Tyler Cowen makes the reasonable, if unfortunate, point that we already have an implicit guarantee of all deposits at the biggest banks. And absent some politically palatable way to remove that guarantee, it may be even worse to only guarantee them instead of guaranteeing all the banks.
The model in my head of a naive regulator looks at the problem you pose, of bad banks outbidding good ones for insured deposits, and says: why not cap the rates that can be paid on insured deposits to prevent that bidding war? "Boring safe places to put your money should not earn high returns" seems like an ok regulatory principle on its face. Now, I can envision plausible sounding lines of counterargument to this, e.g. "the regulators do not know enough to set the cap correctly" or "political pressure on th regulators will too easily make the cap meaningless," but it would be useful to spell out which of these are compelling and why.
It looks to me like the US government was conducting Operation Choke Point 2.0 to punish banks with any crypto clients and ended up causing chained bank runs. Way to go, pushing more risk into banking while unmarked losses on Treasuries were everywhere.
Amusingly, this comes after the attempt to make a full-reserve FDIC-free bank in Wyoming that was killed off by the Fed just a few weeks ago.
Unlike you, I am not an old man, but I like to think I've learned from the past (and my painful memories of '08)... and this is infuriating.
A perfect example of your complaint was summed up by the (normally thoughtful) Noah Smith on his blog: "The only worry regarding a deposit guarantee is that it would create moral hazard. Some people will fret that this sort of move would effectively make all uninsured bank deposits FDIC insured, which will encourage people to put their money in crappy banks in the future."
I have no idea how I can continue to see smart people think that the only side of the banking transaction to worry about is the incentives of the depositors. What the hell do they think the incentives are for the bankers "going to Vegas" with house money as you say?
I was with the FDIC 30 years ago when a number of banks in our state collapsed. The FDIC made sure that no depositor lost any money irrespective of the size of the deposit. Surviving banks were paid by the FDIC to accept all deposits of each failed bank, no matter how large. This policy of avoiding any depositor losses is of long standing.
The notion of “the risk free cost of capital” is itself nonsense. First, the notion that all risks are priced. In particular, not grappling with the difference between risks in the economic system and risks of the economic system. Second, there is no information from the future, which is why we rely on expectations. Hence, the information required for certainty about contingent future events is not possible. Certainty being required for something to be risk free.
A basic investment rule is to diversify your investments. Build a bond ladder. SVB ignored this rule.
The managers of SVB were able to log profits investing in longer dated treasuries. Now due to their greed the public bails them out. Those depositors who had more than $250k should loose their money. Roku who had over $400mil on deposit should loose those funds. Roku shareholders should hold their executives accountable.
Doesn't the saliency of this argument largely hinge on the details of the assets? The S&L crisis got worse because the underlying assets were low quality mortgages. If with "high-quality assets" the Fed means treasury bills then things are less problematic than you suggest. In contrast, if bundled loans secured by seaside real estate in hurricane alley get to be included in the definition, then things are even worse.
Arnold, what would you think of an economist trying to predict inflation by looking at how the prices of many goods and services may evolve in the next 12 months?
Now, what do you think of Jason Furman trying to predict the Fed's Target Rate in late March by looking at how the present and future prices of several assets changed in the past few days? See
Are there good historical examples of governments *not* doing what this one seems about to do (or doing what Kling thinks should be done, however it should be phrased), and things working out the way Kling would expect/predict? I don't know my history on this sort of thing like I should . . . .
The main issue I see is how difficult it is to use regulation to get bank management to do the right thing. For example, it seems to me that SVB should have made bond investments to match the expected withdrawal rates of its depositors AND it should have imposed deposit requirements to minimize the risk of a run on deposits. So for handling the funds of a VC startup, it makes zero sense that the bank would invest those funds in long duration bonds, unless it put the VC depositor under contract it could not access those funds for several years, which would seem to be an unlikely agreement.
As a consumer bank customer, I am well aware of banking polices that protect the bank: Minimum account balances, required direct deposit. This friction makes sense to protect the bank's profitability. What friction did SVB have to prevent a multi-million dollar depositor from withdrawing its money?
The risk of moral hazard in this situation relates to the likelihood of "civilian casualties"- direct victims who did not consent to participation in battle- financial noncombatants, if you will.
There are ALWAYS civilian casualties of the marketplace. Everything is connected (tm).
The only moral line you can draw, then, is the aforementioned consent. You live by the sword, you die by the sword.
These bailouts are not for Peepaw and Mawrmawr's pension funds. We are not airlifting tourists. It's simple cronyism.
Even before yesterday's announcement, bank runs already had very limited power as a check on bad bank management; the preexisting $250k FDIC guarantee and the existence of "bank sweep" accounts that automatically split balances across many banks (yes, this does actually increase your FDIC coverage) had gone a long way toward neutering them.
It is interesting that SVB was vulnerable to one, but we need to refine the incentive set for other banks that were already nearly invulnerable to bank runs. In this sense, yesterday's announcement doesn't change that much.
Very apt post, and I hope you'll consider upping your posting frequency during this vexed time—you've got a valuable skillset here.
Tyler Cowen makes the reasonable, if unfortunate, point that we already have an implicit guarantee of all deposits at the biggest banks. And absent some politically palatable way to remove that guarantee, it may be even worse to only guarantee them instead of guaranteeing all the banks.
