Créer une présentation
Télécharger la présentation

Télécharger la présentation
## The Demand for Home Equity Loans at Bank X*

- - - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - - -

**The Demand for Home Equity Loans at Bank X***An MBA 555 Project Laura Brown Richard Brown Jason Vanderploeg *bank name withheld for proprietary reasons**Introduction**The current market for home equity loans is highly competitive. Due to the massive housing slowdown, demand for equity transactions has also slowed, forcing companies to re-strategize in a changing environment. We have endeavored to develop a model to better equip Bank X decision makers as they pursue strategies for capturing a larger share of the market within the bank’s national footprint.**Project Objective**Construct a demand model for variables affecting the volume of equity loans (demand variable Q), focusing especially on the effect of the bank prime loan rate (demand variable P)**Hypotheses Tested**• H1 = The demand for home equity loans is explained by interest rates offered by banks (prime rate) • H2 = The demand for home equity loans is explained by consumer purchasing power. • H3 = The demand for home equity loans is explained by public consumer economic indicators (stock market) • H4 = The demand for home equity loans is explained by advertising expenses.**Overview of Methodology**• Stage 1 • Collected monthly data sets (2003 to August 2006) • Created independent & dummy variables to test pattern behavior • Used stepwise regression and practical considerations to eliminate variables • Stage 2 • Used OLS to test the 4 basic assumptions of regression analysis • Generated regression charts • Stage 3 • Generated estimation model • Identified and interpreted elasticities • Summarized final results**Stage 1**• Variables Examined • Volume of Home Equity loans • Bank Loan Prime Rate • Federal Funds Rate • # of Houses Sold • Median Price of Houses Sold • Consumer Loans @ Commercial Banks • Total # of Loan Units • Firm Advertising • Residential Energy Consumption • Transportation Energy Consumption • Money Supply (Stocks)**Stage 1**Definition of Remaining Variables VariableTypeHypothesized Sign Demand for Home Equity Loans Dependent Bank Prime Loan Rate (Proxy) Exogenous Negative Money Supply (Stocks) Exogenous Negative Total Units Endogenous Positive Advertising Endogenous Positive Fed Funds Rate Exogenous Negative Consumer Price Index Exogenous Positive Consumer Loans @ Comm’l Banks Exogenous Negative Median Price of Houses Sold Exogenous Positive**Stage 2**• Assumption of Non-Collinearity • Multicollinearity, as measured by VIF, takes place • when an independent variable correlates with other independent variables. VIF under 10 is preferred, and under 5 is ideal. • Parameter VIFs: • Bank Loan Prime Rate – 2.290 • Money Supply (Stocks) – 2.091 • Total Units – 1.142 • Average VIF for model – 1.841**Stage 2**• Assumption of Absence of Autocorrelation • OLS requires that the residual error terms show no discernible pattern. The assumption is violated when the Durbin-Watson test shows either + or - autocorrelation. • The model revealed evidence of auto correlation. • Rho: Pos & Neg Reject • Rho: Positive Do Not Reject • Rho: Negative Reject • Using First Differences to remove the autocorrelation did not improve the model.**Stage 2**• Assumption of Constant Variance • Constant variance means that all random error terms have the same variance and are not correlated to one another. The null hypothesis of White’s test assumes this homeskedasticity is in place. At 95% confidence, a p-value of 0.05 or higher allows us to accept the null hypothesis. • The p-value for White’s in our model was 0.094**Stage 2**• Assumption of Normality • Normality describes the fact that remaining random error terms exhibit a normal distribution. The chart for residual error terms should produce a line angled at approximately 45 degrees. • The model’s correlation for normality was .992, well above the critical value of .977.**Stage 2**Predictive Ability Chart**Stage 2**Confidence Intervals Chart**Stage 2**Constant Variance Chart**Stage 2**Normal Probability Chart**Stage 2**Error Bars Chart**Stage 3**Estimated Model and Results All data sets were entered into WinORS. After a stepwise regression, Ordinary Least Squares and logarithmic transformation, the following model was constructed for quantity of home equity loans: lnQHE = f(-.0268Ln(P) –.2188Ln(M) + .977Ln(U)) The F-statistic which measure the explanatory power of the model was found to be significant at 371.1. The p-value was 0.00001, showing that the model is statistically significant at a 99.9% confidence level.**Stage 3**Estimated Model and Results, cont. The coefficient of determination R2measures the degree of variation in the dependent variable that can be explained by variation in the independent variables. Our model showed excellent scores: R2 = 96.532% Adjusted R2 = 96.272%**Stage 3**• Elasticities • Elasticities measure the % change in the dependent variable, given a 1% change in the independent variable. • In the multiplicative demand model elasticities are revealed to be constant at all points on the demand curve. • Parameter estimates represent the elasticities of the independent variables. • Absolute values < 1 are inelastic, values >1 are elastic.**Stage 3**• Elasticities, cont. • The model reveals that home equity loans are price inelastic: • Parameter estimate for bank loan prime rate = -0.268 • A 10% increase in the prime rate would result in just 2.7% decrease in the demand for home equity loans. This parameter is inelastic. • The model also reveals that home equity loans are elastic • relative to the availability of other funding for consumer • spending. • Parameter estimate for money supply = -2.188 • A 10% increase in the availability of alternative funding would result in a 21.9% decrease in the demand for home equity loans.**Stage 3**• Conclusions • Demand is not heavily affected by interest rate change, so the bank can take advantage of the inelastic relationship and achieve higher revenues. • H1 is accepted • Alternative funding sources for consumer spending has an inverse relationship to the demand of home equity loans. In a bullish market, consumers don’t borrow against their equity. • H3 is accepted • Consumer purchasing power and firm advertising has little statistical significance in the demand for home equity loans. • H2 and H3 are rejected**Appendix A**Excerpts from Industry Literature Home-Equity Borrowing StallsAs the Housing Market Cools Ruth Simon • The slowdown in home-equity borrowing is leading to weaker sales in some markets for autos, building materials and electronics • As rates go up there is unknown future demand for home equity loans • During the housing boom, demand for home-equity lines of credit climbed sharply as property values rose