Showing papers in "Oxford Bulletin of Economics and Statistics in 2009"

Maximum likelihood estimation and inference on cointegration — with applications to the demand for money

Abstract: The estimation and testing of long-run relations in economic modeling are addressed. Starting with a vector autoregressive (VAR) model, the hypothesis of cointegration is formulated as the hypothesis of reduced rank of the long-run impact matrix. This is given in a simple parametric form that allows the application of the method of maximum likelihood and likelihood ratio tests. In this way, one can derive estimates and test statistics for the hypothesis of a given number of cointegration vectors, as well as estimates and tests for linear hypotheses about the cointegration vectors and their weights. The asymptotic inferences concerning the number of cointegrating vectors involve nonstandard distributions. Inference concerning linear restrictions on the cointegration vectors and their weights can be performed using the usual chi squared methods. In the case of linear restrictions on beta, a Wald test procedure is suggested. The proposed methods are illustrated by money demand data from the Danish and Finnish economies.

A Note on the Theme of Too Many Instruments

Abstract: The difference and system generalized method of moments (GMM) estimators are growing in popularity. As implemented in popular software, the estimators easily generate instruments that are numerous and, in system GMM, potentially suspect. A large instrument collection overfits endogenous variables even as it weakens the Hansen test of the instruments’ joint validity. This paper reviews the evidence on the effects of instrument proliferation, and describes and simulates simple ways to control it. It illustrates the dangers by replicating Forbes [American Economic Review (2000) Vol. 90, pp. 869–887] on income inequality and Levine et al. [Journal of Monetary Economics] (2000) Vol. 46, pp. 31–77] on financial sector development. Results in both papers appear driven by previously undetected endogeneity.

PRACTITIONERS’ CORNER: Computing Robust Standard Errors for Within‐groups Estimators*

Abstract: The purpose of this note is to explain how to use standard packages to calculate heteroskedasticity and serial c orrelation consistent standard errors for within-groups estimators of a linear regression model from panel data. The within-groups estimat or is calculated as the least squares estimator in a transformed mult ivariate regression with cross-equation linear restrictions. The Whit e standard errors obtained in this way are the desired ones. Copyright 1987 by Blackwell Publishing Ltd

Finite-sample sizes of johansen's likelihood ratio tests for cointegration

Abstract: This study examines the finite-sample bias of S. Johansen's likelihood ratio tests for cointegration using the Monte Carlo method. Response surface analysis is employed to obtain approximations to the finite-sample critical values and illustrate the individual roles of the sample size, the dimension of the variable system, and the lag order in determining the finite-sample bias of Johansen's tests. The study further shows the importance of lag length selection for Johansen's tests and the performance of standard lag selection criteria in choosing the proper lag length is investigated. Monte Carlo results concerning the sensitivity of Johansen's tests to non-normal innovations are also reported. Copyright 1993 by Blackwell Publishing Ltd

The product cycle hypothesis in a new international environment

Abstract: The last decade has produced a flowering of hypotheses that purport to explain the international trade and direct investment activities of firms in terms of the socalled product cycle. My purpxjse in this paper is to suggest that the power of such hj'potheses has been changing. Two reasons account for that change: one, an increase in the geographical reach of many of the enterprises that are involved in the introduction of new products, a consequence of their having established many overseas subsidiaries; the other, a change in the national markets of the advanced industrialized countries, which has reduced some of the differences that had previously existed between such markets.

PRACTITIONERS CORNER: Tests for Cointegration in Models with Regime and Trend Shifts

Abstract: Recently A. W. Gregory and B. E. Hansen (1996) proposed a number of residual-based tests for cointegration models with the possibility of a structural break. They considered three models: level shift, level shift with trend, and regime shift (both level shift and slope coefficients can change). The authors introduce a more general model that permits a trend shift as well as a regime shift and they provide the critical values appropriate for testing this hypothesis. Copyright 1996 by Blackwell Publishing Ltd

Earnings, independence or unemployment: why become self‐employed?

Abstract: This paper attempts to estimate the role played by three key variables (namely expected earnings, the desire for independence and the ability to find paid employment) on the paid employee/self-employment decision using a simple three stage utility maximization model. The empirical results suggest that individuals are attracted to self-employment because of higher expected earnings relative to paid employment and by the freedom from managerial constraints that it offers. Evidence is also produced supporting the prosperity pull argument for self-employment. Marital status, parents employment status, housing equity and occupational status clearly emerge as significant determinants of labour market choice.

