It's important to make sure that at no point in the process does any uninsured entity have custody of the money. That's the trouble with some "fintech" firms trying to do this. Remember, none of this matters unless the financial system is in severe trouble. Then it really matters.
From the article: “We represent 1,728 insured depositors at the now-failed bank who will really need their money Monday morning. Most had precisely $250,000 on deposit. We live to serve your mission of getting these insured depositors their statutorily mandated money back and after you cut one wire to us you’re done with them; we’ll take it from here.".
The author says you pick a "Too Big To Fail" Custodian, and my take is their tone is intended as serious advice (beyond just tongue-in-cheek) while also being an observation about groupthink in the industry:
What happens if Custodian fails? Well, ahem, plausibly the world ends in fire and blood. This is why Custodian was specifically chosen from the ranks of a count-on-your-hands number of the largest financial institutions in the world. This isn’t even the thousandth most important thing that breaks if Custodian breaks. Custodian cannot be allowed to break. Custodian is Too Big To Fail.
Realistically, Sweep customers are counting on the Custodian to properly maintain all the contracts, diligence, etc. that keep their deposits correctly segregated and protected.
Genuinely curious: How else could you mitigate risk here aside from obliging them to maintain diversified insurance, or using multiple Custodians (which puts you back at having to deal with multiple institutions)?
The father in law of an ex colleague of mine made a ton of money by investing in relatively high risk government bonds from his home country (Italy) many decades ago.
His thought process was basically "if Italy fails I'm gonna be screwed anyway", and it kinda worked for him.
I don’t understand what the fintech is adding to the equation.
The goal of these products is that, to you the user, it feels like "money in your account." You don't need to think of the mechanics of brokered CDs if you e.g. want to keep $750k on deposit at Schwab. You just... keep $750k on deposit at Schwab, and Schwab does some magic you don't care about, and you're insured.
Scale up the numbers a bit and sell it to companies, and that is the fintech pitch.
What happens if you use multiple sweep products? In your example the one from Schwab for 750k$ and the other one for (don’t quite remember) 5m$? Who checks that those two didn’t use overlapping banks?
It's also not a core competency a start-up should have to focus on. For all their talk of adding value, outsourced treasury management would have been a valid one.
This is why real businesses with real money have their bank use ICS or CDARS to get insured deposits, or liquid, short-term CDs, at N different partner banks. No leaving FDIC-insured-land, and it gives you easy withdrawals, too. It’s just a service offered as part of the banking network.
Even small businesses that exceed FDIC-insured balances can also roll short-term T-bills, through various brokerages (most of which don't charge fees on Treasuries) or TreasuryDirect (which is free). That combined with a line of credit should make for decent cash management.
This is why real businesses with real money have CFOs. Anyone who was sweating the SVB collapse should hire a virtual CFO right fucking now because their company has no financial leadership.
I'd think logically, in this hypothetical scenario, that money from the FDIC receivership would be deposited into a new bank such that there is continuous coverage of deposit insurance on all the customers. But yes, reality is that fintech firms don't do everything perfectly, and there are complex constraints that I'm sure we're not aware of.
One key problem is bankruptcy. If the fintech ever touches the money and then goes BK, the court can and will pull back that money and likely pay some to folks who are not the customers, like employees who have senior BK claims.
In the linked hypothetical, fintech never touches the money, but the devil is in the implementation details and shortcuts happen.
This is important: FDIC insurance protects you if the bank goes bust, but it doesn’t protect you if the company/fintech/exchange you’re using goes bust, even if they use an FDIC-insured bank.
This exact issue came up repeatedly in the past few years with crypto companies whose FAQs and even executives falsely told people their deposited money was FDIC-insured simply because the company itself banked with an FDIC bank. That’s not how it works.
An assurance of 100% backing with no limit is only as valuable as you think the aggregate institution is stable, which you should read as 'not very' on a long enough timeline. The CUDIC
does not have a 1 to 1 backing for these assets and it can't print more, so there's risk there.
I will say the Canadian banking system (as well as credit unions, which are regulated differently) is generally quite healthy and stable, I don't think it's imminent risk of collapse. But I would certainly not count on this guarantee, it's very silly.
I don't live in Canada but as your (very) southern neighbor, it seems that banks could be in trouble with the real estate market having been so out of control. As interest rates rise, isn't there a real risk of forced liquidations by leveraged property speculators resulting in banks sitting on a lot of losses?
