Do you have any idea what the balance sheets of your bank(s) look like? I don’t. I don’t know whom they lend to.
I don’t have to. And the reason is deposit insurance. Even when my checking account hits 5 digits, the amount in it is well under the $250,000 limit that is covered by deposit insurance. So no matter how risky the loans my bank makes, I don’t have to worry.
In Canada in the 1870s, by contrast, there was no deposit insurance. Indeed, deposit insurance was not introduced to Canada until 1967. In “Stability in the absence of deposit insurance: The Canadian banking system, 1890-1966, Journal of Money, Credit, and Banking, November 1995, Vol. 2, No. 4, University of Toronto economists Jack Carr and Frank Mathewson and Victoria University (in Wellington, New Zealand) economist Neil Quigley write:
The Canadian Bank Act, 1871, was explicitly designed to prevent the public from thinking that the government was responsible for either the commercial administration of the banks or the ability of individual institutions to pay their creditors. It required banks to make monthly returns to the Department of Finance, but envisaged the self-interests of the shareholders and mandatory double liability as sufficient protection for the creditors.
The related footnote states: “In this period, the banks undertook extensive advertising of their balance sheets in the press. In addition, newspapers reproduced the monthly financial statements that the banks provided to the government.
Incentives matter. Depositors cared about how safe their deposits were and banks and newspapers responded to this concern by reporting balance sheets.
Note: You might wonder what double liability is. Elsewhere in the article, Carr, Mathewson, and Quigley explain it: “Double liability of the shareholders meant that creditors of the bank were secured by both the value of the equity and retained earnings in the bank and a claim against the personal wealth of shareholders equivalent to the subscribed capital.”
READER COMMENTS
Jerry Brown
Apr 27 2021 at 11:05pm
So you don’t have to worry. That is usually a good thing- not worrying.
jj
Apr 28 2021 at 12:26am
But the total amount of worry isn’t gone, it’s just distributed more widely: now the entire nation has to worry (a little bit, each) rather than the bank’s customers (a lot, each). And if this dilution of incentives leads to additional risk-taking by the bank, then the total amount of worry which is due has gone up.
Jerry Brown
Apr 28 2021 at 1:18am
Nah. We pay a few people to worry about it. They are called bank regulators. When they screw up and the banks screw up we just create a lot of money and bail them out. It isn’t fair, but at least you don’t have to worry about your checking account disappearing.
Sriram
Apr 28 2021 at 1:44am
You can still pay people to worry about it in a free banking system, they would just be private regulators who would certify the trustworthiness of a bank rather than public regulators who are a single point of failure.
Jon Murphy
Apr 28 2021 at 7:33am
Nah. You just need to worry about your house disappearing because you got a subprime loan. Or your job disappearing because your firm cannot get the capital it needs from the bank because the bank lost on risky loans. Etc.
There’s more to banking than checking deposits
Andrew_FL
Apr 28 2021 at 9:54am
You’re all missing the point. Nobody worrying about the risks to their deposits means there is no discipline on bank risk taking. Deposit insurance is a huge subsidy for making risky loans. Unsurprisingly banks that are subsidized to make risky loans make riskier loans, and thus are more prone to failure.
Thomas Lee Hutcheson
Apr 28 2021 at 5:31am
That would certainly get more people to pay attention to bank balance sheets, but is that a good use of people’s time? [Wouldn’t that cut into time spent obsessing over how fair the “fair trade” of their kumquats is? :)]
If central banks allow/cause large declines in NGDP, there is going to be a fall in real income whatever the prudential regulation of banks.
Alan Goldhammer
Apr 28 2021 at 9:01am
You would do well to read Anat Admati’s fine book, “The Banker’s New Clothes and What to Do About It.” She is a professor of finance at Stanford Business School. Russ Roberts interviewed her on EconTalk a couple of years ago.
Her thoughts on the banking industry are well reasoned.
Art K
Apr 28 2021 at 10:03am
I wonder why there was an introduction of depositors insurance in 1967. I don’t believe there was any bank run around the time and I thought the economy was going full steam in the 50s and 60s. Would this result from lobbying from the banks?
David Henderson
Apr 28 2021 at 10:49am
Commenters: Please check Andrew_FL’s comment above. I took for granted that people who read this site would immediately get the moral hazard issue. But apparently I shouldn’t have.
Daniel Kling
Apr 28 2021 at 11:33am
My impression is that the commenters above understand the moral hazard issue (at least some of them) but are not much moved by it. You’ve identified a cost of deposit insurance (less prudence). Some folks are bringing up a benefit of deposit insurance (less worry, perhaps mitigated by the introduction of worse tail risks). The post reads like you think the lack of worry is a bad thing, but I think it makes sense to separate the worry from the actual prudence.
