Does Happiness Lie within Trust?

(1)

Does Happiness Lie within Trust?

Happiness and Social Capital in Japan, Mexico and the United States

By

J.A. Hummel

1

Abstract: For many years economists have believed that economic growth and public

well-being would go hand in hand. As of the 1970’s, however, an increasing number of scientists from the fields of economy, philosophy, psychology and sociology have found that after a certain threshold the relationship between income and public happiness becomes very weak over time and across countries. This paper looks into this ‘paradox of happiness’ for the case of Japan, Mexico and the United States. It is argued that a loss in social trust, as a proxy for social capital, could be indicated as one of the explanations for this paradox.

Key words: Happiness, Social capital, Trust JEL Code: A13, A14, N32, N35, N36

Master Thesis

Faculty of Economics

Rijksuniversiteit Groningen

Supervisor: Dr. H.J. de Jong

August 2008

(2)

Does Happiness Lie Within Trust? Happiness and Social Capital in Japan, Mexico and the United States

I Introduction

An economist that is engaged actively and self-critically with the moral aspects of its subject matter cannot help but be more interesting, more illuminating, and ultimately more useful than

one that tries not to be.

Hausman and McPherson (1993, page 723)

In march of 2008 eleven Dutch and Flemish scientists and prominents signed the Declaration of

Tilburg. It contained a drastic cry for a fundamental reorientation of our national economies in

order to protect the world from further ‘health hazards, environmental degradation, a further increase in the global poverty gap, armed conflicts and refugee movements’. One of the main focal points of the declaration is to challenge the ‘continuous celebration of increasing material consumption per capita’. It is their strong belief that – despite of what is commonly believed – lowering national incomes, will not decrease national welfare.

(3)

and consumption. But, is this right? Studies on the relationship between income and happiness did show a positive - but weak – association between the two, but a large part of the composition of happiness still remains unexplained. Implying that material well-being and happiness are not one and the same thing. This leads me to believe that happiness and its determinants should now receive the attention that they deserve from, among others, economists.

This paper aims to shed some light on one possible component of the unexplained part of happiness: the influence social relationships might have on public happiness. As will be clear from the following literature study, differences or changes in social relationships have been put forward as a missing link in the measurement of public happiness which has long focused solely on differences in income. Because researches on the time that people spend engaging in social interactions and the value they ascribe to them are still very limited, this paper will shift its focus to the related concept of social capital of which more data is already available. More specifically, I will look into the matter using data on generalized trust – which has been used in several other papers as a proxy for social capital as well – since trust has been measured for several years now. And since most studies on happiness thus far have focused on developed nations and paid only minor attention to the less developed world, I find it interesting to see whether different cultures might have different interpretations of social capital which might explain some of the residual parts of happiness that differences in income cannot account for. In order to do so I will study the relationship between happiness and social capital for the case of Japan, Mexico and the United States, since these societies represent three different cultural areas, had different growth rates over the years, and all participated in researches on their average happiness levels. The main question this paper will address is whether a difference in generalized trust can (partly)

explain why the relationship between income and average happiness becomes weaker over time.

(4)

(5)

II Literature on Happiness

In this section an inventory of the most recent and some of the older literature on happiness will be presented. The first subsection will discuss some of the different interpretations scientists have ascribed to the concept of happiness. A second subsection will be devoted to the treatment of the so-called paradox of happiness. The third subsection will deal with some of the explanations scholars have found for this paradox.

2.1 Different interpretations of happiness

For the research into the field of happiness we owe a great deal to the Italians. The first time the phrase ‘public happiness’ was introduced, dates back to the eighteenth century. In the year 1749 the Italian writer Ludovico Antonio Muratori published a book titled Sulla Pubblica Felicità (or, as Bruni and Zamagni (2007) point out, the complete title of the book is Sulla pubblica felicità,

oggetto dei buoni principi – which translates to ‘on public happiness, object of good princes’ –

meaning that happiness should be the final goal of politics). Other early Italian economists, like Antonio Genovesi and Giuseppe Palmieri, dug into the theory of value and public policy2. Today there are several scientists who believe that happiness should return to the focus of research once again.

In the field of philosophy, Sumner (1996) makes an inquiry into the nature and value of well-being, welfare and happiness. In defining the concept of happiness, Sumner distinguishes between four dimensions: (i) being happy with or about something, (ii) feeling happy, (iii) having a happy disposition/personality and (iv) being happy/leading a happy life. The first dimension requires an intentional object that has been completed. In other words, there has to be something concrete to be happy with/about (either tangible or intangible). This dimension of happiness is closely related to satisfaction. The second dimension contrasts with the first in the fact that it doesn’t require an intentional object and involves an occurring feeling. This is what we like to call a cheer and it tends to last only briefly. The third dimension is basically a longer lasting version of the second. The happy feeling is associated with an occurring episode in life, rather

(6)

(7)

As for the relationship between happiness and well-being, Uchida et al (2004) point out that (subjective) well-being is commonly described as the broad and cognitive assessment of the quality of personal life. Its emotional attendant is referred to as happiness. Dasgupta (2001) also defines being as the quality of life, but at the same time makes a distinction between well-being and welfare. He denotes well-well-being as the broader notion of the two, since it includes both welfare and on-welfare characteristics (e.g. human rights) at the same time. Personal well-being, therefore, consists of a selection of different objects, like health, happiness, social relationships and freedom.

Happiness in this paper will therefore be placed within the larger context of the well-being literature. For the proceedings of this paper I have chosen to take a more prudential approach to the concept of happiness than most economists have done so far. In my opinion it is crucial that scholars treat research topics in a likewise manner in order to intensify and improve collaboration between different fields of sciences. From above treatment of various interpretations of happiness it will be clear that although it might be easy to denominate happiness as a synonym for e.g. satisfaction, it is actually wrong and should be avoided. Psychologists and philosophers have argued that happiness is an emotional state which can be characterized by different dimensions. When economists write about the subject and use the term very broadly or – even worse – interchangeably with others terms, this could have serious implications for the value of their researches. Most importantly, it would make comparisons between different researches extremely difficult. For it could very well be that two scientists measure two different aspects of the concept. After all, economists have come up with clear definitions of their other research topics (like e.g. GDP) for a good reason. One of the problems that might already be present, is the discrepancy between happiness and life-satisfaction that Pieró (2007) has mentioned. He stresses the difference in the manner in which either is affected by economic conditions. Depending on the definition that is being used, one research is bound to find a larger impact of economic factors when it measures life-satisfaction than it would have done had it measured happiness.

