US Mortgage Research Articles

Racial Disparities in the US Mortgage Market

We study racial disparities in the US mortgage market. Using new data from Hurtado and Sakong (2024), we present three findings. First, we document access disparities between minority and otherwise-identical White borrowers even within the same bank and with the same loan officer. In contrast, cost disparities are nearly zero. Second, the use of automated underwriting algorithms is associated with smaller access disparities but slightly larger cost disparities. Third, individual factors such as loan officers' race and whether borrowers' race is observed at application do not seem to matter much.

Discrete-Time Survival Models with Neural Networks for Age–Period–Cohort Analysis of Credit Risk

Survival models have become popular for credit risk estimation. Most current credit risk survival models use an underlying linear model. This is beneficial in terms of interpretability but is restrictive for real-life applications since it cannot discover hidden nonlinearities and interactions within the data. This study uses discrete-time survival models with embedded neural networks as estimators of time to default. This provides flexibility to express nonlinearities and interactions between variables and hence allows for models with better overall model fit. Additionally, the neural networks are used to estimate age–period–cohort (APC) models so that default risk can be decomposed into time components for loan age (maturity), origination (vintage), and environment (e.g., economic, operational, and social effects). These can be built as general models or as local APC models for specific customer segments. The local APC models reveal special conditions for different customer groups. The corresponding APC identification problem is solved by a combination of regularization and fitting the decomposed environment time risk component to macroeconomic data since the environmental risk is expected to have a strong relationship with macroeconomic conditions. Our approach is shown to be effective when tested on a large publicly available US mortgage dataset. This novel framework can be adapted by practitioners in the financial industry to improve modeling, estimation, and assessment of credit risk.

Paying Too Much? Borrower Sophistication and Overpayment in the US Mortgage Market

Paying Too Much? Borrower Sophistication and Overpayment in the US Mortgage Market

The Effects of Conforming Loan Systems on Housing Finance Stability: A Panel Analysis of US, Japanese, and South Korean Mortgage Loan Systems

The Effects of Conforming Loan Systems on Housing Finance Stability: A Panel Analysis of US, Japanese, and South Korean Mortgage Loan Systems

Advantageous selection with intermediaries: a study of GSE‐securitized mortgage loans

AbstractThis research studies the effects of mortgage subsidies and asymmetric information in the US mortgage market. I exploit discontinuities in interest rates generated by pricing rules and find patterns consistent with advantageous selection. I estimate an industry model that highlights the relationship between mortgage subsidies, intermediary lenders' incentives, and borrowers' advantageous selection. The model shows that mortgage subsidies enable advantageous selection, creating a deadweight loss of $7.90 billion. The counterfactual analysis reveals that pricing borrowers' private information eliminates advantageous selection only if mortgages are not subsidized. Without the mortgage subsidy, pricing borrowers' private information improves efficiency by $728.58 million.

Climate Change, Technology Shocks and the US Equity Real Estate Investment Trusts (REITs)

Given the renewed interest in Real Estate Investment Trusts (REITs), we are keenly focused on exploring the possible connection between climate change and return volatility of US equity REITs, as well as the role of technology innovation for environmental sustainability in the nexus. While climate change might pose some threat to the REIT business, it is necessary to know the direction in which technological innovation can mitigate this impact. As a way to validate our evidence, we offer some additional analyses with alternative measures of technology shocks and the replacement of technology shocks with global economic expansion, as improvement in global economic activity could offer more investment options for investors to diversify their investment portfolio away from climate-prone assets. For completeness, the analyses are replicated for US mortgage REITs. Overall, we show that climate change heightens the return volatility of US equity REITs and that the former contains some predictive content for the latter. When the role of technology is examined, our results show that technology shock indeed reverses the cheering impact of temperature anomaly on the return volatility of US equity REITs. We show that these results are robust to alternative measures of economic shock and that the results equally hold for mortgage REITs. We further document some important implications of our findings for investors and policymakers alike.

