Government conduct and democratic quality

One of the most basic questions to ask about a government is whether or not it works. If we look at a particular country, is the government capable of performing its essential functions well? To Adam Smith, the responsibilities of government could be boiled down to three elements: ‘peace, easy taxes, and a tolerable administration of justice’. All the rest, he argued, flows naturally.Footnote ⁴ Modern theories of government have continued to stress the importance of peace-making, fiscal and legal capacities. These, according to political economists Tim Besley and Torsten Persson, are the three pillars of prosperous states.Footnote ⁵ In recent years, an emerging literature has begun to consider the extent to which differences in these government capacities around the world are also capable of predicting cross-country differences in wellbeing. In this section, we will review the results of these endeavours. Specifically, we will consider the role of government conduct and democratic quality.

Over the last two decades, the World Bank has evaluated governments around the world in terms of six fundamental characteristics: (1) ability to enforce rule of law, (2) effectiveness of service delivery, (3), regulatory quality, (4) control of corruption, (5) political stability and absence of violence and (6) voice and accountability.Footnote ⁶ Each dimension is itself composed of a broad set of individual indicators drawn from international databases on government performance. In the empirical literature, the first four are often aggregated to provide an overall assessment of government conduct while the latter two are considered to be indicators of democratic quality.Footnote ⁷ It is important to note that these features are not necessarily fixed across time. States and government institutions are constantly in flux. Governments can certainly become more or less effective or democratic over time, and in some cases even collapse entirely. Nevertheless, the fundamental quality of institutions tends to remain relatively stable,Footnote ⁸ allowing for broadly reliable cross-country comparisons.

Let’s start with government conduct. When we look around the world, government conduct proves to be highly related to average country life satisfaction.Footnote ⁹ In Figure 16.1, this relationship is shown for a diverse set of 60 countries using data from the Gallup World Poll. With few exceptions, the trendline is clear: countries with higher performing governments have happier citizens. The overall correlation is roughly 0.7. Yet as we’ve seen many times already, correlation does not necessarily imply causation. Countries with better performing governments are also generally richer and able to afford the provision of essential goods and services. To truly isolate the effect of good governance on wellbeing, we need to control for these sorts of potentially confounding variables.

Figure 16.1 Correlation between government conduct and wellbeing

Note: Average life evaluations measured using responses to the Cantril ladder question. Government conduct measure drawn from the Worldwide Governance Indicators and evaluated in terms of (1) rule of law, (2) government effectiveness, (3), regulatory quality and (4) control of corruption. Data from 2005 to 2019. Linear trendline displayed and selected countries highlighted.

Source: Gallup World Poll and Worldwide Governance Indicators

In the literature, many studies have continued to find strong and significant relationships between government conduct and wellbeing across countries, even after controlling for other influences. In one of the first studies of its kind, the economist John Helliwell found a significant relationship between life satisfaction and good governance using data from 46 countries from 1990 to 1998.Footnote ¹⁰ Control variables were included at the individual-level for age, gender, marital status and employment status, among others, as well as societal-level variables including level of economic development, social capital and region were also added. More recent studies have expanded on this result by including additional years and countries and considered indicators of democratic quality and government conduct separately. Most of these generally continue to find a strong and significant relationship between government conduct and life satisfaction.Footnote ¹¹ In one study, even after controlling for income, trust, religiosity and democratic quality, a one standard deviation increase in government conduct predicted a 0.74-point increase in life satisfaction.Footnote ¹² Some researchers have also found that higher levels of government conduct predict lower levels of happiness inequality.Footnote ¹³

These results are typically obtained by comparing different countries to one another. Another way to consider this relationship is to look at the impact of changes in government conduct over time. Readers will recall that this type of longitudinal analysis has the benefit of controlling for time-invariant fixed effects. On an individual level, these could be genes or affective predispositions. On a societal level, variables such as norms, culture and geography may begin to play a role. With the emergence of new waves of international happiness and governance data over time, these types of studies have recently become more feasible. In one recent test of 157 countries using data from the Gallup World Poll and World Bank, a one standard deviation increase in government conduct predicted a subsequent increase in life satisfaction of 0.6 points on a scale from 0 to 10.Footnote ¹⁴ This effect held after controlling for changes in economic development, health outcomes, social support, freedom to make life choices, generosity and democratic quality, as well as country and year fixed effects. Perhaps even most importantly, these effects were observed over the short period of seven years, suggesting that improvements in government functions can have meaningful impacts on wellbeing within policy-relevant timespans.

What about democratic quality? It is perhaps first worth noting that any analysis of the link between democracy and wellbeing is bound to face challenges. In its simplest form, democracy means that citizens have rights and reasonable opportunities to influence legislation or elect representatives to do so on their behalf. Yet even this relatively simple definition defies precise measurement. Relying on any one indicator of democratic quality is likely to present an oversimplified account. At the same time, evaluating degrees of democracy between countries using multiple indicators can become quite a complicated endeavour. Nevertheless, over the last several decades, a number of international organisations and research teams have made impressive attempts to do just that. As mentioned earlier, governance indicators developed by the World Bank for voice and accountability, as well as political stability and lack of violence, are often grouped together to provide an overall assessment of democratic quality. Since 2006, the Economist Intelligence Unit has also published a yearly Democracy Index in which countries are evaluated and ranked as full democracies, flawed democracies, hybrid regimes or authoritarian regimes.Footnote ¹⁵ In the most comprehensive effort to date, the Varieties of Democracy project has rated the degree of democracy in almost every society in the world going back to the late eighteenth century.Footnote ¹⁶

Unfortunately, wellbeing data does not extend so far back.Footnote ¹⁷ Nevertheless, many researchers have linked more recent estimates of democratic governance to average reported levels of wellbeing between and within countries. As it turns out, the relationship is not quite as straightforward as one might expect. The general takeaway from this body of work is represented in Figure 16.2. Average life satisfaction (measured using the Cantril ladder and provided by the Gallup World Poll) is given on the y-axis, while democratic quality (measured in terms of political instability and violence, as well as voice and accountability) is given on the x-axis. Here again we see that both variables are strongly associated, with an overall correlation of 0.7.

