Mar 25, 2023·edited Mar 25, 2023Liked by Arnold Kling
Actually I do think Cochrane wants to eliminate "actual banking". And I agree. Why do fixed value liabilities need to backed by floating value assets? This isn't some law of nature. Banks would be able to issue deposits backed by reserves. These would be used to make payments. To finance their asset portfolio (mortgages, loans, etc.) they would issue floating value securities (primarily equity and risky debt).
I disagree somewhat with the term "public would tolerate the government sitting back while booms and busts take place".
The public doesn't have a clue. In the vaguest sense sudden downturns cause them to punish incumbents vaguely in the next election, and it would take a pretty sustained and sharp downturn for that to pay important political dividends.
I would say that what really happens is that regulators and lawmakers that would normally oppose such actions "tolerate" them when they are panicked and scared. In general I think if they held their ground they would find that the public forced their hand far less then they forced their own hand.
As I've commented before, I agree with Cochrane's take, at least to the extent that we need to alter the current situation in which the Biden regime stumbled into an implicit 100% guarantee of all bank deposits, in contravention to the plain language of the FDIC's authorization, in an effort to make their cronies and contributors whole in the face of SVB and other recent bank failures. We need to firewall that 100% guarantee away from purely investment accounts. We have huge sectors of money management that operate with government regulation to ensure transparency and some protection from fraud but no guarantee of the money deposited in those vehicles. The claim that 'banking' is a mix of payment processing with speculative investing developed as a reaction to disintermediation during the high inflation of 1970s and 1980s, and was a significant contributor to the S&L collapse. Continuing down the path we've been put on in the last month or so is going to endanger the entire banking system if we don't rein in the guarantees before an explosion of 'zombie banks' drives out solvent and prudent banks, which was an excellent description of the potential situation.
If there was a solid narrow banking sector, I am not sure that booms and busts in actual banks (let's call them fat banks) would be such a problem. On the one hand, not being bailed out at the drop of a hat would work somewhat to limit the excesses of risk in the industry, just like in any other. On the other hand, we might not have to care too much if the industry went boom and bust a lot. Most people would have less exposure to its failures, and when it does fail the system we really care about would still be fine.
I suppose an analogy would be fat banking as a very temperamental and unreliable sports car: very fast and exciting when it works, but it tends to break and spend a lot of time not doing anything at all. If that is your only car you have serious problems, as every time it breaks you are missing work, not picking the kids up, all the things you need. It is a big problem and you spend a lot of time worrying about keeping it running. However, if your main car is a narrow bank, boring but eminently reliable, the pressure is off the fat banking mobile to always work. If the fat bank car is up on the lift getting repaired, you can still do everything you need to do, so it doesn't matter as much. It might be less efficient to maintain two cars, but if you want the exciting car you need the reliable one as well to keep up with your obligations.
It seems downright facile to suggest deregulation would be getting the government out of banking if the Federal Reserve still exists and the Federal government is still driving the market as the only producer of safe assets.
From a literal perspective, they are the parts of the system that are most irregular with standard practice and most responsible for the increasing risk in the system.
“Actual banking is when the bank issues liabilities that are riskless and short term (like deposits) while holding assets that are risky and long term (like home mortgages or commercial loans). “
Why is form of banking necessary? Just because people demand to indulge in the fantasy of risk-free returns does not mean that their demands must be placated.
The absolute magnitude and variance of the inflation rate is what drove the real interest rate dramatically negative and made the financial system unstable.
Has the government/central bank (money printer) anywhere ever stopped inflation without making the real interest rate positive? I don't know of any examples.
" Done, zero risk." John Cochrane, like most economists & regulators & auditors, part of the problem.
There is ALWAYS risk, even if it minimal. We need better words and metrics for risk, and comparing risks.
How can you write about SVB and risk and NOT talk about it's duration risk? (because nobody else is talking about it, either -- tho I'm reading and admiring only a few like John here.)
Do you even know what it's duration risk positions were at end of years 2020, 2021, 2022? (If not, why not?)
--
Arnold, you've said duration risk is well understood. Please tell me what the SVB numbers have been over the last couple years. I'm convinced lack of good language to talk about risk has exacerbated all risk related regulatory issues (also with COVID).
"To an economist, no action is really a mistake, it's just an optimal answer to a different question". Very true and the question that the Fed and other regulators seemed to want answered only concerned bank solvency in a recession or some other similar event; no modeling of a stagflation environment, a rising interest rate and how that would affect banks and creditor view of banks for those assets marked to market.
