Facebook recently announced a new
cryptocurrency, Libra. I had earlier speculated about what a Facebook
cryptocurrency might look like here
for Breakermag.
I think this is great news. MasterCard, Visa, and the various national
banking systems (many of which are oligopolies) need more competition. With a
big player like Facebook entering the market, prices should fall and service
improve, making consumers better off.
The most interesting thing to me about Facebook's move into payments is that
rather than indexing Libras to an existing unit of account, the system will
be based on an entirely new unit of account. When you owe your friend 5 Libras,
or ≋5, that will be
different from owing her $5 or ¥5 or £5. Here is what the
white paper has to say:
"As the value of Libra is effectively linked to a
basket of fiat currencies, from the point of view of any specific currency,
there will be fluctuations in the value of Libra."
So Libra will not just be a new way to pay, but also a new
monetary measurement. Given how Facebook describes it, the Libra unit will be
similar to other unit of account baskets like the IMF's special drawing
right (SDR), the Asian
Monetary Unit (AMU), or the European Currency Unit (ECU), the predecessor
to the euro. Each of these units is a "cocktail" of other currency
units.
Facebook's decision to build its payments network on top of a new unit of account
is very ambitious, perhaps overly so. When fintechs or banks introduce new
media of exchange or payments systems, they invariably piggy back off of the
existing national units of account. For instance, when PayPal debuted in
2001, it didn't set up a new unit called PayPalios. It used the dollar (and
for the other nations in which is is active, it used the local unit of
account). M-Pesa didn't set up a new unit of account called Pesas. It indexed
M-Pesa to the Kenyan shilling.
I couldn't find a good explanation for why Facebook wants to take its own
route. But I suspect it might have something to do with the goal of providing
a universal monetary unit, one that allows Facebook users around the globe to
avoid all the hassles of exchange fluctuations and conversions.
Global monetary harmony an old dream. In the mid 1800s, a bunch of
economists, including William Stanley Jevons, tried to get the world to adopt
the French 5-franc coin as a universal coinage standard. Jevons pointed
out that the world already had international copyright, extradition,
maritime codes of signals, postal conventions—so why not international money
too? He wrote of the "immense good" that would arise when people
could understand all "statements of accounts, prices, and
statistics." It would no longer be necessary to employ a skilled class
of foreign exchange specialists to take on the "perplexing" task of
converting from one money to the other.
But the plan to introduce international money never worked out. (I wrote about
this episode for Bullionstar).
Global money like Libra might seem like a great idea. But ultimately, I
suspect that the decision to introduce a new unit of account will prevent
Libra from ever reaching its full potential. Units of account are a bit like
languages. If you are an English speakers, not only do you communicate to
everyone around you in English, but you also think in English. Likewise with
the dollar or yen or pound or euro. If you live in France, you're used to
describing prices and values to friends and family in euros. You also plan
and conceptualize in terms of them.
It's hard to get people to voluntarily switch to another language or unit of
account once they are locked into it. For instance, in the 1800s L.L.
Zamenhof attempted to get the world to adopt Esperanto as a language in
order to promote communication across borders. To help facilitate adoption,
Zamenhof designed it to be easy to learn. But while around 2 million speak
Esperanto, it never succeeded in becoming a real linguistic standard. The
core problem is this: Why bother learning a new language, even an easy one,
if everyone is using the existing language?
Facebook's Libra project reminds me of Zamenhof's Esperanto project.
Nigerians already talk and compute in naira, Canadians in dollars,
Indonesians in rupiahs, and Russians in rubles. Why would any of us want to
invest time and effort in learning a second language of prices?
Let me put it more concretely. I do most of my families grocery shopping.
Which means I keep track of an evolving array of maybe 30 or 40 food prices
in my head. When something is cheap relative to my memory of it, I will buy
it—sometimes multiple versions of it. And when it is expensive, I avoid it.
But this array is entirely made up of Canadian dollar prices. I don't want to
have to re-memorize that full array of prices in Libra terms, or keep two
arrays of prices in my head, a dollar one and a Libra one. I'm already fluent
in the Canadian dollar ones.
Nor will retailers like Amazon or the local corner store relish the prospect
of having to advertise prices in both the local unit of account and Libra,
plus whatever unit Google and Netflix choose to impose on us.
So Facebook is inflicting an inconvenience on its users by forcing us to
adopt a new unit of account. To make for a better user experience, it should
probably index the Libra payments network to the units of account that we're
all used to.
If not, here is what is likely to happen. We'll all continue to think and
communicate in terms of local currency. But at the last-minute we will have
to make a foreign exchange calculation in order to determine out how much of
our Libra to pay at the check-out counter. To do this calculation, we'll have
to use that moment's Libra-to-local currency exchange rate. This is already
how bitcoin transactions occur, for instance.