Figure 16.2 Correlation between democratic quality and wellbeing

Note: Average life evaluations measured using average responses to the Cantril ladder question. Democratic quality measured in terms of (1) voice and accountability and (2) political stability. Data from 2005 to 2019. Linear trendlines displayed.

Source: Gallup World Poll and Worldwide Governance Indicators.

However, the association between democratic quality and wellbeing tends to be stronger in more developed countries. One study found that the effect of democratic quality is insignificant in countries with incomes below the global average, even after controlling for societal trust and religiosity.Footnote ¹⁸ Using an alternative set of democratic indicators, another study also found the quality of legal institutions to be more predictive of life satisfaction than democracy in low-income countries. The reverse was true in high-income countries.Footnote ¹⁹ In a longitudinal within-country analysis controlling for country fixed effects, John Helliwell and colleagues also found that increases in democratic quality significantly predicted increases in Cantril ladder scores in countries with high government conduct but not in countries with low government conduct.Footnote ²⁰

These results are suggestive of increasing marginal returns to democratic institutions at higher levels of socioeconomic development. For countries with less advanced economies, citizens may be more reliant on government for basic and essential services. In these regions, it may therefore be unsurprising that government conduct and service delivery would supersede the importance of democratic quality for wellbeing. Yet, past a certain threshold of development, democratic quality appears to play a more important role in determining national happiness.

However, it is also important to keep in mind the limitations of this body of work. For one, macro-level analyses generally have much greater data limitations. In previous chapters, when we considered the impact of employment status on wellbeing, for example, we could evaluate hundreds of thousands or even millions of individual respondents per year. When making cross-country comparisons, we often only have data on roughly 150 countries to compare in any given year. At the same time, institutional change generally happens very slowly, or all at once. Both of these limitations can make cross-sectional and longitudinal comparisons of national governments somewhat difficult to make.

At the same time, governments do not evolve in a vacuum. Even with the inclusion of fixed effects in longitudinal models, separating out the individual effect of good government from economic development or democratic quality is not only empirically difficult but also hard to meaningfully interpret. Given the interplay and even interdependence of all of these dynamic processes, knowing what to control for and what not to control in a regression is not always obvious. For example, given the substantial correlation between absence of violence and effective delivery of public services, it can be difficult to determine what percentage of citizens’ wellbeing is attributable to either one, independent of the other. Using linear regressions, the standard approach is to include both indicators as right-hand-side independent variables in a regression equation; yet because they are so interrelated, doing so can make it difficult to interpret what the resulting coefficients actually represent.Footnote ²¹

A separate strand of research has investigated the relationship between democratic processes and wellbeing by looking at institutional differences within country borders. This can be somewhat challenging in countries where similar democratic rules govern the entire society. But this is not always the case. In Switzerland, for example, democratic processes and procedures vary considerably between the 26 cantons (states) that make up the country. This fact, coupled with the availability of large-scale wellbeing data, has made Switzerland a uniquely suitable environment for micro-level research into the relationship between democracy and wellbeing.

In one widely cited study, the political economists Alois Stutzer and Bruno Frey created an index of political participation across Swiss cantons.Footnote ²² Cantons were rated on a scale from 1 to 6 depending on the degree of decision-making that relies on public referenda.Footnote ²³ In a cross-sectional regression controlling for a number of individual characteristics including age, gender, income, marital status, citizenship and health, among others, the authors found that residents of cantons with higher levels of public political participation were significantly more satisfied with their lives than counterparts in less democratic cantons. This benefit also accrued almost entirely to Swiss citizens and not foreigners, suggesting that opportunities for political participation are more likely to positively affect those who are able to directly take advantage of them. Overall, a one-point increase in the index of political participation possibilities predicted an increase in the number of people reporting high life satisfaction (10 out of 10) by 3.4 percentage points. Controlling for other factors, the difference in wellbeing between the most and least democratic cantons was 1.2 points on a 10-point scale.

Because these effects are estimated for residents of the same country, the potential for confounding variable bias is expected to be much smaller than in international comparison studies. Nevertheless, cross-sectional comparisons between cantons may still leave out important regional differences and time invariant fixed effects.

Another study attempted to overcome these concerns by considering the impact of political reforms that centralised political decision-making within different Swiss cantons.Footnote ²⁴ These reforms meant that democratic referenda at more local (municipality) levels lost some degree of influence as more decisions were passed up to political processes at the level of the canton itself. They were intended to increase government efficiency but also had the effect of reducing direct opportunities for political participation.

Importantly, reforms were also adopted by different cantons at different times, resulting in something of a natural experiment. By comparing changes in life satisfaction following reforms in some cantons to life satisfaction levels in cantons where reforms were not adopted, the causal effect of centralisation on wellbeing could be estimated. In the empirical literature, this is referred to as a difference-in-difference approach, since changes (or differences) in wellbeing arising from an intervention in one place are compared to changes (or differences) in wellbeing over time in another place where no intervention took place. By implication, for this empirical approach to get off the ground, there has to first be evidence of parallel trends prior to the event in question. In this case, before a given reform, life satisfaction levels in the reforming canton and comparison canton should have been following similar trajectories. If wellbeing levels were already decreasing in the former before the reforms, attempting to estimate their effects on wellbeing by comparing cantons would produce biased results.

Using this method, the study found a small but significant negative effect of centralisation reforms on wellbeing. Relative to those living in cantons in which reforms were not introduced, residents of cantons where democratic decision-making became more centralised experienced a decline in life satisfaction of 0.06 points on a 0 to 10-point scale.Footnote ²⁵ This effect was twice as severe in cantons with relatively low levels of direct democracy to begin with.

