![Sample our Economics, Finance,Business & Industry journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5931/TOC-Subject-Sample-2021-Economics-Finance-Business-Industry.jpg"})
ABSTRACT
Unconventional monetary policy (UMP) influences inequality through two channels that work in opposite directions – a labour market channel (more employment, higher wages) and a financial market channel (higher asset prices). In an earlier paper, covering UMP through 2014, we found that UMP in Japan had contributed to greater income inequality through its effects on asset prices. With a longer time period, a richer dataset including labour market data, and a structural vector autoregression (SVAR) we confirm that these results continue to hold, and investigate why UMP’s impact on inequality in Japan differs from some other countries. We argue that Japanese structural issues may mute the labour market channel, especially: (i) labour market rigidity; and (ii) the large share of the population that is older than 65 years old or retired. The older cohort’s capital gains and dividends are re-saved in other financial assets, instead of being consumed or used for starting businesses. At the same time, wages have not increased despite the severe labour shortage, due to the frictions in Japan’s labour market. We conclude that these factors may make the inequality created by UMP in Japan unique by international comparison.
Acknowledgments
The views expressed here are solely those of the authors and do not necessarily reflect the views of the BIS or DNB. We thank Pierre Monnin (Center of Economic Policy), Takayuki Tsuruga (Osaka University), Gabriele Galati (DNB), Dietrich Domanski (Financial Stability Board), Ken Watanabe (Musashino University) and Nao Sudo (BoJ) and an anonymous referee for valuable comments.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1 Mio Tomita (Citation2019)l, ‘Bank of Japan to be top shareholder of Japan stocks’, Nikkei Asian Review, April 17.
2 The reason we did not include this channel in the previous study is that there was no significant change in the labour market conditions during our earlier sample period. However, in the last several years, the unemployment rate declined substantially (2.4% as of April 2019), reflecting labour shortage.
3 The ratio of the number of people whose income falls below the poverty line (defined as half the median household income of the total population).
4 In June 2016, Governor Kuroda noted at Keio University ‘it is my understanding that inequality has not risen in Japan.’
5 Inui et al. also replicate our study for the period Q4 2008 to Q2 2016 and find that the results are of a similar magnitude, but not (quite) statistically significant. Notably, this does not control for the consumption tax increase in April 2014 (see below).
6 Feldkircher and Kakamu (Citation2018) obtained a similar solution. However, we think using shadow interest rates, which require many assumptions to construct, is less preferable to using the monetary base, which does not require assumptions and is an actual tool of unconventional policy.
7 Regarding wealth inequality, Adam and Tzamourani (Citation2016) find that after UMP in the euro area, the capital gains from bond price and equity price increases turn out to be concentrated among relatively few households, while the median household strongly benefits from housing price increases. Guerello (Citation2018) finds there is significant heterogeneity in the impact of monetary policy on distribution across countries. Finally, there is a new strand of literature on how macroprudential policy affects income inequality; Frost and van Stralen (Citation2018) find a positive association between some macroprudential policies and market or net income inequality in a sample of 69 countries between 2000–2013.
8 The World Bank, World Development Indicators 2018.
9 We use 100 minus the unemployment rate, to be intuitively understandable and take account of quarterly fluctuations. In addition, since many ‘discouraged workers’ (housewives, elderly, ‘lost-generation’ workers) came back to labour market, this does not diverge much from the (lower-frequency) employment rate in Japan’s context.
10 For more detailed description of the survey, see https://www.stat.go.jp/english/data/kakei/1560.html (retrieved in June, 2019). The entire data can be downloaded from the website as well.
11 For households with two or more people, the survey is taken every month, with less granular contents.
12 Especially due to the additional restrictions from the SVAR model, the results become unstable or could not be estimated in our 2014 paper due to the smaller sample size.
13 A first difference is used due to a unit root in this variable.
14 The Tohoku Earthquake occurred on March 11th, 2011.
15 Source: Statistics from The Ministry of Wealth, Labour and Welfare.
16 According to the Tokyo Stock Exchange, TOPIX is ‘a free-float adjusted market capitalization-weighted index that is calculated based on all the domestic common stocks listed on the TSE First Sector.’
17 More precisely, we indexed each index so that the value of 2001Q1 will be 100, and weighted them accordingly, to control for the unit difference.
18 Not reported. Available upon request.
19 Ministry of Internal Affairs and Communication, Japan Statistic Yearbook.
20 According to the Income and Expenditure Survey of Japan Cabinet Office.