Hedging House Price Risk with CME Futures Contracts: The Case of Las Vegas Residential Real Estate

1
Hedging House Price Risk with CME Futures
ContractsThe Case of Las Vegas Residential Real
Estate

• Mark Bertus, Harris Hollans, and Steve Swidler
• Department of Finance, College of Business,
  Auburn University, Auburn AL 36849 USA

By Latoya Owens
2
CME

• Chicago Mercantile Exchange (CME) trades several
  types of financial instruments interest rates,
  equities, currencies, and commodities.
• It also offers trading in alternative investments
  such as weather and real estate derivatives.
• It has the largest options and futures contracts
  open interest (number of contracts outstanding)
  of any futures exchange in the world.
• CME made it possible to hedge house price risk
  directly through futures contracts

3
Futures Contracts

• In finance, a futures contract is a standardized
  contract, to buy or sell a specified commodity of
  standardized quality at a certain date in the
  future, at a market determined price (the futures
  price)
• The price is determined by the instantaneous
  equilibrium between the forces of supply and
  demand among competing buy and sell orders on the
  exchange at the time of the purchase or sale of
  the contract.
• To exit the commitment prior to the settlement
  date, the holder of a futures position has to
  offset his/her position by either selling a long
  position or buying back (covering) a short
  position, effectively closing out the futures
  position and its contract obligations.

4
Futures contracts vs. Options Contract

• A futures contract gives the holder the
  obligation to make or take delivery under the
  terms of the contract,
• An option grants the buyer the right, but not the
  obligation, to establish a position previously
  held by the seller of the option.
• The owner of an options contract may exercise the
  contract, but both parties of a "futures
  contract" must fulfill the contract on the
  settlement date. The seller delivers the
  underlying asset to the buyer, or, if it is a
  cash-settled futures contract, then cash is
  transferred from the futures trader who sustained
  a loss to the one who made a profit.

5
Introduction

• Ownership of residential real estate creates risk
  to investor due to price.
• There are currently ten city futures contracts on
  the CME.
• Each city is based on a SP/Case Shiller Metro
  area Home Price Index.
• These cities include Boston, Chicago, Denver,
  Las Vegas, Los Angeles, Miami, New York, San
  Diego, San Francisco, and Washington

6
Purpose

• The purpose of this paper is to look at the Las
  Vegas market and examine whether metropolitan
  house price risk can effectively be hedged with
  the corresponding CME futures contract.
• The Las Vegas market was selected because it is a
  growing, dynamic area and during the period of
  study expanded at one of the fastest rates in the
  country.
• This growth rate implies that there was a
  substantial mix of both new and existing home
  sales in the area.
• Las Vegas experienced quarterly price volatility
  between -.2 and 17.5.

7
Investor Classes

• For purposes of this analysis, hedging was
  considered from the following viewpoints
• Investment groups hold equity stakes in
  property for income and appreciation generated
  returns
• Mortgage portfolio investors face risks
  associated with falling home prices and
  increasing mortgage default rates
• Local real estate developer/broker could be
  exposed to potential losses due to price drops
  during various stages of construction while
  houses are still on their books
• Individual homeowners concerned about the
  amount of equity in the home

8
Data

• Data was obtained from Clark County, Nevada tax
  records (region shaping the metropolitan area of
  Las Vegas)
• Analysis looks at house prices from 1994 mid
  2006
• Six different tax districts were examined
• Las Vegas City (district 200), North Las Vegas
  (district 250), Sunrise Manor (district 340),
  Spring Valley (district 417), Paradise (district
  470), and Henderson (district 505)
• For the analysis four housing attributes were
  used price, square footage, bedrooms, bathrooms
• To account for outliers, houses outside two
  standard deviations of the mean were excluded

9
Characteristics of Median House

• Las Vegas Sample, by Sales Year

10
Characteristics of Median House

• During 1994 1999, Sales Year and Year Built
  Equal
• New homes made up more than ½ of transactions
• New house pace continued strong but the number of
  transactions began to decline in the later years
• By 2006, median purchase was for a house built in
  1999
• During the beginning of the sample houses
  appreciated 2 3 annually and jumped
  significantly to 36.7 in 2004.
• In the following years it began to moderate
  around 19 and fell significantly during 2006 at
  1.4
• From 1994 2003, square footage increased for
  the most part and began declining in 2004, lot
  sizes decreased
• For the analysis period, the median home had
  3bd/2ba

11
Characteristics of Six Tax Districts

• In Las Vegas

12
Characteristics of Six Tax Districts

• Las Vegas largest city in Clark County (roughly
  600,000 residences)
• Henderson next largest (home of Nellis Air
  Force Base)
• Three unincorporated areas
• Paradise (includes significant portion of the
  Strip)
• Spring Valley
• Sunrise Manor
• During 2000 -2006 Paradise grew 2 (almost built
  out), population in North Las Vegas grew 75,
  Henderson grew 46)
• Population density Low end North Las Vegas
  (2580 people/sq. mile) High end Spring Valley
  (5257 people/sq. mile)

13
Methodology

• The analysis considers the problem of hedging the
  representative home in Clark County using
  SP/Case-Shiller Las Vegas Real Estate Index
• While some investors may own property covering
  the entire metropolitan area, others may limit
  their investments to a particular region.
• They examine hedging effectiveness for each of
  six tax districts in Clark County as well as the
  metropolitan area as a whole.