How are we to explain these results? To be clear, the above analyses suggest that (at least in more developed countries) increased opportunities for democratic involvement in decision-making have a positive effect on wellbeing, regardless of the actual decisions made. This implies that the benefits of democracy extend beyond simply producing better outcomes for citizens. Rather, there appears to be inherent value in being able to participate in the democratic process – what Bruno Frey and Alois Stutzer have called ‘procedural utility’.Footnote ²⁶ This has crucial implications for politics and government but also in domains outside of the public sector. Affording people more agency and voice in work and educational settings, for example, may also help to promote and support wellbeing. Put simply, it is not only the what, but also the how that matters.Footnote ²⁷

The size of government

We now turn from the nature of government to the size of government. The debate over ‘small’ or ‘big’ government has dominated political conversations for centuries. Despite the impressive array of opinions on this issue, the debate often centres around the same fundamental questions. At what size is the state best equipped to provide for better lives? Are larger and more activist governments better able to support societal welfare? Does the route to prosperity demand reducing the size and scale of government interventions? In this section, we will consider the answers to these questions emanating from empirical wellbeing science.

To begin, we should first define what we mean by the size of government. There are two components to consider. The first is the scale of welfare expenditures (as a share of GDP). These include state pensions, unemployment benefits, family allowances and the like – all of them cash transfers of various kinds. The second element is government expenditure on goods and services (like education, healthcare, law and order and defence – otherwise known as government consumption). We can now investigate how these measures affect wellbeing, both across countries and within countries over time. In doing so, we shall hold constant things like the educational level and health of the population, thereby somewhat underestimating the total effects of government activity.

Let’s begin with welfare expenditures. One early study used WVS data from 1981 to 2001.Footnote ²⁸ It found significant effects of welfare expenditures upon average wellbeing in a country. In addition, governments were also rated in terms of the ease by which citizens can access these expenditures.Footnote ²⁹ If we combine these two ratings, the gap in life satisfaction between the highest and lowest rated country was found to be 1.8 points out of 10.Footnote ³⁰ Over time, countries that expanded their welfare states also experienced subsequent increases in average life satisfaction.

Other more recent studies have extended these results. One again found positive wellbeing effects of welfare expenditures.Footnote ³¹ Specifically, a 1% point increase in welfare expenditures (as a share of GDP) raised average wellbeing by 0.03 points (out of 10). Moreover, this effect was found to hold for rich people as much as for those who are poorer and therefore more likely to benefit.Footnote ³²

However, one potential limitation of the early literature is the small number of observations it relies on. Given this, a handful of potential outliers – for example, the Nordic countries with high happiness levels and large welfare states – can skew the estimated relationships. One recent study dealt with this worry by expanding the sample to 107 countries using the Gallup World Poll, including many low income and lower-income countries. Even in this expanded sample, positive and significant relationships between welfare expenditures and life satisfaction remained apparent, after controlling for a host of potentially confounding variables.Footnote ³³ These effects were still significant after Nordic countries and other potential outliers were dropped from the sample. This would appear to suggest that welfare spending can benefit not only those in the most developed countries but also those in developing regions. More cautiously, the results of this literature indicate that countries with very limited welfare expenditures may struggle to promote social welfare.

So what about government expenditure on goods and services? Most studies considering the relationship between government consumption and national wellbeing find a positive relationship. One analysis showed that a 1percentage-point increase in the share of government consumption in GDP increases average wellbeing by 0.04 points (out of 10);Footnote ³⁴ and a 1percentage-point increase in the share of taxes in GDP increases average wellbeing by 0.03 points (out of 10). Another analysis using data from the Gallup World Poll found a positive and significant relationship between wellbeing and the level of progressive taxation.Footnote ³⁵

Like those in the previous section, some of these results may seem surprising. In almost all of the papers discussed, control variables were included for health, trust, economic stability, employment status and a host of other personal characteristics and societal conditions. One could argue that these are in fact some of the most important channels by which the state can impact wellbeing. In this section, we have left them mostly aside. Yet even after controlling for many of the most fundamental ways in which governments are presumed to positively impact citizens lives, most studies continue to find positive effects of government programs on wellbeing. At the very least, the evidence would seem to suggest that forgoing or severely limiting public expenditures seems unlikely to promote wellbeing.

Political orientation

A natural question arising from the discussion so far is which political program is most conducive to wellbeing. In a majority of countries around the world, including almost all high-income countries, governments are elected by some form of democratic political process in which political parties compete for the popular vote. The spectrum of political ideas tends to be represented as a continuum from left to right. The specifics vary between countries, although many of the most fundamental political viewpoints and policy goals of these two opposing sides are common around the world. While there are numerous ethical and philosophical differences between proponents of each position that cannot be easily adjudicated with data, some disagreements do lend themselves to empirical analysis. Here again, the tools of empirical wellbeing science can help shed light on which political project is more likely to succeed at making people happier. That question will be the focus of this section.

There are a few ways we can go about trying to answer this question. One way would be to simply look at differences in wellbeing between people who identify as left-wing or right-wing. This task is easy enough to accomplish. Many large-scale datasets used in happiness research including the European Social Survey and Gallup World Poll also contain information on political ideology. The findings of this body of research constitute an impressively large literature.Footnote ³⁶ The results are remarkably consistent: conservatives (right-wing) are generally happier than liberals (left-wing).Footnote ³⁷ A number of possible explanations have been put forth to explain these gaps. They have included conservatives’ higher levels of perceived personal agency, more transcendent moral beliefs, greater perceptions of the fairness of the world and more positive life-outlooks.Footnote ³⁸ Other studies have qualified these relationships, noting that conservatism only predicts higher levels of wellbeing in the most developed countriesFootnote ³⁹ or in countries with higher levels of perceived national threat.Footnote ⁴⁰ Relative to liberals, conservatives are also generally less likely to be unemployed, more satisfied with their finances and more likely to own homes.Footnote ⁴¹ Part of the reason they are happier may therefore be because prevailing societal institutions in many countries are relatively supportive of their interests.Footnote ⁴² Along similar lines, some research has shown that the wellbeing of partisans is also dependent on whichever party is in power.Footnote ⁴³ However, as interesting as these results are, they are somewhat unhelpful for our present purposes. Observing that proponents of one set of political ideas are happier than those of another tells us little about whether or not the political program they put forth is conducive to wellbeing on a societal scale.

