In an op-ed for the New York Times, economist Paul Krugman wrote that “cryptocurrency enthusiasts are effectively celebrating the use of cutting-edge technology to set the monetary system back 300 years.”
In Krugman's view, the sweep of monetary history has been one of increased efficiency by a transition away from commodity money (e.g. gold and silver) towards fiat money (e.g. the US dollar).
We began with gold and silver, which were heavy, required a lot of security, and consumed a lot of resources. We moved to bank notes, backed by fractional reserves, that were easier to transport but still required banks to hold gold and silver backing up those bank notes.
Central banking caused regular banks to stop holding gold and silver themselves and hold their reserves as deposits in central banks. This eliminated the use of commodity money in day-to-day transactions, but the paper money was still backed by gold and silver in a central bank vault. The shift to full fiat money under Nixon eliminated the “inefficiency” of commodity-backed money entirely with the value of money being dependent upon trust in the national government (or, to be more precise, that U.S. citizens have to pay their taxes in U.S. dollars and if they don’t then men with guns will show up at their homes.)
In my view, Krugman fails to understand the tradeoffs made in the name of efficiency. In our complex and interdependent world, much of what seems efficient is not efficient, it is hiding risk somewhere out of sight (at least for the moment).
In the mid-2000s, it looked like the financial engineers on Wall St. had figured out a “more efficient” way to create more AAA securities and let more people own their own homes. We saw how that worked out.
Allow me to tell a similar story of “increased efficiency” and how it impacted Tom, the turkey.
Tom the Turkey (Problem)
Tom was born a wild turkey in the forests of North Georgia. Life is hard for Tom. Tom must go out every day and find his own food and avoid the hunters who would like to have him for dinner.
It is a struggle to survive and very, very inefficient.
One day, Tom wanders into Bailey the banker’s backyard and finds a huge pile of food.
“No more pecking or avoiding hunters!” thinks Tom.”It’s just eating!”
Tom stays and Bailey brings out more food for Tom everyday. One day, Bailey goes to Tom and explains that it is becoming too hard for her to care for the turkey because she has to do all the work herself. To make the process more efficient, she takes Tom to her landlord’s sprawling estate, Christopher Banks the central banker.
Christopher already has a few turkeys that other tenants have brought over so he can provide even more food on a more regular basis.
True to Bailey’s promise, Christopher’s facility is much more efficient: The food always gets brought out on time and in even larger quantities.
After a while, Christopher comes out and tells the turkey that he has found an even more efficient home for him at a beautiful resort called Watergate run by the very efficient Richard. At Watergate, the food is brought around continuously on one of those Japanese sushi restaurant conveyor belts while Tom can sit in a turkey-edition La-Z-boy and eat to his heart’s content. So excited by this even more efficient method, Tom does not even ask questions.
True to Christopher’s promise, Watergate is wonderfully efficient! Richard seems to know Tom’s every need, as if he is listening to everything Tom says! Tom grows fat and happy - everything at Watergate seems just lovely.
Richard, Tom’s new owner and the the town butcher, looks out his window at his turkey. Thanksgiving is coming up soon and the eating will be good.
The day before Thanksgiving, the turkey could look back at its life and see it as one of ever-increasing efficiency: from having to hunt and peck for food, to Bailey (the banker) bringing the food out, to Christopher Banks (the central banker) bringing out even more food, and finally Richard (Nixon’s) conveyor belt setup.
Looking at past data, the turkey can confidently conclude that things are only going to get better as “more efficiencies” are gained. Why would Tom want to go back to pecking for his own food?
Caption: Tom the turkey confidently forecasts the increasing efficiency of future food intake. Image credit: Antifragile, Nassim Nicholas Taleb
From our vantage point, we can see one good reason he’d want to do this: Watergate is not more efficient in the long run, it just seems more efficient in the short run because the risk is hiding from Tom.
Tom’s situation is fragile. He is completely dependent on Richard’s goodwill, but Richard does not have the best incentive structure for promoting Tom’s long term well-being.
Everything keeps getting better until suddenly, it doesn’t.
It wasn’t obvious to Tom that optimizing for “maximum efficiency” would end with him being served au jus with a cornbread stuffing and cranberry sauce.
Are backup generators at nuclear power plants inefficient?
What Tom didn’t realize is that when resources are scarce (as they always are), there is a tradeoff between being more efficient and being more thorough (or one could also say more secure or robust).
Let’s say you are editing articles and you have only ten hours (the scarce resource) to work on the task. If it takes you two hours to edit an article, you can make the decision to either edit one article five times and being extremely thorough, or edit five articles one time and be more efficient.