Evaluating the impacts of human capital stocks and accumulation on economic growth: some new evidence.

Abstract: Various hypotheses have been put forward in recent years concerning the contribution of human capital to economic growth. This paper argues that school enrolment rates — by far the most commonly used human capital measure in growth regressions attempting to test these hypotheses — conflate human capital stock and accumulation effects and lead to misinterpretations of the role of labour force growth. An alternative educationrelated human capital measure is constructed which is capable of distinguishing between stocks and flows. Applying this measure to samples of developed and less developed countries during the 1960–85 period suggests not only that there are important growth effects associated both with ‘initial’ stocks of, and subsequent growth in, human capital, but also that this new measure out-performs the simple school enrolment rates used in previous analyses.

Global Financial Transmission of Monetary Policy Shocks

Abstract: The paper analyses the transmission of US monetary policy shocks to global equity markets and the macroeconomic determinants of the underlying transmission process. We show that there is a substantial cross‐country heterogeneity in reactions across 50 equity markets worldwide, with returns falling on average around 2.7% in response to a 100 basis point tightening of US monetary policy, but ranging from a zero response in some to a reaction of 5% or more in other markets. As to the determinants of the strength of transmission to individual countries, we test the relevance of their macroeconomic policies and the role of real and financial integration. We find that in particular the degree of global integration of countries – and not a country's bilateral integration with the United States – is a key determinant for the transmission process.

A Simple Explanation of the Forecast Combination Puzzle

Abstract: This article presents a formal explanation of the forecast combination puzzle, that simple combinations of point forecasts are repeatedly found to outperform sophisticated weighted combinations in empirical applications. The explanation lies in the effect of finite-sample error in estimating the combining weights. A small Monte Carlo study and a reappraisal of an empirical study by Stock and Watson [Federal Reserve Bank of Richmond Economic Quarterly (2003) Vol. 89/3, pp. 71–90] support this explanation. The Monte Carlo evidence, together with a large-sample approximation to the variance of the combining weight, also supports the popular recommendation to ignore forecast error covariances in estimating the weight.

Public action and the quality of life in developing countries

Abstract: An attempt is made to identify the developing countries that have performed better than others in terms of the indicators of the so called "quality of life" relating this progress to the nature of these economies and to the public policies followed in these countries. A table presents data on life expectation at birth and adult literacy rates for 100 countries that had a gross national product per head of less than $3000 in 1977. The data have some comparability and it is reasonable to use them for a rough international comparison of performance. 38 countries have shown distinction in 1 or both of the fields. There are 10 communist countries in the total list of 100 and 9 of them show some distinction. 8 of the 9 do this despite not having literacy figures reported. The entry is longevity which is arguably a more basic indicator of success than poverty. Many of the communist countries are wealthier than the mean or median developing country. Although the indices are relative ones the richer countries have typically done better on the whole. The longevity performance of the communist countries is typically superior. This applies to the poorer group also. Some of the high growth early capitalist countries also have very good performance in terms of the chosen indicators (e.g. Taiwan South Korea Hong Kong and Singapore). Taiwan and Hong Kong have the best overall performance record in terms of the 2 criteria for those 61 countries for which both sets of data are available. The countries that appear to have done relatively worse in terms of the indicators are those in the "middle" i.e. neither communist nor successfully capitalist. There are some exceptions. Tanzania appears to have been relatively successful in terms of the removal of illiteracy and Sri Lanka has been successful in raising life expectancy. In examining the excellent performance of the Republic of Korea and Taiwan the lesson to be learned from their experience is the great importance of employment expansion in poverty removal. The experiences of Sri Lanka and Tanzania are recounted to illustrate the positive role of state action. Like Sri Lankas program of social welfare Tanzanias literacy program shows how much can be achieved by a determined effort sensibly directed toward specific goals. Poverty removal and related features including longevity enhancement is ultimately dependent on a wide distribution of effective entitlements. This for any given level of per capita income--would tend to be reflected in the low level of inequality in the distribution of income.

