Micro Data Studies on Japanese Tax Policy and Consumption in the 1990s

Part I : Consumer Response to the 1998 Tax Cut: Which of Tax Cut or Coupon Program is More Effective? The Japanese Government implemented two tax cut packages in 1998. A unique feature of the 1998 tax cuts is that the tax cut amount for each household is based on the number of family members regardless of income level. A wage-earner's household receives the tax cut benefit as a decrease in withheld taxes. Wage earners are entitled to receive a national income tax cut of up to 18,000 yen plus 9,000 yen for each dependant (effective from February) as well as a local inhabitant's tax cut of up to 17,000 yen plus 8,500 yen for each dependant (effective from June). Using monthly data on worker's households from the Family Income and Expenditure Survey (FIES), in this study we explore the variation in the impact of the tax cut episode across families with different numbers of family members to identify its impact on consumption. The estimates using monthly change in consumption indicate that consumers responded to the 1998 tax cut with the implied MPC of 0.6 on impact with stimulating services consumption but the MPC declined in subsequent months. A comparison of the impact on consumption with the shopping coupons program administered in the spring of 1999 based on the same specification demonstrates that the 1998 tax cut and the coupon program affected consumption of different types of goods and that the estimated MPC was roughly comparable and followed the similar pattern. Part II : Consumer Response to the 1994 Tax Cut: Evaluating the First Tax Cut in the 1990s. This paper explores the reaction of Japanese consumers to the 1994 tax cut, which was the first tax cut administered by the Japanese government in the 1990s. Micro-level data from the Family Income and Expenditure Survey (FIES) is used to evaluate the effect of the 1994 tax cut on consumption. The estimates weakly suggest that tax cuts undertaken to stimulate the weak economy in 1994 had some effect on consumption of non-durables or semi-durables. An MPC of 0.1 - 0.2 is estimated right after implementation, but the positive effect was substituted immediately in subsequent months. In other words, consumers reacted to the temporary tax cut but the effect was small and transitory. Part III : Consumer Response to the 1995/1996 Tax Cut: Are Temporary Tax Cuts Effective in Anticipation of Tax Increase? This paper tries to evaluate the effects of the 1995/1996 tax cuts administered by the Japanese government. These tax cuts were implemented exactly in the same manner at different timing. Especially, the 1996 tax cut is unique in the timing, i.e. it was implemented only three months before the consumption tax rate increase from 3% to 5%. We use micro-level data of the Family Income and Expenditure Survey (FIES) to evaluate the effect of the 1995 and 1996 tax cuts implemented in December in these two years on individual household's consumption, focusing on the difference between the two episodes. The estimates indicate that these tax cuts actually stimulated consumption on impact, though the effect decreased in following months, and also suggest that the effect was smaller for the 1996 tax cut than the 1995 tax cut. This finding probably suggests that the Ricardian equivalence story might hold and individual households do not respond to a tax cut in anticipation of coming tax increase, when the timing of tax reduction and tax increase is very close.