Improvements in loss given default forecasts for bank loans

An accurate forecast of the parameter loss given default (LGD) of loans plays a crucial role for risk-based decision making by banks. We theoretically analyze problems arising when forecasting LGDs of bank loans that lead to inconsistent estimates and a low predictive power. We present several improvements for LGD estimates, considering length-biased sampling, different loan characteristics depending on the type of default end, and different information sets according to the default status. We empirically demonstrate the capability of our proposals based on a data set of 69,985 defaulted bank loans. Our results are not only important for banks, but also for regulators, because neglecting these issues leads to a significant underestimation of capital requirements.     

Downturn LGD modeling using quantile regression

Literature on Losses Given Default (LGD) usually focuses on mean predictions, even though losses are extremely skewed and bimodal. This paper proposes a Quantile Regression (QR) approach to get a comprehensive view on the entire probability distribution of losses. The method allows new insights on covariate effects over the whole LGD spectrum. In particular, middle quantiles are explainable by observable covariates while tail events, e.g., extremely high LGDs, seem to be rather driven by unobservable random events. A comparison of the QR approach with several alternatives from recent literature reveals advantages when evaluating downturn and unexpected credit losses. In addition, we identify limitations of classical mean prediction comparisons and propose alternative goodness of fit measures for the validation of forecasts for the entire LGD distribution.     

Measuring price discovery: The variance ratio, the R, and the weighted price contribution

We analyze the statistical properties of three price discovery measures: The variance ratio, the weighted price contribution (WPC), and the R² of unbiasedness regressions. We find that, if the price process is a driftless martingale, only the WPC is an unbiased estimator for the return variance explained during a time interval. For autocorrelated processes with a drift, only the R² of the unbiasedness regression is consistent, but it is biased for small samples.     

Volatility and covariation of financial assets: A high-frequency analysis

Using high frequency data for the price dynamics of equities we measure the impact that market microstructure noise has on estimates of the: (i) volatility of returns; and (ii) variance–covariance matrix of n assets. We propose a Kalman-filter-based methodology that allows us to deconstruct price series into the true efficient price and the microstructure noise. This approach allows us to employ volatility estimators that achieve very low Root Mean Squared Errors (RMSEs) compared to other estimators that have been proposed to deal with market microstructure noise at high frequencies. Furthermore, this price series decomposition allows us to estimate the variance covariance matrix of n assets in a more efficient way than the methods so far proposed in the literature. We illustrate our results by calculating how microstructure noise affects portfolio decisions and calculations of the equity beta in a CAPM setting.     

Loss given default for leasing: Parametric and nonparametric estimations

This study employs a dataset from three German leasing companies with 14,322 defaulted leasing contracts to analyze different approaches to estimating the loss given default (LGD). Using the historical average LGD and simple OLS-regression as benchmarks, we compare hybrid finite mixture models (FMMs), model trees and regression trees and we calculate the mean absolute error, root mean squared error, and the Theil inequality coefficient. The relative estimation accuracy of the methods depends, among other things, on the number of observations and whether in-sample or out-of-sample estimations are considered. The latter is decisive for proper risk management and is required for regulatory purposes. FMMs aim to reproduce the distribution of realized LGDs and, therefore, perform best with respect to in-sample estimations, but they show poor performance with respect to out-of-sample estimations. Model trees, by contrast, are more robust and outperform all other methods if the sample size is sufficiently large.     

Estimating loss-given default through advanced credibility theory

The New Basel Accord allows internationally active banking organizations to calculate their credit risk capital requirements using an internal ratings based approach, subject to supervisory review. One of the modeling components is the loss-given default (LGD): it represents the credit loss for a bank when extreme events occur that influence the obligor ability to repay his debts to the bank. Among researchers and practitioners the use of statistical models such as linear regression, Tobit or decision trees is quite common in order to compute LGDs as a forecasting of historical losses. However, these statistical techniques do not seem to provide robust estimation and show low performance. These results could be driven by some factors that make differences in LGD, such as the presence and quality of collateral, timing of the business cycle, workout process management and M&A activity among banks. This paper evaluates an alternative method of modeling LGD using a technique based on advanced credibility theory typically used in actuarial modeling. This technique provides a statistical component to the credit and workout experts’ opinion embedded in the collateral and workout management process and improve the predictive power of forecasting. The model has been applied to an Italian Bank Retail portfolio represented by Overdrafts; the application of credibility theory provides a higher predictive power of LGD estimation and an out-of-time sample backtesting has shown a stable accuracy of estimates with respect to the traditional LGD model.     

The influence of governance on investment: Evidence from a hazard model

Does corporate governance affect the timing of large investment projects? Hazard model estimates suggest strong shareholder governance may deter managers from pursuing large investments. Controlling for investment opportunities, firms with good governance experience longer spells between large investments. However, in the presence of financial constraints or strong CEO incentives (high delta (δ)), we find no such timing differences. Finally, these higher investment hazard firms exhibit significantly negative long-run operating and stock performance. Overall, our findings are consistent with the notion that poor governance associates with overinvestment.     

