‘‘I Think Competition is Better Than You Do: Does It
Make Me Happier?’’ Evidence from the World Value
Surveys
Juan Jose´ Barrios
1 Introduction
1.1 The Issue
Mainstream Economic theory and most Professional Economists postulate that competition drives the forces of development and improves economic well-being. To the extent that happiness is a measure of well-being,1 then competition and happiness should be positively associated. First, competition creates positive incentives for producers to boost technological progress, improve efficiency and optimize resource allocation, thus improving social welfare. Additionally, competition should improve consumers’ wellbeing by putting downward pressure on prices because consumers, for equal quality, should have more opportunities to buy cheaper products, kicking inefficient suppliers out of the market.2
Non-competitive structures, such as oligopoly, may not survive due to the incentives to free ride supply agreements (e.g. cartels). On the other hand, Monopolists, who charge higher prices and produce less than the optimal quantity of product, are threatened by governments when the latter show credible intentions to approve laws to deregulate and liberalize production and factor markets. When governments pass laws to foster competition, they may create the appropriate environments leading to economic growth and development. These considerations suggest that institutional environments aiming at creating more competitive market structures may lead to better economic outcomes. In particular, institutional forms such as competition may have different effects on well being in more developed countries relative to less developed countries (North 1990).
The theoretical arguments developed by mainstream economists about the benefits of market competition are strong and they seem to have percolated into the minds and souls of other social agents, such as politicians, but also appear to have strong influence on the general public.
In sum, most economists have developed theoretical arguments which show competition as favoring efficient outcomes, while they have also developed logical reasoning and opinions which ‘‘explain’’ why competition improves individual wellbeing. A logical presumption derived from these arguments is the following: individuals with more positive views or feelings about competition should report higher levels of happiness, ceteris paribus.
However, the study of competition and its consequences on human behavior and feelings takes place also outside Economics. Psychologists have been interested in this subject for quite a long time and provide us with examples of how competition affects behavior, for example, in the relation between males and females. For instance, Buss’s
¨derogation of competitors¨ shows us ways through which people use verbal tactics to denigrate same-sex rivals to make them less desirable (Buss and Dedden 1990). The effects on overall (i.e. aggregate) well-being are unclear since the consequence of this competition at the individual level is to generate winners and loosers with respect to the desired object.
Individual competition may enhance suffering (i.e. hurt well-being) on losers at the same time that it increases the well-being of those who win.
Additionally, in assessing the relation between competition and trust, some theorists differentiate between competition at the individual level (as in the above paragraph) and at the firm (or group) level (Johnson et al. 2013). Their argument posits that although competition may elicit greed at the individual level, it may create incentives towards cooperative behavior
(i.e. trust) among workers within a firm when competition with other firms is more intense.
The successful firms (i.e. survivors, a la Darwin) will be those where co-workers cooperate with and trust each other more, increasing firm efficiency and competitiveness. More competitive environments enhance cooperation among co-workers but less trust and cooperation towards out-workers. In this case, although competition may lead to pro-social attitudes within the group, it may at the same time lead to grotesque and savage attitudes towards those outside the firm or group, thus disastrous consequences for well being (Frank 2011).
2 In assessing the wellbeing effects of economic choices, economists have been traditionally led by the principle of revealed preference, by which if one observes that an individual chooses consumption bundle A over consumption bundle B, one assumes that the individual does so because he prefers bundle A over bundle B and in choosing bundle A over bundle B, presumably, the individual will maximize his wellbeing.
But this is a logical conclusion derived from appropriate assumptions and it is by no means clear if it constitutes a measure of individual wellbeing.
Last but not least, feelings about competition and its association with well-being may be influenced by cultural aspects which in turn affect economic outcomes. Economists are used to measure culture by considering the religious affiliation of individuals (Guiso et al. 2006).
In other words, some religious beliefs, such as Post-Vatican II Catholicism may affect economic outcomes because they are critical about cooperating with other individuals outside their own beliefs, while the opposite effect occurs with Protestantism which foster cooperative attitudes which lead to better economic outcomes through a more positive trustworthy attitude towards others (Putnam 2000; Daniels and von der Ruhr 2010).
One interesting question I try to answer in this paper deals with to what extent different opinions about competition affect well-being for individuals who have the same cultural
(religious) heritage? Also, are there any differences between e.g. Catholics and Muslims?
In sum, although economists tend to agree that competition favors well-being, psychologists and other researchers cast doubt on this positive relation. In particular, there may be gender and cultural differences that may be fostering negative consequences. This paper attempts to empirically investigate the relation between competition and happiness at the aggregate level while also considering the cases of different religions and gender.
1.2 Related Empirical Literature
Researchers have attempted to approximate individual wellbeing by collecting survey information on self-reported individual satisfaction and individual happiness (e.g. the World
Value Surveys) and by running experimental games to analyze behavioral outcomes.
On the other hand, the study of happiness, which many researchers use as an approximation to individual well-being, originally a domain of psychology, has been making its way into the field of economics over the past decades as researchers have used survey and experimental data to delve into the nature of the association between well-being and political and economic institutions, although some of these studies use self-reported life satisfaction as a measure of well-being (Bjørnskov et al. 2008).
Empirical studies which test the positive association between market competition and individual well-being postulated mainly by mainstream economists find this relationship questionable, particularly studies of behavioral economics, sociology, psychology and political science.
Empirical Psychological studies of children behavior, cited in Kohn (1992) found that children learn better when they are exposed to cooperative environments (65 % of the cases) with respect to competitive environments (7 %). Other studies also cited in Kohn
(1992) test for creativity of children and find that those children who were competing for prizes produced less creative collages than those who were not competing for prizes.
More closely related to Economics, Brandts et al. (2005) use laboratory experiments in which participants play a repeated social dilemma game (played by a fix group of subjects3 with fixed roles) to study the effects of competition on efficiency and material well-being, subjective well-being and individual’s disposition towards others.4
3 The game is played by two players in the Non-Competitive environment and by three players in the
Competitive environment. In the latter, one party has to choose with who of the other two players she will play, thus creating a competitive condition. Games are repeated over 30 rounds.
4 Subjective well-being is measured by computing self-reported hedonic states experienced by the participants, while disposition towards others is measured using a variant of a social value orientation test
(Liebrand 1984).