E.g., it’s nice that you don’t worry about food safety day-to-day. In another setting you might not be able to take food safety for granted and you’d have to expend more effort to make sure you don’t get food poisoning. That’d be a drag! Some people might lament that Americans are too complacent about food safety, but those people are silly. Regardless of the extent to which the safety comes from food-sellers caring about the reputation hit from poisoning customers or from regulation, the fact that you have minimal worry and still get safe food is a huge net positive.
Would you contend that deposit insurance is net negative? Are the marginal risky loans not worth the peace and confidence and additional investment that we get? Do you think it’d be good to go back to double liability? It seems like the way the policies have evolved has led to more and easier capital formation, and those incentives matter, too, but I’d be interested in your take if there’s something I’m missing. Was there some big systematic malinvestment? (If you say tech in 1999 or housing in 2006 we have to fire up the Scott Sumner “no bubbles” bat signal!)
David Henderson
Apr 28 2021 at 12:51pm
You wrote:
That’s possible.
You wrote:
Well put.
You asked:
Yes.
You asked:
I think not. Think about the S&L crisis
You asked:
Maybe.
You write:
But we actually have a relatively small banking sector relative to the size of our economy than, say, Canada, has. Most capital formation is not through banks but through equity investments, I believe. I think you’re missing the S&L crisis.
Daniel Kling
Apr 28 2021 at 2:48pm
You are totally right. I was not including the S&L crisis in my analysis because it was too long ago for me to have experienced (I was born in 1985) but also too recent to get picked up much when I read about ancient history (like the 70’s or earlier).
Some quick reading on the S&L leaves me ambivalent about what to think about deposit insurance. I guess we’ve been on track to have a major banking crisis every 30-50 years or so, right? That doesn’t seem great, but I don’t have a good counterfactual in mind. It seems like we tend treat each crisis as sui generis (and sure, they’re each special – participation trophies for each one), but that it does makes sense to focus on the big picture and long term incentives that lead to different levels of risk tolerance that fatten those bad tail outcomes (as you did here).
Vivian Darkbloom
Apr 28 2021 at 11:03am
I have mixed views on this. While several commenters have mentioned that deposit insurance is a form of “subsidy” for risk-taking banks no one has mentioned the fact that it is mostly banks themselves that are cross-insuring each other. Sure, Treasury is a back-up to the FDIC, but the FDIC is funded by mandatory bank contributions. If Treasury needs to step in, this inevitably means that FDIC premiums will be raised in the future to make up for that contribution made from outside the banking system.
Someone asked why Canada introduced deposit insurance in 1967. I don’t know that answer to that but I do know that the FDIC (and bank deposit insurance) was first introduced in 1933 in the midst of the depression. Roosevelt was initially against it, but it was because of public opinion (not bank lobbying) that this initial reluctance was overcome.
The idea of insurance as a mechanism to spread risk and protect the public is something that, in my experience, was largely the result of economic thought influencing law. For example, I remember clearly the discussions in tort law class in the 1970’s about the influence of the “Chicago School” (e.g. Posner and Epstein) on the expansion of strict liability as a means of spreading risk among the general public. The possibility that the risks of moral hazard would exceed the benefits of spreading of risk is not something I recall from those lessons.
The following chart seems to show that since the introduction of the FDIC there have been fewer bank failures than prior to that year.
https://www.calculatedriskblog.com/2018/01/bank-failures-by-year.html
If moral hazard were a serious factor introduced by deposit insurance, would the trend have been more favorable? Would fewer bank customers have ended up uncompensated for their losses? Would the public have had the same confidence in the banking system? I have my doubts.
I also surmise that one reason mandatory deposit insurance through the CDIC did not come until 1967 was the fact that bank risk was more evenly distributed geographically in Canada (as compared with the US) due to the lack of restrictions in Canada against banks operating across province lines.
Vivian Darkbloom
Apr 28 2021 at 3:21pm
I guess one reason I am not convinced by the moral hazard argument is that I don’t see how deposit insurance would change banking behavior that much. Pretty much by definition, deposit insurance only kicks in once the bank has failed. With or without deposit insurance bankers are still subject to the same ultimate penalties—if my behavior is too risky the bank may fail (I may lose my job, my options might be worthless, etc). The ultimate penalty for bankers is the same with or without deposit insurance. Also, if their behavior is so agregious as to warrant ignoring the corporate solution and/or imposing criminal penalties, deposit insurance doesn’t change any of that. What am I missing?
David Seltzer
Apr 28 2021 at 6:00pm
I suspect moral hazard is greater than we think not only because of FDIC insurance but government’s history of TBTF bailouts of banks in the past. Of course it is really taxpayers eating losses. If banks issued more equity, it would be reflected on financial statements for all publicly traded banks. A self interested depositor could choose from among more solvent banks than less solvent banks.
