Answers are grounded in this paper and cite their sources.
The main thesis of this paper is that people in areas of (expected) population decline vote more populist to express their discontent about the current and future state of their place of residence. In many ways a “populist voting mark-up” could be expected, as declining areas often are associated with being forgotten, fomenting societal discontent and mistrust in established political parties ultimately expected to lead to more populist votes. Using the outcomes of the Dutch national elections in 2012 and 2017, we link shares of populist votes for the PVV (Party for Freedom) and SP (Socialist Party) to indicators of population decline, as well as other demographic (“compositional effects”), local and regional characteristics (“contextual effects”) to appraise what causes higher rates of votes for populist parties in regions of decline. We do not find a “populist voting mark-up” for declining regions when controlling for contextual effects. However, we do find that both the compositional and the contextual circumstances in areas of population decline are in such a way that they provoke discontent expressed in voting. We also conclude that it is very important to distinguish between different parties when their party programs are as contrasting as in the case of the PVV and the SP. Their different focus on immigration (PVV) and jobs (SP) is clearly visible in the results. El objetivo de esta investigación es realizar un estudio empírico sobre los efectos de la segregación de trabajadores altamente cualificados (trabajadores con un título universitario) en la productividad laboral de las ciudades. Mediante el uso de datos de áreas metropolitanas de EE.UU. se pudo encontrar evidencia de un impacto positivo. Se estudiaron los posibles mecanismos y se encontró que los conglomerados de trabajadores altamente cualificados producen spillovers positivos en la productividad de los trabajadores altamente cualificados, que compensan con creces la pérdida de productividad de los trabajadores con escasas cualificaciones, debido a su exclusión. Esto último es especialmente válido en las ciudades especializadas en industrias en las que la complementariedad entre estos dos tipos de trabajadores en la función de producción de la ciudad es baja, como la tecnología o la ciencia. Se utilizaron datos de panel y un enfoque de variables instrumentales como estrategias de identificación. Los resultados son robustos en cuanto a las diferentes especificaciones. 人口減少 (が予想される)地域の人々は、居住地の現在および将来の状態について不満を表明するために、ポピュリストに投票する傾向にあるということが本稿の主旨である。衰退地域は看過されることに繋がることが多いために、社会的不満や既成政党に対する不信を煽ることは最終的にはポピュリスト票につながることが予想されるため、多くの点で「ポピュリスト票のマークアップ」が予想される。2012年と2017年のオランダ総選挙の結果を用いて、自由党 (Party for Freedom:PVV)と社会党 (Socialist Party:SP)のポピュリスト票の割合を、人口減少の指標だけでなく、他の人口統計学的特徴(「組成効果」)、地方と地域の特徴(「コンテキスト効果」)に結びつけ、衰退地域のポピュリスト票のより高い割合の原因を評価する。コンテキスト効果を調整した場合には、衰退地域の「ポピュリスト票のマークアップ」は見られない。しかし、人口減少地域における構造的状況およびコンテキスト的状況は、いずれも投票によって表明される不満を引き起こすものであることは確かである。また、政党プログラムがPVVとSPの場合のように対照的である場合、政党を区別することが非常に重要であると結論する。政党が重点を置く政策の違い (移民 (PVV)と雇用政策 (SP))は、結果に明確に現れている。 Within the current Regional Science literature, there is growing interest in changing spatial voting patterns, in particular the geography of recent surges in populism (e.g., Becker, Fetzer, & Novy, 2017; Dijkstra, Poelman, & Rodríguez-Pose, 2020; Gordon, 2018; Los, McCann, Springford, & Thissen, 2017; McCann, 2018, 2020; Rodríguez-Pose, 2018). These patterns are so interesting because they provide indirect information about the perception of welfare, happiness or discontent. The information is indirect because casting a vote is influenced by a myriad of factors, such as: (i) party programme; (ii) characteristics of a candidate; (iii) regional socio-economic circumstances; and (iv) strategic considerations related to other parties or the government in general. In addition, voting does not provide information about actual levels of welfare, but of the perception of welfare. It might as much be influenced by objective factors (e.g., income or (un)employment levels), as by more subjective factors (e.g., relative income compared to other regions such as to the capital city), resulting in a feeling of inequality and “being left behind” (Dijkstra, Poelman, & Rodríguez-Pose, 2020; Garretsen, Stoker, Soudis, Martin, & Rentfrow, 2018; Gordon, 2018; Rodríguez-Pose, 2018). In this paper, we are particularly interested in regions that face population decline. Population decline can have a negative effect on the stability of the economy and society, especially when at the regional level a feeling of inequality arises and people in declining areas feel left behind by the national government (Ubarevičienė, Van Ham, & Burneika, 2016). This feeling can be fed by higher unemployment levels, a decreasing supply of facilities and properties that are difficult to sell. In some cases, people are “locked in” due to the inability to sell their property and move somewhere else. The inability of local governments to address livability and the (perceived) disinterest of the national government often results in lower levels of government trust and a lack of confidence in politicians (Obbink, 2016). This “societal discontent” has been linked to rising Euroscepticism, with studies on different but related focuses, namely on anti-EU voting or on determinants shaping citizens’ attitudes towards the EU (cf. Dabrowski, Stead, & Mashhoodi, 2019; Dijkstra, Poelman, & Rodríguez-Pose, 2020). Here, effects of demographic developments such as ageing population and low population density is intricate and results have been mixed, meriting further attention. Overall, the Dutch population is growing. Nevertheless, one in five municipalities faces population decline (Van Weezel, 2017). Although the decline should be labeled as “mild” compared to most EU regions, according to the Dutch Statistics, in these declining areas only around 26% of the inhabitants trust the national politics, whereas this is 39% in areas of