Estimating Well-Being in Developing Countries Well-Being 1

Estimating Well-Being in Developing Countries

Well-Being (1) What is well-being? (2) Why should economists be interested in well-being? (3) Estimating well-being equations (4) Empirical Findings

Well-Being (1) What is well-being? Aristotle sees ‘happiness’ and ‘living –well’ as the same thing and that living well consists of doing something. Well-Being and Ill-Being – Jeremy Bentham in 19 th Century. World Health Organisation Quality of Life – Concerned with measuring physical health, psychological health, social relationships, and the environment. “What people have or do not have (material); what people do or cannot do with it (relational); what people think or feel (subjective)”. Wellbeing in Developing Countries We. D (2009).

Well-Being We focus on subjective well-being (SWB). However, SWB will be closely correlated with material and relational factors. If have a house and car then better-off in yourself (SWB) and when compared to others (relational). It is the relative position of an individual that is arguably most interesting in both theoretical and empirical literature.

Well-Being Issue of who to compare your well-being to…. (i) keeping up with the Jones’ – neighbour or village effect. (ii) peer group – fellow worker, average worker, race, gender, age, caste (iii) different time periods – yesterday, last year. (iv) different generations – parents (v) some pre-determined social norm – sociology.

Well-Being (2) Why should economists interested in well-being? be Well-being and satisfaction are similar concepts hence can directly test utility. It is assumed in micro-foundations of neo-classical theory that utility is formed based on consumption of goods which in itself is determined by the budget constraint (income). A Satisfaction=Utility B

Well-Being However Easterlin (1974) found that over time happiness did not increase with income – Easterlin paradox Average Satisfaction and GDP per capita Satisfaction with life Time If increasing income does not increase satisfaction or utility then why be obsessed with this?

Well-Being Richard Layard and others (e. g. Frey and Stutzer, 2002), argue that it is relative income explains why happiness does not that increase significantly beyond (i) a certain level of GDP per capita (Macro) (ii) a certain individual absolute income level (Micro) So if your income increases but so does everyone elses then your relative position is the same. Indeed, if your income increases but at a slower rate than the average income increase then your relative position worsens – income inequality is becoming more skewed towards the very highest decile in the US and UK…. . is this why we’re not more satisfied?

Well-Being Lucas et al (2004) argue that while income, health and family are correlated with life satisfaction that they do not explain much of the variance in satisfaction. Personality variables account for a much larger share of subjective satisfaction – use of twin studies indicates that 80% of the variance in satisfaction is something that comes from within. Genetic. Any exogenous shocks thus have an impact on short-term variation in satisfaction, but in the long-run individuals return to some ‘set-point’.

Well-Being 10 Negative Shock – e. g. unemployment, death, tragic news Older people tend to report higher levels of life satisfaction – Why? Set Point SWB Positive Shock – e. g. inheritance, lottery win, house prices 0 Years

Well-Being Cross-Country Comparisons Many studies compare countries (e. g. Stevenson and Wolfers, 2008) and suggest that GDP per capita is significant in subjective happiness. Other factors that are important include health and unemployment rates. However variations in happiness between countries can be criticised. Does someone who scores 8 in Sweden really have the same level of happiness as someone who scores 8 in India? – reference group.

Well-Being (3) Estimating well-being equations. Using the Likert scale for happiness or life satisfaction as the dependent variable. Economists can test a number of hypotheses. (1) Are income, wealth, assets positively correlated with happiness? In cross-country studies use GDP per capita as measure of income. In national study use income per adult equivalence for household level income, use the squared term too to see if happiness increases with income at a decreasing rate. Same method for information on assets and wealth – may need to create an index or use factor analysis to get a single measure of household assets/wealth e. g. pots and pans, knives, agricultural tools.

Well-Being Endogeneity Issues It could well be that someone of a happier disposition will earn more in the labour market (Ruut Veenhoven). This could be picking up social network effects and being more sociable. It could also be picking up unobservable characteristics of individuals not traditionally found in earnings functions, e. g. optimism, positive attitude, work ethic?

Well-Being (2) Employment and Satisfaction Well known in the labour economics literature that the unemployed are scarred – we would expect that the unemployed would, ceteris paribus, report lower levels of satisfaction than the employed. As well as this basic test the applied literature has used the satisfaction data to test whether the searching and non-searching unemployed report different satisfaction rates. This tests the neo-classical hypothesis that via maximising utility the non-searching unemployed are voluntarily unemployed as they choose not to search. If it is found that the non-searching are as satisfied as the searching unemployed or even less satisfied then this is consistent with the ‘discouraged worker hypothesis’.

Well-Being (3) Satisfaction and Relative Consumption Evidence from Hinks and Davies for Malawi (2008), Copestake et al (2009) for Peru indicates that even amongst very poor communities, as average consumption increases so satisfaction declines. Suggests that relative economic position in the community is important to satisfaction. This can be tested further by calculating whether your household consumption (or income) is above the community average or not. Studies on job satisfaction have found that relative earnings are important – findings tend to confirm that as the average earnings of someone with your skill set increases then, ceteris paribus, job satisfaction declines.

Well-Being Other Hypotheses of interest. Satisfaction and Crime – Powdthavee (2008), Hinks and Davies (2010). Satisfaction and Race or Gender – Hinks & Gruen (2007), Hinks and Davies (2008). Satisfaction and Social Networks – Hinks and Davies (2008), Polygamy, Religion and Satisfaction.

Well-Being (4) Empirical Evidence (i) Well-Being, Income and Relative Income Evidence is conclusive with a positive relationship within country well-being or happiness and income level. For both developing and developed countries. E. g. Powdthavee (2005), Hinks and Gruen (2007), Hinks and Davies (2008, 2010) work using crosssectional data sets for developing countries.

