Political regimes,business cycles,seasonalities, and returns

This paper provides a method for testing for regime differences when regimes are long-lasting. Standard testing procedures are generally inappropriate because regime persistence causes a spurious regression problem – a problem that has led to incorrect inference in a broad range of studies involving regimes representing political, business, and seasonal cycles. The paper outlines analytically how standard estimators can be adjusted for regime dummy variable persistence. While the adjustments are helpful asymptotically, spurious regression remains a problem in small samples and must be addressed using simulation or bootstrap procedures. We provide a simulation procedure for testing hypotheses in situations where an independent variable in a time-series regression is a persistent regime dummy variable. We also develop a procedure for testing hypotheses in situations where the dependent variable has similar properties.     

In search of robust methods for dynamic panel data models in empirical corporate finance

We examine which methods are appropriate for estimating dynamic panel data models in empirical corporate finance. Our simulations show that the instrumental variable and GMM estimators are unreliable, and sensitive to the presence of unobserved heterogeneity, residual serial correlation, and changes in control parameters. The bias-corrected fixed-effects estimators, based on an analytical, bootstrap, or indirect inference approach, are found to be the most appropriate and robust methods. These estimators perform reasonably well even in models with fractional dependent variables censored at [0, 1]. We verify these results in two empirical applications, on dynamic capital structure and cash holdings.     

Inference in models with adaptive learning

Identification of structural parameters in models with adaptive learning can be weak, causing standard inference procedures to become unreliable. Learning also induces persistent dynamics, and this makes the distribution of estimators and test statistics non-standard. Valid inference can be conducted using the Anderson-Rubin statistic with appropriate choice of instruments. Application of this method to a typical new Keynesian sticky-price model with perpetual learning demonstrates its usefulness in practice.     

Estimating dynamic panel models in corporate finance

Dynamic panel models play a natural role in several important areas of corporate finance, but the combination of fixed effects and lagged dependent variables introduces serious econometric bias. Several methods of counteracting these biases are available and these methodologies have been tested on small datasets with independent, normally-distributed explanatory variables. However, no one has evaluated the methods' performance with corporate finance data, in which the dependent variable may be clustered or censored and independent variables may be missing, correlated with one another, or endogenous. We find that the data's properties substantially affect the estimators' performances. We provide evidence about the impact of various data set characteristics on the estimators, so that researchers can determine the best approach for their datasets.     

Will Any q Do?

We find that the relative levels of computationally costly q estimators and simple q estimators, when used as continuous variables, are affected by variations in many firm financial characteristics. In contrast, when the estimators are used as dichotomous variables, they classify the vast majority of firms identically with respect to the unit q breakpoint. Finally, we find that the computationally costly approach may induce sample‐selection bias as a result of data unavailability. Our results suggest that the simple approach is preferable except when extreme precision of the q estimate is of paramount importance and sample‐selection bias is not likely to be an issue.     

Statistical inference using the T index to quantify the level of comparability between accounts

The extent to which the accounts of companies are comparable is considered important to users and regulators. However, prior research has been restricted by a lack of appropriate statistical methods for testing comparability indices. This has made it difficult to assess the true level of comparability from sample data and to test research hypotheses such as whether the level of comparability (a) differs by policy, (b) differs by country, and (c) changes over time. This paper fills this gap by exploring the statistical properties of the T index. The T index generalises the H, C, I and various modifications of these indices and represents a unified framework for the measurement of the extent to which the accounts of companies are comparable. Formulae for the bias and standard error for any index under this framework are provided and proved. The bias is shown to equal zero or be negligible in most practical situations. Using historical data, the standard error is used to illustrate the accuracy with which comparability is estimated and to perform formal statistical inference using confidence intervals and p‐values. Furthermore, the sampling distribution of the T index is assessed for normality. Implications for research design and sample size determination are also discussed.     

Simple estimators for the parameters of discrete dynamic games (with entry/exit examples)

We estimate parameters from data on discrete dynamic games, using entry/exit games to illustrate. Semiparametric first‐stage estimates of entry and continuation values are computed from sample averages of the realized continuation values of entrants and incumbents. Under certain assumptions, these values are easy‐to‐compute analytic functions of the parameters of interest. The entry and continuation values are used to determine the model's predictions for entry and exit conditional on the parameter vector, and the estimates compare these predictions with the data on entry and exit rates. Small‐sample properties are discussed and lead to the simplest of estimators.     

The Cost Channel Reconsidered: A Comment Using an Identification‐Robust Approach

We reexamine the empirical relevance of the cost channel of monetary policy (e.g., Ravenna and Walsh 2006 ), employing recently developed moment‐conditions inference methods, including identification‐robust procedures. Using U.S. data, our results suggest that the cost channel effect is poorly identified and we are thus unable to corroborate the previous results in the literature.     