Subjective well-being and disposition towards others are obtained by surveying the participant’s emotions before and after the experiments. The authors consider specifically the fact that preferences and tastes are no independent of the institutional environment and that economic interactions are contractually incomplete. In an environment with incomplete contracts, they find that the competitive scenario neither leads to an increase in efficiency nor it leads to material gains to the short side of the transaction relative to the non-competitive scenario. Further, competition harms the subjective well-being of the long side. Only the subjective well-being of the short side to the transaction is improved. The short side to the transaction is significantly happier than the other two competing parties on the long side and is also happier than any of the two parties in the non-competitive game.
Moreover, competition appears to adversely affect disposition towards those on the long side of the transaction.
Although Brandts et al. (2005) nicely capture the rivalry aspects of competition (Stigler
1987), the sample they use appears to be non-representative: they collected data on 153 subjects and they do not specify in which way they selected the individuals.5
More related to the spirit of this paper, Fischer (2008) uses data from the third and fourth waves of the World Value Surveys (2013)6 and other sources and finds that competition increases happiness inequality by aggravating the harmful effects on inequality of differences in economic power,7 that is, competition makes those with greater bargaining power relatively happier relative to those with less bargaining power.
Fischer (2008) segregates the happiness-effect of competition in three parts: (1) the financial gains obtained through competition, (2) the intensity of market transactions, and
(3) the degree of bargaining power of the short side, that is, the happiness-empowering effect of having the power of excluding others on the long side from the economic transactions thus augmenting the latter’s’ economic insecurity. The third part may increase or decrease overall happiness depending on the magnitude of the effect on the short side relative to the long side of the transaction. Fischer (2008) hypothesis is that market competition re-enforces the bargaining power effect for participants’ subjective wellbeing
(her Hypothesis 1).
Her dependent variable arises from individuals’ answers to the question: ‘‘All things considered, how satisfied are you with your life as a whole these days?’’ The persons answering this question have 10 options, the first option being ‘‘dissatisfied’’, the tenth being ‘‘very satisfied’’. In addition, Fischer (2008) approximates market competition by using the KOF index of economic globalization, which measures the integration of a national economy to the world market. Fischer (2008) hypothesis is that the degree of market competition in a country can be approximated by the degree of economic integration of that country to the world economy.8 As such, it constitutes a national, aggregated data, which does not represent a measure of individual behavior or opinion.9
5 Since the experiments were run at the University of Amsterdam, the sample may well consist of university students. 6 The surveys cover more than 60 countries and collect information on some 80,000 individuals.
7 Economic or bargaining power is measured by the absolute self-reported income level of each individual.
8 Since Fischer (2008) is interested in the effect of an individual bargaining position on happiness inequality, she interacts her self-reported satisfaction with self-reported (absolute) income level, also from the WVS.
9 Nevertheless, since the KOF index may be correlated with economic growth or income inequality, Fischer
(2008) uses the predicted residuals of a regression of the KOF index on GDP.
Fischer (2008) does not use self-reported happiness as her dependent variable but selfreported satisfaction. Although both are positively correlated,10 they probably do not mean the same thing (Frey and Stutzer 2000). Also, since her measure of market competition is aggregated at the country level at two different points in time, in order to obtain the same number of observations at the individual level, she needs to impute that same measures to all individuals in that country at that moment of time. This may not represent a true measure of market competition, at least does not represents what each individual thinks about competition and it would be probably better to treat it as a country fixed effect.
On the other hand, Fischer (2008) does not show that being more open to international markets must be necessarily associated with increased competition, which may be true for small economies, but not necessarily for larger ones: in fact, more protection may increase the actual number of local competitors. Finally, to assess the effect of market competition through bargaining power, she also imputes the (country) index of market competition to each individual income (which approximates each individual’s bargaining power).
1.3 Object and Contributions of This Paper
This paper studies the association between feelings about competition and self-reported happiness. I do not attempt here to draw any causal relations between competition and happiness, although I do performed a related analysis elsewhere.11,12
Nevertheless, this paper contributes to the literature on the relation between economic institutions and happiness by focusing on one specific economic institution: market competition. I use self-reported opinions about competition derived from the World Value
Surveys (2013). To my knowledge, this is the first paper to use the opinions of individuals about competition rather than using quantitative aggregate measures such as the degree of openness of an economy or the volume of capital inflows and outflows. As a consequence, my measure is an individual measure rather than an aggregate measure which may prove more useful to drawing appropriate conclusions with respect to the association between competition and individual wellbeing.
Although most studies surveyed above control for gender differences in the effect of competition on happiness, none of them assesses whether there are differences within women and men, and if competition affects men’s happiness different than it does with women’s. That is: do women (men) who like competition more declare to be more or less happy than women (men) who like competition less? In other words, are gender specific differences in their self-reported happiness for different appreciations about competition?
As another contribution to the literature on evolutionary theory mentioned above, this article focuses on the gender differences of the association between views about competition and happiness.
Additionally, none of the papers surveyed above considers the issue of how individuals with the same cultural heritage (measured as religious beliefs) assess the happiness consequences of different views about competition. And most importantly, are cultural beliefs relevant to the association between competition and happiness? Does self-reported
10 Their pairwise correlation is low. 0.475.
11 I have used Instrumental variables to gauge into the potential causal relation stemming from competition to happiness. The analysis is available from the author upon request. .
12 First, one may argue that happier individuals may regard competition as a good thing but also argue that more intense competition may make individuals happier, as most economists do. Second, data collected from the World Value Surveys (2013) are subjective and may not represent actual behavior. happiness change for different views of competition within Christians? If so, can we find a similar effect for other religions? Following the tradition on the cultural influence on economic outcomes mentioned above (Guiso et al. 2006), this paper also improves on the existing empirical literature by considering the relevance of cultural variables on the association between competition and happiness.
Using the World Values Surveys (WVS) improves over the Brandts et al. (2005) study in that the WVS build on representative samples and avoids the problem of self-selection typical of experimental studies. It also improves over the Fischer (2008) study in that my paper considers a subjective opinion about whether an individual thinks market competition is good or harmful (see below) which, as argued above, constitutes an individual measure and it may also be considered an ex-ante opinion, independent of the actual competitive environment derived from any aggregated approximation to competition, such as the KOF globalization index (KOF index of Globalization 2013).