(8)

mentioned above. Although I do agree that happiness and life-satisfaction are closely related, I distance myself from any scientist who would argue that it doesn’t matter which definition is being used. I plea for a commitment from scholars from the fields of psychology, philosophy, sociology and economy to find an unambiguous definition of happiness in order to eliminate the mentioned evaluation problems between these different scientific researches. This, however, doesn’t imply that economics will loose its importance or distinctive mark within scientific research (since there are still many important and unique contributions from this science to be made, like the useful tool of cost-benefit analysis for e.g. environmental policies). It only means that when the time comes that psychologists or sociologists will find ways to measure happiness in such a way that it can be fitted in neatly into economic functions – or when economists finds better ways to handle non-monetary and non-numeric parameters – happiness should be a logic and standard part of any well-being analysis. For I cannot see why we should insist on measuring economic progress, if we are not sure if it is indeed progress we are measuring. After the next subsections it should be clear why I have taken on this statement.

2.2 The paradox of happiness

Since the 1970s the so-called ‘paradox of happiness’ has been gaining increasing interest of various scholars from the fields of economy, psychology, and sociology. Many of them have made efforts in trying to explain what causes this paradox, and how to solve it.

(9)

people how they feel, is an attractive method for assessing average happiness. Still, he doesn’t ignore possible reservations about the data. For instance, it could be doubted that cultural differences leave the concept of happiness unaffected. As a counter argument Easterlin puts forward that the translation of the terms (like happiness) in the questionnaires has been carried out with great accuracy. Moreover the responses about hopes and fears are similar in all countries, proving that people do not differ too much in their attitudes towards life in general.

In his analysis Easterlin found that happiness and income are positively related within every country and that the richest inhabitants are, on average, happier than the poorer ones. He could not, however, find any consistent evidence that (inhabitants of) richer countries could be nominated happier than (inhabitants of) poorer countries. Nor did he find proof that higher incomes provided more happiness over a longer period of time. In analysing these puzzling results, Easterlin suggested that while income does seem to matter greatly to people, it is by far not the only thing that is important for them. When asked about their level of happiness, people take their past experiences and their current status into consideration as well. At any given point in time, they compare themselves to a reference standard, which moves up in line with the growth in their income and consumption. This process of never ending desires is what Easterlin regarded as the engine of economic growth, which, in his view, will never lead us to an ‘ultimate state of plenty’.

(10)

time periods as a consequence of e.g. fluctuations in economic conditions, and this too calls out for the need for longer time series with larger data sets in order to estimate long term trends. Thirdly they point to problems already mentioned earlier in this paper, i.e. the fact that researches use different types of data sets (sometimes within one paper) which could trouble the outcomes. To prevent this sort of problematic issues, Veenhoven and Hagerty feel there are two possible solutions. First, one could choose to include only those data sets that have measured the same concepts. Or, second, one could argue that using all the data sets that are available will eliminate the variety in them because of the large quantity. Whereas Easterlin has always opted for the first solution, Veenhoven and Hagerty favour the second. As for me, I would agree with the more careful approach of Easterlin. Although it might be true that the rule of large numbers equals out much of the variety within data sets, this will not hold if those data sets measure two distinctively different concepts. As was earlier discussed, happiness and life-satisfaction are not entirely the same and it could very well be that Veenhoven and Hagerty’s approach is a little like comparing apples with oranges.

In their attempt to prove that happiness did rise in recent years, Veenhoven and Hagerty take a look at the newest time series on happiness data for the United States and eight European countries, as well as some of the poorer nations of the world. In the period 1947-2000 they found an average yearly improvement of happiness in the United States of 0.006 percent. Which corresponds with a 167 years time period to gain one point on a 1 to 10 scale of happiness. And although the authors acknowledge that this is a small increase indeed, they also point out that small changes are not unusual in long-term developments. As an argument to support this, they put forward the finding of Maddison (1995)3 that the world economy only grew at a rate of 0.0004 percent per year in the period 1500-1820. But in my opinion, this is hardly proof that the results Veenhoven and Hagerty have found are therefore significant. After all, the world’s economy and society between 1500 and 1820 and that of the post-war period are hardly comparable. For the European countries, Veenhoven and Hagerty only found significant positive trends in four of the eight countries for the period of 1973-2004. The authors did not compute any

(11)

regressions for the less developed nations due to lack and inconsistencies in the data sets, but they can conclude that the rises in happiness have been larger over time in poorer countries than in richer countries.

In my opinion, the implications of these researches are not to state whether there has been no increase in happiness or a small increase of less than one hundredth of a percent. What is important is that these researches have made clear that the enormous rises in national income and consumption over the last sixty years are not nearly matched by an increase in either happiness or life-satisfaction. Therefore, it is my opinion that economists along with other scientists should look into this matter in order to find out why the relationship between happiness/life-satisfaction on one hand and income on the other becomes so weak over time.

2.3 Explanations for the paradox of happiness

After Easterlin’s (1974) paper, a number of authors have subsequently contributed to the explanation of the paradox of happiness. They all have presented alternative indicators that could possibly explain a little more of what makes people tick. Although their approaches are different, most of them seem to agree on one thing: economics today, with its focus on growth of GDP, cannot accurately measure public happiness because it foregoes on something important. And it is exactly that ‘something’ important that leaves room for different interpretations. It has been sought in unemployment effects, the influence of democracy, freedom, in positional externalities, the role of inflation, the loss of social relationships or altruism, in political participation, or in social aspirations. This subsection discusses some of those alternative approaches for the explanation of the happiness paradox. It starts with the approach of scientists who have primarily focused on explanations within the field of economics, and ends with those approaches that have taken on positions based on various social studies.

