Where you are. You have posted margin. Lesson 1 noted that margin sits in a segregated account and is still the customer’s money, and promised this lesson would ask how solid that separation is. The answer is that it depends entirely on a clause, and that the collateral you posted is very likely somewhere else by now.
Where is the bill you posted?
You pledged a Treasury bill to your prime broker as margin. Ask where it is.
Not, in all likelihood, sitting in an account with your name on it. Your agreement almost certainly permits the broker to re-pledge it, and the broker has done so - posting it to its own lender to raise cash. That lender has pledged it onward in a repo. The money fund at the end of that chain now has it.
Four parties. One bill. Everybody’s records are accurate, every pledge was contractual, and nobody has done anything improper. Follow the claims and there are three of them, on one instrument.
The idea in one paragraph
Rehypothecation is the reuse of pledged collateral by the party that received it. Your broker takes your bill as security and, under a clause you agreed to, pledges it onward to raise its own funding, and that party may do the same. The result is a chain in which one instrument supports several obligations at once, which is genuinely useful - it means the financial system needs far less collateral than the sum of what everyone has pledged - and genuinely dangerous, because the total leverage created does not appear on any single balance sheet. Each firm’s books show one pledge in and one pledge out. Only by tracing the whole chain, which no participant can see, does the multiplication become visible. When somebody in the middle fails, everybody discovers simultaneously that they were relying on the same bill.
Counting the claims
The exercise walks the bill down a four-party chain and records each pledge: hedge fund to prime broker, prime broker to repo desk, repo desk to money fund. Three pledges, three claims, one bill, and only the last party actually holds it.
That is the whole mechanism, and its plainness is the point. There is no exotic instrument here, no derivative, no model. It is a chain of ordinary secured lending, each link individually sensible.
Why anybody permits it
Because collateral is scarce and reusing it is efficient. If every pledged instrument had to sit immobilised, the system would need several times as much high-quality collateral as it has, and secured lending would be far more expensive.
The customer is compensated too: agreeing to a rehypothecation clause typically buys cheaper financing, and it is often the difference between a viable prime brokerage relationship and an unaffordable one. It is a real trade, not a trick, and the customer’s side of it is a lower rate in exchange for a weaker claim on their own collateral.
Leverage nobody can see
The regulatory difficulty here is not that the chain exists but that nobody can measure it. Each firm reports its own position accurately, and no report anywhere contains the length of the chain or the number of claims on a given instrument.
That means the system’s true collateral leverage is not a number anybody has, and estimates of it vary widely. This is the same problem as module 4’s prime brokers, one layer down: exposures that look independent, and are not, because they run through the same instrument.
Check yourself
1. You pledged one bill. Three parties have a claim on it. Who did something wrong?
Nobody. Each pledge was permitted by the agreement between those two parties, and each firm’s records are accurate. The multiplication is a property of the chain rather than of any link in it, which is precisely why it is hard to see and hard to regulate.
2. Why does the total leverage created not appear on any balance sheet?
Because each firm’s books show only its own two transactions: collateral received and collateral pledged onward. The chain’s length and the number of claims on a single instrument exist only across firms, and no participant sees more than their own link.
3. Why would a customer agree to a rehypothecation clause?
Because it buys cheaper financing. The broker can reuse the collateral to fund itself, and it shares that benefit through a lower rate. The customer accepts a weaker claim on their own collateral in exchange, which is a genuine trade rather than a trick.
4. Your broker fails while your collateral is out on a chain. What is your position?
You are an unsecured creditor for the value of the collateral rather than the owner of a specific instrument you can reclaim. The bill is somewhere down the chain backing somebody else’s obligation, and your claim is against a failed firm’s estate - which is why some jurisdictions cap how much may be re-pledged.
Do this
Trace the chain and count the claims.
python3 code/rehypothecation.pyThe starter leaves you pledge_chain, returning the pledgor and pledgee of each link as the bill moves down the four-party chain. The assertions check that three pledges are created and that the chain begins with the hedge fund and ends with the money fund. Success prints each pledge and the line:
nothing was forged and nobody lied: each pledge was legitimate, and the collateral is now promised to three parties at onceThe completed version is in solutions/rehypothecation.py.
What you can now do. You can trace collateral through a pledge chain, count the claims created, and explain why the resulting leverage is invisible to every participant. The next lesson introduces the institution designed to make these webs of exposure legible by replacing them with a hub.