It's possible; there really have to be two pieces for it all to come crashing down. The first is rising interest rates leaving people unable to pay, and the other is prices falling enough that mortgages would be underwater.
I think the first is pretty likely, but not sure how many people that will hit. There are decent stress test rules so anyone who couldn't afford a 1 or 2 percent jump shouldn't have qualified. The second is less likely, I think. We've got an absolute minimum 5% down payment, and almost always more like 10%, and anything under 20% requires default insurance. So I think pretty low chances unless the housing market craters ~30% or more. Which might be what it actually should fall by, honestly, but the government will probably intervene way before that happens.
> Mortgage default insurance protects lenders in the event a borrower defaults on their mortgage. It does not protect the borrower or a guarantor. If a borrower defaults, the insurer may oversee all legal proceedings and payment enforcement. In addition, the insurer compensates the lender should there be a shortfall after the property has been sold and expenses paid. The defaulting borrower remains responsible for any shortfall on the mortgage and the lender or mortgage insurer may pursue the borrower for any deficiency following sale of the property.[1]
I just want to highlight that last part because the insurance is 100% not for the borrower, but for the lender. Just like when any other insurance policy is used, the insurance provider has a strong incentive to reduce how much of the policy's pay out it pays from its own reserves and that means going after the borrower.
There is also a vague idea that because one of the major mortgage insurers is a government corporation (CMHC), they will pay out and not pursue the borrower. This is not true, and if Canadians know anything it's that no one loves taking other people's money as much as the government. To get out of paying you have to declare bankruptcy, much like the US. [2]
So in short, the Canadian system is set up to have the borrower pay for the lender's ability to expeditiously be made whole in the case of default, and for the insurer to then pursue compensation from the borrower if the asset price is below the loan value. It's the most Canadian thing ever: the borrower is getting screwed but thinks it's a good thing.
The 2008 collapse didn’t really happen up here. I think the big five banks are so heavily regulated that the bar is exceptionally high for catastrophe. The stress test rules around mortgages and mandatory default insurance for high ratio mortgages (less than 20% down) also help.
Interest rates were incredibly low, but they still weren’t just handing out mortgages like we saw in the States in the previous collapse.
I don't think this actually monetizable but if they're giving you unlimited free deposit insurance, while other banks aren't, then you could keep a bunch of money there & resell the deposit insurance.
Of course, if you actually pitched someone on this, their first question would be, "who insures you?" At which point you either didn't tell them (and lose their trust and they don't give you their money), tell them and they don't trust the coops (in which case you don't get their money), or tell them and they do trust the coops (in which case, they'll cut you out as a middle man and bank with these coops).
All the insurance does is make sure you get your money back if the credit union ogoes under.
If I put $500,000 of my own money in the credit union I can't see how I can sell my "insurance" to someone else as I need it to protect my own money.
If someone else wants the insurance they have to put their own money into the credit union. I can't sell my insurance to someone else unless I hand them my money but I cant see why i'd do that as it wouldn't make any sense and its certainly not an arbitrage opportunity as the OP suggested.
You'd either write an insurance contract and pray you have enough money in your account when the redemptions come or you'd accept their deposit and put it into your account & basically be a bank with a negative interest rate (that negative rate being your insurance premium).
The arbitrage is that in some places, universal deposit insurance costs >$0 but these banks are giving it out free. So there is a difference here that's exploitable on paper, but presumably the reason it exists is that it isn't in practice.
It fits the strict definition of an arbitrage opportunity in that there's a difference in pricing that you could exploit without taking on market risk, in a frictionless vacuum where wires clear instantly and carry no transaction fees and a bunch of other unrealistic assumptions.
I don't view it as real, but it gave me a chuckle, like when people turn DNS into an ersatz file system or similar hackery. Like a pun in the financial system.
>>>I can't sell my insurance to someone else unless I hand them my money but I cant see why i'd do that as it wouldn't make any sense and its certainly not an arbitrage opportunity as the OP suggested.
It's not an opportunity for an individual, it's one for a bank.
You start SVB2, and say "we charge significant fees and insurance premiums, but also offer 100% deposit insurance, no cap". And then your entire existence as a bank is just as a front-end to those credit union deposits.
edit: and yah, maxbond hit the nail on the head, it's not a real suggestion, it's a joke.