Control-ownership wedge and investment sensitivity to stock price

This study examines whether insiders’ incentives for private control benefits affect investment sensitivity to stock price. While Chen et al. (2007) link stock price informativeness to firms’ learning from the stock market, we offer an alternative agency-cost based explanation. Using a total of 2822 firms from 22 countries in East Asia and Western Europe, we document a strong negative association between control-ownership wedge and investment-q sensitivity, suggesting that insiders’ incentives for private control benefit reduce their propensity to listen to the market. Furthermore, the negative impact of wedge on investment-q sensitivity is primarily driven by sub-optimal investments. Overall, we provide evidence that agency problem is an important factor that determines the learning from the stock market in capital allocation.     

On the (in)feasibility of covered interest parity as a solution to the forward bias puzzle

This paper examines Pippenger's (2011) proposed solution to the forward bias puzzle, which is based on the covered interest parity (CIP) condition. It is argued that the CIP-based approach does not solve this well known and long-standing puzzle in international finance in a meaningful way. Moreover, it is shown that empirical results from such an approach follow mechanically from the identity-like nature of the theory of covered interest parity, which, aside from small deviations due to transaction costs, is assumed to hold in all periods (as if it were an identity). We show that rather than leading to new insights, the simple reconfiguration of CIP to solve for the time t + 1 spot exchange rate leads to tautological expressions that, when estimated, might appear to successfully explain the forward bias, but in actuality are trivial. Results from simple simulation exercises further illustrate the inconclusiveness of the proposed solution method.     

Conditional beta pricing models: A nonparametric approach

We propose a two-stage procedure to estimate conditional beta pricing models that allows for flexibility in the dynamics of asset betas and market prices of risk (MPR). First, conditional betas are estimated nonparametrically for each asset and period using the time-series of previous data. Then, time-varying MPR are estimated from the cross-section of returns and betas. We prove the consistency and asymptotic normality of the estimators. We also perform Monte Carlo simulations for the conditional version of the three-factor model of Fama and French (1993) and show that nonparametrically estimated betas outperform rolling betas under different specifications of beta dynamics. Using return data on the 25 size and book-to-market sorted portfolios, we find that the nonparametric procedure produces a better fit of the three-factor model to the data, less biased estimates of MPR and lower pricing errors than the Fama–MacBeth procedure with betas estimated under several alternative parametric specifications.     

Basel Core Principles and bank soundness: Does compliance matter?

This paper studies whether compliance with the Basel Core Principles for effective banking supervision (BCPs) is associated with bank soundness. Using data for over 3000 banks in 86 countries, we find that neither the overall index of BCP compliance nor its individual components are robustly associated with bank risk measured by individual bank Z-scores. We also fail to find a relationship between BCP compliance and systemic risk measured by a system-wide Z-score.     

Hometown advantage: The effects of monitoring institution location on financial reporting discretion

We examine the impact of institutional ownership on financial reporting discretion, focusing on whether the impact varies with institutions' cost of acquiring monitoring information. Using geographic distance between the firm and the institutional investor as a proxy for the cost of acquiring monitoring information, we find that corporate managers are less likely to use financial reporting discretion in the presence of local monitoring institutions than distant monitoring institutions. We also find that the impact of monitoring institutions on financial reporting discretion varies with the costs and benefits of financial reporting discretion.     

Dividend signaling under economic adversity: Evidence from the London Stock Exchange

The signaling or information content hypothesis is amongst the most prominent theories attempting to explain dividend policy decisions. However, no research has, to date, examined the information content of dividends in conjunction with generalized economic adversity. With the majority of the western economies facing the tough reality of the economic recession since late 2007–early 2008, we focus on the possibility of asymmetrical dividend signaling effects between periods of stability and economic adversity. Using data from the London Stock Exchange (LSE), where earnings and dividend news are released simultaneously, we test the dividend signaling hypothesis and the interaction of earnings and dividends under both steady and adverse economic conditions. We document positive and significant average abnormal stock price returns around the dividend/earnings announcements. We also find a significant interaction between economic conditions and the information content of dividends. After testing the dividend signaling hypothesis under both stable and recessionary economic conditions we find that dividends have less information content than earnings in periods of growth and stability, but more in periods of economic adversity.     