Algorithmic Fairness

This article reviews the recent literature on algorithmic fairness, with a particular emphasis on credit scoring. We discuss human versus machine bias, bias measurement, group versus individual fairness, and a collection of fairness metrics. We then apply these metrics to the US mortgage market, analyzing Home Mortgage Disclosure Act data on mortgage applications between 2009 and 2015. We find evidence of group imbalance in the dataset for both gender and (especially) minority status, which can lead to poorer estimation/prediction for female/minority applicants. Loan applicants are handled mostly fairly across both groups and individuals, though we find that some local male (nonminority) neighbors of otherwise similar rejected female (minority) applicants were granted loans, something that warrants further study. Finally, modern machine learning techniques substantially outperform logistic regression (the industry standard), though at the cost of being substantially harder to explain to denied applicants, regulators, or the courts.

Black swan events and stock market behavior in Gulf countries: a comparative analysis of financial crisis (2008) and COVID-19 pandemic

PurposeThe current study aims to examine the impact of two black swan events on the performance of six stock markets in Gulf Cooperation Council (GCC) economies (Abu Dhabi, Bahrain, Dubai, Oman, Qatar and Saudi Arabia). The two selected black swan events are the US Mortgage and credit crisis (Global Financial Crisis of 2008) and the COVID-19 pandemic.Design/methodology/approachThe performance of all the six stock markets are represented by their return and price volatility behavior, which has been measured by applying ARCH/GARCH model. The comparative analysis is done by employing mean difference models. The data is collected from Bloomberg on a daily frequency.FindingsThe response of two black swan events on the GCC stock markets has been heterogenous in nature. During the financial crisis, the impact was heavily felt on most of the stock markets in the GCC countries. It is revealed that the financial crisis had a negative significant impact on four of the six countries. Whereas during the COVID-19 crisis, it is revealed that there is no significant impact on four of the six selected stock markets. The positive significant impact is felt on two stock markets, namely, the Abu Dhabi stock market and the Saudi stock market.Originality/valueThe present investigation attempts to fill the gap in the literature on the intended topic because it is evident from the literature on the chosen subject that no study has been undertaken to evaluate and contrast the impact of the GFC crisis and COVID-19 on the GCC stock markets.

Banks, non-banks, and the incorporation of local information in CMBS loan pricing

Comparing banks to non-bank lenders, we investigate whether the geographical distance between lenders, borrowers, and their properties is reflected in the pricing of US mortgages that were included in US commercial mortgage-backed security (CMBS) pools during the 2000 to 2017 period. The difference in loan spreads when the bank-borrower distance increases from zero to the median of about 700 miles is 10 basis points, and this effect is more pronounced if the loan is collateralized by a riskier property. On the contrary, geographical distance does not seem to have any effect on the loan spread of mortgages granted by non-bank lenders. The difference in loan pricing across originator types (even after controlling for key mortgage and property characteristics) suggests that banks and non-bank lenders have different incentives, lending technologies, and/or different types of borrowers. Our results contribute to the emerging literature on non-bank lender behavior.

Stabilizing machine learning models with Age-Period-Cohort inputs for scoring and stress testing

Machine learning models have been used extensively for credit scoring, but the architectures employed suffer from a significant loss in accuracy out-of-sample and out-of-time. Further, the most common architectures do not effectively integrate economic scenarios to enable stress testing, cash flow, or yield estimation. The present research demonstrates that providing lifecycle and environment functions from Age-Period-Cohort analysis can significantly improve out-of-sample and out-of-time performance as well as enabling the model's use in both scoring and stress testing applications. This method is demonstrated for behavior scoring where account delinquency is one of the provided inputs, because behavior scoring has historically presented the most difficulties for combining credit scoring and stress testing. Our method works well in both origination and behavior scoring. The results are also compared to multihorizon survival models, which share the same architectural design with Age-Period-Cohort inputs and coefficients that vary with forecast horizon, but using a logistic regression estimation of the model. The analysis was performed on 30-year prime conforming US mortgage data. Nonlinear problems involving large amounts of alternate data are best at highlighting the advantages of machine learning. Data from Fannie Mae and Freddie Mac is not such a test case, but it serves the purpose of comparing these methods with and without Age-Period-Cohort inputs. In order to make a fair comparison, all models are given a panel structure where each account is observed monthly to determine default or non-default.