Finally, competition means different things for different scholars. Mainstream economists conceptualize competition as an ‘‘end-state’’: competitive markets should achieve efficient social outcomes. The opinions collected in the WVS, however, may not coincide with the economist’s vision of competition in that it may be representing a¨process¨ (Blaug
2001) in which firms attempt to maximize their stake of the market, sometimes achieving a zero-sum outcome: what one firm gains, other firm loses. That process may lead to satisfactory outcomes, e.g. lower prices, but may also lead to higher unemployment, lower quality products, or what is commonly denominated a ‘‘race to the bottom’’. Under this second view competition may drive firms to undertaking unfair, unjust and environmentally damaging strategies in order to get a larger share of the market,13 thus a bad thing
(Hahnel 2011). I postulate that this process-view of competition, expressed by the answers collected in the WVS, is the view of what ordinary people do understand by competition
(more below).
If people’s views about competition are consistent with that of mainstream economists,
I should obtain a direct and positive relation between people’s feelings about competition and self-assessed happiness. On the contrary, if the process view of competition prevails, mixed results should obtain, that is, the relation between people’s feelings about competition and happiness could be non-linear, for example. One can conjecture that there may be a ‘‘competition threshold’’, beyond which, negative views of competition may actually be positive for well-being.
Section 2 describes the methodology used in the paper: the data used, and the intuition behind the econometric model employed. Section 3 shows the results of the investigation.
Section 4 concludes and discusses results, limitations of the study, and suggests guidelines for future research.
2 Methods
2.1 Data
I consider data collected in the fourth wave (2005–2008) of World Value Surveys (2013).
The WVS periodically collect self-reported opinions and beliefs about cultural values of
13 These strategies may include deceiving costumers through advertising, for example. Some critics of corporate global capitalism have also argued that multinationals foster environmentally unsustainable growth strategies, which harm us all. representative samples of individuals over dozens of countries around the world. The fourth wave collected the opinions of more than 60.000 individuals from 56 countries.
Among other things, individuals are asked about their perceptions of life, which includes self-assessments of happiness. They are also asked about politics and society in general including a question about what they think about competition. The surveys also collect socio-demographic characteristics of each individual, including sex, income and religious beliefs. As my dependent variable, I use the WVS question about the individual’s state of happiness which arises from the answers to the following question: ‘‘Taking all things together, would you say that you are (1) very happy, (2) quite happy, (3) Not very happy,
(4) Not at all happy?’’, thus a categorical variable which takes four values. This variable is ordered in the sense that each category represents a level of happiness that can be compared with the preceding and following category: when the individual’s answer falls in category 3, such person is less happy than (some) other individual whose answer fell in category 2 but happier than other individual whose answer fell in category 4.
Market competition is approximated by computing the views individuals have about competition. Specifically, individuals are asked the following question: ‘‘How would you place your views on this scale? 1 means you agree completely with the statement on the left; 10 means you agree completely with the statement on the right; and if your views fall somewhere in between, you can choose any number in between. Sentences: Competition is good. It stimulates people to work hard and develop new ideas versus Competition is harmful. It brings the worst in people’’.
Briefly, should a person chooses option 1, it means she believes competition is a good thing, while if she chooses option 10, it means she believes competition is harmful. In sum, competition is also a categorical variable that takes 10 values, from 1 (good) to 10
(harmful). Moving downwards from the first category, each subsequent category represents less sympathy towards competition.
The fourth wave also collects socio-demographic data of each individual some of which
I use as control variables. These include whether the respondent is female or male, his/her years of age, self-reported education level, employment status, income level and social class. Education is a categorical variable ranging from level 1 (no formal education) to level 9 (university education with degree), level 1 being the omitted (reference) category.
As previous empirical studies show, more educated individuals tend to be happier (e.g. Di
Tella et al. 2001).
Employment status is a non-ordered categorical variable since not all the categories imply that the next (previous) category is better (worse). The respondent is given 9 alternatives to answer with respect to her employment status (See Table 1 for details): there is no ex-ante reason to believe that a house wife would be more or less happy than a full time employee or a student. If we assume that having a job (or else, doing something relative to being unemployed) makes a person happier than not having a job, which appears reasonable, there would be reasons to believe that the level of happiness of all categories would be higher than that reported by the unemployed as studies on the relation of unemployment and happiness suggest (e.g. Frey and Stutzer 2002). I will use the latter as the reference category, accordingly.
Self-reported income is measured using deciles, the first representing the lowest income.
I use this lowest income group as the reference category. Within a country, evidence shows that higher income individuals are happier. International comparisons, however, have shown that the average level of happiness does not change much with respect to the average level of income per person, which has been delved the Easterlin Paradox (Easterlin
1974). Additionally, individuals report as belonging to one of five social classes, the first being the upper class. I use this latter category as the reference category. The descriptive statistics of these control variables are shown in Table 1.
Finally, data also contains information on each individual religious denomination which
I use as a proxy for cultural differences. I split the samples on 7 different religious denominations: Protestants, Orthodox, Muslim, Hindu, Evangelist, Buddhist and Catholic
Christians. Individuals within these religious denominations comprise more than 87 % of the observations. I disregarded minor denominations to avoid the small sample problem. I run seven different regressions, one for each denomination and analyze the nature of the association of feelings about competition and happiness for individuals belonging to each religious denomination.
2.2 The Model
Self-reported happiness is a categorical variable that takes 4 values as described above.
Individuals answer according to how happy they regard themselves. In other words, responses are ordered: here a person selecting option 1 declares to be happier than a person who selects option 2 and so on. As Train (2009) explains, one way to conceptualize this process of decision is to think about some level of opinion or utility associated with the answer given. That is, a person whose (unobserved) opinion about happiness is above some level Z1 will choose to answer ‘‘very happy’’ and a person whose opinion about happiness is below Z1 but above Z2 will choose to answer ‘‘quite happy’’, those whose opinions are below Z2 and above cutoff point Z3 will answer ‘‘not very happy’’, while those with opinions below Z3 will respond ‘‘not at all happy’’.
This unobserved level of opinion or utility associated with happiness is affected by observed and unobserved variables as specified next. Assuming a specific distribution for the unobserved variables (e.g. logistic), the probability of each answer for the level of happiness can be determined. The estimated parameters approximate the quantitative association of the explanatory variables on self-reported levels of happiness. If the model uses the logistic distribution for the unobserved variables, this model is called ordered logit, and it is the one I use in this paper.
