Stable cointegrating regressions: Fully-modified estimates for inflation and employment cost indices

This paper investigates the stability of relationships between inflation and changes in employer cost for labor using tests based on the null hypothesis of no cointegration (Gregory and Hansen (1996a)) as well as tests based on the null hypothesis of cointegration (Hansen (1992)). In addition to specifications which include wages or unit labor cost, employment cost indices for compensation and wages are used to eliminate composition bias over the business cycle. Empirical results support stability and fully-modified estimates are obtained using the semiparametric approach of Phillips (1995). In contrast to studies which have found only one-way causality, this paper presents empirical evidence of feedback between inflation and employer cost for labor, consistent with an expectations-adjusted Phillips's Curve. First version received: July 1997/final version received: September 1998     

Testing purchasing power parity in a multivariate cointegrating framework

This paper examines a bilateral PPP(Purchasing power parity) relationship between Australia and the 11 major trading countries by means of two alternative econometric techniques-a multivariate cointegrating framework and a band-spectral regression. It is acknowledged that there is no strong evidence that classical PPP holds in all cases. However, the generalized version of PPP holds in all cases, and provides a better explanation of the long-run relations between exchange rates and relative prices.The use of different price indices, i.e CPI and WPI,lead to different estimates and hence different policy implications.     

Maximum entropy regressions

Maximum entropy (ME) regression is compared to ordinary regression in the case of two observations on two normally distributed variables (one dependent and one explanatory) with correlation coefficient ρ. ME regressions have the smaller risk under quadratic loss if ? lies in the interval ±0.95. In the case of two explanatory variables and two observations, ordinary regression is not possible but ME regressions do exist and have finite risk.     

The tax versus debt controversy in a multivariate cointegrating system

The empirical literature is replete with evidence that budget deficits and government debt affect (traditional view) or do not appreciably affect (Ricardian equivalence) private consumption. However, the validity of the relative empirical estimates can be questioned to the extent that they do not adequately distinguish among countries with varying levels of indebtedness. The present paper attempts to address this problem by using Johansen's cointegration technique and by sorting the sample countries into groups, according to the ratio of debt to GDP. The empirical evidence generated finds support for the debt-illusion hypothesis as a means of expounding variations in consumers' behaviour among countries with varying degrees of indebtedness.     

Short-term eurocurrency rate behavior and specifications of cointegrating processes

This article presents empirical evidence on short-term behavior based on seven Eurocurrency market rates. Empirical analysis suggests that there is two-dimensional cointegration. First, the domestic short-term interest rate is cointegrated with longer-term interest rates within a particular country. Second, the domestic short-term interest rate is cointegrated with the comparable foreign short-term interest rate adjusted for the foreign exchange forward premium (discount). The empirical evidence confirms that an error-correction model combining both dimensional market-correcting processes better explains short-term interest rate movements.     

Spurious regressions with stationary series

A spurious regression occurs when a pair of independent series, but with strong temporal properties, are found apparently to be related according to standard inference in an OLS regression. Although this is well known to occur with pairs of independent unit root processes, this paper finds evidence that similar results are found with positively autocorrelated autoregressive series or long moving averages. This occurs regardless of the sample size and for various distributions of the error terms.     

Endogeneity bias and growth regressions

The problem of regressor endogeneity stemming from reverse casuality is one that has plagued economists working in the field of empirical economic growth for some time. This paper attempts to address the relevant magnitude of this issue in the context of growth regressions based on the Solow growth model. The paper develops a method of running Monte Carlo simulations that allows us to generate simulated data that match the moments of observed real-world data typically used in such regressions while simultaneously allowing us to impose arbitrarily high correlations between the steady-state determinants of the Solow model and the unobserved residual term of the data-generating process. After running simulations that represent a wide sample of the mathematically-possible correlations, we conclude that a between estimator or a random effects estimator will deliever a lower average absolute bias across all coefficients than alternative estimators in almost all of our simulations. Conversely, estimators that use within-country variation will generate lower biases when looking solely at rates of convergence. Furthermore, we conclude that these results are robust when restricting our sample of simulations to several subsets of the assumed parameters and to changing our assumptions about country fixed-effects terms.     

Dualism and cross-country growth regressions

This paper develops empirical growth models suitable for dual economies, and studies the relationship between structural change and economic growth. Changes in the structure of employment will raise aggregate productivity when the marginal product of labor varies across sectors. The models in the paper incorporate this effect in a more flexible way than previous work. Estimates of the models imply sizeable marginal product differentials, and indicate that the reallocation of labor makes a significant contribution to the international variation in productivity growth.     

Time variation in the cointegrating relationship between stock prices and economic activity

The present paper examines whether there exists a long‐run cointegrating relationship between a stock market index and output and interest rates. Moreover, estimation is conducted over the full sample and both a recursive and rolling sample to examine any time variation in the nature of the relationship. The results support evidence of a single cointegrating vector, where stock prices typically exhibit a positive relationship with industrial production and a negative relationship with interest rates. However, there is significant time variation and periods of time where contrary results are observed. As such any model of stock prices needs to account for such time variation     

Bayesian estimation of cointegrating thresholds in the term structure of interest rates

Threshold Error Correction Models are used to analyse the term structure of interest Rates. The paper develops and uses a generalisation of existing models that encompasses both the Band and Equilibrium threshold models of [Balke and Fomby ((1997) Threshold cointegration. Int Econ Rev 38(3):627–645)] and estimates this model using a Bayesian approach. Evidence is found for threshold effects in pairs of longer rates but not in pairs of short rates. The Band threshold model is supported in preference to the Equilibrium model.

