and Pans' protests
To analyse whether the requirement of procedural-equality in political participation was met in the 'Pots and Pans' protests, I use data from ICENES 2009 and run a binary logistic regression analysis.The response variable is whether or not respondents participated in the 'Pots and Pans' protests 2 .Those who took part in the 'Pots and Pans' protests are assigned the value of 1 and are contrasted with those who abstained, which are coded with a 0. Explanatory variables are divided into motivational and capacitating (resources) factors, and my indicators are partly based on those Teorell et al. (2007) use in their analysis on procedural-equality in political participation.For motivational factors, I consider citizens' stake in the issues to be the major factor in mobilizing them to protest.Dissatisfaction with how democracy works 3 and the blame they assigned to the incumbent government for the crisis 4 are used as indicators of citizens' stake in the issues of the

STJÓRNMÁL & STJÓRNSÝSLA
protests.The more dissatisfied they are and the more they blamed the government, the more their motivation to take action should increase.
The most obvious form of physical capital as a resource is respondents' income, and for those purposes I use a question about their household's income 5 .For better consistency with other explanatory and control variables, household income is rescaled to a scale from 0 (lowest income) to 1 (highest income).Respondents' experience of working together with other people who share a similar concern can be taken both as a source for human capital, as Teorell et al. (2007) do, and as a form of social capital (e.g.Putnam 2000).Regardless of whether respondents' experience of working together with other people is a form of social or human capital, it can be taken as a capacitating factor for political participation.The more people have of such experience, the more socialized people are in participating in collective action.For this I use a question about whether the respondents have worked together with people who shared similar concerns over the last five years 6 .Education can be considered a capacitating factor, and it has been shown to have a strong link with political internal efficacy (Morris 2003), which are citizens' beliefs that their participation in politics is important and that by it they can accomplish things in politics.Thus, education can enhance the perceived capability to act and the belief that participation in protests is of importance, whether it is taken as merely a civic duty to do so or as a perception that participation will put pressure on the authorities to respond to the protesters' demands.
For the terms of procedural-equality to be met, peoples' resources for participation should not prevent them from participating if they have the motivation to take part.This means that motivation should predict participation, not the resources respondents can draw from for their participation.Moreover, the terms of procedural-equality in participation mean that motivation for participation should not be a sole function of peoples' resources for participation.To test whether the effect of motivation to participate in the 'Pots and Pans' protests depends on resources for participation, I include interaction terms between those variables measuring motivation to take part (dissatisfaction with how democracy works and blame assigned to the incumbent government) and respondents' resources (income, worked with other people who share a similar concern and education).
Eight additional factors are controlled for, including political interest; ideological placement on left-right; whether the respondent voted and whether he or she voted for one of the incumbent government parties in the 2007 election; whether or not they live in the capital area; and their gender, age, marital status and number of children living in their household -and each are discussed here briefly.
The more interested people are in politics, the more informed they already are, and thus their information cost is lower (e.g.Teorell et al. 2007).Low information cost can be considered a resource one can draw on for his or her participation, and the information itself can act as a motivation factor for participation.Of importance for the purpose of this paper is that political interest is an important factor when it comes to

