Gary R. Saxonhouse and Robert M. Stern
or minus 1% over a three-year horizon. To achieve this end, the Bank of Japan 110
should buy large amount of market-indexed ETFs (Exchange Traded Funds) and 111
market-indexed REITs (Real Estate Investment Trusts). If this measure is not ef- 112
fective, the government should break the nominal interest rate floor by levying a 113
tax on all government guaranteed financial assets. 114
Willem Buiter provides a somewhat different view in “Overcoming the Zero 115
Bound on Nominal Interest Rate: Gesell’s Currency Carry Tax vs. Eisler’s Parallel 116
Virtual Currency.” He notes that despite the zero lower bound on the short-term rate 117
in Japan having become a binding constraint, conventional monetary policy in the 118
form of generalized open-market purchases of government securities of all 119
maturities has not been pushed to the limit where all outstanding government debt 120
and all current and future government deficits are or are confidently expected to be 121
monetized. As long as the risk-free nominal interest on financial instruments of any 122
maturity remains positive, a currency tax, in Buiter’s view is unwarranted. 123
Unlike Buiter, Fukao does not think that the massive open-market purchases of 124
long-term bonds is feasible. If the Bank of Japan holds a massive amount of long- 125
term JGBs when long-time interest rates rise it will experience a very large capital 126
loss. At the very time it is experiencing this loss, it will also be necessary for it to 127
raise short-term interest rates by mopping up excess liquidity. Fukao believes that 128
before it will be able to accomplish this completely the Bank of Japan will run out 129
of assets to sell. If it is to continue with its open-market operation it will be forced 130
to issue interest-bearing promissory notes. According to Fukao, this will result in 131
the Bank of Japan being forced to turn to the government for a subsidy. 132
In contrast, not only does Buiter feel charging negative interest on money is 133
unnecessary, he worries that it will be administratively difficult. The original 134
proposal made by Silvio Gesell that currency be stamped to indicate that it is 135
current on interest rates is feasible, but Buiter feels it will be costly and intrusive, 136
and, therefore, should only be used as a last resort. Buiter notes that a proposal 137
made by Robert Eisler in 1932 overcomes the problems of both cost and in- 138
trusiveness by unbundling the medium of exchange/means of payments function 139
from the numeraire function of money through the creation of a parallel, virtual 140
money. By controlling the exchange rate between the means of payment and the 141
numeraire, specifically by appreciating the value of the currency in terms of the 142
numeraire, monetary authorities such as the Bank of Japan, can achieve a negative 143
interest rate on the numeraire even though the interest rate on currency remains 144
constrained by the lower bound. 145
Eisler’s proposal raises the fundamental issue about who chooses or what 146
determines the numeraire used in private wage and price contracts. According to 147
Buiter, Eisler’s implicit assumption that monetary authorities can credibly deter- 148
mine the numeraire used in private wage and price contracts is shaky both a priori 149
and empirically. What might otherwise be a neat solution to a difficult problem 150
remains suspect. 151
2.5 Supply shocks and Japanese deflation 152
As noted, Hamada and Noguchi, in their paper for this Symposium, dismiss the role 153
of positive supply shocks as a cause of Japan’s deflation. Indeed, they attribute the 154
long delay in Japan’s recovery to the mistaken view that Japan’s deflation reflects 155