22
CROSS-COUNTRY INCOME MOBILITY COMPARISONS UNDER PANEL ATTRITION: THE RELEVANCE OF WEIGHTING SCHEMES Luis Ayala (IEF, URJC) Carolina Navarro (UNED) Mercedes Sastre (IEF, UCM) EPUNet 2006 Barcelona 8-9 May 2006

CROSS-COUNTRY INCOME MOBILITY COMPARISONS UNDER PANEL ATTRITION: THE RELEVANCE OF WEIGHTING SCHEMES Luis Ayala (IEF, URJC) Carolina Navarro (UNED) Mercedes

Embed Size (px)

Citation preview

Page 1: CROSS-COUNTRY INCOME MOBILITY COMPARISONS UNDER PANEL ATTRITION: THE RELEVANCE OF WEIGHTING SCHEMES Luis Ayala (IEF, URJC) Carolina Navarro (UNED) Mercedes

CROSS-COUNTRY INCOME MOBILITY COMPARISONS UNDER

PANEL ATTRITION: THE RELEVANCE OF WEIGHTING SCHEMES

Luis Ayala (IEF, URJC) Carolina Navarro (UNED)

Mercedes Sastre (IEF, UCM)

EPUNet 2006

Barcelona 8-9 May 2006

Page 2: CROSS-COUNTRY INCOME MOBILITY COMPARISONS UNDER PANEL ATTRITION: THE RELEVANCE OF WEIGHTING SCHEMES Luis Ayala (IEF, URJC) Carolina Navarro (UNED) Mercedes

Motivation• Longitudinal Data: ECHP

– Very detailed information – Income distribution studies: Static Analysis Dynamic

Analysis• Income mobility, poverty dynamics, transitions between

economic states (deprivation, unemployment…)

• Limits: “sample attrition”– Is there attrition in the ECHP? – How much? – Is it selective?– Does if have effects on (dynamic) estimates?– How can it be corrected?

Page 3: CROSS-COUNTRY INCOME MOBILITY COMPARISONS UNDER PANEL ATTRITION: THE RELEVANCE OF WEIGHTING SCHEMES Luis Ayala (IEF, URJC) Carolina Navarro (UNED) Mercedes

Objective• Analyse the effect of sample attrition in the ECHP on

different measures of income mobility • Study the consequences of different weighting schemes

used to correct the potential bias introduced by attrition– Extension and distribution of sample attrition– Estimation of different longitudinal weights based on the

inverse selection probability (probit models)– International Comparison

• France, Germany, Italy, Spain, United Kingdom• Differential impact of sample attrition and weighting schemes

on income mobility • Degree in which sample attrition can condition the results of

the comparative analysis

Page 4: CROSS-COUNTRY INCOME MOBILITY COMPARISONS UNDER PANEL ATTRITION: THE RELEVANCE OF WEIGHTING SCHEMES Luis Ayala (IEF, URJC) Carolina Navarro (UNED) Mercedes

Attrition in the ECHP (I)• Sample attrition: lost of a percentage of the initial

sample as new waves of the survey develop • Several studies: Peracchi (2002), Behr et al.(2002), etc. • Extension and trends of attrition in the ECHP

Substantial rates over a few years important differences among countries

• May prevent to follow up a significant part of the sample and influence the distributive results (if attrition implies a lost of representativeness)

Page 5: CROSS-COUNTRY INCOME MOBILITY COMPARISONS UNDER PANEL ATTRITION: THE RELEVANCE OF WEIGHTING SCHEMES Luis Ayala (IEF, URJC) Carolina Navarro (UNED) Mercedes

Attrition in the ECHP (II)

-60.00

-50.00

-40.00

-30.00

-20.00

-10.00

0.00

1995 1996 1997 1998 1999 2000 2001

Spain Germany France United Kingdom Italy UE-5 UE-15

Page 6: CROSS-COUNTRY INCOME MOBILITY COMPARISONS UNDER PANEL ATTRITION: THE RELEVANCE OF WEIGHTING SCHEMES Luis Ayala (IEF, URJC) Carolina Navarro (UNED) Mercedes

Attrition in the ECHP (III)

• Substantial attrition rates in the ECHP: – Attrition: no necessarily affects estimates, only

if has a selective character• If low income households more exit probability previsible effect of inequality reduction

• The estimations of longitudinal processes may be biased

Page 7: CROSS-COUNTRY INCOME MOBILITY COMPARISONS UNDER PANEL ATTRITION: THE RELEVANCE OF WEIGHTING SCHEMES Luis Ayala (IEF, URJC) Carolina Navarro (UNED) Mercedes

Attrition in the ECHP (IV)

• Incidence of attrition by socioeconomic categories: % remaining in the ECHP (balanced panel)– In general similarities in attrition patterns among

countries / some divergences– Relevant Variables:

• Income, household main income source, age, household type, education level and housing tenure

• Divergences “latin model” vs. Rest of countries

Page 8: CROSS-COUNTRY INCOME MOBILITY COMPARISONS UNDER PANEL ATTRITION: THE RELEVANCE OF WEIGHTING SCHEMES Luis Ayala (IEF, URJC) Carolina Navarro (UNED) Mercedes

