Perseid Macro - Understanding Global Liquidity, Part 3: Bond Rehypothecation

Perseid Macro

One small player trying to make sense of the game

Understanding Global Liquidity, Part 3: Bond Rehypothecation

This is Part 3 of a series on Understanding Global Liquidity.

Part 2 covered the Eurodollar system, fractional reserve banking, and the critical role Eurodollars played in providing the money that the world needed to grow after WW2. As with any large, profitable trade, other players are going to want in on the action. That’s what Part 3 is about.

This article is based on what I learned about in Eurodollar University Part 2 on Macro Voices. I’ll be pushing a bit past what Jeffrey Snider covered there, and he’ll probably be covering a lot of this in part 3 of the Macro Voices series—but that doesn’t come out until Thanksgiving. Who can wait that long?

Bond Carry Trades

Let’s say I’m a bond trader in the early ’80s (right around the start of the 30-year bond bull market). What do I do? I execute bond orders for my clients, maybe advising them in ways to drive additional commissions. I might also use company capital to buy a basket of bonds that’s designed to give me the best risk adjusted return for our proprietary trading desk.

I started out with a mix of bonds from various countries, but I quickly discover the basic principle behind risk parity: I can use leverage to hold a larger quantity of less volatile bonds and achieve a higher return for the same volatilty. So instead of buying a mix of bonds with different yields from across the world, I just buy US Treasuries, borrow against them, and use the proceeds to buy even more US Treasuries.

But where do I get the loan?

Enter Bond Rehypothecation

Rehypothecation is the process of pledging the same collateral against multiple loans. When I buy a house, I take out a mortgage and pledge the house as collateral—that’s “hypothecation”. If my lender then pledges the house against a loan he takes out, that’s rehypothecation. Presumably, my lender’s lender might even finance their own activities by pledging the same house as collateral, yet again.

If I default on my loan and my lender defaults on his loan, there’s only one house to serve as collateral for the entire chain of loans—someone’s taking a bath.

A fractional reserve banking system is pretty much a system for rehypothecating a single dollar as collateral for multiple loans. Because of the inherent risks, the official banking systems are government sanctioned, regulated, and monitored. They’re also backed by the money-printing capacity of the respective countries’ central banks. As we talked about previously, as long people leave their base collateral in the system, everything just sort of works.

It’s a totally different game to rehypothecate the same collateral without the regulations, visibility and guarantees. It’s really easy to get yourself in a situation where confidence evaporates and everyone does try to take their collateral out of the system, causing a collapse. The governments of developed market nations have put a lot of measures in place to protect that confidence (after learning the hard way…)

Switching back to the bond market (which has much less oversight), the US at least passed regulations in 1933 to limit rehypothecation of securities. Broker-dealers can only rehypothecate the assets in the customers’ accounts to finance loans of up to 140% of the asset base. Also, they can only use the assets in margin accounts—and technically only if those assets aren’t fully paid for (i.e, the assets bought on margin) and technically only to the degree that they aren’t paid for (i.e., if I have a $10k margin loan on $100k of securities, they can only rehypothecate $10k’s worth).

Those protections only apply to securities held in the United States. If your broker-dealer moves your securities to, say, the United Kingdom, then UK laws apply. None of these protections exist in the UK. There, broker-dealers can rehypothecate securities in customer accounts with no legal limit—the only limits would be those imposed by the market. (As Manmohan Singh and James Aitken talk about in this IMF working paper this was still true as of 2010).

We now have a totally novel system for credit-money creation, which is basically a fractional reserve banking system with slightly different plumbing: the Bond Rehypothecation System.

Levering Up

As a bond-trader in the 1980’s, I’ve found my source of loans in the Bond Rehypothecation System:

  1. I get my customers to buy bonds in a margin account
  2. …move those bonds to a London-based subsidiary (creating an IOU from my London branch to the US branch)
  3. …take out loans in the London financial markets
  4. …use those loans to buy bonds that which yield more than my interest expense, generating positive carry
  5. …and go back and shove more bonds down my customers’ throats so I can lever up as hard as I can.