Another route would be to look at whether or not citizens of countries with more left-leaning governments are happier than those in countries with right-leaning governments. In one of the first and most cited studies conducted along these lines, Benjamin Radcliff analysed the extent to which cross-country differences in socialist, liberal or conservative welfare regimes, as well as the degree of left-dominance in national parliaments, predict national wellbeing levels, using World Values Survey data.Footnote ⁴⁴ Control variables were included for GDP and overall unemployment. Both higher levels of socialism and left-dominance predicted higher levels of wellbeing. However, this result was potentially limited by familiar problems of scale (only 15 countries and one year of data was included), cross-country comparability, and confounding variable bias.

In a more recent test, a team of researchers looked at whether political differences between state governments in the United States predicted differences in wellbeing.Footnote ⁴⁵ The authors relied on representative life satisfaction data collected throughout the country from 1985 to 1998. The two key variables of political interest in this case were (a) the ideological leaning of the state government as determined by an independent rating system and (b) the percentage of the state legislature controlled by the Democratic party. A host of individual, state and regional control variables were included, including region and year fixed effects. The authors found that more left-leaning state governments predicted higher levels of life satisfaction. The effect of Democratic party control was positive but statistically insignificant.Footnote ⁴⁶ The magnitude of the former was also relatively modest. The impact of moving across the full range of the political spectrum from right to left was associated with an increase in life satisfaction equivalent to about half of the magnitude of the individual effect of unemployment.

The results of these studies suggest that left-leaning political programs are (slightly) more conducive to wellbeing than countervailing programs on the right. Nevertheless, even these results seem somewhat uninformative. Given the large diversity of policies put forth by left- and right-wing political programmes, it remains somewhat unclear why or which particular policies are conducive to happiness. The modest size of the coefficients in both studies may suggest that some left-wing policies have stronger effects than others.

Questions for discussion

1. (1) Is there a contradiction between maximising wellbeing and promoting procedural utility?

2. (2) There is some evidence to suggest that democratic quality is more strongly related to wellbeing in high-income countries. Does this imply that governments in low-income countries should not prioritise democracy?

3. (3) What are two primary limitations of making cross-country comparisons when conducting wellbeing research?

4. (4) How can these limitations be overcome using natural experiments?

Voter preferences

In this section, we move from political participation to voter preferences. Our discussion will centre around the following question: does wellbeing play a role in determining how people vote? Before tackling this question head-on, it is worth underscoring its importance. Throughout this textbook, we have highlighted the limitations of relying on economic indicators as proxies for wellbeing. Even the inventor of GDP, Simon Kuznets, himself once remarked: ‘The welfare of a nation can scarcely be inferred from a measurement of national income.’Footnote ¹⁹ Yet in democratic countries, sceptics could argue that the prime goal of politicians is not necessarily to make people happy but to get (re)elected. If, as a democratic strategist for Bill Clinton famously noted, the most important predictor of political success is ‘the economy, stupid’ after all, perhaps politicians could be justified, or at least excused for focusing primarily on economics. Several theories of voter behaviour in political science do suggest that voters support or oppose politicians in accordance with their rational economic self-interest, and these theories have been supported by a considerable degree of evidence.Footnote ²⁰ Overall, governing parties tend to have greater electoral success when the economy is doing well. As a result, while there may very well be strong moral reasons to care about wellbeing, if wellbeing does not affect voting, there may not be as strong political reasons to do so. Fortunately for our purposes, this is precisely the kind of question that lends itself to empirical analysis.

In the literature, this relationship is generally framed in terms of the vote share of the incumbent government. The intuition here is that successful governments would raise wellbeing and therefore be rewarded at the polls. On the other hand, if governments are performing badly and wellbeing is low, incumbents would be more likely to lose elections. These assumptions underlie a number of theoretical models of political behaviour, although relatively few consider the direct influence of wellbeing.Footnote ²¹

One of the largest studies of these dynamics to date looked at Eurobarometer data covering 139 elections in 15 European countries from 1973 to 2014.Footnote ²² In the first set of analyses, the study considered whether national life satisfaction data collected at the time of the Eurobarometer surveys explained outcomes in the next national election.Footnote ²³ The main results are presented in Figure 17.4. Overall, national happiness levels explained roughly 9% of the variance of the incumbent vote share in the European countries surveyed, while leading economic indicators including the GDP growth rate and unemployment rate explained 7% and 4%, respectively. Voters who were most satisfied with their lives (on a 4-point scale) were also found to be roughly 50% more likely to say they would vote for the governing party in the next election than those who were least satisfied.

Figure 17.4 Predictors of government vote share in Europe

Note: Each bar represents the correlation coefficient for four indicators, estimated in separate bivariate regressions with cabinet vote share as the dependent variable using Eurobarometer data. Country fixed effects were also included as controls. National happiness is the average country life satisfaction at the closest year prior to the election. Macroeconomic variables are drawn from the OECD and refer to the country-year of each election. The sample is 139 elections in 15 European countries, 1973–2014.