The t copula with multiple parameters of degrees of freedom: bivariate characteristics and application to risk management

The t copula is often used in risk management as it allows for modeling the tail dependence between risks and it is simple to simulate and calibrate. However, the use of a standard t copula is often criticized due to its restriction of having a single parameter for the degrees of freedom (dof) that may limit its capability to model the tail dependence structure in a multivariate case. To overcome this problem, the grouped t copula was proposed recently, where risks are grouped a priori in such a way that each group has a standard t copula with its specific dof parameter. In this paper we propose the use of a generalized grouped t copula, where each group consists of one risk factor only, so that a priori grouping is not required. The copula characteristics in the bivariate case are studied. We explain simulation and calibration procedures, including a simulation study on the finite sample properties of the maximum likelihood estimators and Kendall's tau approximation. This new copula is significantly different from the standard t copula in terms of risk measures such as tail dependence, value at risk and expected shortfall.     

Granularity adjustment for default risk factor model with cohorts

This paper examines granularity adjustments to parameter estimators in a default risk model with cohorts. The model is an extension of the Vasicek model (Vasicek, 1991) and includes a general factor and cohort specific factors. The granularity adjustments derived in the paper concern the mean and/or the variance of observed default frequencies and are easy to implement in practice. For illustration, the method is applied to the S&P corporate ratings. The Granularity Adjusted (GA) estimators are compared to the unadjusted estimators in terms of their asymptotic properties and in finite sample.     

Orthogonal Frontiers and Alternative Mean-Variance Efficiency Tests

This paper catalogues properties of minimum norm orthogonal portfolios: portfolios which minimize a quadratic objective function and have returns uncorrelated with those of a candidate portfolio that is not mean-variance efficient. The analysis shows that the dollar versions of these portfolios correspond to estimators of zero beta rates based on alternative statistical criteria and grouping procedures while costless orthogonal portfolios represent candidate mean-variance efficiency tests. It also develops inference procedures for zero and unit net investment portfolios of individual securities (instead of grouped portfolios) that have zero expected betas. The resulting mean-variance efficiency tests are reasonably insensitive to the underlying statistical assumptions.     

Fiscal Stimulus in Times of High Debt: Reconsidering Multipliers and Twin Deficits

We investigate the impact of fiscal stimuli at different levels of the government debt‐to‐GDP ratio for a sample of 17 European countries from 1970 to 2010. This is implemented in an interacted panel VAR framework in which all coefficient parameters are allowed to change continuously with the debt‐to‐GDP ratio. We find that responses to government spending shocks exhibit strong nonlinear behavior. While the overall cumulative effect of a spending shock on real GDP is positive and significant at moderate debt‐to‐GDP ratios, this effect turns negative as the ratio increases. The total cumulative effect on the trade balance as a share of GDP is negative at first but switches sign at higher levels of debt. Consequently, depending on the degree of public indebtedness, our results accommodate long‐run fiscal multipliers that are greater and smaller than one or even negative as well as twin deficit and twin divergence behavior within one sample and time period. From a policy perspective, these results lend additional support to increased prudence at high public debt ratios because the effectiveness of fiscal stimuli to boost economic activity or resolve external imbalances may not be guaranteed.     

Robust bootstrap procedures for the chain-ladder method

Insurers are faced with the challenge of estimating the future reserves needed to handle historic and outstanding claims that are not fully settled. A well-known and widely used technique is the chain-ladder method, which is a deterministic algorithm. To include a stochastic component one may apply generalized linear models to the run-off triangles based on past claims data. Analytical expressions for the standard deviation of the resulting reserve estimates are typically difficult to derive. A popular alternative approach to obtain inference is to use the bootstrap technique. However, the standard procedures are very sensitive to the possible presence of outliers. These atypical observations, deviating from the pattern of the majority of the data, may both inflate or deflate traditional reserve estimates and corresponding inference such as their standard errors. Even when paired with a robust chain-ladder method, classical bootstrap inference may break down. Therefore, we discuss and implement several robust bootstrap procedures in the claims reserving framework and we investigate and compare their performance on both simulated and real data. We also illustrate their use for obtaining the distribution of one year risk measures.     

