About EKOS Politics

We launched this website in order to showcase our election research, and our suite of polling technologies including Probit and IVR. We will be updating this site frequently with new polls, analysis and insight into Canadian politics. EKOS's experience, knowledge and sophisticated research designs have contributed positively to many previous elections.

Other EKOS Products

In addition to current political analysis, EKOS also makes available to the public general research of interest, including research in evaluation, general public domain research, as well as a full history of EKOS press releases.

Media Inquires

For media inquires, please contact: Frank Graves President EKOS Research Associates t: 613.235-7215 [email protected]

At the Crossroads of Hope and Fear


Please click here for a PDF copy of this article.

[Ottawa – February 5, 2018] In this unique ‘populism project’ we are conducting with the Canadian Press, we are trying to gain a better understanding about what many are calling the rise of populism. That term is inadequate to deal with the range of experience and attitudes which are underpinning this movement. A partial list of the closely related concepts relevant to this includes authoritarianism, an ordered versus open outlook, nativism, isolationism, and xenophobia. Questions swirl as to what these forces mean, what is driving them, and how they are distributed in Canada.

Our evidence shows that those who feel that Canada is somehow inoculated from these forces are simply wrong. We are going to examine the linkages across economic anxieties, class formation, and cultural insecurity. Whatever the right language for this debate the new locus of conflict in modern society has shifted from left-right to open-ordered. We will try and explain this with some recent evidence.

Charting the end of progress and its consequences

This research is the latest iteration of work that we have been conducting for the past decade. Around a decade ago, we began to notice that some of the typical outlook on the economy and one’s place in it were fundamentally different than it had been in the last century. The basic ideas of progress, shared prosperity, and subscription to the middle-class dream all appeared to be unravelling.

Since that time, these patterns have only become clearer and are starkly different from the recent notion that the upper North American economies are doing splendidly well. In both Canada and the United States, there has been a large hollowing of the middle class as the ranks of the erstwhile middle class deplete so grows the ranks of the burgeoning working class. While those at the top are doing very well, there is a pervasive sense of stagnation and decline elsewhere. The basic middle class bargain which defined the period of shared prosperity in the last half of the twentieth century is in tatters. For many, we have reached the end of progress. Only one in eight think they are better off than a year ago. An identical 13 per cent think the next generation will enjoy a better life. By more than two to one, Canadians agree that if present trends with inequality at the top continue then we are going to see ‘violent class conflicts’.

The (at the time) shocking Brexit result, and even more astonishing election of President Donald Trump vividly reflect some of these forces. They are both reflections of these forces and may well be adding further impetus to them. As if the economic challenge of restarting middle class progress wasn’t daunting enough, we now see mutation expressing itself in higher levels of cultural insecurity which is linked to a disturbing rise in racism and nativism. In a very real sense, the culture wars of the last century are morphing into the class wars of the 21st Century.

We are going to try and chart some of these trends in Canada. The same surprise which accompanied the Trump and Brexit results is steeped in a denial or blind spot in the institutional status quo. The dismissal of these new unsavoury forces as ‘deplorable’ or an expression of moral or intellectual frailties merely adds fuel to the populist fire.

What we have done to try and clarify this picture is to pull together some of the data that we have assembled over the past year. In particular, we are aggregating two very large random probability samples; one of about 5,700 Canadians conducted in June and another of nearly 8,000 that we have just completed. The data have been aggregated to allow us to provide a statistically reliable picture of the locus of these forces in Canada. We have also linked the data to the settlement that people live in so that we can map the distribution of this new open-ordered axis.

Before considering the quantitative components of what we are calling the open-ordered index, it will be helpful to consider the more qualitative meaning of this term. This brief description reflects some of our other work and the broader literature.

We are arguing that the changes in economic outlook and class formation that have occurred in this century have shifted the traditional left-right axis of societal dispute to this new open-ordered axis. There is a clear lineage connecting the left-right and open-ordered, but the latter has become the more relevant term to describe the fundamental contest for what kind of society we want.