STJÓRNMÁL & STJÓRNSÝSLA
The 'Pots and Pans' protests and requirements for responsiveness of the authorities lowering information cost, and it has been established that those who express higher interest in politics are also more likely to take part in protests (Martin & van Deth 2007).Regardless of whether political interest reflects a resource, a motivation or a mix of both, it is an important factor to control for when analysing protest participation.This is, among other things, to prevent possible bias when estimating peoples' motivation to participate and so that this motivation is not driven solely by the fact that they are more interested in politics.
Ideological placement has been shown to be of importance in protest behaviour, together with the issue of the protest.Those who are left-leaning have been shown to be in general more likely to protest (e.g.Bernburg 2015;Teorell et al. 2007).Given that the 'Pots and Pans' protests were against a government led by a right-wing party (a coalition of a right-wing party and a centre-left party) and that it could be argued that the issue of the protests concerned economic security, it can be assumed that left-wing voters had a stronger motivation to take part in the protests.To control for this, I use an indicator about respondents' left-right self-placement 7 .
As a control, the 'home-team' argument is also relevant.The 'home-team' effect refers to when voters of government parties express more satisfaction with how democracy works and support for the government, simply because the party they voted for is in government (e.g.Anderson & Mendes 2005;Holmberg 1999).It could be argued that if the only thing that distinguishes between protesters and abstainers is whether or not they voted for one of the government parties, the protests would be a manifestation of the 'usual' political debate between the opposition and the government.For these reasons, both as a motivation to participate and as a control for support for the government, I add a variable that indicates whether the respondent voted for one of the incumbent government parties at the time of the protests (Independence Party or Social Democratic Alliance).
I also control for whether or not respondents voted.A little under 10% (9.8%) of respondents in ICENES reported that they did not vote.As we do not know whether they would have voted for the government or the opposition, their electoral participation needs to be controlled for, and for that a dummy is used which indicates whether or not they voted in the 2007 election.Residency is a capacitating factor in the sense that it has an obvious link to the opportunity citizens had to take part in the protests.The major bulk of the 'Pots and Pans' protests took place in front of the parliament in Reykjavik, the capital of Iceland.This means that those living away from the capital area did not have the same opportunity to participate as those living within the capital area due to the inconvenience of having to travel to the city to become actively involved.
Considering age, the youngest and the oldest are least likely to protest (e.g.Teorell et al. 2007;Norris 2005) and thus I control for age 8 and age squared, with the latter estimating a curvilinear effect.By adding age as a control factor, I lower the risk of bias due to the possibility that the motivational and capacitating factors of interest are biased because of their relation with age (for example, household income and education).Earlier research has established that females are less likely to protest (e.g.Dalton 2002), and thus

STJÓRNMÁL & STJÓRNSÝSLA
I control for respondents' gender 9 (male or not).Furthermore, as I use respondents' household income as a capacitation factor, the number of children 10 in the household are controlled for, together with respondents' marital status 11 (household income does not distinguish between whether there is one salary earner and how many dependents there are in the household).
Table 1 presents the results of a binary logistic regression comparing the effect of motivational and capacitating factors on protest behaviour.I present five models.The first model incorporates only the control variables and the second and third models the variables which are of main interest in this paper -first the capacitating factors and next the motivational factors.In the fourth and fifth models, I introduce the interaction variables, testing whether participation in the protests for those respondents who had the motivation to take part was repressed by their lack of resources for participation.This I do first for dissatisfaction with how democracy works and second for how much blame respondents assigned to the incumbent government for the economic crisis.I will focus my discussion on the latter three models but would like to note that when adding the capacitating factors in model 2, those factors have a negligible effect on protest participation, as they do also in model 3.However, in models 4 and 5, household income and the interactions between income and motivation to take part are statistically significant, with a p-value lover than .1.I will now turn the discussion to the main results in models 3, 4 and 5.
Considering first the control variables, it is not a surprise to find that those who participated are more interested in politics, more left-wing and more likely to live in the capital area.Those who voted in the previous election are less likely to have participated in the protests, and this might reflect a distinction between conventional political participation, such as voting, and 'unconventional' participation, such as protesting.The effect of the 'home-team' argument, which is that those who voted for one of the incumbent parties in the previous election should be less likely to have participated is negative, supporting the argument.However, this effect is not strong enough to be statistically significant in models 3 and 4. The direction of the effect of electoral participation and to have voted for one of the incumbent government parties indicates that in the 'Pots and Pans' protests the electoral support of the incumbent government could have been in part a mobilizing factor for protest participation -and underscores the importance of controlling for those factors.

