You're basically describing a bank run, or a run on any institution whose business it is to make investments in relatively illiquid assets, earn a return, and pay its investors a share of that return. The model relies on steady inflows and outflows.
A bank run is a very special case, and while it might have similarities, it pays to be cautious when drawing comparisons.
Bank runs are special because of the fractional reserve banking system.
(Usually its at this point that someone launches into a tirade vaguely conspiracy-theory-ish about fractional reserve banking system, and complains that money is debt, but I grok it)
Meaning, rather than let the cash sit in a vault, the bank is allowed to loan that money out to other people. If everyone wanted their money back, the bank would not be able to fulfill the request until they have recalled all those loans or perhaps sold the loans to other banks.
BoE has some news for you:
https://bankunderground.co.uk/2015/06/30/banks-are-not-inter...
http://www.bloomberg.com/view/articles/2014-08-27/lending-cl...
Lending club is an entity that bears no risk because it simply matches capitalized lenders with borrowers in the present. It matches supply and demand right now.
A bank is an entity that bears risk because it matches borrowers with lenders in the future - meaning that the borrower might simply be lending to his or her future self. The bank is willing to spread that risk among all of its shareholders, and is usually backed up by a government agency to provide that service.
But in practice they have
1) operational risk - insofar as they are the middle-man and so are exposed to investors suing them when the loans go south
2) Implicit credit risk - in that if their loans blow up the losses suffered by their levered investors will likely prevent them from coming back to the market place to 'roll the loans'
3) Explicit credit risk - in that they have invested capital in a subsidiary HF (Cirrix) which buys their loans and is on the hook for the losses (which could flow up to the parent)
http://www.inc.com/business-insider/inside-lending-club-scan... https://personalmoneyservice.com/lending-club-fraud/
Key passage:
"As a result, the company may need to use its own funds to purchase these loans in the coming months."
In other words, LendingClub is going to fundamentally shift its business model from taking no risk to taking on the risk of borrowers defaulting. The startup sold itself as simply a marketplace, connecting borrowers with investors, but now it is buying its own product. The equivalent would be Airbnb buying up loads of houses to list on its own platform, to keep it growing."
Historically, eg Canada and Australia had good experiences with that setup. They had a more stable and advanced banking system in the relevant times (around 19th century-ish, I think), than the much more invasively regulated US at the time.
The fund has some cash around, and the rest sits in less liquid assets, too.
As an aside: there's strong pressure to cover losses from money market funds. (Usually on the sponsors of the fund, but I think there was a government bailout recently?) Even though those funds don't have a formal guarantee. But they are supposed to be liquid.
"A money market fund is said to "break the buck" when its NAV falls below $1.00 per share. In the nearly 40-year history of money market mutual funds, this has happened on only two occasions—in 1994, when a fund lost approximately four cents on the dollar, and in September 2008, when the NAVs of money market funds issued by The Reserve Fund fell below $1.00.
Typically, there has been an expectation that when a money market fund reaches a point where it might break the buck, the investment management firm that sponsors the fund will take action to infuse the fund with cash so that the fund can maintain a stable NAV of $1.00 per share. Most money market funds in the U.S. are sponsored by large financial institutions that may provide assistance in the case of instability."
That model is far too simplistic. And for most intents and purposes, especially when talking about bank runs, it fails rather badly.
I'll give the following explanation. Its going to sound like a huge scam, a ponzi scheme, a hugely irrational method, but in actual fact it makes a lot of sense for reasons that I'm not intelligent enough to put into words. But, its a rather elegant way of controlling the money supply.
So....
Imagine that I start a bank called Humber Necks Bank. We start with 0 customers.
Then Adam comes in and deposits $400.
The Ben comes in and asks for a loan of $200. We give him half of the money we got from Adam. Ben deposits this $200 in the bank.
The Charlie comes in and asks for a loan of $100. We give him half of the money we got from Ben. Charlie deposits this $100 in the bank.
The David comes in and asks for a loan of $50. We give him half of the money we got from Charlie. David deposits this $50 in the bank.
Then all of a sudden Adam, Ben, Charlie and David want to buy a bunch of in-game items on Pokemon go.
Adam will come in and ask for $400, Ben will come in and ask for $200, Charlie will come in and ask for $100, and David will come in and ask for $50.
That's a huge problem, because all I actually have is $400 (that Adam originally deposited).
(And in practice, what you describe would apply only if we used eg gold as money. In a fiat currency system banks are even weirder.)