Weighting Schemes (I)

• Attrition Not totally ramdom potential bias• Estimation of longitudinal weights for each observation • Intuition

– Probabilistic assesment of how many attriters a particular observation represents

– Indiv. with greater exit probability greater weight• Estimation

– Based on the inverse selection probability obtained from different probit models using socieconomic characteristics of individuals and households (Fitzgerald et al., 1998, Gradín et al. 2004)

– Probit• Prob. individual not in the sample in wave 8 Pri (Y=0) • Prob. individual remains in the sample in wave 8 Pri (Y=1) • For each observation: longitudinal weight function of the inverse selection

probability of remaining in the sample

Page 9: CROSS-COUNTRY INCOME MOBILITY COMPARISONS UNDER PANEL ATTRITION: THE RELEVANCE OF WEIGHTING SCHEMES Luis Ayala (IEF, URJC) Carolina Navarro (UNED) Mercedes

Weighting Schemes (II)

• Estimates of longitudinal weights Alternative Models:– Type A Household and household head

characteristics (results sensitivity)• Model 1 (1

A): Adjusted income, main income source, household head characteristics (age, sex, marital status and education level (problem)

• Model 2 (2 A): similar to Model 1 excluding

education level• Model 3 (3

A): similar to Model 1 + household size, number of children and number of full time workers in the household

Page 10: CROSS-COUNTRY INCOME MOBILITY COMPARISONS UNDER PANEL ATTRITION: THE RELEVANCE OF WEIGHTING SCHEMES Luis Ayala (IEF, URJC) Carolina Navarro (UNED) Mercedes

Weighting Schemes (III)

• Tipo B individual characteristics (adults with completed interview)– Model 1 (1

B): : Individual income, activity status, age, sex, marital status and individual education level (problem)

– Model 2 (2 B): similar to Model 1 + age

– Model 3 (3 B): similar to Model 2 + health status

• Type A and type B Models :

• In general, results confirm descriptive analysis not random attrition

Page 11: CROSS-COUNTRY INCOME MOBILITY COMPARISONS UNDER PANEL ATTRITION: THE RELEVANCE OF WEIGHTING SCHEMES Luis Ayala (IEF, URJC) Carolina Navarro (UNED) Mercedes

SPAIN E 1 A 2

A 3 A 1

B 2 B 3

B

E 1.0000

1 A 0.0082 1.0000

2 A

-0.0005 0.9915 1.0000

3 A

0.0344 0.9948 0.9859 1.0000

1 B

-0.0125 0.3749 0.3691 0.3765 1.0000

2 B

-0.0217 0.4012 0.3943 0.4038 0.9316 1.0000

3 B

-0.0099 0.3543 0.3485 0.3577 0.8079 0.8680 1.0000

Weighting Schemes (IV)

Page 12: CROSS-COUNTRY INCOME MOBILITY COMPARISONS UNDER PANEL ATTRITION: THE RELEVANCE OF WEIGHTING SCHEMES Luis Ayala (IEF, URJC) Carolina Navarro (UNED) Mercedes

Weighting Schemes (V)

• Wave 1 frequency distribution: initial sample vs. balanced panel– 1) Initial sample, no weighting– 2) Initial sample, cross section Eurostat weights – 3) Initial sample (balanced panel), Eurostat longi.tudinal

weights – 4) Initial sample (balanced panel), estimated longitudinal

weights• Comparisons: 1) vs 4) similarities

– Estimated longitudinal weights some “guarantees”– Comparisons: 2) vs 3) high divergences– Categories with greater attrition rates less adjustment due

to the use of longitudinal weights

Page 13: CROSS-COUNTRY INCOME MOBILITY COMPARISONS UNDER PANEL ATTRITION: THE RELEVANCE OF WEIGHTING SCHEMES Luis Ayala (IEF, URJC) Carolina Navarro (UNED) Mercedes

Weighting Schemes (VI)

• Summary: – Several attrition adjustment possibilities

longitudinal weights– Frequency distribution important differences

according to the weighting scheme

Implications for dynamic analysis

Income Mobility

Page 14: CROSS-COUNTRY INCOME MOBILITY COMPARISONS UNDER PANEL ATTRITION: THE RELEVANCE OF WEIGHTING SCHEMES Luis Ayala (IEF, URJC) Carolina Navarro (UNED) Mercedes

Attrition and Income Mobility (I)

• Income Mobility Analysis– Results may be affected by the possible “balanced panel”

lack of representation of the population• Greater exit of low income individuals possibly ascending

mobility indicators greater than real• Lower attrition of high income individuals potential effect on

wage mobility analysis

• Preliminary Analysis– Differential impact of attrition among countries– Not random attrition (selective)

Page 15: CROSS-COUNTRY INCOME MOBILITY COMPARISONS UNDER PANEL ATTRITION: THE RELEVANCE OF WEIGHTING SCHEMES Luis Ayala (IEF, URJC) Carolina Navarro (UNED) Mercedes

