NORMS, VALUES, AND PREFERENCES: MEASURING THE INTANGIBLE DRIVERS OF ECONOMIC AND SOCIAL OUTCOMES Conal Smith New Zealand Treasury 24 February 2016 Wellington
Measuring the intangible drivers of economic and social outcomes • Outline – Definitions: what do I mean by the intangible drivers of economic and social outcomes? – Evidence: why believe that these things matter? – OECD: what is the OECD doing to get a better handle on the intangibles? • Culture • Trust 2
Definition Low income countries Middle income countries 13% 26% Natural Capital Natural Capital 19% Produced Capital Produced Capital Intangible Capital Intangible Capital 59% 16% 68% High income countries 2% 17% Natural Capital Produced Capital Intangible Capital 80% 3 World Bank, 2006
Definition Low income countries Middle income countries 13% 26% Natural Capital Natural Capital 19% Produced Capital Produced Capital Intangible Capital Intangible Capital 59% 16% 68% To a first approximation, High income countries intangible capital is the only 2% 17% capital stock that matters in the production function. Natural Capital Produced Capital Intangible Capital 80% 4 World Bank, 2006
Definitions 5
Definitions • Intangibles – No direct market prices – Only measured indirectly or through subjective judgments 6
Definitions • Personal characteristics – Evaluations and feelings – Psychological characteristics – Knowledge and skills – Social networks • Inter-personal characteristics – Norms and values – Trust (expectations of others) – Culture 7
Definitions • Personal characteristics – Evaluations and feelings – Psychological characteristics Human Capital – Knowledge and skills – Social networks • Inter-personal characteristics – Norms and values – Trust (expectations of others) – Culture 8
Definitions • Personal characteristics – Evaluations and feelings – Psychological characteristics Human Capital – Knowledge and skills – Social networks • Inter-personal characteristics – Norms and values Social Capital – Trust (expectations of others) – Culture 9
Evidence: subjective evaluations Source: George Ward, 2015 10
Evidence: subjective evaluations Source: George Ward, 2015 11
Evidence: locus of control Source: George Ward, 2015 Heckman, 2006 12
Evidence: locus of control Source: George Ward, 2015 Heckman, 2006 13
Evidence: trust and GDP Source: Algan and Cahuc, 2013 14
Evidence: Trust and TFP Source: Algan and Cahuc, 2013 15
Evidence: Trust and life satisfaction Source: Algan and Cahuc, 2013 16
OECD The raison d’etre of the OECD is better policies for better lives . • Better lives: well-being • Better policies: what drives well-being The broad outlines of how to measure well-being is not a significant issue: there is a general consensus on what you need to measure across countries and from different philosophical starting points 17
OECD 18
OECD At the more detailed level, measurement gaps still exist Key measurement gaps include: • Measures of social capital • Measures of natural capital • Measures of non-cognitive skills • Measures of social contact • Measures of economic security …but we can make progress on these issues • Cultural bias • Social capital 19
Cultural bias in measuring subjective well- being 20
Cultural bias in measuring subjective well- being 21
Cultural bias in measuring subjective well- being 22
Cultural bias in measuring subjective well- being Four possible sources for unexplained country differences in subjective well-being • Unmeasured country circumstances and differences in how life is lived • Differences in how people feel about their lives • Language differences in scale use • Cultural response styles or biases 23
Relatively little evidence for large cultural differences in emotionality 24
Our assumptions about which countries are unusually emotional are wrong… 25
… and emotionality is not driving life satisfaction 26
There is relatively little cross-country correlation in response bias to different types of question 27
Country of birth explains only 18% of variance in life satisfaction 28
The maximum implied size of cultural bias is not large 29
Trustlab • The OECD published Guidelines on Measuring Subjective Well-being in 2013 • In 2017 the OECD Guidelines on Measuring Trust will be published 30
Trustlab • There is strong prima-facie evidence that questions on generalised trust collect valid information (Algan and Cahuc, 2013) • Evidence for measures of institutional trust (i.e. trust in parliament, police, media etc) is less clear, but measures have higher immediate policy relevance • Unlike subjective well-being, the evidence base for the validity of trust measures is rather limited • Falk, Becker, Dohmen, Huffman, and Sunde (2013) – Risk taking – Time discounting – Trust – Altruism – Positive reciprocity – Negative reciprocity • We need high quality experimental measures of trust that can be used to validate survey-based measures 31
Trustlab • Internet survey • Representative national sample of n=1000 • Combines traditional survey questions with experimental games providing both behavioural and subjective information • Games are played with real resources at stake (mean value c$15 NZ) 32
Trustlab Trust Game Core experimental module Public Good Game (every wave) Flexible content Flexible experimental module Generalised trust (Rosenburg), Wallet Core trust question module question, radius of trust, (every wave) Flexible content Flexible question module age, sex, HH composition, marital status, Demographic module (every LF status, HH income, personal income, wave) educational attainment, 33
Trustlab Trust Game Public Good Game Experimental Flexible content Generalised trust (Rosenburg), Wallet question, radius of trust, Flexible content Traditional survey questions age, sex, HH composition, marital status, LF status, HH income, personal income, educational attainment, 34
Trustlab Trust Game Core experimental module Public Good Game (every wave) Implicit Association Test Institutions experimental module Generalised trust (Rosenburg), Wallet Core trust question module question, radius of trust, (every wave) GOV trustworthiness questions, trust in Institutions question module institutions age, sex, HH composition, marital status, Demographic module (every LF status, HH income, personal income, wave) educational attainment, 35
Content – experimental module on generalised trust 36
Trustlab • Timeframes • 2015 – Survey development • 2016 – IT platform finalised – First wave (generalised trust + institutional trust) implemented in Korea, France, and 2 other countries. – Results of first wave published • 2018 – Second wave? 37
Recommend
More recommend
