Liquidationists... You might want to hide your eyes, because what I'm about to say might be a bit disturbing.
The bond market needs liquidity in order to efficiently price securities. Specifically, it needs leveraged investors to have access to leverage.
Wait, wait. Don't fire up that outraged comment just yet. Hear me out.
Let's imagine a world where there are four investors and three asset classes. The three asset classes are...
- High quality bonds, which we'll just call "Quality" from now on.
- Low quality type A, which we'll just call "Type A."
- Low quality type B, which we'll call "Type B."
These three bonds types are perfectly correlated within each class, such that all Type A bonds move exactly the same. They have no fundamental correlation with each other, however. Therefore a credit event in one type doesn't necessarily have any impact on the other types, at least not directly.
The four investors have specific "preferred habitats." This is a term used in several academic works on the bond market. All it means is that investors tend to be involved in certain types of bonds (or certain maturities of bonds) because of their investment objectives. The relative value between the investor's preferred habitat and other types of bonds does not enter into their investment decision. In our case, we will assume that each investor only has expertise in certain bonds, and therefore will not venture into the classes where they have no expertise. I think in the short-term, this matches reality.
- The first we'll call "Mutual Fund." This is a non-leveraged investor, who invests in all three asset types. Mutual Fund's demand for bonds is a function of flows from outside investors. Therefore Mutual Fund's short-term demand curve is perfectly inelastic. By this I mean, it doesn't matter how cheap bonds get (or how rich) Mutual Fund can only invest what it has.
- The second we'll call "BSAM," just to make up an acronym. BSAM is invested equally in Quality and Type A assets and is maximum leveraged at 10x.
- The third we'll call "NLY." NLY invests in Quality assets only. They are currently 10x leveraged, but could be as much as 15x based on current lending conditions.
- The fourth we'll call "Alpha." Alpha invests in equally Type A and Type B assets and is 5x leveraged. They could go up to 8x.
OK. So now let's say there is an exogenous shock to Type A assets, causing them to lose 10% of their value. Nothing has fundamentally changed about any of the other assets.
So BSAM immediately becomes subject to margin calls. 50% of their assets are now down 10%, so to put it in dollar terms, they've suffered $5 in losses for every $100 in assets. Of course, they only had $10 in equity to begin with, so now they're down to $5 in equity. In order to get them back to 10x leverage, they have to sell half their assets. Because liquidity in Type A is poor, most of the sales are in Quality.
Alpha has suffered as well, but they had much more equity. They lost $5 for every $100 in assets, but they had $20 in equity. In order to get down to 8x leverage (the max allowed) they only need to sell $2.50 per $100. Again, since Type A is illiquid, Alpha sells Type B bonds.
So now we have selling in both Quality and Type B assets, but the question is, who is going to buy?
Let's put this in a supply and demand context. The supply of Quality and Type B has shifted outward. Note it isn't a move along the supply curve, because there has been no change in the price of either.
The demand curve is completely flat for Type B assets. Mutual Fund is the only potential buyer, and they have no assets with which to purchase bonds.
What about Quality assets? We know that NLY has some room to add, so there is potential demand there. Unfortunately, NLY's ability to add bonds is no where near the supply BSAM is dumping. NLY has $8.33 in equity for every $100 in assets. They could add about $25 in assets by levering up to 15x. But BSAM needs to sell $50 in bonds.
So in both cases there aren't enough investors capable of buying the bonds that are for sale. So there is no clearing price. At all. I guess the hard core mark to market crowd would write it all down to zero, but that's another topic.
In real life, Wall Street would serve as market maker, so the bonds would clear. But the prices would all be lower. But now it becomes self-feeding. Prices on Quality and Type B fall because of the technical of BSAM liquidating. That puts pressure on Alpha and NLY's leverage levels. Both then continue to sell more assets, and the Street soaks them up, but at still lower prices. Now the Street is over-leveraged themselves, and stops extending their repo lines. Now everyone is in trouble.
Notice that it took one heavily leveraged buyer to take a relatively modest loss to touch off a serious liquidity crunch. The other players had much more responsible leverage, and yet they got caught up in the liquidity crunch all the same.
OK. Now let's say that the Fed wants to try to deal with this. What the system needs is more cash to absorb the forced selling. Once that is absorbed, the extra cash can be taken away. What should they do?
- Extend loans to banks and brokerages. This gives them cash to continue making markets. It also prevents them from pulling credit lines away from investors with performing assets. The Fed actually did this by opening up discount window borrowing as well as allowing Citi and Bank of America to lend money to its brokerage unit through its bank unit.
- Force short-term rates down. This increases carry for banks and brokerages, which increases the profitability of market making. It also discourages investors from leaving money in cash, which over the intermediate term should increase demand for higher-quality bonds. Again, this helps absorb the overhang of bonds for sale.
I know some of you are just busting at the seams to write a comment along the lines of "Let 'em suffer! Why should the Fed bail these people out?!" But who's getting bailed out here? In my example, BSAM saw 50% of its capital wiped out, and the Fed's actions aren't changing that. The Fed has merely made it possible to sell those bonds at some price. In fact, under the scenario I developed, all three leveraged players wound up getting hurt to varying degrees.
I contend that if the Fed cuts rates down 75-100bps for a few months, then rapidly reverses course, the impact on inflation will be minimal. And the bond market is able work through this deleveraging process without too many innocent bystanders getting hurt.