How Inclusive Is Abenomics?; by Chie Aoyagi, Giovanni Ganelli, and Kentaro Murayama; imf working Paper No. 15/54; March 1, 2015

 10 

In the long run, a high child poverty rate may exhibit ‘lock-in’ effects of poverty across 

generations, since poor households are less 

likely to afford education, and low 

educational attainment is likely to affect the 

future income of the children. In 

macroeconomic terms, this would imply a 

decrease in aggregate human capital, and 

therefore in potential growth, once those 

children enter the labor market. For 

instance, Heckman (2000) argues that the 

lack of early education and training can be 

costly as the returns are diminishing with 

age.  

Moreover, not only children (who are part of the dependent population), but the young 

population in general is increasingly poor, according to the historical data on the relative 

poverty rate by age group. These data show a rapid increase in poverty rates among 0-17 and 

18-25 age groups, and less marked increases in poverty for age brackets from 26 to 65, while 

poverty rates for those ages 66-75 has declined over time (text chart). This observation 

echoes the findings in generational differences in income inequalities discussed above. 

 

 

C.   Other Dimensions of Inequality  

The data analyzed so far suggest that income disparities are widening in Japan amongst the 

working-age population. The drivers of such disparities need to be studied more in detail, but 

there are at least two prominent dimensions over which income inequality – or, primarily 

0

0.05

0.1

0.15

0.2

0.25

0.3

All age 

groups

Age group 

0-17

Age group 

18-25

Age group 

26-40

Age group 

41-50

Age group 

51-65

Age group 

66-75

Age group 

76+

By age groups

1985

1995

2000

2006

2009

Poverty Rates by Age Groups

(% of  households  for each  group)

Source: Ministry  of Health,  Labour and Welfare

0

10

20

30

40

50

60

70

One Adult (%)

Two or More Adults (%)

Child Poverty Rates 

by the number of adults in the household

Source: Ministry of Health, Labour and Welfare (2013)

---

 11 

wage inequality for young, working population – are observed: namely, the gender gap and 

labor market duality. 

Labor participation rates, defined as the 

ratio of the labor force to the population 

of age above 16 years old, are declining 

in general due to Japan’s aging 

population. Nevertheless, the Female 

Labor Participation (FLP) rate is lower 

than the Male Participation Rate (MPR) 

by about 20 percentage points (text 

chart).  

Low labor force participation and 

underemployment of women imply a 

lack of inclusiveness in the process of growth, which cannot be fully captured by household-

based poverty measure. As it has been strongly emphasized by the IMF, low FLP is also 

costly in terms of reduced potential growth.

 3

 The problem is compounded by the fact that the 

Japanese economy has been experiencing negative growth of labor input for years and is 

facing labor shortages in more recent 

years.  

Another important driver of inequality 

is labor market duality. According to 

data by the Ministry of Internal Affairs 

and Communications (MIAC), the 

share of non-regular workers 

consistently increased from less than 20 

percent in the 1980s to above 35 

percent by 2011(text chart). Aoyagi and 

Ganelli (2013) stress that such 

excessive labor market duality is likely 

to be holding back growth by reducing productivity. The two factors discussed here, low FLP 

and duality, are interrelated, since, as discussed by Aoyagi and Ganelli (2013), more than 

half of employed women are non-regular workers, with less job security, lower wages, and 

reduced career opportunities.  

                                                 

3

 “The Economic Power of Women’s Empowerment” speech by Christine Lagarde, Managing Director, 

International Monetary Fund, Tokyo, Japan, September 12, 2014. Available online at 

http://www.imf.org/external/np/speeches/2014/091214.htm 

40

50

60

70

80

90

1974

1980

1986

1992

1998

2004

2010

Labor Participation Ratios by Gender (%)

above 15 

years old 

(Female)

15-64 

years old 

(Female)

above 15 

years old 

(Male)

15-64 

years old 

(Male)

Source: Minister of Internal Affairs and Communications

10

15

20

25

30

35

40

Non-regular staff (Shares)

Part-time (Share)

Shares of Non-regular Workers (%)

Sources: Ministry of Health, Labour and Welfare

---

 12 

In summary, the evidence presented in this section shows that both inequality and relative 

poverty have increased in Japan in recent decades, and suggests that, with the bulk of fiscal 

redistribution benefitting the elderly, the economic burden of rising inequality and poverty is 

concentrated in a disproportionate way on children, women and non-regular workers. This 

observation is particularly relevant and important when growth of the economy on average is 

promoted without considering inclusiveness. This begs some questions on the growth that 

implementation of Abenomics reforms is likely to generate. Will such growth be inclusive or 

will it create more inequality? If the latter is true, will the increased inequality be 

compensated by average income growth, so that those who are left behind can still enjoy 

some of the prosperity that comes with the growth of the economy? What would be 

inequality implications of successfully exiting deflation and of structural reforms in the labor 

market? The rest of this paper seeks to address such issues in a systematic way by conducting 

an econometric analysis on the impact of key policy variables on a measure of inclusive 

growth. 

 

IV.   D

ATA AND 

E

MPIRICAL 

S

TRATEGY

 

We use prefectural level longitudinal data. Data on income distribution by prefecture are 

obtained from the National Population Census, which is conducted every 5 years. Income 

distributions are available for aggregate income, which consists of wages, interest, rent, 

social security and other payments to households. Data are compiled for the whole 

population and a subset of working-age households. Incomes observed for each prefecture 

are deflated by the GDP deflators of the corresponding prefecture, which are provided by the 

Cabinet office. 

Our measure of inclusive growth is the one developed by Anand et al. (2013). Intuitively, it 

is a weighted average of growth in average income and of the change in an equity index 

which takes into account income distribution. The equity index is built in a way that it is 

bounded between zero and one, with one being a perfectly equitable income distribution. 

This measure of inclusive growth is equivalent to average income growth in the hypothetical 

case of growth which leaves income distribution unchanged, but deviates upward 

(downward) from average income growth when growth is achieved by making income 

distribution more equal (unequal). In other words, our proxy can be interpreted as a measure 

of growth in average income “corrected” for the equity impact. For a more technical 

discussion, see the appendix.

 4 

 

The distribution of the average real income growth and the growth in our equity index 

growth by prefectures is shown in the text chart. It is clear that observations are clustered by 

years. With some periods (1979-84; 1984-1989)  being  characterized by high growth in 

average income, which tends to be negative in other periods (especially 1999-2004). The 

variation in the equity index growth shows a less clear pattern.  

                                                 

4

 For the definition, derivation, and a more technical discussion, see Appendix A. 

---