This paper investigates the relationship between national identification and social welfare. Society is a direct democracy where representative rich and poor agents vote on their most preferred tax rate. Utilizing the economics of identity, agent utility not only depends on income but also on social identification. We show that national identification increases agent utility and social welfare when the rich implements the most preferred tax rate of the poor, who is also the median voter. Furthermore, using respondent level of happiness as a proxy for social welfare, we show that national pride has a positive impact on welfare through logit regression analysis.
Nationalism has acquired a bad rap in recent times. Some consider it a main source of xenophobia, while others see it as a tool used by majority cultures to forcibly assimilate minority ones in the name of nation-building [
Nonetheless, nationalism has its redeeming qualities. As an instance, national identity has aided states in overcoming colonialism and helped modernize the state [
This paper shows theoretically that national pride raises social welfare, which contributes to the literature where nationalism benefits society. We do so by utilizing the economics of identity where agents not only derive utility from material payoff, but from social identification as well. We support our results empirically through logit regression analysis, and show that national pride is positively correlated with happiness, which we proxy for welfare.
This paper is organized in the following manner. First, we discuss the theoretical model, which is an application of the economics of identity of Akerlof and Kranton [
The nation is a direct democracy with rich (r) and poor agents (p) where agents vote for their most preferred tax rate. Members in each social class are identical and the proportion of rich and poor agents are δ and 1 − δ , while the income of each rich and poor agent is y r and y p . We assume the rich is the minority ( δ < 1 / 2 ) and the rich’s income exceeds the poor’s ( y r > y p ). Let the nation’s average income be y ¯ , then y p < y ¯ < y r . If society implements flat tax τ , the posttax income of agent i ∈ { p , r } is
π i ( τ ) = ( 1 − τ ) y i + ( τ − τ 2 2 ) y ¯ (1)
where ( τ 2 / 2 ) y ¯ is the quadratic deadweight lost following Bolton and Roland [
π ¯ ( τ ) = ( 1 − τ 2 2 ) y ¯ (2)
To model national identity, we apply the economics of identity [
U i ( τ ) = π i + σ [ s + π ¯ ( τ ) ] − w d (3)
where σ ∈ [ 0,1 ] is agent “pride” from national attributes that provide positive utility, such as s > 0 , which is exogenous and summarizes non-material dimensions impacting the nation and π ¯ ( τ ) representing national economic strength. The coefficient w ∈ [ 0,1 ] is the level of “shame” agents suffer from perceived negative national attributes d > 0 , which is determined exogenously. We restrict σ + w = 1 , hence a rise in national pride σ causes shame w to fall.
Substituting (1) and (2) into (3) and noting that w = 1 − σ , the utility of a poor agent is
U p ( τ ) = ( 1 − τ ) y p + ( τ − τ 2 2 ) y ¯ + σ [ s + ( 1 − τ 2 2 ) y ¯ ] − ( 1 − σ ) d (4)
The poor’s most preferred tax rate is defined as the tax rate that maximizes the poor’s post-tax utility in (4). Since y ¯ > y p , the first order condition of the poor’s utility with respect to τ is
τ p = y ¯ − y p ( 1 + σ ) y ¯ > 0 (5)
which is decreasing in σ since
d τ p d σ = − y ¯ − y p ( 1 + σ ) 2 y ¯ < 0 (6)
Note that this outcome is identical to that in Shayo [
U r ( τ ) = ( 1 − τ ) y r + ( τ − τ 2 2 ) y ¯ + σ [ s + ( 1 − τ 2 2 ) y ¯ ] − ( 1 − σ ) d (7)
and since y r > y ¯ , the first order condition of this utility with respect to τ is
d U r ( τ ) d τ = − ( y r − y ¯ ) − ( 1 + σ ) τ y ¯ < 0 (8)
Hence, the most preferred tax rate of the rich is τ r = 0 and we state the equilibrium below, which says that society implements τ p .
Proposition 1 A direct democracy with identical poor and identical rich agents with most preferred tax rates τ p and τ r will implement τ p if the poor is the majority.
Proof: Since the poor is the majority and all poor agents are identical, the median voter is a poor agent. As only two policy choices τ p and τ r exist, the median voter theorem stipulates that society implements the most preferred tax rate of the poor.
We now define a social welfare function in the following to investigate the impact national identification on social welfare.
W ( τ ) = δ U r ( τ ) + ( 1 − δ ) U p ( τ ) (9)
The following proposition demonstrates the main result of this paper, which shows that when τ = τ p , welfare in a direct democracy is increasing in national identification.
