Did Japan's shopping coupon program increase spending?

a r t i c l e i n f o In March 1999, 31 million " shopping coupons " worth 20,000 yen each were distributed to Japanese families with children and to the elderly. The coupons expired after six months and could only be used within the recipient's local community. We use variation in the number of children across families and in the number of recipients across prefectures to measure the effect of the coupons on spending. We find that coupons had a positive effect on spending on semi-durables, but no effect on spending on nondurables or services. The marginal propensity to consume on semi-durables was 0.1–0.2 when the coupons were distributed in March. The results using regional variation provide stronger evidence that spending did not fall after the coupons had been redeemed. In the spring of 1999 the Japanese government distributed shopping coupons worth 20,000 yen (about 200 dollars) to families with children under the age of 15 and to more than half of the elderly population. In total, 620 billion yen (about 6 billion dollars) worth of coupons were distributed to 31 million people. The coupons had to be spent in the recipient's local community and expired within six months. The Japanese government's rationale for the " use-it-or-lose-it " nature of the coupons was that this unusual feature would stimulate more spending than a conventional tax cut. The shopping coupon program is widely viewed in Japan as having been modestly successful in stimulating spending. In 2009, 10 years later, Japan implemented a similar program, this time providing coupons worth 12,000 yen to every resident (regardless of age or income), with an additional 8000 yen for individuals under 18 or over 65. Other countries appear to have drawn similar conclusions from Japan's earlier experience with shopping coupons. Taiwan, for example, distributed coupons worth about 120 dollars to every citizen in February 2009. However, there are good reasons to believe that there is nothing particularly special about Japan's coupon program. First, the fact that the coupons expired if not used would only have an effect on spending if households would have spent less than the amount of the coupons. And even in this case, households that spent more than they would have in the absence of the coupons may offset this spending by reducing spending in the future. Second, the fact that households had to spend the coupons in …