Boy, what an oversight! I would also add that I suspect that bankers taking note of a rising interest rate environment as part of credit analysis probably looked at their borrowers loans and modeled how a rising interest rate would affect creditworthiness; at least once upon a time they did. How bank managers in bank treasury departments did not with regard to bond securities is a bit baffling.
Cochrane seems to just want to see the system be more stable; he knows credit drives economic activity and putting aside fiscal dysfunction ,real growth.
Asset/lLiability mismatch remains the issue as most are now aware of. But banks sell off loans to CLO, CDO, and CMO buyers who do match fund. So, why cannot this be more readily achieved by bank themselves and better safeguard depositors? The gov, as in the past, will always step in to see depositor suffer no loss, so alternatives to what we have at present need to be examined.
In some sense, narrow banking already exists. Narrow banks at present appear in the form of money market mutual funds that hold 100% of their assets in U.S. Treasury obligations or other instruments collateralized by the same (repos). Such MMMFs are highly successful and hold 100's of billions of dollars in assets, if not trillions. As for "actual banks," IMHO, it seems clear that deposit insurance is necessary to maintain liquidity in the monetary system, but that comes with moral hazards as we have seen time and again. That requires some regulation, but the simpler the better. I conclude from this line of thought (but could be proven wrong) that the simplest regulation would be to require actual banks to hold more capital. Since such capital might generate returns that are lower risk and lower reward, it would attract only certain types of investors. But clearly there is a market for such equity, witness the large market for preferred shares and high dividend yielding and/or low-beta common equity.
"we show that the combination of rapid credit and asset price growth over the prior three years, whether in the nonfinancial business or the household sector, ..."
I think I understand what is meant in the generic sense by credit growth and asset price growth but I have no idea how Greenwood et al are measuring either and for sure I don't know how they combine them into one measure. I don't see any discussion of that in their paper. Isn't that something rather important to first verifying their statement is at least factual? Or do we have to take them at their word?
Actually I do think Cochrane wants to eliminate "actual banking". And I agree. Why do fixed value liabilities need to backed by floating value assets? This isn't some law of nature. Banks would be able to issue deposits backed by reserves. These would be used to make payments. To finance their asset portfolio (mortgages, loans, etc.) they would issue floating value securities (primarily equity and risky debt).
I disagree somewhat with the term "public would tolerate the government sitting back while booms and busts take place".
The public doesn't have a clue. In the vaguest sense sudden downturns cause them to punish incumbents vaguely in the next election, and it would take a pretty sustained and sharp downturn for that to pay important political dividends.
I would say that what really happens is that regulators and lawmakers that would normally oppose such actions "tolerate" them when they are panicked and scared. In general I think if they held their ground they would find that the public forced their hand far less then they forced their own hand.
As I've commented before, I agree with Cochrane's take, at least to the extent that we need to alter the current situation in which the Biden regime stumbled into an implicit 100% guarantee of all bank deposits, in contravention to the plain language of the FDIC's authorization, in an effort to make their cronies and contributors whole in the face of SVB and other recent bank failures. We need to firewall that 100% guarantee away from purely investment accounts. We have huge sectors of money management that operate with government regulation to ensure transparency and some protection from fraud but no guarantee of the money deposited in those vehicles. The claim that 'banking' is a mix of payment processing with speculative investing developed as a reaction to disintermediation during the high inflation of 1970s and 1980s, and was a significant contributor to the S&L collapse. Continuing down the path we've been put on in the last month or so is going to endanger the entire banking system if we don't rein in the guarantees before an explosion of 'zombie banks' drives out solvent and prudent banks, which was an excellent description of the potential situation.
If there was a solid narrow banking sector, I am not sure that booms and busts in actual banks (let's call them fat banks) would be such a problem. On the one hand, not being bailed out at the drop of a hat would work somewhat to limit the excesses of risk in the industry, just like in any other. On the other hand, we might not have to care too much if the industry went boom and bust a lot. Most people would have less exposure to its failures, and when it does fail the system we really care about would still be fine.
I suppose an analogy would be fat banking as a very temperamental and unreliable sports car: very fast and exciting when it works, but it tends to break and spend a lot of time not doing anything at all. If that is your only car you have serious problems, as every time it breaks you are missing work, not picking the kids up, all the things you need. It is a big problem and you spend a lot of time worrying about keeping it running. However, if your main car is a narrow bank, boring but eminently reliable, the pressure is off the fat banking mobile to always work. If the fat bank car is up on the lift getting repaired, you can still do everything you need to do, so it doesn't matter as much. It might be less efficient to maintain two cars, but if you want the exciting car you need the reliable one as well to keep up with your obligations.