Proposition 2 W ( τ p ) increases in national pride σ .
Proof: By the chain rule
d U i ( τ p ) d σ = ∂ U i ( τ p ) ∂ τ d τ p d σ + ∂ U i ( τ p ) ∂ σ
Since the first order condition of (4) implies that ∂ U p ( τ p ) / ∂ τ = 0 and because ∂ U p ( τ p ) / ∂ σ = s + ( 1 − τ p 2 / 2 ) y ¯ + d ,
d U p ( τ p ) d σ = s + ( 1 − τ p 2 2 ) y ¯ + d > 0 (10)
For the rich
d U r ( τ p ) d σ > 0 (11)
as d τ p / d σ < 0 in (6), ∂ U r / ∂ τ < 0 in (8)
and ∂ U r ( τ p ) / ∂ σ = s + ( 1 − τ p 2 / 2 ) y ¯ + d > 0 .
Hence
d W ( τ p ) d σ = ( 1 − δ ) d U p ( τ p ) d σ + δ d U r ( τ p ) d σ > 0
We demonstrate empirically that national pride positively impacts welfare. Works such as Frey and Stutzer [
Happiness i c t = β h 0 + β h 1 National Pride i c t + β 2 h Personal Characteristics i c t + β 3 Macro Controls i c t + Country c + Year t + ε i c t
The dependent variable Happinessict is a dummy variable that measures the level of happiness of individual i who resides in country c in period t. For this variable, we use the EVS question “taking all things together how happy are you?” Responses are from a scale of 1 to 4 (“very happy”, “quite happy”, “not very happy”, “not at all happy”). We create a happiness dummy such that “quite happy” and “very happy” are coded 1, and other responses are coded 0. The main independent variable National Prideict gauges how proud individuals are of their nation. For this variable, we use the EVS question “How proud are you to be [e.g., French]?” where individuals answer on a scale of 1 to 4 (“very proud,” “quite proud,” “not very proud,” and “not at all proud”).
The vector Personal Characteristicsict includes respondents’ income level (low, medium, high), age, gender, employment status (“full time”, “Part time”, “self-employed”, “retired”, “housewife”, “student”, “Unemployed”, “Other”), marital status (“single”, “married”, “divorced”, “separated”, “widowed”). We also include the age when respondents completed formal schooling (“below fifteen”, “fifteen to seventeen”, “above seventeen”), respondent’s number of children (“no child”, “one child”, “two children”, “three and more”), and a dummy on whether respondents is religious (“a religious person”, “not a religious person” and “a convinced atheist”). Macro Controlsct refers to a set of country-level variables, which are real GDP per capita from the Expanded Trade and GDP database by Kristian Gleditsch, gini index from The Standardized World Income Inequality Database (SWIID), as well as unemployment and inflation rates using data from the World Bank.
Country dummies are present to account for country-specific traits and year dummies are present to capture any common shocks among countries. The error term is ε i c t , which captures all omitted factors with E ( u i c t ) = 0 for all i, c and t. All standard errors are fully robust to arbitrary heteroskedasticity and are clustered by country to account for serial correlation at the country level. The summary statistics not shown here is in the online Appendix.
Column 1 restricts the analysis to low income members of society and demonstrates that compared to individuals who are “not at all proud of the nation” (base case), respondents that are “not very proud of the nation” are 22.3% more likely to be happy, while individuals who are “quite proud of the nation” and “very proud of the nation” are 121% and 185% more likely to be happy. When the analysis includes both low and medium income groups in Column 2, and just the high income group in Column 3, results are similar to that in Column 1.
Dependent Variable: Level of Happiness | Low Income Group | Low & Medium All Income Group | High Income Group | All Income Group |
---|---|---|---|---|
(1) | (2) | (3) | (4) | |
Not very proud of nation | 1.214+ | 1.203** | 1.390+ | 1.249* |
(0.139) | (0.099) | (0.267) | (2.221) | |
Quite proud of nation | 2.188*** | 2.155*** | 2.452*** | 2.221*** |
(0.214) | (0.167) | (0.433) | (0.201) | |
Very proud of nation | 2.822*** | 2.736*** | 2.776*** | 2.760*** |
(0.299) | (0.237) | (0.495) | (0.258) | |
Countries | 33 | 33 | 33 | 33 |
Observations | 22592 | 48790 | 21248 | 70038 |
Personal characteristics | Yes | Yes | Yes | Yes |
Macroeconomic controls | Yes | Yes | Yes | Yes |
Pseudo R2 | 0.14 | 0.15 | 0.12 | 0.16 |
+p < 0.1; *p < 0.05; **p < 0.01; ***p < 0.001.