Following the intuition outlined above, my econometric model specifies individual’s
‘‘i’’ self-reported happiness (Happyi) as a function of how individual ‘‘i’’ feels about competition (Compi), other observed socio-demographic variables (Xi), country fixed effects and other unobserved variables. This relation can be expressed as follows:
Happyi ¼ a Compi þ b Xi þ ei where ei is an individual-specific error term which is assumed to be distributed logistic.
Since the dependent variable is categorical, OLS results may be biased and inefficient.
Nevertheless I first run OLS since I wish to gauge for any non-linear relation between competition and happiness. Following the OLS regression, I apply an order logit technique to take account of the nature of the variables involved. Since the results of this type of regressions are difficult to evaluate, I analyze the marginal effects of competition on the probability of being very happy (value = 1), which also will help in the evaluation of nonlinearities.
Control variables include self-reported income, social class, educational level, gender, age and country fixed effects as outlined in the preceding section and described in Table 1.
To analyze gender and cultural effects, I use the same model described in this section but I split the samples to consider women and men separately when analyzing gender relevance, and use 7 different samples, one for each selected religious affiliation.
Table 1 Descriptive statistics
Variable Description Observations Mean SD Minimum value Maximum value
Happiness Answer to question: Taking all things together, would you say that you are (1) very happy, (2) quite happy,
(3) not very happy, (4) not at all happy?
66,610 1.9134 72,683 1 4
Competition Answer to question: How would you place your views on this scale? 1 means you agree completely with the statement on the left; 10 means you agree completely with the statement on the right; and if your views fall somewhere in between, you can choose any number in between. Sentences: competition is good. It stimulates people to work hard and develop new ideas versus competition is harmful. It brings the worst in people
64,210 3.7532 0.4806 1 10
Competition 2 Square of competition 64,210 20.240 4.265 1 100
Female Category: 1 = male; 2 = female 67,222 1.5217 0.49952 1 2
Age Years 67,050 41.777 6.544 15 98
Education Category: 1: no formal education, 2: incomplete primary, 3: complete primary; 4: incomplete technical secondary; 5: complete technical secondary; 6: incomplete university secondary; 7: complete university secondary; 8: some university; 9: university with degree
66,794 5.1705 0.5093 1 9
Employment Category 1: full time; 2: part time; 3: self-employed; 4: retired/pensioned; 5: house wife; 6: student; 7: unemployed; 8: other
65,018 3.4319 0.2008 1 8
Social class Category 1: upper; 2: upper middle; 3: lower middle; 4: working; 5: lower
61,615 33.3763 99,982 1 5
Income Deciles 1; lower; 2: 2nd; 3: 3rd; 4: 4th; 5: 5th; 6: 6th; 7:
7th; 8: 8th; 9; 9th; 10: upper
60,541 4.5979 0.2781 1 10
3 Results
3.1 General Model
The results from the OLS regression are shown in Table 2. I check for robust standard errors. Results show a positive but nonlinear relation between feelings about competition and self-reported happiness. Although individuals who think competition is relatively good are associated with higher self-reported happiness (and those who think competition is harmful are associated with less self-reported happiness), the relationship appears to have a minimum,14 as shown by the negative coefficient of the square of competition.
Figures 1 and 2 below show some support for this preliminary result. These graphs plot the predicted levels of happiness (vertical axis) against the opinions about competition
(horizontal axis). Those who feel competition is good are, on average, as happy as those who feel competition is harmful. The inverted-U form of the relationship suggests that as individuals feel competition is less beneficial, they also regard themselves as less happy.
But this goes up to a point beyond which the relationship becomes negative: happier persons are associated with the view that competition as more harmful than beneficial.
Next, Table 3 shows the results of the ordered logit regressions. Country fixed effects are considered but not shown. Country fixed effects comprise variables such as the competitive environment of a country or its institutional structure. In this sense, the coefficients of competition (the a’s) reflect the individual association of feelings about competition and self-reported happiness.
Interpretation of the meaning of the coefficients is difficult, but a positive relation between stronger aversion to competition and less happiness appears to be supported. In other words, happier persons are associated with the view that competition is good rather than harmful. Observation of the magnitude of coefficients and significance levels, also suggest that, beyond some threshold, individuals who regard competition as less beneficial are also those who report higher levels of happiness. This nonlinear relationship is manifested by the fact that individuals who think competition is harmful self-report statistically similar levels of happiness than those who think that competition is beneficial.
Table 2 Competition and happiness-OLS regression
Happiness Coef. SE t P[t [95 % CI]
Competition 0.0344 0.0045 7.59 0.000 0.0255 0.0433
Competition square -0.0030 0.0004 -6.84 0.000 -0.0041 -0.0023
Female -0.0160 0.0061 -2.69 0.007 -0.0286 -0.0045
Age 0.0020 0.0001 10.59 0.000 0.0016 0.0024
Education 0.0019 0.0013 1.43 0.154 -0.0007 0.0047
Employment 0.0092 0.0014 6.37 0.000 0.0064 0.0121
Social class 0.0869 0.0038 22.50 0.000 0.0793 0.0945
Income level -0.0470 0.0016 -28.98 0.000 -0.0508 -0.0444
_cons 1.682 0.0273 61.58 0.000 1.6291 1.7362
Observations: 52.699; R2: 0.06; F(8, 52690) = 378.38; Prob[F = 0.0000; fixed effects: yes
14 Note that increasing values of both categorical variables, happiness and feelings about competition indicate lower happiness and less appreciation for competition, respectively.
Controls have the expected effects on happiness: individuals earning more money are associated with increased happiness in line with previous literature (Ferrer-i-Carbonell
2005; Brandts et al. 2005), as well as individuals of higher social classes. Also, more educated persons tend to be happier than less educated. The only unusual result is the relationship between older persons and happiness: they appear to feel lees happy than their younger fellows.
The analysis of marginal effects (Table 4) confirms the nature of the association between feelings about competition and happiness, The negative coefficients show that, compared to the reference group 1 (individuals who feel competition is good), becoming an individual who likes competition less is associated with less happiness, except for the last representative individual, that is, the one who believes competition is harmful, who shows no statistically significant difference in self-reported happiness with respect to group 1.
