In 2014 the government of Ecuador, under then-President Rafael Correa, announced with great fanfare that the Ecuadorian Central Bank (BCE) would soon begin issuing an electronic money (dinero electrónico, or DE). Users would keep account balances on the central bank's own balance sheet and transfer them using a mobile phone app. Enabling legislation was passed in September, qualified users could open accounts beginning in December, and the accounts became spendable in February 2015. A headline on CNBC's website declared: "Ecuador becomes the first country to roll out its own digital cash."
The subsequent fate of the electronic money project has received less attention in the American press. Less than three years after opening, the system is now shutting down. In December 2017 Ecuador's National Assembly, at the urging of President Lenin Moreno, Correa's hand-picked successor who took office earlier in the year, passed legislation to decommission the central bank electronic money system. The legislation simultaneously opens the market to mobile payment alternatives from the country's private commercial banks and savings institutions. As described below, the state system had failed to attract a significant number of users or volume of payments. Account holders now have until the end of March 2018 to withdraw their funds. Complete deactivation is scheduled for mid-April.
The substitution of open competition for state monopoly in mobile money is an important victory for the people of Ecuador. The entire episode is important internationally for the lesson it teaches us about the limits to a central bank's ability to launch a new form of money when the public prefers established forms. The lesson provides an instructive contrast to "the case for central bank electronic money" recently made by Aleksander Berentsen and Fabian Schär in the pages of the Federal Reserve Bank of St. Louis Review.
The Birth of the Project
There is an important backstory to the episode: Ecuador had suffered a hyperinflation of its domestic currency, the sucre, in 1999, prompting residents to dollarize their own payments and finances. In January 2000 the government, bowing to the popular verdict, announced that it would officially dollarize, fixing a parity of the sucre to the US dollar and retiring all sucres from circulation by September. (Concerning Ecuador's dollarized system see my earlier post here.)
The electronic money project was born in 2014 legislation that gave the state a monopoly in electronic money. Only the central bank could issue electronic dollars, and only the state-owned mobile phone company CNT could provide mobile payment services. The law barred the private mobile phone companies and private financial institutions from providing competing systems. The legislature also banned cryptocurrencies.
Because President Correa (in office 2007-2017) had often complained about the discipline that dollarization imposed on his government, observers wondered whether the electronic money system was intended merely as a way for the government to gain some monopoly profits, or was a first step toward de-dollarization. To calm fears that the electronic money would become a forced currency to be followed by de-dollarization, the law declared that use of the electronic money would be voluntary, and that even public employees and state contractors would not be obliged to accept it in payments from the state. (Everyone knew that the Assembly could later revise that provision of the law, of course.)
The government was quite optimistic that the system would rapidly prove popular. The leading newspaper El Comercio reported on Christmas Day of 2014: "Fausto Villavicencio, responsible for the new payment mechanism in Ecuador, said that the authorities expect that some 500,000 people will use e-money in 2015." The actual number of accounts opened in 2015 turned out to be less than 5000. The economist Diego Grijalva of the Universidad de San Francisco de Quito, citing the Ecuadorian central bank's balance sheet, noted in early 2016 that "the Ecuadorian Electronic Money System is already implemented, but it has an uncertain future. In particular, financial institutions are not obliged to use it and the use thereof (less than US $ 800,000 for the end of January 2016) corresponded to less than 0.003% of the monetary liabilities of the Ecuadorian financial system."
Its Failure to Achieve Popularity
One had to be skeptical of the stated rationale for the central bank electronic money project, to benefit the unbanked. Invited to speak in Ecuador about the dollarization regime in November 2014 (working paper in English, later published in Spanish) at events organized by the USFQ and the think-tank IEEP, I added some critical comments on the new project:
There is no reason to believe that a national government can run a mobile payment system more efficiently than private firms … If the government sincerely wishes to help the poor and unbanked, it should let private providers enter the competition, which will drive down the fees that the poor and unbanked will have to pay.