It seems downright facile to suggest deregulation would be getting the government out of banking if the Federal Reserve still exists and the Federal government is still driving the market as the only producer of safe assets.
From a literal perspective, they are the parts of the system that are most irregular with standard practice and most responsible for the increasing risk in the system.
A trenchant essay.
Re: "the financial policies that survive are the ones that increase the bureaucrats’ control over the financial sector."
Let's call this *Kling's Law* -- A robust corollary to *Wagner's Law of Increasing State Activity*:
https://en.wikipedia.org/wiki/Wagner%27s_law
PS: I see what you did there, turning the tables on John Cochrane's phrase.
"To an economist, no action is really a mistake, it's just an optimal answer to a different question".
It's a cute statement but doesn't seem all that valuable.
1 - Some questions have little or no importance. There is no value in answering them.
2 - Some actions aren't optimal to any question.
I suppose it could be argued these two statements are different way of saying the same thing.
“Actual banking is when the bank issues liabilities that are riskless and short term (like deposits) while holding assets that are risky and long term (like home mortgages or commercial loans). “
Why is form of banking necessary? Just because people demand to indulge in the fantasy of risk-free returns does not mean that their demands must be placated.
The absolute magnitude and variance of the inflation rate is what drove the real interest rate dramatically negative and made the financial system unstable.
Has the government/central bank (money printer) anywhere ever stopped inflation without making the real interest rate positive? I don't know of any examples.
My unpublished comment there:
" Done, zero risk." John Cochrane, like most economists & regulators & auditors, part of the problem.
There is ALWAYS risk, even if it minimal. We need better words and metrics for risk, and comparing risks.
How can you write about SVB and risk and NOT talk about it's duration risk? (because nobody else is talking about it, either -- tho I'm reading and admiring only a few like John here.)
Do you even know what it's duration risk positions were at end of years 2020, 2021, 2022? (If not, why not?)
--
Arnold, you've said duration risk is well understood. Please tell me what the SVB numbers have been over the last couple years. I'm convinced lack of good language to talk about risk has exacerbated all risk related regulatory issues (also with COVID).
"To an economist, no action is really a mistake, it's just an optimal answer to a different question". Very true and the question that the Fed and other regulators seemed to want answered only concerned bank solvency in a recession or some other similar event; no modeling of a stagflation environment, a rising interest rate and how that would affect banks and creditor view of banks for those assets marked to market.
Boy, what an oversight! I would also add that I suspect that bankers taking note of a rising interest rate environment as part of credit analysis probably looked at their borrowers loans and modeled how a rising interest rate would affect creditworthiness; at least once upon a time they did. How bank managers in bank treasury departments did not with regard to bond securities is a bit baffling.
Cochrane seems to just want to see the system be more stable; he knows credit drives economic activity and putting aside fiscal dysfunction ,real growth.
Asset/lLiability mismatch remains the issue as most are now aware of. But banks sell off loans to CLO, CDO, and CMO buyers who do match fund. So, why cannot this be more readily achieved by bank themselves and better safeguard depositors? The gov, as in the past, will always step in to see depositor suffer no loss, so alternatives to what we have at present need to be examined.
In some sense, narrow banking already exists. Narrow banks at present appear in the form of money market mutual funds that hold 100% of their assets in U.S. Treasury obligations or other instruments collateralized by the same (repos). Such MMMFs are highly successful and hold 100's of billions of dollars in assets, if not trillions. As for "actual banks," IMHO, it seems clear that deposit insurance is necessary to maintain liquidity in the monetary system, but that comes with moral hazards as we have seen time and again. That requires some regulation, but the simpler the better. I conclude from this line of thought (but could be proven wrong) that the simplest regulation would be to require actual banks to hold more capital. Since such capital might generate returns that are lower risk and lower reward, it would attract only certain types of investors. But clearly there is a market for such equity, witness the large market for preferred shares and high dividend yielding and/or low-beta common equity.
In an unregulated "banking" system the Fed would still be trying to "stabilize the economy" (keep inflation on target).
"we show that the combination of rapid credit and asset price growth over the prior three years, whether in the nonfinancial business or the household sector, ..."
I think I understand what is meant in the generic sense by credit growth and asset price growth but I have no idea how Greenwood et al are measuring either and for sure I don't know how they combine them into one measure. I don't see any discussion of that in their paper. Isn't that something rather important to first verifying their statement is at least factual? Or do we have to take them at their word?