Defining Daily Value in a Non-Commodity Market

Transcription

1 Defining Daily Value in a Non-Commodity Market Enabling Property Derivatives: The Radar Logic Approach Presented by Michael A. Feder, President and CEO at Real Estate Derivatives World 2007, New York, NY

2 Real Estate Is One of The World s Largest Asset Classes But You Can t Trade It U.S residential housing assets exceed $20 trillion There is no spot market, nor a spot price There is no ticker to watch, nor a stream of data on your Bloomberg To date there have been surrogates, but they are several steps removed: REITs: represent equity in trusts invested in real estate; shares trade like equities ETFs: the Dow Jones U.S. Real Estate Index Fund is an ETF that invests wholly in REITs Homebuilder index: diluted by corporate issues not necessarily reflective of the real estate itself Government data: OFHEO includes only conforming product ($417,000 limit)

3 There Is No Good Correlation The influence of factors other than pure real estate prices causes the obvious lack of direct correlation among these vehicles DJ Home Builder DJ REIT OFHEO DJ U.S RE Index S&P 500 1/2/01 3/2/01 5/2/01 7/2/01 9/2/01 11/2/01 1/2/02 3/2/02 5/2/02 7/2/02 9/2/02 11/2/02 1/2/03 3/2/03 5/2/03 7/2/03 9/2/03 11/2/03 1/2/04 3/2/04 5/2/04 7/2/04 9/2/04 11/2/04 1/2/05 3/2/05 5/2/05 7/2/05 9/2/05 11/2/05 1/2/06 3/2/06 5/2/06 7/2/06 9/2/06 11/2/06 1/2/07 3/2/07

4 However, Real Estate Is a Market U.S. residential housing has a greater valuation than either the U.S. taxable bond markets or the U.S. equity market 124 million units in U.S. housing inventory 7.5 million residential transactions in 2006 = $1.8 trillion value Market Comparison Trillions $25 $20 Value of Assets $15 $10 $5 $- U.S. Housing Market 2006 Value of the U.S. Taxable Bond Market 2006 Value of the U.S. Equity Market Radar Logic's 25 Initial MSAs Total U.S. Commercial Real Estate Market * * $1.3 trillion of U.S. commercial real estate market assets are residential.

5 Radar Logic s 25 Initial MSAs No Texas due to insufficient data Radar Logic s 25 initial MSAs represent over $10 trillion in assets; we will produce a composite index as well.

6 How Radar Logic Looks at the Market: Calculating Intelligible Values from Actual Data Radar Logic calculates intelligible values from the actual data We combine title transfer prices with available data on square footage The public source data comes from municipal offices Unlike other property derivative entrants, we include as many transactions as we can (e.g., condos, new homes, foreclosure sales, etc.) By using maximum data with the power of advanced statistical analysis: We are able to be more robust than other indices We illuminate market trends and volatility, and Our daily prices are highly representative of the actual property markets This sophisticated analysis was accomplished through the development of proprietary algorithms by our affiliate, Ventana Systems, one of the leading mathematical consulting firms in the world

7 How Radar Logic Looks at the Prices We considered using simple statistics like average or median Our analysis showed that there is a pattern in the data that holds across different times and different MSAs The pattern segments the ppsf spectrum into three ranges which roughly correspond to the low, mainstream, and high ends of the market We apply a sophisticated statistical technique and generate a more robust output We use a year s worth of transaction data to understand the pattern of the underlying housing price in the MSA We use a given day s transactions to identify the daily price within that landscape In mathematical terms, we generate a probability distribution function with parameters derived from a year of data, and then calculate the daily prices using maximum likelihood We do not eliminate any trade that we are confident is legitimate

8 Chaotic Data Distribution Sample Distribution of Transaction Data in the New York MSA In its raw form, the distribution of transaction observations appears to be chaotic