Bias correcting adjustment coefficients in a cointegrated VAR with known cointegrating vectors

The maximum likelihood estimator of the adjustment coefficient in a cointegrated vector autoregressive model (CVAR) is generally biased. For the case where the cointegrating vector is known in a first-order CVAR with no intercept, we derive a condition for the unbiasedness of the maximum likelihood estimator of the adjustment coefficients, and provide a simple characterization of the bias in case this condition is violated. A feasible bias correction method is shown to virtually eliminate the bias over a large part of the parameter space.     

S-values and Bayesian weighted all-subsets regressions

This paper compares and contrasts Bayesian variable-exclusion methods proposed by Eduardo Ley and coauthors with methods proposed by Raftery and Sala-i-Martin et al. and with the s-values proposed by myself. A distinction is drawn between estimation uncertainty which is the focus of Ley׳s research and model ambiguity which arises in Ley׳s work and is the focus of my own recent proposal. The discussion is organized around the prior covariance matrix, which needs to be diagonal to support all-subsets regressions. The basic question addressed here is: what aspects of the prior covariance matrix can be taken as known, what aspects can be estimated and what aspects require a sensitivity analysis because they are neither known nor estimable. When diagonality is in doubt, we are more-or-less forced into a model ambiguity sensitivity mode because the data are never rich enough credibly to estimate the full prior covariance matrix. When diagonality is assumed, the data evidence, though very limited, can help to estimate the diagonal elements, but this literature has not yet produced a compelling conventional treatment which will necessarily include both estimation uncertainty and model ambiguity as they relate both to the diagonal values and to the rest of the prior covariance matrix. But there has been a lot of progress.     

Instrumental variables regressions involving seasonal data

Various approaches to handling seasonal data are known to be equivalent in the context of least squares estimation of a fixed-regressor linear model. This note extends these results to models which have stochastic regressors and are estimated by the method of Instrumental Variables.     

Spurious regressions driven by excessive volatility

This paper shows that the excessive volatility results in spurious regressions. The spuriousness can be driven by persistency in the error variances unlike the conventional spurious regressions that are generated by the persistency in the level of regression errors.     

Long-run money demand in Taiwan revisited: evidence from a cointegrating STR approach

Economic growth over the past two decades has failed to reduce income inequality. We contend that major reasons for this are the slowdown and bias in technological change (productivity growth). Given the complexity of the many interactions that take place, this phenomenon is best addressed in a general equilibrium context. For this purpose, we have developed a computable general equilibrium (CGE) model with advanced features relating to income distribution. We perform a series of simulations based on recent overall productivity changes, but under various forms of technological change bias, factor mobility, and government budgetary balance. We find the labour-augmenting technological change cases to be most consistent with recent experience.     

A Monte Carlo study on the pitfalls in determining deterministic components in cointegrating models

Recently the use of the so-called ‘Pantula principle’ has been suggested as a means of determining deterministic components in cointegrating models (see Ahking in Journal of Macroeconomics 24, 2002, and Hatemi-J in Economic Modelling 19, 2002). Moreover, the procedure is suggested in the widely used CATS in RATS program (see [Hansen, H., Juselius, K., 1995. CATS in RATS. Estima, United States]). In this paper, we examine, by means of Monte Carlo simulation, the properties of the ‘Pantula principle’. We investigate the five models contained within the Johansen methodology and find that the ‘Pantula principle’ is heavily biased towards choosing the model with an unrestricted constant when the model with a restricted trend is the true one. We suggest a modification that reduces this bias to an important extent.     

On the constancy of bribe-proof solutions

Summary. We consider the problem of choosing one point in a set of alternatives when monetary transfers are possible. In this context, Schummer (2000) shows that a social choice function must be a constant function if manipulation through bribes is ruled out. But he requires two kinds of domain-richness conditions. One is either smooth connectedness or the finiteness of the set of alternatives and the other is monotonical closedness. However, dispensing with the former condition, we alternatively prove the same result under a weaker condition than monotonical closedness. Received: April 11, 2000; revised version: February 25, 2002 RID="*" ID="*" This paper received the Osaka University Institute of Social and Economic Research Moriguchi Prize in January 2001. I am grateful to Prof. Ryoichi Nagahisa, Prof. Tatsuyoshi Saijo, Prof. Ken-ichi Shimomura, Prof. Ken Urai, and especially two anonymous referees for their useful and helpful comments and suggestions. I am a Research Fellow of the Japan Society for the Promotion of Science.     

Seemingly unrelated regressions with spatial error components

This article considers various estimators using panel data seemingly unrelated regressions (SUR) with spatial error correlation. The true data generating process (DGP) is assumed to be SUR with spatial error of the autoregressive or moving average type. Moreover, the remainder term of the spatial process is assumed to follow an error component structure. Both maximum likelihood (ML) and generalized moments (GM) methods of estimation are used. Using Monte Carlo experiments, we check the performance of these estimators and their forecasts under misspecification of the spatial error process, various spatial weight matrices, and heterogeneous versus homogeneous panel data models.     

Resolving spurious regressions and serially correlated errors

This study investigates the spurious regression phenomenon for two independent stationary and non-stationary processes and illustrates, using a Monte Carlo analysis, that estimation of the spurious regression in first differences or with a lagged dependent variable eliminates the spurious regression problem. Moreover, the results also apply in eliminating the problem of serially correlated errors as well as the problem of ARCH(1) errors.