STJÓRNMÁL & STJÓRNSÝSLA
The 'Pots and Pans' protests and requirements for responsiveness of the authorities
In regard to one of the main concerns in this paper, I find for citizens' stake in the issue in models 3 and 4 that those who express greater dissatisfaction with how democracy works are more likely to have participated in the protests.Model 3 also shows that the more the respondents blame the government for the economic crisis, the more likely they are to have participated in the protests.However, when interacting blame with household income, which is one of the resources respondents can draw on for their participation, the effect of blame on participation becomes negligible, whereas household income becomes significant as well as the interaction between income and blame.A similar pattern is found for dissatisfaction with how democracy works; when interacting it with household income in model 4, the main effect of dissatisfaction becomes weaker (but is still statistically significant), and both the interaction term with household income and the main effect of household income are significant.Before examining these interaction effects graphically, it should be noted that all other interaction terms in models

STJÓRNMÁL & STJÓRNSÝSLA
The 'Pots and Pans' protests and requirements for responsiveness of the authorities 4 and 5 are not statistically significant.This indicates that protest participation is not repressed among those who have the motivation to participate due to lack of resources in terms of their experience in working together with other people who share similar concerns and their education level.
Figures 1 and 2 show the marginal effects of citizens' stake in the issue, dissatisfaction with how democracy works and how much responsibility respondents assigned to the government for the crisis 12 , and how the effect of those on participation interact with household income 13 .Focusing first on dissatisfaction with how democracy works in Figure 1, the general trend is that those who are more dissatisfied are more likely to have participated.But there is a difference in protest participation depending on household income, which is that income has a stronger effect on participation among those who express the most dissatisfaction (not at all satisfied) when compared to those who are less dissatisfied.The difference in probabilities to have participated in the protests between respondents in the lowest and highest quintile among those who were not at all satisfied with how democracy works is .09,where higher-income voters are more likely to have participated, with a margin of .30,compared to low household income respondents, with a margin of .21.The difference between the same income groups among those who were very satisfied is much smaller, or .04(high household income respondents are slightly more likely to have participated).Household income:

STJÓRNMÁL & STJÓRNSÝSLA
Interestingly, the interaction effect between household income and the responsibility the respondents assign to the former government for the crisis is reversed.The general trend is still (but not significant) that the more the government was blamed for the crisis, the higher the probability that respondents took part in the protest -and household income still has a stronger impact among those who assigned more responsibility compared to those who blamed the government less.The difference is that, when compared to dissatisfaction with how democracy works, low household income respondents are those who are more likely to have participated in the protest compared to high household income respondents, after controlling for how much they blamed the government for the crisis.Furthermore, those differences are bigger the more the government was blamed for the crisis.Among those who assign much responsibility (+.5 st.dev.above the mean) to the government for the crisis, the difference between the lowest and highest income groups in probability to have participated is .19,where low household income respondents are more likely to have participated, with a margin of .31,whereas high household income respondents have a margin of .12.The difference between the same income groups among those who assign low responsibility to the government for the crisis (-.1.5st.dev.below the mean) is .12,where low household income respondents are still more likely to have participated.Household income:

STJÓRNMÁL & STJÓRNSÝSLA
The 'Pots and Pans' protests and requirements for responsiveness of the authorities The reason for the reversed interaction effects between, on one hand, household income and, on the other hand, dissatisfaction with democracy and blame assigned to the government can only be suggested here.It could be that there are partly different reasons that drive motivation between high and low household income respondents; for example, high income respondents are less likely to target the government as the culprit for the economic crisis and more likely to blame the democratic system, whereas for low-income voters this is reversed (they target the government to a greater extent and the democratic system to a lesser extent).However, what can be argued is that these differences in the interaction effects indicate that the two motivation factors used in the models presented in this paper do partly reflect different motivations to participate in the 'Pots and Pans' protests.