Source: Adapted from Ward (Reference Ward2020)

From a political perspective, these results alone provide strong reasons for elected officials to care about the wellbeing of their constituents. For our purposes though, it is worth pressing on. While these results are at least suggestive of causal dynamics – in that pre-existing happiness levels are used to predict future election outcomes – there may still be a number of confounding variables at play. In later specifications, the study also controlled for societal-level variables including incumbent party seat share and party system fractionalisation, as well as individual-level variables including past voting behaviour and personal finances. Even after accounting for all of these effects, wellbeing levels continued to emerge as meaningful and significant predictors of both incumbent vote shares at the national level and voter preferences at the individual level. In one analysis in particular, an increase in national wellbeing of one standard deviation predicted an increase in the incumbent vote share of 6 percentage points in the next election, while the same increase in the economic growth rate predicted an increase of 3 percentage points. Taken together, these results strongly suggest that wellbeing plays an important role in determining election outcomes.

An analysis in the United Kingdom provides additional evidence of this relationship.Footnote ²⁴ The authors in this case relied on 18 years of panel data from 1991 to 2008 collected by the British Household Panel Survey (BHPS), a period covering four national elections. The dataset also contained yearly information on respondents’ voting intentions ‘if the general election were held tomorrow’. Because the same respondents are interviewed every year, the authors are able to control for time-invariant individual fixed effects. Overall, being more than averagely satisfied with life predicted an increase in incumbent party support of 1.6%. This effect held even after controlling for personal financial situation – widely considered to be a fundamental driver of voting behaviour. For comparison, a 10% increase in family income predicted an increase in incumbent support of only 0.18%.

However, these results may be at least partially attributable to reverse causation. As discussed earlier, voters tend to be happier when the party they support is in power.Footnote ²⁵ The positive effect of wellbeing on voter preferences may therefore simply be a side-effect of political partisanship. In other words, even if happier voters are more likely to support the incumbent party, they may also be happier because the party they support is in power in the first place. In this case it wouldn’t necessarily be wellbeing that drives voter preferences but rather voter preferences that drive wellbeing.

To account for this potential bias, the UK study ran two additional empirical tests. In the first, the authors limited their sample to swing voters, defined as (a) those declaring that they do not favour one particular political party over the other or (b) those who consistently voted for different parties in various elections. Even within these groups, wellbeing still proved to be a significant predictor of incumbent support. In fact, this effect among swing voters was even stronger than it was in the full sample.Footnote ²⁶ In the second test, the authors split their sample not in terms of partisanship but in terms of exogenous shocks to wellbeing. They first selected out respondents who recently became widowed, and then used propensity score matching techniques to compare the voting preferences of these respondents to those who are similar to them in all other relevant respects, except for the fact that they did not recently become widowed. This approach is intended to resemble a randomised control trial, though in this case the treated and control groups are divided by (presumably random) variation in the recent death of a spouse and not by random assignment on the part of researchers. Using this procedure, ‘treated’ respondents who recently became widowed were 8% less likely to support the incumbent party than controls.

A third analysis related to the United States. It used county-level data from the Gallup Daily Poll and found that wellbeing levels were significantly predictive of incumbent party support in the 2012 and 2016 US elections.Footnote ²⁷ In 2016, low life satisfaction today and low expected life satisfaction in five years explained 28% and 61% of the variation in Donald Trump’s vote share, respectively. The effect of the latter proved to be larger than any other variable under consideration, including race, age, racial animus, education or population density. In subsequent regressions, a one standard deviation increase in life satisfaction was associated with a 7-percentage-point reduction in Trump voter support in 2016, while a similar increase in expected future life satisfaction was associated with a 12-percentage-point decrease. The authors also found similar results for the 2012 election, in which present and expected future life satisfaction scores predicted decreases in support for Mitt Romney, the Republican challenger to Barack Obama, of 6 and 10 percentage points respectively.

A handful of other analyses have used other forms of exogenous shocks to wellbeing to explain voting outcomes. In one of the most entertaining tests of this sort, election outcomes in counties across the United States were linked to the outcome of local sports games.Footnote ²⁸ The authors found that counties in which local college football teams had won games in the ten days leading up to the election were 1.6 percentage points more likely to support incumbent parties in Senate, gubernatorial and national elections. The authors suggested that this result was likely explainable in terms of higher levels of wellbeing in counties with victorious teams, although the analysis did not contain a direct measure of wellbeing.

Overall, the results of this section imply a strong link between happiness levels and incumbent party support. While all of these results are representative of effects in high-income countries, some analyses have also found wellbeing to be a significant predictor of incumbent party support in Latin AmericaFootnote ²⁹ and in Malaysia.Footnote ³⁰ While this literature is still very much in its infancy, these findings underscore the important role that voter wellbeing can play in determining election outcomes. In this section, we have focused on the role of wellbeing and incumbent support, though there are of course many more lenses through which this relationship could be analysed. In the next and final section of this chapter, we will consider one more of those perspectives in particular: the association between wellbeing and populism.

Populism

In the previous section, we introduced evidence indicating that US voters with lower wellbeing were more likely to vote for Donald Trump in the 2016 election. These results are largely in line with related evidence showing dissatisfaction as a predictor of non-incumbent party support. Yet they may also be indicative of another phenomenon: the rise of populism. While populist political parties are nothing new;Footnote ³¹ in recent years, many of them have gained traction in Western countries. In Europe, populist parties have more than doubled their share of the vote in national elections since 1960, from 5% to 13%, while their share of parliamentary seats has tripled.Footnote ³² A number of explanations have been put forward to explain these developments, though perhaps the most common narrative in popular discourse has been the rise of ‘discontent’.Footnote ³³ In this section, we will look at the empirical evidence for this relationship in greater detail. Specifically, we will consider whether or not low wellbeing predicts support for populism.