Выбор приоритетного пути повышения уровня конкурентоспособности экономики России в условиях неопределенности мировой конъюнктуры

This article is debatable and is devoted to the consideration of the problem of choosing ways to reform the Russian economy in the conditions of the crisis of the world economic system. The relevance of the proposed discussion is determined by the need to find effective instruments of economic regulation that allow the national economy of Russia to take a worthy competitive place in the changing geo-economic space. The article analyzes the main cause of crises in the world capitalist economy on the basis of the theory of K. Marx, which consists in the contradiction between the social nature of production and the private capitalist form of appropriation of the results of labor. The statistics of the world cyclical crises for the period from 1825 to the present are given. Cyclical crises, generated mainly by economic factors, are repeated periodically through 7–9–11 years with minor deviations. The author considers the cause of the US mortgage crisis and its impact on the economy of other countries, including Russia. Three possible ways of development of the Russian economy are shown: 1) pro-Western; 2) the development of the economy in the interests of the national bourgeoisie; 3) the NEP period of Soviet power, which is successfully used by China. The third way is objectively justified. This is evidenced by the statistics provided in this article. The necessity of changing in the near future the disastrous liberal course of managing the country’s economy, of carrying out a left-centrist financial and economic course towards justified socialism, is justified. The conditions for the transition to a managerial revolution that provides a leap in the development of the country are outlined. The directions for reducing the dependence of the Russian economy on the conditions of the capitalist market are proposed.

Monitoring System and Instruments for Risk Management in Commercial Banks

As social рractice shows, shortcomings in the system of management and control of the banking system can cause not only a worsening of the crisis state of the country's banking system but also the emergence of a systemic crisis at the national level due to the transfer of financial resources to other segments. In particular, as a result of allowing an unreasonable increase in the amount of non-рerforming loans (NРL) in the case of NPL in the case of loans from developed countries, US commercial banks and mortgage organizations led to a sharр increase in the number of non-рerforming loans in the grouр of unsecured borrowers, a sharр decrease in the investment attractiveness of mortgage securities on a financial basis, and the global financial and economic crisis has led to its exрansion around the world. In the course of the study, the effective organization and implementation of the “monitoring of a bank account” activity, which is an integral part of the activity of a bank account, as well as the scientific and theoretical foundations of the concept of “bank account” were studied scientifically and conceрtually.

Bowling alone, buying alone: The decline of co-borrowers in the US mortgage market

This paper documents stylized and empirical facts associated with co-borrowers in the US mortgage market since the early 1990s. The share of mortgages with a co-borrower has declined dramatically across different income and demographic groups. We show that this decline, despite being a universal phenomenon across the US, evinces significant regional heterogeneity which contributes to the divergence in local mortgage markets outcomes. Regions with a lower co-borrower share have higher mortgage default rates. Further, in an event of an adverse shock, regions with a low share of mortgages with a co-borrower experience persistently lower house price growth, and lower purchase and refinance mortgage growth.

Joint models for longitudinal and discrete survival data in credit scoring

The inclusion of time-varying covariates into survival analysis has led to better predictions of the time to default in behavioural credit scoring models. However, when these time-varying covariates are endogenous, there are two major problems: estimation bias of the survival model and lack of a prediction framework for future values of both the event and the endogenous time-varying covariates. Joint models for longitudinal and survival data is an appropriate framework to model the mutual evolution of the survival time and the endogenous time-varying covariates. To the best of our knowledge, this paper explores for the first time the application of discrete-time joint models to credit scoring. Moreover, we propose a novel extension to the joint model literature by including autoregressive terms in modelling the endogenous time-varying covariates. We present the method via simulations and by applying it to US mortgage loans. The empirical analysis shows, first, that discrete joint models can increase the discrimination performance compared to survival models. Second, when an autoregressive term is included, this performance can be further improved.

Impact of mortgage soft information in loan pricing on default prediction using machine learning

AbstractWe analyze the impact of soft information on US mortgages for default prediction and provide a new measure for lender soft information that is based on the interest rates offered to borrowers and incremental to public hard information. Hard and soft information provide for a variation in annual default probabilities of approximately 3%. Soft information has a lesser impact over time and time since origination. Lenders rely more on soft information for high‐risk borrowers. Our study evidences the importance of soft information collected at loan origination.

The use of digital footprints in the US mortgage market

AbstractWe study the rise of digital footprint (DF) users in the U.S. residential mortgage market. The proportion of lenders that use a borrower's DF has witnessed remarkable growth from 6% in 2013 to 34% in 2018 in a short span – according to our analysis. We show that the use of DFs can significantly reduce a lender's overall risk and that the use of DFs can result in considerable societal benefits, by reducing the overall discriminatory forces. In sum, we provide evidence that the informational advantage associated with using a borrower's DF far outweighs that of existing traditional lending methods.