Private bankers in Ecuador made similar arguments during the life of the project. In its December 2017 legislation, the government conceded the case. According to one news account, "the Government hopes that with the transfer to the private financial system the means of payment can reach more unbanked population."
I attributed a fiscal rationale to the project:
[W]hy does the government want to issue mobile payment credits as a monopolist? It seems likely that the project is meant as a fiscal measure. One million dollars held by the public in the form of government-issued credits is a million-dollar interest-free loan from the public to the government.
From the fact that the government is now closing its service, I infer that the central bank failed to make a profit even as a statutory monopolist. Float was smaller, and expenses higher, than had been hoped (see below). The new administration had no fiscal reason to keep it open.
Although I did not foresee the system's failure to achieve sustainability, I did add one final dig at the system's low trustworthiness:
Personally, I would find dollar-denominated account credits that are claims on [the leading private mobile phone companies] Movistar or Claro more credible than claims on the government of Ecuador. After all, unlike the government, neither company defaulted on its bonds in the past 12 years.
Trust, it turned out, was the crucial issue.
Unlike what is usually envisioned under the rubric "central bank electronic money," the BCE was not creating nominally default-risk-free accounts denominated in its own domestic fiat money. It was issuing claims to US dollars that it might become unable or unwilling to repay. The government under Correa had in fact defaulted on sovereign dollar-denominated bonds in 2008. Although the sucre hyperinflation of 1999 had brought with it a banking crisis, since dollarization the commercial banks had by all indications become stable and prudently run.
Consequently it was reasonable for an informed citizen in 2014-17 to think that dollars on deposit at a private commercial bank in Ecuador were less risky than dollars on deposit at the central bank. The private banks had better incentives to behave prudently than the BCE had. A private bank could be taken to court if it failed to pay, but not so the government central bank with its sovereign immunity. The enabling legislation specified no limit on the volume of electronic dollars the BCE could create, and no prudential requirement that the central bank hold adequate assets to redeem them.
The Ecuadorian public recognized a risk of default or devaluation with the central bank's electronic money accounts, and stayed away from them, defying to the optimistic projections of government officials promoting the system. In June 2016 President Correa recognized that the project had critics, but he dismissed them as merely members of the opposition party and certain private bankers annoyed that the business was not going to them. In fact mistrust in the system was much more widespread.
At least one print commentator at the time pointed to the BCE's lack of trustworthiness. El Comercio's economic columnist Gabriela Calderón de Burgos in a June 2016 column clearly predicted that because of public mistrust the DE system would not succeed. She noted that, unlike the private bankers with their own wealth at stake, the BCE could behave irresponsibly, and would be pressured to do so by the Treasury with its chronic financing problems. Thus the electronic claims on the BCE are "a currency that does not inspire confidence," and as a result "the DE will not work because it will not enjoy widespread acceptance. It would only achieve this if the government declares it to be a curso forzoso [forced tender]." But the government knows that such a move "would lead to chaos."
Later that month, in a column on the "antics" of the central bank, she observed: "The government has intensified its campaign for people to deposit their dollars in the BCE and use 'electronic money.' I suspect that the campaign will have little success because of the justified distrust that the government and the BCE have earned in terms of their ability to take care of others' funds."
In May 2017 Calderón de Burgos returned to the theme, in a column entitled "Only the Dollar is Trusted." The project for the Ecuadorian government to issue its own digital currency, denominated in US dollars, she wrote, "faces an insurmountable inconvenience: people will not voluntarily accept the new currency. That's why they've been trying to convince us to use electronic money for three years and still few use it." The US dollar itself, because it is something that Ecuadorian politicians cannot devalue, "generates much more confidence than any alternative that can occur to our politicians, even and particularly in times of financial crisis."