Populist movements have sprung out of both left-wing and right-wing political movements, making them somewhat difficult to classify. Settling on a universally agreed-upon definition can be challenging. Nevertheless, most researchers generally agree on certain key shared features between all populist movements. Three in particular are: (1) valuing ‘the people’ in opposition to ‘the elite’, (2) opposition to the political establishment and (3) support for popular sovereignty.Footnote ³⁴ Using these characteristics as a starting point, several classifications of European political parties have been developed to rate their degree of populist rhetoric, platforms and policies.Footnote ³⁵ Armed with these data, some researchers have begun examining the extent to which wellbeing is predictive of populist party support.

In one recent analysis of roughly 180,000 European adults across 29 countries, lower levels of wellbeing were significantly associated with higher levels of populist support.Footnote ³⁶ After controlling for a number of personal characteristics including age, gender, race, education, employment status, income, residential area and other related variables, respondents who were the most satisfied with their lives were 3.7 percentage points less likely to have voted for a populist party in the previous election than those who were least satisfied with their lives. To put this figure into context, it was larger than the effect of anti-immigrant sentiment on populist support.Footnote ³⁷

Another approach is to consider populist support not in terms of voting preferences but in terms of political attitudes. An analysis of this sort used representative data on more than 350,000 respondents in 100 countries and estimated the extent to which life satisfaction is predictive of political attitudes associated with populism, after controlling for age, gender, income, education, marital status and country fixed effects.Footnote ³⁸ The results are presented in Figure 17.5. Overall, wellbeing proves to be highly related to political attitudes across the board. Happier voters are more likely to have confidence in political parties, faith in the political system, maintain a positive opinion of democracy and consider themselves to be citizens of the world. They are also less likely to view having a strong leader as a good thing for their country. The starkest differences relate to political institutions. Compared with the least satisfied respondents, those reporting a 10 out of 10 on a life satisfaction scale are roughly 5% more likely to have confidence in the political parties and 13% more likely to have a positive opinion of the political system.

Figure 17.5 Difference in political attitudes depending on life satisfaction

Note: Based on results from Ward (Reference Ward, Helliwell, Layard and Sachs2019). Bars represent percent differences in political opinion from those with the lowest life satisfaction level Estimated using OLS linear regressions controlling for household income (quintiles), education level, marital status, gender, age and its square; 95% confidence intervals displayed.

Source: World Values Survey

Both of the analyses thus far suggest that the rise of populism in Western countries is at least partly related to unhappiness. A separate strand of literature has sought to address this issue by closely examining notable political successes for populists in recent years, including the 2017 national elections in France, the Brexit referendum in the United Kingdom, and Donald Trump’s presidential victory in the United States. Here, the results tend to be less straightforward.

In France, Marine Le Pen’s populist National Front party outperformed traditional left- and right-wing political rivals to reach the final round of the runoff national election in 2017 against Emmanuel Macron. Macron eventually won by a comfortable margin, but the stark rise and success of Le Pen’s party platform, which was partially fuelled by populist anti-establishment and elite-opposition sentiment, warrants further attention. One study sought to examine the drivers of French populist support using a unique dataset of roughly 17,000 surveys of voters in the lead-up to the election.Footnote ³⁹ The authors found that lower levels of life satisfaction were strongly predictive of votes in favour of Marine Le Pen. Her voters were less satisfied with their lives than supporters of any other candidate. Even after controlling for income, education, ethnicity and other sociodemographic variables, low life satisfaction remained significantly predictive of populist support, while high life satisfaction predicted support for the eventual winner Emmanuel Macron. At the same time, Le Pen voters were also less trusting of others (including their family and neighbours) and less optimistic about their future than any other group across the political spectrum.

While Marine Le Pen did not win the French national election in 2017, populist movements for Brexit in the United Kingdom and Donald Trump’s presidential candidacy in the United States proved successful. However, in both cases, the evidence seems to be somewhat mixed. In the United Kingdom, two studies in particular have examined the link between support for leaving the European Union and dissatisfaction. While both studies find dissatisfaction with income in particular to be strongly predictive of Brexit support – in fact, even more predictive than the actual level of income itself – life satisfaction was found to have a much smaller and largely insignificant effect.Footnote ⁴⁰

One study in the United States also looked at county-level data on wellbeing and voting patterns for Donald Trump and Bernie Sanders in the Republican and Democratic primaries.Footnote ⁴¹ Because both candidates promoted populist messages and policies, we might expect that dissatisfaction would be predictive of support for both. The authors do in fact observe this to be the case. In two separate regressions controlling for income, employment status, religion, economic growth, residential area population density, and region fixed effects, Republican voters with low life satisfaction today or low expectations for future life satisfaction in five years were more likely to vote for Donald Trump in the primary, while Democratic voters with the same characteristics were more likely to vote for Bernie Sanders.

A related study also relied on high-frequency Gallup data to examine the extent to which changes in wellbeing from 2012 to 2016 could explain electoral swings in favour of Donald Trump.Footnote ⁴² The authors compiled information on life satisfaction, future predicted life satisfaction and affective wellbeingFootnote ⁴³ for roughly 177,000 US respondents in 2012 and 353,000 respondents in 2016. The study found that counties that swung from supporting Barack Obama in 2012 to supporting Donald Trump in 2016 were significantly more likely to have experienced average declines in wellbeing over the same period. Specifically, in counties in which the vote share from Democrats to Republicans swung by at least 10%, the rate of respondents reporting severely low life satisfaction (1 to 4) had more than doubled from 3.4% to 7.1%, while the rate of respondents reporting high life satisfaction (7 to 10) had declined from 73% to 61%. Similar trends were observed for future expected life satisfaction and satisfaction with the area in which the respondent lived. Residents of these counties were also significantly more likely to report feeling sadness and less likely to report feelings of happiness and enjoyment. For comparison, changes in income over the same time were not significantly predictive of vote shifts.

