An Introduction to RPX INTRODUCTION Radar Logic is a real estate information company based in New York. We convert public residential closing data into information about the state and prospects for the US housing market. Unlike many analytic and information providers, we focus on the state of housing at the point of sale giving us unique and early insight into the housing economy in the US. Early on we discovered a bifurcation in the market of homes for sale. Homeowners in distress, and the lenders who held their mortgages, began to be a driving force in housing valuation. This motivated inventory led the price points down between 35 and 50 percent in most major housing markets, a dynamic which, thereafter, led to early signs of stability in the spring of 2009. As we look for stability to convert into a recovery, two issues remain which have a potentially dampening effect on housing values. First is the potential for many more distressed sales as a result of the current and growing level of foreclosures. Second is the potential for capital constraint as lenders struggle to carry large amounts of delinquent or defaulted mortgages. These two factors are closely related. August, 2011
Introduction Originally conceived and designed to be used as an underlying valuation for property derivatives, RPX prices reflect a true and dynamic spot market valuation metric for housing. Unlike repeat sales analyses, RPX is based on a proprietary probability density function chosen for accurate representation of the empirical spectrum of price per square foot in closed home sales. The result is a high frequency view of what prices are actually being achieved in the market. Historically, housing was a steady, low growth asset. The boom and bust cycle of the past four years has introduced a brand new level of risk and dynamics to housing. Inventory has become the key element in value trend. Investors can no longer rely solely on traditional valuation methods to analyze risk exposure. With distressed sales making up about 40% of the market, on average, traditional repeat sales methodologies arguably compare apples to oranges. To accurately forecast home price appreciation, and estimate current loan to value ratios, modelers must look at the distressed and non-destressed market segments separately. RPX data let you do this. As home prices continue to decline, loan to value ratios are moving fast. Timely analysis of the impact of collateral on investments is critical. Without it, investors do not really know what their underlying risk is. 2
Using RPX Data Radar Logic data are available in a number of formats for a wide range of coverage areas: We capture data on roughly 2,000 counties which can be mapped however you need it. We offer snap shot and time series data packages that can be custom tailored to your specific needs. We offer a number of standard outputs. 3
Data Characteristics As our data are based on public records, we are highly confident of their accuracy. By using public records we capture real points of value everyday. RPX is based on what buyers and sellers actually do. We do not include data for which we do not believe an audit is possible. RPX prices are based on all transactions in a given market and therefore provide the most inclusive view of housing values. Unlike other housing price indicators, we do not systematically exclude any housing transactions. Repeat-sales indices exclude new-home sales, and FHFA data exclude sales of homes that do not conform to federal loan limits. The spot market is a function of all asset types, not just some, and they are interrelated. Inventory matters. And it shifts. RPX prices show market movements earlier than other data sources. Our data is published daily, the most frequent data available. RPX prices are calculated using data from transactions that occur over 1, 7, and 28 days, whereas other home price indices are calculated based on transactions that occur over 30 to 90days. As we use only the most recent data, RPX prices provide the best possible indication of current price trends. RPX is always lagged 63 days. Some other housing value metrics are released more quickly, but these metrics are usually revised making their true release slower than RPX. 6