Our argument is that the end of progress and has been critical in fuelling the rise of this recent expression of populism. The collapse of the middle class bargain of shared prosperity and progress has produced profound shifts in the belief in the current economic model operating in upper North America. This shift has been reinforced by a growing consensus that the external world has become a more dangerous place and this too is fuelling a greater apprehension about the merits of trade liberalization, globalization, immigration, and a general belief that those who have been running things are not really concerned about the public interest. This new outlook has fuelled the Brexit and Donald Trump victories and it is very much at play in Canada, albeit with a somewhat different and less intense expression (for now).

Consider some of the terms that capture either end of this continuum. The essential ingredients are both economic (particularly economics security or insecurity) and cultural (broad value orientations and beliefs). The open side of the spectrum is the paradigm under siege. It is future-oriented, welcomes change, and is receptive to an open approach to trade and immigration. It tends to be optimistic and progressive and is most evident in those currently seeing themselves as members of the middle and upper class. We shall see that there several other important features of this outlook. The ‘ordered’ side of the continuum is basically a darker mirror image of the open side; it questions the current institutional order, which has been seen to have abandoned those taking this view. It is both economically and culturally insecure. Support for trade is more tepid as is support for immigration. The outlook is associated with a more authoritarian view which sees order and obedience as more important than creativity and reason. It tends to be nostalgic for a previous (perhaps apocryphal) era where comfort and security were enjoyed. Those attracted to this outlook tend to be in the working and poor classes, many of whom have experienced downward class mobility.

We note that there is an important place for both the ordered and open world views. Healthy societies require both. What is new is the engagement of what Theodor Adorno and others (such as Will Davies and Amanda Taub) called authoritarianism (closely linked to the ordered outlook). While societies without order would be chaotic the problem seems to occur when the ordered type becomes the dominant political power.

With this brief conceptual overview let’s consider the distribution and trajectory of these forces in Canada. We have several different empirical indicators that we have been testing over the past several years. For this exercise we have created an index which measures this axis based on a few key indicators drawn from this research. While an imperfect measure it does seem to capture a large portion of what we talked about above.

The index has strong demographic and attitudinal correlates which perform in a theoretically plausible manner.

Despite strong debate about whether economic insecurity or cultural insecurity is more important, our research suggests that both are inextricably woven together to shape these orientations. It is most likely that economic insecurity triggered this movement. It is, however, arguable whether economic insecurity was more important than this new closing of outlook and the desire to abandon the open society in favour of a more inward-looking, isolationist society which is more allergic to the idea that diversity and multiculturalism are good things.

The open-ordered index

We now have an index which combines (the actual scoring system and index ingredients are presented in Annex I):

  • Class mobility (moving upward downward of staying the same);
  • Economic outlook on the future (optimistic, stable, pessimistic);
  • Ethnic fluency (in ballpark, moderately off, severely exaggerated);
  • Racial tolerance (too many, too few, or about the right number of visible minorities in Canada); and
  • Pro-populism (good thing, indifferent, or a bad thing).

In the rest of this release, we will show this continuum is distributed in Canada and how it links to broader attitudes to politics and big policy questions. While there are some intriguing features to the geography of populism in Canada, the more profound connections are to social class, economic outlook, and cultural outlook. The connections to partisanship and political ideology are also very striking.

A few notes on the ingredients of the index:

  1. If we chart class membership through time, the patterns are decidedly downward (from middle to working class) (see Figure 6). Asking people their class status of ten years ago and comparing to it their current status provides a more mixed measure with an overwhelming lean to staying the same. But the process of falling backward is strongly linked a more ordered outlook.
  2. Economic outlook may be good on Bay Street but it sucks on Main Street. Few Canadians see themselves moving forward in the future and this sense of decline and stagnation is strongly linked to an ordered outlook.
  3. A pretty stringent measure of racial intolerance would be the view that too many Canadians simply aren’t white. About one in four hold this view, but it is much stronger among those subscribing to an ordered outlook. Notably, this measure has shown declining racial intolerance over the past few years (as has the more widespread view that there are too many visible minority immigrants) (see Figure 1).
  4. Ethnic fluency is pretty sketchy in Canada; we tend to overstate the presence of different minorities. The degree to which the exaggeration exists is strongly linked to racial intolerance and a more ordered world view (see Figure 2).
  5. Contrary to the generalized horror that populism seems to engender in established Canada, overall the public aren’t certain whether it’s a good or bad thing. A plurality think it’s indifferent but, of those who have a view, more think it is a bad thing than a good thing (only one in three believe this rise in populism to be a bad thing). This pro-populism ethic is strongly connected to a more ordered outlook.

