Negative interest rates are in the news. In response to imploding economies and very low inflation rates, some commentators have argued that central banks should cut nominal interest rates below zero to try and stimulate economic activity. Others including Federal Reserve Chairman Jerome Powell are less enthusiastic, noting that that there is little evidence that negative interest rates have a substantial stimulatory effect. Either way, the whole proposition raises several interesting economic issues that concern the way that monetary policy affects the distribution of income and consumption.
The normal case against negative interest rates on money is that they are impractical. If banks offer a negative interest rate, depositors have an incentive to withdraw their deposits as cash and deposit the cash in safety deposit boxes or hide it in a freezer. Under a fiat monetary system, there is no limit to how much cash people with demand deposits can withdraw, or how much cash governments can produce.
It seems plausible that interest rates below -1 percent will spark cash withdrawals. Unless the central bank replaces the general public as the holder of the banking systems’ short-term liabilities, these withdrawals will be contractionary rather than expansionary. Since the central bank can be expected to lend cash to banks in response to widespread withdrawals, a negative interest rate policy results in the government paying banks for the privilege of temporarily lending them cash, so that their former customers can withdraw cash and place it in a safety deposit box. It is nice work if you can get it. Of course, this policy could be stimulatory if banks expand their balance sheet by lending to customers who purchase newly produced goods and services to take as much advantage of the government transfer as possible.
However, there is a different case against negative money interest rates that stems from the economics of commodity interest rates.
Commodity interest rates
What is a commodity interest rate? It is the explicit or implicit interest rate denominated in terms of a commodity (e.g. oil, wheat or gold) when one party explicitly or implicitly borrows or lends a commodity. An example of an explicit commodity interest rate is the interest rate on nuclear grade uranium that is borrowed and lent by nuclear power stations. A power company with a temporary surplus of uranium will lend 1000 grams to another power company, and have a different amount, say 1020 grams returned at a later date. A common example of an implicit interest is a wheat hedge, where a flour mill may simultaneously buy wheat on the spot market and sell it on the forward market, effectively borrowing wheat until the forward market is due. In this case the wheat interest rate is equal to the spot price divided by the forward price and multiplied by the money interest rate: this formula calculates the kilograms of wheat that are delivered in the future as a function of the kilograms of wheat purchased immediately. Commodity interest rates were extensively analysed by Sraffa and Keynes in the 1930s and have been subject to periodic research ever since.
When are commodity interest rates negative? Normally when the spot price is less than the forward or future price. Usually this occurs when the spot price falls relative to the future price, and the lower spot prices induces higher consumption of the commodity. Temporarily low spot prices occur if there is a temporary supply glut in the market, or a temporary dearth in demand. A glut that happens every year occurs in the strawberry market: in summer, large quantities of strawberries are produced and sold at low prices, whereas in winter few strawberries are available and the price is high. The implicit negative strawberry interest rate that occurs every summer suggests most people would be willing to swap a kilo of strawberries in summer for 500 grams in winter. More to the point, in commodity markets negative interest rates not only signal higher future prices, but they are also usually associated with temporary price declines.
The same combination – negative real interest rates and temporarily low current prices – is also associated with temporary demand shortfalls. When demand temporarily collapses in a commodity market, the price of the commodity temporarily falls and the implicit commodity interest rates becomes negative. This price decline stimulates additional demand to ensure the quantity demanded is equal to the quantity for sale; consumers get a discount, and although producers suffer lower prices they at least sell their produce.
Commodity interest rates and money interest rates
This is not the mechanism that is being proposed by those advocating negative money interest rates. Rather, negative interest rates are seen as an alternative to temporary price declines. Most firms spend a lot of time and marketing effort to establish profit-maximising prices, and are reluctant to cut them even in the face of temporary demand declines. From their perspective, it may be better to reduce and output and lay off people rather than cut prices, particularly if they think it will be difficult to raise prices back to normal levels when demand recovers. When lots of firms act in the same manner, reduced demand leads to reduced output rather than lower prices. Negative interest rates on money are proposed as a means of stimulating demand without firms cutting their prices.
