How Population Aging
Affects the Macroeconomy
Ronald Lee
University of California at Berkeley
Demography and Economics
rlee@demog.berkeley.edu
PANEL ON DEMOGRAPHICS
Jackson Hole, August 22, 2014
Ronald Lee, UC Berkeley, 8/22/14, Jackson Hole
1
NTA
• NTA goes beyond national accounts in two important new
ways.
• breaks down national accounts by age.
• estimates transfers within families and households, between
households, and through the public sector.
• Andy Mason and I co-direct NTA
• 45 countries have NTA teams working on their accounts.
• Based on existing surveys and administrative data
• Centralized methods, quality control, training, workshops.
Ronald Lee, UC Berkeley, 8/22/14, Jackson Hole
2
Countries with NTA Teams
Ronald Lee, UC Berkeley, 8/22/14, Jackson Hole
3
Recent publications, both free downloads
(see ntaccounts.org)
Also coming: special NTA issue of Journal of Economics of Aging
Ronald Lee, UC Berkeley, 8/22/14, Jackson Hole
4
The basic budget identity: Inflows = Outflows at each age
for individual or for generation
Income inflows
=
Expenditure outflows
Asset Inc + Labor Inc + Transfers Rcvd = Cons + Transfers Given + Saving
OR, rearrange to
Life cycle deficit
=
Reallocations
Cons - Labor Inc = Transfers Rcvd - Transfers Given + Asset Inc - Saving
NTA estimates these flows, and subcomponents, public and
private, and by specific type. “Age Profiles”.
Ronald Lee, UC Berkeley, 8/22/14, Jackson Hole
5
Per Capita Value, Relative to Avg Labor income
Average Consumption and Labor Income for 6 High Income and 6
Developing Countries
1.2
Developing: Kenya, Nigeria, India,
Philippines, China, Indonesia
High Income: Germany, Japan,
Austria, Finland, Sweden, United
1.0
Cons-Hi Inc
0.8
Cons-Developing
0.6
0.4
Labor Inc-Developing
0.2
Labor Inc-Hi
0.0
0
10
20
30
40
50
60
70
80
90
Age
Ronald Lee, UC Berkeley, 8/22/14, Jackson Hole
6
Age profiles may change: US cons over past half century:
1960, 1981 and 2007 (Ratio to labor income ages 30-49).
1960
1981
2007
Public Education
1
1
1
Public
Health
Private Education
Private Health
Owned Housing
0.5
0.5
0.5
Private Other
0
0
0
10
20
30
40
50
60
70
80
90
Public Other
0
0
10
20
30
40
50
60
70
80
90
0
10
20
30
40
50
60
70
80
90
Source: US National Transfer Accounts, Lee, Donehower and Miller,
2011
Ronald Lee, UC Berkeley, 8/22/14, Jackson Hole
7
Ronald Lee, UC Berkeley, 8/22/14, Jackson Hole
8
Population age distributions and support ratios
• How many working age people are available to support a society’s
consumers?
• Calculate hypothetical workers and consumers multiplying changing
population age distributions times baseline NTA age profiles.
• “Support ratio” is:
hypothetical w orkers
hypothetical consum ers
• Resources available per capita are proportional to this support ratio.
Ronald Lee, UC Berkeley, 8/22/14, Jackson Hole
9
Ronald Lee, UC Berkeley, 8/22/14, Jackson Hole
10
Support ratios based on the average developing country
profiles and UN Population Projections
Trough to Peak
Peak to 2100
Rate of change of support ratio
China
India
Nigeria
Costa Rica
0.67
0.37
0.27
0.67
-0.26
-0.17
na
-0.31
Ronald Lee, UC Berkeley, 8/22/14, Jackson Hole
11
Support ratios based on the average rich country
profiles and UN Population Projections
Rate of change of support ratio , 2010-2050 (%)
Germany
-0.66
Japan
-0.66
Spain
-0.78
US
-0.34
Ronald Lee, UC Berkeley, 8/22/14, Jackson Hole
12
How the “life cycle deficit” is financed at each age (US
2003)
Financing the Lifecycle Deficit
Components at Each Age
2
Units of Avg YL 30-49
1.5
Private transfers
Public transfers
Asset income - savings
1
Pub Trans
ABR
0.5
0
Priv Trans
0
5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90
-0.5
-1
Ronald Lee, UC Berkeley, 8/22/14, Jackson Hole
13
For Elderly (65+): How is consumption net of labor income funded? Shares of Family
Transfers, Public Transfers and Asset income not saved sum to 1.0
Ronald Lee, UC Berkeley, 8/22/14, Jackson Hole
14
The “general support ratio” (GSR) reflects both labor income
and asset income: how dependent are the elderly?
• Do the elderly actually depend on workers to fund their
consumption?
• Suppose elderly use their own savings for consumption?
• GSR reflects use of asset income by elderly to fund own consumption
(asset inc – saving).
• GSR isolates the impact of population aging on transfers (public and
private).
• Change in GSR over time shows consequences of pop change if age
profiles of consumption, labor income, and asset income remain
constant.
Ronald Lee, UC Berkeley, 8/22/14, Jackson Hole
15
Standard support ratios (blue) and general support ratio (red),
scaled to 1.0 in 2010
Ronald Lee, UC Berkeley, 8/22/14, Jackson Hole
16