Hedging with Two Futures Contracts: Simplicity Pays

We propose to use two futures contracts in hedging an agricultural commodity commitment to solve either the standard delta hedge or the roll‐over issue. Most current literature on dual‐hedge strategies is based on a structured model to reduce roll‐over risk and is somehow difficult to apply for agricultural futures contracts. Instead, we propose to apply a regression based model and a naive rules of thumb for dual‐hedges which are applicable for agricultural commodities. The naive dual strategy stems from the fact that in a large sample of agricultural commodities, De Ville, Dhaene and Sercu (2008) find that GARCH‐based hedges do not perform as well as OLS‐based ones and that we can avoid estimation error with such a simple rule. Our semi‐naive hedge ratios are driven from two conditions: omitting exposure to spot price and minimising the variance of the unexpected basis effects on the portfolio values. We find that, generally, (i) rebalancing helps; (ii) the two‐contract hedging rules do better than the one‐contract counterparts, even for standard delta hedges without rolling‐over; (iii) simplicity pays: the naive rules are the best one–for corn and wheat within the two‐contract group, the semi‐naive rule systematically beats the others and GARCH performs worse than OLS for either one‐contract or two‐contract hedges and for soybeans the traditional naive rule performs nearly as well as OLS. These conclusions are based on the tests on unconditional variance ( Diebold and Mariano, 1995 ) and those on conditional risk ( Giacomini and White, 2006 ).     

The risk in hedge fund strategies: Theory and evidence from long/short equity hedge funds

Theory suggests that long/short equity hedge funds' returns come from directional as well as spread bets on the stock market. Empirical analysis finds persistent net exposures to the spread between small vs large cap stocks in addition to the overall market. Together, these factors account for more than 80% of return variation. Additional factors are price momentum and market activity. Combining two major branches of hedge fund research, our model is the first that explicitly incorporates the effect of funding (stock loan) on alpha. Using a comprehensive dataset compiled from three major database sources, we find that among the three thousand plus hedge funds with similar style classification, less than 20% of long/short equity hedge funds delivered significant, persistent, stable positive non-factor related returns. Consistent with the predictions of the Berk and Green (2004) model we find alpha producing funds decays to “beta-only” over time. However, we do not find evidence of a negative effect of fund size on managers' ability to deliver alpha. Finally, we show that non-factor related returns, or alpha, are positively correlated to market activity and negatively correlated to aggregate short interest. In contrast, equity mutual funds and long-bias equity hedge funds have no significant, persistent, non-factor related return. Expressed differently, L/S equity hedge funds, as the name suggests, do benefit from shorting. Besides differences in risk taking behavior, this is a key feature distinguishing L/S funds from long-bias funds.     

OLIVE: A SIMPLE METHOD FOR ESTIMATING BETAS WHEN FACTORS ARE MEASURED WITH ERROR

We propose a simple and intuitive method for estimating betas when factors are measured with error: ordinary least squares instrumental variable estimator (OLIVE). OLIVE performs well when the number of instruments becomes large, whereas the performance of conventional instrumental variable methods becomes poor or even infeasible. In an empirical application, OLIVE beta estimates improve R² significantly. More important, our results help resolve two puzzling findings in the prior literature: first, the sign of average risk premium on the beta for market return changes from negative to positive; second, the estimated value of average zero‐beta rate is no longer too high.     

Modeling and forecasting expected shortfall with the generalized asymmetric Student-t and asymmetric exponential power distributions

Financial returns typically display heavy tails and some degree of skewness, and conditional variance models with these features often outperform more limited models. The difference in performance may be especially important in estimating quantities that depend on tail features, including risk measures such as the expected shortfall. Here, using recent generalizations of the asymmetric Student-t and exponential power distributions to allow separate parameters to control skewness and the thickness of each tail, we fit daily financial return volatility and forecast expected shortfall for the S&P 500 index and a number of individual company stocks; the generalized distributions are used for the standardized innovations in a nonlinear, asymmetric GARCH-type model. The results provide evidence for the usefulness of the general distributions in improving fit and prediction of downside market risk of financial assets. Constrained versions, corresponding with distributions used in the previous literature, are also estimated in order to provide a comparison of the performance of different conditional distributions.     

Forecasting the yield curve with linear factor models

In this work we compare the interest rate forecasting performance of a broad class of linear models. The models are estimated through a MCMC procedure with data from the US and Brazilian markets. We show that a simple parametric specification has the best predictive power, but it does not outperform the random walk. We also find that macroeconomic variables and no-arbitrage conditions have little effect to improve the out-of-sample fit, while a financial variable (Stock Index) increases the forecasting accuracy.     

Trading frequency and asset pricing on the London Stock Exchange: Evidence from a new price impact ratio

This study proposes a new price impact ratio as an alternative to the widely used Amihud’s (2002) Return-to-Volume ratio. We demonstrate that the new price impact ratio, which is deemed Return-to-Turnover ratio, has a number of appealing features. Using daily data from all stocks listed on the London Stock Exchange over the period 1991–2008, we provide overwhelming evidence that this ratio, while being unequivocal to construct and interpret, is also free of a size bias. More importantly, it encapsulates the stocks’ cross-sectional variability in trading frequency, a relatively neglected but possibly important determinant of stock returns given the recently observed trends in financial markets. Overall, our findings argue against the conventional wisdom that there is a simple direct link between trading costs and stock returns by strongly suggesting that it is the compound effect of trading frequency and transaction costs that matters for asset pricing, not each aspect in isolation.