But this means that Libra users will lose one of the greatest services
provided by money: money's interval of certainty. This is one of
society's best free lunches around. It emerges from a combination of two
fact. First, most of us don't live in a Libra world in which we must make
some sort of last-minute foreign exchange calculation before paying. Rather,
we live in a world in which the instruments we hold in our wallet are indexed
to the same unit of account in which shops set prices.
Monetary economists call this a wedding of the medium-of-exchange and
unit-of-account functions of money. This fusion is really quite convenient.
It means that we don't have to make constant foreign exchange conversions
every time we pay for something. A bill with a dollar on it is equal to the
dollars emblazoned on sticker prices.
Secondly, shops generally choose to keep sticker prices fixed for long
periods of time. Even with the growth of Amazon and other online retailers,
Alberto Cavallo (who co-founded the Billion Prices Project) finds
that the average price in the U.S. has a duration of around 3.65 months
between 2014-2017. So for example, an IKEA chair that is priced at $15 will
probably have this same price for around 3.65 months. This is down from 6.48
month between 2008-10. But 3.65 months is still a pretty long time.
Prices are getting less sticky thanks to
online shopping: https://t.co/AVikQrJjJ6
A price change endured for 7 months back in 2008. It now lasts just 3.7
months. This effect is particularly strong in sectors where Amazon has high
market share i.e. electronics & household goods. pic.twitter.com/ncAYkagA0b
— JP Koning (@jp_koning) June
25, 2019
Why do businesses provide sticky pricing? In the early 1990s Alan Blinder
asked businesses this very question. He found that the most common reason was
the desire to avoid "antagonizing" customers or "causing them
difficulties." Blinder's findings were similar to Arthur Okun's earlier
explanation for sticky prices whereby business owners maintain an implicit
contract, or invisible handshake, with customers. If buyers view a price
increase as being unfair, they might take revenge on the retailer by looking
for alternatives. (I explore these ideas more here).
Anyways, the combination of these two factors—sticky prices and a wedding of
the unit of account and medium of exchange—provides all of us with an
interval of certainty (or what I once
called money's 'home advantage'). We know exactly how many items we can
buy for the next few weeks or months using the banknotes in our wallet or
funds in our account. And so we can make very precise spending plans. In an
uncertain world, this sort of clarity is quite special.
Given Libra's current design, the interval of certainty disappears. Store
keepers will still keep prices sticky in terms of the local unit of account,
but Libra users do not benefit from this stickiness because Libras aren't
indexed to the same unit as sticker prices are. Anyone who has ≋100 in their account won't
know whether they can afford to buy a given item two weeks from now. But if
they hold $100, they'll still have that certainty, since dollar prices are
still sticky.
If money's interval of certainty is important, it is particularly important
to the poor. The rich have plenty of savings that they can rely on to ride
out price fluctuations. The fewer resources that a family has, the more it
must carefully map out the next few day's of spending. The combination
of sticky prices and a wedding of the unit-of-account and medium-of-exchange
affords a vital planning window to those who are just barely getting by.
This clashes with one of Libra's founding principles: to help the world's 1.7
billion unbanked. Here is David Marcus, Libra's project lead:
3/ Because it's time for the internet to
have a protocol for money, and it's time to try something new for the 1.7
billion people who are still unbanked 30 years after the invention of the
web.
— David Marcus (@davidmarcus) June
18, 2019
Most of the world's unbanked people are poor. But Libra won't be doing the
poor much of a favor by choosing to void the interval of certainty that they rely
on. If Facebook and David Marcus truly wants to help the unbanked, it seems
to me that it would better to index Libras to the various local units of
account.
I suppose there is an argument to be made that Libras could provide poor
people in nations with bad currencies a haven of sorts. Better Libras than
Venezuelan bolivars, right? But the nations with
the world's largest unbanked populations—places like India, Nigeria,
Mexico, Ethiopia, Bangladesh, and Indonesia—all have single digit inflation,
or close to it. Extremely high inflation is really just a problem in a few
outliers, like Zimbabwe and Venezuela.
Besides, providing those who endure high inflation with a better unit of
account isn't the only way to help them. Offering locally-denominated Libras
that offer a compensating high rate of interest would probably be more
useful. Not only would these types of Libra offer inflation protection, but
they would preserve the interval of certainty.
Thankfully, I suspect that Libra is very much a work-in-progress. The current
whitepaper seems to give only a hint of what the project might become. If so,
one of the changes I suspect it will have to make as it goes forward is
choosing to link its system to already-existing units of account. A new unit
of account is just too Utopian.
