So people don’t spend when they win a lottery?

According to the leading market monetarist, Scott Sumner, having the state print fresh money and spend it (and/or cut taxes) won’t have much effect. Well strikes me that if the state prints extra money and for example uses the money to employ an extra thousand teachers, then employment will rise by one thousand all else equal. Or perhaps I’ve missed something. I’m assuming of course that the relevant economy is not short of skilled labour, teachers in particular. That is, if the relevant country WERE SHORT of skilled labour, then increased demand for teachers could just cause extra inflation.

Moreover, that extra money ends up in the bank accounts of sundry households, which has a secondary effect: it induces those households to spend more. After all, peoples’ spending does rise when they win a lottery, doesn’t it? Or perhaps, again, I’ve missed something.

Japan.

By way of trying to substantiate his point, Scott Sumner cites Japan, which according to Scott Sumner “did an almost unbelievably large money-financed fiscal expansion over the past 20 years and got DEFLATION…”.

Now the big problem there is that the Japanese are the world’s champion savers: you could say they are obsessed by saving. Thus Japan is PRECISELY the country in which the “print and spend” policy will have the least effect per dollar printed and spent. Thus if Scott Sumner is looking for evidence to back his point, Japan is the LEAST SUITABLE country to choose. That is, Scott Sumner’s argument is a bit like looking at death rates in a hospital for the terminally ill, and concluding that hospitals don’t do much to improve peoples’ health.

As to the above mentioned “unbelievably large money-financed fiscal expansion”, the ACTUAL AMOUNTS involved were modest, far as I can see. Briefly, the addition to the Japanese monetary base over the 20 year period was the equivalent of 750 US dollars per person per year. Now it wouldn’t surprise me if that’s about the amount of saving the Japanese would actually WANT TO DO, and if I’m right there, then Sumner’s point starts to look shaky.

I arrived at that $750 figure thus (and I may well have got something wrong).

According to the “Long term time series data” file here, the Japanese monetary base rose by about 2¹⁴ Yen between 1994 and 2014. Taking the population of Japan as 130million, 2¹⁴/130,000,000 is 1,500,000. Divide that by the above mentioned 20 (years) and by 100 for the Yen/USdollar exchange rate and you get $750.