(12)

that Easterlin (1974) and others following him, have used: the Eurobarometer Survey Series for Europe and the General Social Surveys for the United States. From these it can be concluded that life satisfaction (not happiness) in the European countries has on average been rising, but only slightly. The same holds for the United States.

(13)

contact) should be quantified or estimated as accurate as possible. That way, economic factors would be determinants within the determinants of happiness and (part of) the happiness equation would have the following form:

Happiness = f(Income, Health,…, Social relationships f(Employment,…))

Because obtaining data on happiness can be quite difficult, Oswald also feels that it could be useful to look into data on extreme unhappiness, i.e. suicide data. Although some scientists state that suicides are not a proper indication of public happiness (many suicides are the outcome of mental illness), Oswald states that it may be wrong to consider every suicide to be irrational. He claims there is evidence that a frequent number of suicides occur among people who have very little in life to live for (in an objective sense) and among those who have experienced a negative event in their lives – including harsh economic situations. Oswald continues to state that historical data sustain the fact that economic events can have a downward effect on people’s happiness, because successful suicides reached their peak during the Great Depression of the 1930s. Research has also found evidence that suicide death rates are in large independent of the social class to which the victim belonged. Thus indicating that poor people do not commit suicide more often than rich people do. From the data on suicide Oswald found that suicidal behaviour is more common among men, people who are unemployed, or those going through marital problems. It also seems that richer countries show more suicide death rates than poorer countries do (but I feel this result might be a bit biased, since poor countries might not report as accurate as richer countries on death or suicide rates). Furthermore, almost all Western countries report rising male suicides since the 1970s, despite booming national incomes.

(14)

have even dropped. And although incomes have increased, the number of male suicides has been rising in the industrialised countries since the 1970s. These findings, however, do not mean that Oswald fully supports Easterlin’s (1974) conclusion that economic growth is worthless. He admits that due to the difficulties of measuring well-being, his paper should be read critically. It is, for instance, not to be said for certain that the answers obtained from questionnaires are reliable. Nor it is proven without a doubt that suicide rates are a good indicator of people’s happiness, or mental state. But Oswald defends his position by stating that these researches are the only information that is available for now, and that we should look into the doubts they raise about routine beliefs concerning the often heard statements of politicians and economists that higher growth rates or lower levels of inflation automatically imply higher levels of happiness. And that conclusion I fully support.

Other reasons for the paradox of happiness have been sought in the effects of welfare distribution on happiness. In a paper by Alesina, Di Tella and MacCulloch (2001) it is investigated whether or not Europeans and Americans differ in their attitudes towards inequality. The background of this question stems from the fact that European governments have been more intensely involved in redistribution matters over the last decades than American governments. And since democratic governmental policies reflect the will of the people, a large part of the European population must support these programs more than their American contemporaries do. The authors use happiness data collected in the United States’ General Social Surveys (1972-1994) and the Euro-Barometer Survey Series (1975-1992). For the data on inequality, Alesina et al use Gini coefficients of gross family incomes.

To compare happiness and inequality, along with some macroeconomic variables and several individual characteristics, Alesina et al have come up with the following regression formula:

ict t c ict ct ct

ict Inequality Macro Micro

Happy ₌_α ₊_β ₊_δ ₊_η ₊ _µ ₊_ε

where Happyict is the answer of individual i living in country c in year t to the question “How

happy are you?”. Macroct refers to some aggregate macroeconomic variables that have been

(15)

Microict on the other hand, contains personal factors that affect individual happiness like e.g. age,

employment status, or income. The authors have also included a dummy variable for cross sectional units, ηc (for the United States this accounts for the different states, for the case of

Europe these are individual countries), and a dummy variable µt for each year. Finally they

included an error term εict as well.

After running some regressions the authors find no evidence that Americans dislike inequality, while it seems that European happiness is strongly affected by it. Moreover, the European aversion towards inequality seems to be concentrated within the leftists groups of the population. The authors also find that the happiness levels of poor European citizens are negatively affected by inequality, while it has no effect on the rich. In the United States, only a group of rich leftists are affected. Alesina et al propose social mobility as a possible explanation for these results. The American society is a much more mobile society than European societies are. People in the United States have more opportunities to climb up the social ladder and that way, present-day inequality might have little influence on individual happiness. These results imply that inequality in itself might not be a source of discontent. It depends on the cultural values present in a society whether people suffer losses in their happiness levels from unequal income distributions. In my opinion, therefore, it could be useful to look into some of the literature about the effects of culture on public happiness and I will do so later in this section.

(16)

involvement also increases social cohesion (at least for some part), which is believed to enhance average happiness levels.

Frey and Stutzer do point out that GDP has one great advantage, since it is based on the idea that the value of goods and services is given by the utility for the consumer. When the economy finds itself in equilibrium, market prices are thus equal to marginal utility. But as a sole indicator of well-being, GDP faces some problematic difficulties. First of all, there is the problem that the value of goods consumed earlier than margin doesn’t equal market prices, but is in fact much higher. That points to an underestimation of the value of the overall bundle of goods and services, and thus of national well-being. Related to this first problem is that fact that measuring GDP in terms of goods and services produced, neglects those goods that have no monetary value but do yield utility, like personal relationships. This too leads to an underestimation of well-being. A third difficulty arises when GDP incorporates an actual destruction of well-being. This is the case, for example, with accidents from which an increase in national product arises through the expenditures for e.g. hospitalisation. The neglecting of income distribution in measures of GDP is the final problem facing GDP as a sole indicator of well-being.

Frey and Stutzer, therefore, find need for some improvements. One of which could be incorporating non-market activities, such as housework, into the accounts. But the authors also applaud the recent concept of sustainability, which defines an utility level that doesn’t decrease over time (due to e.g. asset depletion). Taking the income distribution, the voluntary sector and the existence of a shadow economy into consideration would, according to Frey and Stutzer, be a serious improvement as well.

(17)

continue to be high, even though it could be expected that they would decline in economies that are as rich as they are now. The same holds for the labour supply: participation rates keep on rising and working time is increasing. Moreover, the massive increase in labour productivity has been allocated to output increases and not towards more leisure time. Bartolini provides some reasons why people are still, as he calls it, enslaved to money and are not as happy as they should be.