1.9
1.9
1.94
Happier
Less Happy
Competition 2 4 6 8 is good
Competition
Is Harmful
Happiness
Feelings about Competition
Fig. 1 Association between feelings about competition and happiness
1.88
1.9
1.92
Less Happy
Happier
Happiness
Competition 2 4 6 8 is Good
Competition
Feelings about Competition is Harmful
Fig. 2 Association between feelings about competition and happiness
Table 3 Competition and happiness-ordered logit regression
Dependent variable happiness Coefficient Robust SE Z P[z [95 % CI]
Competition is almost good 0.1587 0.030 5.16 0.000 0.0984 0.2189
Competition 3 0.2530 0.028 8.76 0.000 0.1964 0.3097
Competition 4 0.3270 0.030 10.64 0.000 0.2668 0.3873
Competition 5 0.1996 0.029 6.84 0.000 0.1424 0.2569
Competition 6 0.2507 0.037 6.6 0.000 0.1762 0.3252
Competition 7 0.2298 0.042 5.35 0.000 0.1455 0.3140
Competition 8 0.2496 0.047 5.27 0.000 0.1568 0.3424
Competition 9 0.1641 0.067 2.44 0.015 0.0322 0.2960
Competition is harmful 0.0160 0.059 0.27 0.787 -0.100 0.1322
Female -0.0123 0.018 -0.66 0.509 -0.0491 0.0243
Age 0.0065 0.000 8.58 0.000 0.005 0.0080
Incomplete primary education -0.0683 0.052 -1.3 0.193 -0.1712 0.0344
Complete primary education -0.1421 0.047 -3.01 0.003 -0.2347 -0.0496
Incomplete technical secondary -0.1143 0.053 -2.12 0.034 -0.2200 -0.0086
Complete technical secondary -0.2033 0.047 -4.28 0.000 -0.2965 -0.1101
Incomplete university secondary -0.1617 0.056 -2.89 0.004 -0.2715 -0.0519
Complete university secondary -0.2012 0.046 -4.3 0.000 -0.2930 -0.1094
Some University -0.0982 0.055 -1.76 0.078 -0.2074 0.0109
University w/some degree -0.1620 0.049 -3.28 0.001 -0.2589 -0.0651
Employment: full-time 0.0717 0.037 1.91 0.057 -0.0020 0.1455
Employment: pert-time 0.0066 0.030 0.21 0.830 -0.0537 0.0670
Employment: self-employed -0.0492 0.035 -1.4 0.160 -0.1178 0.0194
Employment: retired -0.1851 0.032 -5.72 0.000 -0.2485 -0.1217
Employment: housewife -0.1189 0.038 -3.06 0.002 -0.1951 -0.0426
Employment: student 0.2714 0.037 7.29 0.000 0.1984 0.3443
Employment: other -0.0290 0.055 -0.52 0.602 -0.1381 0.0801
Social class 2: upper middle 0.3281 0.087 3.74 0.000 0.1560 0.5001
Social class 3 0.5557 0.087 6.35 0.000 0.3841 0.7274
Social class 4 0.6142 0.088 6.93 0.000 0.4404 0.7879
Social class 5: Lower 1.0921 0.092 11.83 0.000 0.9112 1.2730
Income 2: second lower -0.1516 0.045 -3.32 0.001 -0.2410 -0.0621
Income 3 -0.2681 0.042 -6.24 0.000 -0.3523 -0.1839
Income 4 -0.4396 0.043 -10.15 0.000 -0.5245 -0.3547
Income 5 -0.6106 0.042 -14.44 0.000 -0.6935 -0.5277
Income 6 -0.7226 0.044 -16.15 0.000 -0.8103 -0.6349
Income 7 -0.8128 0.046 -17.33 0.000 -0.9047 -0.7209
Income 8 -0.9639 0.051 -18.65 0.000 -1.0652 -0.8626
Income 9 -0.9708 0.069 -13.98 0.000 -1.1070 -0.8347
Income 10: upper -0.9189 0.071 -12.79 0.000 -1.0597 -0.7781
Number of obs = 52,699; Wald Chi2(82) = 7,625.02; Prob[Chi2 = 0.0000; log pseudolikelihood =
- 51,474.486; pseudo R2 = 0.075; fixed effects: yes
Starting from category 1, the probability of feeling very happy if one person likes competition a bit less (that is, moving from category 1 to category 2) decreases by almost
3 % and it is statistically significant. The probability of being very happy decreases for subsequent categories up to category 5, beyond which it increases (i.e. it is less negative).
We therefore reach a situation where the effect of moving from category 1 to the second is almost identical as the effect of moving from category 1 to category 9. In other words, the effect of moving one step further (feeling that competition is harmful) is statistically insignificant with respect to the first category, meaning that those who really like competition and those who really dislike it, show no significant differences in their selfreported levels of happiness.
This finding is different than that observed by Fischer (2008), who reports a positive relation between competition and happiness but does not report a nonlinear relation among the variables. Moreover, her study does not analyze a direct relation between competition and happiness (more below).
The general findings of this paper are also different to those of Brandts et al. (2005) who report a negative relation between competition and happiness (at least for some parties involved in the transaction). Although my general conclusion is positive, there are negative consequences too, since, as noted above, individuals with greater aversion to competition report higher levels of happiness.
To sum up, my results challenge the mainstream view that competition is always a good thing and gives support for an alternative view that people may feel happier with less competition, probably because they view competition as a harmful processwhere a fewwinners win at the expense of a majority of losers, hurts the environment and produces inefficient results.
3.2 Gender Differences
When it comes to gender differences, a first approximation indicates that there may not be significant differences with respect to the association between competition and happiness: data from the WVS considered in this paper show that 34 % of males who feel competition is good self-report as very happy compared to an almost identical 33 % of females.
