Consumption Consumption Amonthep Chawla Amonthep Chawla ( - PowerPoint PPT Presentation

Consumption Consumption Amonthep Chawla Amonthep Chawla ( Adjusted from Sang- -Hyop Lee Hyop Lee’ ’s s ( Adjusted from Sang Summer Seminar 2008 Presentation) Summer Seminar 2008 Presentation) 1 N ational T ransfer A ccounts

_ _ The Economic Lifecycle (per capita) (per capita) The Economic Lifecycle _ _ _ _ Consumption and Labor Income, Per Capita 1.200 _ _ 1.000 Labor Income 0.800 _ _ Consumption 0.600 _ _ 0.400 _ _ 0.200 0.000 0 10 20 30 40 50 60 70 80 90+ _ _ Age Note. Values normalized on per capita labor income of persons 30-49. _ _ 2 N ational T ransfer A ccounts

Outline Outline Private Consumption Public Consumption ► Education ► Education ► Education ► Education ► Health Health ► Health Health ► ► ► Owner occupied ► Other Other ► Owner occupied ► housing (imputed rent) housing (imputed rent) ► Durables Durables ► ► Other ► Other 3 N ational T ransfer A ccounts

Private Consumption Private Consumption ► Standard approach of a ► Standard approach of allocating llocating h household ousehold consumption among the onsumption among the m members embers did not provide did not provide c reasonable results reasonable results � Engel method � Engel method: f : food share is used to measure ood share is used to measure households’ ’ well well- -being being households � Rothbarth method � Rothbarth method: w : welfare measured by expenditure on elfare measured by expenditure on adult goods per adult adult goods per adult ► Alternative method ► Alternative method ( (NTA NTA) ) � Estimate � Estimate education and health education and health consumption directly consumption directly � Estimate private capital consumption (rental value of � Estimate private capital consumption (rental value of owner occupied housing + flow of services from owner occupied housing + flow of services from durables) durables) � Allocate other consumption indirectly (using Equivalence � Allocate other consumption indirectly (using Equivalence Scale) Scale) 4 N ational T ransfer A ccounts

Allocating Private Education Private Education Allocating Consumption Consumption ∑ ∑ = α + β edu C ( ) a E ( ) a ( ) a NE ( ) a j j j • Household education consumption is Household education consumption is • regressed on the number of enrolled (E) and regressed on the number of enrolled (E) and non- -enrolled (NE) in each age group. enrolled (NE) in each age group. non • The age groups included will vary with the The age groups included will vary with the • country and its enrollment rates. country and its enrollment rates. • Use unsmoothed profile Use unsmoothed profile. . • 5 N ational T ransfer A ccounts

Allocating Private Private Health Care Health Care Allocating Consumption Consumption ► Often v ► Often very complex in part due to various source ery complex in part due to various source of financing, which includes , which includes of financing � Private out � Private out- -of of- -pocket expense pocket expense � Private insurance � Private insurance � Public sector � Public sector ► Available sources of data vary across countries ► Available sources of data vary across countries. . ► There are differences between NHA and NTA ► There are differences between NHA and NTA � E.g. NHA document expenditures rather than � E.g. NHA document expenditures rather than consumption. Thus it includes profits of insurance consumption. Thus it includes profits of insurance companies. companies. ► Estimate using one of four approaches. Estimate using one of four approaches. ► 6 N ational T ransfer A ccounts

Approach 1: Method based on Approach 1: Method based on individual utilization measures from individual utilization measures from expenditure survey data expenditure survey data ∑ ∑ = α + β health C ( ) a IN ( ) a ( ) a OUT ( ) a j j j ► Private health consumption is regressed on ► ______ ______ ___________ __ ► ► Private health consumption is regressed on ______ ______ ___________ __ the number of members using inpatient _________ __ ___ ______ __ _______ ___ ______ __ _______ _________ __ the number of members using inpatient services (IN) and outpatient services (OUT) _____ _________ ________ ___ _____ _________ ________ ___ services (IN) and outpatient services (OUT) in each age group. __________ ________ __ ____ ___ __________ ________ __ ____ ___ in each age group. _____ _____ 7 N ational T ransfer A ccounts