growth. These lower levels of trust can lead to more voters for populist parties as a way of protesting against the current system and its politicians. According to Vossen (2012, p. 32), inhabitants of shrinking regions “feel that the current establishment fails to serve the interest of the people.” Since population decline can encompass both sorting of people and their socio-economic context, it is therefore apposite to raise the question: to what extent do regions in decline face higher levels of populist voting when controlling for “compositional effects” (e.g., demographic factors) and “contextual effects” (e.g., regional economic conditions)? Like more European countries, the Netherlands has also faced a rise in parties labelled as populist. These the PVV (Party for the SP (Socialist and more the for their of a and these parties party In this we have to be in votes for these parties in a can also be a to in other with different political of the we distinguish between the party (PVV) and the most populist party The outcomes the 2012 and national elections to appraise the lack of population decline, with a of other The between populist voting in national elections and population decline has not been the of populism in the different of political parties in this it is of important to determinants have different on the voting for these and the The of this paper is as we the related literature, also in to the of declining regions in the spatial voting patterns and of the populist we the factors, and in by the results and in and we provide and for further do vote in the voting for by the of for a & we demographic factors important by both the to vote and the to voting. with a higher level are more to vote and have a higher & the of a to the political system the to people vote more because they have more at in the elections & a are more to vote when they are in their the of people vote the have more information about the political system and therefore are more to could be that this of has the most at in the of the elections & is to not have a effect when controlling for income and & due to in some might vote to have effect as In addition, factors The to for is often (i) a of by local and and (ii) with and in the local of local and are often associated with a more population of and and towards voting. to also of a in of or characteristics is to lead to higher due to the on and people are more a for by they are more to the of voting population density a Since a higher population density often to a of and are lower in density important is the trust in the local and national political that the system is it lower their to in are to to vote and in & & and find that in particular European are to vote when populist parties with negative about and more However, this is not as the population could also have a to vote as and find for European is by as to against both the established of and the and of politicians to the of the the that in their interest of to the people The in between people and also the with the populist the the established political that fails to serve the of the people can be in & do people vote for populist vote for because they with the populist parties & The does not what the people important and their This feeling of is one important for voting for a populist is a vote & The is the and in populist to the between the people and the state by a that what the people This is a people can to and the interest of the In the the parties that are often labelled as populist are the SP and the PVV and 2012 and 2017, the populist parties SP and PVV of has a focus on (un)employment of the and of and The party has a local and regional The party to have a in Since could not the party for has been around one with local In the PVV is anti-EU and they are and more for the Population decline is It can be by or causes have different The Dutch government a between and areas on the 2018). In declining regions, the population is expected to by at the of be decreasing by around In addition, there are areas the population is declining but a decline is in the In these it is expected that the of inhabitants or the of by at also in the It is that there is much in population developments the Dutch decline is in more with the of areas with to a (e.g., in the The effect of decline by the of to the of a or decreasing rates be very different & 2018). decline both the of areas and the socio-economic of due to that of that have the to it is particularly the and often lower people that In many declining regions, the of the population of people and more & Population decline also has a effect on the & & & It results in a lower for the often to a in and in The and have effects for property can and the & Van & also much as in with recent that are for in much lower in municipalities more at the especially in the compared to to the important contextual is related to the of the and the of the it as that the the of the people In declining regions, this can be by a decline in and due to lower of In the are the of and when this so does the (Van de it can be a to a level of facilities and with a decreasing & Van 2018). facilities to be it often that the are difficult to be by of people and that the decline in most the of & The decline in and might also local to move or their resulting in higher unemployment These might not only have a but also indirect effect when they people and in areas that are labelled as or This might the a a negative of and economic However, population decline is not only a and might in the municipalities do have a of with a and levels of & a the of the in shrinking municipalities is higher in the other of the Netherlands & and and also find that inhabitants of shrinking municipalities are with their important of livability is with and a feeling of being of the livability objective of economic and subjective of not a growing of to at the of In the 2018, the national government a of in the not only but also other indicators trust and In addition, the