Attrition and Income Mobility (II)• What are the extent and type of income mobility in

a society? – The answer depends greatly on how well panel data

represent the population

• Does attrition introduce bias on income mobility estimations?– Assess the effect of attrition on income mobility

measures – Estimate if the bias can be corrected using the estimated

longitudinal weights

Page 16: CROSS-COUNTRY INCOME MOBILITY COMPARISONS UNDER PANEL ATTRITION: THE RELEVANCE OF WEIGHTING SCHEMES Luis Ayala (IEF, URJC) Carolina Navarro (UNED) Mercedes

Attrition and Income Mobility (III)• Does attrition introduce bias on income mobility

estimations? (Fitzgerald et al. 1998, Behr et al. 2003)– Split the sample according to attrition behaviour of individuals on

future waves and compare the mobility results for the two subsamples

• Estimating medium-term income mobility : Wave 1 to Wave 4– ECHP 8 waves – Subsample P balanced panel individuals remaining all waves

(from wave 1 to wave 8) – Subsample K individuals remaining in the sample at least the

four first waves

Np < Nk

Page 17: CROSS-COUNTRY INCOME MOBILITY COMPARISONS UNDER PANEL ATTRITION: THE RELEVANCE OF WEIGHTING SCHEMES Luis Ayala (IEF, URJC) Carolina Navarro (UNED) Mercedes

Attrition and Income Mobility (IV)• Income mobility indices summarize changes in

economic status from one time period to another• Fields y Ok 1999:

– Mobility as a normative concept / value judgments – Various approximations different dimensions No consensus

• Mobility as inequality reduction as the accounting period is extended Shorrocks Rigidity Index

• Mobility as origin independence of last period income (statistic associación) Hart Index

• Mobility as transition among different classes on the income distribution (matrices) Shorrocks Index, Bartholomew Index

• Mobility as income movement (Fields and Ok Index)

Page 18: CROSS-COUNTRY INCOME MOBILITY COMPARISONS UNDER PANEL ATTRITION: THE RELEVANCE OF WEIGHTING SCHEMES Luis Ayala (IEF, URJC) Carolina Navarro (UNED) Mercedes

Attrition and Income Mobility (V)• Effect of attrition on income mobility

measures: compare the mobility results of subsample P and subsample K

– Attrition: no bias on international comparisons No country rerankings.

– Comun pattern: subsample P shows lower mobility than subsample K greater mobility of attriters

Page 19: CROSS-COUNTRY INCOME MOBILITY COMPARISONS UNDER PANEL ATTRITION: THE RELEVANCE OF WEIGHTING SCHEMES Luis Ayala (IEF, URJC) Carolina Navarro (UNED) Mercedes

Attrition and Income Mobility (VI)

• Mobility decomp. = Growth Mobility + Transfer Mobility – Changes in the relative contribution of each component

• Mobility decomposition by population subgroups– Important differences in group mob. measures and

mobility contribution between subsample p and subsample k

Attrition no high influence on aggregate resultsCertain bias on income mobility decomposition by components and population groups

ADJUSTMENT PROCEDURES

Page 20: CROSS-COUNTRY INCOME MOBILITY COMPARISONS UNDER PANEL ATTRITION: THE RELEVANCE OF WEIGHTING SCHEMES Luis Ayala (IEF, URJC) Carolina Navarro (UNED) Mercedes

Mobility and weighting schemes (I)• Attrition adjustment schemes:

– No adjustment : li=1

– Eurostat longitudinal weights: l i= Ei

– Estimated longitudinal weights: l i= A2i

• Does the weighting scheme affect the estimations? – Correlation among different weights questions the

robustness of analysis

ECHP Mobility from wave 1 to wave 8

Page 21: CROSS-COUNTRY INCOME MOBILITY COMPARISONS UNDER PANEL ATTRITION: THE RELEVANCE OF WEIGHTING SCHEMES Luis Ayala (IEF, URJC) Carolina Navarro (UNED) Mercedes

Mobility and weighting schemes (II)• ECHP Mobility Wave 1 to wave 8 • Hart Index sensitive to the weighting scheme

– Country rerankings• Shorrocks Rigidity Index/Transition Matrices

sensitive to the weighting scheme (moderate)– Country rerankings

• Fields and Ok Index Decomposition sensitive to the weighting scheme– Growth mobility and transfer mobility– Descomp. by population subgroups highly sensitive

• Income mobility measures sensitive to the weighting scheme, specially on the disaggregated analysis

Page 22: CROSS-COUNTRY INCOME MOBILITY COMPARISONS UNDER PANEL ATTRITION: THE RELEVANCE OF WEIGHTING SCHEMES Luis Ayala (IEF, URJC) Carolina Navarro (UNED) Mercedes

Main Results• Attrition in the ECHP certain non-randomness• Estimation of longitudinal weights to adjust for

selective attrition (probit models)– Estimated longitudinal weights dif. with Eurostat weights

• Results Sensitiviness– Attrition no high effect on income mobility indices– Income mobility decomposition Greater sensitivity to the

weighting scheme “Correction” Adjustment throught longitudinal weights

• Adjustment with longitudinal weights– Some degree of sensitiviness: especially important for

some population subgroups