There are only a few limits on my activity:

  • What do my lenders allow as collateral?
  • How much leverage will my lenders give me against that collateral?
  • How much of that collateral can I get my hands on?
  • How much of that available collateral-supply can I convince my customers to buy?

Unwinding the Leverage

Fast forward 30 years, and this dynamic directly leads to the 2008 financial crash. We’ll cover the details later, but let’s look at the unwind now.

If I’m a broker-dealer, technically, the bonds I’ve pledged as collateral are still “owned” by my customers, but they have no idea what’s going on. I provide discounts for allowing this, and even my sophisticated hedge fund clients signed the boilerplate paperwork without realizing what it all meant. No one knows how levered up I am—they all assume I’m a safe haven.

When things fall apart, my US customers that opened accounts with my US branch find out that their securities are actually held in my London-based subsidiary (and totally overcommitted as collateral). In the ugly reality of bankruptcy proceedings, my lender(s) quickly lays claim to all assets in all my customers’ margin accounts (since I moved all margin account holdings to the UK). My customers become unsecured creditors as far as the bankruptcy proceedings are concerned. Political backlash might get them “made whole” eventually, but it isn’t pretty—the funds arrive slowly, causing my customers to miss out on the profitable post-crash recovery.

(As a small guy, the only way to protect yourself is to refuse margin accounts and instead open “cash” accounts. I have to keep your cash-account securities separate from the soup of margin securities—they are totally owned by you. In cash accounts you have to wait for the funds from selling a position to settle before you can use the proceeds to enter a new position. Trade settlement takes days, and for swing traders that’s a deal-breaker…So to protect against my tail risk, you’d have to alter your trading strategies and let many potentially profitable strategies go.)

Rehypothecation in itself isn’t evil, but it adds risk. If the risks are monitored and managed, the benefits of faster growth can outweigh the risk of collapse. But it’s not given that the benefits outweigh the risks, as the world saw in 2008.

The Evidence

Okay, that’s a nice story. Where’s the proof? Jeffrey P. Snider provides the proof in Eurodollar University, Part 2 on Macro Voices.

From the interview: In April 1991, Salomon Brother’s bond trader, Paul Mozer, placed above-market bids for the entire US Treasury auction. The SEC told him “Quit it! That’s not allowed.” Yet Mozer kept at it. The SEC went so far as to make a rule that no single trading organization can buy more than 35% of the auction—and names the rule after Mozer—but he did it again anyway. (I have no idea how he thought this was a good idea.) It was later discovered he even faked customer orders and generally ignored all the laws in getting as many bonds as possible stuffed into this customers’ accounts. The SEC gets pissed and pulls some license or other, sending Solomon Brothers in to a death spiral from which Warren Buffet has arrange a rescue.

Why would Mozer make these illegal bids on US Treasuries and go far as to fake orders? He was in the middle of a web of incredibly profitable carry trades. Whoever’s customers bought those bonds in NYC could borrow a lot in London for cheap and make a killing buying up higher yielding bonds with the credit-money (I haven’t been able to find exact ratios yet).

I don’t know the exact economics of the time. But it’s clear that—if Mozer had an unlimited appetite for US government bonds at above-market rates—there was a profit to be made there.

Basically, Mozer was running a fractional reserve banking system, where the collateral consisted of US government bonds. He was in the business of borrowing short, creating money via rehypothecation (skirting US regulations in the process) and using that money to lend long. He made himself a shadow-bank.

That’s the Bond Rehypothecation System.

The Problem

The 1991 situation shows that there was an incredible demand for debt financing. There just wasn’t enough money to loan out, even with the 2nd derivative fractional reserve banking system that is the Eurodollar Banking System running full steam. Why?

The Eurodollar system, despite its potential for growth, still required a growth in collateral (US Bank Credit Dollars shipped overseas) to grow itself. US banks could only add to their international liabilities at a certain rate. Plus—regardless of the lack of oversight—the banks creating Eurodollars could only lever up their collateral so much. They still had to consider liquidity in case people wanted their US Bank Credit Dollars back. Between those factors, the growth rate of the Eurodollar Banking System was just not fast enough to meet the growing global demand for credit.