For all three income groups in Column 4, individuals that are “not very proud of nation” are 25% more likely to report happiness, while respondents that fall under the categories of “quite proud of nation” and “very proud of nation” are 123% and 177% more likely to express happiness in their lives. Hence increasing national pride implies a higher likelihood of happiness or welfare.
Robustness checks of the results for the low as well as low and medium income groups are illustrated in
When considering all income groups in columns 3 and 4 of
Dependent Variable: Level of Happiness | Low Income Group | Low & Medium All Income Group | ||
---|---|---|---|---|
(1) | (2) | (3) | (4) | |
Not very proud of nation | 1.205* | 1.201* | 1.180* | 1.158* |
(0.114) | (0.108) | (0.091) | (0.077) | |
Quite proud of nation | 1.961*** | 1.978*** | 1.963*** | 1.926*** |
(0.170) | (0.165) | (0.153) | (0.130) | |
Very proud of nation | 2.360*** | 2.585*** | 2.284*** | 2.478*** |
(0.204) | (0.215) | (0.175) | (0.181) | |
Countries | 36 | 36 | 36 | 36 |
Observations | 30518 | 28594 | 65988 | 61641 |
Personal characteristics | No | Yes | No | Yes |
Macroeconomic controls | No | No | No | No |
Pseudo R2 | 0.10 | 0.15 | 0.10 | 0.16 |
+p < 0.1; *p < 0.05; **p < 0.01; ***p < 0.001.
Dependent Variable: Level of Happiness | High Income Group | All Income Group | ||
---|---|---|---|---|
(1) | (2) | (3) | (4) | |
Not very proud of nation | 1.221 | 1.196 | 1.225** | 1.167* |
(0.165) | (0.171) | (0.0872) | (0.083) | |
Quite proud of nation | 1.995*** | 1.991*** | 1.963*** | 1.934*** |
(0.274) | (0.284) | (0.155) | (0.149) | |
Very proud of nation | 2.231*** | 2.308*** | 2.210*** | 2.442*** |
(0.278) | (0.301) | (0.166) | (0.183) | |
Countries | 36 | 36 | 36 | 36 |
Observations | 28109 | 26249 | 114325 | 87890 |
Personal characteristics | No | Yes | No | Yes |
Macroeconomic controls | No | No | No | No |
Pseudo R2 | 0.10 | 0.13 | 0.09 | 0.16 |
Country List used in EVS from 1981 to 2014 | |
---|---|
Albania | Kosovo |
Armenia | Latvia |
Austria | Lithuania |
Belarus | Luxembourg |
Belgium | Macedonia |
Bosnia-Herzegovina | Malta |
Bulgaria | Moldova |
Croatia | Montenegro |
Cyprus | Netherlands |
Northern Cyprus | Norway |
Czech Republic | Poland |
Denmark | Portugal |
Estonia | Romania |
Finland | Russia |
France | Serbia |
Georgia | Slovak Republic |
Germany | Slovenia |
Great Britain | Spain |
Greece | Sweden |
Hungary | Switzerland |
Iceland | Turkey |
Ireland | Canada |
Northern | Ukraine |
Italy | USA |
+p < 0.1; *p < 0.05; **p < 0.01; ***p < 0.001.
We show theoretically that social welfare in direct democracies is increasing in national identity through the economics of identity. We substantiate this result by applying logit regression analysis on integrated data from the European Value Survey by demonstrating that happiness (proxy for social welfare) is positively correlated with national pride (proxy for national identity).
A number of studies using statistical methods have demonstrated that national pride is positively correlated with measures for subjective well-being such as happiness and satisfaction of life. However, to the best of our knowledge, none have tried to formulate a theoretical paper examining the connection between national pride and welfare/happiness. There are, however, potential limitations to this work. First, measuring social welfare through agent utility maybe a common approach taken by some, but it remains uncertain whether this approach is without fault since welfare is subjective. Second, even though some studies consider happiness a suitable representation for utility, some works dispute this assertion. Hence, proxying welfare with happiness is not without controversy.
Sincere thanks go to colleagues in the Nottingham University Business School at Malaysia for their support on this project and the anonymous reviewer for valuable feedback.
Tan, C.M., Lee, Y.H. and Ng, P.K. (2018) National Identity and Social Welfare. Theoretical Economics Letters, 8, 1665-1673. https://doi.org/10.4236/tel.2018.810107