Intergenerational Conflict over Fiscal Consolidation:
Theory and Evidence from Japan
Real Arai
1Ryosuke Okazawa
2Katsuya Takii
31Kochi University of Technology
2Osaka City University
3Osaka University
January 19, 2018 Very Preliminary
1 / 31
Introduction
Consumption tax in Japan
Abe cabinet postponed an increase of consumption tax by 10%
on May 2016.
The increase of consumption tax has been postponed, although Japan’s public debt is severely large.
Why is it difficult to raise consumption tax?
→A political reason probably affects the situation.
2 / 31
Introduction
Political Economics of public deficit
“Inter-generational conflict” is one of important issues.
Ihori (2016, in book, translated by the authors)
“As the difference in preference by age increases, the current beneficiary tends to hope for an increase in benefit payments but is reluctant to bear their costs, and the reforms tend to be postponed. As a result, the fiscal burden is imposed on the future beneficiary. If the policy is determined by “silver democracy,” it causes the accumulation of fiscal deficits and an increase in social security benefits.”
3 / 31
Introduction
Inter-generational conflict and politics
Should we lower the political powers of older people? Election system reform is discussed, e.g.,
reallocating representatives’ seats to each generational group (Ihori and Doi, 1998)
reallocating representatives’ seats based on voter’s life expectancy (Takeuchi, 2012)
Demeny voting (Demeny, 1986)
c.f. The voting rights in the parliamentary election in Japan extended to 18 and 19 year-old citizens in 2016.
4 / 31
Introduction
Objectives of this study
1 Using microdata of JGSS 2010 and 2012, we show features of political preference on a consumption tax increase.
→We empirically verify the prevailing view on
inter-generational conflict on a consumption tax increase.
2 We propose and test three hypotheses that can explain the different preferences among generations.
5 / 31
Introduction
Objective 1: Empirical results
The older voter tends to support an increase of consumption tax, which is contrast to the prevailing view and to the prediction of the basic model.
The people who have higher education or income tend to support a consumption tax increase.
Women tend disagree with a consumption tax increase.
Whether a voter has a child does not affect the preference on a consumption tax increase.
6 / 31
Introduction
0.1.2.3.4.5Share of Supporters
20−30 31−40 41−50 51−60 61−70 70−
Support for Consumption Tax Increase By Age
Figure:Support for Consumption Tax Increased by Age (source: JGSS2010, 2012)
7 / 31
Introduction
Objective 2: Why is older voter more likely to support for an increase of consumption tax?
The baseline theoretical model predicts that “older people tend to disagree with an increase of consumption tax”.
We consider three hypotheses: older voter tends to
1 prefer “a large government” because they obtain more benefits from public goods and services than young voter.
2 be afraid of risks of “an inflation tax” because they hold more financial assets.
3 understand well the current situation about Japan’s public finance.
Our empirical results support hypotheses 2 and 3.
8 / 31
Literature Review
Empirics: Preference for fiscal consolidation with individual survey data
Blinder and Krueger (2004) Stix (2013)
Hayo and Neumeier (2017)
Theory 1: Welfare analysis of fiscal consolidation in multi-overlapping generations models
Ihori et al. (2006) Yamada (2011) Okamoto (2013)
Theory 2: Politico-economic theories on public debt with inter-generational conflict
Tabellini (1991) Song et al. (2012)
9 / 31
Data and Facts
We analyse preferences for an increase of consumption tax with microdata of JGSS 2010 and 2012.
The central question: “At what level do you think the consumption tax rate should be?”
→If a person choose the alternatives with tax more than 5%, he/she supports for the consumption tax increase.
JGSS includes rich dataset, e.g., income, education, family, opinions for policies
Sample size: 5,140.
10 / 31
Data and Facts
0.1.2.3.4.5Share of Supporters
20−30 31−40 41−50 51−60 61−70 70−
Support for Consumption Tax Increase By Age
Figure:Age and Support for an Increase of Consumption Tax
11 / 31
Data and Facts
0.1.2.3.4.5Share of Supporters
0−350 350−650 650−1000 1000−
Support for Consumption Tax Increase By Income
Figure:Household Income and Support for an Increase of Consumption Tax
12 / 31
Data and Facts
OLS Probit Ordered Probit
age 0.0051***0.0141*** 0.0135***
female -0.1140***-0.3081*** -0.1983***
marry 0.0370** 0.1039** 0.0942**
havechildren -0.0097 -0.0315 0.0057 highschool 0.1178***0.3248*** 0.3102***
university 0.2804***0.7572*** 0.7343***
household income 0.0109***0.0296*** 0.0304***
N
5140 5140 5140
(We control the employment status and survey year. )
13 / 31
Data and Facts
Results of regression analysis:
Older voter is more likely to support a consumption tax increase.
Women significantly tend to disagree with an increase of consumption tax.
Married people significantly tend to agree with an increase of consumption tax. However, the tendancy is not affected by the People who experienced higher education or receive higher incomes tend to agree with an increase of consumption tax.
14 / 31
Data and Facts
Robustness Check 1
The result on the relationship between age and support for consumption tax does not change, even though
we use cohort dummies instead of age itself.
we control living city size, health status, house owner, and experiences of charity or volunteer.
15 / 31
Data and Facts
Robustness Check 2
Because of older people’s altruism (toward the future generations)?
To test, we use “opinions for the free education policy in senior public high schools”.
“Are you in favor of policies to make the tuition of public or private school free?”
→Older people are significantly less likely to agree with the public subsidy for education.
→Altruism does not explain the result on age and the support for consumption tax.
16 / 31
Data and Facts
Robustness Check 3: Using the UTAS data in 2012 and 2013.
Survey for 3,000 citizens randomly sampled just after the parliamentary election in 2012.