Vivian Darkbloom
Apr 29 2021 at 9:32am
“Of course it is really taxpayers eating losses”.
If we are talking about the FDIC deposit insurance, I’m not sure that’s true. As I noted earlier, the FDIC is funded by assessments to insured financial institutions. On the whole, it is the banks collectively who bear the burden of those premiums and not the federal government. The term “federally insured” is largely a misnomer. Rather, the program is federally mandated and administered. Of course, like all other costs to corporations, this is largely passed on to shareholders and customers. Currently, the FDIC fund stands at about $120 billion, all funded by assessments to banks. We can argue over accounting methods, but under straight-forward economic accounting even the TARP program returned a very modest profit.
As far as moral hazard is concerned, I would admit the *theoretical* possibility that the existence of FDIC insurance makes *depositors* less likely to do extensive research on the financial condition of the banks they entrust funds to and perhaps then indirectly that banks don’t have to compete as much on safety. However, insurance does nothing to prevent the curious and concerned from doing so (and those typically more saavy that have more than $250K on deposit). And, I submit that one is greatly overstating here the likelihood of an average depositor doing *any* due diligence. As others have noted, part of the FDIC program is doing the kind of oversight that the average depositor is not capable of or likely to carry out. I can also see the potential of lack of insurance causing damage flowing the other way–absent insurance, unwarranted rumors of bank financial instability causing bank runs and leaving those too late to bail with actual uninsured losses.
One further point: Under a system of mandatory contributions to an insurance fund, it is theoretically possible that an individual bank undertake more risky behavior with the knowledge that losses will be picked up *by other banks*. But, as I noted above, this does little to alter behavior to potentially bad actors–the potential damage to *them* is the same. Also, not mentioned anywhere previously, is the fact that FDIC insurance assessments are not flat rate—the assessment system is risk adjusted so that this does mitigate moral hazard concern.
I searched in vain for the current accounting treament of FDIC assessment contributions. I ran across a publication from the early 1990’s which argued that contributions should be treated as bank equity; however, I suspect today that FASB treats them as expenses. When you think about it, the FDIC reserve is clearly akin to bank equity. Requiring more bank equity (in the form of risk reserves) to protect depositors against overly-risky behavior strikes me as not a bad thing even if the accounting doesn’t treat it that way.
David Seltzer
Apr 29 2021 at 11:07am
Vladimir, I was talking about the TBTF bailouts as socializing losses across the taxpayers as well as share holders. My point, there is an incentive on the part of the banker to take more risk because losses will be socialized. An example of the Peltzman effect. Those losses should be born by the shareholder alone. After the dot com crash, it was the primarily shareholders who lost not the taxpayer.
Vivian Darkbloom
Apr 29 2021 at 11:12am
David,
I would be grateful if you were to provide some support and numbers for the idea that the TARP program cost was borne by taxpayers. Further, how is this relevant to FDIC deposit insurance?
David Seltzer
Apr 29 2021 at 11:24am
https://www.nationalreview.com/corner/real-cost-tarp-veronique-de-rugy/
David Seltzer
Apr 29 2021 at 11:27am
In the interest of scientific inquiry, it seems you would take the counterfactual and demonstrate that taxpayers DID NOT cover those losses or TARP.
Vivian Darkbloom
Apr 29 2021 at 11:53am
David,
The “counterfactual” is the same GAO report that you indirectly linked to. From the Conclusions (as of the date of the report):
”
CPP repayments and other income surpassed the program’s original investment disbursements, and as institutions continue to exit CPP, this surplus continues to grow. Furthermore, Treasury’s latest estimate (November 30, 2011) projects CPP’s lifetime income to be $13.5 billion. However, a growing number of the remaining institutions have missed scheduled dividend or interest payments or appeared on FDIC’s problem bank list. As a result, there is increased concern regarding the speed at which institutions will be able to repay remaining funds and how much of
these funds Treasury will ultimately recover. In particular, our analysis showed that institutions remaining in CPP were generally less profitable, held riskier assets and less regulatory capital, and had lower reserves for covering losses compared with institutions that repaid their CPP investment and those that never participated in the program. Despite the noticeably different financial profiles for remaining and former CPP institutions, Treasury’s quarterly analysis of CPP institutions does not distinguish between these two groups. As we have indicated in past reports on TARP, transparency remains a critical element in the government’s unprecedented assistance to the financial sector. Such transparency helps clarify to policymakers and the public the costs of TARP assistance and the government’s intervention in various markets. Enhancing the quarterly CPP analysis by distinguishing between remaining and former CPP participants will help Treasury provide Congress and the public with a more transparent and comprehensive understanding of the status of CPP and the institutions that participate in it.”
