I posted a blog on Friday that described how one might use HPHF Ratio Agreements to hedge the relative performance of a region versus the 10-city index. That post referenced Atlanta, which, as illustrated again below, has had a fairly stable historical ratio vs the Case Shiller 10-city index. An HPHF Ratio Agreement could be structured with boundaries that have not be violated in years.
What of the other indices?
The graph to the right shows the ratio of all twenty regions (including one for the CS 20-city index) versus the Case Shiller 10-city index. The graph may seem like spaghetti thrown against wall, but it shows that some areas saw higher price gains than the 10-city index (hence rising ratio values) while falling ratios are consistent with smaller gains.
The legend for each label also shows the percent each ratio changed over the last three years. While index values were up everywhere, Phoenix (PHX) was the biggest over-achiever, while Chicago (CHI) had the slowest gains.
Notably, 15 of the 20 regional indices outperformed the 10-city index, but two of the laggards (CHI and NYM) constitute ~>37% of the 10-city index. That regional prices outpaced the 10-city index is seen in performance of the 20-city ratio improving by 1.18%. Recall that the "other' ten cities only comprise ~26% of the 20-city index, so that the outperformance of the "other ten" components, was a lot more, as evidenced by the gains in Phoenix ratio (POX) +12.82%, Seattle (SEX) +6.64%, and Tampa (TPX) 5.27%. Should COVID incent more moves from the big cities to second-tier cities (e.g. Austin, Nashville, Salt Lake), I'd expect these ratios to have even more pronounced trends.
Net, trends in ratios can be used as another tool to identify out-performers (All areas "won").
Quoting ratio agreements for regions with higher or lower gains in home prices relative to the 10-city index, can still be done. However, the boundaries need to be shifted up (or down) to create a balanced tradeoff between forward ratios.
In the example below, I calculated the ratio of BOS/10-city index futures for the Feb '22 (G22) and Feb '23 (G23) expirations, and used those as bid and offered levels on ratio agreements. The historical BOS ratio has been steadily rising since Feb '15 and futures prices are at levels that are consistent with further gains in BOS relative to 10-city index futures (i.e. HCI,CUS). (The same would be true for CHI ratios and comparisons of forward ratios, but with the need to lower boundaries).
As such, while the BOS ratio spent many years below 0.95, a more neutral view (based on forward ratios) for an HPHF BOS Ratio Agreement might have boundaries for G22 and G23 of 0.95-1.05. Given an upward trend in the ratio, boundary conditions on longer-dated expiration could be widened by raising the cap.
As noted in the last blog, the boundaries here are +/- 5% so the out-of-pocket cost (premium) would probably be 5.5-6.0%.
There would be no need to fear possible increase in margins (on these ratio agreements) as the payouts are capped at the 0.95 and 1.05 boundary conditions.
Finally, unlike a CME intercity spread trade, the same (to be negotiated) notional amount would apply to both indices. (By contrast CME intercity spread trades are for one lot versus one lot. As such, a CHI/10-city intercity trade would have one contract worth ~$40,000 (CHI @ index of 160 *$250/point) hedging another index worth $67,000 (HCI @268 * $250/point). In the past, I've done such trades to include "extra" CHI contracts to even out exposures. However, given the ratios, the notional values to have equal exposure might not equal the amount someone wants to hedge.
With ratio agreements, if someone wants $243,000 notional exposure, that amount (ro any other) can be written into the agreement. (That said, if someone wants to add 10-city futures exposure to have more of an outright view, recall that each lot has a notional value of ~$67,000.
While I'm open to entering a few ratio agreements on modest notional amounts to get things started, in the long run I'd like to build a book of inquiries (across ~50 cities) where users might want to take long/short exposure to the relative performance of one region vs. another, while steering users to take their absolute price risk in the same-expiration 10-city index futures.
Future blogs will address other cities (referencing Freddie Mac home price indices) and will review the supporting assumption of how correlated (or not) the index values are for larger cities).
Feel free to contact me if you have any questions about this blog, have any cities that you might have an interest in expressing a view vis a Ratio Agreement, or if you just want to learn more about hedging strategies involving home price index derivatives.
Thanks,
John