9 Logical Results: Our Pattern Recognition Techniques Enable Visibility of Actual Market Dynamics Radar Logic Index: New York (7-day) $350 $300 Data are preliminary and presented "AS IS" without warranty of any kind. $250 Radar Logic Index (PPSF) $200 $150 $100 $50 $0 10/31/ /31/ /31/ /31/ /31/ /31/ /31/2006 Transaction Date

10 The Application-centric Approach Application-centric approach: start with what has practical value to end user customers Financial customers: investment alternatives, enhanced project finance Enterprise customers: opportunity and risk management Defining the ideal application What s the right metric? What s the right frequency? Which transactions should be captured? Metric: price per square foot Frequency: daily Transactions captured Every transaction is a trade THE MARKET IS THE MARKET

11 The Application-centric Approach The next three charts show our results for the New York MSA on a 1, 7 and 28 day basis. 1 day: This is the daily price observation exhibited as each day s actual observations, one day at a time. 7 day: Exhibits our output as a daily calculation of the price observed as a result of 7 rolling calendar days worth of observations 28 day: Exhibits our output as a daily calculation of the price observed as a result of 28 rolling calendar days worth of observations

12 New York Daily Price $350 Data are preliminary and presented "AS IS" without warranty of any kind. $300 $250 Radar Logic Index (PPSF) $200 $150 $100 $50 $0 10/31/ /31/ /31/ /31/ /31/ /31/ /31/2006 Transaction Date

13 New York (7-day) $350 $300 Data are preliminary and presented "AS IS" without warranty of any kind. $250 Radar Logic Index (PPSF) $200 $150 $100 $50 $0 10/31/00 10/31/01 10/31/02 10/31/03 10/31/04 10/31/05 10/31/06 Transaction Date

14 New York (28-day) $350 $300 $250 $200 $150 $100 $50 $0 10/31/ /31/2000 2/28/2001 4/30/2001 6/30/2001 8/31/ /31/ /31/2001 2/28/2002 4/30/2002 6/30/2002 8/31/ /31/ /31/2002 2/28/2003 4/30/2003 6/30/2003 8/31/ /31/ /31/2003 2/29/2004 4/30/2004 6/30/2004 8/31/ /31/ /31/2004 2/28/2005 4/30/2005 6/30/2005 8/31/ /31/ /31/2005 2/28/2006 4/30/2006 6/30/2006 8/31/ /31/2006 Data are preliminary and presented "AS IS" without warranty of any kind. Radar Logic Index (PPSF) Transaction Date

15 What the Data Can Tell Us: Seasonality Financial customer Does it matter when your trade settles? Enterprise customer Does it matter when you sell your product? Does it matter how you hedge? If there is no seasonality in a given market, the answer to these questions may be NO However, if there is meaningful seasonality, the answer is clearly YES Let s look at Boston Seasonality is evident in both price observation and volume observation Other vehicles do not exhibit the existence of seasonality

16 Boston (7-day) $300 Data are preliminary and presented "AS IS" without warranty of any kind. $250 Radar Logic Index (PPSF) $200 $150 $100 $50 $0 10/31/00 10/31/01 10/31/02 10/31/03 10/31/04 10/31/05 10/31/06 Transaction Date

17 Boston (7-day): Transaction Volume Number of Transactions /31/00 10/31/01 10/31/02 10/31/03 10/31/04 10/31/05 10/31/06 Transaction Date

18 Boston (7-day) vs. OFHEO Radar Logic Index OFHEO Home Price Index (1) Index: 3/1/2001 = Data are preliminary and presented "AS IS" without warranty of any kind Source: Office of Federal Housing Enterprise Oversight 0.0 3/1/01 3/1/02 3/1/03 3/1/04 3/1/05 3/1/06 Reporting Date

19 What the Data Can Tell Us: Frequency Does the frequency of reporting matter? If markets are stable then NO If markets are rapidly moving, then YES Seeing the market turn Let s look at Phoenix By viewing the data daily, we see market trends sooner and more clearly

