The Great Recession and the Rise of Populism

The success of populist rhetoric in Europe is worrisome as it presents a threat to national and European institutions, the rule of law, and other fundamental free-market democratic institutions, such as freedom of the press and the independence of the judiciary system.


Forum
homogeneous and antagonistic groups, the pure people versus the corrupt elite, and which argues that politics should be an expression of the volonté générale of the people." 2 Clearly, the views on populism have evolved. It is important to acknowledge the recent developments, which are relevant for today's populist debate. First, the traditional confl ict between the left and the right has been replaced by a new tension: the divide between the socially conservative and the socially progressive. 3 Second, the support for traditional social democratic parties is dramatically declining, while new parties that campaign on anti-establishment platforms are rapidly emerging. 4

Cultural and economic drivers of populism
Several explanations have been offered and numerous factors have been identifi ed as potential drivers of populism. For analytical purposes, one can bundle the various drivers into two broad explanations. The fi rst explanation revolves around deeply rooted cultural attitudes; the second focuses on economics. The cultural view asserts that populism is an outcome of the growing chasm in values between the parochial and the cosmopolitan. The economic standpoint, conversely, sees populism as the result of heightened economic concerns. Recent work has pinned these economic anxieties down to various secular trends, namely immigration, 5 rising income inequality, 6 the effects of foreign trade, in particular the China shock, 7 and automation. 8 Other works stress the impact of the fi nancial crises, 9 which in many countries (United Kingdom, European periphery) have been amplifi ed by austerity. 10 Although the cultural and economic views seemed inconsistent for some time, this is not necessarily the case, as there are evident interactions. 11 In this note, I briefl y revisit the driving factors of the recent wave of European populism. First, there are long-standing political frictions between the European 'core' and the European 'periphery'. These frictions may refl ect deeply rooted cultural and institutional traits, though, as Alesina et al. show, differences in trust, beliefs, and attitudes across European countries and broad regions are much less than the public debate suggests. 12 Second, European countries have been the recipients of sizable immigration waves as of late. Third, the recession hit the European periphery badly, leading to underinvestment, unemployment, and rising extreme poverty in some regions. Although the crisis has been short-lived, it appears to have had long-lasting implications. Fourth, like the rest of the world, Europe has experienced the effects of globalisation, skill-biased technical change and automation, as well as the general trend for non-unionised and not wellprotected jobs. Finally, Europeans, too, have been exposed to the rapid global diffusion of the Internet and social media.
Both economic and cultural aspects appear to be at play. Any comprehensive analysis should not dismiss the complexity of the interdependencies between the two. However, when one looks for the catalyst behind the recent surge in populism and its latest electoral success across a large number of European countries, the cultural argument falls short of explaining such a quick and concurrent reaction. Immigration might be an important part of the story, but its impact is hard to quanti- Forum fy as it transmits via both the cultural and the economic channel. Additionally, the recent large waves of immigration cannot explain the populist victories in rural areas, as migrants tend to concentrate in big cities, unless one acknowledges the huge discrepancies between reality and beliefs. 13 From a timing perspective, the current wave of European populism was most plausibly triggered by the global fi nancial crisis and the recent diffusion of social networks. 14 In the following, it is argued that economic factors appear to be important for the rise of populism, on their own and/or by amplifying pre-existing cultural traits and slow-changing social dynamics. 15

The European trust crisis and the rise of populism
In Algan et al., we look at the implications of the fi nancial crisis for voting, political attitudes and trust in Europe. 16 Our Elias Papaioannou on the recent rise of populism in Europe. 16 Y. A l g a n , S. G u r i e v, E. P a p a i o a n n o u , E. P a s s a r i , op. cit. before-and-after analysis is carried out at the regional level. We exploit time variation in unemployment, actual voting and beliefs-norms attitudes across 220 regions (NUTS2) in 26 European countries between 2000 and 2017.
Our results show that the economic crisis, measured by the rise in unemployment, unfolded alongside an institutional crisis of trust and a rise in populist and Eurosceptic voting. The regional level of the analysis allows us to compare the regions that were more severely hit by the crisis to the less affected ones, controlling for pan-European and country group-specifi c time trends. We fi nd that electoral successes for populists follow increases in unemployment; in particular, a one percentage point increase in unemployment is followed by a one percentage point increase in the vote for populist parties. We show that what matters for electoral outcomes is predominantly the change in unemployment, rather than its level (see Figure 1).
Our estimates linking unemployment to populist voting do not automatically imply a causal relationship. Although we control for time-invariant features driving populist voting and unemployment, one cannot rule out the possibility that omitted, time-varying regional factors could drive this correlation. Other potential problems include reverse causation and nonnegligible errors in variables; for example, unemployment statistics may fail to precisely measure part-time employment.