So, using straight accounting methods, TARP didn’t lose money. One can argue that a different, risk-adjusted cost should have been applied to those loans *ex-ante*; however, I don’t see that the evidence you linked to supports an *ex-post* conclusion that those investments cost taxpayers money. In fact, the GAO report you cited was as of 2012. Pro-Publica (hardly a banker’s association) now reports that the government made $110 billion on TARP as of February 2021: https://projects.propublica.org/bailout/
Am I being “un-scientific” here?
David Seltzer
Apr 29 2021 at 1:12pm
No you are not being unscientific. You presented your argument well. My take is that those returns to TARP have costs that are ignored by risk adjustment, as you point out, and time value. Please see Measuring the Cost of Bailouts, November, 2018, Deborah Lucas, MIT. I still think it is better that banks issue more equity so as to absorb large losses and indemnify depositors than having the tax payer front run bailouts. A moral hazard issue just waiting to happen. I suspect cost-benefit would yield something of a solution. Good conversation.
Vivian Darkbloom
Apr 29 2021 at 2:14pm
Thanks, David. It’s always nice to have a good civil discussion. In that spirit, I have to respond to this:
“I still think it is better that banks issue more equity so as to absorb large losses and indemnify depositors than having the tax payer front run bailouts.”
In principle, I don’t disagree. Although I’m for factual accuracy, I’m not for “taxpayer funded bailouts”. However, please recall that the main discussion was about mandatory FDIC deposit insurance. I don’t see how this constitutes “taxpayer front run bailouts”. To the contrary, as I’ve taken pains to point out, the program is funded by bank contributions, not taxpayer funds. And, as I’ve also argued, those reserves *are* a form of equity and those bank supplied funds are used to to “indemnify depositors”.
David Seltzer
Apr 29 2021 at 2:42pm
Vivian, I don’t disagree as to FDIC. My point, as was DH’s, is moral hazard. As for taxpayers front running bailouts, the tax payer is “asked” to take the risk with taxes paid upfront, to rescue those failing banks. The risk is that some banks or insurance companies will not repay those taxpayers. The second risk, as you pointed out, is not getting a risk adjusted return on those tax dollars. We can assess risk with CAPM. I suspect the betas for those failed banks are far greater than one. If Beta is determined to be two, for instance, the taxpayer should get twice the market return which would compensate him/her for assumed risk. Any thing less is an opportunity cost to the taxpayer.
Scott Sumner
Apr 29 2021 at 2:44pm
The fees assessed on FDIC insured banks are a tax, and the cost gets passed along to the public.
David Seltzer
Apr 29 2021 at 2:53pm
Thank you Scott.
Jerry Brown
Apr 29 2021 at 10:01pm
So what? The cost may possibly get passed to the public that has a bank account (and that is far from clear) but so does the benefit. And the benefit is clear. If the benefit is greater than the maybe cost wouldn’t that be a good thing overall?
Thomas Sewell
May 2 2021 at 1:25pm
The problem is that the costs are redistributed by the FDIC tax and pay process from banks which are run riskily/poorly to banks which are run safely/well. That provides incentives for all banks to be run more risky/less well.
Consider a forced auto insurance scheme where you and your neighbor’s sixteen-year-old son with a Camaro were required to jointly share via taxation the expected costs of your combined future auto accidents. The total cost/benefits can roughly match overall for the scheme, and it still be a bad idea from a moral hazard/risk management/fairness perspective.
Vivian Darkbloom
May 2 2021 at 2:15pm
I doesn’t strike me that you have a clear understanding of the nature of insurance. The very nature of insurance is that everyone in the insurance pool shares risks with the rest of the pool. In fact, unlike real taxes, the premiums paid into the insurance pool of the FDIC can’t be diverted to other uses, such as buying missles or funding schools. The funds are used exclusively to insure the customers of those banks against loss of their deposits and clearly to the indirect benefit of those contributing financial institutions as well. You are undoubtedly required to have liability insurance for you and the car you drive. That doesn’t mean that the premiums you pay are “taxes”, much less that they can be used to fund schools or build roads. Whether you like it or not, you *are* sharing risks with that 16 year-old. He may pay higher premiums than you, but as I noted above, the premiums banks pay are risk-adjusted as well.
Jerry Brown
Apr 28 2021 at 6:19pm
My first comments were a bit sarcastic and I apologize for that. I like to think I understand some of the moral hazards here. It bothers me intensely that the Fed basically bailed out many banks in 2008. That being said- I doubt that a lack of deposit insurance would have prevented that from occurring. And I value not having to worry too much about the safety of my meager savings in the bank down the street.
Comments are closed.