Approach 2: Based on age profile of per Approach 2: Based on age profile of per capita utilization measures capita utilization measures = ∑ β h e a l t h C ( a ) U ( a ) M ( a ) j j ∑ ∑ = β + β h e a l t h C U ( a ) M ( a ) a U ( a ) M ( a ) j 0 j 1 j ∑ + β 2 a U ( a ) M ( a ) 2 j ► ______ ______ ___________ __ _________ __ ► Private health consumption is regressed on the ___ ______ __ _______ ___ ___ ______ number of members (M) and per capita utilization measure by age (U) ___________ _______ __ ___ ► Could be linear (the former) or non-linear (the ► ____ __ ______ ___ ______ __ ___ ______ ___ ______ latter) 8 N ational T ransfer A ccounts

Approach 3: Based on non- -parametric parametric Approach 3: Based on non iterative method iterative method ► Assign health expenditure equally to each Assign health expenditure equally to each ► household member and then tabulate the per household member and then tabulate the per capita profile. capita profile. ► The per capita profile is then used as weights to The per capita profile is then used as weights to ► allocate health expenditure to household members allocate health expenditure to household members producing a new per capita profile. producing a new per capita profile. ► Repeat until the weights do not change much. ► Repeat until the weights do not change much. ► Unlike regression approach, it does not produce Unlike regression approach, it does not produce ► negative coefficients for some age groups. negative coefficients for some age groups. 9 N ational T ransfer A ccounts

Estimated C after iteration Estimated C after iteration Age True C 1 2 3 4 5 6 Age True C 1 2 3 4 5 6 5 10 14.44 11.90 10.83 10.36 10.16 10.07 5 10 14.44 11.90 10.83 10.36 10.16 10.07 35 30 25.00 27.42 28.74 29.40 29.72 29.87 35 30 25.00 27.42 28.74 29.40 29.72 29.87 65 20 22.22 21.54 20.85 20.43 20.21 20.10 65 20 22.22 21.54 20.85 20.43 20.21 20.10 HH No H No I d No Age HH C C 1 2 3 4 5 6 H I d No Age HH 1 2 3 4 5 6 1 1 1 1 5 5 70 70 23.33 23.33 15.69 15.69 12.48 12.48 11.09 11.09 10.49 10.49 10.22 10.22 1 1 2 2 35 35 70 70 23.33 23.33 27.16 27.16 28.76 28.76 29.45 29.45 29.76 29.76 29.89 29.89 1 1 3 3 35 35 70 70 23.33 23.33 27.16 27.16 28.76 28.76 29.45 29.45 29.76 29.76 29.89 29.89 2 1 35 80 26.67 27.69 28.72 29.35 29.69 29.85 2 1 35 80 26.67 27.69 28.72 29.35 29.69 29.85 2 2 35 80 26.67 27.69 28.72 29.35 29.69 29.85 2 2 35 80 26.67 27.69 28.72 29.35 29.69 29.85 2 3 65 80 26.67 24.62 22.56 21.30 20.63 20.30 2 3 65 80 26.67 24.62 22.56 21.30 20.63 20.30 3 1 65 40 20 20.00 20.00 20.00 20.00 20.00 3 1 65 40 20 20.00 20.00 20.00 20.00 20.00 3 2 65 40 20 20.00 20.00 20.00 20.00 20.00 3 2 65 40 20 20.00 20.00 20.00 20.00 20.00 4 4 1 1 5 5 20 20 10 10 10.00 10.00 10.00 10.00 10.00 10.00 10.00 10.00 10.00 10.00 4 4 2 2 5 5 20 20 10 10 10.00 10.00 10.00 10.00 10.00 10.00 10.00 10.00 10.00 10.00 10 N ational T ransfer A ccounts

Approach 4: Based on simple regression Approach 4: Based on simple regression = ∑ β health C ( ) a M ( ) a j j ► Private health consumption is regressed on the number of household members (M). ► Could have negative coefficients—replace with zero. ► The least recommended approach. 11 N ational T ransfer A ccounts

Estimating Other Household Estimating Other Household Consumption Consumption β = − ≤ ( ) a 1 0.6 (for a 4) β = − − ( ) a 1 [0.6*(20 a )]/16 (for 4< <20) a β = ≥ ( ) a 1 (otherwise, i.e., a 20) ► Assumed to be proportional to an equivalence Assumed to be proportional to an equivalence ► scale that is equal to 1 for adults aged twenty scale that is equal to 1 for adults aged twenty or older, declines linearly from age 20 to 0.4 or older, declines linearly from age 20 to 0.4 at age 4, and is constant at 0.4 for those age at age 4, and is constant at 0.4 for those age 4 or younger. 4 or younger. 12 N ational T ransfer A ccounts