The Bank Rehypothecation System, run by people like Mozer, was a new 2nd derivative fractional reserve system created to profit from the demand that the Eurodollar Banking System was leaving unmet. Like the Eurodollar Banking System, it was using US Bank Credit Dollars as collateral and multiplying that collateral to create new credit-money. It’s advantage was that it had an independent transmission system for getting at the US Bank Credit Dollars it used as collateral.

But the new system faced its own constraints. I don’t have a good history on reserve requirements in the ’80s—so correct me if I’m wrong here. The picture, as far I can see it, is that European banks were the ones lending against the collateral. Those banks had to meet their reserve requirements, and by regulation, only US government bonds counted towards those requirements. So only US government bonds would do as collateral.

This all mean that, in the ’80s, the collateral for this system had to meet very specific requirements. The collateral had to be:

  • US Treasury Bonds
  • …that were owned by the customers of broker-dealers
  • …in margin accounts where the customers consented to rehypothecation
  • …and held London-based subsidiaries to be under UK (lack of) laws

By the early 1990’s, the US government wasn’t issuing debt fast enough to feed the Bond Rehypothecation Systems’ demand for collateral. Even though the collateral was the debt issued by the world’s largest economy, while that economy engaged in deficit spending to fund a war—and the collateral was being multiplied into an unknown amount of credit-money—that still wasn’t fast enough to meet the demand for credit. Even with the Eurodollar system working along side it!


Eurodollar University Part 3 hasn’t been released. But Part 2 teased that the Eurodollar system was changed in the 90’s by Basel I. I think the story here is that Basel I opened up the classes of bonds that qualified as reserve assets for European banks, using a tiered system.

Depending on “riskiness” of the bonds, bank regulators determine the ratio at which different bonds qualify as reserves. From the Basel I article on Wikipedia:

Assets of banks were classified and grouped in five categories according to credit risk, carrying risk weights of 0% (for example cash, bullion, home country debt like Treasuries), 20% (securitisations such as mortgage-backed securities (MBS) with the highest AAA rating), 50% (municipal revenue bonds, residential mortgages), 100% (for example, most corporate debt), and some assets given No rating.

Basel I would allow banks to make loans against more than just US government bonds—and have those bonds still qualify as “reserves” (rehypothecation be damned). Money supply growth could then explode by using corporate bonds, municipal bonds, and Mortage Backed Securities (MBS’s) as collateral for the bond rehypothecation-based fractional reserve system.

It’s likely that the reason the US spent the ’80s in the top-right quadrant of the Credit Conditions Matrix is because the Eurodollar Banking System and the Bond Hypothecation System (when it was based on US Treasuries) supplied enough credit without over-supplying credit. Lots of people could qualify for debt financing, but there was still more demand than supply—causing interest rates to remain elevated, which meant that lenders demanded reasonable IRR hurdle rates, preventing malinvestment.

I suspect, once Basel I takes effect, the rate of credit-money creation went parabolic, and we moved into the top-left quandrant of the matrix. Interest rates started falling because capital-owners lost pricing pressure. IRR hurdle rates dropped (or were totally abandoned), and we got malinvestment like we’d just abandoned the gold standard (because we kind of did…)

Quite simply, from the standpoint of money-creation and credit (and as a result culturally), 2000–2007 were a repeat of the roaring 20’s, with a similar boom after the world actually dropped the gold standard in 1914.

Next Up

The Bond Rehypothecation System is a 2nd derivative fraction reserve system running in parallel to the Eurodollar Banking System. It sits “alongside” the Eurodollar Banking System conceptually, but in the real world, it’s impossible to draw bright lines between the three systems (US Banking, Eurodollar Banking, Bond Rehypothecation). They all generate credit-money, which gets transacted, transferred and turned into collateral for the other systems (and themselves).

While these two system are huge, critical and largely unknown, there are other key pieces of global liquidity to cover still. Part 4 will explore the Petrodollar.

DISCLAIMER: None of this intended as investment advice. This is a place for me to put down my ideas and share them with others, whom I fully expect will tear these ideas apart. I have no professional training or certifications, and I've probably already changed my mind on whatever you just read.