Formation of Aspirations

1

Formation of Aspirations – An Empirical Analysis

Tanguy Bernard1, Stefan Dercon2, Fanaye Tadesse1, Alemayehu Seyoum Taffesse1 and Ibrahim Worku1

1International Food Policy Research Institute2 University of Oxford

March 20, 2012CSAE Conference 2012 ‘Economic Development in Africa’

Oxford, UK

13/04/23

• Development discourse and praxis focus on ‘opportunities’

• Why do the poor not ‘invest’? – Ethiopians and fatalism?

• A variety of mostly complementary explanations have been forwarded over the years. – Low returns to investments – Example: no schools, lack of credit

(market failures); – Unexploited opportunities due to lack of information/knowledge

about the opportunities/returns – Example: insufficient observable cases;

– Social constraints dampen the economic attractiveness of investment opportunities – Example: egalitarian norms, very high taxation, excessive regulations’

Motivation – why aspirations

13/04/23 2

• Focus - ‘external circumstances’ and ‘opportunities’,

• Shift in focus - constraints associated with the manifested attributes of decision makers– Identity issues: People’s choices are conditioned by their

sense of self – Example: stereotype roles;– Psychological issues: impatience, commitment, and

psychological barriers and similar reasons identified by the behavioural economics literature – Example: loss aversion and the consequent preference for the status quo;


13/04/23 3

13/04/23 4


• the aspirations failure approach:

– attempts to blend external constraints that the poor face with the potential effect these constraints may have on the internal logic governing choice by them;

– affords an alternative characterisation of what appears to be fatalism (Ethiopia)

13/04/23 5

Elements of the Aspirations Perspective

What are aspirations? Aspirations have two distinctive features:• Future-oriented - are goals that can only be satisfied at some

future time (differ from immediate gratifications);• Motivators - are goals individuals are willing, in principle, to

invest time, effort or money in to attain (different from idle daydreams and wishes)

Note: the ‘willingness to invest’ is ‘potential’, or ‘conditional’ Ref: Simon (1977), Selten (1999), Appadurai (2002), Ray (2003)


Why are aspirations important/useful?

Aspirations (or the capacity to aspire): • Reflect bounded rationality;• Are socially determined (social interaction); • Are distributed unevenly within communities.

– Condition individual behaviour and well-being– Useful device in analysing and/or addressing poverty

13/04/23 6

Framework

13/04/23 7

Wealth Aspiration - DataRound 7 Question: We would now like you to think of your own wealth. Thinking of a scale from 1 (the lowest or worst level) to 10 (the highest or best level):

Q39a. At what level do you believe you are currently?

Q39b. At what level would you like to be?

Estimation

•Use responses to Q39b as the dependent variable;

•Responses to the wealth aspiration question are ordered –

– ordered probit model;

– Generalized ordered probit model

– Semi-nonparametric estimation of extended ordered probit model

•Clustered standard errors

13/04/23 8

Data - ERHS Survey Rounds

• Spatial coverage: 15 Kebeles (villages);

• Temporal coverage: 1993/94-2008/9 (7 rounds the last three roughly one every 5 years)

13/04/23 9

Variables – All survey roundsWealth Aspiration

Level of wealth the respondent would like to achieve (measured using a 10-point scale)

Age Respondent's age in number of yearsAge squared Respondent's age in number of years (squared)Gender (male=1) Gender of respondent

Non-farm activities (yes=1) Dummy - the respondent engages in non-farm activities (Yes=1)

Iddir (member=1)Dummy - the respondent is a member of an Iddir (funeral association) (Member=1)

Married (yes=1) Marital status of the respondent (married=1)

ShocksShock faced by respondent in the last 5 years - index calculated using principal components

Neighbourhood sizeNumber of neighbours within a radius of 500m from the respondent's residence

Average income growth of neighbours (round7-round6)

Average growth in the real consumption per capita (in 1994 prices) of the respondent's neighbours between Rounds 6 and 7 (%).

Income growth (round 2 - round 1) - Income growth (round 7 - round 6)

Growth in the real consumption per capita (in 1994 prices) of the respondent's household between successive survey Rounds (%)

PoorNumber of rounds the respondent's family was classified as poor (using the poverty line in 1994)

13/04/23 10

Internal Locus of Control – Round 7

13/04/23 11

.2.4

.6.8

1

0 2 4 6 8selfreported_wealth

Pr(internal_loc) 95% CIFitted values

Marginal (Partial) Effects – All Rounds

dy/dx per one SD change (%)

Married (yes=1)

Average income

growth of neighbors

(R6-R7)

Income growth

(R6 – R7)

