4 ms·
The Fed is an association of banks, it's directors directly picked, by bank directors, from bank directors. So you're not exactly denying the GP's point : the s
by iofj 11y ago
The Fed is an association of banks, it's directors directly picked, by bank directors, from bank directors. So you're not exactly denying the GP's point : the source of your data isn't a neutral party, but is controlled by banks. For all intents and purposes, it is a bank.
But even if there was no conflict of interest in the data you presented. (Fractional reserve) banks are built on trust. Not on money, not on reserves, not on loans, insurance or credit default swaps, not on any of that. If the trust goes, reserves won't last long, and can't cover the losses. This is by design, not because something weird happened.
The banks in the US in theory should have 0, 3 or 10% of the money needed to cover depositors, depending on size, bigger banks having more. However the laws state that if a bank goes bankrupt that money won't actually be used to reimburse depositors, but rather lenders (ie. mostly other banks). So from your perspective as a depositor, money you have a legal claim against that's likely to be fulfilled (the original meaning of reserve ratio, decades ago), reserve ratio is 0%. In Europe due to bail-in legislation introduced during and after the 2008 crisis and the Greek debt crisis, the reserve ratio, while nominally 1%, is effectively 0 from the perspective of depositors. If a European bank goes belly-up, depositors will see what the government gives them, if anything.
You might think, but there's FDIC insurance, so why worry ? But in history there has never been a big payout resulting from FDIC insurance, either in the US, or anywhere. The big examples are the Greek and Cypriot debt crisis. People lost access to their money, and their banks were bankrupt. The result : they couldn't transfer money anywhere, nor get money from ATMs. Then they were "bailed in" (the depositor's money was used to pay off other debtors) and they took haircuts on the money in their accounts. You might think the ECB's equivalent of the FDIC insurance would have made up the difference, but this was of course explicitly excluded in the legislation.
In the Cyprus examples, depositors < 100k euro found their funds blocked for years and halfway through them being blocked, the amounts on the accounts were reduced by 6.75%. > 100k euro accounts, same, but 9.9%.
Therefore, banks will default when (enough) people believe they will default. Why ? Exactly because enough people believe they will default. That's the nature of the beast. When a large bank fails, the government will refuse to honor FDIC insurance. Granted the US government hasn't yet proven that it would do this, it has never been forced to make that choice. But it came damn close in 2005 and again in 2008, and we all know what choice they made.
- snowwrestler 11y agoThe government will not refuse to honor FDIC insurance. In the first place, they have no legal basis to do so. From a practical perspective, they have no reason to do so, because all it takes to make a bank solvent is cash, and unlike in Greece, the U.S. government can create as much cash as it needs.
- neffy 11y agoThe problem with these kinds of analysis is that they forget one key element in the entire deposit creation mechanism - loans are also repaid. (The economists just leave that part out). In fact a bank lives in the middle of continuous streams of loan repayments coming back on its loan book. Banks default when their annual losses are more than ~1% of their loan book. Anything else is illiquidity, and that is solvable.