Opinion

In Search of a Stable Electronic Currency

Robert J. Shiller — March 2014

In a New York Times op-ed, Professor Robert J. Shiller writes that Bitcoin is a speculative bubble with a doubtful future, but its legacy should be that we move toward a system of stable economic units of measurement backed by sophisticated forms of electronic payment.

Originally published in the New York Times on March 3, 2014.

Bitcoin, an experiment with a radically new kind of electronic money, has exhibited many of the characteristics of a speculative bubble. That was clear long before the collapse of the Bitcoin exchange Mt. Gox last week.

Bitcoin’s future is very much in doubt. Yet whatever becomes of it, something good can arise from its innovations—even if the results are very different from its current form or its numerous competitors. What I have in mind isn’t another wave of price speculation. Instead, I believe that electronic forms of money could give us better pricing, contracting, and risk management.

The Bitcoin phenomenon seems to fit the basic definition of a speculative bubble—that is, a special kind of fad, a mania for holding an asset in expectation of its appreciation. Further, a bubble is publicized and amplified by news of price increases, often justified by some kind of inspiring “new era” story that attracts more attention as the price rises. In this case, the narrative was that a computer whiz invented a new kind of money in the form of electronic currency units, as part of a decentralized computer-driven system for a world economy that extends beyond the reach of any single government.

The central problem with Bitcoin in its present form, though, is that it doesn’t really solve any sensible economic problem. Nor should it substitute for banks and the governmental institutions that regulate them. They are reasonably effective institutions, despite their flaws, and should not just be scrapped and replaced by a novel electronic system.

Unfortunately, the Bitcoin success story has been tied intrinsically with instability, with excitement and envy for those who have become rich through investing in it—rich for a while at least, because the value of the electronic currency has fluctuated wildly. The instability of Bitcoin’s value in dollars is a measure of failure, not success. It means that any commerce using Bitcoin or its competitors would be buffeted by enormous inflation and deflation.

But if we go back to the electronic-money drawing board, we may conclude that Bitcoin has been focused on the wrong classical functions of money, as a medium of exchange and a store of value. Bitcoin offers a way of “mining” electronic coins that can replace our dollar bills and bank accounts. Yet there is no fundamental need for this. Money, as we’ve known it for decades, works quite well in these respects. It would be much better to focus on another classical function: money as a unit of account—that is, as a basic standard of economic measurement. Scientists spend a lot of time thinking about ways to improve systems of measurement. Businesspeople should, too.

This has already begun to happen. History shows that this unit-of-account function of money has been separated from the other two, and to good purpose. For example, since 1967 in Chile, an inflation-indexed unit of account called the unidad de fomento (U.F.), meaning unit of development, has been widely used. Financial exchanges are made in pesos, according to a U.F.-peso rate posted on the website valoruf.cl. One multiplies the U.F. price by the exchange rate to arrive at the amount owed today in pesos. In this way, it is natural and easy to set inflation-indexed prices, and Chile is much more effectively inflation-indexed than other countries are.

Consider rents. Increases may seem unfair to tenants, yet they may be needed to offset inflation. In Chile, a landlord can easily set the monthly rent for the tenant in U.F.s and then never have to change it, reducing the potential for errors, delays, and misunderstandings. The name “U.F.” reframes people’s thinking so that keeping real economic values stable is natural and easy.

With electronic software in the background, we can improve on the Chilean idea and make it more useful. First, we should use more evocative words and avoid jargon. I’d suggest calling the unit of account a “basket.” A basket is a simple description for something rather complicated. In its basic form, it is the equivalent of one day’s common consumption items, like the local marketbasket of typical purchases that already underlies the Consumer Price Index. The value from one base year would be chosen so that, with inflation, the price of a basket in local currency terms would rise.

A second improvement could be achieved by a credit card company or an entity like PayPal or Square Cash: Set up the electronic machinery to allow people to easily send conventional money, denominated in baskets, to anyone in the world. You would just check off a box indicating whether your payment was in dollars or pesos or euros—or baskets. If you check “baskets,” the computer would calculate the amount of local currency the recipient would need to buy that basket at that time, and it would transmit it, too.

A third improvement would be to move beyond just one new unit of account to a whole system of them, so that we could have baskets for different purposes. There should be senior baskets representing items consumed by older people in one day in a given country, as well as subsistence baskets representing the consumption of the poor. There should also be a day-wage unit of account representing a day’s work by an average unskilled wage earner.

And there could be a “trills” unit—a concept that Mark Kamstra of York University and I have been advocating—that represents one trillionth of a country’s most recently estimated annual G.D.P. There should also be a unit that grows or retreats with per-capita daily consumption. This could be used for pension and Social Security payments as a form of intergenerational risk-sharing: The idea is that payments to older people would rise and fall with overall consumption.

With many kinds of baskets, it will be easier to set prices and make contracts that are sensible for the long term.

These are ideas about simplification of complex economic calculations. But they are borrowed from behavioral economics, which has taught us that our actions are profoundly affected by framing—by the words we use and the approaches we take. And they are also ideas from cognitive linguistics, which implies that improved language, with new meanings for words like basket or day-wage, will change our thinking.

I have been arguing for almost 20 years that we should not only copy the Chileans but also carry their ideas further into such improvements. All the recent interest in new electronic money could mean that now is the time to reconsider the foundation of our standards of value.

Bitcoin has been a bubble. But the legacy of the Bitcoin experience should be that we move toward a system of stable economic units of measurement—a system empowered by sophisticated mechanisms of electronic payment.