Bartolini first mentions the decreasing marginal utility of money. This idea states that people might not be able to obtain what they need, despite the affluent society they live in. There may be needs that cannot be satisfied by marketable goods at all, like e.g. social relationships. Then there is the problem of adaptation. People adapt to new circumstance and therefore rising wealth has only temporary effects on their happiness. Finally, it’s the relative position of a person that is important to his happiness. If everyone’s situation is improving, that person feels no better off despite rising income. According to this ‘relative-position-matters’ theory the trends in work and savings are based on a failure in co-ordination. People are induced to work harder and accumulate more money by increases in other people’s position. Those efforts generate economic growth, which in turn generates an improvement in the absolute position of all. Given the fact that it is the relative position that matters, growth cannot provide a rise in overall happiness. Thus, it is the quest for rank that ensures perpetual growth and stagnant happiness levels. For me, this research provides another reason to broaden the economic point of view on growth and well-being.

(18)

importance of studying happiness once more. She points out the fact that the economics of happiness has one great advantage over traditional economics: this new science no longer relies on revealed choices which cannot provide information about non-monetary factors, but on expressed preferences instead. That way, scientists can begin to measure effects of governmental policies or institutions which cannot be controlled by individuals, like inequality or environmental degradation. To me that would indeed seem like important progression within the field of welfare economics.

An interesting recent book on happiness that draws on insights from economics, philosophy, psychology and sociology, is the work of Layard (2005). He discusses what things in life should make people happy and some explanations for why they might not be. Layard believes that people in general are meant to be fairly happy, even though they may differ in their natural state of mind, ranging from optimistic to doomsayers. This, combined with a necessary dose of anxiety, is needed for the survival of our species. But that doesn’t explain why we, since we have wiped out a lot of the causes of former misery (like poverty and diseases) aren’t happier than before. Sceptics might say that we are indeed happier, but just don’t recognise it because of

rising expectations. Layard provides us with three arguments against this theorem to prove that happiness has not risen in the West since the second World War.

(19)

third argument, finally, contains evidence about the recent trends in alcoholism and crime. All of these have rapidly increased since the second World War. For alcoholism, the rates have been up since the second half of the twentieth century. But the evidence of crime tells a surprising story. During the Industrial revolution of the early nineteenth century crime rates in the cities were high. But as incomes rose, the social order regained itself and crime rates fell. Until the post-war period arrived, after which the numbers have been rising ever since.

After establishing that having a higher income is not a definite guarantee for greater happiness, Layard turns to investigate why. He claims there are two forces at work: you compare your situation to that of the people around you, which is known as social comparison; and secondly, you compare what you have today to what you’re used to having, called habituation.

As for social comparison, Layard reasons that this is why over time richer countries are no happier than poorer ones. People in rich countries compare themselves with other rich people, instead of their poorer fellowmen. One thing that can seriously affect one’s happiness is when people change their reference group upward. Whenever someone was used to comparing himself to people who are less fortunate in life, that person is now looking at people who are richer. Consequently lowering his own happiness. This is for example the case in East Germany, where people are nowadays comparing themselves to (richer) West Germans in stead of other former Soviet residents. The best place to be, therefore, is at the top of the richest group. That way all other people are poorer, thus raising your happiness. This partly explains why on average rich people are happier than poor people. Habituation on the other hand, seems to be an integral part of our lives. In order to survive it is necessary for men to possess the ability to adapt to hardships. There are, however, a number of miseries that people never seem to fully adjust to like e.g. widowhood. But, at the other side, there are also positive things that we do not fully adapt to, like sex or friendship.

(20)

patterns in the areas of family relationships, communities and personal values. Since the second World War divorce laws have been eased, causing a dramatic increase in broken families. Crime rates have also been rising and trust has subsequently deteriorated drastically. According to Layard, these are all negatively influenced by changes in science and technology.

Ever since living standards and medical care have improved in the western countries, child mortality rates have declined and women had fewer children. Combined with technological inventions concerning housework, this opened up opportunities for women to join the labour force. At the same time women continued to do more of the housework and parenting, while men felt they got less attention from their wives. This left room for dissatisfaction for both men as well as women. Together with eased divorce laws, marriages have broken down much sooner than before. At the same time new inventions in the control of childbirth gave way to the sexual revolution, leaving marriage even more unstable.

With the introduction of the television during the 1950s and 1960s people’s life altered in various ways. Before the introduction people spent their leisure time with friends and family, or sporting. But now, most people devote much of their free hours watching television, leaving community ties weakened. At the same time television floods people with violence and sex at the one hand and wealth and beauty at the other. While an abundance of violence seems to trigger crime, an abundance of wealth and beauty affects our standards of comparison. We are less happy with the way we look or with what we have, when confronted every day with rich and beautiful people – fictitious or not. It would seem then, that the recent accumulation of luxuries has some downsizes as well. Psychological research has made progress in explaining why this might be.

(21)

percentage of variance in subjective well-being caused by a change in e.g. income. The hedonic treadmill theory is a metaphor for this type of argument: people are continuously running but always remain in the same place, because the treadmill is spinning at the same speed but in the opposite direction. Recently, the treadmill theory has been divided into two types: the hedonic treadmill, as just explained, and the satisfaction treadmill. Where the first type depends on adaptation, the latter depends on aspiration. The satisfaction treadmill theory states that people need more and more pleasures in order to sustain the same amount of happiness: if you have more, you’ll expect more too.

In a chapter by Raghunathan and Irwin in Bruni and Porta (2007), the hedonic treadmill theory is applied to a specific area: i.e. product experience. They relate these experiences to overall happiness and find that consumers derive far more happiness from the consumption of a product if that consumption follows a less pleasant experience in the same product category, and if it is more pleasant than experiences from other product categories. This implies that if past product experiences have been pleasant, future experiences need to be even better in order to sustain at least the same level of happiness. Consumers are thus constantly running in a hedonic product treadmill.