This blend is almost equally composed of indicators of both economic and cultural insecurity. It connects very strongly to the key debates around immigration, diversity, and race, the role of the state, globalization, and trade and class membership. It is also strongly connected to partisanship and political ideology. Before exploring those connections, we will look at the geography of open-ordered outlook in Canada. Which places are most open and welcoming? Which are most wary of openness?

These two preliminary charts show two important things: 1) racial tolerance is clearly a factor in Canada, but it may well be declining, which is part of a hypothesis that Canada is experiencing an open pivot in response to these forces; and 2) our ethnic fluency isn’t particularly strong and, more notably, the more you exaggerate the influence of ethnic groups, the greater your tendency to racial intolerance.

I. The geography of populism: Mapping the open-ordered outlook in Canada

The relative emphasis on the ordered or open side of our index varies depending on where one lives. We begin by looking at regional patterns, then consider settlement size, the actual city one lives in and finally how the index varies within the largest megapolitan cities in Canada.

While the effects are modest there are statistically significant differences in how open or ordered different regions of Canada are. Note that our aggregate sample of 12,604 respondents provides an unusual capacity to look at these geographic factors.

Regional differences

The East and West coasts are the most open regions in Canada. In the Atlantic, this is despite the fact that their populations are older, less diverse and somewhat lower educated. All those factors are linked to less open outlooks so there are obviously other forces at work; perhaps the cultural outlook? This greater openness could be a product of the demographic and economic exigencies of Canada’s oldest region. Simply put, the Atlantic region is in more urgent need of bolstering its population – particularly its older population – than any other region in Canada. British Columbians are equally open. Perhaps it is the oceanic influence with its conduit to the broader world.

All of the other regions of Canada – except Alberta and Saskatchewan – are also quite open (as is Canada as a whole). Alberta and Saskatchewan are significant less open than the other regions but those differences are modest. Perhaps the continued sense that their economies are suffering from lower oil and gas prices or perhaps their values and ideological orientations are more congenial with the ordered outlook.

Impact of settlement size

Whether one lives in the densest or most sparsely populated communities has a modest but clear effect on open-ordered.

We see a clear monotonic decline in openness as we move from the largest metropolitan centres (Toronto, Montreal, Vancouver) to those living in rural Canada. This would favour the notion that big cities, which are more diverse and connected to global economies, exert a cosmopolitan influence on their denizens.

Ranking cities

The most open city in Canada is St. John’s, with Victoria, Kelowna, Ottawa, Winnipeg, and Vancouver not far behind. Looking at the least open, we can group Oshawa, Calgary, Edmonton, Saskatoon, Regina, London, Hamilton, Barrie, and Windsor as the most ordered cities in Canada. These cities tend to reflect places where manufacturing jobs have disappeared and many workers who once enjoyed middle class status by virtue of a strong back and union card now find themselves in working class status. This would not apply to Calgary or Edmonton, but they too have seen slippage in their economic outlook due to falling carbon energy markets. It is quite clear that the process of stagnating or falling backward has a corrosive impact on one’s openness.

Exploring differences within cities

While the variations by settlement type are interesting, our analytic tools allow us to look at differences within the largest cities in Canada. The results are very revealing and suggest that the differences within largest cities are bigger than the differences across cities.

Let’s begin with Toronto, by far the largest city in Canada and the 4th largest in North America. Toronto may well be the most diverse city in North America and leads to a quite open posture. As one can clearly see from the mapping of the index within Toronto there are vivid differences in this overall outlook depending on where one lives in the Toronto census metropolitan area.