Well-Being (ii) The Case of Employment Status and Well-Being Lucas et al (2004, pp. 11) find that (i) satisfaction begins to decline before the worker is unemployed (ii) life satisfaction is reduced massively when unemployed and (iii) satisfaction increases but not back to pre-unemployment level.

Well-Being Case Study of South Africa – Kingdon and Knight (2006). First issue is that of endogeneity – are the unhappiest people more likely to be unemployed? Evidence from longitudinal studies by psychologists is that this reverse causality is ‘doubtful’ – Veenhoven suggests that happier people are more likely to be employed first than unhappier people. Happiness is on a 5 -point scale, 0 is very dissatisfied and 5 is very satisfied. Only collected for head of households – so not individual level. The searching unemployment rate = (No. of searching unemployed in HH/No. of broad labour force participants). The non-searching unemployment rate = (No of non-searching unemployed in HH/No. of broad labour force participants).

Well-Being No significant difference in life satisfaction between the searching and non-searching unemployed. Implication is to reject the hypothesis that the non-searching unemployed are happier than the searching and accept that worker discouragement is at work in South Africa. Given the strict rate of unemployment is high (still officially 24% in 2 nd quarter of 2009) this is perhaps unsurprising but importantly adds to the evidence against voluntary unemployment.

Well-Being An abridged version of an estimated happiness equation taken from Hinks and Davies (2008). Analysis of some findings: (i) Find that larger consumption raises life satisfaction. In developing countries there is still a great need to increase economic growth to raise income levels. (ii) If relatively better off in terms of consumption then more satisfied. The importance of relative position is confirmed. (iii) The perceptions of your neighbours’ wealth positively predicts more satisfaction for you. Capturing some of the log consumption variable as this coefficient decreases in size. (iv) Wealthier households (assets) are more satisfied (vi) The salaried employed and self-employed are more satisfied than farmers.

Model 1 Model 2 Model 3 Model 4 Model 5 ln(Per Capita Consumption) 0. 338*** 0. 173*** 0. 184*** 0. 138*** 0. 072** (13. 905) (6. 115) (7. 044) (4. 811) (2. 322) ln(Mean Community Consumption) -0. 193*** (-5. 267) Relative Household Consumption 1 0. 152*** 0. 174*** 0. 164*** (4. 910) (5. 586) (5. 212) Relative Household Consumption 2 0. 061*** (5. 671) Neighbours' Subjective Wealth 0. 108*** 0. 101*** (8. 627) (7. 996) Asset Score 0. 096*** (5. 564) Salaried Employment Dummy 0. 144*** 0. 136*** 0. 135*** 0. 124*** 0. 091*** (4. 442) (4. 234) (4. 199) (3. 848) (2. 737) 0. 151*** 0. 145*** 0. 146*** 0. 137*** 0. 130*** (4. 417) (4. 244) (4. 266) (4. 002) (3. 797) N 11264 11248 11205 Pseudo R-Squared 0. 068 0. 070 0. 071 1802. 312 1793. 891 1812. 681 1850. 342 1869. 723 Self-Employed Dummy Chi 2

Some Useful Websites for Economics of Wellbeing, Quality of Life and Happiness (1) World Health Organisation Quality of Life, http: //www. who. int/substance_abuse/research_tools/whoqolbref/en/ (2) Wellbeing in Developing Countries We. D, http: //www. welldev. org. uk/ (3) World Values Survey http: //www. worldvaluessurvey. org (4) World data base of Happiness – Ruut Veenhoven http: //worlddatabaseofhappiness. eur. nl/ (5) International Society for Quality of Life Studies http: //www. isqols. org/

References Easterlin, R. , (1974), ““Does Economic Growth Improve the Human Lot? , ” in Nations and Households in Economic Growth: Essays in Honor of Moses Abramovitz, ed. by P. A. David, and M. W. Reder. New York: Academic Press. Stevenson, B, and Wolfers, J. , (2008), “Economic Growth and Subjective Well-Being: Reassessing the Easterlin Paradox”, NBER Working Paper Series, No. 14282 Layard. R. , (2005), Happiness: Lessons from a new science, Penguin, London. Frey, B. , and Stutzer, A. , (2002), Happiness and Economics: How the Economy and Institutions Affect Human Well-Being, Princeton University Press, NJ. Lucas, R. , Clark, A. , Georgellis, Y. , and Diener, E. , (2004), “Unemployment Alters the Set Point for Life Satisfaction”, Psychological Science, 15(1): 8 -13. Powdthavee, N. (2005). Unhappiness and crime: Evidence from South Africa. Economica, 72, 531– 547. ), Hinks, T. , & Gruen, C. (2007). What is the structure of South African happiness equations? : Evidence from quality of life surveys. Social Indicators Research, 82(2), 311– 336. Hinks, T. , &Davies, S. (2008). Life satisfaction in Malawi. Journal of International Development, 20, 888– 904. Davies, S. , and Hinks, T. , (2010), “Crime and Happiness amongst heads of households in Malawi”, Journal of Happiness Studies, 11: 457 -476. Copestake, J. , Guillen-Royo, M. , Chou, W. J. , Hinks, T. , Velazco, J. , (2009). The relationship between economic and subjective wellbeing indicators in Peru. Applied Research in Quality of Life, 4 (2), pp. 155 -177. Kingdon, G. , and Knight, J. , (2006), “The measurement of unemployment when unemployment is high”, Labour Economics 13(3): 291 -315.