Hedging the Value of a Home with S&P/Case-Shiller Futures

I’m currently looking for a rental house in Fairfield County, CT, but have been totally disgusted with the rents… they are high enough that it currently makes more sense to buy (in terms of a basic time-value of money calculation)… if you assume that you’re going to stay there for a few years and housing prices won’t depreciate further in the meantime.

Unfortunately, I’m not comfortable assuming that house prices will not drop further, so I looked into the idea of hedging my expected exposure to the New York housing market with a short position in futures on the S&P/Case Shiller New York Real Estate Index.

These Futures contracts trade on the CME with a notional value of $250 per point. With the index currently trading at around 170, each contract is enough to hedge $42,500 worth of housing in Fairfield County (which is included in the calculations of the New York index.) So a $425,000 generic house in the New York area could be hedged with 10 contracts. (In fact, you’d probably need 20-30% more than that, because profits and losses on the futures would be taxable, while capital gains and losses on the house would not be, at least up to a limit of a $250,000 gain if you had lived in it 2 out of the last 5 years.

The problem lies in market liquidity. It currently looks like the longest dated futures are trading at 170, which means that they are predicting the market will not rise or depreciate significantly over the next four years (although shorter-term contracts are predicting a short-term decline followed by a recovery. But to hedge that $425,000 house if you’re subject to a 30% tax rate, you would need about 13 contracts. As I look at the quotes today, the current bids for even the closest November 2010 contracts are for only 3 contracts, and the bid-ask spread is 10 points, or $2500 per contract.

There is no bid for the November 2014 contracts that I’d be most interested in, meaning that I can be confident that my order to sell would move the market significantly. Assuming unrealisticly optimisticly that an ask for 13 November 2014 contracts moved the market only as much as the spread on the closest-in contract, that means a transaction cost of 10 x $250 x 13 = $32,500, which is about a full year’s rent in a comparable house.

I’m going to have to file this one under "it seemed like a good idea at the time."


Comments are closed.

%d bloggers like this: