Publications
Forthcoming
Authors: William N. Goetzmann, Liang Peng, and Jacqueline Yen
This paper argues that econometric analysis of housing price indexes before 2006 generated forecasts of future long-term price growth and low estimated probabilities of extreme price decreases. These forecasts of future increases in home-loan collateral values may have affected both the demand and the supply of mortgages. Standard time series models using repeat-sales indexes suggest that positive trends had a long-half-life. Expectations based on such models support expectations that could lead to an asset bubble.
Analysis of data from the HMDA loan database and LoanPerformance.com at the MSA level and at the loan level substantiates the effects of past price trends on the demand and supply of subprime mortgages. On the demand side, at the MSA level, past home price increases are associated with more subprime applications, higher loan to income ratios and lower loan to value ratios of applications for both prime and subprime mortgages. This is consistent with the notion that households not only borrowed more but also invested more in home equity conditional on greater past house price increases. On the supply side, past home price appreciation had a significantly greater impact on the approval rate of subprime applications than the approval rate of prime applications. Loan level analysis indicates that past home price appreciation increased the approval rate of subprime applications but did not affect the approval rate of prime applications. Further, approved HMDA subprime loans had higher loan to income ratios in MSAs with greater past house price trends.
Read/Download
Working Paper
August 2011
Author: Liang Peng and Thomas G. Thibodeau
This paper empirically examines the segmentation of house price risk across 99 zip-code delineated neighborhoods in metropolitan Denver. The house price risk in each neighborhood is measured with the temporal variation of quarterly appreciation rates of the neighborhood house price index over the 2002 to 2007 period. Cross sectional regressions of neighborhood house price risk on the median household income and the percentage of population in poverty from the 2000 census data for the same neighborhood provide strong evidence that the house price risk is significantly higher in low-income/poor neighborhoods. Sub-period analyses further indicate that the risk segmentation exists in both a booming period (pre 2005:2) and a busting period (post 2005:3). The results indicate that homeownership can be a much riskier investment for low-income/poor households.
Read/Download
Working Paper
August 2011
Authors: Liang Peng and Thomas G. Thibodeau
This paper empirically analyzes the non-monotonic influence that interest rate changes have on irreversible investment in income producing properties. Using the complete history of quarterly capital improvements for 1,416 commercial properties over the 1978 to 2009 period, we find strong evidence of the non-monotonic effect for apartment, office, and retail properties, but not for industrial properties. For the first three property types, a decrease in the Treasury yield dramatically increases capital improvements when property values are high, but has a weak or negative effect when property values are low. This result has important implications for monetary and fiscal policies.
Read/Download
Forthcoming
Authors: Liang Peng and Rainer Schulz
This paper examines the dynamics of the covariance matrix of return rates for securitized real estate, other company stocks, and government bonds for a cross- section of eight countries. In-sample analysis establishes that in all countries the covariance matrix is time-varying and reacts stronger to bad than to good news. Using a realistic out-of-sample exercise, we find that portfolios selected with a forecasted dynamic covariance matrix are less risky than portfolios constructed with the static matrix. However, benefits of using the dynamic covariance matrix for active portfolio management are mostly offset by rebalancing cost. Passive buy-and-hold investors benefit, because the forecasted dynamic covariance matrix provides better risk assessment.
Forthcoming
Authors: Marcel Arsenault, Jim Clayton, and Liang Peng
This paper provides strong evidence for a positive feedback loop between property prices and mortgage supply, using data from the U.S. commercial property and mortgage markets over the 1991 to 2011 period. The empirical analyses control for the endogeneity of property prices, mortgage flows, mortgage interest rates, and loan to value ratios, and provide two main findings. First, exogenous increases in mortgage supply, measured with the growth of the CMBS market, significantly reduce property cap rates. Second, volatility of past price changes and the "biggest loss" in property values in the past significantly affect mortgage supply. This positive feedback loop may be an important driving force for real estate cycles.
Read/Download
Publication
August 2011
Authors: Norman Miller, Liang Peng, and Michael Sklarz
Using quarterly data for all 379 metropolitan statistic areas (MSAs) in the U.S. from 1980:1 to 2008:2, this paper empirically studies the effect of house prices on local Gross Metropolitan Product (GMP). We compare the effects of predictable and unpredictable house price changes, which we use to capture the collateral and wealth effects of house prices respectively. We further analyze the relationship between the effects and household borrowing constraints, as well as the temporal pattern of the effects. Our analysis provides the following findings. First, house price changes have significant effects on GMP growth, and the effect of predictable changes (the collateral effect) is about three times stronger than the effect of unpredictable changes (the wealth effect). Second, the persistent component of predictable changes has a stronger collateral effect than the novel component. Third, when households are more financially constrained, the collateral effect is stronger, the wealth effect is weaker, and the total effect remains unchanged. Finally, the effects last for eight quarters, and peak on the fourth quarter after house price changes take place.
Read/Download
Publication
May 2011
Authors: Norman G. Miller, Liang Peng, and Michael A. Sklarz
If realized house prices have the wealth effect and the collateral effect on the economy, anticipated house price changes should have similar economic effects. This paper empirically analyzes the effects of single family home sales, which are shown to be able to predict house prices in the literature, on economic production, using 372 Metropolitan Statistic Areas in the U.S. from the first quarter of 1981 to the second quarter of 2008 in a panel Vector Error Correction Model. Changes in home sales are found to Granger cause the growth rate of per capita Gross Metropolitan Product, and the dynamic effects are visualized with impulse response functions. Supporting evidence for the economic impact of home sales is also found in contemporaneous regressions.
Read/Download
Working Paper
August 2010
Author: Liang Peng
This paper analyzes the risk and returns of direct commercial real estate investments at the property level. A novel regression is developed to use property level cash flows instead of index returns to estimate the sensitivity of real estate returns to economic variables. Monte Carlo simulations suggest that this regression is more accurate than the conventional index approach. Applying this regression to 3,125 commercial properties held between 1978 and 2009, this paper finds that commercial real estate risk premium is positively related to GDP growth and the change in the credit spread, and negatively related to inflation, the stock market risk premium, and the change in the term spread. The sensitivities vary across property types and time. This paper also finds that the risk characteristics of commercial real estate, such as loadings on Fama French factors, vary across property types and time.
Read/Download
Forthcoming
Author: Liang Peng
This paper proposes a generalized repeat sales regression (GRSR) that uses repeat sales from the entire market, in which properties may have heterogeneous value appreciation processes, to estimate price indices for not only the entire market, but also submarkets or customized portfolios of properties that only have small numbers of value observations. Monte Carlo simulations provide strong evidence that the GRSR indices more accurately measure the index for the entire market as well as individual property value appreciation than conventional RSR indices. This paper also proposes a Chi-square test to detect the heterogeneity in property value appreciation across submarkets/portfolios, and use simulations to show that the test is powerful in small samples. This paper finally illustrates the application of the GRSR using a historical dataset of the Chicago housing market from 1970 to 1986.
Read/Download
Forthcoming
Authors: Liang Peng and Thomas G. Thibodeau
Municipal governments in China established direct control of the supply of urban land in August 2004. This paper examines whether this government action mitigates the efficiency of the residential land market. Using a unique data set of detailed land and residential community transactions with manually collected location information for residential land lots in seven Chinese cities, this paper analyzes the relationship between the land lease prices and residential property prices from the first quarter of 2001 to the fourth quarter of 2007. Results indicate that property prices determined land prices both before and after 2004:3, but the effect was significantly weaker after 2004:3. This is consistent with the hypothesis that the market for residential land became less efficient after municipal governments gained direct control of the land supply
Read/Download
Publication
January 2010
Authors: Jim Clayton, Norman Miller, and Liang Peng
Housing market cycles are featured by a positive correlation of prices and trading volume, which is conventionally attributed to a causal relationship between prices and volume. This paper analyzes the housing markets in 114 metropolitan statistical areas in the United States from 1990 to 2002, treats both prices and volume as endogenous variables, and studies whether and how exogenous shocks cause comovements of prices and volume. At quarterly frequency, we find that, first, both home prices and trading volume are affected by conditions in labor markets, the mortgage market, and the stock market, and the effects differ between markets with low and high supply elasticity. Second, home prices Granger cause trading volume, but the effects are asymmetric—decreases in prices reduce trading volume, and increases in prices have no effect. Third, trading volume also Granger causes home prices, but only in markets with inelastic supply. Finally, we find a statistically significant positive price-volume correlation; which, however, is mainly explained by co-movements of prices and volume caused by exogenous shocks, instead of the Granger causality between prices and volume.