Income growth

(R5–R6)Poor

Pr(Wealth Aspiration = 1) -0.25 -0.13 -0.09 -0.10 0.18






Pr(Wealth Aspiration = 7) 0.15 -0.21 -0.14 -0.16 0.19

Pr(Wealth Aspiration = 8) 9.01 6.50 5.37 5.97 -5.84

Mean (SD) -18.2 (32.5) -18.3 (53.7) 2.2 (59.7) 2.8 (1.8)

Note: Figures in red are statistically significant at least at 10% level of significance.13/04/23 12

Variables – Rounds 6-7

Wealth AspirationLevel of wealth the respondent would like to achieve (measured using a 10-point scale)

Age Respondent's age in number of yearsAge squared Respondent's age in number of years (squared)Gender (male=1) Gender of respondent

Non-farm activities (yes=1) Dummy - the respondent engages in non-farm activities (Yes=1)

Iddir (member=1)Dummy - the respondent is a member of an Iddir (funeral association) (Member=1)

Married (yes=1) Marital status of the respondent (married=1)

ShockShock faced by respondent in the last 5 years - index calculated using principal components

Neighbourhood sizeNumber of neighbours within a radius of 500m from the respondent's residence

Log of asset valueLogarithm of the value of assets (including farm implements, furniture, jewelry) the respondent’s household owned in survey round 7 (2008/09)

Log of average asset value of neighbors

Logarithm of the average value of assets owned by the respondent's neighbours (residents within 500m radius of the respondent's house) excluding the respondent’s household13/04/23 13

Marginal (Partial) Effects – Rounds 6-7

dy/dx per one SD change (%)

Gender (male=1)

Log of asset value

Log of average asset value of

neighborsPr(Wealth Aspiration = 1) 0.01 -0.04 -0.04

Pr(Wealth Aspiration = 2) 0.01 -0.04 -0.04

Pr(Wealth Aspiration = 3) 0.05 -0.12 -0.10

Pr(Wealth Aspiration = 4) 0.11 -0.30 -0.34Pr(Wealth Aspiration = 5) 0.30 -0.85 -1.02Pr(Wealth Aspiration = 6) 0.45 -1.34 -1.43Pr(Wealth Aspiration = 7) 0.74 -2.20 -2.45Pr(Wealth Aspiration = 8) 0.74 -2.07 -2.45Pr(Wealth Aspiration = 9) 0.18 -0.50 -0.57

Pr(Wealth Aspiration = 10) -2.63 7.44 8.51Mean (SD) 7.1 (1.2) 7.7 (0.7)

Note: Figures in red are statistically significant at least at 10% level of significance.13/04/23 14

Observations

Poorer individuals seem to have lower aspirations;

Correlations but note that most of the relevant variables lagged;

Results persist across models;

Encouraging result since data not designed for the purpose;

13/04/23 15

Aspiration window

Aspiration leveland gap

Aspiration failure

Aspired

Gap

Efforts to improve

Attained

Well-being


13/04/23 16

Elements of the Aspirations PerspectiveMeasurement Issues

• Aspirations are not directly observable

– Revealed by observed behaviour: interpretation issues (linking aspirations and behaviour)

– Elicited using subjective questions: measurement issues

• Limits to subjective assessment:

– subjects’ willingness to report private knowledge, evaluation apprehension, and subject role playing

– attributes of the instrument used: prior questions (anchoring), the number of categories on the rating scale, the adjectives that are used as the endpoints of the rating scale, and the adverbs that describe scale categories.

(e.g. Delavande et al. (2009), Bertrand and Mullainathan (2001) for some reviews)

13/04/23 17


Identification issues

• individual characteristics affect aspirations, aspiration windows and behaviour (e.g. schooling levels, wealth, and family background),

Particularly the endogeneity of the aspiration window a key hurdle

• Investment ‘cause’ aspirations (e.g. the successful investor may in turn revise his/her aspiration to a higher level), or

• aspirations ‘cause’ investment – the one we aim to identify.

An experimental design helpful to test the link between aspiration window, aspirations and aspirations failures.

13/04/23 18

Aspiration formation – A simple formulation

Central Proposition: ((Bandura (1993, 1994) in psychology, Appadurai (2001) in sociology , Ray (2006) in economics):

Aspirations reflect: an individual’s aspiration window – peers and economic

opportunities of the local environment An individual’s own characteristics and past experience including

shocks;

13/04/23 19

1

, , , ,1

1( ) ( ) ( )

1

t

it i i i w w w w w w ia u s u s u s

t t t t tt

q q q e-

- - -=

é ù= + + +ê úë û- å

Formation of Aspirations - An Empirical Analysis

Technology

aspirations importantuseful

aspirations perspectivewhy

aspirations perspectivewhat

wealth aspiration question

wealth aspiration dataround

aspirations failure

focus constraints