Quadruple wavelet coherence and pairwise cointegration: comparative time-frequency analysis of bank deposit rate

PurposeThis study aims to attempt to understand the joint co-movement of bank deposit rate and its main underlying determinants (foreign exchange rate (FX) rate, cross-currency swap rate and implied forward rate). The authors also compare time and frequency variant approaches in this dynamic.Design/methodology/approachThe authors examine bank deposit rates where multiple variables jointly interact, and the integration is time and frequency variant. The study applies both cointegration and wavelet coherence methods and conducts a comparative analysis. It investigates eight markets over 2005–2020 aiming to capture the impact of changing market conditions and degree of development.FindingsThe results are in line with cross-country interdependence, where we observe more robust evidence for co-movement during adverse economic conditions with higher correlation compared to other periods such as the 2007–2009 US mortgage crises, 2010–2012 Euro crises and 2019 pandemic. Moreover, wavelet analysis suggests deposit rate lags FX rate and leads cross-country swap rate. The USA arguably leads the co-movement accompanied briefly by Japan and followed closely by other developed markets and later the developing markets. Heat maps suggest clustering of countries.Practical implicationsThe wavelet coherence's ability to indicate the periods and the frequencies of the relationship is essential to capture the true nature of the relationship. Such additional insight would enable the practitioners to determine the true price of the deposit rate.Originality/valueThe study captures the long suggested collective nature of three main underlying determinants of bank deposit. The results shed light on the order of dynamics in a complex bank deposit environment. Comparative analysis further highlights the valuable insight quadruple wavelet coherence provides.

Feeling the Heat: Mortgage Lending and Central Bank Options

Extreme heat periods due to the changing climate are having a negative impact on many areas of the economy. In our study, we look at how much US mortgage lending is originated in the areas that are most vulnerable to future heat waves, and what central bank and supervisory authority actions could mitigate the resulting risk. From our results, we see that proportionally more lending flows into the areas that are likely to be most exposed to heat in the future. Population and economic output are relatively higher in these areas, and thus climate risk is less of a factor in lending decisions. However, lenders reject proportionately slightly more mortgage applications in the counties that are expected to be the hottest. Central bank and supervisory authority measures to support climate objectives are therefore of key importance. These could include, for example, central bank asset purchase programmes to support sustainable construction, especially in areas vulnerable to climate change, or the development of collateral management along similar lines. Coordination between the different authorities is also important because of the limitations of central bank actions.

Monetary easing and the lending concentration channel of monetary policy transmission

I use loan-level data on US mortgage loan applications to identify the effect of lending concentration on the pass-through of the 2008 monetary easing to the volume of lending. Lenders eased credit conditions but less so in counties with higher lending concentration. Furthermore, within a county, the pass-through was lower for lenders with higher local market power. The channel is active also during the 2005 monetary tightening episode. It is distinct from the deposits channel of monetary policy transmission, and its influence is ubiquitous: it affects both new loans and refinances, depository and non-depository lenders, and market leaders and laggards.

Spillovers of US unconventional monetary policy: quantitative easing, spreads, and international financial markets

This study investigates the international spillover effects of US unconventional monetary policy (UMP)—frequently called large-scale asset purchases or quantitative easing (QE)—on advanced and emerging market economies, using structural vector autoregressive models with high-frequency daily data. Blinder (Federal Reserve Bank of St. Louis Rev 92(6): 465–479, 2010) argued that the QE measures primarily aim to reduce US interest rate spreads, such as term and risk premiums. Considering this argument and recent empirical evidence, we use two spreads as indicators of US UMP: the mortgage and term spreads. Based on data from 20 emerging and 20 advanced countries, our empirical findings reveal that US unconventional monetary policies significantly affect financial conditions in emerging and advanced countries by altering the risk-taking behavior of investors. This result suggests that the risk-taking channel plays an important role in transmitting the effects of these policies to the rest of the world. The extent of these effects depends on the type of QE measures. QE measures such as purchases of private sector securities that lower the US mortgage spread exert stronger and more significant spillover effects on international financial markets than those that reduce the US term spread. Furthermore, the estimated financial spillovers vary substantially across countries and between and within the emerging and advanced countries that we examine in this study.