Data Comparisons Index RPX CSI FHFA Geographic Coverage 25 MSAs 20 MSAs 363 MSAs Publication Frequency Daily Monthly Quarterly (MSAs); Monthly (U.S. and Census Divisions) Publication Lag 63 days 2 months 2 months Data Collection Period 1, 7 and 28 days 1 Quarter 1 Quarter (MSAs); 1 Month (U.S. and Census Divisions) Data History Available from 2000 1987 1975 (MSAs); 1991 (U.S. and Census Divisions) Methodological Approach Probability Density Function Parameterization Weighted-Repeat Sales Weighted-Repeat Sales New Home Sales Included Not Included Not Included Condominiums / Multifamily Included Not Included Not Included Transaction Weighting None Adjustments for changes in property quality between sales. Reduced weights for idiosyncratic transactions. Increased weights for more expensive homes Adjustments for changes in property quality between sales. Prices Included in Index Purchase Price Purchase Price Purchase-Only Index: Purchase Price; All-Transaction Index: Purchase Price and Refinance Appraisals Data Source Local governments Local governments Fannie Mae and Freddie Mac Tradable in Derivatives Markets Yes Yes No Traded * Beginning Sept 2007 May 2006 N/A Since the CSI is published on a 2-month lag based on housing data collected over 3 months, it reflects property values in transactions that closed 60 to 150 days prior to the publication date. By way of comparison, the 28-day RPX reflects property values in transactions that occurred 63 to 91 days prior to the publication date. In light of this difference, we classify the RPX as a "spot" price to highlight its focus on recent data and the CSI as a "trend" index to highlight its longer historical perspective. We have been accurately forecasting the CSI in our RPX Monthly Housing Market Report for more than a year. 7
Data Comparisons: RPX is Most Timely View 190 Directional changes in home price trends appear in the the RPX TM two to six months before they are visible in the S&P/Case-Shiller index. RPX Composite S&P/Case-ShilIer 20-City Composite 185 180 Median Home Price ($000) 175 170 165 160 155 150 145 140 Stability returns to home prices RPX CSI Seasonal weakness sets in RPX CSI End of winter declines RPX CSI Spring strength wanes as tax credit expires Winter weakness recedes Jul-08 Sep-08 Nov-08 Jan-09 Mar-09 May-09 Jul-09 Sep-09 Nov-09 Jan-10 Mar-10 May-10 Jul-10 Sep-10 Nov-10 Jan-11 Mar-11 May-11 RPX CSI RPX CSI Transaction Date 8
Housing as We See It 2000-2011 $300 RPX 25-MSA Composite Transaction Count* RPX 25-MSA Composite Price 210,000 $250 175,000 $200 140,000 $150 105,000 $100 70,000 $50 35,000 $0 Jan-00 Jan-01 Jan-02 Jan-03 Jan-04 Jan-05 Jan-06 Jan-07 Jan-08 Jan-09 Jan-10 Jan-11 - * Prior to 1/1/06 all transactions are captured. Thereafter, counts reflect a 63 Day cut off 9
$280 RPX Composite Prices, January 2008 May 2011 $260 RPX Motivated Price $240 $220 RPX Other Price RPX MSA Price Price Per Square Foot $200 $180 $160 40% Discount $140 $120 $100 Jan-08 Apr-08 Jul-08 Oct-08 Jan-09 Apr-09 Jul-09 Oct-09 Jan-10 Apr-10 Jul-10 Oct-10 Jan-11 Apr-11 Transaction Date 10
RPX 25-Metro-Area Transaction Counts, January 2008 May 2011* 100,000 Motivated Transaction Count Other Transaction Count 90,000 80,000 70,000 Transactions 60,000000 50,000 40,000 30,000 20,000 10,000 0 Jan-08 Apr-08 Jul-08 Oct-08 Jan-09 Apr-09 Jul-09 Oct-09 Jan-10 Apr-10 Jul-10 Oct-10 Jan-11 Apr-11 11 Transaction Date *Transaction counts reflect the number of transactions included in the calculation of the 28-Day RPX Composite price and, while correlated with actual sales volumes, may differ from such volumes.
Summary Radar Logic Data are uniquely presented to allow users to develop timely and insightful analyses of their real estate tt related ltdinvestments. t Our data can be configured to match the specific market segments defined by geography, home size or home value included in users proprietary portfolios. Custom data configurations can be produced allowing users to have both trend line and moving value model inputs on specific holdings. Custom financial instrument engineering is possible, and Radar Logic is prepared to support the efforts with qualified parties. 12
For More Information and Pricing Contact: Radar Logic Incorporated 180 Varick Street, Suite 502 New York, New York 10014 Phone: (212) 965-0300 www.radarlogic.com info@radarlogic.com Joseph W. Gonsalves II National Sales Director Direct: (212) 965-9981 Email: jgonsalves@radarlogic.com Quinn W. Eddins Director of Research Direct: (212) 965-9982 Email: qeddins@radarlogic.comcom 13