Finally, another study found that, controlling for other factors, both feelings of worry and racial animus significantly predict higher levels of Trump support. However, once a measure of relatedness (social connection to others) was introduced, the effect of worry becomes significantly weaker, while the effect of racial animus becomes insignificant. The authors interpreted this result as an indication that Trump support in the 2016 election was driven primarily by a desire for in-group affiliation to buffer against the economic and cultural anxieties that had led to unhappiness. In other words, voters who felt disconnected from their communities channelled their anxieties towards Trump support. These dynamics echo those observed among Marine Le Pen’s supporters in France.

All of these studies provide suggestive evidence that dissatisfaction and disconnectedness precede and predict populist victories. Taken together, they underscore the role of social connection and general wellbeing in explaining the recent resurgence of populism in Western countries. However, it is also important to recognise the limitations of these results. While the longitudinal dynamics observed in these latter studies are suggestive that unhappiness drives populist support, the issue does not easily lend itself to causal inference. As of yet, no studies to our knowledge have sought to run randomised controlled experiments in which treated respondents are induced to feel more or less happy and then asked about their level of populist support. Exploiting natural experiments or quasi-experimental designs in the future to predict populist party vote shares may prove fruitful.

However, overall time series cast some doubt on the interpretation that the rise of populism can be entirely explained by declining wellbeing. As discussed in previous chapters, in many countries around the world, including those in Europe and North America, average levels of life satisfaction have remained remarkably flat.Footnote ⁴⁴ Given the dramatic increase in populist party support over recent years, one might expect similarly dramatic declines in life satisfaction. This does not appear to be the case. At the same time, in many countries, social isolation and negative affect (a sense of ‘worry’ in particular) have been on the rise.Footnote ⁴⁵ Inasmuch as this phenomenon reflects growing disconnectedness and anxieties about the future, it could help to explain populists’ appeal to voters. These issues remain open to future research.

Overall, the results of this section call out for further research and experimentation. Despite a handful of recent electoral defeats in both Europe and the United States, populist political parties on both the left and right of the political spectrum retain considerable influence in mainstream politics. Understanding the primary drivers of their support, and in particular the role of wellbeing in explaining them, will remain a central and urgent challenge for social scientists in the years to come.

Questions for discussion

1. (1) Research has shown that wellbeing predicts incumbent party support. If you were advising an elected official, how would you make use of this knowledge?

2. (2) So far, there has been limited experimental research to examine whether or not low wellbeing leads to populist support. Can you think of an experimental design (either in the lab or in the real world) to test this assumption? What are the main benefits and drawbacks of your approach?

Relation to traditional cost-benefit analysis

A natural question is, How does this approach compare with traditional economic cost-benefit analysis, where benefits are measured in units of money (rather than of wellbeing)? The answer is that the results of existing cost-benefit analysis can be incorporated very easily into the wellbeing framework. For we know the impact that money income has on wellbeing, and we can always therefore convert any benefits measured in money into benefits that are measured in units of wellbeing. To be precise, if income is Y and X is a policy variable, the effect of the policy on wellbeing is given by

$\frac{\mathrm{dW}}{\mathrm{dX}}=\frac{\mathrm{dY}}{\mathrm{dX}}.\frac{\mathrm{dW}}{\mathrm{dY}}$(2)

where dY/dX is the effect of X on income, and dW/dY is the marginal utility of money.

This approach is often very useful, because for some policies it is easier to measure their effects initially in monetary units. This applies not just to direct effects on wages or other incomes (due, for example, to educational investment), but also to intangible benefits (like reduced journey times). These intangible benefits are in this case valued by what people’s behaviour shows they would be willing to pay for them (by their revealed preference).Footnote ⁵

But willingness to pay works only when people can show by their choices how much they value different outcomes. Sometimes they can do this, but very often they cannot. They can do it for things like transport, industrial production, education and some aspects of the environment. But many outcomes are not things that people can choose – they are things that just happen to people through outside influences – what economists call external effects. People fall sick, children get abused, elderly people get abandoned and people get mugged. We cannot learn about how much people value these experiences by observing choices. So how are we to evaluate policies like vaccination, or child protection, or family courts, or elderly care, or police protection? Measuring benefits in units of wellbeing is an obvious solution.

Critics might argue that, even though people can’t show their values by their choices, we can ask them hypothetical questions about how much they would in principle be willing to pay to promote these goods? Unfortunately, however, it has been shown repeatedly that asking people hypothetical questions about how they value things produces nonsensical answers.Footnote ⁶

So data on the happiness effects of activities may offer a better route to evidence-based policy making. But why not then translate those wellbeing estimates back into units of money? Thus, suppose we know the effect of a policy upon wellbeing. We could compute the equivalent gain in income that would increase wellbeing as much as the policy would. This equivalent variation in income could be computed by dividing the change in wellbeing by the marginal utility of income. Thus, if policy X improves wellbeing by dW/dX, the equivalent variation in income is given by

$\frac{\mathrm{dY}}{\mathrm{dX}}=\frac{\mathrm{dW}}{\mathrm{dX}}/\frac{\mathrm{dW}}{\mathrm{dY}}.$(3)

This is the reverse operation to that in equation (2). The resulting aggregate benefit measured in units of money could then be compared with the cost.

There are, however, two overwhelming objections to this approach. First, it automatically makes changes in happiness less important if they occur to poor people – it treats a dollar as the same whether it belongs to a Trump or a tramp. To avoid this, the results could be analysed separately for different income groups, applying a different marginal utility of income to each group. This re-establishes wellbeing as the measure of benefit, so why not simply stay with it in the first place?

Second, we might not want to simply add the ∆Ws, but rather to give extra weight to those with low initial happiness. If the monetary valuation procedure is followed, there is no way to do this, since the happiness level of each individual has become invisible. We may or may not want to give extra weight to those who are most miserable (see Chapter 2), but it is helpful to retain the ability to do so.

