While I may not be swayed by one trade, I can clearly use it to talk about it as a market-clearing event that others might want to discuss, and possibly (hopefully) react to.
Last week there was a calendar spread trade in SFRX16/X17 at 8 points. Recall that a calendar spread is where a trader enters an order to simultaneously go long one contract conditional on being able to short another at a given spread. For calendar spreads this allows someone to enter a trade to express a view on the difference in value between two expirations (among other reasons). A trader entering such a trade may have a view not on the absolute level of future prices, but on the rate of change between two dates. (Similar logic can also be one of the rationales traders use to enter IC (intercity) spread orders.)
While calendar spread trades are quoted in points, I find it useful to translate price differences into percentage differences. The chart below takes the prices for outright SFR contracts as well as calendar spread orders and tries to translate those into percentage differences. For example, the difference between the mid-market prices of the SFRX16 and X17 contracts (of 7.3 points) is 5.80% of the X16 mid-market level. However a trader can’t trade on mid-market levels. If one wanted to enter a long in one contract and a short in the other a trader could lift the offer on X16 (233.6) and hit the bid on X17( 237.0) for a spread of 3.4 points, or lift the offer on X17 (240.8) and hit the bid on X16 (229.6) for a spread of 11.2 points.
A much more efficient way is the calendar spread markets quoted on the right side. There the 7.2 bid (i.e. willingness to sell X16 7.2 points lower than X17) is consistent with implied HPA of 3.20%, while the 9.2 offer is consistent with HPA of 3.97%. Note that some futures platforms reverse the sign on these quotes.
But wait a minute -hasn’t HPA in SFR been averaging 10%. How can 3-4% implied HPA make sense?
The two main reasons might include: 1) that market prices are consistent with SFR HPA slowing dramatically, and 2) that the contracts are so thinly traded (and the quotes have no depth) that calendar spreads quotes are just wrong.
The following charts show year-on-year (YOY) changes to both the CUS and SFR indices (in black). It also shows the percent difference between forward contracts and either earlier indices (e.g. May 16 futures vs. CS index released in May 2015) or differences between two contracts (e.g. Nov ’17 vs. Nov ’16 calendar spreads). Bids (blue) and offers (in red) are highlighted.
The CUS chart shows that YOY gains have been running in the 4-6% range but that forward spreads are consistent with gains of 2-4%. This is consistent with many analysts who’ve called for smaller (but still positive) gains in home prices.
The SFR chart shows the 10% HPA gains from May 2015, but then shows a much more pronounced drop in forward HPA. While most forward calendar spreads have been mine (prompting traders for a reaction) last week’s trade of SFR X16/X17 at 8 points is consistent with the notion that at least one other trader thought that implied HPAs of 3-4% made sense (again a) for only one lot, and b) the trader may have had other reasons).
So, fans of SFR/ worriers of an SFR bubble, what will/should the HPA be (using the Case Shiller index) for the period between the X16 and X17 contracts? That debate has been going on for a while. We now have a data point in that discussion.
Please feel free to contact me (firstname.lastname@example.org) if you’d care to discuss this or weigh in with a trade.