So far it seems that growth in income and consumption did not bring along the kind of happiness or life-satisfaction that has been assumed for long. Theories like the hedonic treadmill theory make clear that buying luxuries might be a non-fulfilling prophecy (despite of advertisers’ efforts to make us think otherwise4). But if economic growth is not the sole component of happiness, there must be others too. In this subsection some of those possible other components have been put forward. What should also be clear from this section, is the possibility that happiness might consist of different components within different cultural areas (or – which in my opinion is more plausible – those different components might have different weights assigned to them in different cultures). It would therefore be interesting to briefly look into the possible relationship between culture and happiness.

(22)

It is not my intention, however, to fully dig into the concept of culture and its determinants. For that I gladly refer to the works of e.g. Diener and Suh (2000) or Schimmack et al (2002). For the purpose of this paper I am merely interested in the possible implications that different cultures might have for public happiness. One paper that looks into this matter is the work of Uchida et al (2004). In this paper it is argued that different cultures might label different types of positive events and experiences as forms of happiness. Each culture has its own set of values about hopes for one’s own life or the kind of social relationships that one needs to engage in. Different cultures might also have different motivations underlying their quests for happiness. And therefore there should be a difference in factors that predict happiness as well. Uchida et al suggest that those factors in European and American cultures involve personal achievements and self-esteem as primary sources of happiness, whereas the realization of social harmony is more applicable for East Asian cultures. The Euro-American cultures are therefore generally regarded as individualistic, while Asian cultures can be described as collectivist cultures. Schimmack et al (2004) have provided a short summary of the differences between individualistic and collectivist cultures regarding happiness and life-satisfaction. In general, individualistic societies place a large emphasis on the independence of individuals, the freedom of choice and on individual needs, and ascribe more importance to emotions. Whereas collectivist cultures greatly value the interdependence of individuals on others, and emphasize duties, other persons’ needs and the acceptance of one’s faith. Following the present cultural norms is also more important in collectivist societies than in individualistic cultures which are more focused on maximizing pleasure. It would be valuable then, to investigate whether indeed different cultural areas experience different effects of social interactions on their happiness levels. Section four will therefore look into this matter for the case of Japan, Mexico and the United States.

(23)

(24)

III Happiness and social relationships

The bird, a nest; the spider a web; man, friendship William Blake (1790)

This section will discuss some of the theories on the effects that social relationships might have on public happiness. I will first look into the concept of social interactions, and later on the focus will be shifted towards the topic of social capital. As for the validity of discussing the topic of social relations, I refer to the work of Haller and Hadler (2006) who argue that happiness is closely associated with social relationships (while life-satisfaction will be more related to material living conditions). The authors also argue that in some cases social relationships can produce happiness or unhappiness directly, but in most cases they will only facilitate their emergence. They claim that happiness will be higher among those persons, groups or societies, where people are more involved in close social relationships than among those where weaker social relationships are common. Another work that deals with the relationship between happiness and social interactions is the paper by Diener and Seligman (2002). They have investigated the difference between very happy people and people with average or low levels of happiness. After studying the responses of more than two hundred undergraduates the authors found that the ones who reported themselves to be very happy, could be characterized as very social individuals who were all engaged in strong social and romantic relationships. In contrast, they found that very unhappy persons had much fewer and less stronger social relationships than average. From the results of their study Diener and Seligman conclude that social interactions seem to be a necessary ingredient to achieve greater happiness (although some of the unhappy people participated in social networks as well). Possible explanations for this trade-off between happiness and human interactions have been sought in the fact that people within increased social networks face more opportunities to discuss important matters with others, thereby reducing loneliness and feelings of powerlessness. This topic however, still needs to be researched more thoroughly.

(25)

made out of relationships which are not used for other purposes but the relationship itself, and that come forward from intrinsic motivations. The relational good has some resemblance with local public goods in the sense that it has to be consumed together, and also with externalities because they too emerge sometimes unintentionally. But they are not the same. Relational goods are, for instance, not only nonrival, they are anti-rival as well.

The authors analyse a simple model linking happiness to relational goods. They come up with the following function: HA = f(IA, RA,B) in which HA stands for the happiness of person A, IA for

income and RA,B for the genuine relationship with other people, i.e. relational goods. To keep

things simple they assume that happiness is a measurable variable. That way, the relationship will have the following shape:

Figure 1: The relationship between happiness and income Happiness (HA)

Income (IA) (time invested in working) Source: Bruni and Zamagni (2007)

(26)

negative impact of loosing of relational goods. But after a certain threshold, Bruni and Zamagni find that the reverse seems to be true.

After establishing this result, the authors shift their focus to the question of why people, who are thought of as rational agents, insist on investing too much time in the accumulation of income whilst paying less and less attention to personal relationships. One possible explanation could be that people are in fact not rational. People make mistakes about what will make them happy and

this is for a large part worsened by present market forces. One of the arguments in favour of this thesis originated from Scitovsky (1976). He came up with the concept of comfort and creativity goods. Comfort goods are those type of goods that make us feel comfortable for a small period of time by saving us time and effort and because they require few skills (like television). Creativity goods on the other hand do require time and skills, but they yield pleasure for a much longer period because they stimulate our creativity (like cultural activities). In making this distinction between comfort and creativity goods (Bruni and Zamagni count relational goods to the latter type), Scitovsky puts forward that nowadays we are consuming too many comfort goods because they are continually offered to us under false pretences of being creativity goods. That way, we invest too much in the wrong type of goods and we end up not being happier than before the consumption. Another explanation of the diminishing time spent on relational goods is the growing relative costs of them. With technological progress, market goods (like e.g. cars or televisions) have become cheaper over time. Relational goods, however, require the same time and effort as they have always done, making them relatively more expensive.

Another interesting work is the research carried out by Konow and Earley (2007), who have studied the relationship between happiness and generosity. Their paper is essentially a response to the proposition posed in Adam Smith’s The Wealth of Nations (1776) and The Theory of Moral Sentiments (1759) that the actions of homo economicus, the rational man who is only interested in

(27)

their paper they answer two questions: do more generous people report higher levels of happiness, and, if so, what might cause this relationship? For their research the authors use a so-called dictator experiment, in which one participant, the dictator, needs to decide whether or not to share (part of) a sum of money with another unknown participant. The outcome of the experiment is then related to responses of the participants in questionnaires concerning several measures of subjective, material and psychological well-being. The results of this experiment showed that participants that shared (and could thus be regarded as generous) showed positive long-run feelings, like higher levels of overall happiness (but less positive short-run feelings compared to participants who did not share).