20 Phoenix (7-day) vs. OFHEO 2.5 Radar Logic Index OFHEO Home Price Index (1) 2.0 Index: 3/1/2001 = Data are preliminary and presented "AS IS" without warranty of any kind. 1 Source: Office of Federal Housing Enterprise Oversight 0.0 3/1/01 3/1/02 3/1/03 3/1/04 3/1/05 3/1/06 Reporting Date

21 What the Data Can Tell Us: Inclusive Does the inclusion of all valid transactions matter? If markets include only homogenous assets, then NO If markets have a wide variety of housing stock, then YES Let s look at San Diego and New York Both of these markets have experienced substantial non-conforming condominium development As a result, our broad data inclusion shows true price movements in these markets

22 San Diego (7-day) vs. OFHEO 2.5 Radar Logic Index OFHEO Home Price Index (1) 2.0 Index: 3/1/2001 = Data are preliminary and presented "AS IS" without warranty of any kind. 1 Source: Office of Federal Housing Enterprise Oversight 0.0 3/1/01 3/1/02 3/1/03 3/1/04 3/1/05 3/1/06 Reporting Date

23 New York (7-day) vs. OFHEO 2.5 Radar Logic Index OFHEO Home Price Index (1) 2.0 Index: 3/1/2001 = Data are preliminary and presented "AS IS" without warranty of any kind. 1 Source: Office of Federal Housing Enterprise Oversight 0.0 3/1/01 3/1/02 3/1/03 3/1/04 3/1/05 3/1/06 Reporting Date

24 What the Data Can Tell Us: Inclusive Does the inclusion of all valid transactions matter? If markets include only homogenous assets, then NO If markets have a wide variety of housing stock, then YES Let s look at San Francisco There s been a distinct difference in the performance of condos versus non-condos in San Francisco Radar Logic s daily price enables tracking of specific asset performance

25 San Francisco (7-day) $500 $450 Data are preliminary and presented "AS IS" without warranty of any kind. $400 $350 Radar Logic Index (PPSF) $300 $250 $200 $150 $100 $50 $0 10/31/00 10/31/01 10/31/02 10/31/03 10/31/04 10/31/05 10/31/06 Transaction Date

26 San Francisco (7-day) Condo vs. Non-Condo Median [SF Condo, Over 7 Days] RL Index [San Francisco, Over 7 Days] Median [SF NotCondo, Over 7 Days] $700 $600 Data are preliminary and presented "AS IS" without warranty of any kind. $500 $400 PPSF $300 $200 $100 $0 10/31/00 10/31/01 10/31/02 10/31/03 10/31/04 10/31/05 10/31/06 Transaction Date

27 A Unique Perspective Prices + Tools = Utility Radar Logic s comprehensive analytic tools present the ability to segment markets with granularity which has previously been unavailable to investors. Our analytic tools allow segmentation by: Area specificity down to zip code levels Asset filtering by property type, and Segmentation by price range and size Understanding basis risk Our web-based analytics enable users to examine user-defined elements within MSAs

28 A Unique Perspective: Radar Logic Analytics

29 Tribeca (28-day) vs. New York (28-day) RL Index [ New York, Over 28 Days] Median [ Tribeca, Over 28 Days] $1,400 $1,200 Data are preliminary and presented "AS IS" without warranty of any kind. $1,000 $800 PPSF $600 $400 $200 $0 10/31/00 10/31/01 10/31/02 10/31/03 10/31/04 10/31/05 10/31/06 Transaction Date

30 South Beach (7-day) vs. Miami (7-day) RL Index [Miami, Over 7 Days] Median [South Beach, Over 7 Days] $700 $600 Data are preliminary and presented "AS IS" without warranty of any kind. $500 $400 PPSF $300 $200 $100 $0 10/31/00 10/31/01 10/31/02 10/31/03 10/31/04 10/31/05 10/31/06 Transaction Date

31 What Will Make This Market Work Utility/benefit Breadth and depth Opportunity for balance Ultimately, it s about liquidity Natural longs vs. natural shorts The ability to customize products Bankable productivity for all involved

32 Contact Us Radar Logic Incorporated 379 West Broadway, Suite 401 New York, New York Phone: (212)