Discussion
Using Beetham's (2003) five criteria for when the authorities should consider to defer to the demands of protesters -that the issue(s) should be of major importance, that the issue(s) should be of national concern, that the number of protest participants should be high, that their demands should be supported by the general public, and that the protest issue(s) should have been scrutinized in public debate -I show that those conditions were all met in the 'Pots and Pans' protests.Moreover, I add the criterion that the terms of procedural-equality in protest participation should be met as a necessary justification for when it is justifiable that the government defers to the demands of protesters.The terms of procedural-equality in protest participation are important to prevent a situation where the protesters' can force their demands to be taken into consideration against the wishes and interests of those who do not participate in the protests if participation were to be solely be driven by resources for participation.This part of the terms procedural-equality coincides directly with Beetham's requirement, that the protest issue(s) should be supported by the general public.Moreover, the terms of procedural-equality are violated if the abstainers have a stronger motivation than participants to participate but are inhibited to take part due to their lack of resources.In this paper, I have established that the conditions of procedural-equality between participants and abstainers in the 'Pots and Pans' protests were indeed met -and this together with Beetham's five requirements indicates that the authorities justified in giving in to the main demands of the protester participants.
Of importance for the terms of procedural-equality in political participation is that the main difference between those who are active and those who abstain should be first and foremost be viewed in terms of motivational factors, not the resources they have to participate.Examining the differences in capacitating factors between participants and abstainers in the 'Pots and Pans' protests, such as in their experience in working with other people who share similar concerns and education, there are no signs that the participants differ from abstainers when it comes to the resources they can draw upon to be actively involved.Protest participants did not have more human and/or social capital in terms of being better educated or of having had more practice in using civic skills when compared to abstainers.

STJÓRNMÁL & STJÓRNSÝSLA
When it comes to physical capital, respondents' household income does seem to have an effect on participation when interacted with dissatisfaction with how democracy works and the blame respondents assign to the government for the crisis -but those interaction effects are still within the terms of procedural-equality for two reasons.First, in the case of dissatisfaction with how democracy works, high household income seems to accelerate the probability for participation the more dissatisfied one is.Because the main effect of dissatisfaction with democracy is still positive, this means that motivation still drives participation.Thus household income seems to heighten participation.However, due to the finding that among those who have low household income, those who are more dissatisfied are more likely to have participated compared to those who are less dissatisfied -participation is still within the terms of procedural-equality.The terms of procedural-equality would have been violated if low household income were to have been accompanied with a stronger motivation to participate when compared to those with high household income.Second, in the case of the blame assigned to the government for the crisis, the interaction is reversed.In this case, stronger motivation to take part in terms of how much the government is blamed has a stronger effect on low household income voters, which are also more likely to have participated, when compared to high household income voters.This means that even if high house-income has a stronger effect when it comes to dissatisfaction with how democracy works as a motivation to participate, the reverse interaction between income and the blame assigned to the government, where the blame is a stronger mover of low income voters, indicates that the differences between household income groups balance each other out depending on which motivation is focused upon.Thus it can be argued that my results indicate that in the 'Pots and Pans' protests, the terms of procedural-equality were not violated -that inequality in resources did not prevent participation among those who had the motivation to participate.
Even though the data used in this analysis were gathered after the 'Pots and Pans' protests took place (but still within eight months since the end of January 2009 when the government resigned), it is not a plausible assumption that the protests created procedural-equality in terms of protest participation.The protests can possibly have intensified the motivational factors, but I assume that participants were motivated to protest from the start.The 'Pots and Pans' protests can probably be considered an exception in the sense that it was a rare type of circumstances that made it justifiable for the authorities to give in to the protesters' demands.In this sense these protests can be considered to have been effective, as all the main demands of the protesters were met, even if results were not as revolutionary as some protest organizers might have liked.The government was held accountable and resigned, an early election was called and the directors of the two main financial regulatory institutions -the Central Bank and the Financial Supervisory Authority -also resigned.Even if it is true that the government could not have prevented the economic recession, it was still blamed for how severely it hit Iceland, echoing Kayser's (2005) point that there are numerous examples that voters hold politicians accountable for events that are beyond their control "… that they can