After analysing the results of their experiment, Konow and Earley find that happiness and generosity are caused by psychological factors of well-being. People with high levels of psychological well-being are more likely to experience long-run happiness as well as higher peaks in the affective dimensions, leading to more generous behaviour. Their findings link up with other scientific researches that have proven that altruism contributes to long term subjective well-being. The authors thus encourage any policies that would reinforce e.g. volunteerism, charitable behaviour or community activities.

(28)

brought about an increase in divorces and broken families, and neighbours often refrain from contact as a result of a dislike of either party. Those relationships that do hold, however, tend to be strong because people are more intrinsically motivated to attend to them.

It would seem then, that being involved in social relationships can bring along happiness. But at the same time, loosening social interactions (for instance as a consequence of present market persuasions) or a diminishing urge for altruistic behaviour, could bring happiness levels down or cause them to stagnate. If that is the case, a lessened care for social involvement could be one of the important factors causing the paradox of happiness.

One field of research that is strongly related to the concept of social relationships is the – among economists more commonly accepted – concept of ‘social capital’. Social capital, according to the World Bank5, refers to ‘the institutions, relationships, and norms that shape the quality and quantity of a society’s social interactions’. It is not my intention to commence an overview of all the recent theories concerning social capital. For that I gratefully refer to the work of e.g. Fukuyama (1995). For the proceedings of this paper, I am interested in this particular topic because of the close association with social relationships. Or more particularly, because social relations form a cornerstone of the broader definition of social capital. Since this topic has been studied by economists for a longer period (and with more elaboration) it will be easier to relate it to the trade-off between income and happiness.

As a research topic, social capital has some measurement difficulties that correspond to the challenges facing happiness research. Social capital is defined by concepts like trust, community and networks, all of which are difficult both to quantify and qualify. Because of that, a number of studies have used trust as a proxy for civic associations. As a consequence, trust has been one of the elements of which data is available on a larger scale. For that reason the next section will make use of studies on trust as well.

(29)

In this section it has been argued that recent trends in increased working hours and lessened relational activities can lower happiness levels after a certain threshold, since both income and relational goods constitute a person’s happiness. There seems to be a fragile equilibrium between the two factors. When that equilibrium is disturbed in favour of either income or social relations, happiness will not reach its optimum. Several authors have indicated that marketing activities focused on the purchase of comfort goods in stead of creativity goods, might be a reason for disturbances of the equilibrium. Because these forces induce people to opt for the (cheaper) type of goods and trigger them to work harder in order to gain more income to buy even more comfort goods (since comfort goods yield only short lasting pleasures). Other researches have indicated that a loss of altruism in today’s societies can be an explanation for the fragile relationship between income and happiness. It has been found that altruistic behaviour can cause an increase in psychological well-being. A larger focus on individualistic material well-being might undermine this effect. It has also been found that economic growth in itself can cause the breakdown of social relationships. Growth brings along financial independence and can cause social relationships to break down much sooner, because people are no longer dependent of each other. All the concerns discussed in this section have in common that they point to the diminishing emphasis on social interactions as a reason for the stagnating average happiness levels in industrialized countries after the second World War. Related to the decreased strength of social relationships is the topic of trust. People in close relationships tend to trust each other and the larger those relational networks are, the higher the level of trust within a society. Whenever those networks fall apart, one would intuitively feel that the percentage of people trusting others might also drop (since there will be more unfamiliar people and less familiar ones). There are of course differences between generalized trust (trust towards strangers) and trust within existing social networks (and even some cultures tend to be more trusting than others)6_{. But since the}

research into this discrepancy and the relationship with individual and public happiness is still in its infancy, I will use data on generalized trust as an indication of the quantity of social networks within a society. Still, I do acknowledge there could be reservations concerning this assumption. The next section will look into the matter of social capital to find out whether increasing levels of mistrust might explain the discrepancy between happiness levels and growth in income.

(30)

IV The Case of Japan, Mexico and the United States

This section will investigate whether a difference in generalized trust can explain why the relationship between happiness and income becomes weaker over time. It will look into the matter from the point of view of three countries from three different cultural areas with an entirely different emphasis on, and interpretation of social capital. I have chosen Mexico and Japan since they can both be qualified as collectivist cultures, and the United States for its individualistic nature. Furthermore, both Japan and the Unites States have displayed high growth rates since the second World War, while Mexico can function as a control case since this nation does not yet belong to the rich industrialized countries. This section has been divided into four subsections. The first subsection will give a brief overview of what has been treated thus far in this paper. The second subsectionwill portray the levels of average happiness and real GDP per capita for the period 1975-2004 for the case of Japan, Mexico and the United States. Subsection three will discuss the level and characteristics of social capital within these three countries. In subsection four, finally, the relationship between happiness, income and generalized trust will be discussed.

4.1 Résumé

(31)

three has been entirely devoted to this interplay between social relationships and happiness. It has been put forward that a primary focus on gaining (more) income can eventually decrease happiness, because time is allocated towards working in stead of interacting socially. Research has also found evidence that generosity – as a proxy for altruism – is closely related to happiness. Because social relationships and the way in which people trust each other are closely intertwined and since data on social interactions are still scarce, this paper will use the broader and more commonly researched topic of social capital as a substitution for social relations. Within the concept of social capital, generalized trust has been documented for multiple countries for several years and therefore I will use trust as a proxy for the level of social capital. Since many authors have proposed the possible influence of cultural differences on social relationships and happiness

and because most happiness research so far has primarily focused on Western economies, this

paper will look into the matter for three different countries from three different cultural areas: Japan, Mexico and the United States.

4.2 Happiness levels and real income per capita in Japan, Mexico and the United States

This subsection will relate the levels of average happiness for each of the countries to the pattern in real GDP per capita over a number of years. All of the data on real GDP per capita are subtracted from the Penn World Tables, while the data on happiness are obtained from the World Database of Happiness7. For all of the countries real GDP per capita is measured in US dollars. Average happiness reflects the replies in questionnaires to the question: “In general, how happy would you say you are?”. For reasons of comparability, all answers have been converted to 1 to

10 scales (ranging from not at all happy to very happy). I have specifically chosen to incorporate

only those data sets that measured happiness and not the ones measuring life-satisfaction, to prevent the problematic issues mentioned earlier in this paper. I also used the same data set for all three countries to eliminate further measurement problems.