The quality and efficiency of private and public education: a case‐study of urban india†

Abstract: Recent evidence of a substantial link between quality of schooling and individual productivity suggests that, from an economic efficiency perspective, quality aspects of education deserve attention. This paper presents empirical evidence on the relative quality and efficiency of private and government-funded schools in urban India, using data from Uttar Pradesh. The results suggest that standardizing for home background and controlling for sample selectivity greatly reduces the raw average achievement advantage of private school students over public school students, but does not wipe it out. Private schools' standardized achievement advantage (or better quality) is complemented by their lower unit costs to enable them to be more efficient. The results support much of the existing international evidence on the relative efficiency of private and public schools.

Seasonality and the order of integration for consumption

Abstract: This paper investigates the univariate orders of integration for the variables in a consumption function, emphasizing the role of seasonality in integration testing. Using seasonally-unadjusted quarterly data, the authors find that nondurable consumer expenditure appears to require two levels of differencing, whereas each of its explanatory variables needs to be differenced only once. Consumption is, however, highly seasonal and conventional one-quarter or annual differencing may not be appropriate in such a case. Generalizing the time series process to allow seasonally-varying parameters, only one level of generalized first differencing is needed for this variable. This is consistent with a seasonally-varying form for the consumption function. Coauthors are A. P. L. Chui, Jeremy P. Smith, and C. R. Birchenhall. Copyright 1988 by Blackwell Publishing Ltd

The impact of nutritional status on agricultural productivity: wage evidence from the philippines*

Abstract: This paper examines the impact of individual nutritional status on agricultural wage rates in southern Philippine province. Empirical results concur with other studies which have shown a positive relationship between nutritional status and labor productivity, as measured by wages for agricultural laborers and/or own-farm output, after controlling for simultaneity and a number of other effects. However, the authors' result from better height, a cumulative measure of the absence of poor diets and infection in early childhood, rather than from short-run (calorie intake) or medium-run (weight-for-height) proxies of nutritional status. This implies that short-to-medium-run policies designed to improve calorie intakes and weights of adults will have little impact on agricultural productivity. Rather, productivity increases through better health and nutrition will be more fully realized with a substantial lag as better nourished, healthier children attain better adult height. Copyright 1991 by Blackwell Publishing Ltd

On weak exogeneity in error correction models

Abstract: In this paper, the author discusses the appropriateness of treating explanatory variables of a single-equation error correction model as being weakly exogenous. He restates the derivation of the error correction model, and particularly the single-equation conditional error correction model, both in reduced and structural form. These allow the author to point out that the condition of weak endogeneity implicitly tested by usual orthogonality tests is generally not sufficient for single-equation conditional error correction models. The potential pitfalls of working within incomplete simultaneous dynamic equation models for exogeneity tests are also pointed out. The author illustrates this on some empirical studies. Copyright 1992 by Blackwell Publishing Ltd

PRACTITIONERS CORNER: A Note on the Empirical Power of Unit Root Tests under Threshold Processes†

Abstract: Unit root tests are frequently used to test long-run economic relationships. However, many of these relationships are not expected to follow simple time-series processes due to transaction costs boundaries or thresholds. This paper finds that the power of unit root tests in detecting long-run equilibrium relationships can be significantly diminished by the existence of transaction costs. The authors' results call into question the efficacy of unit root tests in the presence of transactions costs. Copyright 1993 by Blackwell Publishing Ltd

The returns to endogenous human capital in pakistan's rural wage labour market

Abstract: This paper estimates the private returns to human capital in Pakistani rural labour markets. The rich data used permit inclusion of several dimensions of human capital and control for endogeneity resulting from investment in human capital. The results suggest that, without data on determinants of human capital, it would not be possible to disentangle the separate effect of each dimension of human capital on wage differentials nor to distinguish human capital explanations for wage differentials from signalling and credentialist models. With control for endogeneity of human capital and selectivity into wage employment, cognitive achievement, but not schooling attainment separate from cognitive achievement nor long-run health status, affects wage differentials.

Testing parameter constancy and super exogeneity in econometric equations

Abstract: The paper is concerned with testing super exogeneity in a single equation that can either be linear or partially nonlinear. A joint test for testing both weak exogeneity and a form of invariance, which together amount to a form of super exogeneity, is presented and its properties discussed. The considerations also include testing parameter constancy and modelling parameter nonconstancy by an example in which testing super exogeneity in a consumption function for Norway, based on a previously published model is considered. The results are compared with the previous conclusions and found not to disagree with them.