Minyan Mailbag - Mortgage Backed Securities
Note: Our goal in Minyanville is to remove intimidation from the financial markets and encourage an interactive dialogue among the Minyanship. We share this next discussion with that very intent.
Could you give the Minyans an overview of mortgage backed securities/collateralized mortgage obligations, how these derivatives are priced and the markets in which they're traded? Without trying to answer my own question, my gist is that there's a big market out there, oft-neglected by equity investors, that redistributes the various risks associated with investing in mortgages (especially prepayment risk), by dividing mortgage payment streams into interest-only and principal-only "tranches" that have unique pricing behaviors. The pricing models for CMOs really haven't been tested in the market under extremely dynamic conditions that might exist during a housing price bubble deflation, or a very rapid rate-of-change in interest rates that undermines prepayment assumptions. Am I correct in fearing that one result of big firms' investing in large CMO portfolios (think: Fannie Mae (FNM:NYSE), Freddie Mac (FRE:NYSE), maybe banks like WaMu (WM:NYSE) and big insurers with long time horizons, and now maybe the Federal Home Loan Banks) is that some firms may have either deliberately or inadvertently taken on huge risks, proportionally greater than the risks of investing in baskets of whole mortgages, that may threaten the stability of the financial system under a sufficiently bad shock?
I don't think this area has been addressed on Minyanville.
Minyan Ben Richter
Mortgage Backed Securities (MBS) are very complicated instruments for one basic reason: the pay-offs and risks are patterned like options. Remember, an option is a contingent liability; the actual liability is dependent on a future event. It is necessary then for the probability of that future event to be estimated. This estimate is based on linear statistics (extrapolating past data into the future), a methodology that has inherent flaws. When this probability turns out to be wrong, very bad things can happen.
The estimated event in the case of MBS's is pre-payments. When a mortgage is bought as an investment, the investor plans on getting all the cash flows dictated by the contract (interest and principal payments). When interest rates drop, debtors re-finance and consequently pay off those old mortgages to get new ones. The investor that bought those mortgages suddenly does not have the asset they thought they had. When that investor borrowed money to buy that asset, the investor is forced to pay off their liability, usually at a loss.
So the probability of pre-payment is estimated like an option: the volatility and distribution of interest rates is the primary variable in estimating this probability.
So what is an MBS in the first place? Starting with a pool of conventional residential mortgages with similar coupon rates and maturities, one can create a simple pass-through security, say FNMA 6% with a maturity of 30 years, which will pass through the interest and principal payments from the original mortgages to the security investor. The investor doesn't get cash-flows from one mortgage, but a pool of mortgages.
To make things more interesting, one can slice the cash flow stream from both interest and principal to create various CMO (Collateralized Mortgage Obligation) tranches, collateralized by the same pass-through. A CMO deal has in general PAC (Planned Amortization Classes) tranches and support tranches. These tranches have combinations of both principal and interest cash flows. There are also specialized tranches of just PO's (the cash flows are derived exclusively from principal payments) and IO's (cash flow from interest only).
Cash flows of a PAC tranche are guaranteed if the prepayment speed is within its definition band. Support tranches will receive cashflow after PAC cash flows are satisfied. In other words, if the pre-payment of mortgages is faster than anticipated, the support tranches lose their value first before anything is eaten away in the PAC tranches.
One can further break up a PAC tranche or a support tranche into different pieces, such as a floater (Libor plus 1%) and an inverse floater (5% minus Libor). The total between the two is the same, but when rates go up the floater tranche wins what the inverse floater tranche loses. This is just a Wall Street game to attract investors who like to bet on rates (or are hedging that risk away).
One rule of thumb: no matter how you slice a tranche, the total interest and total principal for the newly created tranches have to equal the interest and principal for the original tranche under all scenarios.
Today, the majority of the conventional mortgages are securitized in the form of pass-throughs. MBS's have made mortgages far more liquid. Liquidity, however, is an enigma. In normal times it can cause excesses to be built up, then in extreme situations it can dry up just at the wrong times.
In order to price a MBS/CMO tranche, one needs an OAS model (Option Adjusted Spread, which employs a Monte-Carlos simulation methodology), which has three different pieces: an interest rate model, which generates various rate scenarios for both short and long rates; a prepayment model, which projects pre-payment rates for each and every rate scenario generated by the interest rate model; and a cash flow distribution model, which will calculate cashflows for each and every rate scenario using projected prepayment speeds.
One can run all types of sensitivity simulations to gauge the risk numbers, but negative convexity is not an easy thing to manage when rates are extremely volatile.
And here is the main point. The thing that makes the whole pass-through arrangement fall apart is a mis-estimation of the volatility of interest rates. If rates drop much faster than anticipated, first the support tranches lose money and then the PAC tranches. Fast pre-payments make the PO's make a lot of money since the principal is paid sooner, while the IO's lose money, since there are less interest payments. In this way MBS's are essentially short option portfolios.
FNM, FRE, and GNM (Ginnie Mae which works with the government on low-income family mortgages) are the primary issuers of MBS's: they buy individual mortgages from banks and brokers (who also trade in MBS's) and package and sell them to mutual funds and other investors. We estimate that FNM would sell around 20% of their portfolio out through MBS's.
In 1993 as head of equity derivatives at PaineWebber I was called in to manage and unwind the firm's inventory of CMO's; everyone else had been fired.
The mortgage market has grown from $3 trillion to $7 trillion over that time. All this pre-payment convexity must be hedged in the tradeable government debt market which has only risen by $1 trillion over the same period. This situation not only drives up the volatility of bonds due to this re-hedging activity, it also leaves MBS holders susceptible to any exogenous event that creates extraneous volatility.
The information on this website solely reflects the analysis of or opinion about the performance of securities and financial markets by the writers whose articles appear on the site. The views expressed by the writers are not necessarily the views of Minyanville Media, Inc. or members of its management. Nothing contained on the website is intended to constitute a recommendation or advice addressed to an individual investor or category of investors to purchase, sell or hold any security, or to take any action with respect to the prospective movement of the securities markets or to solicit the purchase or sale of any security. Any investment decisions must be made by the reader either individually or in consultation with his or her investment professional. Minyanville writers and staff may trade or hold positions in securities that are discussed in articles appearing on the website. Writers of articles are required to disclose whether they have a position in any stock or fund discussed in an article, but are not permitted to disclose the size or direction of the position. Nothing on this website is intended to solicit business of any kind for a writer's business or fund. Minyanville management and staff as well as contributing writers will not respond to emails or other communications requesting investment advice.
Copyright 2011 Minyanville Media, Inc. All Rights Reserved.
Daily Recap Newsletter