The impressive rally in the bond markets has pushed the 10-year yield to all-time lows, and MBS prices to record high levels. Front-month prices of Fannie 3.5s closed on 6/1 at 105-10, the same day that 10-year yields closed for the first time below the 1.50% level. However, the strength in bonds has served to obscure a number of details. The first is that the rally has been concentrated in longer maturities. Since May 1st, the yield on the 2-year Treasury has barely budged, pushing the 2-10 spread to its tightest level in years (as shown in Chart 1 below). The flattening of the curve has also impact the forward LIBOR rates; using Eurodollar futures, the forward LIBOR curve has also exhibited a noticeable flattening, as shown in Chart 2.
There are a number of ways to interpret these developments. One is that the longer end of the curve (i.e., 7-years and longer) is the only place where there is room to move; at less than 0.70%, the 5-year has little room to move. It also suggests that investors expect the Fed to continue to buy longer-dated Treasuries even after the end of the Twist operation next month. This will be especially true if the Fed initiates another round of asset purchases in order to offset the risks of a renewed slowdown emanating from the European turmoil.
Despite the rally, “realized volatility” (or “vol) is quite low. Bond traders generally watch both realized and so-called implied volatility. Realized vol represents a statistical measure of how volatile bond prices or yields have been over some defined period of time, while implied vols are market-implied numbers derived from an option’s price. Chart 3 indicates that 60-day realized volatility on the 10-year is close to its lowest level since last summer. Despite the steady decline in rates since early April, yields had been inching downward at a steady rate, although the market became quite volatile last week (which was reflected in the uptick in the shorter 40-day lookback volatility).
This is also reflected in implied volatilities on swaptions, which have also declined steadily throughout April and May and plunged last week. The tame volatility of the bond markets and the decline in implied vols both are generally supportive of MBS levels despite their stratospheric prices; since mortgages and MBS are implicitly short options, their prices tend to benefit from lower levels of volatility.
It’s worthwhile to revisit trading and discuss recent developments in 30-year 3s, an item of strong interest to lenders. I discussed the illiquidity and lack of sponsorship of Fannie 3s in a writeup a few weeks ago, prior to the latest downdraft in mortgage rates. With consumer rates trending below 3.80%, lenders now have an even greater incentive to attempt to sell 30-year 3s. It’s interesting to note that the price behavior of Fannie 3s, as illustrated by the Fannie 3.5/3.0 swap, has recently exhibited better and more rational behavior. Chart 4 contains a scatterchart showing the Fannie 3.5/3.0 swap versus the level of 10-year Treasury yields.
The chart indicates that the pricing of Fannie 3s) has been much better behaved since mid-March than it was earlier in the year. The R-squared of the fitted values (basically a measure of how closely the fitted values represent the data points) is 0.42 for the entire period in question; since March 9th, however, the R-square is 0.84, implying a much closer fit and more predictable relationship between the swap and Treasury yields.
While I continue to believe that 30-year 3s have no natural private buyer, the Fed appears to be filling the void by buying roughly 100-200mm in Fannie 3s each week. (In the last report, the Fed bought $250mm FN 3s in July.) While small relative to the purchases of other coupons, Fed buying has been enough to create at least the appearance of reasonable liquidity in the coupon, and makes it a decent sale for mortgage bankers looking to hedge their pipeline of sub-4% 30-year conventional loans. It’s also understood that the coupon will perform poorly in the event of a broad market selloff, which makes it a good sale in the current environment.
Originators need to be cautious, however, and aware of a number of potential pitfalls. At times, the coupon can seem to trade poorly, with fairly volatile price movements and very wide bid/ask spreads. For example, Treasuries opened sharply lower on Monday 5/21; early in the day, Fannie 3s had a quarter-point bid/ask spread before the market regained its footing.
Given its inherent illiquidity, originators using the coupon as a hedging vehicle should plan to assign loans to the trades; expecting to pair these trades off could backfire under a variety of circumstances. These could include a short squeeze, too many hedgers looking to pair out of trades at the same time, or the lack of sustained buying of the coupon by the Fed.