Here's why blockchains won't end fractional reserve banking
Commentators often argue that replacing cash with "Fedcoin"—a Fed-administered cryptocurrency like bitcoin—would end fractional reserve banking. JP Koning, who wrote the cannonical piece about the idea of Fedcoin, writes
Fedcoin has the potential to tear down the private banking system...As the public shifted out of private bank deposits and into Fedcoin, banks would have to sell off their loan portfolios, the entire banking industry shrinking into irrelevance. One way to prevent this from happening would be for the Fed to make an explicit announcement that any bank could be free to create its own competing copy of Fedcoin, say WellsFargoCoin.
Koning argues that fractional reserve banking is doomed unless commercial banks are able to launch separate side-chains that build off of the official blockchain maintained by the Fed. The idea that Fedcoin would wipe out fractional reserve banking also underlies much of John Cochrane's post where he quotes Raskin and Yermack (2014) who also take this as a given:
Depositors would no longer have to rely on commercial banks to hold their checking accounts, and the government could get out of the risky deposit-insurance business. Commercial banks that wished to keep making loans would raise long-term capital in the debt and equity markets, ending the mismatch between demand deposits and long-term loans that can cause liquidity problems.
This aspect probably appeals to Cochrane because he wants to abolish fractional reserve banking as a matter of policy, to end the possibility of bank run equilibria.
But Koning, Raskin, Yermack, and Cochrane have overthought this. Taking a step back, it's clear that blockchains pose no threat to fractional reserve banking (for better or worse).
Here's the problem I think Koning had in mind when he proposed side-chains: Suppose you own some Fedcoin and you want to deposit them into a bank. One way you could "deposit" Fedcoin is by sharing your private keys for your Fedcoin addresses with the bank. This way, the bank would be able to use your keys to send some of your Fedcoin to borrowers, while you would also be able to use your private keys to use some of your Fedcoin to buy stuff. But there's an obvious problem: you can't spend Fedcoin from the same addresses the bank is lending out to others, because only one of those two transactions will clear in the blockchain, regardless of whether, in principle, the bank has enough Fedcoin deposits to cover these liabilities.
To avoid this, Koning proposes that instead of sharing your private keys with the bank, you'd sign over the Fedcoins to the bank and receive WellsFargoCoin in exchange, which stores would accept as payment because they trust Wells Fargo's guarantee of redeemability. This way, Wells Fargo is free to lend out your Fedcoin deposit. That's one way to do it.
But there's actually a much simpler option that doesn't require side-chains. When you make a deposit you'd sign over your Fedcoin to the bank, as in Koning's version, but you don't get WellsFargoCoin. Instead, just as you do with cash, you'd withdraw whatever amount of Fedcoin you want as spending money. You are then free to spend the Fedcoin you withdrew, while the bank is free to lend out the rest of your deposits.
In practice you wouldn't even be aware of the need to "withdraw" Fedcoins before spending them. Your digital Fedcoin wallet could just do this automatically. You pay for a cup of coffee by pressing a button on your phone, the digital wallet sends an order to your bank to withdraw the funds, then sends the funds to the coffee shop's Fedcoin address. No side-chains necessary.
In this post, I explained why fractional-reserve banking is simple to implement in a blockchain even without sidechains. In a future post, I'll explain why it's likely that most consumers will prefer to keep their Fedcoin in banks.