of and the in a also happiness and a regional The that in the and the of the levels of are very the also low on of welfare, that does not for the is one of the regions in many with unemployment levels it is also to that there is not a (e.g., and This it to with more economic subjective on to Van & in the populist voting is often as a of against the against people that are against growing in voting is also associated with the feeling of “being left behind” as a as a by the national this can further negative effects that population decline can have on the stability of the economy and (Ubarevičienė, Van Ham, & Burneika, 2016). Population decline, in can encompass a myriad of factors such as to population decline results in more votes for populist we a that the between of populist votes in a by (expected) population decline, as well as compositional and contextual characteristics of the Population decline is on the of the government and This is often in and the and as such can the feeling of discontent of the In addition, we also the population of the in the we on the of votes the for the 2012 and of the Dutch Using the of these we linked to the the on the PVV and the SP in both around of the for the SP in 2017, the PVV to of votes. they of the votes in rising to in these are only national when at the for in 2017, the is for the PVV and for the SP. population decline, we the national and municipalities 2018). In the of the population in these areas to around and areas are with of population decline population decline of at decline of in areas are not population decline population decline of at decline of in but are towards population decline It that these to regions at the some between this (expected) decline and population developments in can be also areas are expected to population decline and more often of people (Ubarevičienė, Van Ham, & Burneika, 2016). In of demographic to (expected) population we population and as to the population decline we other compositional namely income and unemployment in is not as it is with and level is not factors that we are shares of and and of or and to The of are as a for with a as in the to we a with to of to a to a to and to The of that the is to a in one and of the The of the between and The as a In addition, at the level can local and livability more We also it as important due to its to These both for the elections of and as well as for the national elections of 2012 and Since they are and we the for the national elections due to the effects of both but not when the in for income or we more people and lower levels of populist voting. these are expected to a of we the the five to the national on growing to and on we also a that livability by factors of at the regional level We also the of the to The to for spatial factors related to the between and the Dutch political We this voting outcomes many in The vote for the more (e.g., in to vote more for the SP. the results for 2012 and 2017, we focus on the more recent national This also more of the effects of the of as well as more recent surges in especially there are these be for 2012 are in we at the results in and both the shrinking and at the to be in higher levels of PVV and SP votes. when not controlling for compositional or contextual people in areas of (expected) population decline do vote more populist. However, due to the intricate of population decline, it is to this regional decline controlling for the related compositional and contextual effects. for the of the is low in the especially for the of votes for the This and and a is that the shrinking and population decline indicators are The population is as We find that effects are Nevertheless, there is a between with higher shares of people between and and populist votes for both the the this also for with higher shares of population the 2012 there is a negative between areas with higher shares of people and populist voting. is the of the ageing for a negative effect on SP votes for the 2012 national to income and unemployment are in the case of the for the the of income is negative and but only and for the SP to on the to a negative the further the lower the of populist voting. the expected to have a negative with populist voting does not as a compared to other local factors, with only the expected being and for of votes for SP. at the different we do PVV voters are most often in and the SP voters also often in This is The national is negative and in the of votes for also for PVV in local is It that with a of has a negative on the of votes for with a higher of there is a effect for both PVV and SP. It is also important to appraise more regional contextual effects to areas in the This is very as it and as such the economic and in the place of or for a developments do not to be a in regional results. It is and for the PVV in 2017, but negative and in Regional economic also has a negative in for the SP. The on the other more it is and negative in with of votes for PVV in both and and for the SP votes. This that a vote for PVV is more a of this is not the case for SP. controlling for and welfare, we find that to the political of The has a effect on populist voting for a is further The to vote more for the SP on results for it has a negative effect on shares of votes for The main thesis of this paper is that people in areas of (expected) population decline vote more populist to express their discontent about the current and future state of their place of residence. In many ways a “populist voting mark-up” could be expected, as declining areas often are associated with being forgotten, fomenting societal discontent and mistrust in established political parties ultimately expected to lead to more populist votes. this (expected) population decline to areas this also that and municipalities in of and & & & these declining areas often face resulting in and to the feeling of being left behind by the national as well as effects such as on the and of local & Van 2018; & & Although we do find