You can choose where on the efficiency thoroughness trade-off (ETTO) spectrum you fall, but you can’t get off the spectrum. Given a scarce set of resources, you cannot be both highly thorough and highly efficient.
If we’re deciding where on the ETTO spectrum to fall, we need to consider the impact of something going wrong. Is an error costly or cheap?
The web developer building a puppy photo sharing app can “move fast and break things” because the impact of making a mistake is cheap. No one freaks out if their puppy photo won’t load.
The nuclear power plant engineer does not “move fast and break things” in the name of efficiency. The impact of making a mistake is catastrophic.
The backup to the backup to the backup to the backup generator may be idle 99.9999% of the time. Having it around is very “inefficient,” but I, for one, want my nuclear power plants to be incredibly inefficient.
Krugman is effectively saying that “the main generator on this nuclear power plant is working just fine, why waste resources on all those backups?”
Money should optimize for thoroughness and security, not efficiency
Money is more like a nuclear power plant than a dog photo sharing app. Major shocks to monetary systems have catastrophic consequences.
The German currency was relatively stable at about 90 marks per dollar during the first half of 1921. By November 1923, the US dollar was worth 4,210,500,000,000 German marks. The same thing that made the money “efficient,” that it wasn’t backed by some form of commodity money also meant central bankers could inflate the supply.
You do not want to have highly efficient money if you have to sacrifice its thoroughness or soundness. (In the context of money, a word like soundness capture the idea of thoroughness as it’s used in ETTO better).
You want money to be maximally sound, and only more efficient if it doesn’t affect that soundness. Would you rather be able to send your Weimar Republic German Marks quickly and easily as they hyperinflated or would you rather store your wealth in some other currency which was more secure?
Critics would argue that the US Dollar has been both efficient and sound over the past fifty years and that’s certainly true. The question is whether or not that will continue to be the case. It is possible the same will be true in fifty years, but there is some evidence that it won’t.
Image credit: Skin in the Game, Nassim Nicholas Taleb
It was the efficiency of central banks printing presses which drove Bitcoin’s creation and early supporters. Global macro investors looking at central bank balance sheets found it harder to foresee a scenario where the U.S. dollar and other major fiat currencies could continue to be both efficient and sound. It’s hard to see a scenario in which the central banks don’t have to make the money less sound in order to pay off their liabilities.
In the short term, the central bank action (extended low interest rates and money printing like QE) taken over the last two decades has looked very efficient: The U.S. stock market is in one of the longest bull runs in history and recently minted the first ever trillion dollar company.
But it seems unlikely to me that this efficiency comes without a cost. Arguably, the huge increase in asset prices (stocks, bonds and real estate) is inflationary, it’s just not being labeled as such because we are used to seeing inflation in consumer prices. Is Apple really twenty two times more valuable than it was in December of 2008 or is there just a bunch of money sloshing around that has to go somewhere?
Recall: Much of what seems efficient is not efficient, it is hiding risk somewhere out of sight (at least for the moment).
Tom the turkey was blind to the hidden risk of his domestication. The nuclear power plant can create efficiencies by not having backups, but it also hides risk that does not show up, usually until it is too late.
What problem does cryptocurrency solve? It makes the opposite tradeoff.
Bitcoin maximizes for soundness at the cost of efficiency. Compared to PayPal or Chase Bank or the Federal Reserve, Bitcoin seems highly inefficient. It is slow, highly redundant and has a terrible user experience. The entire ledger is only a couple hundred gigabytes and yet maintaining it costs over $20 million in electricity per day!
This seems dramatically “inefficient” and “wasteful” until it isn’t, just like running the quadruple backup generator at the nuclear power plant.
Bitcoin introduces a necessary and healthy counterweight to Krugman’s “efficient” fiat money.
Currently the data supports two hypotheses: Krugman is right and the current fiat system is simply superior, both more efficient and more sound, than a system based on commodity money. By ignoring the tradeoff between the efficiency and soundness of money, the monetary system has accumulated hidden risk with potentially catastrophic consequences.
Cryptocurrency creates a free market for money that allows individuals to vote, with their money, for the second hypothesis.
The central bankers are starting to take note. A 2018 IMF report noted:
To fend off potential competitive pressure from crypto assets, central banks must continue to carry out effective monetary policies. They can also learn from the properties of crypto assets and the underlying technology and make fiat currencies more attractive for the digital age.
For those wary of the wonderful efficiency at Richard’s Watergate Resort, Bitcoin is the hole in the backyard fence.
Maybe that won’t be necessary, but I bet Tom would sure like to have the option.
Thanks to Alex Hardy and Julia Hess for reading early drafts.