What are the economic consequences of negative nominal interest rates when prices do not fall? We don’t know for sure because empirical evidence is limited. There are not many examples of substantially negative money interest rates because people can withdraw bank deposits as cash. But the idea raises conceptually troubling issues.
First, negative interest rates punish lenders if they save rather than spend. This is intended to stimulate spending in the same way that temporarily low prices reward spending rather than saving. Unfortunately, there is no good evidence that low price carrots and negative interest sticks have the same effect. A temporary price cut creates substitution and income effects that raise spending: a person buys more not just because the price is temporarily cheap (the substitution effect) but because once they have bought the item they still have left over money which can be spent on other things (the income effect).
In contrast, negative interest rates change the relative price of current and future consumption without increasing overall purchasing power. A negative interest rate also has distributional consequences that are different from a temporary price cut. When a negative demand shock results in lower prices on commodity markets, buyers gain and producers lose, although the producers do get to sell their goods. If central banks make interest rates negative because firms are reluctant to cut prices, the losses are imposed on lenders rather than producers or borrowers. Perhaps lenders deserve these losses for the “sin” of saving – although in New Zealand, where the prices of so many goods and services are high by international standards, you cannot always blame them for their reluctance to spend.
From a different perspective, negative interest rates without a spot price declines appear bizarre. An interest rate is merely a short hand summary of a schedule of future payments made from one party to another. Consider vendor finance on a new car. When interest rates are positive, a person can take possession of a car and pay cash immediately, or they can take possession of a car and make a higher payment to the vendor at a later date. Firms offer vendor finance to induce immediate sales to those who expect to receive cash in the future but don’t have it in the present.
In contrast, a negative interest rate means that a person who purchases a new car today can choose between paying a large amount of cash immediately or a smaller amount if they pay later. Unless cash hoarding is very expensive, people have an incentive to withdraw their deposits in a bank as cash and hoard them in a safe deposit box until the delayed payment falls due.
But why would a vendor offer such a deal? Why would they prefer to accept a smaller amount of cash in the future that the full amount immediately? Maybe they fear the government plans to confiscate some or all of their cash or bank deposits if they receive payment immediately. Maybe a negative interest allows them to offer a customer a discount without formally reducing the price.
You can imagine the sales pitch: ‘Have we a deal for you. You can pay the full price now, or if you pay later we will give you a five percent discount and a free set of steak knives. Is the manager crazy or what? You better buy now because the deal can’t last long.’
Or maybe negative interest rates are simply a coordination device, allowing all firms to offer implicit discounts without changing their headline prices. Whatever the reason, it is still curious that lenders rather than borrowers are being asked to take a redistributive hit because firms are reluctant to cut prices to induce higher demand.
Negative interest rates can be expected to have other effects as well. Loans or delayed payment arrangements can finance asset exchange as well as new activity. If delayed payment makes the purchase of firms or second-hand houses cheaper, there is the risk that negative interest rates will raise the current price of assets. The rationale for doing this when asset prices are near record highs is far from obvious. Negative interest rates on loans used to finance asset exchange also have distributional consequences, favouring people who have previously chosen to purchase assets rather than those who lent money to others. The reason for redistributing wealth to people who have equity rather than debt claims on assets is not particularly clear.
Is there an alternative to negative money interest rates? If the government is concerned about the decline in output and employment that occurs when firms do not cut prices in the face of temporary adverse demand shocks, it could always try to induce a temporary reduction in prices by another means. One obvious way is to temporarily reduce its GST tax, to reward those who spend now rather than to try and punish those who save by accumulating bank deposits. Clearly this results in temporarily lower revenue, which will need to be offset by higher taxes in later more buoyant times. This policy will also result in a different distribution of income and consumption than the effect of negative interest rates.
I have no idea about the relative merits of the redistributive effects of negative interest rates and a temporary cut in the consumption taxes. I don’t even know if those advocating negative interest rates have attempted to understand the relative distributive implications of these different policies. I hope they have, because it is not clear that negative money interest rates are an effective substitute for temporary price cuts. This suggests careful thought is in order before attempts are made to crash through the zero bound barrier. As Jerome Powell has argued, there is little evidence that negative interest rates without price falls are a panacea for the problems caused by low demand.