higher levels of populist voting in for compositional or contextual this as as we do more socio-economic This can that (expected) population decline so does not for populist voting factors, compositional or more contextual The for population also that population decline, in is or This could have to do with the low levels of population decline we face in the related has also results. In a of in the Dijkstra, Poelman, and for find that economic decline votes for anti-EU parties much more other factors such as and population decline. Although not on actual Dabrowski, Stead, and find that the effects of shrinking regional not a in shaping EU could be due to the of In they find that declining regional economic citizens’ on the However, of the related local and regional factors to be at compositional we find effects of the in the Dijkstra, Poelman, and find the for areas with a of population to be at it that ageing is a of the populist vote compared to other factors 2020). In the should be with it 2012 results and also that areas with higher shares of voting and to vote for populist and for people between and income and unemployment it is particularly important to the parties due to their very different expected, with a higher income population also of higher to vote for populist parties at both of the However, the lower income is not this could be that more more to might be more unemployment this lack of jobs can be due to unemployment the on the of SP votes can be expected (un)employment and are the this does not in votes for the the other local factors, the results are in with The negative effect of on the of populist voting for most of that feel and trust the national The results for have more also the intricate of this Although is not to be as when controlling for income and (cf. & in the of populist it can be more so especially in parties with It is important to that the of the can have effects. with are more to vote for PVV but it can also be the in areas with a higher of and that this as a for voting for It can also be that a of the population vote for parties such as the This in as with a of has a negative on the of votes for with a higher of there is a effect for both PVV and but much more for the is the party to the contextual we expected lower to to have a on populist can be in with the results of and In their on livability and they also find a negative between of and that for these the level of with does not a to their level of the other the for SP in the case of lower especially in areas SP has a local (e.g., in there be more to vote for SP. low unemployment levels this faced a discontent about the of due to related to for the of unemployment at the regional is that unemployment unemployment in a to the place of residence. with higher levels and inequality are often more they are in In with due to lower a rise in unemployment unemployment and the of & the local in this case to link more to societal discontent. regional economic for the the could be related to the negative often between economic and unemployment the is more of a for 2017, but could the focus of the party the is not on (un)employment of the and it is that is a of welfare. This is also in the Dutch of a of in the Netherlands 2018, other indicators the trust and In the is and negative in with of votes for PVV in both and and for the SP votes. This that a vote for PVV is more a of this is not the case for SP. interesting of the for the to the political The negative and for PVV that in and to The to vote for this populist the regional of SP are in other areas such as the (e.g., This is important spatial The main this to what extent do regions in decline face higher levels of populist voting when controlling for (e.g., demographic factors) and (e.g., regional economic The is and these regions do not face higher levels of populist voting when compositional and contextual there is such as a “populist voting mark-up” in these this could also be due to the low levels of population decline faced in the Netherlands compared to other such as However, when only between shrinking and regions, the shrinking regions as in of populist voting. This that both the compositional and contextual circumstances are in such a way that they provoke discontent expressed in voting. In regions of decline, income is often as are unemployment levels are The of these negative factors in regions of population decline can the higher shares of populist votes. In other we do not find a “populist voting people in these regions do express a feeling of discontent to be by regional and national in this is to not populist voting as voting. can be a myriad of as to people vote for populist It is to more what the This to a that we can the it is very important to distinguish between different This is particularly when their party are as contrasting as in the case of the PVV and SP. Their different focus on immigration (PVV) and jobs (SP) is clearly visible in the results. with more vote more for the areas with vote for the with more people on unemployment vote more often in of SP. interesting between the parties is that the regional a more important in the case of the in regions is and is as well as regions to the political of The people vote more in of the This that PVV votes more a feeling of discontent of people that feel they do not and that face lower levels of welfare, people voting for the SP particularly do so because of their about the of important for further both by and is to focus more on of discontent. in the voting is only as it might be on factors, do not to local socio-economic circumstances as we find in this and the most objective and subjective factors are to regional discontent and to The the the regional and the for very and that have to the The also to the especially Dijkstra, and de of the on the geography of discontent at the for are also to for the voting
No takes yet. Share an insight, caveat, or question.
Leeuwen et al. (2020) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: