Why Negative Nominal Interest Rates Miss the Point

Willem Buiter, Greg Mankiw, and Scott Sumner have all recently proposed negative nominal interest rates on reserves or currency as a way to stimulate consumer spending and bank lending. It may be nothing more than a coincidence, but the Swedish Riksbank just set the rate it pays banks on reserve balances at -0.25%, effectively taxing banks for holding reserve balances. But I think they are all missing the point, and here’s why.

The classic example of a negative nominal interest rate—long suggested by a number of economists for avoiding deflation—is a tax on currency, which can be summarized in an example Mankiw provides:

“Imagine that the Fed were to announce that, a year from today, it would pick a digit from zero to 9 out of a hat. All currency with a serial number ending in that digit would no longer be legal tender. Suddenly, the expected return to holding currency would become negative 10 percent. That move would free the Fed to cut interest rates below zero. People would be delighted to lend money at negative 3 percent, since losing 3 percent is better than losing 10. Of course, some people might decide that at those rates, they would rather spend the money — for example, by buying a new car. But because expanding aggregate demand is precisely the goal of the interest rate cut, such an incentive isn’t a flaw — it’s a benefit.”

Well, I don’t know about you, but my response would be pretty clear if the Fed made such an announcement: I’d stop holding currency and just use my debit card (most major purchases already aren’t done with currency, anyway). And so would most everyone else. Banks would then sell their currency back to the Fed and the Fed would pay banks reserve balances in exchange (as it usually does). I highly doubt this would lead any bank to cut its lending rates, as Mankiw thinks.

The response and result are in essence unchanged if instead you require individuals to periodically pay to have their currency “stamped” for validation (which Buiter also describes)—I would use a debit card or go to the ATM machine on the day I was going to spend.

The idea of the currency tax is to incentivize people to spend their idle balances, instead of save them. But it doesn’t do this since we can simply switch to holding financial assets that don’t have such a tax. And no matter how far up the chain of liquid financial assets you want to go with the tax (deposits, money market funds savings accounts, etc., as Buiter notes are also options for the tax), I just keep moving my balances to the next most liquid and non-taxed financial asset (or buy everything with a credit card and then pay it off at the end of the month) while financial institutions will also have a substantial incentive to continually design and redesign special liquid accounts (or credit card-type arrangements) that can avoid or otherwise minimize the tax.

In a sense we already tried this in the early 1980s with high interest rates that were essentially a tax on non-interest earning balances, with the end result (for liquidity portfolios, not the economy as a whole) that individuals just started more actively managing their liquidity to minimize the opportunity cost of holding transaction balances while it became less and less clear precisely what was and was not “money.” The currency tax and taxes on transaction account balances similarly just widen the spread between what is earned on non-transaction balances versus transaction balances.

The currency tax also begs the more important point. The problem IS NOT that people have idle balances and aren’t spending them (and it almost never is the problem in a recession). The problem IS that people don’t have enough income (or don’t have the certainty that their current income will be sustained) and/or savings to make them comfortable to spend or to borrow to spend. Indeed, the household sector as a whole is trying to deleverage, not sit on idle balances. And in fact, a tax on currency or transaction balances REDUCES income, obviously, for those holding the types of financial assets being taxed . . . so instead of a monetary stimulus, what you really have with this tax increase is a fiscal TIGHTENING.

What about a tax on bank reserve accounts? (That is, the Fed paying negative interest on reserve balances held by banks.) Sumner argues that, aside from the Fed publicly committing to a nominal GDP growth path (the “expectations about monetary policy are all that matter” view of neoclassical economists. . . which I won’t address here), “the second most effective option would be a modest interest penalty on excess reserves, perhaps 2%.” But this is perhaps even more misguided than a currency tax. And this is what Sweden’s central bank just decided to do (though with a smaller penalty of 0.25%).

In short, the proposal assumes that banks either need the reserve balances (if you tax all reserve balances) for the actual act of creating the loan or (as appears to be the case in Sumner’s proposal to tax only excess balances) they need the reserve balances because of reserve requirements. So, in their view, the tax would give incentive to banks to either get the reserve balances off their balance sheets by “lending” them (so as to avoid a tax on any balances), or to lend in order to transform the balances from excess balances into required balances (so as to avoid an excess reserves tax).

This is the money multiplier view that doesn’t apply to a non-gold standard monetary system like ours (or Sweden’s) since it assumes banks are constrained in their abilities to create loans and deposits by the quantity of reserve balances or excess balances. I’ve explained this at length in my previous posts to this blog, so I won’t go into detail here. In short, though, banks create loans and deposits simultaneously out of thin air. A bank short of balances to meet reserve requirements AUTOMATICALLY receives an overdraft in its reserve account at the Fed’s stated penalty rate, which it then clears with the lowest cost liability it can find. Neither reserve requirements nor the quantity of reserve balances banks hold have anything to do with their abilities to lend.

What will happen, then? Instead of providing an incentive for banks to lend, banks instead will have an incentive to rid their balance sheets of reserve balances. So they will try to trade them in the interbank markets. But in the aggregate, banks only trade the existing quantity of balances among themselves as only a change in the central bank’s balance sheet alters the quantity of balances circulating. So unless the central bank takes action to drain the reserve balances, the undesired excess quantity will just lead banks to bid the overnight rate down to the rate paid on reserve balances, if it’s not already there (as in the US; in Sweden this would be -0.25% unless the balances are drained to hit a higher target rate).

And what is the effect? For banks holding the extra balances not drained by the central bank, the effect is a reduction in their income, as they have to transfer income to the central bank to pay the tax. This then, in the case of the US, would raise the Fed’s profits (all things equal) and thereby raise the amount the Fed turns over to the Treasury (since the Fed turns its profits after paying a fixed percent dividend to member banks). So . . . instead of helping banks lend via monetary easing, we again have a fiscal tightening, as the government’s budget deficit has been reduced, however marginally. Further, by reducing bank profits, you reduce bank capital . . . again probably marginally, but nonetheless, not the most intelligent thing to do in the midst of a banking crisis.

But what about the fact that interest rates people can borrow at may have fallen? Doesn’t this help the economy? First, note that this is NOT the primary reason anyone has advocated negative nominal rates . . . the primary reasons have been to encourage spending idle balances and bank lending, which as I’ve explained here, they do neither. Second, this all depends, since in the private sector, for every borrower paying a lower rate, there is a lender/saver receiving a lower rate, while the private sector also sees reduced income from government securities it is holding. So any improvement in the incentive to borrow (which currently is not very interest sensitive as the private sector deleverages) must be weighed against the offsetting effect of reduced income for lenders and savers.

Returning to remedies for the economy, if the problem is a lack of spending, why not a simple payroll tax holiday, for example, as my fellow bloggers have proposed? This increases spending by increasing incomes and savings, rather than by the attempt at intimidation of a negative interest rate on people already short of income and savings that is the root cause of low spending in the first place. Unfortunately, those recommending penalties on currency, deposits, or reserves don’t fully understand monetary operations given that their basic framework is inapplicable to a modern monetary system such as ours.