Happy Brazil,

Unhappy São Paulo

Adam Okulicz-Kozaryn theaok.github.io �

In collaboration with Rubia Valente, Baruch College - CUNY

(was happy Rio, but we don’t have Rio)

Brazil

​

5th in the world (area) nearly half of South America 8.5/18 m km2

some of the world’s largest cities (population):

São Paulo, SP 12m [metro 21-23m]

Rio de Janeiro, RJ - 7m [metro 12-14m]

Brasília, Federal District - 3m

Salvador, Bahia - 3m

Fortaleza, Ceará - 2.7m

Belo Horizonte, Minas Gerais - 2.4m

the concrete jungle - São Paulo City

Inequality

side by side in

São Paulo

Literature: Brazil’s urban-rural similar to other developing countries

​

• economic activity concentrated in large urban centers (Cohen 2006)
• socio-economic development tends to rise with city size (more/better education, jobs, public services, income, etc)
• infrastructure and basic services improve linearly with size of place in Brazil (Meirelles et al. 2018)
• high inequality particularly in cities (lots of favelas or shanty towns)

literature

• urban-rural swb gradient:

happiness raises from low in largest cities to high in smallest areas (Berry and Okulicz-Kozaryn 2011; theaok.github.io)�

• Latin America
• no urban-rural swb gradient (Valente and Berry 2016).
• but LA vast and heterogenous. Need country-specific studies.

(and regional conferences like this one!)

​

Motivation: Urban-rural gradient in Brazil? Large cities less happy?

​

​

​

literature: few studies on rural-urban gradient in Brazil specifically

​

• Nepomuceno et al (2016) - mental health and wellbeing in urban and rural contexts in Northeast.
• Amorim et al (2017) - predictors of happiness among retired urban and rural areas.
• Burger et al. 2022 - Using gallup world poll for 2010 examines SWB.

Data: WVS

Dependent Variable: SWB/happiness

• Mostly cognitive and affective measure, respondent evaluates her life as a whole globally:
• “All things considered, how satisfied are you with your life as a whole these days?

Using this card on which 1 means you are ‘completely dissatisfied’ and 10 means you are ‘completely satisfied’ where would you put your satisfaction with your life as a whole?”

Urbanicity (interviewer reported; unlike Gallup), latest 2018 wave only

Method

• Ordinary least squares (ols)
• Default method in swb research (Blanchflower and Oswald 2011) and results substantially the same as those from discrete models (Ferrer-i-Carbonell and Frijters 2004).

​

• All variables recoded so that higher values means “more.”

Then “traditional” regression with size of a place

• Now with 1991, 1997, 2006, 2014, 2018 dataset
• Urbanicity “X049” measures urbanicity in 8 categories:

<2k, 2-5k, 5-10k, 10-20k, 20-50k, 50-100k, 100-500k, 500k>

• Controls: age, age2, male, married, divorced, income unemployed, race, and year dummies

non-linear effect: only top 2 categories (common in lit)

but we pooled the three smallest categories into one <10k:

• -2k
• 2-5k
• 5-10k

​

If we didn’t pool:

but note very wide CI on -2k, only 72 obs; 5-10k the happiest

​

Findings

• regional centers happier than São Paulo But villages happier yet! 0.9!!
• largest places least happy, and nonlinear effect: only 100-500k and >500k, not much difference across smaller ones
• yet the very smallest places <2k seem to be the least happy (only 72 obs)--possible reason is that remote places in Brazil lack basic necessities such as drinking water and other commodities (still healthcare universal in Brazil and available widely)
• 5-10k seem to be the happiest

Bonus: by province

• Limitation: sample runs small and WVS is not representative of provinces

2021 HDI (color)

numbers is rank

sp

rj

• red dot is Brasilia
• Sao Paulo federative unit red too (and couple others)
• Green north west (Amazonas etc)
• Huge difference 6.3 v 8.7

Thank you!

adam.okulicz.kozaryn@gmail.com

https://theaok.github.io

​

rubia.valente@baruch.cuny.edu

www.rubiavalente.com

​