Tables 5 and 6 show the results for the association between feelings about competition and self-reported happiness for females and males, respectively. Results confirm our first
Table 4 Competition and happiness. Marginal effects on different feelings about competition on the probability of being very happy. General model
Variable dy/dx SE z P[z [95 % CI] X
Competition is almost good -0.0287 0.0054 -5.32 0.000 -0.0393 -0.0181 0.1289
Competition 3 -0.0451 0.0049 -9.18 0.000 -0.0548 -0.0355 0.1498
Competition 4 -0.0572 0.0050 -11.39 0.000 -0.0671 -0.0474 0.1223
Competition 5 -0.0360 0.0050 -7.1 0.000 -0.0459 -0.0260 0.1535
Competition 6 -0.0442 0.0063 -7.01 0.000 -0.0566 -0.0318 0.0660
Competition 7 -0.0406 0.0071 -5.66 0.000 -0.0547 -0.0265 0.0480
Competition 8 -0.0439 0.0078 -5.62 0.000 -0.0592 -0.0286 0.0394
Competition 9 -0.0294 0.0115 -2.55 0.011 -0.0521 -0.0067 0.0206
Competition is harmful -0.0029 0.0110 -0.27 0.786 -0.0245 0.0185 0.0322
Omitted category: competition is good
*P\0.01; **P\0.05; ***P\0.10 approximation of the last paragraph: there are no significant gender differences as to how competition is associated with happiness. For both, lower levels of happiness are associated with greater dislike for competition, reproducing the general pattern observed in the general model.
Results show small differences between females and males who think competition is
‘‘almost’’ harmful (category 9 in Tables 5, 6): here males are statistically as happy as males who think competition is good while females are significantly less happy.
3.3 Cultural Differences
Figure 3 plots, for the seven different religions considered, the percentage of individuals declaring to be very happy and who feel competition is good. At first sight, there are
Table 5 Competition and happiness. Marginal effects of different feelings about competition on the probability of being very happy. Females
Variable dy/dx SE z P[z [95 % CI] X
Competition is almost good
-0.0246696 0.00789 -3.12 0.002 -0.040143 -0.009197 0.121088
Competition 3 -0.0440048 0.00699 -6.29 0.000 -0.057713 -0.030297 0.14869
Competition 4 -0.0636608 0.00702 -9.07 0.000 -0.077419 -0.049902 0.124487
Competition 5 -0.0394254 0.00706 -5.59 0.000 -0.05326 -0.025591 0.162177
Competition 6 -0.0491006 0.00868 -5.66 0.000 -0.066112 -0.032089 0.069098
Competition 7 -0.051781 0.00981 -5.28 0.000 -0.070999 -0.032562 0.049182
Competition 8 -0.0499538 0.01052 -4.75 0.000 -0.070571 -0.029337 0.041533
Competition 9 -0.0355983 0.01475 -2.41 0.016 -0.064515 -0.006681 0.022762
Competition
is harmful
0.0014962 0.01499 0.10 0.921 -0.027892 0.030884 0.035842
Omitted category: competition is good
*P\0.01; **P\0.05; ***P\0.10
Table 6 Competition and Happiness. Marginal Effects of different feelings about competition on the probability of being very happy. Males
Variable dy/dx SE z P[z [95 % CI] X
Competition is almost good
-0.0313957 0.00744 -4.22 0.000 -0.045987 -0.016804 0.13719
Competition 3 -0.0466561 0.00693 -6.73 0.000 -0.06024 -0.033073 0.150999
Competition 4 -0.0519194 0.00719 -7.23 0.000 -0.066002 -0.037836 0.120066
Competition 5 -0.0343623 0.00729 -4.71 0.000 -0.048654 -0.020071 0.144406
Competition 6 -0.0401712 0.00922 -4.36 0.000 -0.058244 -0.022098 0.06288
Competition 7 -0.0292377 0.01051 -2.78 0.005 -0.049845 -0.008631 0.04677
Competition 8 -0.0367138 0.01173 -3.13 0.002 -0.059708 -0.013719 0.037174
Competition 9 -0.0199223 0.01854 -1.07 0.283 -0.056257 0.016413 0.018412
Competition
is harmful
-0.0079264 0.01634 -0.49 0.628 -0.039955 0.024103 0.028476
Omitted category: competition is good
*P\0.01; **P\0.05; ***P\0.10 differences between religions: percentages go from only 18 % of Orthodox individuals reporting to be very happy up to 49 % in the case of Evangelists. Figure 3 gives us a crude approximation to cultural differences, but I need to control for other potential explanatory variables to reach more definite conclusions.
Accordingly, using religious affiliation as an approximation for cultural preferences,
Table 7 shows the results for the marginal effects on how self-reported happiness of individuals from different religious denominations is associated with how those individuals feel about competition. More specifically, for each religion, the table shows how a move from an individual who feels competition is good (the omitted category) to an individual who has a different feeling (less favorable) about the goodness of competition affects the probability of being very happy, controlling for other socio-demographic variables and country fixed effects.
I find significant cultural differences between individuals in the way competition is associated with happiness. Protestants and Roman Catholics appear to be driving the general results discussed above: a significant (but in general decreasing) positive relation between feelings about competition and self-reported happiness on the probability of being very happy. For Protestants who feel increasingly less compelled by competition, the probability of being very happy decreases within a range of 2 and 3.6 %. This probability decreases more for Roman Catholics than for Protestants., reaching a negative 6 % for those in the middle of the feelings-about-competition scale.
For both Protestants and Roman Catholics, individuals who think competition is harmful are as happy as those who feel competition is good, showing the non-linear relation discussed for the general case. Moreover, although not significant, the sing of the relation changes for Protestants in the last category: their probability of being very happy increases. On the other hand, Muslims and Hindus who increasingly dislike competition are as happy as those who believe competition as being good, while some individuals with other religious backgrounds (Orthodox, Evangelist and Buddhist) are more or less as happy as their competition fans of their same religion.
4 Conclusions and Discussion
This paper investigates the relation between competition and happiness. I use individual measures of feelings about competition and self-reported happiness derived from the WVS.