Note that throughout this chapter we assume that the total budget is determined by political considerations. We do not use the wellbeing approach to determine the total of public expenditure. The same thing happens with traditional cost-benefit analysis – projects are often rejected even if their monetary benefits exceed their costs. The reason is that there is not enough public money to finance all projects whose benefits exceed their costs. In consequence, the only projects that get through are those with a high enough ratio B/C, with the cut-off value being frequently higher than 1.

The alternative approach would be to allow wellbeing cost-effectiveness analysis to determine the total of public spending. But this would lead to much higher public expenditure. For example, if W = 0.3 log Y, the value of a healthy life year (with W = 7.5) is in England $750,000. But England’s NICE does not sanction the expenditure of more than $40,000 per additional life-year.Footnote ⁷

Taxes and regulations

Going on, there are other important public policy problems besides how to spend a given budget total. There is the issue of how to structure the taxes. The approach here can be straightforward. If we envisage a self-financing tax change, we would simply evaluate how this alters the happiness of each member of the population and aggregate these changes (assuming we are simply maximising the sum of wellbeing across all members of the population). We should also use wellbeing as the criterion for whether to introduce a new regulation or to abolish an old one.Footnote ⁸

Five major issues

It is time to address some thorny issues. First, there is the issue of the discount rate. We discussed this in Chapter 15, where we suggested something like 1.5% per annum.

Then there is the issue of the length of life. If we know the quality of any additional life-year that a policy would produce, we would value the extra life-year by the quality of life it produces. But otherwise we would value changes in life-years by the average level of wellbeing (in the country in question).

Then there is the issue of how we treat the birth-rate. If we increase discounted WELLBYS by encouraging the birth of more children, would that count as a benefit? If it did, it would almost certainly be the most cost-effective way of increasing the sum of future wellbeing. For example, by switching expenditure from healthcare to child subsidies, we could surely increase the birth rate by more (in %) than we reduced the length of life. This would increase the number of future WELLBYs. But most people would not support the policy. We would therefore propose that, when evaluating the effect of policies, we ignore any effect on total WELLBYs coming from changes in the number of people born.Footnote ⁹

Next, there is the issue of whose wellbeing counts. In principle, it should cover at least the whole of humanity. Every person is equally important. But for private ethics, there are some people who are easier for us to help than others. So, in practice, human society works through a division of labour. People take especial care of people close to them, and this also satisfies the need that humans have for a kind of affection they can only give to a small number of people.

But, when it comes to public policy or charitable activity, the circle of concern has to be widened to include people we do not know, including people in far-off parts of the world.Footnote ¹⁰ Ideally, each government would choose to do whatever it could for the good of humanity. In practice, democratic governments inevitably feel that their main responsibility is for their own electorates. But this responsibility would hopefully include two other important considerations:

• the altruistic desire to help less happy nations and

• the necessity of collaborating with other countries to secure global public goods like fighting climate change and securing world peace.

And what about the wellbeing of other sentient beings, besides humans? They must surely count. There is ample evidence that birds and mammals (at least) have feelings of pleasure and pain. For example, researchers have offered injured birds or mammals the choice between food that includes standard pain-killers and food that does not. Injured animals prefer the food with pain-killers. And, even more important, when they’ve taken the painkiller, they stop whimpering or calling out. This shows that the choice of the painkiller is not simply an automatic reaction to a wound but a reaction to an emotional feeling.Footnote ¹¹

Finally, there is the issue of equity. As we said in Chapter 2, the starting point for public policy analysis could be the Benthamite approach of adding up all changes in WELLBYs, regardless of who they accrue to. But most people would probably wish to give extra weight to improving the wellbeing of those who are least happy. The problem is how to secure agreement on the weights. One obvious approach is to conduct a representative survey of the views of the public, and this is a high priority for future research. In the meantime, one natural approach is through sensitivity analysis, examining how far the results are reversed when different weights are used. Equally, when considering new policy initiatives, it seems natural to focus on those areas of life that account for the greatest amount of total misery.

Experiments

But this is only the beginning of the search for new policies. The next step is to identify specific policy changes that might be considered. Once a plausible policy option has been identified, it would ideally be the subject of a proper randomised experiment in the field. Where it is impractical or unethical to randomise across individuals, it is often possible to randomise across areas or across institutions (schools, hospitals, etc.). From such an experiment would come information on the short-run wellbeing benefits and budgetary costs of the experiment – and then it can ideally be projected into the longer term using a model.

The final result of such an evaluation would be an estimate of the change in WELLBYs per dollar of expenditure – or alternatively the number of people removed from misery per dollar. Or these estimates can be expressed the other way round – as the cost per WELLBY or the cost per person removed from misery. Table 18.2 is a crude illustration of the latter approach. It is a back-of-the-envelope calculation and we only include it to provoke thought and discussion and to encourage the reader to do better. In the table, we have taken four standard methods proposed in the UK for reducing misery. We then estimate what public costs would need to be incurred each year to ensure that there was one less person in misery. (The assumptions are in Annex 18.2.) The outcome is somewhat surprising. Better mental health care is the most cost-effective of the four polices, followed by active labour market policy and physical health care, with income redistribution the least effective. Because redistribution is so expensive and relatively few of those in misery are also poor, it is often more effective to spend public money on services in kind – helping people to help themselves. So in this analysis, the top priority is building up the social infrastructure (health care, skills development, employment, and community services).

Questions for discussion

1. (1) How if at all does the wellbeing approach to public expenditure improve on traditional cost-benefit analysis. Can the two approaches be reconciled? How?

2. (2) How can equity considerations be best incorporated in the design of policies aimed at wellbeing?

3. (3) Would we ignore effects on the number of births?

4. (4) What is your reaction to Table 18.2? How could you improve the analysis?