Forum
To progress on causality, we develop an instrumental variables approach that links the pre-crisis industrial composition with the change in post-crisis populist voting. The regional share of construction in total value added is used as an (excludable) instrument for unemployment given the critical role of the construction sector both before and during the fi nancial crisis.
The instrumental variable estimates show that the effects of the rise in unemployment on populist voting, driven by the pre-crisis share of construction, are considerable and if anything somewhat larger than the correlational analysis: A one percentage point increase in the unemployment rate is associated with a two to four percentage point increase in populist support.
The reduced form link between the share of construction and voting patterns and the strong relationship between construction and unemployment do not necessarily imply a causal relationship between construction, unemployment and populist voting. The necessary condition for causality is that construction does not affect voting directly or via other channels aside from unemployment. Although it is not possible to formally test for the exclusion restriction, alternative mechanisms that could be at play are explored.
These alternative explanations relate to corruption, education and immigration. In all three cases, it is shown that the documented relationship between construction share, unemployment and populist voting does not seem to refl ect these timevarying regional features. Pre-crisis populist voting is also not linked to the pre-crisis share of construction.

A European trust crisis?
We try to shed light on the channels through which economic insecurity affects voting by looking at individual-level data on Europeans' beliefs and attitudes from the European Social Survey covering the period between 2000 and 2014. 17 An increase in regional unemployment leads to a signifi cant decline in trust towards national and European political establishments. Europeans are no longer happy with their parliaments and the European Union. Additionally, there appears to be a loss of trust towards courts and the countries' judiciary system. At the same time, the association between unemployment and interpersonal trust is weak. The rise in unemployment does also not seem to correlate with distrust towards the police or the United Nations, again suggesting that Europeans fault the national and European political institutions for the rise in unemployment. Our results further 17 The results remain unaltered after the inclusion of rounds 8 and 9 of the European Social Survey and the new release of unemployment statistics from Eurostat.
indicate that European respondents increasingly feel that no political party is close to them, suggesting a political isolation that sets the stage for the populist parties.
The instrumental variable estimates are similar; the component of the rise in unemployment due to the pre-crisis share of construction is a signifi cant correlate of distrust towards European and national institutions.

Attitudes towards immigration
The information from the European Social Survey has been exploited in order to assess whether the rise in unemployment has also affected Europeans' attitudes towards immigration. This exercise is of particular interest because populists often portray immigrants as a cultural and economic threat to the 'people' and run campaigns that support policies designed to restrict immigration, combat cultural dilution and avoid adverse economic outcomes (like in the case of the Lega Nord in Italy, the Front National in France and the UKIP in the UK).
Differences in unemployment during the crisis correlate with views that immigration harms the country's economic life. However, we uncover no evidence that the rise in unemployment relates to views on the immigrants' role in a country's cultural life. Our fi ndings seem to point to the importance of economic, rather than cultural, explanations; it is the former that appear to nurture the support for anti-immigration parties.

Heterogeneity
Despite some minor differences across subsamples, these patterns are present for both men and women, and for both younger and older cohorts. We obtain stronger and more accurate estimates for older cohorts, consistent with the works of Eichengreen and Norris and Inglehart. 18 The link between unemployment and distrust toward political institutions is also stronger for non-college graduates. 19

Economic consequences of the populist wave in Europe
Populists tend to support erratic policies. In their effort to please their electoral base, they often engage in myopic ac-Forum trust in political institutions. Economic factors seem to be an important part of the populism story, and a key determinant of its recent pan-European electoral success. This presents an exceptional opportunity to address populism through the implementation of countercyclical macroeconomic and fi scal policies that will fi ght unemployment and foster growth.
Moreover, throughout the crisis, national and European policymakers had to implement unpopular structural reforms in the labour, capital and product markets and enforced austerity policies in order to stimulate the economy. This resulted in a further loss of trust, especially in countries with inadequate systems of social security. It is, therefore, imperative to preserve the trust of citizens facing economic insecurity by creating safety nets, especially for the unemployed workers.
There is a need for policies that promote social and economic mobility and reduce inequalities by offering equal opportunities for education and providing healthcare access for all. Both national governments and the European Union should ultimately prioritise research, innovation and public infrastructures.
Additionally, policy reforms need to be backed by electoral support and enjoy people's trust. Europeans need to be reassured that the gains from the painful reforms will be widely distributed rather than extracted by a privileged few. National governments and the European Union need to cater to the unskilled workers and those who are 'left behind' by investing and promoting programmes of life-long training and support, thus limiting the scope of the supply side of populism.
tions and undertake radical reforms, which usually prove to be damaging for national economies.
The rise of populism along with the documented 'European trust crisis,' pose a serious threat to the European project. Urgent policy issues call for international coordination and deeper integration. Populists, on the contrary, push to abolish/weaken international organisations and oppose the European Union membership outright on the basis that it undermines national sovereignty, encourages high levels of migration and is responsible for austerity as well as the members' economic misfortunes.
On an alarming note, it appears that the rise in unemployment also directly refl ects on attitudes towards the European Union project, albeit in a heterogeneous manner. In the European South, despite the growing institutional distrust, people still hope for deeper integration. In contrast, in the crisis-hit regions of the European core, respondents believe that the European project has gone too far. This shift in attitudes provides a fertile ground for anti-establishment politicians who, in turn, fuel the anti-European sentiment creating a vicious circle.

Policy implications
If populism is entrenched in deeply rooted cultural differences, addressing it is a non-trivial endeavor. Our results, however, show that the rise in unemployment results in increased support for anti-establishment parties via the loss of