Read/Download
Publication
March 2009
Authors: Gina Nicolosi, Liang Peng, and Ning Zhu
After analyzing retail investors' stock trades for potential learning behavior, we present evidence that individual investors learn from their trading experience. Initially, we question whether investors' previous forecasting ability (inferred from prior purchases' subsequent risk-adjusted performance) affects their future trade profitability and activity. Indeed, as an investor's inferred ability increases, so does her ensuing trade profitability and intensity. Further, because additional investment experience allows more accurate ability inference, we posit that trading experience should help investors obtain better investment performance. Consistent with this hypothesis, not only do excess portfolio returns improve with account tenure, but we also find that trade quality (i.e., average raw and excess buy-minus-sell returns) significantly increases with experience (i.e., calendar time and account tenure). In sum, individual stock investors do learn, and they consequently adjust their behavior and thus effectively improve their investment performance.
Publication
June 2008
Authors: Jim Clayton, Greg MacKinnon, and Liang Peng
This paper proposes and empirically evaluates competing models to explain the time variation in private real estate market liquidity documented in Fisher et al. (2003). We test three classes of models. In the first, that seller estimates of property value lag market conditions because of an asymmetric information problem. Sellers, at least in part, base their estimates of value on observations of signals from the market, but the presence of noise means a change in signal is not fully reflected in sellers' updated value estimates. The second class of models incorporates the value of waiting or opportunity cost of not transacting, recently introduced by Krainer (2001) and Nov-Marx (2004), into seller's optimal valuation strategy. In the third, we allow for the possibility of noise traders, or investors who are not fully rational in the sense that they trade on market sentiment. We follow Baker and Stein (2003) and consider a formal model that links stock market-wide liquidity to investor sentiment with higher liquidity being due to the presence of irrationally over-optimistic traders. In this model measures of aggregate liquidity act as an indicator of the relative presence (or absence) of sentiment-based traders in the market place and therefore the divergence of asset price from fundamental value. Empirical findings are generally consistent with models of optimal valuation with rational updating and provide support for the opportunity cost.
Read/Download
Publication
June 2006
Authors: Norman Miller and Liang Peng
This paper uses GARCH models and a panel VAR model to analyze possible time variation of the volatility of single-family home value appreciation and the interactions between the volatility and the economy, using a large quarterly data set that covers 277 MSAs in the U.S. from 1990:1 to 2002:2. We find evidence of time varying volatility in about 17% of the MSAs. Using volatility series estimated with GARCH models, we find that the volatility is Granger-caused by the home appreciation rate and GMP growth rate. On the other hand, the volatility Granger causes the personal income growth rate but the impact is not economically significant.
Publication
March 2006
Authors: William Goetzmann and Liang Peng
We analyze a bias in transaction-based price indices due to the presence of seller reservation prices. We develop a model in which the ratio of seller's reservation prices to the market value affects trading volume and biases of observed transaction prices: when trading volume decreases (increases), index returns are estimated with an upward (downward) bias. We propose a new econometric procedure to mitigate the bias, and use simulations to demonstrate the effectiveness of the procedure. We construct a reserve-conditional unbiased index for the Los Angeles housing market, which substantially differs from a traditional repeat sale index.
Read/Download