About the same, in that the vast majority of Canadians perceive both to be incredibly safe. Which they probably are, the biggest risk would be a huge housing crash, which the government has a lot of incentive to prevent.
> Section 271 of the Financial Institutions Act provides that the Provincial Government may provide financial support to the CUDIC Fund if the CUDIC Fund is impaired (see the CUDIC Fund Target Policy for further details about the CUDIC Fund).
The federal deposit insurance in Canada is the same (cdic). The relevant legislation explicitly says the government has no obligation to bail out the insurance fund.
Practically the fund doesn't have sufficient funds to bail out a big bank failure - let alone multiple - and everyone expects a government bailout.
No, not really (the law says they "may" provide support to the deposit insurance fund), but even if it was, the province can't mint money, so that could only get so far.
What were SVB customers getting in exchange for encouraging every company to keep their $ in uninsured deposits earning low interest?
Was it 2.5% 30yr mortgages for founders? Was it liquidity, in the form of loans against illiquid pre-IPO stock?
Were the interests of the companies aligned with those of the people who were getting the perks? (At 4-5% for parking excess cash in simple T bill it seems like a pricey perk)
> What were SVB customers getting in exchange for encouraging every company to keep their $ in uninsured deposits earning low interest?
Sometimes the issue is that there is nothing better, that is known to be available.
I can't recall being in a treasury management meeting where the CFO or VP of finance raised a concern about maxing coverage by FDIC insurance. In fact, many transactions that even a $2-3M per year company does require pooling more than FDIC covered amounts in a bank account just to cover receiving payments or making payments.
Incidentally SVB did have very comparable treasury management feature to other banks.
Finally, to address the perk part of the post, at no time did I see SVB offer anything other than market rates for mortgages - the difference is that they would not reject founders immediately because they were a business owner with no W-2 income.
> can't recall being in a treasury management meeting where the CFO or VP of finance raised a concern about maxing coverage by FDIC insurance
This is standard issue for corporate America. People who cut their teeth in the last decade's tech boom didn't learn it. But managing counterparty risk is one of the core jobs of corporate treasurers.
"Sometimes the issue is that there is nothing better, that is known to be available."
Where SVB customers money in a money market fund that failed, or was it simply a "bank deposit" ?
There are plenty of money market funds that are ungated and that are based on very short term treasuries that can be used as cash: write checks,ue credit cards, wire money, schedule payments, use autopay, etc, etc.
This is what I do - I keep $0 in the "bank", have almost all the services a bank can provide, and all my "cash" is guaranteed by the federal government.
Maybe I am misunderstanding something here. I am not, and never have been, a CFO.
Frankly I don't think the deposit insurance thing is a big deal -- It's really really hard to imagine a world where a big enough bank fails and people don't get $1.
But the interest rate spread ... Seems like a lot now? Wasn't, for a long time?
Talk to your accountant (Most of you here on HN probably make enough to warrant having one of these), Talk to your banker, and if you have more that 50 people working for you then you have a firm or a CIO...
You don't call a plumber to fix your electrical wiring... Get advice from a professional and rest easy at night. Take this article and the comments here as you being an informed consumer when you go to your accountant or CFO... speaking their language and understanding what they have to say will benefit both of you.
If I was a billionaire I would readily pay 1-5 million dollars a year to a bank that would guarantee an entire 100 million dollar account by not lending out any of the money. This would of course have to be a bank that didn't lend out any money at all and existed entirely to server individuals that didn't want their money held in a bank that used fractional reserve banking.
It really makes no sense. The bank's "cash" is really just database entries with the Fed anyway, why not loan the money to the Fed overnight (reverse repo) at 5%? The reason GP's dream doesn't exist is because MMs do effectively the same thing and pay you.
Not even a great solution either. For zero risk and zero out-of-pocket cost, you could just find a broker on commission to ladder you in 30-day T-Bills and manage the purchases.
Massachusetts has an optional Depositors Insurance Fund[1] that banks can join which insures all their deposits above 250k, seems like if you were really worried you could just open an account in one of those banks.
Still surprising to me how tech is trying to spin this as some novel event.
Fractional reserve banking was a known quantity. Risk management of your cash matters in a system like that. Full stop. Whether you think fractional reserve banking is dumb idea is a different discussion, but to willfully pretend that the depositors weren't at least partially culpable for this collapse is something else.
What even do you mean by “tech”? Tech has no humility? Nonsense.