Questions on citizens’ opinions for policies:
“For the time being, the government should increase spending to stimulate economic activities rather than cut spending for fiscal consolidation. ”
“In the long run, it is inevitable that the consumption tax rate will be higher than 10%.”
We regress on age, gender, education, and employment status (ordered probit).
Note: There is no information on income in the UTAS.
17 / 31
Why Does the Old Support Tax Increase?
Why are older people more likely to support a consumption tax increase?
The baseline model cannot explain the result.
We propose the following three hypotheses.
1 Older people prefer “a large government”: they want more public goods/services.
2 Older people are more afraid of falling the value of government bond.
3 Older people know the situation about public finance more.
→
To derive the hypotheses, we extend the theoretical model in our paper.
18 / 31
Empirical Test
Empirical analysis with preference for government spending
We test the hypothesis 1.Do older people support the consumption tax increasebecause they demand public goods/services more than young?
We test the following two statements to check hypothesis 1.
1 Do older people have a significant demand for government spending?
2 Do people with a high demand for public goods/services tend to support the consumption tax increase?
19 / 31
Empirical Test
Empirical Analysis with preference for government spending
We use the results on the following questions.policy on redistribution:
“Are you in favor of the statement that the government should adopt policies to reduce the income gap between the rich and the poor?”
opinion for life security of the elderly:
“Is the life security of the elderly the responsibility of individuals or families? Or is it the responsibility of the state or local government?”
(“responsibility of individuals or families” = “1”,
“responsibility of the government” =“5”) opinion for health care for the elderly:
“Is health care for the elderly the responsibility of individuals or families? Or is it the responsibility of the state or local government?”
20 / 31
Empirical Test
0.2.4.6.8Share of Supporter
20−30 31−40 41−50 51−60 61−70 70−
Policy Preference and Age
Redistribution Economic Support for the Elder Medical Care for the Elder
Figure:Age and Policy Preferences
21 / 31
Empirical Test
redistribution healthcare safetynet
age 0.0039*** -0.0015 -0.0035***
female -0.0761** -0.0091 -0.0843**
marry 0.0259 0.0554 0.0736*
havechildren 0.0881* 0.0103 -0.0149
highschool -0.1751*** 0.0314 -0.0662 university -0.3791*** -0.0883 -0.1941***
household income -0.0394*** -0.0143*** -0.0166***
N
5129 5091 5091
22 / 31
Empirical Test
(1) (2) (3) (4) (5)
age 0.0052*** 0.0051*** 0.0050*** 0.0051*** 0.0048***
redistribution -0.0278*** -0.0193***-0.0195***
healthcare -0.0294*** 0.0006 0.0001
safetynetforold -0.0390***-0.0358***-0.0349***
ldp support 0.0336*
dp support 0.1033***
N 5129 5091 5091 5072 5060
23 / 31
Empirical Test
Empirical results with preference for government spending
There is no evidence that older people tend to demand public goods/services more.People who demand public goods/servicesmore tend to disagreewith a consumption tax increase.
The positive relationship between age and support for the consumption tax increase still remains even though policy preferences are controlled.
→
Hypothesis 1 is not supported.
24 / 31
Empirical Test
Empirical analysis with financial asset holdings
We test the hypothesis 2.We use the qustions on the following financial asset holdings in JGSS 2012 (only sheet B).
“Do you have stocks / bonds / an investment-trust / foreign deposits?”
→We test whether support for the consumption tax increase is associated with financial asset holdings.
25 / 31
Empirical Test
Empirical analysis with acquisition of information
We test the hypothesis 3.We use “frequency of reading newspaper” as a proxy of acquisition of information.
“How frequently do you read newspapers?”
“almost everyday”=“3”, “several times a week”=“2”, “about once a week”=“1”, “never”=“0”.
We test the relationship between reading newspapers and support of the consumption tax increase.
26 / 31
Empirical Test
0.05.1.15Share of Asset Hold
20−30 31−40 41−50 51−60 61−70 70−
Financial Assets and Age
mean of stock mean of bond
mean of investmenttrust mean of foreigndeposit
Figure:Age and Financial Asset Holdings
27 / 31
Empirical Test
0.2.4.6.8Read Newspaper Everyday
20−30 31−40 41−50 51−60 61−70 70−
Newspaper and Age
Figure:Age and Frequency of Reading Newspapers
28 / 31
Empirical Test
(1) (2) (3) (4)
age 0.0051***0.0042***0.0044***0.0034***
bond 0.0781* 0.0807*
stock 0.0705* 0.0675*
investmenttrust 0.0773* 0.0750*
foreigndeposit 0.0285 0.0198
newspaper 0.0233***0.0291***
N 5140 1720 5126 1716
29 / 31
Empirical Test
Empirical results with financial asset holdings and frequency of reading newspapers.
People who holds domestic financial assets tend to support for consumption tax increase.
Foreign deposit holding does is not associated with the support for the tax.
People who read newspapers more tend to support for a consumption tax increase.
Controlling financial asset holdings and frequency of reading newspapers decreases the coefficient of age.
→Risks of government bond holdings and information can partly explain the effects of age on support for consumption tax increase.
30 / 31
Conclusion
Conclusion:
Using JGSS data, we find that older people are more likely to support the consumption tax increase, which is inconsistent with the standard view and theoretical model.
We cannot find that the hypothesis, “older people prefer a larger government”, explains the relationship between age and the support.
We find that risks of holding public debt and acquisition of information partly explain the relationship between age and the support.
We should consider the commitment effect of public debt holding in order to consider the inter-generational conflict on fiscal consolidation.
→Our analysis can apply to investigation on fiscal consolidation plans in other countries (e.g. EU countries).
31 / 31