14
Methodology

• To test for hedging effectiveness, they follow
  the methodology in Kolb and Overdahl (2003) and
  estimate the following regression for a given
  region
• Ik,t ak ßk Ft ek,t (1)
• where, Ik,t the percentage change in the median
  home price in region k in quarter t,
• Ft the return on the CME Las Vegas Real
  Estate futures contract in quarter t,
• ak the constant regression parameter,
• ßk the slope coefficient for the risk
  minimizing hedge for region k using the CME Las
  Vegas Real Estate futures contract, and
• ek,t an error term with 0 mean.

15
Interpreting the Equation

• CME real estate contracts only began trading in
  2006, therefore they proxy Ft, the return to the
  futures contract, by calculating the quarterly
  percentage change in the SP/Case-Shiller Las
  Vegas Real Estate Index.
• Given convergence between spot and futures
  prices, this provides a reasonable futures return
  estimate and is consistent with the methodology
• ßk is the risk minimizing hedge ratio and
  exhibits the relationship between spot prices
  (the median home) and futures prices.
• As part of the analysis, they examine hedging
  performance when investors enter into a naïve
  hedge. In that case, ßk equals 1 and the investor
  takes an equal but opposite position to their
  stake in the cash market.

16
Interpreting the Equation

• A regressions R2 represents the percentage of
  the dependent variables variance explained by
  the independent variables variance.
• Within the context of equation (1), the
  regressions coefficient of determination can be
  interpreted as the percentage of house price risk
  eliminated by taking an opposite position in the
  real estate futures contract.
• Regressions R2 is interpreted as the measure of
  how effective futures contracts are in hedging
  house price volatility.

17
Results

• Hedge Ratio vs. Las Vegas Index represents the
  minimum variance hedge ratio for each of the
  sample tax districts
• Hedge ratios range from low of .7065 in district
  340 to high of .9291 in district 470
• All districts have beta lt 1, therefore entire
  sample is lt 1
• For districts 340 and 417 and the entire sample,
  95 confidence intervals imply minimum variance
  hedge is below 1.
• For hedging effectiveness, futures contracts
  reduce house price volatility a little more than
  50 in district 340 and as much as 84 in
  district 200.
• For all homes hedging effectiveness is nearly 89

18
Results

• Hedge Ratio vs. Las Vegas Index (LVXR) All
  Homes

19
Results

• Results imply for a representative home, prices
  advance at a slower rate than SP/Case-Shiller
  Las Vegas Real Estate Index.
• Real estate and mortgage investors wishing to
  hedge need to sell futures contracts with a
  notional value below the value of porfolio held.
• If they enter naïve hedge they will have sold too
  many futures contracts and hedging effectiveness
  will be less than the reported R2 values.
• Under best circumstances investors holding
  portfolio of homes located in Sunrise Manor
  (district 340) can only reduce half their price
  volatility
• Its most difficult to form an effective hedge in
  the three unincorporated areas

20
Results

• New homes also have relatively low correlation
  with the SP/Case-Shiller Las Vegas Real Estate
  Index. Hedging effectiveness as measured by R2 is
  uniformly lower in all areas compared to the
  results in the previous table. In the case of
  unincorporated Paradise, CME futures contracts
  reduce volatility only 4.
• Out of 5 of the six districts the highest
  reduction of price risk is in North Las Vegas.
  Once again, for the complete sample (All),
  hedging effectiveness is greater than in any one
  district nevertheless, futures hedging only
  reduces 60 of new house price risk.
• Not only is minimum variance hedging
  effectiveness low, but if builders enter into a
  naïve hedge instead (beta equal 1), there may be
  virtually no reduction in volatility. In fact, a
  naïve hedge runs the risk that it may actually
  increase overall price volatility.

21
Minimum Variance Hedge Ratio (beta)
22
Effectiveness of Naïve hedge
23
Hedging Effectiveness New Homes
24
Hedging Effectiveness Existing Homes
25
Conclusion

• CME makes it possible to hedge house price risk
  in ten metropolitan areas
• The results over the analysis presented are
  decidedly mixed
• For investment groups/mortgage holders with real
  estate assets spread across the Las Vegas metro
  area, hedging with CME futures would have reduced
  house price risk by more than 88 over sample
  period.
• If real estate were localized in one of the six
  tax districts house price volatility would have
  been lower

26
Conclusion

• For builders/developers of new homes in Las
  Vegas, it may be very difficult to manage house
  price risk with CME futures
• Due to repeat sales there is little correlation
  between house appreciation and the index
• Individual homeowners and others trying to hedge
  price appreciation of existing homes are likely
  to be most successful
• The empirical evidence finds that hedge ratios as
  well as hedging effectiveness have not been
  stable over time. Over certain periods, an
  investor that entered into a naïve hedge with a
  beta of 1 would have found that the hedged
  portfolio had greater volatility than the real
  estate assets themselves.

27
Conclusion

• Instability of hedge ratios also suggests that
  out of sample results might differ significantly
  from in sample findings.
• In the future one may consider how effective real
  estate hedging might be if minimum variance hedge
  ratios are continually changing.
• A second area of future research might consider
  longer hedging horizons. For many investors,
  price risk issues are longer term than one
  quarter.
• Any investor wanting to insure against price risk
  for long periods of time, might enter into a
  rollover strategy.
• For others wishing to hedge long term,
  idiosyncratic movements in their portfolio might
  wash out in time.