There appears to be a correlation between ordered outlook and ethnic homogeneity. Areas with a high proportion of visible minorities or a high White population tend to be more closed. In contrast, areas that are more mixed tend to be more open.

For instance, in Montreal’s South Shore and North Shore regions, residents tend to be more closed. The Island of Montreal and Laval, meanwhile, tend to be more open. The unexpectedly ordered outlook of residents in the Central Montreal and Westmount regions came as something of a surprise; however, delving into the data reveals that the area’s score is being inflated by the Le Sud Ouest and Verdun boroughs, which tend to be more ethnically homogenous than the rest of the city.

In the Greater Toronto Area, openness seems to rise with proximity to the city centre. Within the City of Toronto itself, the more closed areas tend to be found in ‘Ford-nation’ (i.e., areas where Rob and Doug Ford did well in the last two mayoral elections, such as Etobicoke, North York, York, and Scarborough). These areas are more working class, but also have a high visible minority population. The wealthier areas – particularly Eastern North York and Midtown – tend to be more open, as do the East End and East York regions, which are known to be among the more progressive parts of the city. Most of the area outside of the city proper is more closed than open. Brampton and Mississauga are quite closed, and both have a rather high visible minority population. It was, however, Toronto’s exurban fringes (which include areas like Caledon, Bradford, Orangeville, and Georgina) that are the most closed, and tend to be the least diverse areas of the Greater Toronto Area.

In Vancouver, the entire metro area is more open than closed with the City of Vancouver itself ranking as the most open part of the region. Meanwhile, the city’s inner suburbs (such as Burnaby, Surrey and Richmond), which have a lot of visible minorities were the most closed. The wealthier North Shore region and the outer eastern suburbs (which are also whiter than the inner suburbs) are more open than the inner suburbs.

So where you live matters. The key factors appear to be the ethnic diversity of where you live and the economic outlook in your community.

II. Links to economic outlook and social class

In the 20th century, the concept of social class became somewhat abandoned. The favoured approach was to think of how order was maintained across a continuum of socioeconomic status (SES). As inequality has burgeoned, the more conflict rooted concept of social class has become more relevant to today’s societal realities.

There are hundreds of operational definitions of class, most driven by income characteristics. We use a much simpler approach. We ask respondents to locate themselves in either the upper, middle, working, or poor classes. This links strongly to income but even more strongly to health and overall well being. More importantly the evidence is that the proportion of Canadians seeing themselves as members of the middle class has declined sharply in this century in both Canada and the United States. This is not a product of upward movement. This is large a pattern of either stagnation or decline. The nearly 70/20 per cent gap between middle and working class that we saw at the outset of the 21st century has dramatically transformed such that the middle class is now only slightly larger than the swelling working class. This has left a lot of citizens angry and fearful and is the principal fuel propelling the rise of populism and resurgent authoritarianism.

Education has powerful influence on whether one leans open or ordered. Although there is a progression from least to most educated, the big divide is university educated versus college and high school. Interestingly, this divide disappeared in the final stages of the last federal election but it has reappeared with a vengeance. Among the ‘not so educated’ to paraphrase John Kenneth Galbraith’s notion of the war of the not so rich against the rich, there is broad disdain of the professional class and a view that the elite prescription that they have authored (trade, globalization immigration etc.) has worked fine for them but left others behind. Confidence in the recipe for success offered by the educated establishment has declined in lockstep with the general erosion of the economic position of those lacking such human capital.

The effects of education are very large but the effects of class membership are even stronger. Basically, one’s enthusiasm for an open outlook declines dramatically as we move from the privileged to impoverished classes. Upper and middle-class favour openness, while working class and poor are much more receptive to an ordered view. The ‘proof’ of the wrong headedness of the elite prescription is their real personal stagnation and decline. This is the principal engine underpinning populism and a more ordered outlook.

Another selected indicator is the relationship with retrospective experience with the economy. This chart shows that the ordered are over three times more likely to have fallen behind over the past five years. Open respondents are more than twice as likely to have moved forward as non-open respondents.

As one can see in the exhibits in Annex II, the open and ordered citizens of Canada occupy two different worlds. The open think national and federal directions are grand; the ordered think we are overwhelmingly moving in the wrong direction.

III. Political and public policy expressions

Whether one leans open or ordered is strongly correlated with one’s position on many of the key political and policy debates of the day. In this final section, we will connect the open-ordered axis to the world of politics and public policy. There are vivid differences in how these different orientations express themselves in those worlds.

Let’s begin by breaking this down by partisanship.

The differences in open-ordered outlook across party preference are profound and undoubtedly reflect political ideologies as well. Liberal, NDP, and Green supporters are quite open with the Liberals being the most so. Liberal supporters are nearly three times more likely to be open than ordered. On the other end of political spectrum are the Conservative and Bloc supporters. The plurality of Conservative supporters are ordered and close to the same splits are evident among Bloc supporters. The Bloc supporters appear to have traded their traditional progressive posture for this more ordered outlook and now find themselves as bedfellows with Conservative supporters. The clear ordered lean among Conservative supporters seems disconnected from the rhetorical style of the new leaders of the federal and Ontario Conservatives.

Not only are the ordered more likely to be very upset with the direction of the country and federal government, they also are much less convinced that government is the solution. Support for larger government is much higher in the open cadre whereas other citizens strongly favour smaller government.

What about some of the big policy questions of the day? We know that ordered are far less receptive to immigration, but it also clear that their enthusiasm for NAFTA is much weaker. While overall support for NAFTA is positive throughout Canada, there is far more opposition and much less ‘strong’ agreement outside of the open category.

The reason that NAFTA is once again centre stage is because of the election of President Donald Trump. While most Canadians strongly disapprove of the new President, it is notable that, among those leaning ordered, his approval rating is three times higher.

One final illustration of the differences across open-ordered is response to Bill C-62 in Quebec. With a pretty balanced forced choice, we see Canadians opting for a positive view of this bill banning face coverings in some public settings. This support is much stronger among ordered respondents who almost unanimously see this as a sensible thing to do. Support is still positive in open Canada, but by a much narrower margin.

In conclusion, we note that there is a clear expression of the ordered version of populism evident in the United States and elsewhere in Canada. It tends to be much more pessimistic, skeptical of the role of the state and professionals, less supportive of trade liberalization and more wary of immigration and other ethnicities. The ordered outlook is most strongly linked to social class and economic outlook and also strongly linked to political outlook. There are also important regional, urban, and neighbourhood differences which may well be connected to economic outlook and ethnic homogeneity of the settlement one lives in.

The more open outlook is more common and the country seems to be tilting in that direction, perhaps an open pivot in response to the Donald Trump presidency. It is unclear where this will evolve, but it does seem to capture the essence of the new divide separating Canada into the open and ordered camps.

Annex I: The Index

Respondents are assigned points based on their responses to various social and economic questions. A respondent’s score is simply the total number of points they receive. A respondent may receive up to six points. Respondents who skip any of these questions are dropped from the analysis. The actual scoring system and index ingredients are presented below.

Would you describe you and your household as poor, working class, middle class, or upper class? / How would you have described you and your household ten years ago?

  • If the combination of the respondent’s answers suggests they have moved ahead in the past 10 years, they receive no points.
  • If the combination of the respondent’s answers suggests they have stayed the same in the past 10 years, they receive 1 point.
  • If the combination of the respondent’s answers suggests they have fallen behind in the past 10 years, they receive 2 points.

Thinking ahead over the [ROTATE – HALF-SAMPLE EACH: NEXT YEAR/NEXT FIVE YEARS] or so, do you think your personal financial situation will be better or worse than it is today?

  • If the respondent says they will be better off (5, 6, or 7 on a 7-point scale), they receive no points.
  • If the respondent says they will be about the same (4 on a 7-point scale), they receive 1 point.
  • If the respondent says they will be worse (1, 2, or 3 on a 7-point scale), they receive 1 point.

If you were to guess, roughly what proportion of Canadians would you say are [ROTATE – QUARTER-SAMPLE EACH: members of a visible minority group / indigenous / Muslim / Jewish]?