This gives me a direct link between the two variables, which contrasts with Fischer (2008)
Fig. 3 Percentage of individuals who declare to be very happy and feel competition is good. By religious affiliation Table 7 Competition and happiness. Marginal effects of different feelings about competition on the probability of being very happy by religious affiliation
Variable Protestant Orthodox Muslim Hindu Evangelists Buddhist Roman Catholic dy/dx SE dy/dx SE dy/dx SE dy/dx SE dy/dx SE dy/dx SE dy/dx SE
Competition is almost good
-0.0226* 0.007 -0.0269 0.0188 -0.0089 0.0120 -0.0372 0.0658 -0.0257 0.0419 -0.0397 0.0344 -0.0404* 0.0130
Competition 3 -0.0216* 0.007 -0.0749* 0.0173 -0.0127 0.0113 -0.0362 0.0435 -0.0620*** 0.0363 -0.0166 0.0307 -0.0490* 0.0117
Competition 4 -0.0348* 0.006 -0.1089* 0.0176 -0.0147 0.0126 0.1086 0.0879 -0.0753** 0.0363 -0.0773* 0.0289 -0.0527* 0.0123
Competition 5 -0.0198* 0.007 -0.0487** 0.0206 -0.0128 0.0120 0.0126 0.0347 -0.0562 0.0362 -0.0233 0.0297 -0.0330* 0.0122
Competition 6 -0.0197** 0.008 -0.0012 0.0270 -0.0078 0.0160 -0.0334 0.0937 -0.0603 0.0399 -0.0687** 0.0304 -0.0625* 0.0138
Competition 7 -0.0361* 0.008 -0.0753* 0.0251 -0.0014 0.0192 -0.0185 0.0482 -0.1177* 0.0413 -0.0194 0.0369 -0.0207 0.0182
Competition 8 -0.0234** 0.011 -0.0286 0.0327 -0.0204 0.0195 -0.0373 0.1497 -0.0668 0.0459 0.0206 0.0435 -
0.0414**
0.0178
Competition 9 -0.0228 0.014 -0.0561 0.0380 -0.0357 0.0296 -0.0184 0.1347 -0.0138 0.0746 0.0942 0.0687 -0.0231 0.0255
Competition is harmful 0.0117 0.019 -0.0256 0.0390 0.0375 0.0289 -0.0312 0.0505 -0.0737 0.0494 0.0585 0.0775 -0.0328 0.0230
Omited category: competition is good
* P\0.01; ** P\0.05; *** P\0.10 who studies the effect of an aggregated measured of competition (KOF index of Globalization
2013) on happiness mainly through the relation between competition and income.
Although results do not support a direct negative relation I find support for a non-linear association: people with the highest aversion to competition report higher levels of happiness, suggesting that competition may be exerting negative effects on at least, some individuals. This finding is in line with those of Brandts et al. (2005) in their experiments with players, who suggest that under certain institutional environments, players experience negative emotions when competition rises, possible due to higher ‘‘social stress’’. Although my regressions do not control for specific institutional environments, they do consider country fixed effects, which partially address the issue.
It is also in line with evolutionary theory, e.g. Buss’s ¨derogation of competitors¨ hypothesis mentioned in the introduction: losers in a competition may suffer the negative consequences of that derogation leading to a loss of utility or well-being.
However, Buss’s hypothesis on the different effect of competition on men relative to women’s’ well-being is not supported: there are no significant well-being effects differences of competition when I run regressions for men and women. One reason for this apparent contradiction may be related with the measure of competition used: Buss and others use observed measures of competition while my study uses subjective opinions about it. The two may differ: I may actually be experiencing negative consequences due to market competition and may nevertheless have a positive view with respect to it (more below).
The Academic economic view postulates a positive relationship but researchers of other disciplines disagree. This paper finds, that, other things equal, feelings about competition are positively related to self-reported happiness, although at a decreasing rate. Since I take account of country fixed effects, I am also partially addressing (Fischer 2008) concerns about the potential effects of institutional environments and the nature of contracts, which are the fundamental assumptions the (economic) theoretical view of competition makes
(more below).
The finding that individuals with high aversion to competition report being as happy as those who like competition suggests that individuals regard competition more as a process than an equilibrium (efficient) ‘‘end-state’’. The process view considers producers as rivals attempting to obtain a larger share of the pie at the expense of other producers, a situation where efficient outcomes are not always achieved.15 However the positive but decreasing relationship between competition and happiness may well be describing both of those views about competition.
These findings are consistent (but not identical to) with the opinion of economic historian
Blaug (2001), a strong supporter of the so called ‘‘process-view’’ of competitive capitalism: The man-in- the-street favours capitalism because it is ultimately responsive to consumers’ demands, technologically dynamic and produces the goods that are wanted at low cost; of course, it also suffers from periodic slumps, more or less chronic unemployment even in booms, and frequently generates a highly-unequal distribution of income. Still, on balance the good outweighs the bad and without becoming Panglossian, he or she votes for capitalism – and so do virtually all economists.
One could conjecture about the behavior of individuals who increasingly dislike competition: because they see competition as a bad thing or because they have experienced
15 For a lucid review of both views, see Blaug (2001). the bad outcomes of competition [e.g. being on the short side in Brandts et al. (2005) experimental set up], they may shy away from it and may choose to live and work in less competitive environments, thus achieving a higher level of happiness. Moreover, too much competition may lead to situations where people are hurt in their self esteem and are prisoners of jealously to other persons’ success (Boehm and Lyubmirsky 2008).
Cornell’s Economist Robert Frank argues that the appropriate view of competition should be the one based on Darwin’s principles rather than on Adam Smith’s lines.
Essentially Smith argues that competition reveals good for society although each individual pursues only his own, limited interests. Darwin’s natural selection process argues that competition selects those who are more fit to it. The basic difference relies on the potential contradiction between individual and social outcomes: while competition may prove satisfactory for a few winners, it may result in frustration for a vast majority, the losers. Theoretically, mainstream economists have solved this potential problem by postulating appropriate compensations from the winners to the losers. Anyways, followers of
Smith argue that there is no contradiction, while Darwinists support the opposite view.16
In addition, the findings of this paper suggest that the pessimistic view of competition expressed by researchers outside the economic profession may be overstated, e.g. Kohn
(1992). Maybe what is driving these pessimistic results is the fact that their evidence relies on experiments on cooperative behavior where the subjects are only children, suggesting that competition affects only adults and not children.
Competition may indeed be one of the factors that affect the behavior and the feelings of adults significantly different than the behavior and feelings of children, but not just the only one. What Kohn (1992) shows is not that competition is necessarily bad, but that adult behavior is different than child behavior in relation to cooperation, which appears to be a different issue. Moreover, experimental studies on trust and ultimatum games show that adults trust and cooperate more than what is assumed by economic theory (Ca´rdenas et al.
2008a, b).
