Consider Beta versus VHS in video recording, or GSM versus CDMA and TDMA in cellular telephony. When new technologies develop, very often competing methodologies vie to become the standard which all participants must recognize and respect, even if it is not a particular participant's preferred solution. Why did the Eurodollar contract, long a weaker sibling to the T-Bill contract, surpass and ultimately bury its brother? The flexibility and utility of the simple contract design were crucial advantages. The portability of the design to other non-dollar money markets was helpful. The method of cash settlement, rather than of physical delivery, removed limitations that constrained the T-bill contract. The Eurodollar contract became the standard for all of these reasons, among numerous others.
In the young U.S. inflation swap market, a single standard has not yet developed. Instead, two competing structures--zero-coupon inflation swaps and TIPS asset swaps--jockey back and forth for supremacy. A single dominant standard ranks high in the list of things this market needs.
In last week's Learning Curve, we discussed the pros and cons of these two competing structures. Now, we will examine the effect of having two competing standards on the market.
Loss Of Liquidity
The two different standards in the U.S. CPI market reflect a fundamental disagreement over what is the elemental underlying transaction that is being captured. This leads to occasional losses of liquidity when certain underlying market developments imply different derivative market prices depending on the lens through which they are viewed.
For example, consider the case of a significant widening in LIBOR swap spreads. Suppose for the sake of this illustration that LIBOR swap rates rise while Treasury and TIPS yields remain unchanged. To those market participants who consider zero-coupon inflation swaps as derivative of breakeven inflation rates, the zero-coupon swap curve shouldn't move since, after all, the breakeven inflation-which is the difference between Treasury and TIPS yields--hasn't moved. Whatever relationship between breakeven inflation and zero-coupon swaps prevailed in the moment before LIBOR rates rose should prevail afterwards, since LIBOR is not an element of zero-coupon swaps. The implication is that TIPS asset swap spreads should have increased as much as, or nearly as much as, Treasury asset swap spreads have. In both cases, it is the introduction of LIBOR flows into a pure TIPS or Treasury flow that adds volatility to the asset swap structure.
But in the same circumstance, inflation market participants who focus on TIPS asset swaps will see the zero-coupon swap market rates as having shifted higher by approximately the same amount as the movement in LIBOR rates. Why? These traders insightfully observe that the TIPS asset swap level reflects the expected financing cost of TIPS relative to LIBOR, which--since TIPS are almost always general collateral in the repo market--doesn't move much. But if the asset swap doesn't move much, then since the value of the LIBOR side of that swap has changed the value of the inflation-indexed side must also change, and the mechanism for that change is through the zero-coupon swap curve.
Thus, when LIBOR swap spreads widen the zero-coupon swap disciples see zero-coupon swap rates as unaffected while asset swap aficionados see zero-coupon swap rates as rising. This leads in such a circumstance to at least a temporary widening of bid/offer spreads in the inflation market as the two sides "negotiate" through prices in the market.
Who is right? It depends on whether the widening of LIBOR swap spreads was caused by (1) a general richening of Treasuries in repo, in which case the asset swappers are right, (2) a widening of general collateral repo/LIBOR spreads, in which case the zero-coupon camp is right, or (3) other LIBOR-related dynamics such as new issue hedging or mortgage paying, in which case the zero-coupon camp is probably closer to right since the dynamic is on the LIBOR side. But the issue for the purposes of our discussion is that the fact there is not a singular inflation swaps standard means that pricing conditions in the LIBOR market affect the liquidity conditions in the CPI swaps market.
Seasonality
A second difference in the perceptions of the CPI swaps market participants involves seasonality. If the asset swap market merely reflects relative financing rates, then the shape of the asset swap spread curve should be smooth when all bonds trade at general collateral repo rates. But then what to do about the April 2010 and April 2011 TIPS issues, which trade idiosyncratically expensive to normal seasonal patterns? The implication, seen through the asset swap lens, is that there must be a very complex pattern to implied future seasonality, which makes pricing forward inflation trades very tricky. It seems much more reasonable to postulate that these issues are expensive for other reasons and that therefore they should trade at wider asset swap spreads than surrounding issues. Here the zero-coupon camp has won a partial victory: these issues do trade wide to surrounding issues although they finance the same, implying that the swap market is "right" and the bond market is "wrong" about the regularity of the expected seasonal pattern in the future.
The competition between zero-coupon and asset swaps continues. One will win out as the market standard, eventually, because the gain in efficiency from having a single standard outweighs the loss in specificity from having multiple standards. In inflation swaps, the cleaner structure is the zero-coupon swap, and the resulting ease with which inflation curves can be constructed, inflation swaps can be understood, and higher-order derivatives--such as swaptions--can be priced will probably lead to that becoming the standard. This doesn't mean that both types, and other types of inflation swaps, won't all trade: after all, television producers still prefer Beta for their purposes.
This week's Learning Curve was written byMichael Ashton, inflation product manager at IXIS Capital Marketsin New York.