  • If the respondent provides a correct estimate, they receive no points.
  • If the respondent provides an estimate that is ‘somewhat’ off (either higher or lower), they receive 1 point.
  • If the respondent provides an estimate that is severely exaggerated, they receive 2 points.

Of those who live in Canada, would you say there are too few, too many, or the right amount who are members of visible minorities?

  • If the respondent answers ‘too few’, they receive no points.
  • If the respondent answers ‘about right’, they receive 1 point.
  • If the respondent answers ‘too many’, they receive 2 points.

Many people talk about the rise of populism in the United States and Europe which, among other things, includes growing opposition to trade and globalization, growing support for more restrictive immigration policies, and growing distrust of those considered to be elite. Overall, do you think this rise of populism is a good thing or a bad thing?

  • If the respondent answers ‘bad thing’, they receive no points.
  • If the respondent answers ‘neither’, they receive 2 points.
  • If the respondent answers ‘good thing’, they receive 4 points.

Annex II: Index by demographics

Annex III: Attitudinal differences between open and ordered


This report draws on data from two separate surveys. Both surveys were conducted using High Definition Interactive Voice Response (HD-IVR™) technology, which allows respondents to enter their preferences by punching the keypad on their phone, rather than telling them to an operator. In an effort to reduce the coverage bias of landline only RDD, we created a dual landline/cell phone RDD sampling frame for this research. As a result, we are able to reach those with a landline and cell phone, as well as cell phone only households and landline only households.

The field dates for the first survey are June 1-19, 2017. In total, a random sample of 5,658 Canadian adults aged 18 and over responded to the survey. The margin of error associated with the total sample is +/- 1.3 percentage points, 19 times out of 20.

The field dates for the second survey are November 7-December 10, 2017. In total, a random sample of 7,882 Canadian adults aged 18 and over responded to the survey. The margin of error associated with the total sample is +/- 1.1 percentage points, 19 times out of 20.

Please note that the margin of error increases when the results are sub-divided (i.e., error margins for sub-groups such as region, sex, age, education). All the data have been statistically weighted by age, gender, and region to ensure the sample’s composition reflects that of the actual population of Canada according to Census data.

Please click here for a PDF copy of this article.

4 comments to At the Crossroads of Hope and Fear

  • Karen Spencer

    I think your paper is very interesting. However, I see 3 of the 9 questions used in creating the index are economic-oriented, and certainly for Alberta, Saskatchewan and NFLD, these will skew the index downward due to oil price impacts and the depth of the recession in these regions, with residents’ correspondingly glum view of the world. I also do not agree that 5 of your questions are about cultural concerns or uncertainties. 4 of the cultural questions are factual – only one is attitude, so only one question really shows bias and potential prejudices. Also, by having 3 economic questions out of 9, you clearly link recession areas with closed viewpoints, while I would like to leave those questions out entirely and see how the index holds up between regions. With 4 knowledge questions, you also link level of education with the index, similarly, which I don’t think you meant to do. Any chance I could get the data and do more manipulation with it? Thank you! Very interesting.

  • Frank Graves

    write me at [email protected] . You can have the data for non commerical application

  • Marg Ruttan

    Very interesting study. However, here is the problem. What the heck can and do we need to DO about this? We need to find out how to shift the mood of the country back to a more open one. Where can we find information on how to do that?

  • Mike Beaudry

    “The unexpectedly ordered outlook of residents in the Central Montreal and Westmount regions came as something of a surprise; however, delving into the data reveals that the area’s score is being inflated by the Le Sud Ouest and Verdun boroughs, which tend to be more ethnically homogenous than the rest of the city.”

    Having lived in several places in the Montreal region, I can assure you that Sud Ouest and Verdun are not “ethnically homogenous” at all.

    What is certain is the southern and western sections of that part of town are less economically well-off than the northern and eastern (Westmount, Outremont, Plateau) regions. The dividing line would be roughly rue Sherbrooke or rue Ste Catherine; south of that line tends to be poorer: St Henri, Verdun, Griffintown, lower Westmount, eastern Lasalle; though more recently those regions are slowly being transformed by gentrification due to proximity to downtown.

Leave a Reply