Two-pass estimation of risk premiums with multicollinear and near-invariant betas

This paper investigates the reliability of the two-pass (TP) estimators of factor risk prices when betas (multifactor loadings) have high levels of cross-sectional correlation (multicollinearity) and/or when some of them have small cross-sectional variations (near-invariance). Our simulation results show the following. First, the TP estimators can have biases larger than 100% of true risk prices when data are generated by the betas with high levels of multicollinearity and invariance that can be observed from actual data. Second, the t-tests for hypotheses related to risk prices and pricing intercepts have only limited power. The levels of multicollinearity and invariance of betas can vary depending on the assets and sample periods used in estimation. Thus, we propose use of two pre-diagnostic statistics to measure these levels. Many previous studies have investigated the finite-sample properties of the TP estimators using the data generated with the estimated betas from actual data. Our results indicate that simulation outcomes can lead to quite different conclusions, depending on the levels of multicollinearity and invariance of the betas used to generate the data.     

Evidence on Simulation Inference for Near Unit-Root Processes with Implications for Term Structure Estimation

The high persistence of interest rates has important implicationsfor the preferred method used to estimate term structure models.We study the finite-sample properties of two standard dynamicsimulation methods—efficient method of moments (EMM) andindirect inference—when they are applied to an first orderautoregressive (AR[1]) process with Gaussian innovations. Whensimulated data are as persistent as interest rates, the finite-sampleproperties of EMM differ both from their asymptotic propertiesand from the finite-sample properties of indirect inferenceand maximum likelihood. EMM produces larger confidence boundsthan indirect inference and maximum likelihood, yet is muchless likely to contain the true parameter value. This is primarilybecause the population variance of the data plays a much largerrole in the EMM conditions than in the moment conditions foreither indirect inference or maximum likelihood. These resultssuggest that, under Gaussian assumptions, indirect inference(if practical) is preferable to EMM when working with persistentdata such as interest rates. EMM's emphasis on the populationvariance strongly enforces stationarity on the underlying process,so this same reasoning suggests that EMM may be preferable insettings where stability and stationarity are important anddifficult to impose.     

Dynamic risksharing in the United States and Europe

We use a panel VAR model to improve upon the existing methodologies to analyze interregional risksharing and consumption smoothing channels. First, we endogenize the output process within a more general multi-equation framework, capturing the dynamic feedback between output and various smoothing channels. Second, in line with dynamic general equilibrium open economy models of risksharing, we exploit impulse response functions to trace the role of each smoothing channel over time, in the presence of different structural shocks (temporary vs. permanent and output vs. smoothing channels). In the application to the US and OECD countries, we find different dynamic properties of different smoothing channels. We compare our results with the predictions of standard risksharing and consumption theories, and tackle some of the puzzles in the literature, such as the “international risksharing puzzle” and the “consumption-output correlation puzzle.” We are also able to address such policy issues as whether fiscal stabilizers have been substitutes or complements for financial market diversification activities and whether further financial market integration is likely to provide countries with more shock-absorption tools. A key result is the strong substitutability between capital and credit smoothing in the US, and between fiscal and credit smoothing in the OECD.     

Fundamentals,Misvaluation, and Business Investment

Does stock market misvaluation affect business fixed investment? To answer this question, we provide evidence based on U.S. firm‐level panel data. We examine the orthogonality conditions for the investment Q and Euler equations, and our qualitative tests reject the null hypothesis that investment is unaffected by misvaluation (this result is not driven exclusively by the late 1990s). To measure the quantitative effects on investment, we introduce a measure of misvaluation into standard investment equations. Our estimates imply that a one‐standard‐deviation increase in misvaluation increases investment between 20% and 60% relative to the mean level of investment in the sample.     

Do Large Banks Have Lower Costs? New Estimates of Returns to Scale for U.S. Banks

This paper presents new, fully nonparametric estimates of ray‐scale and expansion‐path scale economies for U.S. banks based on a model of bank costs. Unlike prior studies that use models with restrictive parametric assumptions or limited samples, our methodology uses local polynomial estimators and data on all U.S. banks over the period 1984–2006. Our estimates indicate that as recently as 2006, most U.S. banks faced increasing returns to scale, suggesting that scale economies are a plausible (but not necessarily only) reason for the growth in average bank size and that the tendency toward increasing scale is likely to continue unless checked by government intervention.     

Is the Short Rate Drift Actually Nonlinear?

Aït-Sahalia (1996) and Stanton (1997) use nonparametric estimators applied to short-term interest rate data to conclude that the drift function contains important nonlinearities. We study the finite-sample properties of their estimators by applying them to simulated sample paths of a square-root diffusion. Although the drift function is linear, both estimators suggest nonlinearities of the type and magnitude reported in Aït-Sahalia (1996) and Stanton (1997). Combined with the results of a weighted least squares estimator, this evidence implies that nonlinearity of the short rate drift is not a robust stylized fact.