(32)

Happiness and real income per capita in Japan

For Japan, happiness data was not available for every year in the period 1975-2004, but in recent years the number of researches conducted has been rising. From figure 2 it follows that Japan has experienced an enormous increase in real GDP per capita. But at the same time, average happiness levels show only a minor improvement over the years. For the last years in this sample, average happiness is even declining (but this result needs to be validated in a larger time series research).

Figure 2: Happiness and real per capita income in Japan for the period 1975-2004

Source: World Database of Happiness and Penn World Tables

Happiness and real income per capita in Mexico

For the case of Mexico, problems arise when only data from happiness surveys are incorporated in stead of questionnaires on both happiness as well as life-satisfaction. As said, however, for the sake of clarity I will only use happiness data even though this implies a limited data set, because only a few of this kind of researches have been carried out in the period 1975-2004. From these, however, some careful conclusions can be derived. From figure 3 it can be found that real GDP per capita has increased by some thirty percent and that average happiness has been slightly

1000 6000 11000 16000 21000 26000 31000 1975 1980 1985 1990 1995 2000 Year R eal GD P per c apit a ($) 0 1 2 3 4 5 6 7 8 9 A verage H appines s

(33)

rising (with the notion that there are only a handful observations available). This result was to be expected for Mexico as a less developed country, because these nations tend to show a larger effect of income on happiness.

Figure 3: Happiness and real per capita income in Mexico for the period 1975-2004

Happiness and real income per capita in the United States

Happiness research has take place in the United States as early as 1946 and these researches became regular from the 1970’s onwards. This makes the US one of the most thoroughly documented countries in the world when it comes to happiness data. From figure 4 it can be seen that, although real GDP per capita has almost doubled in a period of thirty years time, average happiness lagged behind with most data points centred around an average of 7,2 on a scale of 1 to 10. 5000 6000 7000 8000 9000 1975 1980 1985 1990 1995 2000 Year R eal GD P p er c apit a ($) 0 1 2 3 4 5 6 7 8 9 Av erage H appine ss

(34)

Figure 4: Happiness and real per capita income in the United States for the period 1975-2004

All of the fitted trend lines for Japan, Mexico and the United States lie between an average level of happiness of 6 and 7,5. It follows then, that happiness shows no significant difference for countries with a relatively low level of GDP per capita. Nor is it higher in countries with high growth rates. So higher incomes, or higher growth levels do not provide that much more happiness. Or, looking at it from the opposite perspective, low incomes or low growth rates, are not equivalents for low levels of happiness. In the next subsection I will turn to see whether an explanation could be found in a (cultural) difference regarding social capital between the three countries.

4.3 Social capital within the Japanese, Mexican and American society

This subsection will look into the levels of generalized trust for the three cultural areas, as well as some of the characteristics of the social relationships within those countries. Generalized trust (as opposed to the also existing particularized trust) refers to the trust towards strangers of which no information is available from direct experience and is measured as the responses of people in

0 10000 20000 30000 40000 1975 1980 1985 1990 1995 2000 Year R ea l G D P p er ca p ita ( $ ) 6,5 7 7,5 8 8,5 Av erag e happ in es s

(35)

questionnaires conducted for the World Values Survey8 to the question “Generally speaking, would you say that most people can be trusted or that you need to be very careful in dealing with people?”.

Bjørnskov (2003) has investigated the relationship between happiness and social capital and found that at the national level social capital brings about more happiness (although he also – wrongly – uses happiness and life-satisfaction interchangeably) over and above the effects of economic conditions. Figure 5 shows this relationship between social capital and happiness for several countries of the world.

Figure 5: Social capital scores and public happiness

SWI LAT RUS BUL SLV ARG MEX POR TRA IT CZ HUN JAP AUS BEL GER UK IRE USICE DK CAN NOR 0 10 20 30 40 50 60 70 80 90 100 -2 -1,5 -1 -0,5 0 0,5 1 1,5 2 2,5 3

Social capital score

pu bl ic h appi nes s

Source: Adapted from Bjørnskov (2003)

Rahn and Transue (1998) have suggested that widespread social trust can be regarded as an important element of social capital. In their view, social, or generalized, trust is a standing decision to give most people – even the ones you don’t know from direct experience – the benefit

(36)

of the doubt. This subsection will therefore use generalized trust as a proxy for social capital and for that matter, for social relationships as well.

Social capital within the Japanese society

In Lu and Gilmour (2004) the Asian culture is described as a culture of collectivism. Individualism and collectivism have been the topic of many researches in different fields of sciences, and the dimension between the two is one of the important aspects to distinguish between cultures. Individualism has been defined by Hofstede (1980) as an emotional independence of groups, organisations and other collectivities, while collectivism focuses on the group and the social interdependence that comes along with it. The Asian ‘self’ is described by Lu and Gilmour as a ‘connected, fluid, flexible, committed being who is bound to others’. The characterization of the self is an inherent part of his or her social relationships. Within this collectivist society, the social customs, institutions and media all promote the relationship with others. The greatest concern of the Asian self is to achieve socially and culturally desirable goals. Uchida et al (2004) also discuss the construction of the self within Asian cultures. For most of these cultures (among which Japan) they find that the self in relationship with others is the center of all thoughts, actions and motivations. The boundary between the self and others is thin in these cultures, and is subjected to a continuous negotiation within their social relationships. Thus making the self interdependent of others. Within the norms of these social relationships, individuals are constantly motivated to commit to the social roles, obligations and expectations of the interdependent groups to which they belong.