STJÓRNMÁL & STJÓRNSÝSLA
The 'Pots and Pans' protests and requirements for responsiveness of the authorities at best mitigate but cannot prevent" (pp. 21-22).Moving the focus away from the 'Pots and Pans' protests to protests in general, there are reasons to believe that protests similar to the 'Pots and Pans', which fulfil the requirements that make it justifiable for governments to consider to defer to the demonstrators' demands, are rare.In this paper, I have demonstrated that the 'Pots and Pans' protests were well within the framework of democratic principles, and I have established the importance of taking into account the terms of procedural-equality in protest participation.A research topic that can be developed from that presented here is how participation and widespread support for major protests such as the 'Pots and Pans' can add to citizens' learning experience in political participation, especially if the feedback effects are positive.Whether the authorities should consider taking account of demands of those who actively participate should depend on, among other things, the nature and the scope of the issue, that extensive public discussion about the protest issues have taken place, that the protesters' demands are supported by the general public and that participation in the protests is first and foremost driven by a motivation to take part, which is not repressed due to a lack of resources people can draw upon for their participation.bank collapse last October?' -was a follow-up question from 'Over the past five years or so, have you taken part in a protest or a march?'.In both questions, the response categories were either yes or no.In this paper, I use respondents' replies to the question about whether they have participated after the bank collapse in October'.3 Question asked: 'On the whole, are you very satisfied, fairly satisfied, not very satisfied or not at all satisfied with the way democracy works in Iceland?' 4 Question asked: 'Various agents have been mentioned as being responsible for the bank collapse and the economic crisis that followed.Now I will name a few of them and ask you to give each of them a number from 0 to 10, where 0 means that the agent in question did not bear any responsibility at all and 10 that the agent in question bears very great responsibility.Where would you place... the Government of Geir H. Haarde?'. 5 Question asked: 'What was your/your and your spouse's total income in the last month before taxes and other deductions approximately?'.Open response.6 Question asked: 'Over the past five years or so, have you worked together with people who shared the same concern?'.Response categories were yes or no.7 Question asked: 'Where would you place yourself on this scale?' (…on a scale from 0 to 10, where 0 means the left and 10 means the right).8 Information about age registered from sample.9 Information about gender registered from sample.

STJÓRNMÁL
& STJÓRNSÝSLA with a mean of 8.1 and st.dev. of 2.1, on a scale from 0 (no responsibility) to 10 (a lot of responsibility).13 Household income is divided into five evenly distributed quintiles, and the predicted probabilities in Figures 1 and 2 are computed using the mean of each quintile.

Figure 1 .
Figure 1.Marginal effects of dissatisfaction with how democracy works, household income and protest participation -calculated based on model 4 in Table1.

Figure 2 .
Figure2.Marginal effects of the government's perceived responsibility for the bank crisis, household income and protest participation, calculated from model 5 in Table1.
Figure2.Marginal effects of the government's perceived responsibility for the bank crisis, household income and protest participation, calculated from model 5 in Table1.
s responsibility for the crisis (0=none, 10=a lot) Mean of first quintile (lowest) Mean of second quintile Mean of third quintile Mean of fourth quintile Mean of fifth quintile (highest)

Notes1
Those same 25 members of the Constitutional Council had been elected to the Council in the fall of 2010.The election was invalidated by the Supreme Court on 25 January 2011 due to technicalities in how the election was carried out.The parliament decided on 24 March 2011 to appoint the 25 elected candidates to the Constitutional Council. 2 The question about participation in the 'Pots and Pans' protests -'But have you done so after the

Table 1 .
Participation and support for the 'Pots and Pans' protests.
The government coalition of the Independence Party and the Social Democratic Alliance.