Nobody has humility. Everyone trying to build a business fakes it. Your HVAC guy is highly likely to be a bozo (and overcharging besides). This isn’t “bad” it just is.
Right. It is incredible to me that a lot of "thought leaders" are acting like this situation hasn't been handled in a structured way since the 30s by the FDIC.
It has really made me believe that a lot of tech is just grifters reinventing wheels everywhere, but with great marketing.
I’m still confused by this. Aren’t they making depositors mostly whole by selling all the banks assets? I believe they are at least doing that. But are they also injecting money to make up the difference?
Banks cut sweetheart deals with large depositors to keep their deposits with them. A sweep account is less lucrative than one you can YOLO into long-term Treasuries. I'm not surprised most start-ups didn't have a treasurer; I am surprised e.g. Roku or Circle didn't.
Most depositor networks demand reciprocity for their members, so a dollar that leaves the bank in a sweep comes back from some other bank. They are technically more expensive than a large depositor that wanders in off the street in that the network takes a bit but from a total customer acquisition cost it really depends on the bank.
For some banks brokered deposits subsidize all other deposits because their customers otherwise are more expensive. Many small banks that serve underprivileged communities are in this category.
For SVB it seems this was decidedly not true! They had low cost to acquire large depositors and had no market pressure to engage with a deposits network.
This is called full reserve banking, which is a step further than narrow banking (where you only invest in essentially zero risk assets like short term government bonds or let money sit in a federal reserve account).
That says narrow banking means you park it at the fed. In theory, couldn’t you keep the value of all deposits in cash in a vault and not break any rules?
To be clear, I’m not suggesting this is a good idea. I’m just trying to clarify if this is allowed. Because at face value the idea that keeping all deposits liquid is a systemic risk seems counter intuitive. It only begins to make sense with the specific context of parking that money at the Fed.
Right, so you can think of it as a spectrum where full reserve banking (all deposits held directly, you're most likely going to have to pay for the privilege of secure storage and transactions) is on one end, fractional reserve banking (a small portion of deposits are held liquid, the rest is loaned out to generate interest and pay for operations) is on the other end, and narrow banking (park money in short term treasury notes and at the fed) is in the middle.
In our current system, every bank or bank-like institution does fractional reserve banking. The systemic risk posed by a narrow bank or a full reserve bank is that during any potential banking crisis, there is a strong prisoner's dilemma style incentive for depositors to "defect" by withdrawing all of their money from normal banks and moving it to narrow or full reserve banks all of a sudden. This is broadly why regulators won't let you run one.
Now, to answer your core question AIUI: "what stops me from running a full reserve bank [equiavalent]?"
In theory, you could offer a secure storage service like safety deposit boxes etc and not be subject to finance regulatory controls. In fact, most cities do have businesses that offer secure storage facilities that you can rent space in to store your property and retrieve it as needed. This is perfectly legal and honestly such a business doesn't usually know or even need to know the contents of such storage lockers etc.
The problem comes when you want to integrate with the larger financial system. Finance regs are far reaching in a "viral" sort of way; they don't just restrict how you do business, they also typically restrict whom you can do what business with based on how and what kind of business they do. The result of this is that you end up under the purview and subject to the approval of regulators one way or another, at which point they disallow your business.
As a simple example, if you want to make it so your customers can easily send or receive money, you'll probably want to be able to process ACH and wire transactions, at which point you fall under extensive regulation administered primarily by the Treasury and the Federal Reserve. I imagine the same will be true for things like debit cards on the standard networks since they're already subject to regulation of a viral nature and so on and so on.
Even if there were it wouldn’t obviate the need for orderly receiverships for banks. Operational risk is a common reason for holes in a banks balance sheet (eg fraud by a bank employee).
Deposit insurance is a useful tool to manage the orderly unraveling of a bank independent of how it goes bust.
That makes sense, but how many US banks have failed due to fraud or error? Counting if deposits = money at the fed seems vastly easier managing loans that differ from your deposits.
What’s the point of doing this anymore? It seems FDIC will insure anything too big to fail. You might incurs more risks by spreading your money like this.
From the article: “We represent 1,728 insured depositors at the now-failed bank who will really need their money Monday morning. Most had precisely $250,000 on deposit. We live to serve your mission of getting these insured depositors their statutorily mandated money back and after you cut one wire to us you’re done with them; we’ll take it from here.".
What could possibly go wrong with that?