(37)

Tokuda and Inoguchi (2008) have investigated the relationship between this interpersonal mistrust on the one hand and unhappiness among Japanese citizens on the other. They define mistrust as ‘the cognitive habit of interpreting the intentions and behaviour of others as dishonest, unsupportive and self-seeking’. It is based on a belief that other people are merely seeking to pursue their own goals and that they might even do so at the expense of others. In researches the Japanese society has been found to show lower levels of interpersonal trust than other countries and it is said to be one of the most unhappy countries of the industrialized world. While at the other hand, Japan also has a high degree of income equality and belongs to the richest countries in the world. Using data from the Asia Barometer Survey for the period 2003-2006 Tokuda and Inoguchi find that 55 percent of the participants answered that ‘you can’t be too careful in dealing with people’. After adjustments for other important factors like marital status, income, and self-rated health, the authors conclude that interpersonal mistrust is more common among people who are unhappy. This significant relationship between mistrust and unhappiness is, according to Tokuda and Inoguchi, one of the reasons for the fact that Japanese are among the most unhappy people within the developed world.

Figure 6: Trust in the Japanese society

0 10 20 30 40 50 60 70 1981 1982 1990 1995 1996 1999 2000 Year P er cent age

Most people can be trusted Can't be too careful

(38)

Figure 6 shows the levels of trust within the Japanese society for the period 1981-2000. It can be seen that over the years, the majority of the people who were questioned believed that most people cannot be trusted.

Social capital within the Mexican society

Shkodriani and Gibbons (1995) have measured the levels of individualism and collectivism of university students from Mexico and the United States. And although thorough investigations of the differences among Latin American nations were unavailable for their research, Shkodriani and Gibbons do point to researches that have found that Latin American societies in general can be regarded as collectivist societies. This also holds for Mexico, which has been found to have a far less individualistic tendency than the United States. The Mexican society can be distinguished as one in which family and solidarity are highly valued, and where the interdependence within generations is still far more important compared to the United States’ customs of self-reliance. For their research Shkodriani and Gibbons questioned students from colleges in Mexico and the United States. They found that, in general, Mexican students are far more collectivist than American students are. Furthermore, their results reveal that parents, spouse and classmate relationships are much stronger among Mexican students.

Diaz-Loving and Draguns (1999) find that in Mexico social relationships are dictated by gender roles and social hierarchies and that Mexicans prefer groups over individuals. Other researches have established that family constitutes an important fundament in Mexican society, and that status is based on who a person is rather than on what a person has. One of the basic purposes of

interpersonal relationships (especially within the family) is to make other people happy. In this, Mexicans focus more on the needs of others than Americans, who are much more self-interested. With regard to the community as a whole, Mexicans perceive society as a union of people linked together by interpersonal relationships.

(39)

Figure 7: Trust in the Mexican society

0 10 20 30 40 50 60 70 80 90 1981 1982 1990 1995 1996 1999 2000 Year Pe rc en ta g e

Source: World Values Survey

Social capital within the American society

In a paper by Lu and Gilmour (2004) the Euro-American culture is defined as a culture of individualism. The Western self is viewed as a ‘bounded, stable, coherent, autonomous and free entity’. All social customs, institutions and media are determined by the free and individual will. In some researches the American culture has even been described as a culture of ‘rugged’ individualism, which promotes the pursuit of individual happiness and personal successes more than it values interpersonal relationships.

(40)

would theoretically foster these kinds of memberships. Putnam therefore concludes that American social capital measured by civic associations has drastically decreased within the time frame of one generation.

In a response to Putnam’s (1995) paper, Rahn and Transue (1998) have investigated the attitudes of young Americans towards social trust. They analysed data from a project called Monitoring the Future, in which American high school seniors participate annually. The authors find that increasing materialistic values have decreased the levels of social trust for young Americans: over the years, the number of youngsters claiming that having a lot of money is the most important goal in life has drastically increased.

The level of trust within the American society over the years can be found in figure 8. As a first result it can be seen that the percentage of respondents stating that most people cannot be trusted has clearly increased over the years. While at the same time, the percentage of people that do trust others is declining. Making the level of mistrust in the American society larger.

Figure 8: Trust in the American society

0 10 20 30 40 50 60 70 1981 1982 1990 1995 1996 1999 2000 Year P er cent age

(41)

4.4 Discussion

In this subsection the results of the previous two subsections concerning average happiness levels, growth in real GDP per capita and the levels of generalized trust for Japan, Mexico and the United States will be related to each other. Given the small number of observations, it makes little sense to compute regression results. For that reason I have composed a scatter plot showing the different relationships between happiness and trust for the three nations over time.

Figure 9: Average happiness and generalized trust in Japan, Mexico and the United States

Mexico 1990 Mexico 1996 Mexico 2000 Japan 1981Japan 1990 Japan 1995 Japan 2000 USA 1982 USA 1990 USA 1995 USA 1999 6 6,5 7 7,5 8 8,5 20 25 30 35 40 45 50 55 Generalized Trust A ver ag e H a p p in ess

Source: World Values Survey and World Database of Happiness

(42)

Strengthened by the literature on the relationships between social capital and happiness, I feel that indeed a loss in social trust (and by that a decline in strength of social relations) could be one of the factors causing the paradox of happiness.

As for the case of Mexico, the results seem to tell a different story. Real GDP per capita has been increasing over the past years, but not as much as in Japan or the United States. The level of Mexican income lies far below those of the other two countries, and Mexico can therefore still be regarded as a lesser developed country. From the literature study in this paper, it then follows that for Mexico different rules of engagement are applicable. In poorer countries happiness does seem to be strongly related to rises in income, and that is exactly what can be found for the results obtained for Mexico in this paper. Although generalized trust has also been declining in Mexico, the effects of it on the levels of average happiness might be overcome due to the positive effects of the growth in income.

One striking result of this analysis is the fact that in all three cultural areas (either individualistic or collectivist) the levels of generalized trust have been declining in recent years. A possible explanation for this could in my opinion be the fact the economic growth – and with that changing consumption behaviour and increasing financial independence – changes societies into more individualistic nations. Since the United States were already an individualistic society, the effects of this on the levels of mistrust might be stagnating. But for Japan as well as for Mexico the difference between people who do not trust others and people that do, is growing. Since the levels of trust and the strength and size of social networks could very well be related, this could imply that these cultures might be slowly losing their sense of collectivity and are involving into more individualistic societies as a result of growing wealth. This reasoning, however, needs to be verified in more elaborate researches including more nations over a larger time period.