A news article in December 2017 reported the answers that ordinary people gave when asked them directly why they weren't using the BCE's electronic money. Their answers confirm that many found the system not creditworthy. For example: "Mistrust is among the reasons, says Frank Guijarro, owner of a tire network." And: "I do not trust opening an account with the Central Bank, so I pay in cash and sometimes with a debit card when I cannot get out," says Katherine Alcivar, 26." The president of the association of cooperative savings banks gave a similar answer in an interview: "The greatest confidence we can give is that your resources are in your financial institution and not in the BCE." The BCE system was haunted by the "ghost" of the previous government's default. In addition, the BCE did too little to promote acceptance by shopkeepers and other businesses: "Not enough strength was given to the reception channels."
As a result of these shortcomings, the system peaked at only $11.3 million in account balances, less than 5 hundredths of 1% of the country's narrow money stock M1 ($24.5 billion). According to the deputy general manager of the central bank electronic money system, before the announcement of the coming shutdown ironically raised the average level of activity due to withdrawals, the system averaged only about 1,100 transactions per day. The total value transacted over the entire life of the system was only about $65 million. Only 7,067 businesses ever conducted transactions with the electronic money. While a total of 402,515 accounts were eventually opened, the BCE found in retrospect that only 41,966 were ever used to acquire goods and services or to make payments. Another 76,105 were used only to upload and download money. The remaining 286,207 accounts (71%) that were opened were never used. (I do not know why the three reported component figures do not sum exactly to the reported total.)
Lessons from the Failure of the Project
We can make a back-of-the-envelope calculation of the Ecuadorian government's profit from its monopoly electronic money project. Between 2014 and the present, the Ecuadorian government has been paying roughly 8% interest on the bonds it sells in international markets. Replacing $11.3 million of 8% bonds with zero-interest liabilities of the central bank provides an annual debt-service savings of less than $1 million, specifically $904,000. From the BCE's 2014 income statement (the most recent that seems to be available), its "administrative expenses" (presumably payroll) were roughly $38 million. Thus the project would have turned a loss if it enlarged the BCE payroll by as little as 2.4%, even leaving aside non-salary expenditures on promoting and operating the project.
An accounting report on the DE project issued by GPR, an Ecuadorian government accounting office, puts the government's expenditures on the project at $7,967,553.78. Comparing that figure to the estimated debt service savings of only $904,000, the fiscal loss is clear. My thanks to Luis Espinosa Goded and Santiago Gangotena of the USFQ Department of Economics for pointing me toward the GPR report and helping me read it.
It is instructive to contrast the outcome in Ecuador with the optimistic picture of central bank electronic money drawn by Berentsen and Schär, who write:
We believe that there is a strong case for central bank money in electronic form, and it would be easy to implement. Central banks would only need to allow households and firms to open accounts with them, which would allow them to make payments with central bank electronic money instead of commercial bank deposits. As explained earlier, the main benefit is that central bank electronic money satisfies the population's need for virtual money without facing counterparty risk.
The BCE deposits, by contrast to the scenario they have in mind, were not free of counterparty risk. More generally, in a sound banking system a commercial bank's counterparty risk for depositors can be negligible, very close to zero, so that the central bank's zero default risk need not be a big draw. In episodes where the central bank and the commercial banks simultaneously circulate banknote liabilities (e.g. today's Scotland or Northern Ireland), no public concern about a risk difference is evident.
The Ecuadorian case also shows that implementation of a central bank electronic money system isn't so easy. It requires more than merely setting up a website (the US federal government has sometimes proven not even competent at that) and letting households and firms open deposits. A convenient point-of-sale deposit-transfer mechanism, requiring both hardware and software, must be provided to many thousands of merchants. Consumer service and marketing are part of the business of providing retail payments. There is no reason to think that central banks are or would be good at a commercial business operation. In short, it is far from clear that asking bureaucrats to build a "public option" electronic money system would have benefits in excess of its cost.
 All quoted statements from Ecuadorean sources are my own translations, assisted by Google Translate.