Richard Koo on fiscal policy and interest rates

[Skip to the end]

Met Richard years ago. Seems he’s still confused on fiscal policy:

Bond issues to fund capital injections will not lead to higher interest rates

Right! The CB sets rates. Too bad he didn’t stop here rather than try to explain the process.

Japan’s second round of capital injections was four times the size of the first, and some question the ability of US capital markets to absorb such a large emission of government debt. However, the 1989 S&L crisis demonstrated that funds raised for the purpose of rescuing the financial sector will not lead to higher interest rates.


This is because, unlike fiscal outlays for public works, money spent to rescue the financial system does not reduce the amount of investment funds available in the financial markets.


Assume, for example, that the government issues $100 of Treasury bonds to recapitalize a troubled bank.

And then makes a payment to the bank.

The bank receiving the capital injection would credit its capital account by $100 and then invest that $100. In effect, there will be $100 in the market to be invested regardless of whether the government issues debt to rescue the bank.

The $100 gets credited to the banks account at the CB. The bank can leave it there or look for alternatives.

Purchases of alternatives in the private sector cause the banks $100 to be ‘wire transferred’ to another bank.

That means the bank’s account at the CB is reduced by $100 and another bank’s account at the CB is increased by $100.

Because the $100 represents capital, the bank’s investment should be liquid and easily convertible into cash. The asset that best fills this bill is government securities


If the bank decides to buy government debt with the money, the government will have another $100 to fund a capital injection.

I assume he means new government debt as he started with the government issuing $100 of bonds and recapitalizing the bank.

Two rubs.


The government would only issue additional bonds if it wanted to (deficit) spend additional funds.

And it if wanted to issue bonds and (deficit) spend new funds, it would do so whether this particular bank wanted to buy the bonds or not. That is, the bank wanting to buy bonds is not the enabling force for (deficit) spending.


The sale of the original $100 of bonds reduced total bank reserves by $100 and the payment of the $100 to the bank added $100 to total bank reserves. So the initial bond issue and the recapitalization left bank reserves offset each other leaving total bank reserves unchanged. Institutionally, issuing new bonds starts a new series of transactions, and, again, that particular bank is not the enabling force.

If, on the other hand, the government uses that $100 to build bridges or roads, that money will leave the capital markets and be spent on wages or construction materials, producing a corresponding decrease in the amount of investment funds available.

I don’t follow this distinction at all.

In this case, as before, the Treasury borrowing $100 reduces bank balances at the CB by $100, and the Treasury spending $100 as above adds $100 to bank balances at the CB, leaving total bank balances (reserves) unchanged.

In short, money spent on public works projects leads to higher interest rates because it does not find its way back to the capital markets.

Not the case, interest rates go to where the CB sets them, one way or another.