The model in my head of a naive regulator looks at the problem you pose, of bad banks outbidding good ones for insured deposits, and says: why not cap the rates that can be paid on insured deposits to prevent that bidding war? "Boring safe places to put your money should not earn high returns" seems like an ok regulatory principle on its face. Now, I can envision plausible sounding lines of counterargument to this, e.g. "the regulators do not know enough to set the cap correctly" or "political pressure on th regulators will too easily make the cap meaningless," but it would be useful to spell out which of these are compelling and why.
It looks to me like the US government was conducting Operation Choke Point 2.0 to punish banks with any crypto clients and ended up causing chained bank runs. Way to go, pushing more risk into banking while unmarked losses on Treasuries were everywhere.
Amusingly, this comes after the attempt to make a full-reserve FDIC-free bank in Wyoming that was killed off by the Fed just a few weeks ago.
Unlike you, I am not an old man, but I like to think I've learned from the past (and my painful memories of '08)... and this is infuriating.
A perfect example of your complaint was summed up by the (normally thoughtful) Noah Smith on his blog: "The only worry regarding a deposit guarantee is that it would create moral hazard. Some people will fret that this sort of move would effectively make all uninsured bank deposits FDIC insured, which will encourage people to put their money in crappy banks in the future."
I have no idea how I can continue to see smart people think that the only side of the banking transaction to worry about is the incentives of the depositors. What the hell do they think the incentives are for the bankers "going to Vegas" with house money as you say?
I was with the FDIC 30 years ago when a number of banks in our state collapsed. The FDIC made sure that no depositor lost any money irrespective of the size of the deposit. Surviving banks were paid by the FDIC to accept all deposits of each failed bank, no matter how large. This policy of avoiding any depositor losses is of long standing.
The notion of “the risk free cost of capital” is itself nonsense. First, the notion that all risks are priced. In particular, not grappling with the difference between risks in the economic system and risks of the economic system. Second, there is no information from the future, which is why we rely on expectations. Hence, the information required for certainty about contingent future events is not possible. Certainty being required for something to be risk free.
It’s a sandwich alright. But not “baloney”. But then I acknowledge you have certain rhetorical standards which I applaud😏
A basic investment rule is to diversify your investments. Build a bond ladder. SVB ignored this rule.
The managers of SVB were able to log profits investing in longer dated treasuries. Now due to their greed the public bails them out. Those depositors who had more than $250k should loose their money. Roku who had over $400mil on deposit should loose those funds. Roku shareholders should hold their executives accountable.
Doesn't the saliency of this argument largely hinge on the details of the assets? The S&L crisis got worse because the underlying assets were low quality mortgages. If with "high-quality assets" the Fed means treasury bills then things are less problematic than you suggest. In contrast, if bundled loans secured by seaside real estate in hurricane alley get to be included in the definition, then things are even worse.
Arnold, what would you think of an economist trying to predict inflation by looking at how the prices of many goods and services may evolve in the next 12 months?
Now, what do you think of Jason Furman trying to predict the Fed's Target Rate in late March by looking at how the present and future prices of several assets changed in the past few days? See
https://twitter.com/jasonfurman/status/1635268094502531074
Are there good historical examples of governments *not* doing what this one seems about to do (or doing what Kling thinks should be done, however it should be phrased), and things working out the way Kling would expect/predict? I don't know my history on this sort of thing like I should . . . .
The main issue I see is how difficult it is to use regulation to get bank management to do the right thing. For example, it seems to me that SVB should have made bond investments to match the expected withdrawal rates of its depositors AND it should have imposed deposit requirements to minimize the risk of a run on deposits. So for handling the funds of a VC startup, it makes zero sense that the bank would invest those funds in long duration bonds, unless it put the VC depositor under contract it could not access those funds for several years, which would seem to be an unlikely agreement.
As a consumer bank customer, I am well aware of banking polices that protect the bank: Minimum account balances, required direct deposit. This friction makes sense to protect the bank's profitability. What friction did SVB have to prevent a multi-million dollar depositor from withdrawing its money?
Think Big:
1. Deposit way under par bonds with the Feds
2. They give you par value in cash
3. Buy many more under par bonds
4. Repeat until all the money is yours
thank you for writing this, it has been driving me crazy
The risk of moral hazard in this situation relates to the likelihood of "civilian casualties"- direct victims who did not consent to participation in battle- financial noncombatants, if you will.
There are ALWAYS civilian casualties of the marketplace. Everything is connected (tm).
The only moral line you can draw, then, is the aforementioned consent. You live by the sword, you die by the sword.
These bailouts are not for Peepaw and Mawrmawr's pension funds. We are not airlifting tourists. It's simple cronyism.
Even before yesterday's announcement, bank runs already had very limited power as a check on bad bank management; the preexisting $250k FDIC guarantee and the existence of "bank sweep" accounts that automatically split balances across many banks (yes, this does actually increase your FDIC coverage) had gone a long way toward neutering them.
It is interesting that SVB was vulnerable to one, but we need to refine the incentive set for other banks that were already nearly invulnerable to bank runs. In this sense, yesterday's announcement doesn't change that much.