Last but not least, cultural differences appear to be important as a mediating structural factor. Following the lead of Guiso et al. (2006) my findings show some cultural traits may significantly matter for how individuals are affected by competition and as a consequence, for economic and social outcomes. Catholics and Protestants seem to be particularly sensitive to feelings about competition. Other cultures, however, appear to influence individuals in such a way that their happiness seems to be immune to how they feel about competition. This finding is particularly important in light of the potential consequences of the current ‘‘westernization’’ (globalization) of tastes and preferences around the world. To the extent that this process may lead to a significant spread of catholic and protestant beliefs, competition may no longer be indifferent to individuals, as this paper suggests.
This study has limitations. One is related to the data used: in their review of the literature, Frey and Stutzer (2002) argue that the usefulness of survey data in measuring behavioral outcomes is limited by measurement error and by the questionability of their behavioral relevance. The findings of this paper should be complemented by experimental studies which may attempt to measure the same variables used here.
Another limitation relates to the measure of competition used: I collect information on subjective opinions about competition, which may not represent the competitive environment of the location where the individual lives or the competitive position of the respondent. This issue is partially addressed by the inclusion of country fixed effects in the
16 In fact, Smith’s view of competition may be closer to Darwin’s than Frank suggests (see Blaug 2001). regressions, but a more direct measure of the competitive environment would be a nice improvement to the paper.
The approximation to cultural traits used in this paper can also be considered from a different angle. For example, one can differentiate between socialist (collectivist) and capitalist countries. The data of this WVS allows for the differentiation between OECD countries and non-OECD countries: preliminary results (not reported here) indicate no significant differences on the effects of feelings about competition between individuals from the two types of countries.
Other studies may improve on this paper by considering specific institutional environments, such as the degree of openness of the economy, the quality of public institutions, etc. Such environments may directly affect what different people think about competition and would allow reaching more specific conclusions about its relation to well-being.
The policy implications of this paper are tentative. On the one side, happier persons like competition more, but persons with the same level of happiness reject it as harmful. Competition may lead to a ‘‘race to the bottom’’ situation, where only some corporate interests are benefited, while the general public may not. Rough competition in the so-called labor market, both inside and outside firms, hurt rather than benefit workers, for example.
Since competition cannot be avoided in capitalistic societies, governments can actually manage institutional structures, as Fischer (2008) suggests. But the effect of competition may run deeper and just better (Blaug 2001) capitalistic institutional environments may not suffice. References
Bjørnskov, C. B., Dreher, A., & Fischer, J. A. V. (2008). On decentralization and life satisfaction. Economics
Letters, 99, 147–151.
Blaug, M. (2001). Is competition such a good thing? Static efficiency versus dynamic efficiency. Review of
Industrial Organization, 19, 37–48.
Boehm, J. K., & Lyubmirsky, S. (2008). Does happiness promote career success? Journal of Career
Assessment, 16, 101.
Brandts, J., Riedl, A., & van Winden, F. (2005). Competition and well-being. IZA discussion papers 1769.
Buss, D. M., & Dedden, L. (1990). Derogation of competitors. Journal of Social and Personal Relationships,
7, 395–422.
Ca´rdenas, J. C., Chong, A., & N˜ opo, H. (2008a). To what extent do Latin-Americans trust and cooperate?
Field experiments on social exclusion in six Latin American countries. Working paper 635, Inter-
American Development Bank, Research Department.
Ca´rdenas, J. C., Chong, A., & N˜ opo, H. (2008b). Stated social behavior and revealed actions: Evidence from six Latin American countries using representative samples. Working paper 634, Inter American
Development Bank. Research Department.
Daniels, J., & von der Ruhr, M. (2010). Trust on others: Does religion matter? Review of Social Economy,
68(2), 163–186.
Di Tella, R., MacCulloch, R. J., & Oswald, A. J. (2001). Preferences over inflation and unemployment:
Evidence from surveys of happiness. American Economic Review, 91(1), 335–341.
Easterlin, R. (1974). Does economic growth improve the human lot? Some empirical evidence. In
P. A. David & M. W. Reder (Eds.), Nations and households in economic growth: Essays in honor of
Moses Abramovitz. New York: Academic Press.
Ferrer-i-Carbonell, A. (2005). Income and well-being: An empirical analysis of the comparison income effect. Journal of Public Economics, 89, 997–1109.
Fischer, J. A. (2008). Is Competition good for trust? Cross-country evidence using micro data. Economics
Letters, 100(1),56–59.
Frank, R. H. (2011). The Darwin economy: Liberty, competition and the common good. Princeton, NJ:
Princeton University Press.
Frey, B., & Stutzer, A. (2000). Happiness, economy and institutions. The Economic Journal, 110(2),
918–938.
Frey, B., & Stutzer, A. (2002). What can economists learn from happiness research? Journal of Economic
Literature, 40(2), 402–435.
Guiso, L., Sapienza, P., & Zingales, L. (2006). Does culture affect economic outcomes? Journal of Economic
Perspectives, 20(2), 23–48.
Hahnel, R. (2011). Green economics: Confronting the ecological crisis. New York: M.E. Sharpe.
Johnson, D., Price, M., & Van Vugt, M. (2013). Darwin’s invisible hand: Market competition, evolution and the firm. Journal of Economic Behavior and Organization, 90:(Supplement 1–13).
KOF index of Globalization. (2013, May 31). KOF index of globalization. http://globalization.kof.ethz.ch.
Kohn, A. (1992). No contest: The case against competition. Revised Edition. Boston, New York: Houghton
Mifflin Company.
Liebrand, W. B. (1984). The effect of social motives, communication and group size on behavior in a n-person multi-stage mixed motive game. European Journal of Social Psychology, 14(3), 239–264.
North, D. C. (1990). Institituions, institutional change and economic performance. Cambridge: Cambridge
University Press.
Putnam, R. (2000). Bowling alone: The collapse and revival of American community. New York: Simon and
Schuste.
Stigler, G. (1987). Competition. In M. Milgate, J. Eatwell (Eds.), The new Palgrave: A dictionary of economics. London: The MacMillan Press Limited.
Train, K. (2009). Discrete choice methods with simulation (2nd ed.). Cambridge: Cambridge University
Press.
World Value Surveys. (2013). World Value Surveys. www.worldvaluesurveys.org.